- Email: [email protected]
Does ketoprofen or diclofenac pose the lowest risk to fish?
Journal of Hazardous Materials 229–230 (2012) 100–106
Contents lists available at SciVerse ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Does ketoprofen or diclofenac pose the lowest risk to fish? Filip Cuklev a , Jerker Fick b , Marija Cvijovic a,c , Erik Kristiansson c , Lars Förlin d , D.G. Joakim Larsson a,∗ a
Institute for Neuroscience and Physiology, the Sahlgrenska Academy, University of Gothenburg, Göteborg, Sweden Department of Chemistry, Umeå University, Umeå, Sweden c Chalmers University of Technology and University of Gothenburg, Göteborg, Sweden d Institute of Biological and Environmental Sciences, University of Gothenburg, Göteborg, Sweden b
h i g h l i g h t s
Ketoprofen bioconcentrates poorly (<0.05) from water to trout blood plasma. No pharmacological responses were identified by microarray or qPCR at 100 g/L. Based on single drug exposure data, ketoprofen is less likely to affect wild fish. Bioconcentration factors for NSAIDs were much higher in effluent-exposed fish. Single drug exposures may underestimate field bioconcentration and thus risks.
a r t i c l e
i n f o
Article history: Received 9 January 2012 Received in revised form 30 April 2012 Accepted 21 May 2012 Available online 30 May 2012 Keywords: Ketoprofen NSAID Fish Bioconcentration Gene expression
a b s t r a c t Ketoprofen and diclofenac are non-steroidal anti-inflammatory drugs (NSAIDs) often used for similar indications, and both are frequently found in surface waters. Diclofenac affects organ histology and gene expression in fish at around 1 g/L. Here, we exposed rainbow trout to ketoprofen (1, 10 and 100 g/L) to investigate if this alternative causes less risk for pharmacological responses in fish. The bioconcentration factor from water to fish blood plasma was <0.05 (4 for diclofenac based on previous studies). Ketoprofen only reached up to 0.6‰ of the human therapeutic plasma concentration, thus the probability of targetrelated effects was estimated to be fairly low. Accordingly, a comprehensive analysis of hepatic gene expression revealed no consistent responses. In some contrast, trout exposed to undiluted, treated sewage effluents bioconcentrated ketoprofen and other NSAIDs much more efficiently, according to a metaanalysis of recent studies. Neither of the setups is however an ideal representation of the field situation. If a controlled exposure system with a single chemical in pure water is a reasonable representation of the environment, then the use of ketoprofen is likely to pose a lower risk for wild fish than diclofenac, but if bioconcentration factors from effluent-exposed fish are applied, the risks may be more similar. © 2012 Elsevier B.V. All rights reserved.
1. Introduction Non-steroidal anti-inflammatory drugs (NSAIDs) are frequently found in surface waters. Several NSAIDs are well known to the gen-
Abbreviations: BCF, bioconcentration factor; HTPC, human therapeutic plasma concentration; NSAID, non-steroidal anti-inflammatory drug; qPCR, quantitative polymerase chain reaction; RTGI, rainbow trout gene index; STP, sewage treatment plant. ∗ Corresponding author at: Department of Physiology/Endocrinology, Institute of Neuroscience and Physiology, The Sahlgrenska Academy at the University of Gothenburg, Box 434, SE-405 30, Göteborg, Sweden. Tel.: +46 31 7863589; fax: +46 31 7863512. E-mail addresses: fi[email protected] (F. Cuklev), jerker.fi[email protected] (J. Fick), [email protected] (M. Cvijovic), [email protected] (E. Kristiansson), [email protected] (L. Förlin), [email protected] (D.G.J. Larsson). 0304-3894/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jhazmat.2012.05.077
eral public as they are frequently used to treat common symptoms, such as fever, pain conditions and inflammation. The mechanism of action is not entirely known, though the primary targets are the prostaglandin G/H synthases 1 and 2 (ptgs1 and ptgs2), also known as cyclooxygenase 1 and 2. Both ptgs1 and ptgs2 are involved in the arachidonic acid pathway, and inhibited activities of the enzymes leads to reduced peripheral prostaglandin synthesis. Prostaglandins are pain-receptor sensitizers, hence the inhibition of their synthesis provides the analgesic effects of NSAIDs. In addition, prostaglandins suppress blood coagulation and regulate vasodilatation and vascular permeability. However, not all NSAIDs are used for the same indications; ibuprofen, for example, is most commonly used for light pain conditions and for its antipyretic effects, whereas diclofenac and ketoprofen are more commonly used to suppress inflammation, stronger pain as well as for rheumatic diseases. Although ketoprofen and diclofenac are used in comparable manners there are still differences in their
F. Cuklev et al. / Journal of Hazardous Materials 229–230 (2012) 100–106
pharmacokinetic and pharmacodynamic properties. Both are considered non-selective blockers of ptgs1 and ptgs2, though diclofenac is slightly more selective for ptgs2, whereas ketoprofen tends to be more selective for ptgs1. Diclofenac has a higher lipophilicity (log P of 4.26) compared to ketoprofen (3.61), which in the ecotoxicological context discussed below would suggest a higher potential for bioconcentration for diclofenac [1–3]. One of the most substantial environmental effects caused by any pharmaceutical is the dramatic decline in vulture populations in India and Pakistan. Diclofenac was given to cattle in order to relieve pain during the last time of their lives. Carcasses of treated livestock were subsequently eaten by Gyps vultures, which rapidly developed visceral gout and renal failure, a known sideeffect of high doses of diclofenac in mammals [4,5]. Subsequently, in 2006 the use of diclofenac for veterinary purposes was banned in India, Nepal and Pakistan [6]. However, the recommended alternative, meloxicam, is expensive and other alternatives, such as ketoprofen, is often used instead [7]. Unfortunately, the use of ketoprofen has recently been reported to cause mortalities in Gyps vultures. Captive and wild-caught vultures were fed ketoprofentreated cattle-tissue containing concentrations similar to the levels found in deceased cattle available to wild vultures [7]. These findings stand in some contrast to previous indications that ketoprofen is rapidly eliminated from livestock tissues and does not cause mortality after exposure to Gyps vultures and other scavenging birds at therapeutic doses [8]. Also, ketoprofen-related mortality has been reported in male eider ducks given the drug [9]. The symptoms were identical to those found in the Gyps vultures exposed to diclofenac, i.e. renal failure and visceral gout. The occurrences of both ketoprofen and diclofenac in effluents from sewage treatment plants (STPs) are typically at similar levels, up to approximately 1 g/L [10–13]. Several studies have reported effects on fish at these concentrations of diclofenac, ranging from cytological and histological effects [14–17] to effects on gene expression [15,18]. Reported bioconcentration factors (BCFs) for diclofenac in rainbow trout varies between studies [10,11,16,18,19]. However, the most recent studies are in relatively close agreement, showing a stable BCF from water to blood plasma of approximately 4 [18,19]. We have recently shown effects on the global hepatic gene expression at exposure concentrations of diclofenac down to 1.6 g/L, even though the corresponding plasma concentration of exposed fish was relatively low (approximately 6 ng/mL) in comparison with the human therapeutic plasma concentrations (HTPC) of >420 ng/mL [20,21]. The number of affected genes as well as their fold change increased with an increasing exposure concentration up to a blood plasma concentration just below the HTPC [18]. Based on its frequent occurrence in the environment together with its risks to affect wildlife, diclofenac was very recently included in the substance priority list within the Water Framework Directive [22]. As ketoprofen and diclofenac often are used for similar indications, we wanted to investigate whether the use of ketoprofen could pose lower risks for fish. However, to the best of our knowledge, studies on the effects of ketoprofen exposure on fish are very scarce. Thibaut et al. [23] have studied the potential of ketoprofen impact on metabolism in carp liver in vitro and there are a few studies on the bioconcentration potential of ketoprofen, though the BCF to fish blood plasma is reported to vary quite much from 0.1 to 48 [10,11]. Here we studied the water to plasma BCF of rainbow trout during controlled flow-through exposure experiments with multiple water concentrations of ketoprofen. Measurements of bioconcentration to blood plasma provide several advantages to whole body bioconcentration (traditional BCF). Importantly, HTPCs are most often available, therefore data on fish plasma levels provide an opportunity for read across between species and thus an estimate of the risk for a pharmacological response in the fish
101
[2,3,24]. Another reason to focus on blood plasma levels is that they conceptually provide a better measure of exposure at the targets compared with whole body levels as some substances are mainly stored in fat and thus not readily available. We have also put our findings in perspective by performing a meta-analysis including previously reported data on BCFs between water and blood plasma for three different NSAIDs. Finally, similar to our previous study on diclofenac [18] we searched for effects on global hepatic gene expression in rainbow trout exposed to ketoprofen as an additional strategy to determine if water concentrations found in the environment are likely to evoke detectable pharmacological responses in exposed fish. 2. Materials and methods 2.1. Fish exposure and sampling Ketoprofen (purity ≥ 98%) (Sigma–Aldrich, Steinheim, Germany) was dissolved in water (500 mg/L concentrated solution), stirred vigorously and diluted stepwise to obtain stock concentrations (0.5 mg/L, 5 mg/L and 50 mg/L). Juvenile rainbow trout (Oncorhynchus mykiss) of both sexes (age: approximately six months, weight: 39.4 ± 6.6 g) were obtained from Vänneåns fiskodling AB, Sweden, and kept in 500 L holding tanks with sand-filtered, aerated fresh water in a flow-through system for one week prior to the experiment. Eight experimental 48 L glass aquaria, semi-covered in black plastic bags to reduce visual stress from the outside (12:12 h light:dark photocycle), were supplied with filtered, aerated freshwater at a flow rate of 0.25 L/min. At the onset of the experiment 80 trout were randomly divided among the aquaria (ten fish in each). Peristaltic pumps (500 L/min) from each stock solution were used to reach target ketoprofen concentrations of 0 (control), 1 (low), 10 (intermediate) and 100 (high) g/L with two aquaria for each concentration. Water samples were taken from each aquarium at day 1, 7 and 14 and stored at −20 ◦ C until analysis. To minimize biological variation due to variable food intake, the fish were not fed during the exposure period. It should be stressed that trout readily cope with food deprivation for much longer periods at these temperatures [25]. To ensure a stable water quality, temperature (12.9 ± 0.1 ◦ C), pH (7.4 ± 0.2) and oxygen saturation (88.3% ± 1.3) were monitored every third day. The total organic carbon content of the supply water was 2.5 mg/L. At the experiment termination on day 14, the fish were killed by a rapid blow to the head and blood was collected from the caudal vessels by the use of heparinized syringes. The plasma was instantly separated by centrifugation at 10,000 rpm for 2 min and snap frozen in liquid nitrogen. The length and weight of the fish were measured and their sex determined by macroscopical observation of their gonads. Liver samples for gene expression analyses were collected and snap frozen in liquid nitrogen. Both liver and plasma samples were stored at −70 ◦ C until analysis. All animal experiments were approved by the local animal committee in Gothenburg (permission number 216-2010). 2.2. Chemical analyses Acetonitrile, methanol and water (Lichrosolv) and sulfuric acid (puriss pa) were obtained from Merck and formic acid (puriss pa) was obtained from Fluka. Labeled naproxen (methyl-13 C, 99%; methyl-D3 , 98%) was used as surrogate standard and was obtained from Cambridge Isotope Laboratories (Andover, MA, USA). Diluted plasma samples (0.5 mL with 1.5 mL 1% aqueous formic acid and 50 ng of surrogate standard) and aqueous samples (100 mL, acidified to pH 3 with sulfuric acid, with 50 ng surrogate standard) were filtered through a 0.45 m MFTM-membrane filter (Millipore,
102
F. Cuklev et al. / Journal of Hazardous Materials 229–230 (2012) 100–106
Sundbyberg, Sweden). Solid phase extraction (SPE) columns (Oasis HLB, 200 mg; Waters Corporation, Milford, MA, USA) were preconditioned and equilibrated with 5.0 mL of methanol and 5.0 mL of de-ionized water. Water with 5% methanol was used to wash the SPE column before eluting with 5 mL of methanol. Eluate was concentrated and reconstituted in water/methanol (80:20) containing 0.1% formic acid to a total volume of 100 L for the fish plasma samples and 1 mL for the water samples. Recoveries of the sample pre-treatments were investigated by spiking 100 ng ketoprofen to tap water and to fish plasma from non-exposed fish (n = 6 for each pre-treatment). Triple stage quadrupole MS/MS TSQ Quantum Ultra EMR (Thermo Fisher Scientific, San Jose, CA, USA) coupled with Accela LC pump (Thermo Fisher Scientific, San Jose, CA, USA) and PAL HTC autosampler (CTC Analytics AG, Zwingen, Switzerland) were used as analytical system for the LC–MS. Aliquots (10 L) were injected onto a Hypersil GOLD aQ column (50 mm × 2.1 mm ID × 5 m particles, Thermo Fisher Scientific, San Jose, CA, USA). Mobile phase consisting of methanol, acetonitrile and water (all solvents buffered by 0.1% formic acid) was used for elution of analytes. The gradient was programmed as follows: 200 L/min 20% methanol in water for 2 min isocratically, then composition is changed to 40% of ACN and 60% of methanol with flow of 300 L/min for 6 min. These parameters were kept for 2 min and then they were switched to starting conditions and let 5 min to equilibrate for next run. Atmospheric pressure photo ionization (APPI) in positive ion mode was used for ionization of ketoprofen and surrogate standard. The setting of key parameters was as follows: sheath gas 50 and auxiliary gas 15 arbitrary units respectively, vaporizer temperature 450 ◦ C, capillary temperature 350 ◦ C, collision gas was argon at 1.5 mL/min. Two SRM transitions were monitored for ketoprofen (255 → 209.1, collision energy 13; 255 → 105.2, collision energy 36) and one for the surrogate standard 13 C, D3 -labeled naproxen (235 → 189, collision energy 13). To reduce matrix effect, resolution at first quadrupole was set to 0.4 FWMH. Quantification based on 5 point calibration curves and calibration standards were run in the beginning and end of each run, with additional standards in the middle of the run (after every 10th sample). Several different blanks are also run at each run, including instrumental blanks to evaluate carry over effects and procedural blanks to evaluate possible laboratory contamination. The maximum differences between results at quantification and qualification masses/mass transitions were set to 30% as criterion for positive identification. For the purpose of simplicity, the term BCF in this paper refers to the relation between blood plasma concentrations in fish and the exposure concentration in water, unless otherwise stated. To assess whether the BCF of ketoprofen depends on water concentration, a statistical analysis was performed using linear regression. Meta-analysis on reported BCFs of ketoprofen was performed by collecting available data from the literature [10,11,19,26]. To make fair comparisons, only BCFs to fish blood plasma were included. The selection of the two other NSAIDs besides from ketoprofen (naproxen and ibuprofen), was based on whether data was available from different types of exposures. Diclofenac was not included here as a similar compilation has been performed in a previous study of ours [18]. 2.3. Microarray design A 15k rainbow trout gene expression microarray was designed for the RT analyzer platform (febit, Heidelberg, Germany) using The Institute for Genomic Research (TIGR) Rainbow Trout Gene Index (RTGI) database version 7.0 (http://compbio.dfci.harvard.edu/tgi/). Details on the probe design strategy, but for eelpout (Zoarces viviparus), and transcript selection strategy are described elsewhere [18,27,28]. When available, transcripts at GenBank
(http://www.ncbi.nlm.nih.gov/nucleotide) were used. Results from similar microarrays, performed by our research group using the same platform, have shown good correlation with qPCR data [18,27–30]. The design, including all probe sequences, and data from the complete microarray experiment is available at GEO database (accession number [GEO:GSE37628]), http://www.ncbi.nlm.nih.gov/geo/, according to the minimum information about a microarray experiment (MIAME) guidelines. 2.4. Microarray synthesis, hybridization and sample preparation The frozen liver tissue was homogenized using Tissuelyser (Qiagen) and total hepatic RNA was isolated using QIAcube and RNeasy® Plus Mini Kit (Qiagen). The RNA quantity was determined with spectophotometric measurements (Nanodrop 1000, NanoDrop Technologies) and the quality was assessed by Experion automated electrophoresis using RNA StdSense chip (Bio-Rad, Sundbyberg, Sweden). Biotinylated aRNA was synthesized using MessageAmpTM II-Biotin Enhanced Single Round aRNA Amplification Kit (Ambion® ). The aRNA samples (20 g) were vacuum dried in a vacuum centrifuge, dissolved in 10 L water and fragmented according to the manufacturer’s protocol. The following steps described in this subchapter were all performed by febit. Oligonucleotide arrays were synthesized by photo-controlled in situ synthesis using the Geniom One system (febit). Each biochip consists of eight individually accessible microchannels, each of which is referred to as a microarray in this manuscript. Four individuals (of both sexes) from each control aquarium and each high exposure concentration aquarium were included in the analysis, i.e. eight fish per exposure concentration. In total, two biochips were analyzed corresponding to 16 microarrays. Pre-hybridization and hybridization were performed based on a customized protocol, described in detail by Cuklev et al. [18]. The samples were randomly allocated on the biochips. Signals were detected using the internal CCD-camera system of the RT analyzer instrument (febit) and quantified using the Geniom Wizard software. Integration times were 266 and 273 ms, determined automatically by the instrument software. 2.5. Microarray analysis All statistical calculations were done in R-2.12.2 (www.rproject.org) [31]. The quality of pre- and post-normalized arrays was verified with Box and MA plots. The data analysis was performed in the R-package LIMMA [32,33]. Data were normalized using the ‘quantile’ method. Moderated t-statistics and adjusted p-values of differential expression were calculated using the empirical Bayes model. As we found no clear differences between the two control aquaria or between the two aquaria with the highest concentration of ketoprofen (100 g/L), all fish from the same treatment were treated as a single group, i.e. we compared one control group to one group of fish exposed to 100 g/L of ketoprofen. 2.6. Quantitative RT-PCR To verify the microarray results, gene expression analyses with quantitative real-time polymerase chain reaction (qPCR) analyses were performed with mRNA from fish not included on the array, four individuals from each control aquarium and four from each high exposure concentration aquarium. The idea of analyzing different fish than those analyzed on the array is that it reduces the risk for incorrectly identifying genes as differentially expressed between treatment groups, a major challenge when thousands of endpoints are analyzed in parallel. The samples of total RNA were reverse transcribed to cDNA with a mixture of oligo(dT)
F. Cuklev et al. / Journal of Hazardous Materials 229–230 (2012) 100–106
103
Table 1 Primer sequences used to validate the microarray results by quantitative PCR. Annotation fancl mid1 fth ptgs1 ptgs2 cox1 cox2 ubq actb a b
Accession number a
TC164519 TC156087a CB486370a NM 001124361.1b NM 001124348.1b NA NA NA NA
Forward sequence 5 –3
Reverse sequence 5 –3
agcatcttgcaaaacctccg aggccaagaaccagacctacct cctgcacaagattgcctctga aaggttaccaacatgtcgcgg tggcgtggacttaaaccatgtt caaacgagaggtagcatcaatc ccaggacgattaaaccaaacag acaacatccagaaagagtccac tctcagtctcattggcatggc
aggcgtaacagatcccacactc catgacagagaaaagccccct gtctcccagcttcttaatggcc aaacctgacacacacaccccag tagacctcgccgttcaaaacc gttcttcaaatgtgtggtaggg accgcttcaacaacgatagg aggcgagcgtagcacttg gctgtttcaccgttccagttgt
Transcript from the Institute for Genomic Research (TIGR) Rainbow Trout Gene Index (RTGI) database version 7.0 (http://compbio.dfci.harvard.edu/tgi/). Transcript from GenBank.
and random hexamers, using iScriptTM cDNA Synthesis Kit (Biorad) according to the manufacturer’s instructions. In addition, no reverse transcriptase (NoRT) controls were synthesized for each sample. The following genes were analyzed: prostaglandin G/H synthase 1 (ptgs1), prostaglandin G/H synthase 2 (ptgs2), Fanconi anemia complementation group L (fancl), ferritin H-3 (fth) and MID1 interacting protein 1 (mid1). Also included were cytochrome c oxidase I (cox1) and cytochrome c oxidase II (cox2) from the study by Mehinto et al. [15], which were not present on the array. Ubiquitin (ubq) and -actin (actb) were chosen as potential housekeeping genes. Primers were designed using Primer Express software (Applied Biosystems, Warrington, UK; not cox1 and cox2 for which the primer sequences given by Mehinto et al. [15] were used) and purchased from Eurofins MWG Operon (Ebersberg, Germany) (Table 1). For each primer pair, qPCR runs with a standard curve with six cDNA dilutions as starting material were performed to ensure optimized assays. Each qPCR reaction was carried out in triplicates and contained 1× Power SYBR® Green mastermix (Applied Biosystems), 100 nM of each primer pair and 12.5 ng cDNA in a final reaction volume of 6 L. The qPCR analysis was performed on a ABI 7900HT with 10 min initial denaturation at 95 ◦ C, followed by 40 cycles of 95 ◦ C for 15 s and 60 ◦ C for 1 min. After each run a melting curve analysis was performed to verify the specificity of the amplification (95 ◦ C for 15 s, 60 ◦ C for 15 s and 95 ◦ C for 15 s). To ensure that no genomic DNA, primer-dimers etc. could inflict the results, NoRT controls were run for each sample. None of the NoRT controls had a threshold cycle (Ct ) value below 30 and all had Ct values more than 5 cycles after the actual sample. However, due to technical errors, one sample from control group A was excluded from ptgs1, fth and mid1 analyses; one sample from control group B was excluded from the runs with mhc1, cox1, cox2, and fancl primers. The expression of both housekeeping genes was stable and showed no tendencies to be differentially regulated between the groups and were thus used for normalization by subtracting the average Ct value of both genes for each sample. The initial microarray analysis showed no consistent differences between fish from the duplicate aquaria for each exposure, hence the fish from replicate tanks were considered as one group for the following qPCR analyses. The resulting Ct values were used for the statistical analyses where tests for differential expression between control and exposed fish were performed using a single-sided Student’s t-test assuming equal variances. Furthermore, the correlation between blood plasma concentration of ketoprofen and gene expression on individual levels were analyzed for these seven genes using linear regression and Bonferroni correction.
and R2 above 0.99 for all standard curves. No ketoprofen was detected in the instrumental or procedural blanks. Absolute recoveries of the solid phase extraction of the fish plasma and water samples were 127% (17% RSD, n = 6, spiking level 100 ng) and 103% (8% RSD, n = 6, spiking level 100 ng) respectively. The following results are all displayed in the order of low, intermediate and high exposure concentration, respectively. The average water concentrations of ketoprofen throughout the exposure were 1.2 ± 0.21, 7.2 ± 2.59 and 90.5 ± 32.98 g/L. The bioconcentration potential of ketoprofen was in this study low compared to e.g. diclofenac, hence the blood plasma concentrations, 0.05 ± 0.02, 0.19 ± 0.08 and 0.62 ± 0.49 ng/mL, were all well below human therapeutic levels (HTPC > 1000 ng/mL) [34]. No residues of ketoprofen could be found in the controls (detection limit = 0.005 ng/mL). There was a significant decrease (p < 0.0001) in the BCF with increasing water concentration, from 0.042 down to 0.007 (Fig. 1).
3.2. Meta-analysis The compilation of reported BCFs for ketoprofen, naproxen and ibuprofen revealed that BCFs measured in fish exposed to undiluted effluents tend to be higher than BCFs measured in fish exposed to a single substance (or a mixture of very few substances) in pure water (Fig. 2). For ketoprofen the BCF for effluent-exposed fish was 35–6857 times higher, for ibuprofen 2–13333 times higher and for naproxen 5.5–40 times higher than the reported BCFs in studies where fish were exposed to water with a lower complexity. To the best of our knowledge, no BCF data from wild fish blood plasma is available for any of the three selected substances.
3. Results 3.1. Bioconcentration Ketoprofen concentration was successfully measured in water and blood plasma samples, with stable and reproducible results
Fig. 1. Bioconcentration factor of ketoprofen in blood plasma of rainbow trout exposed to waterborne ketoprofen for 14 days at different concentrations. The bioconcentration factor decreased significantly (p < 0.0001) with an increasing water concentration of ketoprofen. Statistical analysis was performed using linear regression.
104
F. Cuklev et al. / Journal of Hazardous Materials 229–230 (2012) 100–106
3.3. Gene expression
Fig. 2. A comparison of bioconcentration factors of three NSAIDs to blood plasma in fish exposed to single substances or a mixture of a few substances in pure water under controlled lab conditions (closed symbols) versus fish exposed to undiluted sewage effluents (open symbols). Data were collected from the present study and other studies [10,11,19,26]. Only nominal concentrations were used for the lab exposure in the study by Brown et al. [10].
Due to the low bioconcentration potential of ketoprofen and thereby low plasma concentrations in all exposure groups, the gene expression pattern was only analyzed in the fish exposed to the highest concentration. The microarray analysis revealed 59 genes with p < 0.05, however the adjusted p-value was >0.99 for all genes (Supplementary Table A.1). To confirm these results, qPCR was performed on different individuals from the ones present on the arrays. A repertoire of genes were selected based on different sources: fancl and mid1 were top-ranked genes based on p-value in the microarray analysis; fth was present among the top-ranked genes in the study by Cuklev et al. [18] as well as in the upper layer in the present study; ptgs1 and ptgs2 are the primary targets of ketoprofen and other NSAIDs; cox1 and cox2 were analyzed by Mehinto et al. [15]. There were no significant differences between treatment groups for any of the genes analyzed (Fig. 3). The linear regression analysis of the correlation between blood plasma concentration of ketoprofen and gene expression on individual levels did not reveal any significance for the genes analyzed by qPCR (p > 0.48) except for fth (p = 0.0497 without Bonferroni correction). However, when adjusting the p-value for multiple testing (n = 8) fth was not significantly regulated (Supplementary Fig. B.1). Neither the microarray nor the qPCR analysis resulted in any significant differences in responses between males and females.
4. Discussion
Fig. 3. Hepatic gene expression of selected genes in rainbow trout after ketoprofen exposure (90.5 g/L) analyzed by qPCR (white bars) or microarray (black bars). Values are expressed as log fold change (log2 ) compared with control fish. Statistical analyses were performed on the qPCR data using a Student’s t-test assuming equal variances (single-sided, based on the direction of regulation on the microarray) followed by Bonferroni correction. None of the genes were significantly regulated.
In the present study we show that waterborne ketoprofen bioconcentrates considerably less in fish than diclofenac under controlled laboratory conditions. Measured plasma levels of ketoprofen in fish at an exposure concentration about 100 times higher than levels found in undiluted sewage effluents reached less than 1‰ of human therapeutic plasma levels. In accordance no effects on the global hepatic gene expression were found, in sharp contrast to previous reports for diclofenac [15,18]. These results support our hypothesis that the use of ketoprofen rather than diclofenac may pose lower risks for exposed wild fish. Results from other exposure experiments with complex, undiluted treated sewage effluent, however, complicate the interpretation somewhat. During exposure to undiluted effluent, ketoprofen as well as ibuprofen and naproxen appear to bioconcentrate considerably more than in controlled lab experiments with drugs diluted in pure water, whereas for diclofenac the difference is not as clear. It is therefore not entirely straightforward how to extrapolate the results from the present study (and other lab studies on NSAIDs) to the situation for wild fish. Ketoprofen and diclofenac are often used for similar indications, though diclofenac is often used to a higher extent. Over the last four years, the number of defined daily doses (DDD) of diclofenac sold in Sweden has been approximately four times higher than for ketoprofen (43.4 M and 11.1 M DDD in 2010 for diclofenac and ketoprofen respectively) [35]. Both compounds are mainly excreted as metabolites (>99%; http://www.fass.se) of which some are glucuronide conjugates that are likely to be cleaved during sewage treatment, thus regenerating the parent compound [12,36]. Both are found at approximately the same concentration in STP influents (typically just below 1 g/L) [37–39]. Removal efficiency of both compounds is low to moderate (5–70%), and they are often found in similar concentrations in effluents, with an average from four studies of 1.4 g/L and 0.83 g/L for ketoprofen and diclofenac respectively [11,12,38,39]. Fernandez et al. [40] reported slightly higher concentrations of ketoprofen than diclofenac in a Spanish river (0.3–991 ng/L and 0.7–156 ng/L respectively), though in German rivers the maximum concentration of 2.1 g/L diclofenac
F. Cuklev et al. / Journal of Hazardous Materials 229–230 (2012) 100–106
was higher than ketoprofen levels of 0.38 g/L [12]. Given similar water levels, their bioconcentration potential is therefore crucial for assessing differences in their potential to affect aquatic organisms. Here, rainbow trout was exposed to waterborne ketoprofen at three different concentrations, where the lowest one (measured concentration of 1.2 g/L) corresponds to concentrations found in undiluted STP effluents. The bioconcentration potential of ketoprofen was relatively low (BCF < 0.05) compared with approximately 4 for diclofenac based on a very similar exposure setup [18]. The slightly lower lipophilicity of ketoprofen indeed suggests a lower bioconcentration potential of ketoprofen, though not quite to the extent observed here [1–3]. The BCF decreased significantly from 0.042 down to 0.007 with increasing water concentration (1.2–90.5 g/L). A variable BCF for diclofenac from water to fish liver was reported by Schwaiger et al. [16] whereas a stable BCF to both liver and plasma for diclofenac was reported by us [18]. In the study by Schwaiger et al. [16], the BCF dropped from >2000 at the lowest exposure concentration down to approximately 10 at a 500 times higher exposure concentration i.e. a 200-fold BCF drop. Here, the drop was 6-fold over a 100-fold exposure concentration. It should also be noted that, in contrast to other studies with diclofenac [15,18,19] as well as for the present study, Schwaiger et al. [16] used dimethyl sulfoxide (DMSO; 0.12‰) as a solvent. Most predictive models for bioconcentration assume a stable BCF over exposure concentrations [41]. Accordingly, decreasing BCFs with increasing water concentrations are not commonly reported, though the phenomenon has been observed for other chemicals in other species (Oryzias latipes, Perna viridis and Dreissena polymorpha) [41–43]. Such patterns could for example be explained by an increased activity of detoxification systems at higher exposures, if there is a saturation of binding sites for the bioconcentrating chemical in the organism or if the energy required to bind the chemical is not sufficient at higher concentrations [41]. However, the latter explanations seem less likely as considerably higher plasma concentrations of ketoprofen in trout have been reported than those detected here [10,11]. Previous studies have been somewhat contradictive regarding the BCF of ketoprofen, though the experimental conditions differed. The measured BCF in the present study differ considerably from studies where rainbow trout were exposed to undiluted sewage effluents (BCF = 3.5–48) [11], but is more similar to the BCF of 0.1 reported in the study by Brown et al. [10] where rainbow trout were exposed to nominal concentrations of ketoprofen together with four other pharmaceuticals in pure water under controlled lab conditions. Indeed, lower BCFs in exposure situations with single drug(s) in pure water compared with exposure to undiluted effluents is a collective trend for ketoprofen, ibuprofen and naproxen, and also for diclofenac but to a lesser extent [18]. It is therefore possible that the concentration of other constituents of treated sewage effluents can influence the uptake, distribution, metabolism or excretion of NSAIDs and possibly other pharmaceuticals, and thus the expected effect/toxicity, to a considerable extent. This possibility is not taken into account in current environmental risk assessment procedures, such as those for medicinal products in the EU [44] and the USA [45]. It should be stressed that neither the exposure to single drugs in pure water, nor exposure to 100% undiluted effluents represent the regular situation for fish in the wild, but we still do not know which experimental setup provides the best approximation. At the highest tested exposure concentration in the present study (90.5 g/L), plasma concentrations corresponding to 0.62‰ of the HTPC were achieved. Thus, one would presuppose minor, if any, changes in the gene expression pattern compared to non-exposed fish [46]. The microarray analysis in the present study revealed 59 genes with p < 0.05. However, given the large
105
number of probes present on the array, it is likely that some genes are falsely identified as regulated. Accordingly, our assessment shows an adjusted p-value > 0.99 for all genes analyzed on the array suggesting that all of them could very well be false positives. To verify these results, seven genes were analyzed using qPCR on other individuals than on the array, i.e. verification by biological replication. Firstly, the two top-ranked genes fancl [RTGI: TC164519] and mid1 [RTGI: TC156087] both had a p-value <0.03 (not adjusted for multiple testing) on the array, though neither was significantly regulated according to the qPCR. The list of the 1000 most likely regulated genes on the array was compared with the ranking lists at different exposure concentrations of diclofenac from the study by Cuklev et al. [18] and fth ([RTGI: CB486370] and [RTGI: TC171322] in the present study and in Cuklev et al., respectively) was one of the genes present on all lists. Ferritin is a ubiquitous intracellular protein that stores iron [47] and fth was down regulated in both studies, though only significantly on the diclofenac array. None of the main target genes, ptgs1 [GenBank: NM 001124361.1] and ptgs2 [GenBank: NM 001124348.1], were significantly regulated on neither the array nor the qPCR. If the highest reported BCF for ketoprofen [11] is applied to an effluent dominated stream with a surface water concentration of 1 g/L of ketoprofen, the predicted blood plasma concentration in exposed fish would be 48 ng/mL, i.e. about 100 times higher than the levels found here in fish exposed to 90.5 g/L. This is still only about 5% of the HTPC, however plasma concentrations of diclofenac, considerably lower than 5% of the corresponding HTPC, have been reported to affect gene expression in fish [15,18]. Thus, we cannot yet rule out a pharmacological response at water concentrations of ketoprofen similar to those concentrations found in STP effluents. If we in a similar way apply the highest reported BCF from effluent to fish blood plasma for diclofenac, i.e. 29 [11] and a surface water concentration of 1 g/L, that would correspond to a blood plasma concentration of about 7% of the HTPC, thus quite similar to the predicted ratio for ketoprofen. Also, several NSAIDs are regularly found together in surface water, and as they are acting via the same or similar modes of action their effects are expected to be additive. Ibuprofen, for example, have quite recently been shown to cause a decrease in gill tissue prostaglandin E2 levels, but only significantly at relatively high exposure concentrations (≥50 g/L), most likely a consequence of the quite poor bioconcentration potential of ibuprofen (BCF = 1.3) [48]. Whether the gene expression is affected at lower exposure concentrations is not known. Given these premises, it is not entirely clear what alternative pose the lowest risk for wild fish. However, if controlled exposure conditions with a single chemical in pure water, as applied here, reflect the environment, ketoprofen would pose lower risk than diclofenac to fish. Acknowledgements The authors thank Lina Gunnarsson, Bethanie Carney-Almroth and Hannah Svedlund for help with experiments and analyses. This research was supported by the Swedish Foundation for Strategic Environmental Research (MISTRA), the Swedish Research Council (VR) and the Adlerbertska Research Foundation. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jhazmat. 2012.05.077. References [1] P.N. Fitzsimmons, J.D. Fernandez, A.D. Hoffman, B.C. Butterworth, J.W. Nichols, Branchial elimination of superhydrophobic organic compounds by rainbow trout (Oncorhynchus mykiss), Aquat. Toxicol. 55 (2001) 23–34.
106
F. Cuklev et al. / Journal of Hazardous Materials 229–230 (2012) 100–106
[2] D.B. Huggett, J.C. Cook, J.F. Ericson, R.T. Williams, A theoretical model for utilizing mammalian pharmacology and safety data to prioritize potential impacts of human pharmaceuticals to fish, Hum. Ecol. Risk. Assess. 9 (2003) 1789–1799. [3] R. Schreiber, U. Gündel, S. Franz, A. Küster, B. Rechenberg, R. Altenburger, Using the fish plasma model for comparative hazard identification for pharmaceuticals in the environment by extrapolation from human therapeutic data, Regul. Toxicol. Pharmacol. 61 (2011) 261–275. [4] J.L. Oaks, M. Gilbert, M.Z. Virani, R.T. Watson, C.U. Meteyer, B.A. Rideout, H.L. Shivaprasad, S. Ahmed, M.J. Iqbal Chaudhry, M. Arshad, S. Mahmood, A. Ali, A. Ahmed Khan, Diclofenac residues as the cause of vulture population decline in Pakistan, Nature 427 (2004) 630–633. [5] S. Shultz, H.S. Baral, S. Charman, A.A. Cunningham, D. Das, G.R. Ghalsasi, M.S. Goudar, R.E. Green, A. Jones, P. Nighot, D.J. Pain, V. Prakash, Diclofenac poisoning is widespread in declining vulture populations across the Indian subcontinent, Proc. Biol. Sci. 271 (Suppl. 5) (2004) S458–S460. [6] D.J. Pain, C.G.R. Bowden, A.A. Cunningham, R. Cuthbert, D. Das, M. Gilbert, R.D. Jakati, Y. Jhala, A.A. Khan, V. Naidoo, J.L. Oaks, J. Parry-Jones, V. Prakash, A. Rahmani, S.P. Ranade, H.S. Baral, K.R. Senacha, S. Saravanan, N. Shah, G. Swan, D. Swarup, M.A. Taggart, R.T. Watson, M.Z. Virani, K. Wolter, R.E. Green, The race to prevent the extinction of South Asian vulture, Bird Conserv. Int. 18 (2008) S30–S48. [7] V. Naidoo, K. Wolter, D. Cromarty, M. Diekmann, N. Duncan, A.A. Meharg, M.A. Taggart, L. Venter, R. Cuthbert, Toxicity of non-steroidal anti-inflammatory drugs to Gyps vultures: a new threat from ketoprofen, Biol. Lett. 6 (2010) 339–341. [8] R. Cuthbert, J. Parry-Jones, R.E. Green, D.J. Pain, NSAIDs and scavenging birds: potential impacts beyond Asia’s critically endangered vultures, Biol. Lett. 3 (2007) 91–94. [9] D.M. Mulcahy, P. Tuomi, R.S. Larsen, Differential mortality of male spectacled eiders (Somateria fischeri) and king eiders (Somateria spectabilis) subsequent to anesthesia with propofol, bupivacaine, and ketoprofen, J. Avian Med. Surg. 1 (7) (2003) 117–123. [10] J.N. Brown, N. Paxéus, L. Förlin, D.G.J. Larsson, Variations in bioconcentration of human pharmaceuticals from sewage effluents into fish blood plasma, Environ. Toxicol. Pharmacol. 24 (2007) 267–274. [11] J. Fick, R.H. Lindberg, J. Parkkonen, B. Arvidsson, M. Tysklind, D.G.J. Larsson, Therapeutic levels of levonorgestrel detected in blood plasma of fish: results from screening rainbow trout exposed to treated sewage effluents, Environ. Sci. Technol. 44 (2010) 2661–2666. [12] T.A. Ternes, Occurrence of drugs in German sewage treatment plants and rivers, Water Res. 32 (1998) 3245–3260. [13] C. Tixier, H.P. Singer, S. Oellers, S.R. Müller, Occurrence, Fate of carbamazepine, clofibric acid, diclofenac, ibuprofen, ketoprofen, and naproxen in surface waters, Environ. Sci. Technol. 3 (7) (2003) 1061–1068. [14] B. Hoeger, B. Kollner, D.R. Dietrich, B. Hitzfeld, Water-borne diclofenac affects kidney and gill integrity and selected immune parameters in brown trout (Salmo trutta f. fario), Aquat. Toxicol. 75 (2005) 53–64. [15] A.C. Mehinto, E.M. Hill, C.R. Tyler, Uptake and biological effects of environmentally relevant concentrations of the nonsteroidal anti-inflammatory pharmaceutical diclofenac in rainbow trout (Oncorhynchus mykiss), Environ. Sci. Technol. 44 (2010) 2176–2182. [16] J. Schwaiger, H. Ferling, U. Mallow, H. Wintermayr, R.D. Negele, Toxic effects of the non-steroidal anti-inflammatory drug diclofenac. Part I: Histopathological alterations and bioaccumulation in rainbow trout, Aquat. Toxicol. 68 (2004) 141–150. [17] R. Triebskorn, H. Casper, A. Heyd, R. Eikemper, H.R. Kohler, J. Schwaiger, Toxic effects of the non-steroidal anti-inflammatory drug diclofenac. Part II: Cytological effects in liver, kidney, gills and intestine of rainbow trout (Oncorhynchus mykiss), Aquat. Toxicol. 68 (2004) 151–166. [18] F. Cuklev, E. Kristiansson, J. Fick, N. Asker, L. Förlin, D.G.J. Larsson, Diclofenac in fish: blood plasma levels similar to human therapeutic levels affect global hepatic gene expression, Environ. Toxicol. Chem. 30 (2011) 2126–2134. [19] M. Lahti, J.-M. Brozinski, A. Jylhä, L. Kronberg, A. Oikari, Uptake from water, biotransformation, and biliary excretion of pharmaceuticals by rainbow trout, Environ. Toxicol. Chem. 3 (0) (2011) 1403–1411. [20] J.G. Hardman, L.E. Limbard, A.G. Gilman, The Pharmacological Basis of Therapeutics, McGraw-Hill, NY, USA, 2001. [21] T. Juhlin, S. Björkman, B. Gunnarsson, Å. Fyge, B. Roth, P. Höglund, Acute administration of diclofenac but possibly not long term low dose aspirin, causes detrimental renal effects in heart failure patients treated with ACE-inhibitors, Eur. J. Heart. Fail. 6 (2004) 909–916. [22] European Union, Proposal for a Directive of the European Parliament and of the Council Amending Directives 2000/60/EC and 2008/105/EC as Regards Priority Substances in the Field of Water Policy, Water framework directive, Brussels, Belgium, 2012. [23] R. Thibaut, S. Schnell, C. Porte, The interference of pharmaceuticals with endogenous and xenobiotic metabolizing enzymes in carp liver: an in vitro study, Environ. Sci. Technol. 40 (2006) 5154–5160.
[24] V. Roos, L. Gunnarsson, J. Fick, D.G.J. Larsson, C. Rudén, Prioritising pharmaceuticals for environmental risk assessment: towards adequate and feasible first-tier selection, Sci. Total. Environ. 421–422 (2012) 102–110. [25] A. Kullgren, L.M. Samuelsson, D.G.J. Larsson, B.T. Björnsson, E.J. Bergman, A metabolomics approach to elucidate effects of food deprivation in juvenile rainbow trout (Oncorhynchus mykiss), Am. J. Physiol: Regul. Integr. Comp. Physiol. 299 (2010) R1440–R1448. [26] G.C. Nallani, P.M. Paulos, L.A. Constantine, B.J. Venables, D.B. Huggett, Bioconcentration of ibuprofen in fathead minnow (Pimephales promelas) and channel catfish (Ictalurus punctatus), Chemosphere 84 (2011) 1371–1377. [27] E. Kristiansson, N. Asker, L. Forlin, D.G.J. Larsson, Characterization of the Zoarces viviparus liver transcriptome using massively parallel pyrosequencing, BMC Genomics 10 (2009) 345. [28] F. Cuklev, L. Gunnarsson, M. Cvijovic, E. Kristiansson, C. Rutgersson, B. Björlenius, D.G.J. Larsson, Global hepatic gene expression in rainbow trout exposed to sewage effluents: a comparison of different sewage treatment technologies, Sci. Total. Environ. 427–428 (2012) 106–114. [29] L. Gunnarsson, E. Kristiansson, C. Rutgersson, J. Sturve, J. Fick, L. Förlin, D.G.J. Larsson, Pharmaceutical industry effluent diluted 1:500 affects global gene expression, cytochrome P4501A activity and plasma phosphate in fish, Environ. Toxicol. Chem. 28 (2009) 2639–2647. [30] A. Lennquist, N. Asker, E. Kristiansson, A. Brenthel, B.T. Björnsson, P. Kling, M. Hultman, D.G.J. Larsson, L. Förlin, Physiology and mRNA expression in rainbow trout (Oncorhynchus mykiss) after long-term exposure to the new antifoulant medetomidine, Comp. Biochem. Physiol. C: Pharmacol. Toxicol. 154 (2011) 234–241. [31] R Development Core Team, R: A Language and Environment for Statistical Computing, R Foundation for Statistical Computing, Vienna, Austria, 2010. [32] G. Smyth, Limma, Linear models for microarray data, in: R. Gentleman, V.J. Carey, W. Huber, R.A. Irizarry, S. Dudoit (Eds.), Bioinformatics and Computational Biology Solutions Using R and Bioconductor, Springer, New York, 2005, pp. 397–420. [33] G.K. Smyth, Linear models and empirical bayes methods for assessing differential expression in microarray experiments, Stat. Appl. Genet. Mol. Biol. 3 (2004) Article3. [34] M. Schulz, A. Schmoldt, Therapeutic and toxic blood concentrations of more than 800 drugs and other xenobiotics, Pharmazie 58 (2003) 447–474. Apoteket, Selling Statistics of Human Pharmaceuticals, [35] A.B. http://www.apotekensservice.se/Om-Apotekens-Service/ 2010, Infrastrukturtjanster/Statistik/. [36] N. Lindqvist, T. Tuhkanen, L. Kronberg, Occurrence of acidic pharmaceuticals in raw and treated sewages and in receiving waters, Water Res. 39 (2005) 2219–2228. [37] C.G. Daughton, T.A. Ternes, Pharmaceuticals and personal care products in the environment: agents of subtle change? Environ. Health Perspect. 107 (1999) 907–938. [38] H. Matsuo, H. Sakamoto, K. Arizono, R. Shinohara, Behavior of pharmaceuticals in waste water treatment plant in Japan, Bull. Environ. Contam. Toxicol. 87 (2011) 31–35. [39] R. Rosal, A. Rodríguez, J.A. Perdigón-Melón, A. Petre, E. García-Calvo, M.J. Gómez, A. Agüera, A.R. Fernández-Alba, Occurrence of emerging pollutants in urban wastewater and their removal through biological treatment followed by ozonation, Water Res. 44 (2010) 578–588. [40] C. Fernández, M. González-Doncel, J. Pro, G. Carbonell, J.V. Tarazona, Occurrence of pharmaceutically active compounds in surface waters of the HenaresJarama-Tajo River system (Madrid, Spain) and a potential risk characterization, Sci. Total. Environ. 408 (2010) 543–551. [41] C. Liu, K.Y.H. Gin, V.W.C. Chang, B.P.L. Goh, M. Reinhard, Novel perspectives on the bioaccumulation of PFCs – the concentration dependency, Environ. Sci. Technol. 45 (2011) 9758–9764. [42] V. Contardo-Jara, C. Lorenz, S. Pflugmacher, G. Nützmann, W. Kloas, C. Wiegand, Molecular effects and bioaccumulation of levonorgestrel in the non-target organism Dreissena polymorpha, Environ. Pollut. 159 (2011) 38–44. [43] T. Kondo, H. Yamamoto, N. Tatarazako, K. Kawabe, M. Koshio, N. Hirai, M. Morita, Bioconcentration factor of relatively low concentrations of chlorophenols in Japanese medaka, Chemosphere 61 (2005) 1299–1304. [44] Guideline on the Environmental Risk Assessment of Medicinal Products for Human Use, 2006, http://www.ema.europa.eu/docs/en GB/document library/ Scientific guideline/2009/10/WC500003978.pdf. [45] U.S. EPA, Guidelines for Ecological Risk Assessment, 1998, http://www.epa. gov/raf/publications/guidelines-ecological-risk-assessment.htm. [46] J.P. Berninger, B.W. Brooks, Leveraging mammalian pharmaceutical toxicology and pharmacology data to predict chronic fish responses to pharmaceuticals, Toxicol. Lett. 193 (2010) 69–78. [47] E.C. Theil, Ferritin: structure gene regulation, and cellular function in animals, plants, and microorganisms, Ann. Rev. Biochem. 5 (6) (1987) 289–315. [48] K. Bhandari, B. Venables, Ibuprofen bioconcentration and prostaglandin E2 levels in the bluntnose minnow Pimephales notatus, Comp. Biochem. Physiol. C: Toxicol. Pharmacol. 153 (2011) 251–257.
Contents lists available at SciVerse ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Does ketoprofen or diclofenac pose the lowest risk to fish? Filip Cuklev a , Jerker Fick b , Marija Cvijovic a,c , Erik Kristiansson c , Lars Förlin d , D.G. Joakim Larsson a,∗ a
Institute for Neuroscience and Physiology, the Sahlgrenska Academy, University of Gothenburg, Göteborg, Sweden Department of Chemistry, Umeå University, Umeå, Sweden c Chalmers University of Technology and University of Gothenburg, Göteborg, Sweden d Institute of Biological and Environmental Sciences, University of Gothenburg, Göteborg, Sweden b
h i g h l i g h t s
Ketoprofen bioconcentrates poorly (<0.05) from water to trout blood plasma. No pharmacological responses were identified by microarray or qPCR at 100 g/L. Based on single drug exposure data, ketoprofen is less likely to affect wild fish. Bioconcentration factors for NSAIDs were much higher in effluent-exposed fish. Single drug exposures may underestimate field bioconcentration and thus risks.
a r t i c l e
i n f o
Article history: Received 9 January 2012 Received in revised form 30 April 2012 Accepted 21 May 2012 Available online 30 May 2012 Keywords: Ketoprofen NSAID Fish Bioconcentration Gene expression
a b s t r a c t Ketoprofen and diclofenac are non-steroidal anti-inflammatory drugs (NSAIDs) often used for similar indications, and both are frequently found in surface waters. Diclofenac affects organ histology and gene expression in fish at around 1 g/L. Here, we exposed rainbow trout to ketoprofen (1, 10 and 100 g/L) to investigate if this alternative causes less risk for pharmacological responses in fish. The bioconcentration factor from water to fish blood plasma was <0.05 (4 for diclofenac based on previous studies). Ketoprofen only reached up to 0.6‰ of the human therapeutic plasma concentration, thus the probability of targetrelated effects was estimated to be fairly low. Accordingly, a comprehensive analysis of hepatic gene expression revealed no consistent responses. In some contrast, trout exposed to undiluted, treated sewage effluents bioconcentrated ketoprofen and other NSAIDs much more efficiently, according to a metaanalysis of recent studies. Neither of the setups is however an ideal representation of the field situation. If a controlled exposure system with a single chemical in pure water is a reasonable representation of the environment, then the use of ketoprofen is likely to pose a lower risk for wild fish than diclofenac, but if bioconcentration factors from effluent-exposed fish are applied, the risks may be more similar. © 2012 Elsevier B.V. All rights reserved.
1. Introduction Non-steroidal anti-inflammatory drugs (NSAIDs) are frequently found in surface waters. Several NSAIDs are well known to the gen-
Abbreviations: BCF, bioconcentration factor; HTPC, human therapeutic plasma concentration; NSAID, non-steroidal anti-inflammatory drug; qPCR, quantitative polymerase chain reaction; RTGI, rainbow trout gene index; STP, sewage treatment plant. ∗ Corresponding author at: Department of Physiology/Endocrinology, Institute of Neuroscience and Physiology, The Sahlgrenska Academy at the University of Gothenburg, Box 434, SE-405 30, Göteborg, Sweden. Tel.: +46 31 7863589; fax: +46 31 7863512. E-mail addresses: fi[email protected] (F. Cuklev), jerker.fi[email protected] (J. Fick), [email protected] (M. Cvijovic), [email protected] (E. Kristiansson), [email protected] (L. Förlin), [email protected] (D.G.J. Larsson). 0304-3894/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jhazmat.2012.05.077
eral public as they are frequently used to treat common symptoms, such as fever, pain conditions and inflammation. The mechanism of action is not entirely known, though the primary targets are the prostaglandin G/H synthases 1 and 2 (ptgs1 and ptgs2), also known as cyclooxygenase 1 and 2. Both ptgs1 and ptgs2 are involved in the arachidonic acid pathway, and inhibited activities of the enzymes leads to reduced peripheral prostaglandin synthesis. Prostaglandins are pain-receptor sensitizers, hence the inhibition of their synthesis provides the analgesic effects of NSAIDs. In addition, prostaglandins suppress blood coagulation and regulate vasodilatation and vascular permeability. However, not all NSAIDs are used for the same indications; ibuprofen, for example, is most commonly used for light pain conditions and for its antipyretic effects, whereas diclofenac and ketoprofen are more commonly used to suppress inflammation, stronger pain as well as for rheumatic diseases. Although ketoprofen and diclofenac are used in comparable manners there are still differences in their
F. Cuklev et al. / Journal of Hazardous Materials 229–230 (2012) 100–106
pharmacokinetic and pharmacodynamic properties. Both are considered non-selective blockers of ptgs1 and ptgs2, though diclofenac is slightly more selective for ptgs2, whereas ketoprofen tends to be more selective for ptgs1. Diclofenac has a higher lipophilicity (log P of 4.26) compared to ketoprofen (3.61), which in the ecotoxicological context discussed below would suggest a higher potential for bioconcentration for diclofenac [1–3]. One of the most substantial environmental effects caused by any pharmaceutical is the dramatic decline in vulture populations in India and Pakistan. Diclofenac was given to cattle in order to relieve pain during the last time of their lives. Carcasses of treated livestock were subsequently eaten by Gyps vultures, which rapidly developed visceral gout and renal failure, a known sideeffect of high doses of diclofenac in mammals [4,5]. Subsequently, in 2006 the use of diclofenac for veterinary purposes was banned in India, Nepal and Pakistan [6]. However, the recommended alternative, meloxicam, is expensive and other alternatives, such as ketoprofen, is often used instead [7]. Unfortunately, the use of ketoprofen has recently been reported to cause mortalities in Gyps vultures. Captive and wild-caught vultures were fed ketoprofentreated cattle-tissue containing concentrations similar to the levels found in deceased cattle available to wild vultures [7]. These findings stand in some contrast to previous indications that ketoprofen is rapidly eliminated from livestock tissues and does not cause mortality after exposure to Gyps vultures and other scavenging birds at therapeutic doses [8]. Also, ketoprofen-related mortality has been reported in male eider ducks given the drug [9]. The symptoms were identical to those found in the Gyps vultures exposed to diclofenac, i.e. renal failure and visceral gout. The occurrences of both ketoprofen and diclofenac in effluents from sewage treatment plants (STPs) are typically at similar levels, up to approximately 1 g/L [10–13]. Several studies have reported effects on fish at these concentrations of diclofenac, ranging from cytological and histological effects [14–17] to effects on gene expression [15,18]. Reported bioconcentration factors (BCFs) for diclofenac in rainbow trout varies between studies [10,11,16,18,19]. However, the most recent studies are in relatively close agreement, showing a stable BCF from water to blood plasma of approximately 4 [18,19]. We have recently shown effects on the global hepatic gene expression at exposure concentrations of diclofenac down to 1.6 g/L, even though the corresponding plasma concentration of exposed fish was relatively low (approximately 6 ng/mL) in comparison with the human therapeutic plasma concentrations (HTPC) of >420 ng/mL [20,21]. The number of affected genes as well as their fold change increased with an increasing exposure concentration up to a blood plasma concentration just below the HTPC [18]. Based on its frequent occurrence in the environment together with its risks to affect wildlife, diclofenac was very recently included in the substance priority list within the Water Framework Directive [22]. As ketoprofen and diclofenac often are used for similar indications, we wanted to investigate whether the use of ketoprofen could pose lower risks for fish. However, to the best of our knowledge, studies on the effects of ketoprofen exposure on fish are very scarce. Thibaut et al. [23] have studied the potential of ketoprofen impact on metabolism in carp liver in vitro and there are a few studies on the bioconcentration potential of ketoprofen, though the BCF to fish blood plasma is reported to vary quite much from 0.1 to 48 [10,11]. Here we studied the water to plasma BCF of rainbow trout during controlled flow-through exposure experiments with multiple water concentrations of ketoprofen. Measurements of bioconcentration to blood plasma provide several advantages to whole body bioconcentration (traditional BCF). Importantly, HTPCs are most often available, therefore data on fish plasma levels provide an opportunity for read across between species and thus an estimate of the risk for a pharmacological response in the fish
101
[2,3,24]. Another reason to focus on blood plasma levels is that they conceptually provide a better measure of exposure at the targets compared with whole body levels as some substances are mainly stored in fat and thus not readily available. We have also put our findings in perspective by performing a meta-analysis including previously reported data on BCFs between water and blood plasma for three different NSAIDs. Finally, similar to our previous study on diclofenac [18] we searched for effects on global hepatic gene expression in rainbow trout exposed to ketoprofen as an additional strategy to determine if water concentrations found in the environment are likely to evoke detectable pharmacological responses in exposed fish. 2. Materials and methods 2.1. Fish exposure and sampling Ketoprofen (purity ≥ 98%) (Sigma–Aldrich, Steinheim, Germany) was dissolved in water (500 mg/L concentrated solution), stirred vigorously and diluted stepwise to obtain stock concentrations (0.5 mg/L, 5 mg/L and 50 mg/L). Juvenile rainbow trout (Oncorhynchus mykiss) of both sexes (age: approximately six months, weight: 39.4 ± 6.6 g) were obtained from Vänneåns fiskodling AB, Sweden, and kept in 500 L holding tanks with sand-filtered, aerated fresh water in a flow-through system for one week prior to the experiment. Eight experimental 48 L glass aquaria, semi-covered in black plastic bags to reduce visual stress from the outside (12:12 h light:dark photocycle), were supplied with filtered, aerated freshwater at a flow rate of 0.25 L/min. At the onset of the experiment 80 trout were randomly divided among the aquaria (ten fish in each). Peristaltic pumps (500 L/min) from each stock solution were used to reach target ketoprofen concentrations of 0 (control), 1 (low), 10 (intermediate) and 100 (high) g/L with two aquaria for each concentration. Water samples were taken from each aquarium at day 1, 7 and 14 and stored at −20 ◦ C until analysis. To minimize biological variation due to variable food intake, the fish were not fed during the exposure period. It should be stressed that trout readily cope with food deprivation for much longer periods at these temperatures [25]. To ensure a stable water quality, temperature (12.9 ± 0.1 ◦ C), pH (7.4 ± 0.2) and oxygen saturation (88.3% ± 1.3) were monitored every third day. The total organic carbon content of the supply water was 2.5 mg/L. At the experiment termination on day 14, the fish were killed by a rapid blow to the head and blood was collected from the caudal vessels by the use of heparinized syringes. The plasma was instantly separated by centrifugation at 10,000 rpm for 2 min and snap frozen in liquid nitrogen. The length and weight of the fish were measured and their sex determined by macroscopical observation of their gonads. Liver samples for gene expression analyses were collected and snap frozen in liquid nitrogen. Both liver and plasma samples were stored at −70 ◦ C until analysis. All animal experiments were approved by the local animal committee in Gothenburg (permission number 216-2010). 2.2. Chemical analyses Acetonitrile, methanol and water (Lichrosolv) and sulfuric acid (puriss pa) were obtained from Merck and formic acid (puriss pa) was obtained from Fluka. Labeled naproxen (methyl-13 C, 99%; methyl-D3 , 98%) was used as surrogate standard and was obtained from Cambridge Isotope Laboratories (Andover, MA, USA). Diluted plasma samples (0.5 mL with 1.5 mL 1% aqueous formic acid and 50 ng of surrogate standard) and aqueous samples (100 mL, acidified to pH 3 with sulfuric acid, with 50 ng surrogate standard) were filtered through a 0.45 m MFTM-membrane filter (Millipore,
102
F. Cuklev et al. / Journal of Hazardous Materials 229–230 (2012) 100–106
Sundbyberg, Sweden). Solid phase extraction (SPE) columns (Oasis HLB, 200 mg; Waters Corporation, Milford, MA, USA) were preconditioned and equilibrated with 5.0 mL of methanol and 5.0 mL of de-ionized water. Water with 5% methanol was used to wash the SPE column before eluting with 5 mL of methanol. Eluate was concentrated and reconstituted in water/methanol (80:20) containing 0.1% formic acid to a total volume of 100 L for the fish plasma samples and 1 mL for the water samples. Recoveries of the sample pre-treatments were investigated by spiking 100 ng ketoprofen to tap water and to fish plasma from non-exposed fish (n = 6 for each pre-treatment). Triple stage quadrupole MS/MS TSQ Quantum Ultra EMR (Thermo Fisher Scientific, San Jose, CA, USA) coupled with Accela LC pump (Thermo Fisher Scientific, San Jose, CA, USA) and PAL HTC autosampler (CTC Analytics AG, Zwingen, Switzerland) were used as analytical system for the LC–MS. Aliquots (10 L) were injected onto a Hypersil GOLD aQ column (50 mm × 2.1 mm ID × 5 m particles, Thermo Fisher Scientific, San Jose, CA, USA). Mobile phase consisting of methanol, acetonitrile and water (all solvents buffered by 0.1% formic acid) was used for elution of analytes. The gradient was programmed as follows: 200 L/min 20% methanol in water for 2 min isocratically, then composition is changed to 40% of ACN and 60% of methanol with flow of 300 L/min for 6 min. These parameters were kept for 2 min and then they were switched to starting conditions and let 5 min to equilibrate for next run. Atmospheric pressure photo ionization (APPI) in positive ion mode was used for ionization of ketoprofen and surrogate standard. The setting of key parameters was as follows: sheath gas 50 and auxiliary gas 15 arbitrary units respectively, vaporizer temperature 450 ◦ C, capillary temperature 350 ◦ C, collision gas was argon at 1.5 mL/min. Two SRM transitions were monitored for ketoprofen (255 → 209.1, collision energy 13; 255 → 105.2, collision energy 36) and one for the surrogate standard 13 C, D3 -labeled naproxen (235 → 189, collision energy 13). To reduce matrix effect, resolution at first quadrupole was set to 0.4 FWMH. Quantification based on 5 point calibration curves and calibration standards were run in the beginning and end of each run, with additional standards in the middle of the run (after every 10th sample). Several different blanks are also run at each run, including instrumental blanks to evaluate carry over effects and procedural blanks to evaluate possible laboratory contamination. The maximum differences between results at quantification and qualification masses/mass transitions were set to 30% as criterion for positive identification. For the purpose of simplicity, the term BCF in this paper refers to the relation between blood plasma concentrations in fish and the exposure concentration in water, unless otherwise stated. To assess whether the BCF of ketoprofen depends on water concentration, a statistical analysis was performed using linear regression. Meta-analysis on reported BCFs of ketoprofen was performed by collecting available data from the literature [10,11,19,26]. To make fair comparisons, only BCFs to fish blood plasma were included. The selection of the two other NSAIDs besides from ketoprofen (naproxen and ibuprofen), was based on whether data was available from different types of exposures. Diclofenac was not included here as a similar compilation has been performed in a previous study of ours [18]. 2.3. Microarray design A 15k rainbow trout gene expression microarray was designed for the RT analyzer platform (febit, Heidelberg, Germany) using The Institute for Genomic Research (TIGR) Rainbow Trout Gene Index (RTGI) database version 7.0 (http://compbio.dfci.harvard.edu/tgi/). Details on the probe design strategy, but for eelpout (Zoarces viviparus), and transcript selection strategy are described elsewhere [18,27,28]. When available, transcripts at GenBank
(http://www.ncbi.nlm.nih.gov/nucleotide) were used. Results from similar microarrays, performed by our research group using the same platform, have shown good correlation with qPCR data [18,27–30]. The design, including all probe sequences, and data from the complete microarray experiment is available at GEO database (accession number [GEO:GSE37628]), http://www.ncbi.nlm.nih.gov/geo/, according to the minimum information about a microarray experiment (MIAME) guidelines. 2.4. Microarray synthesis, hybridization and sample preparation The frozen liver tissue was homogenized using Tissuelyser (Qiagen) and total hepatic RNA was isolated using QIAcube and RNeasy® Plus Mini Kit (Qiagen). The RNA quantity was determined with spectophotometric measurements (Nanodrop 1000, NanoDrop Technologies) and the quality was assessed by Experion automated electrophoresis using RNA StdSense chip (Bio-Rad, Sundbyberg, Sweden). Biotinylated aRNA was synthesized using MessageAmpTM II-Biotin Enhanced Single Round aRNA Amplification Kit (Ambion® ). The aRNA samples (20 g) were vacuum dried in a vacuum centrifuge, dissolved in 10 L water and fragmented according to the manufacturer’s protocol. The following steps described in this subchapter were all performed by febit. Oligonucleotide arrays were synthesized by photo-controlled in situ synthesis using the Geniom One system (febit). Each biochip consists of eight individually accessible microchannels, each of which is referred to as a microarray in this manuscript. Four individuals (of both sexes) from each control aquarium and each high exposure concentration aquarium were included in the analysis, i.e. eight fish per exposure concentration. In total, two biochips were analyzed corresponding to 16 microarrays. Pre-hybridization and hybridization were performed based on a customized protocol, described in detail by Cuklev et al. [18]. The samples were randomly allocated on the biochips. Signals were detected using the internal CCD-camera system of the RT analyzer instrument (febit) and quantified using the Geniom Wizard software. Integration times were 266 and 273 ms, determined automatically by the instrument software. 2.5. Microarray analysis All statistical calculations were done in R-2.12.2 (www.rproject.org) [31]. The quality of pre- and post-normalized arrays was verified with Box and MA plots. The data analysis was performed in the R-package LIMMA [32,33]. Data were normalized using the ‘quantile’ method. Moderated t-statistics and adjusted p-values of differential expression were calculated using the empirical Bayes model. As we found no clear differences between the two control aquaria or between the two aquaria with the highest concentration of ketoprofen (100 g/L), all fish from the same treatment were treated as a single group, i.e. we compared one control group to one group of fish exposed to 100 g/L of ketoprofen. 2.6. Quantitative RT-PCR To verify the microarray results, gene expression analyses with quantitative real-time polymerase chain reaction (qPCR) analyses were performed with mRNA from fish not included on the array, four individuals from each control aquarium and four from each high exposure concentration aquarium. The idea of analyzing different fish than those analyzed on the array is that it reduces the risk for incorrectly identifying genes as differentially expressed between treatment groups, a major challenge when thousands of endpoints are analyzed in parallel. The samples of total RNA were reverse transcribed to cDNA with a mixture of oligo(dT)
F. Cuklev et al. / Journal of Hazardous Materials 229–230 (2012) 100–106
103
Table 1 Primer sequences used to validate the microarray results by quantitative PCR. Annotation fancl mid1 fth ptgs1 ptgs2 cox1 cox2 ubq actb a b
Accession number a
TC164519 TC156087a CB486370a NM 001124361.1b NM 001124348.1b NA NA NA NA
Forward sequence 5 –3
Reverse sequence 5 –3
agcatcttgcaaaacctccg aggccaagaaccagacctacct cctgcacaagattgcctctga aaggttaccaacatgtcgcgg tggcgtggacttaaaccatgtt caaacgagaggtagcatcaatc ccaggacgattaaaccaaacag acaacatccagaaagagtccac tctcagtctcattggcatggc
aggcgtaacagatcccacactc catgacagagaaaagccccct gtctcccagcttcttaatggcc aaacctgacacacacaccccag tagacctcgccgttcaaaacc gttcttcaaatgtgtggtaggg accgcttcaacaacgatagg aggcgagcgtagcacttg gctgtttcaccgttccagttgt
Transcript from the Institute for Genomic Research (TIGR) Rainbow Trout Gene Index (RTGI) database version 7.0 (http://compbio.dfci.harvard.edu/tgi/). Transcript from GenBank.
and random hexamers, using iScriptTM cDNA Synthesis Kit (Biorad) according to the manufacturer’s instructions. In addition, no reverse transcriptase (NoRT) controls were synthesized for each sample. The following genes were analyzed: prostaglandin G/H synthase 1 (ptgs1), prostaglandin G/H synthase 2 (ptgs2), Fanconi anemia complementation group L (fancl), ferritin H-3 (fth) and MID1 interacting protein 1 (mid1). Also included were cytochrome c oxidase I (cox1) and cytochrome c oxidase II (cox2) from the study by Mehinto et al. [15], which were not present on the array. Ubiquitin (ubq) and -actin (actb) were chosen as potential housekeeping genes. Primers were designed using Primer Express software (Applied Biosystems, Warrington, UK; not cox1 and cox2 for which the primer sequences given by Mehinto et al. [15] were used) and purchased from Eurofins MWG Operon (Ebersberg, Germany) (Table 1). For each primer pair, qPCR runs with a standard curve with six cDNA dilutions as starting material were performed to ensure optimized assays. Each qPCR reaction was carried out in triplicates and contained 1× Power SYBR® Green mastermix (Applied Biosystems), 100 nM of each primer pair and 12.5 ng cDNA in a final reaction volume of 6 L. The qPCR analysis was performed on a ABI 7900HT with 10 min initial denaturation at 95 ◦ C, followed by 40 cycles of 95 ◦ C for 15 s and 60 ◦ C for 1 min. After each run a melting curve analysis was performed to verify the specificity of the amplification (95 ◦ C for 15 s, 60 ◦ C for 15 s and 95 ◦ C for 15 s). To ensure that no genomic DNA, primer-dimers etc. could inflict the results, NoRT controls were run for each sample. None of the NoRT controls had a threshold cycle (Ct ) value below 30 and all had Ct values more than 5 cycles after the actual sample. However, due to technical errors, one sample from control group A was excluded from ptgs1, fth and mid1 analyses; one sample from control group B was excluded from the runs with mhc1, cox1, cox2, and fancl primers. The expression of both housekeeping genes was stable and showed no tendencies to be differentially regulated between the groups and were thus used for normalization by subtracting the average Ct value of both genes for each sample. The initial microarray analysis showed no consistent differences between fish from the duplicate aquaria for each exposure, hence the fish from replicate tanks were considered as one group for the following qPCR analyses. The resulting Ct values were used for the statistical analyses where tests for differential expression between control and exposed fish were performed using a single-sided Student’s t-test assuming equal variances. Furthermore, the correlation between blood plasma concentration of ketoprofen and gene expression on individual levels were analyzed for these seven genes using linear regression and Bonferroni correction.
and R2 above 0.99 for all standard curves. No ketoprofen was detected in the instrumental or procedural blanks. Absolute recoveries of the solid phase extraction of the fish plasma and water samples were 127% (17% RSD, n = 6, spiking level 100 ng) and 103% (8% RSD, n = 6, spiking level 100 ng) respectively. The following results are all displayed in the order of low, intermediate and high exposure concentration, respectively. The average water concentrations of ketoprofen throughout the exposure were 1.2 ± 0.21, 7.2 ± 2.59 and 90.5 ± 32.98 g/L. The bioconcentration potential of ketoprofen was in this study low compared to e.g. diclofenac, hence the blood plasma concentrations, 0.05 ± 0.02, 0.19 ± 0.08 and 0.62 ± 0.49 ng/mL, were all well below human therapeutic levels (HTPC > 1000 ng/mL) [34]. No residues of ketoprofen could be found in the controls (detection limit = 0.005 ng/mL). There was a significant decrease (p < 0.0001) in the BCF with increasing water concentration, from 0.042 down to 0.007 (Fig. 1).
3.2. Meta-analysis The compilation of reported BCFs for ketoprofen, naproxen and ibuprofen revealed that BCFs measured in fish exposed to undiluted effluents tend to be higher than BCFs measured in fish exposed to a single substance (or a mixture of very few substances) in pure water (Fig. 2). For ketoprofen the BCF for effluent-exposed fish was 35–6857 times higher, for ibuprofen 2–13333 times higher and for naproxen 5.5–40 times higher than the reported BCFs in studies where fish were exposed to water with a lower complexity. To the best of our knowledge, no BCF data from wild fish blood plasma is available for any of the three selected substances.
3. Results 3.1. Bioconcentration Ketoprofen concentration was successfully measured in water and blood plasma samples, with stable and reproducible results
Fig. 1. Bioconcentration factor of ketoprofen in blood plasma of rainbow trout exposed to waterborne ketoprofen for 14 days at different concentrations. The bioconcentration factor decreased significantly (p < 0.0001) with an increasing water concentration of ketoprofen. Statistical analysis was performed using linear regression.
104
F. Cuklev et al. / Journal of Hazardous Materials 229–230 (2012) 100–106
3.3. Gene expression
Fig. 2. A comparison of bioconcentration factors of three NSAIDs to blood plasma in fish exposed to single substances or a mixture of a few substances in pure water under controlled lab conditions (closed symbols) versus fish exposed to undiluted sewage effluents (open symbols). Data were collected from the present study and other studies [10,11,19,26]. Only nominal concentrations were used for the lab exposure in the study by Brown et al. [10].
Due to the low bioconcentration potential of ketoprofen and thereby low plasma concentrations in all exposure groups, the gene expression pattern was only analyzed in the fish exposed to the highest concentration. The microarray analysis revealed 59 genes with p < 0.05, however the adjusted p-value was >0.99 for all genes (Supplementary Table A.1). To confirm these results, qPCR was performed on different individuals from the ones present on the arrays. A repertoire of genes were selected based on different sources: fancl and mid1 were top-ranked genes based on p-value in the microarray analysis; fth was present among the top-ranked genes in the study by Cuklev et al. [18] as well as in the upper layer in the present study; ptgs1 and ptgs2 are the primary targets of ketoprofen and other NSAIDs; cox1 and cox2 were analyzed by Mehinto et al. [15]. There were no significant differences between treatment groups for any of the genes analyzed (Fig. 3). The linear regression analysis of the correlation between blood plasma concentration of ketoprofen and gene expression on individual levels did not reveal any significance for the genes analyzed by qPCR (p > 0.48) except for fth (p = 0.0497 without Bonferroni correction). However, when adjusting the p-value for multiple testing (n = 8) fth was not significantly regulated (Supplementary Fig. B.1). Neither the microarray nor the qPCR analysis resulted in any significant differences in responses between males and females.
4. Discussion
Fig. 3. Hepatic gene expression of selected genes in rainbow trout after ketoprofen exposure (90.5 g/L) analyzed by qPCR (white bars) or microarray (black bars). Values are expressed as log fold change (log2 ) compared with control fish. Statistical analyses were performed on the qPCR data using a Student’s t-test assuming equal variances (single-sided, based on the direction of regulation on the microarray) followed by Bonferroni correction. None of the genes were significantly regulated.
In the present study we show that waterborne ketoprofen bioconcentrates considerably less in fish than diclofenac under controlled laboratory conditions. Measured plasma levels of ketoprofen in fish at an exposure concentration about 100 times higher than levels found in undiluted sewage effluents reached less than 1‰ of human therapeutic plasma levels. In accordance no effects on the global hepatic gene expression were found, in sharp contrast to previous reports for diclofenac [15,18]. These results support our hypothesis that the use of ketoprofen rather than diclofenac may pose lower risks for exposed wild fish. Results from other exposure experiments with complex, undiluted treated sewage effluent, however, complicate the interpretation somewhat. During exposure to undiluted effluent, ketoprofen as well as ibuprofen and naproxen appear to bioconcentrate considerably more than in controlled lab experiments with drugs diluted in pure water, whereas for diclofenac the difference is not as clear. It is therefore not entirely straightforward how to extrapolate the results from the present study (and other lab studies on NSAIDs) to the situation for wild fish. Ketoprofen and diclofenac are often used for similar indications, though diclofenac is often used to a higher extent. Over the last four years, the number of defined daily doses (DDD) of diclofenac sold in Sweden has been approximately four times higher than for ketoprofen (43.4 M and 11.1 M DDD in 2010 for diclofenac and ketoprofen respectively) [35]. Both compounds are mainly excreted as metabolites (>99%; http://www.fass.se) of which some are glucuronide conjugates that are likely to be cleaved during sewage treatment, thus regenerating the parent compound [12,36]. Both are found at approximately the same concentration in STP influents (typically just below 1 g/L) [37–39]. Removal efficiency of both compounds is low to moderate (5–70%), and they are often found in similar concentrations in effluents, with an average from four studies of 1.4 g/L and 0.83 g/L for ketoprofen and diclofenac respectively [11,12,38,39]. Fernandez et al. [40] reported slightly higher concentrations of ketoprofen than diclofenac in a Spanish river (0.3–991 ng/L and 0.7–156 ng/L respectively), though in German rivers the maximum concentration of 2.1 g/L diclofenac
F. Cuklev et al. / Journal of Hazardous Materials 229–230 (2012) 100–106
was higher than ketoprofen levels of 0.38 g/L [12]. Given similar water levels, their bioconcentration potential is therefore crucial for assessing differences in their potential to affect aquatic organisms. Here, rainbow trout was exposed to waterborne ketoprofen at three different concentrations, where the lowest one (measured concentration of 1.2 g/L) corresponds to concentrations found in undiluted STP effluents. The bioconcentration potential of ketoprofen was relatively low (BCF < 0.05) compared with approximately 4 for diclofenac based on a very similar exposure setup [18]. The slightly lower lipophilicity of ketoprofen indeed suggests a lower bioconcentration potential of ketoprofen, though not quite to the extent observed here [1–3]. The BCF decreased significantly from 0.042 down to 0.007 with increasing water concentration (1.2–90.5 g/L). A variable BCF for diclofenac from water to fish liver was reported by Schwaiger et al. [16] whereas a stable BCF to both liver and plasma for diclofenac was reported by us [18]. In the study by Schwaiger et al. [16], the BCF dropped from >2000 at the lowest exposure concentration down to approximately 10 at a 500 times higher exposure concentration i.e. a 200-fold BCF drop. Here, the drop was 6-fold over a 100-fold exposure concentration. It should also be noted that, in contrast to other studies with diclofenac [15,18,19] as well as for the present study, Schwaiger et al. [16] used dimethyl sulfoxide (DMSO; 0.12‰) as a solvent. Most predictive models for bioconcentration assume a stable BCF over exposure concentrations [41]. Accordingly, decreasing BCFs with increasing water concentrations are not commonly reported, though the phenomenon has been observed for other chemicals in other species (Oryzias latipes, Perna viridis and Dreissena polymorpha) [41–43]. Such patterns could for example be explained by an increased activity of detoxification systems at higher exposures, if there is a saturation of binding sites for the bioconcentrating chemical in the organism or if the energy required to bind the chemical is not sufficient at higher concentrations [41]. However, the latter explanations seem less likely as considerably higher plasma concentrations of ketoprofen in trout have been reported than those detected here [10,11]. Previous studies have been somewhat contradictive regarding the BCF of ketoprofen, though the experimental conditions differed. The measured BCF in the present study differ considerably from studies where rainbow trout were exposed to undiluted sewage effluents (BCF = 3.5–48) [11], but is more similar to the BCF of 0.1 reported in the study by Brown et al. [10] where rainbow trout were exposed to nominal concentrations of ketoprofen together with four other pharmaceuticals in pure water under controlled lab conditions. Indeed, lower BCFs in exposure situations with single drug(s) in pure water compared with exposure to undiluted effluents is a collective trend for ketoprofen, ibuprofen and naproxen, and also for diclofenac but to a lesser extent [18]. It is therefore possible that the concentration of other constituents of treated sewage effluents can influence the uptake, distribution, metabolism or excretion of NSAIDs and possibly other pharmaceuticals, and thus the expected effect/toxicity, to a considerable extent. This possibility is not taken into account in current environmental risk assessment procedures, such as those for medicinal products in the EU [44] and the USA [45]. It should be stressed that neither the exposure to single drugs in pure water, nor exposure to 100% undiluted effluents represent the regular situation for fish in the wild, but we still do not know which experimental setup provides the best approximation. At the highest tested exposure concentration in the present study (90.5 g/L), plasma concentrations corresponding to 0.62‰ of the HTPC were achieved. Thus, one would presuppose minor, if any, changes in the gene expression pattern compared to non-exposed fish [46]. The microarray analysis in the present study revealed 59 genes with p < 0.05. However, given the large
105
number of probes present on the array, it is likely that some genes are falsely identified as regulated. Accordingly, our assessment shows an adjusted p-value > 0.99 for all genes analyzed on the array suggesting that all of them could very well be false positives. To verify these results, seven genes were analyzed using qPCR on other individuals than on the array, i.e. verification by biological replication. Firstly, the two top-ranked genes fancl [RTGI: TC164519] and mid1 [RTGI: TC156087] both had a p-value <0.03 (not adjusted for multiple testing) on the array, though neither was significantly regulated according to the qPCR. The list of the 1000 most likely regulated genes on the array was compared with the ranking lists at different exposure concentrations of diclofenac from the study by Cuklev et al. [18] and fth ([RTGI: CB486370] and [RTGI: TC171322] in the present study and in Cuklev et al., respectively) was one of the genes present on all lists. Ferritin is a ubiquitous intracellular protein that stores iron [47] and fth was down regulated in both studies, though only significantly on the diclofenac array. None of the main target genes, ptgs1 [GenBank: NM 001124361.1] and ptgs2 [GenBank: NM 001124348.1], were significantly regulated on neither the array nor the qPCR. If the highest reported BCF for ketoprofen [11] is applied to an effluent dominated stream with a surface water concentration of 1 g/L of ketoprofen, the predicted blood plasma concentration in exposed fish would be 48 ng/mL, i.e. about 100 times higher than the levels found here in fish exposed to 90.5 g/L. This is still only about 5% of the HTPC, however plasma concentrations of diclofenac, considerably lower than 5% of the corresponding HTPC, have been reported to affect gene expression in fish [15,18]. Thus, we cannot yet rule out a pharmacological response at water concentrations of ketoprofen similar to those concentrations found in STP effluents. If we in a similar way apply the highest reported BCF from effluent to fish blood plasma for diclofenac, i.e. 29 [11] and a surface water concentration of 1 g/L, that would correspond to a blood plasma concentration of about 7% of the HTPC, thus quite similar to the predicted ratio for ketoprofen. Also, several NSAIDs are regularly found together in surface water, and as they are acting via the same or similar modes of action their effects are expected to be additive. Ibuprofen, for example, have quite recently been shown to cause a decrease in gill tissue prostaglandin E2 levels, but only significantly at relatively high exposure concentrations (≥50 g/L), most likely a consequence of the quite poor bioconcentration potential of ibuprofen (BCF = 1.3) [48]. Whether the gene expression is affected at lower exposure concentrations is not known. Given these premises, it is not entirely clear what alternative pose the lowest risk for wild fish. However, if controlled exposure conditions with a single chemical in pure water, as applied here, reflect the environment, ketoprofen would pose lower risk than diclofenac to fish. Acknowledgements The authors thank Lina Gunnarsson, Bethanie Carney-Almroth and Hannah Svedlund for help with experiments and analyses. This research was supported by the Swedish Foundation for Strategic Environmental Research (MISTRA), the Swedish Research Council (VR) and the Adlerbertska Research Foundation. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jhazmat. 2012.05.077. References [1] P.N. Fitzsimmons, J.D. Fernandez, A.D. Hoffman, B.C. Butterworth, J.W. Nichols, Branchial elimination of superhydrophobic organic compounds by rainbow trout (Oncorhynchus mykiss), Aquat. Toxicol. 55 (2001) 23–34.
106
F. Cuklev et al. / Journal of Hazardous Materials 229–230 (2012) 100–106
[2] D.B. Huggett, J.C. Cook, J.F. Ericson, R.T. Williams, A theoretical model for utilizing mammalian pharmacology and safety data to prioritize potential impacts of human pharmaceuticals to fish, Hum. Ecol. Risk. Assess. 9 (2003) 1789–1799. [3] R. Schreiber, U. Gündel, S. Franz, A. Küster, B. Rechenberg, R. Altenburger, Using the fish plasma model for comparative hazard identification for pharmaceuticals in the environment by extrapolation from human therapeutic data, Regul. Toxicol. Pharmacol. 61 (2011) 261–275. [4] J.L. Oaks, M. Gilbert, M.Z. Virani, R.T. Watson, C.U. Meteyer, B.A. Rideout, H.L. Shivaprasad, S. Ahmed, M.J. Iqbal Chaudhry, M. Arshad, S. Mahmood, A. Ali, A. Ahmed Khan, Diclofenac residues as the cause of vulture population decline in Pakistan, Nature 427 (2004) 630–633. [5] S. Shultz, H.S. Baral, S. Charman, A.A. Cunningham, D. Das, G.R. Ghalsasi, M.S. Goudar, R.E. Green, A. Jones, P. Nighot, D.J. Pain, V. Prakash, Diclofenac poisoning is widespread in declining vulture populations across the Indian subcontinent, Proc. Biol. Sci. 271 (Suppl. 5) (2004) S458–S460. [6] D.J. Pain, C.G.R. Bowden, A.A. Cunningham, R. Cuthbert, D. Das, M. Gilbert, R.D. Jakati, Y. Jhala, A.A. Khan, V. Naidoo, J.L. Oaks, J. Parry-Jones, V. Prakash, A. Rahmani, S.P. Ranade, H.S. Baral, K.R. Senacha, S. Saravanan, N. Shah, G. Swan, D. Swarup, M.A. Taggart, R.T. Watson, M.Z. Virani, K. Wolter, R.E. Green, The race to prevent the extinction of South Asian vulture, Bird Conserv. Int. 18 (2008) S30–S48. [7] V. Naidoo, K. Wolter, D. Cromarty, M. Diekmann, N. Duncan, A.A. Meharg, M.A. Taggart, L. Venter, R. Cuthbert, Toxicity of non-steroidal anti-inflammatory drugs to Gyps vultures: a new threat from ketoprofen, Biol. Lett. 6 (2010) 339–341. [8] R. Cuthbert, J. Parry-Jones, R.E. Green, D.J. Pain, NSAIDs and scavenging birds: potential impacts beyond Asia’s critically endangered vultures, Biol. Lett. 3 (2007) 91–94. [9] D.M. Mulcahy, P. Tuomi, R.S. Larsen, Differential mortality of male spectacled eiders (Somateria fischeri) and king eiders (Somateria spectabilis) subsequent to anesthesia with propofol, bupivacaine, and ketoprofen, J. Avian Med. Surg. 1 (7) (2003) 117–123. [10] J.N. Brown, N. Paxéus, L. Förlin, D.G.J. Larsson, Variations in bioconcentration of human pharmaceuticals from sewage effluents into fish blood plasma, Environ. Toxicol. Pharmacol. 24 (2007) 267–274. [11] J. Fick, R.H. Lindberg, J. Parkkonen, B. Arvidsson, M. Tysklind, D.G.J. Larsson, Therapeutic levels of levonorgestrel detected in blood plasma of fish: results from screening rainbow trout exposed to treated sewage effluents, Environ. Sci. Technol. 44 (2010) 2661–2666. [12] T.A. Ternes, Occurrence of drugs in German sewage treatment plants and rivers, Water Res. 32 (1998) 3245–3260. [13] C. Tixier, H.P. Singer, S. Oellers, S.R. Müller, Occurrence, Fate of carbamazepine, clofibric acid, diclofenac, ibuprofen, ketoprofen, and naproxen in surface waters, Environ. Sci. Technol. 3 (7) (2003) 1061–1068. [14] B. Hoeger, B. Kollner, D.R. Dietrich, B. Hitzfeld, Water-borne diclofenac affects kidney and gill integrity and selected immune parameters in brown trout (Salmo trutta f. fario), Aquat. Toxicol. 75 (2005) 53–64. [15] A.C. Mehinto, E.M. Hill, C.R. Tyler, Uptake and biological effects of environmentally relevant concentrations of the nonsteroidal anti-inflammatory pharmaceutical diclofenac in rainbow trout (Oncorhynchus mykiss), Environ. Sci. Technol. 44 (2010) 2176–2182. [16] J. Schwaiger, H. Ferling, U. Mallow, H. Wintermayr, R.D. Negele, Toxic effects of the non-steroidal anti-inflammatory drug diclofenac. Part I: Histopathological alterations and bioaccumulation in rainbow trout, Aquat. Toxicol. 68 (2004) 141–150. [17] R. Triebskorn, H. Casper, A. Heyd, R. Eikemper, H.R. Kohler, J. Schwaiger, Toxic effects of the non-steroidal anti-inflammatory drug diclofenac. Part II: Cytological effects in liver, kidney, gills and intestine of rainbow trout (Oncorhynchus mykiss), Aquat. Toxicol. 68 (2004) 151–166. [18] F. Cuklev, E. Kristiansson, J. Fick, N. Asker, L. Förlin, D.G.J. Larsson, Diclofenac in fish: blood plasma levels similar to human therapeutic levels affect global hepatic gene expression, Environ. Toxicol. Chem. 30 (2011) 2126–2134. [19] M. Lahti, J.-M. Brozinski, A. Jylhä, L. Kronberg, A. Oikari, Uptake from water, biotransformation, and biliary excretion of pharmaceuticals by rainbow trout, Environ. Toxicol. Chem. 3 (0) (2011) 1403–1411. [20] J.G. Hardman, L.E. Limbard, A.G. Gilman, The Pharmacological Basis of Therapeutics, McGraw-Hill, NY, USA, 2001. [21] T. Juhlin, S. Björkman, B. Gunnarsson, Å. Fyge, B. Roth, P. Höglund, Acute administration of diclofenac but possibly not long term low dose aspirin, causes detrimental renal effects in heart failure patients treated with ACE-inhibitors, Eur. J. Heart. Fail. 6 (2004) 909–916. [22] European Union, Proposal for a Directive of the European Parliament and of the Council Amending Directives 2000/60/EC and 2008/105/EC as Regards Priority Substances in the Field of Water Policy, Water framework directive, Brussels, Belgium, 2012. [23] R. Thibaut, S. Schnell, C. Porte, The interference of pharmaceuticals with endogenous and xenobiotic metabolizing enzymes in carp liver: an in vitro study, Environ. Sci. Technol. 40 (2006) 5154–5160.
[24] V. Roos, L. Gunnarsson, J. Fick, D.G.J. Larsson, C. Rudén, Prioritising pharmaceuticals for environmental risk assessment: towards adequate and feasible first-tier selection, Sci. Total. Environ. 421–422 (2012) 102–110. [25] A. Kullgren, L.M. Samuelsson, D.G.J. Larsson, B.T. Björnsson, E.J. Bergman, A metabolomics approach to elucidate effects of food deprivation in juvenile rainbow trout (Oncorhynchus mykiss), Am. J. Physiol: Regul. Integr. Comp. Physiol. 299 (2010) R1440–R1448. [26] G.C. Nallani, P.M. Paulos, L.A. Constantine, B.J. Venables, D.B. Huggett, Bioconcentration of ibuprofen in fathead minnow (Pimephales promelas) and channel catfish (Ictalurus punctatus), Chemosphere 84 (2011) 1371–1377. [27] E. Kristiansson, N. Asker, L. Forlin, D.G.J. Larsson, Characterization of the Zoarces viviparus liver transcriptome using massively parallel pyrosequencing, BMC Genomics 10 (2009) 345. [28] F. Cuklev, L. Gunnarsson, M. Cvijovic, E. Kristiansson, C. Rutgersson, B. Björlenius, D.G.J. Larsson, Global hepatic gene expression in rainbow trout exposed to sewage effluents: a comparison of different sewage treatment technologies, Sci. Total. Environ. 427–428 (2012) 106–114. [29] L. Gunnarsson, E. Kristiansson, C. Rutgersson, J. Sturve, J. Fick, L. Förlin, D.G.J. Larsson, Pharmaceutical industry effluent diluted 1:500 affects global gene expression, cytochrome P4501A activity and plasma phosphate in fish, Environ. Toxicol. Chem. 28 (2009) 2639–2647. [30] A. Lennquist, N. Asker, E. Kristiansson, A. Brenthel, B.T. Björnsson, P. Kling, M. Hultman, D.G.J. Larsson, L. Förlin, Physiology and mRNA expression in rainbow trout (Oncorhynchus mykiss) after long-term exposure to the new antifoulant medetomidine, Comp. Biochem. Physiol. C: Pharmacol. Toxicol. 154 (2011) 234–241. [31] R Development Core Team, R: A Language and Environment for Statistical Computing, R Foundation for Statistical Computing, Vienna, Austria, 2010. [32] G. Smyth, Limma, Linear models for microarray data, in: R. Gentleman, V.J. Carey, W. Huber, R.A. Irizarry, S. Dudoit (Eds.), Bioinformatics and Computational Biology Solutions Using R and Bioconductor, Springer, New York, 2005, pp. 397–420. [33] G.K. Smyth, Linear models and empirical bayes methods for assessing differential expression in microarray experiments, Stat. Appl. Genet. Mol. Biol. 3 (2004) Article3. [34] M. Schulz, A. Schmoldt, Therapeutic and toxic blood concentrations of more than 800 drugs and other xenobiotics, Pharmazie 58 (2003) 447–474. Apoteket, Selling Statistics of Human Pharmaceuticals, [35] A.B. http://www.apotekensservice.se/Om-Apotekens-Service/ 2010, Infrastrukturtjanster/Statistik/. [36] N. Lindqvist, T. Tuhkanen, L. Kronberg, Occurrence of acidic pharmaceuticals in raw and treated sewages and in receiving waters, Water Res. 39 (2005) 2219–2228. [37] C.G. Daughton, T.A. Ternes, Pharmaceuticals and personal care products in the environment: agents of subtle change? Environ. Health Perspect. 107 (1999) 907–938. [38] H. Matsuo, H. Sakamoto, K. Arizono, R. Shinohara, Behavior of pharmaceuticals in waste water treatment plant in Japan, Bull. Environ. Contam. Toxicol. 87 (2011) 31–35. [39] R. Rosal, A. Rodríguez, J.A. Perdigón-Melón, A. Petre, E. García-Calvo, M.J. Gómez, A. Agüera, A.R. Fernández-Alba, Occurrence of emerging pollutants in urban wastewater and their removal through biological treatment followed by ozonation, Water Res. 44 (2010) 578–588. [40] C. Fernández, M. González-Doncel, J. Pro, G. Carbonell, J.V. Tarazona, Occurrence of pharmaceutically active compounds in surface waters of the HenaresJarama-Tajo River system (Madrid, Spain) and a potential risk characterization, Sci. Total. Environ. 408 (2010) 543–551. [41] C. Liu, K.Y.H. Gin, V.W.C. Chang, B.P.L. Goh, M. Reinhard, Novel perspectives on the bioaccumulation of PFCs – the concentration dependency, Environ. Sci. Technol. 45 (2011) 9758–9764. [42] V. Contardo-Jara, C. Lorenz, S. Pflugmacher, G. Nützmann, W. Kloas, C. Wiegand, Molecular effects and bioaccumulation of levonorgestrel in the non-target organism Dreissena polymorpha, Environ. Pollut. 159 (2011) 38–44. [43] T. Kondo, H. Yamamoto, N. Tatarazako, K. Kawabe, M. Koshio, N. Hirai, M. Morita, Bioconcentration factor of relatively low concentrations of chlorophenols in Japanese medaka, Chemosphere 61 (2005) 1299–1304. [44] Guideline on the Environmental Risk Assessment of Medicinal Products for Human Use, 2006, http://www.ema.europa.eu/docs/en GB/document library/ Scientific guideline/2009/10/WC500003978.pdf. [45] U.S. EPA, Guidelines for Ecological Risk Assessment, 1998, http://www.epa. gov/raf/publications/guidelines-ecological-risk-assessment.htm. [46] J.P. Berninger, B.W. Brooks, Leveraging mammalian pharmaceutical toxicology and pharmacology data to predict chronic fish responses to pharmaceuticals, Toxicol. Lett. 193 (2010) 69–78. [47] E.C. Theil, Ferritin: structure gene regulation, and cellular function in animals, plants, and microorganisms, Ann. Rev. Biochem. 5 (6) (1987) 289–315. [48] K. Bhandari, B. Venables, Ibuprofen bioconcentration and prostaglandin E2 levels in the bluntnose minnow Pimephales notatus, Comp. Biochem. Physiol. C: Toxicol. Pharmacol. 153 (2011) 251–257.