Aquatic Toxicology 95 (2009) 164171

Contents lists available at ScienceDirect

Aquatic Toxicology
journal homepage: www.elsevier.com/locate/aquatox

Fluoxetine treatment affects nitrogen waste excretion and osmoregulation in a marine teleost sh
Michael B. Morando, Lea R. Medeiros, M. Danielle McDonald
Rosenstiel School of Marine and Atmospheric Science, University of Miami, 4600 Rickenbacker Causeway, Miami, FL 33149-1098, USA

a r t i c l e

i n f o

a b s t r a c t
Measurable quantities of the selective serotonin reuptake inhibitor (SSRI), uoxetine, have been found in surface waters and more recently in the tissues of sh. This highly prescribed pharmaceutical inhibits the reuptake of the monoamine, serotonin (5-HT; 5-hydroxytryptamine), causing a local amplication of 5-HT concentrations. Serotonin is involved in the regulation of many physiological processes in teleost sh including branchial nitrogen excretion and intestinal osmoregulation. Since the gill and intestine are directly exposed to the environment, environmental exposure to uoxetine has the potential of affecting both these mechanisms. In the present study, we test the potential sensitivity of these processes to uoxetine by implanting gulf toadsh, Opsanus beta, intraperitoneally with different concentrations of uoxetine (0 (control), 25, 50, 75 and 100 g g1 ). Fluoxetine treatments of 25 and 50 g g1 were sublethal and were used in subsequent experiments. Fish treated with both 25 and 50 g g1 uoxetine had signicantly higher circulating levels of 5-HT than control sh, suggesting that any 5-HT sensitive physiological process could potentially be affected by these two uoxetine doses. However, only sh treated with 25 g g1 uoxetine showed a signicant increase in urea excretion. A similar increase was not measured in sh treated with 50 g g1 uoxetine, likely because of their high circulating levels of cortisol which inhibits urea excretion in toadsh. Intestinal uid absorption appeared to be stimulated in sh treated with 25 g g1 uoxetine but inhibited in 50 g g1 treated sh. Despite these differing responses, both doses of uoxetine resulted in lowered plasma osmolality values, which was expected based on the stimulation of uid absorption in the 25 g g1 uoxetine-treated sh but is surprising with the 50 g g1 treated sh. In the case of the latter, the corresponding stress response invoked by this level of uoxetine may have resulted in an additional osmoregulatory response which accounts for the lowered plasma osmolality. Our ndings suggest that branchial urea excretion and intestinal osmoregulation are responsive to the SSRI, uoxetine, and further investigation is needed to determine the sensitivity of these processes to chronic waterborne uoxetine contamination. 2009 Elsevier B.V. All rights reserved.

Article history: Received 19 January 2009 Keywords: Gulf toadsh Opsanus beta Selective serotonin reuptake inhibitors Pharmaceuticals Toxicants Serotonin Cortisol Urea transport Intestine

1. Introduction Fluoxetine, the active compound in ProzacTM , is a selective serotonin reuptake inhibitor (SSRI) that works to prevent the reuptake of serotonin (5-hydroxtryptamine, 5-HT) by 5-HT specic transporters located in neurons, human platelets (Lesch et al., 1993), lymphocytes (Faraj et al., 1994; Lima and Urbina, 2002; Yang et al., 2007) and the gastrointestinal tract (Wade et al., 1996; reviewed by Gershon, 2003; Martel, 2006). It is used most commonly for the treatment of illnesses such as depression and anxiety as well as panic and obsessive-compulsive disorders. After consumption, uoxetine is metabolized to noruoxetine, which in itself, can act as

DOI of original article:10.1016/j.aquatox.2009.03.011. Corresponding author. Tel.: +1 305 421 4856. E-mail address: dmcdonald@rsmas.miami.edu (M.D. McDonald). 0166-445X/$ see front matter 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.aquatox.2009.10.015

an SSRI with similar potency to that of uoxetine (Fuller et al., 1992; Hiemke and Hrtter, 2000). Fluoxetine and noruoxetine are then excreted from the human body primarily through urine, with up to 11% of the administered uoxetine dose being excreted as the unchanged parent compound (Altamura et al., 1994; Hiemke and Hrtter, 2000). Both compounds enter wastewater treatment facilities in human waste and through the disposal of unused drug down the sink or toilet, and these treatment facilities are ill-equipped to remove these compounds. Consequently, low levels of uoxetine and its potent metabolite have been continuously released into the environment, resulting in now measurable quantities in freshwater (Kolpin et al., 2002) and marine coastal (Vassog et al., 2008) ecosystems. While the potential human health consequences through the consumption of tainted sh or drinking water are considered very low (Brooks et al., 2005), the effects of chronic uoxetine exposure on aquatic organisms are still not completely understood.

M.B. Morando et al. / Aquatic Toxicology 95 (2009) 164171

165

Because uoxetine treatment results in an increase in 5-HT concentrations, any physiological process that is regulated by 5-HT may be susceptible to environmental uoxetine exposure; particularly those that are present in organs exposed directly to the environment. In the case of marine teleost sh, these organs would include the gill and the gastrointestinal tract (reviewed by Marshall and Grosell, 2005). The gulf toadsh, Opsanus beta, is a benthic, marine teleost sh that is found along the coast of Florida and within the Gulf of Mexico. This marine teleost may be a particularly useful model for studying potentially sensitive physiological endpoints for uoxetine exposure, not only because ongoing research on the toadsh has been systematically characterizing the physiological processes (i.e., urea excretion, intestinal ion transport, cardiovascular physiology, respiration, stress hormone secretion) that are regulated by 5-HT (Wood et al., 2003; McDonald and Walsh, 2004; Sloman et al., 2005; McDonald, unpublished) but because high uoxetine levels have been measured in sediment, where benthic organisms are found (Fent et al., 2006). Arguably, the best studied serotonergic process in toadsh is its mechanism of urea excretion across the gill. In brief, urea is not excreted continuously across the gill but in distinct pulses that occur once or twice daily. This pulsatile urea excretion is attributed to a urea transport protein (tUT) in the gill (Walsh et al., 2000) that is believed to be periodically activated or inserted into the gill membrane. The regulation of tUT involves the hormone, cortisol, which is the primary glucocorticoid released in times of stress, and 5-HT; with elevated circulating cortisol levels inhibiting urea excretion (Wood et al., 2001; McDonald et al., 2004, 2009) while high 5-HT concentrations appear to stimulate urea excretion (Wood et al., 2003; McDonald and Walsh, 2004). This mechanism is not only important for nitrogen waste excretion, a necessary physiological process, but appears also to have a role in social interaction (Sloman et al., 2005). Evidence suggests that toadsh may have cognitive control over their ability to pulse; when two sh were placed in the same tank, they pulse within 1 h of one another (Sloman et al., 2005). Another important mechanism that is regulated by 5-HT in teleost sh is gastrointestinal tract (GIT) function. In marine teleosts, the intestine plays a major role in osmoregulation: marine sh drink the surrounding seawater and the uid is absorbed across the intestine as a way to replace the water continuously lost to the environment by osmosis (reviewed by Marshall and Grosell, 2005). Many aspects of GIT function have been shown to be regulated by 5-HT using mammalian in vitro preparations including gut motility (Butler et al., 1988, 1990) and gastric acid secretion (LePard and Stephens, 1994). In terms of marine teleost osmoregulation, Mori and Ando (1991) determined that 5-HT lowered the absorption of Na+ , Cl , and water across the eel intestine. Thus, direct exposure of the GIT to uoxetine in the environment through drinking could have implications on intestinal ion absorption, which is a crucial element of marine osmoregulation. The goal of this current study is to determine whether urea excretion and osmoregulation, two physiological processes that are regulated by 5-HT, are sensitive to treatment with the SSRI, uoxetine. Sensitivity to uoxetine treatment would suggest that these physiological endpoints had the potential to be responsive to environmental uoxetine exposure. To test the hypothesis that urea excretion and osmoregulation would be sensitive to uoxetine treatment, gulf toadsh were implanted intraperitoneally with different concentrations of uoxetine and survival, plasma hormone levels, nitrogen excretion and various osmoregulatory and acid base parameters were measured. In addition, the effect of uoxetine treatment on toadsh urea transporter (tUT) transcript levels, drinking rate and intestinal uid absorption was determined.

2. Materials and methods 2.1. Experimental animals Gulf toadsh (O. beta) were captured by roller trawl used by commercial shrimpers in Biscayne Bay, Florida in the summer of 2007. The toadsh were then immediately transferred to the laboratory where they were held for up to one month. Fish were treated with a dose of malachite green (nal concentration 0.05 mg l1 ) in formalin (15 mg l1 ) (AquaVet, Hayward, CA, USA) on the day of transfer to the laboratory in order to prevent infection by the ciliate Cryptocaryon irritans (Stoskopf, 1993). Initially, the sh were kept in 50-l glass aquaria with owing, aerated seawater in relatively uncrowded conditions (10 g sh l1 ). At least 48 h prior to experimentation, the sh were placed in a small crowding tank (10 l) at a much higher density (50 g sh l1 ) to initiate a switch to ureotelism (Walsh et al., 1994). The water temperature was 2429 C. Fish were fed weekly with squid up until the time of implantation.

2.2. Experimental protocol 2.2.1. Series i: the effect of uoxetine treatment on survival, nitrogen excretion, plasma and intestinal uid composition and tUT mRNA expression Toadsh (N = 78; mean weight 0.030 0.004 kg, range 0.0170.050 kg) were chosen at random from the crowding tanks. Toadsh were anesthetized with 1 g l1 MS 222 and wrapped with wet towels. Fluoxetine (uoxetine-HCl; SigmaAldrich Chemicals) at ve different concentrations (0 (control), 25, 50, 75 and 100 g g1 ) was mixed with warmed coconut oil for a total implant volume of 5 l g1 sh and was injected into the peritoneal cavity of the sh with an 18 G needle. The intraperitoneal injection of compounds (such as uoxetine) in oil vehicles (i.e., peanut oil, coconut oil, corn oil) mediates the slow release of these substances into the circulation (Vijayan and Leatherland, 1989; Christensen et al., 1999; McDonald et al., 2004). After injection, the sh were put on ice for one minute so the implant could solidify after which time they were added randomly to their own individual, shielded ux chamber and allowed to recover overnight undisturbed. After recovery, water ow to the chambers was stopped and without disturbing the sh, the water level quickly dropped to a volume mark between 1.1 and 1.6 l through a small hole in the ux chamber. Vigorous aeration maintained thorough mixing and the oxygen partial pressure (PO2 ) close to air saturation. A water sample (2.5 ml) was taken for the measurement of initial urea and ammonia concentrations. Thereafter, water samples were taken every hour for up to 8 h; at 24 h a nal water sample was taken to characterize branchial urea excretion. At 24 h, blood was sampled by caudal puncture. Briey, toadsh were netted out of their individual boxes and wrapped with wet towels. Using a 1 ml syringe with a 23 G rinsed with heparinized saline (50 i.u. ml1 ; SigmaAldrich), blood was quickly drawn from the caudal vein. This sampling process took less than 2 min; thus an increase in circulating cortisol due to this handling stress would be undetectable in the given sampling time. There was no difference in handling time between treatment groups. A fraction of the whole blood was analyzed for pH and hematocrit. The remaining blood was immediately centrifuged for 10 min at 13,000 g to separate the plasma; the plasma was frozen in liquid N2 and maintained at 80 C for later analysis of serotonin, cortisol, protein, urea, and osmolality. Toadsh were then anesthetized in concentrated MS 222 (2 g l1 ) and gill was removed, frozen in liquid N2 and maintained at 80 C for analysis of toadsh urea transporter mRNA expression. Intestinal uid was also sampled, frozen in liquid N2 then maintained at 80 C for later analysis of osmolality.

166

M.B. Morando et al. / Aquatic Toxicology 95 (2009) 164171

2.2.2. Series ii: potential effects of sublethal uoxetine treatment on intestinal osmoregulation A separate group of toadsh (N = 30; mean weight 0.025 0.001 kg, range 0.0170.040 kg) were implanted with 0 (control), 25 or 50 g g1 uoxetine as describe above and the sh of the different treatment groups (N = 10 in each) were separated into three 10 l tanks that were supplied with ow-through aerated seawater. Toadsh were left undisturbed to acclimate to their new surroundings overnight. After acclimation, water ow to the tanks was stopped and an exact volume mark of 4.0 l was reached by siphoning the water from the tanks. To initiate the drinking rate experiment, 10 Ci [3 H]-polyethylene glycol-4000 (PEG-4000; specic activity: 2050 Ci g1 ; American Radiochemicals Inc.), an extracellular marker, was added to each of the tank (staggered by 1 h to allow for sampling) as previously described by Grosell et al. (2004). Water samples (5 ml) were taken for radioactivity measurements at 0.25, 3.0 and 6.0 h after addition of the PEG-4000. After the nal water sample, MS-222 was added from a neutralized stock solution to yield a nal concentration of 0.2 g l1 in the exposure tanks. Subsequently, individual sh were netted out of the exposure tanks, rinsed in non-radioactive seawater, dried and weighed. Then, the GIT of the sh was exposed, clamped at the start of the esophagus and immediately anterior to the anus by hemostats to prevent loss of contents, carefully placed in a vial and weighed. An aliquot of intestinal uid (25 l) was removed for analysis of PEG-4000 activity. The entire GIT and remaining intestinal uid was then homogenized in 4 vol. of 10% perchloric acid and the homogenate centrifuged (500 g for 5 min). Aliquots of 1.5 ml of clear, protein-free supernatant were used to determine radioactivity in the GIT. 2.2.3. Quantitative (Realtime) PCR Total RNA was isolated from whole gill within one month of obtaining the sample following the protocol provided by Trizol reagent (Invitrogen) and treated with DNAse to remove potential residual genomic DNA (TurboDNA-free kit, Ambion). cDNA synthesis using random hexamers was performed according to the protocol provided with the SuperScript First-Strand Synthesis System for RT-PCR (Invitrogen). Quantitative (Realtime) PCR (qPCR) was performed on tUT (gene of interest; GOI) and 18S (normalizer gene) using a Mx4000 Multiple Quantitative PCR system (Stratagene) with SYBR Green and primers as described by McDonald et al. (2009). The amount of cDNA for the GOI was expressed relative to the amount of cDNA from the normalizer gene. The stability of normalizer gene expression with treatment was tested by comparing normalizer gene expression in sh from different treatment groups using cDNA obtained from normalized quantities of total RNA as described in Schmittgen and Zakrajsek (2000). To determine that the amplication/detection efciencies of the GOI and the normalizer gene were similar, a standard curve was generated with known quantities of cDNA for the GOIs and the normalizer gene plotted versus their Ct value. The standard curves of the GOI and normalizer gene in the present study gave PCR efciencies of 100% and 80.6%, respectively. To ensure that the amplication of only one PCR product was contributing to the measured Ct value, a dissociation curve was established for each product, revealing only a single peak signifying only one amplied product. No template controls were also run to ensure that primer concentrations were optimized and primer dimers were not contributing to the uorescence. Lastly, no RT (reverse transcriptase) controls, completed during the cDNA synthesis step, were run to ensure the lack of genomic DNA contamination. To verify that the correct product for each primer pair was being amplied, the size of the PCR product for a subset of samples was determined using gel electrophoresis. Gel extracted PCR products (with Qiaex II from Qiagen) were then cloned in Escherichia coli according to the protocol provided with

the TOPO TA Cloning Kit for sequencing using TOP10 Chemically Competent One Shot Cells (Invitrogen). The plasmid cDNA was isolated (Miniprep, Qiagen) and the clones sequenced and identied by Gene Gateway LLC, CA, USA. 2.3. Analytical techniques Plasma cortisol concentrations were measured using a commercial [125 I] radioimmunoassay kit (MP Biomedicals Inc.) with standards diluted to the same protein range as toadsh plasma. Plasma protein concentration was measured by Bradford assay (SigmaAldrich). Serotonin (5-HT) concentrations were measured by ELISA (ALPCO Diagnostics). Urea concentrations in plasma and water were measured using the diacetyl monoxime method of Rahmatullah and Boyde (1980), with appropriate adjustments of reagent strength for the different urea concentration ranges in water and plasma. Ammonia concentrations in the water were measured by the indophenol blue method (Ivancic and Degobbis, 1984). Osmolality was analyzed in both plasma and intestinal uid using a Wescor vapor pressure osmometer. Hematocrit was measured by adding blood to a hematocrit tube and then spinning it down in a microcapillary centrifuge (International Equipment Company) for 2 min. The pH of whole blood was measured using a small bore pH electrode (Accumet) and an Orion pH meter. The majority of urea excretion is via the gills (>90%); the kidney of toadsh contributes to only a small percentage (<10%; McDonald et al., 2000) and the permeability of toadsh skin to urea is extremely low (5.07 0.56 (8) 107 cm s1 ; Part et al., 1999). The branchial excretion rate (in mol-N kg1 24 h1 ) of urea-N and ammonia-N was calculated from changes in concentration in the external seawater (mol-N l1 ) multiplied by volume (l) and divided by the mass (kg) and corrected for time (24 h). [3 H] PEG-radioactivity in the water samples from the drinking rate experiment was determined using liquid scintillation counting. Non-radioactive seawater was added to intestinal uid and homogenate samples to a total volume of 5 ml; to which 10 ml Ecolume uor (MP Biomedicals) was added. Radioactivity was determined using a Beckman LS3801 liquid scintillation counter and quench correction was performed manually by internal standardization. Drinking rates (ml kg1 h1 ) were calculated from the radioactivity recorded in the GIT taking into account the weight of the sh, the PEG-4000 incubation time, and the PEG-4000-specic activity of the water. 2.4. Statistics Data were reported as means 1 S.E.M. (N = number of sh). The signicance of differences between treatments was determined using a one-way ANOVA followed by a HolmSidak test for multiple comparisons. If the data was not normally distributed upon log transformation, a one-way ANOVA based on ranks was used. 3. Results Fish implanted with the highest concentrations of uoxetine (75 and 100 g g sh1 ) had a very low percentage of toadsh survive (<25%, N = 16) while the lowest concentration (25 g g1 ) had a survivorship almost identical to that of the control (94%, N = 16) (Fig. 1). Fish treated with 50 g g1 showed a decreased survivorship rate (69%, N = 16) but not as great as the higher concentrations. Based on these ndings, the sublethal concentrations of uoxetine implant that were used for subsequent analysis of physiological parameters were 25 and 50 g g1 . Plasma 5-HT concentrations in sh implanted with 25 g g1 (3.8 0.21 ng ml1 ; N = 11) and 50 g g1 (3.7 0.5 ng ml1 ; N = 9) uoxetine were higher than sh treated with 0 g g1

M.B. Morando et al. / Aquatic Toxicology 95 (2009) 164171

167

Fig. 1. A comparison of mortality rates (% survival) of the different implant concentrations versus time (h). Fish treated with the two highest concentrations (75 and 100 g g1 ) of uoxetine showed a signicant reduction in % survival compared to the other treatments. Values are means S.E.M.; * P < 0.05 denotes a signicant difference compared to value at 12 h.

(3.0 0.1 ng ml1 ; N = 13) (Fig. 2A) but no signicant difference was found between the two treatment groups. Fish treated with 25 g g1 uoxetine had circulating cortisol concentrations (76.0 23.1 ng ml1 , N = 11) that were not signicantly different from control (107.1 20.4 ng ml1 , N = 12) while the sh treated with 50 g g1 uoxetine had circulating cortisol levels (368.5 96.6 ng ml1 , N = 8) that were more than 3-fold greater than the control and 25 g g1 treated sh (Fig. 2B).

Fig. 3. (A) Urea excretion rates of sh treated with 25 g g1 uoxetine were signicantly higher than the other two treatments. (B) There was not a signicant difference in the rate of ammonia excretion between the three treatment groups. (C) The relative mRNA expression in toadsh urea transporter (tUT) in the gill was consistent amongst all three treatment groups. Values are means S.E.M.; * P < 0.05 denotes a signicant difference compared to control, t P < 0.05 denotes a signicant difference compared to 50 g g1 .

Fig. 2. (A) Plasma serotonin concentrations are signicantly higher in sh treated with 25 and 50 g g1 uoxetine compared to control. (B) Circulating cortisol concentrations are signicantly elevated only in sh treated with 50 g g1 uoxetine. Values are means S.E.M.; * P < 0.05 denotes a signicant difference compared to control.

Toadsh treated with 25 g g1 uoxetine had rates of ureaN excretion (10,539 862 mol-N kg1 24 h1 ; N = 14) that were almost twice as high as both the control (5761 523 molN kg1 24 h1 ; N = 14) and toadsh treated with 50 g g1 uoxetine (5520 1190 mol-N kg1 24 h1 ; N = 8) (Fig. 3A) with no signicant difference between the control and the 50 g g1 uoxetine groups. The rate of ammonia excretion was similar amongst controls and sh treated with either 25 or 50 g g1 uoxetine (Fig. 3B). There was no signicant difference in plasma protein or plasma urea concentrations amongst the three groups of sh (Table 1). Plasma osmolality in sh treated with 25 g g1 uoxetine (290.3 2.8 mosm l1 ; N = 11) and 50 g g1 uoxetine (287.4 4.0 mosm l1 ; N = 8) was signicantly lower than measured values in control sh (305.1 4.0 mosm l1 ; N = 13) (Fig. 4A).

168

M.B. Morando et al. / Aquatic Toxicology 95 (2009) 164171

Table 1 The effect of uoxetine treatment on various blood parameters. Control Plasma protein (mg ml ) Plasma urea (mmol-N l1 ) Hematocrit Plasma pH
1

25 g g1 1.4 1.2 3.8 0.08 13.4 4.8 29.8 7.71 1.1 0.5 3.4 0.04

50 g g1 12.6 5.4 20.5 7.55 0.6 2.0 1.6* 0.09

15.4 7.4 35.4 7.68

Values are means 1 S.E.M. * P < 0.05 compared to control value.

A similar trend was observed in intestinal uid osmolality; however, there was no signicant difference found between any of the groups (Fig. 4B). Drinking rate was also not signicantly different amongst the three different treatment groups (Fig. 5A); however, sh treated with 25 g g1 uoxetine had signicantly higher intestinal uid:seawater PEG 4000 ratios, signifying a greater absorption of water across the intestine than controls (Fig. 5B). In contrast, sh treated with 50 g g1 uoxetine had signicantly lower intestinal uid:seawater PEG 4000 ratio (Fig. 5B). Fish treated with 50 g g1 uoxetine (20.5 1.6%; N = 8) had a signicantly lower hematocrit than control sh (35.4 3.8%; N = 12); however there was no signicant difference between the control sh and those treated with 25 g g1 uoxetine (29.8 3.4%; N = 15) (Table 1). There was no signicant difference in the blood pH of any of the groups (Table 1). 4. Discussion The primary effect of a selective serotonin reuptake inhibitor (SSRI), such as uoxetine, is to increase synaptic 5-HT levels by selectively inhibiting a 5-HT transporter that functions to remove 5-

Fig. 5. (A) No signicant changes in drinking rate were measured in sh treated with uoxetine. (B) The ratio of intestinal uid PEG: seawater PEG concentration was signicantly elevated, suggesting an increased intestinal uid absorption, in sh treated with 25 g g1 uoxetine compared to controls and 50 g g1 -treated sh. The opposite effect was measured in sh treated with 50 g g1 uoxetine, indicating a reduced uid absorption at the level of the intestine. * P < 0.05 denotes a signicant difference compared to control, t P < 0.05 denotes a signicant difference compared to 50 g g1 .

Fig. 4. (A) A signicant reduction in plasma osmolality is measured in both groups of uoxetine-treated sh compared to control sh. (B) A signicant difference was not measured in intestinal uid osmolality. Values are means S.E.M.; * P < 0.05 denotes a signicant difference compared to control.

HT from the synaptic cleft back into the presynaptic neuron. These 5-HT transporters (SERT) are not only found on neurons, but are believed to be present on blood platelets (Lesch et al., 1993), lymphocytes (Faraj et al., 1994; Lima and Urbina, 2002; Yang et al., 2007) and on enterocytes of the gastrointestinal tract (Wade et al., 1996; reviewed by Gershon, 2003; Martel, 2006). Thus, treatment with uoxetine could easily result in a systemic increase in circulating 5-HT concentrations by inhibiting SERT present on nonneural cells in addition to the hypothetical spill-over from increases in 5-HT within the neural synaptic cleft. Furthermore, elevations in circulating 5-HT levels after acute uoxetine treatment have been measured in mammals (Rothman et al., 2008). At the same time, there is evidence in mammals that chronic treatment with uoxetine has the opposite effect on circulating 5-HT concentrations, likely due to a depletion in 5-HT stores within non-neural cells (Rothman et al., 2008). In the present study, plasma 5-HT levels were found to be elevated in both groups of sh that were implanted with uoxetine. While we believe the present study is the rst to measure the effects of uoxetine on circulating 5-HT in sh, a study by Gaworecki and Klaine (2008) measured brain 5-HT levels in hybrid striped bass exposed for 6 days to waterborne uoxetine levels ranging from 0 to approximately 100 g l1 . In this study, brain 5-HT and 5-HT turnover rates were shown to decrease as exposure time increased (Gaworecki and Klaine, 2008). While the data from the two studies cannot be directly compared, it is evident from both studies that uoxetine will result in alterations in 5-HT concentrations of blood and tissues. Thus, the

M.B. Morando et al. / Aquatic Toxicology 95 (2009) 164171

169

potential consequences of uoxetine treatment are widespread and any physiological process that is sensitive to 5-HT is potentially susceptible. In the gulf toadsh, one of these 5-HT sensitive processes is pulsatile urea excretion, which involves a urea transport protein, tUT, located in the gill (Walsh et al., 2000). Because pulsatile urea excretion is activated by arterial injection of 5-HT into the circulation (Wood et al., 2003; McDonald et al., 2004), one would predict that uoxetine treatment, which results in an increase in circulating levels of 5-HT, would cause an increase in urea excretion compared to control sh. In fact, the implantation of uoxetine caused an almost a 2-fold increase in the excretion of urea, but only in sh treated with the lowest level of uoxetine (25 g g1 ). We believe that this effect on urea excretion is on the pulsatile urea excretion mechanism itself, as there was not a signicant change in ammonia excretion, plasma urea concentrations or plasma protein levels, which could indicate changes in nitrogen metabolism. Surprisingly, a dose-dependent increase in urea excretion was not measured in sh treated with the highest level of uoxetine (50 g g1 ). However, this result could be explained by the different stress reactions, in terms of cortisol release, between the two groups and the role that cortisol plays in regulating urea excretion in toadsh. The stress hormone, cortisol, is believed to play an important but complicated role in the regulation of pulsatile urea excretion. On the acute level, high plasma levels of cortisol appear to inhibit urea excretion (Wood et al., 1998, 2001; McDonald et al., 2004, 2009; Rodela et al., 2009). Specically, a urea pulse is preceded by a drop in circulating cortisol levels, the urea pulse occurs and then circulating cortisol levels return to pre-pulse concentrations. If the drop in cortisol is prevented by exogenous cortisol loading through infusion, pulse size dramatically decreases, suggesting that there are fewer functional tUTs (McDonald et al., 2004). We have recently ruled out transcriptional inhibition of tUT by cortisol; in fact, sh infused with cortisol show a paradoxical increase in tUT transcription compared to sh infused with saline alone (McDonald et al., 2009). Instead, we believe that the drop in circulating cortisol levels is permissive to tUT activation with high circulating cortisol levels interfering on the post-transcriptional level with tUT activation by 5-HT. The results of the present study support this idea, as despite elevated circulating 5-HT levels similar to the 25 g g1 treated sh, toadsh treated with 50 g g1 uoxetine do not show a stimulation in urea excretion; the only difference between the two groups being the high circulating cortisol concentrations in the latter. While cortisol elevation has been shown to stimulate tUT transcription (McDonald et al., 2009), levels of tUT mRNA expression were unchanged in the 50 g g1 treated toadsh. This is likely due to the time frame; toadsh used in the study by McDonald et al. (2009) were stressed for >96 h. In the present study, toadsh were treated with uoxetine for only 36 h and it is possible that cortisol levels were not elevated for the entire treatment period. Interestingly, 5-HT regulates the hypothalamic pituitaryinterrenal (HPI) axis, the neuroendocrine axis that is responsible for the teleost stress response. Essentially, the HPI axis involves the release of corticotropin releasing hormone (CRH, also called corticotropin releasing factor, CRF) from the hypothalamus which stimulates the secretion of adrenocorticotropin (ACTH) from the pituitary (reviewed by Mommsen et al., 1999). ACTH then stimulates the secretion of cortisol from the interrenal cells, which participates in homeostasis and the response of an organism to a stressor. In mammals, the release of CRH from the hypothalamus is under serotonergic control; neurons in the hypothalamus express various 5-HT receptors and a direct synaptic connection exists between serotonergic nerves from the raphe nucleus and CRHcontaining neurons in the hypothalamus (Carrasco and Van De Kar, 2003; Liposits et al., 1987; Sawchenko et al., 1983). Furthermore, 5-HT directly stimulates both ACTH and glucocorticoid release,

independent of its action on CRH (Jrgensen et al., 1999; Van De Kar et al., 2001). While not as well understood, a similar relationship is believed to exist in sh and injection with the 5-HT1A receptor agonist, 8-OH-DPAT, has been shown to have effects on ACTH and cortisol secretion (Winberg et al., 1997; Hglund et al., 2002; Medeiros and McDonald, unpublished). That being said, it is not surprising that uoxetine treatment, which stimulates an increase in circulating levels of 5-HT in toadsh, also results in an increase in cortisol secretion under certain conditions. It is slightly unexpected that only sh treated with 50 g g1 uoxetine showed a stimulation in the HPI axis despite having circulating levels of 5-HT similar to sh treated with 25 g g1 . However, it is not unlikely that the 50 g g1 injected sh experienced a spike in circulating 5-HT concentrations early on in the treatment that might have stimulated cortisol secretion, which was missed with our terminal blood sampling protocol (i.e., a single 36 h post-uoxetine injection blood sample). More work including serial blood sampling would be necessary to conrm this hypothesis. Regardless, these data do suggest that uoxetine does have the capacity to stimulate the teleost stress response by increasing cortisol secretion, a hormone that has widespread physiological effects as evident by its apparent role in regulating toadsh urea excretion. In terms of osmoregulation, marine teleosts are hypoosmotic to the environment and consequently are continuously plagued with diffusive ion gain and osmotic water loss. Therefore, to maintain osmotic and ionic balance marine sh must replace the water lost by drinking the surrounding seawater and absorbing monovalent ions and uid across the intestine (reviewed by Marshall and Grosell, 2005). Marine teleosts will then actively pump out monovalent ions across the gill (reviewed by Marshall and Grosell, 2005). We hypothesized that sh treated with uoxetine would experience a disruption in osmoregulation, since treatment with 5-HT has been shown to lower the absorption of Na+ , Cl , and water across an in vitro eel intestine preparation, resulting in a decrease in the serosa-negative transepithelial potential difference (Mori and Ando, 1991). Furthermore, preliminary evidence using in vitro toadsh intestine preparation shows a similar reduction in transepithelial potential in response to 5-HT, suggesting that the monoamine may also be lowering the absorption of monovalent ions and water in toadsh (McDonald and Grosell, unpublished). If this were to translate to the whole animal, we would predict that high circulating levels of 5-HT caused by uoxetine treatment may result in a reduction in intestinal uid absorption and a consequent increase in plasma osmolality. The action of uoxetine on toadsh intestinal uid absorption appears to be complicated; uid absorption appeared to be stimulated in toadsh treated with 25 g g1 uoxetine based on elevated intestinal uid: seawater PEG 4000 ratios and sh treated with 50 g g1 uoxetine displayed an inhibition in intestinal uid absorption, as originally hypothesized. Despite these differing responses, both doses of uoxetine resulted in lowered plasma osmolality values, a nding that may have been expected based on the stimulation of uid absorption in the 25 g g1 uoxetinetreated sh but is surprising with the 50 g g1 treated sh. In the case of the latter, the higher dose of uoxetine or perhaps the corresponding stress response invoked by this level of uoxetine, may have resulted in an additional osmoregulatory response which could account for the lowered plasma osmolality. For example, in addition to its role in the stress response, cortisol is a key osmoregulatory hormone in marine teleost sh and high levels result in an upregulation in branchial Na+ , K+ ATPase activity, which drives Na+ and Cl extrusion at the gill (reviewed by Evans et al., 2005). Interestingly, there appeared to be a dose-dependent decrease in hematocrit in sh treated with uoxetine, with this trend being signicant in the 50 g g1 -treated sh. This decrease in hematocrit is consistent with a decrease in plasma osmolality and a dilution

170

M.B. Morando et al. / Aquatic Toxicology 95 (2009) 164171 Grosell, M., McDonald, M.D., Walsh, P.J., Wood, C.M., 2004. Effects of prolonged copper exposure in the marine gulf toadsh (Opsanus beta) II: copper accumulation, drinking rate and Na+ /K+ -ATPase activity in osmoregulatory tissues. Aquat. Toxicol. 68, 263275. Hiemke, C., Hrtter, S., 2000. Pharmacokinetics of selective serotonin reuptake inhibitors. Pharmacol. Therapeut. 85, 1128. Hglund, E., Balm, P.H.M., Winberg, S., 2002. Stimulatory and inhibitory effects of 5-HT1A receptors on adrenocorticotropic hormone and cortisol secretion in a teleost sh, the Arctic charr (Salvelinus alpinus). Neurosci. Lett. 324, 193196. Ivancic, I., Degobbis, D., 1984. An optimal manual procedure for ammonia analysis in natural waters by the indophenol blue method. Water Res. 1B, 11431147. Jrgensen, H., Knigge, U., Kjr, A., Warberg, J., 1999. Adrenocorticotropic hormone secretion in rats induced by stimulation with serotonergic compounds. J. Neuroendocrinol. 11, 283290. Kolpin, D.W., Furlong, E.T., Meyer, M.T., Thurman, E.M., Zaugg, S.D., Barber, L.B., Buxton, H.T., 2002. Pharmaceuticals, hormones, and other organic wastewater contaminants in US streams, 19992000: a national reconnaissance. Environ. Sci. Technol. 36, 12021211. LePard, K.J., Stephens, R.L., 1994. Serotonin inhibits gastric acid secretion through a 5-hydroxytryptamine1-like receptor in the rat. J. Pharmacol. Exp. Therapeut. 270, 11391144. Lesch, K.-P., Wolozin, B.L., Murphy, D.L., Riederer, P., 1993. Primary structure of the human platelet serotonin uptake site: identity with the brain serotonin transporter. J. Neurochem. 60, 23192322. Lima, L., Urbina, M., 2002. Serotonin transporter modulation in blood lymphocytes from patients with major depression. Cell. Mol. Neurobiol. 22, 797804. Liposits, Z., Phelix, C., Paull, W.K., 1987. Synaptic interaction of serotonergic axons and corticotropin releasing factor (CRF) synthesizing neurons in the hypothalamic paraventricular nucleus of the rat, a light and electron microscopic immunocytochemical study. Histochemistry 86, 541549. Marshall, W.S., Grosell, M., 2005. Ion transport, osmoregulation and acid-base balance. In: Evans, D.H., Claiborne, J.B. (Eds.), Physiology of Fishes. CRC Press, pp. 177230. Martel, R., 2006. Recent advances on the importance of the serotonin transporter SERT in the rat intestine. Pharmacol. Res. 54, 7376. McDonald, M.D., Walsh, P.J., 2004. 5-HT2A -like receptors are involved in triggering pulsatile urea excretion in the gulf toadsh Opsanus beta. J. Exp. Biol. 207, 20032020. McDonald, M.D., Wood, C.M., Wang, Y., Walsh, P.J., 2000. Differential branchial and renal handling of urea, acetamide and thiourea in the gulf toadsh, Opsanus beta: evidence for two transporters. J. Exp. Biol. 203, 10271037. McDonald, M.D., Wood, C.M., Grosell, M., Walsh, P.J., 2004. Glucocorticoid receptors are involved in the regulation of pulsatile urea excretion in toadsh. J. Comp. Physiol. 174B, 649658. McDonald, M.D., Vulesevic, B., Perry, S.F., Walsh, P.J., 2009. Urea transporter and glutamine synthetase regulation and localization in gulf toadsh gill. J. Exp. Biol. 212, 704712. Mommsen, T.P., Vijayan, M.M., Moon, T.W., 1999. Cortisol in teleosts: dynamics, mechanisms of action and metabolic regulation. Rev. Fish. Biol. Fish. 9, 211268. Mori, Y., Ando, M., 1991. Regulation of ion and water transport across the eel intestine: effects of acetylcholine and serotonin. J. Comp. Physiol. 161B, 387392. Part, P., Wood, C.M., Gilmour, K.M., Perry, S.F., Laurent, P., Zadunaisky, J., Walsh, P.J., 1999. Urea and water permeability in the ureotelic gulf toadsh (Opsanus beta). J. Exp. Zool. 283, 112. Rahmatullah, M., Boyde, T.R., 1980. Improvements in the determination of urea using diacetyl monoxime; methods with and without deproteination. Clin. Chim. Acta 107, 39. Rodela, T.M., McDonald, M.D., Walsh, P.J., Gilmour, K.M., 2009. The regulatory role of glucocorticoid and mineralocorticoid receptors in pulsatile urea excretion of the gulf toadsh, Opsanus beta. J. Exp. Biol., in press. Rothman, R.B., Zolkowska, D., Baumann, M.H., 2008. Serotonin (5-HT) transporter ligands affect plasma 5-HT in rats. Ann. N. Y. Acad. Sci. 1139, 268284. Sawchenko, P.E., Swanson, L.W., Steinbusch, H.W., Verhofstad, A.A., 1983. The distribution and cells of origin of serotonergic inputs to the paraventricular and supraoptic nuclei of the rat. Brain Res. 277, 355360. Schmittgen, T.D., Zakrajsek, B.A., 2000. Effect of experimental treatment on housekeeping gene expression: validation by real-time, quantitative RT-PCR. J. Biochem. Biophys. Methods 46, 6981. Sloman, K.A., McDonald, M.D., Barimo, J.F., Lepage, O., Winberg, S., Wood, C.M., Walsh, P.J., 2005. Does pulsatile urea excretion serve as a social signal in the gulf toadsh, Opsanus beta? Physiol. Biochem. Zool. 78, 724735. Stoskopf, M.K., 1993. Fish Medicine. W.B. Saunders Co., Philadelphia, p. 882. Van De Kar, L.D., Javed, A., Zhang, Y.H., Serres, F., Raap, D.K., Gray, T.S., 2001. 5-HT2A receptors stimulate ACTH, corticosterone, oxytocin, renin and prolactin release and activate hypothalamic CRF and oxytocin-expressing cells. J. Neurosci. 21, 35723579. Vassog, T., Anderssen, T., Pedersen-Bjergaard, S., Kallenborn, R., Jensen, E., 2008. Occurrence of selective serotonin reuptake inhibitors in sewage and receiving waters at Spitsbergen and in Norway. J. Chromatogr. 1185A, 194 205. Vijayan, M.M., Leatherland, J.F., 1989. Cortisol-induced changes in plasma glucose, protein, and thyroid hormone levels, and liver glycogen content of coho salmon (Oncorhynchus kisutch Walbaum). Can. J. Zool. 67, 27462750. Wade, P.R., Chen, J., Jaffe, B., Kassem, I.S., Blakely, R.D., Gershon, M.D., 1996. Localization and function of a 5-HT transporter in crypt epithelia of the gastrointestinal tract. J. Neurosci. 16, 23522364.

of extracellular uid, which likely masks an even greater impact of the 50 g g1 treatment on circulating cortisol levels. However, it is inconsistent with the decrease in intestinal uid absorption measured in the 50 g g1 -treated sh. Whether this decrease in hematocrit is a reection of altered salt extrusion at the gill or a direct effect on red blood cell proliferation is a question that clearly warrants further investigation. In conclusion, the present study set out to investigate the potential uoxetine sensitivity of two important physiological processes, namely branchial urea excretion and intestinal osmoregulation, as these mechanisms may be particularly susceptible in terms of environmental uoxetine contamination because of their direct exposure to the environment. Our ndings suggest that the both these processes are sensitive to uoxetine treatment and should be further investigated as potential endpoints in terms of chronic exposure to low levels of waterborne uoxetine. In addition, this study also illustrates the widespread impact of uoxetine treatment on physiology, due to the alteration in circulating 5-HT and, in particular, cortisol concentrations, which can invoke secondary responses on nitrogen waste excretion, osmoregulation and potentially many other aspects of physiology. While the doses of uoxetine given to sh of the present study (25 and 50 g g1 ) are difcult to compare directly to environmental doses without measuring tissue levels, it is evident that the doses of uoxetine given to sh of the present study are several orders of magnitude higher than the tissue concentrations of sh that are exposed to uoxetine through the water (<0.008 g g1 ; Brooks et al., 2005). The logical next step, therefore, will be to determine whether chronic exposure to and/or treatment with environmentally realistic concentrations of uoxetine (0.001 g l1 ) have similar physiological effects as those found in the present study.

Acknowledgements This project was funded by an NSF grant (IOS-0455904) to M.D.M. Special thanks to Alexander Gonzalez for his technical assistance. Our sincere gratitude also goes out to Mr. Ray Hurley and Ms. Debbie Fretz for their supply of toadsh.

References
Altamura, A.C., Moro, A.R., Percudani, M., 1994. Clinical pharmacokinetics of uoxetine. Clin. Pharmacokinet. 26, 201214. Brooks, B.W., Chambliss, C.K., Stanley, J.K., Ramirez, A., Banks, K.E., Johnson, R.D., Lewis, R.J., 2005. Determination of select antidepressants in sh from an efuent-dominated stream. Environ. Toxicol. Chem. 24, 464469. Butler, A., Hill, J.M., Ireland, S.J., Jordan, C.C., Tyers, M.B., 1988. Pharmacological properties of GR38032F, a novel antagonist at 5-HT3 receptors. Br. J. Pharmacol. 94, 397412. Butler, A., Elswood, C.J., Burridge, J., Ireland, S.J., Bunce, K.T., Kilpatrick, G.J., Tyers, M.B., 1990. The pharmacological characterization of 5-HT3 receptors in three isolated preparations derived from guinea-pig tissues. Br. J. Pharmacol. 101, 591598. Carrasco, G.A., Van De Kar, L.D., 2003. Neuroendocrine pharmacology of stress. Eur. J. Pharmacol. 463, 235272. Christensen, L.J., Karsgaard, B., Bjerregaard, P., 1999. The effect of 4-nonylphenol on the synthesis of vitellogenin in the ounder Platichthys esus. Aquat. Toxicol. 46, 211219. Evans, D.H., Piermarini, P.M., Choe, K.P., 2005. The multifunctional sh gill: dominant site of gas exchange, osmoregulation, acid-base regulation and excretion of nitrogen waste. Physiol. Rev. 85, 97177. Faraj, B.A., Olkowshi, Z.L., Jackson, R.T., 1994. Expression of a high-afnity serotonin transporter in human lymphocytes. Int. J. Immunopharmacol. 16, 561567. Fent, K., Weston, A.A., Caminada, D., 2006. Ecotoxicology of human pharmaceuticals. Aquat. Toxicol. 76, 122159. Fuller, R.W., Snoddy, H.D., Krushinski, J.H., Robertson, D.W., 1992. Comparison of noruoxetine enantiomers as serotonin uptake inhibitors in vivo. Neurpharmacology 31, 9971000. Gaworecki, K.M., Klaine, S.J., 2008. Behavioral and biochemical responses of hybrid striped bass during and after uoxetine exposure. Aquat. Toxicol. 88, 207213. Gershon, M.D., 2003. Plasticity in serotonin control mechanisms in the gut. Curr. Opin. Pharmacol. 3, 600607.

M.B. Morando et al. / Aquatic Toxicology 95 (2009) 164171 Walsh, P.J., Tucker, B.C., Hopkins, T.E., 1994. Effects of connement/crowding on ureogenesis in the gulf toadsh, Opsanus beta. J. Exp. Biol. 191, 195 206. Walsh, P.J., Heitz, M.J., Campbell, C.E., Cooper, G.J., Medina, M., Wang, Y.S., Goss, G.G., Vincek, V., Wood, C.M., Smith, C.P., 2000. Molecular characterization of a urea transporter in the gill of the gulf toadsh (Opsanus beta). J. Exp. Biol. 203, 23572364. Winberg, S., Nilsson, A., Hylland, P., Soderstom, V., Nilsson, G.E., 1997. Serotonin as a regulator of hypothalamicpituitaryinterrenal activity in teleost sh. Neurosci. Lett. 230, 113116. Wood, C.M., Gilmour, K.M., Perry, S.F., Part, P., Walsh, P.J., 1998. Pulsatile urea excretion in gulf toadsh (Opsanus beta): evidence for activation of

171

a specic facilitated diffusion transport system. J. Exp. Biol. 201, 805 817. Wood, C.M., Warne, J.M., Wang, Y., McDonald, M.D., Balment, R.J., Laurent, P., Walsh, P.J., 2001. Do circulating plasma AVT and/or cortisol levels control pulsatile urea excretion in the gulf toadsh (Opsanus beta)? Comp. Biochem. Physiol. 129A, 859872. Wood, C.M., McDonald, M.D., Sundin, L., Laurent, P., Walsh, P.J., 2003. Pulsatile urea excretion in the gulf toadsh: mechanisms and controls. Comp. Biochem. Physiol. 136B, 667684. Yang, G.B., Qiu, C.L., Aye, P., Shao, Y., Lackner, A.A., 2007. Expression of serotonin transporters by peripheral blood mononuclear cells of rhesus monkeys (Macaca mulatta). Cell. Immunol. 248, 6976.