Interference of Pharmaceuticals with the Cytochrome P450 family in rainbow trout (Oncorhynchus mykiss)
Daphné Behrens
Degree project inbiology, Master ofscience (2years), 2011 Examensarbete ibiologi 45 hp tillmasterexamen, 2011
Biology Education Centre and Institutionen för organismbiologi, Uppsala University Supervisors: Björn Brunström and Kristina Beijer
2
Table of contents
Table of contents ... 2
Abstract ... 3
Introduction ... 4
Pharmaceuticals ... 4
MistraPharma ... 6
EROD activity and biological effects ... 7
Aims ... 8
Project 1 ... 8
Project 2 ... 8
Material and Method ... 8
The fish ... 8
The pharmaceuticals ... 9
EROD inhibition study ... 9
EROD induction study ... 10
Development of the method ... 10
The exposure ... 11
Gene expression ... 11
Exposure ... 11
Extraction of mRNA and cDNA synthesis ... 11
Quantitative real-time PCR ... 12
Statistics ... 12
Results ... 14
Induction in gill filaments in vitro ... 14
EROD inhibition by the pharmaceuticals ... 17
EROD induction by the pharmaceuticals ... 18
Effects of indigo on EROD activity and gene expression ... 19
Discussion ... 21
The new method ... 21
EROD inhibition/induction by the pharmaceuticals ... 22
The indigo/miconazole interaction ... 23
Conclusion ... 24
Acknowledgements ... 25
References ... 26
Appendix ... 29
3
Abstract
Due to their very high and still increasing consumption, pharmaceuticals have become a real issue regarding environmental pollution. Toxic effects on aquatic organisms of estrogenic and androgenic drugs are well studied but very little is known about drugs having other mechanisms of action such as the neuro-active pharmaceuticals. Ethoxyresorufin O- deethylase (EROD) activity is a widely used biomarker for exposure to aryl hydrocarbon receptor (AhR) ligands. The first aim of this study has been to develop an in vitro gill filament-based method to study EROD induction. Compared with in vivo methods such a method will reduce the amount, waste and costs of experimental chemicals. Rainbow trout (Oncorhynchus mykiss) gill filaments were exposed to -naphthoflavone (BNF) dissolved in different solvents at different temperatures and during different exposure times. Optimal experimental conditions were found to be 6 h of exposure at 19°C while using DMSO as a solvent.
The EROD inducing and inhibiting properties of nine different active pharmaceutical ingredients frequently found in the aquatic environment were then analysed in rainbow trout gill filaments. Five of them were neuro-active drugs (sertraline hydrochloride, fluvoxamine maleate, chlorpromazine hydrochloride, oxazepam, and haloperidol), a broad-spectrum antibiotic (ciprofloxacin), an antifungal (miconazole nitrate), a cholesterol lowering agent (fenofribrate) and a blood pressure lowering agent (losartan potassium). In an ex vivo inhibition study, oxazepam, fenofibrate and losartan did not show any inhibition potency and increasing inhibition was observed for the other compounds as follows: ciprofloxacin <
haloperidol sertraline fluvoxamine chlorpromazine < miconazole. In the in vitro induction study, only oxazepam showed any induction, but it was very weak.
Finally a preliminary experiment on interaction of a readily metabolised compound (indigo) and a very potent cytochrome P450 (CYP) inhibitor (miconazole) was performed. Gene expression analysis showed a clearly higher induction of transcripts for the three CYP1C genes by 1 nM indigo/50 µM miconazole than by 10 nM indigo. The combination of indigo and miconazole caused similar mRNA expressions of the two CYP1A genes and the CYP1B gene as obtained after treatment with 10 nM indigo.
Keywords: Rainbow trout, Aryl hydrocarbon receptor, Cytochrome P450 1A, 1B and 1C, pharmaceuticals, EROD activity, mRNA expression.
Abbreviations:
AhR Aryl hydrocarbon receptor
BNF -naphthoflavone
CYP Cytochrome P450
CYP1A, CYP1B, CYP1C Cytochrome P450 1A, 1B and 1C
EROD 7-ethoxyresorufin O-deethylase
Real-time qPCR Real-time quantitative polymerase chain reaction
4
Introduction
Pharmaceuticals
In our modern society, pharmaceutical products and drugs are widely used in daily life to cure or prevent diseases. Their consumption does not concern only humans but pharmaceuticals are also widely used in veterinary medicine. The worldwide consumption of these products has been reported to be between 100,000 and 200,000 tonnes (Wise, 2002). These compounds can enter the aquatic environment via their manufacturing process or via excretion by users (Larsson et al., 2007), as well as from direct disposal of unused drugs (Daughton and Ternes, 1999). Wastewater treatment plants (WWTPs) remove dissolved toxicants by biodegradation and sorption to sewage sludge. However, most of the pharmaceuticals are hydrophilic and therefore their sorption to sludge is limited (Vieno et al., 2007). Many pharmaceuticals are therefore very poorly removed in WWTPs and are usually found in surface and effluent water at concentrations between low ng/L and low µg/L (Lindberg et al., 2005; Nikolaou et al., 2007). Even though the concentrations measured in the environment are below the therapeutic concentrations (Fent et al., 2005) and proved to be harmless for humans, very little is known about their potential effects on aquatic ecosystems.
Moreover the input of pharmaceuticals to aquatic ecosystems is considered as constantly ongoing; therefore they can be described as “pseudopersistent” compounds (Daughton, 2002).
Several studies following OECD guidelines have shown that acute toxicity to aquatic organisms is very unlikely to happen (Fent et al., 2005), but combined effects of various pharmaceuticals as well as effects of their metabolites remain unclear.
The pharmaceuticals used in this study are shown in figure 1 and include nine different active pharmaceutical ingredients (APIs) from different groups of pharmaceuticals and a dye also used in Chinese medicine:
- Two antidepressants from the Selective Serotonin Reuptake Inhibitor class (SSRI);
Sertraline Hydrochloride (a) and Fluvoxamine Maleate (b). Sertraline is an antidepressant used to treat major depressions (DailyMed homepage sertraline hydrochloride) whereas fluvoxamine is mainly used in the treatment of obsessive-compulsive disorder (DailyMed homepage fluvoxamine maleate).
- One broad-spectrum antibiotic from the fluoroquinolone drug class; Ciprofloxacin (c) (DailyMed homepage ciprofloxacin)
- Two antipsychotic drugs from the first generation, one from the phenothiazine class;
Chlorpromazine Hydrochloride (d) and one from the butyrophenone class; Haloperidol (e).
Chlorpromazine is used to treat manic-depressive disorder and hyperactivity. It is also used to decrease pre-surgical stress (Inchem homepage chlorpromazine) and haloperidol is a tranquilizer used mainly in schizophrenia and Gilles de la Tourette´s syndrome (Inchem homepage haloperidol).
- One anxiolytic drug from the benzodiazepine derivative class; Oxazepam (f) used mainly in the treatment of anxiety associated with depression (DailyMed homepage oxazepam) - One vasoconstrictor from the angiotensin II receptor antagonist class; Losartan Potassium (g) (DailyMed homepage losartan potassium)
- One antifungal from the imidazole class; Miconazole Nitrate (h) (Inchem homepage miconazole)
5 - One cholesterol and triglyceride lowering agent; Fenofibrate (i) (DrugBank homepage fenofibrate)
- Indigo (j) is a dye extracted from several different plants originally used to dye denim clothes. It is also used in Chinese medicine as a component of a cure for certain type of leukemia (Wang et al., 2007).
Seven out of the nine studied compounds are known to be substrates and/or inhibitors of several human CYP genes including CYP1A2, CYP3A4, CYP2D6, CYP2C9, CYP2C19 and CYP2B6. It is however not known whether they interact with CYPs in fish.
(http://www.healthanddna.com/Druglist.pdf)
(a)
N
O CH3 O
N H2
CF3
(b)
N N H
N F
O
OH O
(c) N
S N CH3
CH3
Cl
(d)
F
N
O
Cl OH
(e)
N NH
O
H O
Cl
(f)
N H
CH3
Cl
Cl
H H
6
N Cl N
OH
C H3
N N H
N N
(g)
Cl Cl
O
N N
Cl
Cl (h)
O
O C H3
CH3
Cl
O
CH3 C H3
O
(i)
N H O
N H
O (j)
Figure 1. Structure of the pharmaceuticals used in this study. Sertraline Hydrochloride (a), Fluvoxamine Maleate (b), Ciprofloxacine (c), Chlorpromazine Hydrochloride (d), Haloperidol (e), Oxazepam (f), Losartan Potassium (g), Miconazole Nitrate (h), Fenofibrate (i), Indigo (j). The molecules were drawn using ChemSketch free software version 11.01 built 22419 from ACD/Labs, Ontario, Canada.
MistraPharma
MistraPharma is a Swedish program implemented by the Swedish Foundation for Strategic Environmental Research (MISTRA) in 2008, which aims to identify and reduce the environmental risks caused by the use of pharmaceuticals. The mission of the project has 3 major goals:
-Assess the size and the nature of risks that APIs can pose to the aquatic environment
-Identify and evaluate efficient methods to reduce emission of APIs from wastewater treatment plants
-Generate new methods to identify problematic APIs at an early stage during drug development. (MistraPharma homepage)
7 This master thesis is a contribution to the first goal of the MistraPharma program and aims to give a first indication on the toxicity of several pharmaceuticals in Rainbow trout (Oncorhynchus mykiss).
EROD activity and biological effects
The EROD (7-ethoxyresorufin O-deethylase) activity is a commonly used biomarker to determine aryl hydrocarbon receptor (AhR) agonist exposure. The uptake of AhR agonists in fish is reflected by the rate of deethylation of the substrate 7-ethoxyresorufin by cytochrome P450 (CYP) to give the product resorufin (figure 2).
O
N
O O
C H3
Cytochrome P450
O
N O
H O
7-Ethoxyresorufin Resorufin
Figure 2. The 7-ethoxyresorufin deethylation to resorufin catalyzed by cytochrome P450.
The AhR is a cytosolic unit bound to two chaperone proteins, hsp90 homodimer and a p37 protein, in its inactivated state. When an AhR agonist enters a cell, it binds to the AhR- chaperone complex, which translocates into the nucleus. Once in the nucleus, chaperone proteins leave the complex and the AhR still bound to its ligand forms a transcription factor complex with an aryl hydrocarbon nuclear translocator protein (ARNT). This heterodimer then recognizes specific sequences on the DNA called dioxin responsive elements (DREs) and induces transcription of several genes.
Many genes are induced upon activation of the AhR, but the most studied one is CYP1A.
Activation of the receptor also leads to induction of CYP1B and CYP1C subfamilies (Hankinson, 1995; Savas et al., 1993). Even if the identity of AhR regulated genes involved in toxicity is not yet clear, CYP enzymes are proved to be important for both detoxification and formation of toxic metabolites.
In fish gills, first-pass metabolism can occur and therefore reduce the bioavailability and the biological half-life of readily metabolised pollutants (Jönsson et al., 2010). Gill EROD activity may therefore be a more sensitive biomarker than liver EROD activity for rapidly metabolized AhR agonists since the concentration of parent compound is likely to be higher in the gill than in the liver.
The rainbow trout gill filament EROD activity primarily reflects the CYP1A level and to a lower extent the CYP1B and CYP1C levels (Jönsson et al., 2010). In rainbow trout, two forms of CYP1A genes have been found by Berndston and Chen (1994), CYP1A1 and
8 CYP1A3, and one form of CYP1B (CYP1B1) as well as three forms of CYP1C (CYP1C1, CYP1C2 and CYP1C3) have been found by Jönsson et al. (2010).
Aims
Project 1
The first part of my thesis work has been to implement a new in vitro exposure method to study the induction of CYP in fish gills. Such an in vitro method is expected to reduce the amount of pharmaceutical needed for exposure compared with amounts used in vivo. Several temperatures, exposure times and solvents were tested in order to optimize the reaction and therefore the EROD response.
Project 2
The second part of this thesis aimed to test the EROD induction and inhibition potency of the pharmaceuticals listed above. In the inhibition study the in vivo EROD assay already published by Jönsson et. al. (2002) was used. The induction study was performed using the in vitro EROD assay developed in project 1.
Furthermore, the effects of indigo and miconazole on the mRNA expression of several CYP genes were studied. Indigo is a potent inducer of CYP1A. However it has been shown that it is also rapidly metabolized via first-pass metabolism in the gills resulting in its elimination.
(Jönsson et al., 2006). Miconazole is proved to be a strong CYP inhibitor in stickleback gills (Beijer et al., 2010) and therefore co-treatment with miconazole might decrease the metabolization/excretion of indigo and increase its effects at low concentration.
Material and Method
The fish
The rainbow trout (Oncorhynchus mykiss) (figure 3) were obtained from Näs fiskodling AB, By Kyrkby, Sweden, and kept in the aquarium facility of the Department of Organismal Biology, Uppsala University. The fish used for each experiment were killed before their daily feeding with pellets (Dan-ex 1352) from Dana Feed A/S (Horsens, Denmark). Their weight was 125 ± 26 g (mean weight ± standard deviation of the mean; SD).
9
Figure 3. Rainbow trout (Oncorhynchus mykiss)
The pharmaceuticals
Fenofibrate (CAS No 49562-28-9, 99%), Chlorpromazine hydrochloride (CAS No 69-09-0,
>98%), Haloperidol (CAS No 52-86-8), Fluvoxamine maleate (CAS No 61718-82-9, solid), Sertraline hydrochloride (CAS No 79559-97-0, ≥98%), Oxazepam (CAS No 604-75-1), Miconazole nitrate (CAS No 22832-87-7, >98%) and Indigo (CAS No 482-89-3, synthetic, dye content 95 %) were obtained from Sigma-Aldrich (St. Louis, MO, USA). Ciprofloxacin (CAS No 85721-33-1, >98%) and Losartan potassium (CAS No 124750-99-8, ≥99.5%) were obtained from Fluka (Sigma- Aldrich, St. Louis, MO, USA). All pharmaceuticals were stored at 3˚C and protected from light.
Each pharmaceutical was dissolved in dimethyl sulfoxide (DMSO) to a final concentration of 100 mM, except for ciprofloxacine where the final concentration was 2 mM due to the compound’s low solubility in DMSO. The stock solutions were stored at room temperature and covered with aluminium foil to protect them from light. Indigo was dissolved in DSMO into two different stock solutions (20 µM and 2 µM), covered with aluminium foil and stored at -20˚C.
EROD inhibition study
Inhibition of EROD activity was studied according to the protocol for EROD determination established by Jönsson et al. (2002) and modified by Beijer et al. (2010). Six fishes were exposed for 24 h in transparent polyethylene bags disposed into boxes (45 cm x 30 cm x 15 cm). The bags were filled with 20 L of Uppsala tap water and β-naphthoflavone (BNF) dissolved in acetone (50 mM) in a final concentration of 1 µM. A flow of tap water was running continuously around the boxes to maintain the temperature at 14°C.
10 The fish were then anaesthetized with benzocain, measured, weighed and killed by decapitation. The gill arches were cut off and stored in HEPES-Cortland buffer (HC buffer – see appendix). For each fish, duplicate groups of ten 2 mm long filament tips were placed in a 12-well plate containing HC buffer. The stock solution for each pharmaceutical was diluted with HC buffer to a final concentration of 50 µM - except for ciprofloxacine for which the final concentration was 10 µM. After removal of the HC buffer, the wells were replenished with 0.5 mL of HC buffer with pharmaceutical and the plate was covered with aluminium paper and left at room temperature for 110 minutes for the pre-exposure step. A solution of each pharmaceutical in a final concentration of 50 µM (except for ciprofloxacin = 10 µM) in reaction buffer (see appendix) was prepared. The pre-exposure solution was then replaced with 0.5 mL of reaction buffer containing pharmaceutical, and the plates were pre-incubated for 10 min under constant shaking. After 10 min the buffer was removed and replaced with 0.7 ml of fresh reaction buffer with the pharmaceutical. Samples of 200 µL were taken after 30 min and 40 min of incubation and transferred to a 96-well plate. Standard solutions of resorufin (Sigma-Aldrich, St. Louis, MO, USA) from 0 to 250 nM in reaction buffer were also sampled. Fluorescence was assessed using a mutliwell-plate reader (Victor 3, PerkinElmer, Boston, MA, USA) using 544 nm as the excitation wavelength and 590 nm as the emission wavelength. The EROD activity was then expressed in pmol resorufin/filament tip/min.
Figure 4. Preparation of gill filament tips.
EROD induction study
Development of the method
This in vitro method is an alternative to the protocol established by Jönsson et al. (2002) in order to use less chemical for each experiment while keeping a pretty high concentration of the compound.
11 The fish were removed from the tank in the aquarium facility in the department before their daily feeding. They were anaesthetized in benzocain solution, measured, weighed and killed by decapitation in the same way as in the inhibition study. The gill arches were excised and stored in HC buffer. For each fish, duplicate groups of ten 2 mm long filament tips were placed in a 12-well plate containing HC buffer. When optimizing the method in terms of exposure time and temperature, BNF (50 mM in acetone) in a final concentration of 1 µM was used as an inducer. Five different solvents were tested to compare their effects on the control activity: ethanol, acetone, DMSO, HC buffer and distilled water. The HC buffer was then removed from the well and replaced by 3 mL of the BNF solution or buffer containing the solvent (distilled water was tested without any HC buffer). Different exposure times and temperatures were tested: 3 h, 4.5 h, 6 h, 9 h, and 17 h; 8°C, 16°C, 19°C, and 22°C. After exposure in vitro the method proceeded the same way as after exposure of fish in vivo (Jönsson et al., 2002) and as described above for the inhibition study.
The exposure
The induction study was done using the experimental conditions determined above. The selected experimental time and temperature were 6 h and 19°C. The selected solvent for the pharmaceuticals was DMSO.
Expression of mRNA
Exposure
Exposure of gills to indigo was done following the protocol described above. Based on the results from EROD induction with different doses of indigo, two concentrations were selected for this study. Indigo at a concentration of 1 nM was combined with miconazole and 10 nM indigo was used to give a maximal response. Three different solutions were therefore prepared: the mixed solution was prepared with a final concentration of 50 µM of miconazole and 1 nM of indigo in HC buffer (0.1 % DMSO), the solution supposed to give maximal response was prepared with a final concentration of 10 nM indigo in HC buffer (0.05 % DMSO) and the negative control was prepared with 0.1% DMSO to make sure that DMSO did not influence the final result. Three mL of the different solutions were added to the different wells with 10 gill filament tips and the plates were placed in an incubator at 19°C for 6 h. At the end of the incubation time, the gill filament tips were collected from each well into Eppendorf tubes – duplicates from the same fish together in the same tube. After centrifugation (30 sec, 12 000 x g at 4°C), the HC buffer was removed from the tubes and the tubes were instantly frozen in liquid nitrogen. The samples were stored at -80°C.
Extraction of mRNA and cDNA synthesis
RNA was isolated from the samples using the Aurum total RNA fatty and fibrous tissue kit according to the manufacturer’s instructions (Catalog #732-6830, Bio-Rad, Laboratories, CA, US). Because of the small amount of material, the purified samples were eluted only 1 time
12 with only 30 µL elution buffer. The concentration of newly extracted mRNA and its purity were analysed with a NanoDrop 2000c spectrophotometer (Saveen & Werner AB, Malmö, Sweden). A standard volume of 1 µg of mRNA for each sample was reversely transcribed into cDNA using the iScript cDNA Synthesis kit (Catalog #170-8891, Bio-Rad). cDNAs were then diluted 20 times with nuclease-free water and stored at -20˚C.
Quantitative real-time PCR
Quantitative real-time PCR (qPCR) reactions were conducted on a Rotor-Gene 6000 (Corbett Research, Sydney, Australia). The reagents were acquired from Bio-Rad. The reactions were carried out using the iQ SYBR Green Supermix kit (Catalog #170-8885, Bio-Rad) in a final volume of 20 µl. The thermal profile for the SYBR green reactions was as described in the following protocol: 95˚C for 10 min, followed by 40 cycles of 95˚C for 15 s and 62˚C for 60 s. Primers used for the PCRs have already been published (Jönsson et. al., 2010).
Table 1. Sequences (5′–3′) of the gene-specific real-time PCR primers used in the experiment.
Transcript Forward primer (5´- 3´ ) Reverse primer (3´- 5´ ) GenBank Accession number
RbCYP1A1 GGAAACTAGATGAGAACGCCAACA GTACACAACAGCCCATGACAG AAB69383.1 RbCYP1A3 GAAACTAGATGAGAACGCCAACG CTGATGGTGTCAAAACCTGCC AAD45966.1 RbCYP1B1 CATTCTGATACTTGTGAGGTTTCC CAACTGAGACTGGTCTTCCAT GU325707 RbCYP1C1 GCAGCACAGAGAAACCTTCAAC GTCCTTTCCGTGCTCAATCACA GU325708
RbCYP1C2 GAGCACAGGGAGACATTTGAC GGTATCACTGTCCGCCTTG GU325709
RbCYP1C3 CATGAGTGATGCCATCATTAACGC AGGTCTGTGACTGTTCCTTCAACAA GU325710 RbEF1
salmon18S Rbβactin
GCAGGTACTACGTCACCATCAT CTCAACACGGGAAACCTCAC CTTACGGATGTCCACGTCACA
CACAATCAGCCTGAGATGTACC AGACAAATCGCTCCACCAAC CCTCGGTATGGAGTCTTGC
CF752904 AF243427 FJ975145
Elongation factor 1 (EF1 ) was selected as the reference gene because it is proved to be among the most stable genes for tissue comparison and treatment studies in fish (Jorgensen et al., 2006 and McCurley and Gallard, 2008).
Relative CYP mRNA expressions were calculated using the EΔΔCt method described by Livak and Schmittgen, (2001). PCR efficiencies for each primer were determined using the LinRegPCR program (Ruijter et al., 2009).
Statistics
One-way ANOVA with Dunnett’s post test was performed for the EROD activity in the indigo study. Paired t-test and Mann-Whitney test were performed on the EROD activity data in the study on the effects of pharmaceuticals on EROD. Kruskal-Wallis test followed by Dunn’s post test were performed on the data from the gene expression study. Data were log
13 transformed when the variances differed between groups. All tests were performed using GraphPad Prism version 5.00 for Windows, GraphPad Software, San Diego California USA, www.graphpad.com.
14
Results
Induction in gill filaments in vitro
In this study BNF was selected as a positive control because of its well-known induction of CYP1A. Moreover, induction of EROD by BNF has been studied in rainbow trout in vivo and therefore in vitro data obtained in the present study can be compared with the in vivo data. In the first experiment an exposure time to BNF of 4.5 h and three temperatures (8°C, 16°C and 22°C) were tested. EROD activity increased with increasing temperatures from 0.039 ± 0.003 pmol/filament/min at 8°C to 0.096 ± 0.021 pmol/filament/min at 22°C (0.024 ± 0.003
<control <0.031 ± 0.002) (Table 2). In the next experiment different exposure times were tested (3 h, 6 h and 17 h) at 16°C and 22°C. After 3 h and 6 h of exposure a similar pattern was observed: increasing EROD activity with increasing temperature, from 0.072 ± 0.002 to 0.116 ± 0.017 pmol/filament/min after 3 h and from 0.18 ± 0.002 to 0.231 ± 0.019 pmol/filament/min after 6 h. After 17 h a higher EROD activity was observed at 16°C than at 22°C: 0.188 ± 0.0001 and 0.130 ± 0.011 pmol/filament/min, respectively. Moreover the EROD activity was similar after 6 h and 17 h of exposure at 16°C and markedly decreased between 6 h and 17 h after exposure at 22°C. In these experiments controls were between 0.018 ± 0.002 and 0.040 ± 0.004 pmol/filament/min. Finally, exposure times of 6 h and 9 h were tested at 19°C. High EROD activities were observed after both experimental times but controls were also high (Table 2).
15
Table 2. Mean EROD activity (pmol/filament/min) obtained during the screening study with BNF for different exposure times and temperatures. For each combination time/temperature both BNF-induced EROD activity and negative control are presented (n=6).
Mean EROD activity (pmol/filament/min) ± SD
Temperatures 8˚c 16˚c 19˚c 22˚c
Times
BNF induced
control BNF induced control BNF induced control BNF induced control
3h 0.072 ± 0.002 0.034 ± 0.001 0.116 ± 0.017 0.028 ± 0.001
4.5h 0.039 ± 0.003 0.028 ± 0.002 0.074 ± 0.003 0.031 ± 0.002 0.096 ± 0.021 0.024 ± 0.003
6h 0.180 ± 0.002 0.037 ± 0.003 0.235 ± 0.025 0.118 ± 0.011 0.231 ± 0.019 0.027 ± 0.002
9h 0.257 ± 0.014 0.138 ± 0.003
17h 0.188 ± 0.0001 0.040 ± 0.004 0.130 ± 0.011 0.018 ± 0.002
16 The in vivo methodology developed by Jönsson et. al. (2002) used acetone as a solvent for BNF. However, in the development of the in vitro assay, controls were observed to be very high and therefore other solvents were tested. In a first experiment, DMSO, ethanol and acetone in HC buffer, and HC buffer only, were tested. Ethanol gave the highest basal activity (0.039 ± 0.002 pmol/filament/min), followed by acetone (0.020 ± 0.002), DMSO (0.014 ± 0.002) and HC buffer (0.014 ± 0.002) (figure 2). Another experiment was conducted to confirm the results with acetone, DMSO, HC buffer and distilled water. The values were different than in the first experiment, but the same pattern was observed. Addition of acetone resulted in the highest basal activity (0.041 ± 0.006), followed by DMSO (0.037 ± 0.003) and HC buffer (0.027 ± 0.004). Gills exposed to distilled water instead of HC buffer were colourless and denatured after 6 h of exposure and did not show any EROD activity. In the two experiments where DMSO in HC buffer was compared with HC buffer only, no statistically significant difference was obtained (figure 5).
EtOH Acetone
Acetone(2) DMS
O DMS
O(2) HC
buffer HC
buffer(2) O dH2
0.00 0.02 0.04 0.06
**
** ns
ns ++
Solvents
EROD activity (pmol/filament/min)
Figure 5. Gill EROD activity (pmol/filament/min, mean ± SD, n=6) in rainbow trout (Oncorhynchus mykiss) exposed to ethanol, acetone (two times) or DMSO (two times) in HC buffer and to HC buffer and distilled water without solvent. Gills were exposed for 6 h at 19°C. Significant differences were examined by a paired t-test and Mann-Whitney test (** or ++ P < 0.01; *compared to HC buffer (1); + compared to HC buffer (2)).
17
EROD inhibition by the pharmaceuticals
Five different experiments were performed in this part: each time two pharmaceuticals were tested versus the positive control. Statistical analysis was therefore performed for each experiment: the EROD activity when pharmaceutical was added versus the positive control performed on the same day. The five BNF-induced controls (PCs) showed EROD activities of 0.324 (PC 1), 0.293 (PC 2), 0.349 (PC 3), 0.352 (PC 4), and 0.330 pmol/filament tip/min (PC 5). Oxazepam, fenofibrate and losartan did not cause any statistically significant differences compared to their respective BNF-induced controls (PC 1, 3 and 4). Ciprofloxacin showed a rather weak inhibition of the BNF-induced EROD activity (P < 0.01) compared to its BNF- induced control (PC 2) with a mean EROD activity of 0.230 pmol/filament tip/min.
Haloperidol, chlorpromazine, fluvoxamine and sertraline caused more significant inhibitions of the BNF-induced EROD activity compared to their respective BNF-induced controls (PC 3, 2 and 5) with respective activities of 0.182, 0.111, 0.129 and 0.147 pmol/filament tip/min.
Finally miconazole showed a very strong inhibition of the BNF-induced EROD activity with an activity of 0.011 pmol/filament tip/min.
Negative control Positive control 1
Oxazepam Miconazole Positive control 2
Ciprofloxacine Chlorprom
azine
Positive control 3 Haloperidol
Fenofibrate Positive control 4
Losartan Positive control 5
Fluvoxam ine Sertraline 0.0
0.1 0.2 0.3 0.4 0.5
ns
****
**
***
***
ns ns
******
Treatments
EROD activity (pmol/filament/min)
Figure 6. EROD activity (pmol/filament tip/min, mean ± SD, n=6) in rainbow trout (Oncorhynchus mykiss) gill filament tips exposed to 50 µM of different pharmaceuticals (oxazepam, miconazole, ciprofloxacin (10 µM due to low solubility), chlorpromazine, haloperidol, fenofibrate, losartan, fluvoxamine and sertraline). Induction with β-naphthoflavone was performed in vivo for 24 h and inhibition of the catalytic activity was studied in vitro at 19°C. Significant differences were examined by a paired t-test or a Mann-Whitney test (* P < 0.05, ** P <
0.01, *** P < 0.001, **** P < 0.0001).
18
EROD induction by the pharmaceuticals
Six different experiments were performed in this part: each time one or two pharmaceuticals were tested versus the negative control. In the same way as for the inhibition study, statistical analysis was therefore performed for each experiment: the EROD activity after exposure to a pharmaceutical versus the negative control (NC) of the same day. Only oxazepam showed a statistically significant (P < 0.05) induction (0.018 pmol/filament tip/min) versus its negative control (NC 2) (figure 7). In most of the experiments, the negative control showed the highest EROD activity (figure 7). Exposure to miconazole, chlorpromazine, fenofibrate, haloperidol fluvoxamine and sertraline resulted in statistically lower EROD activity than in the HC buffer control (negative control). EROD activity after ciprofloxacin and losartan exposure were not significantly different compared to the activity of their respective negative control with respective activities of 0.022 versus 0.018 pmol/filament tip/min (NC 3) and 0.017 versus 0.017 pmol/filament tip/min (NC 5).
Positive control Negative control 1
Mic onazole
Negative control 2 Oxazepam Chlorproma
zine
Negative control 3 Fenofibrate
Ciprofloxacine Negative control 4
Haloperidol Negative control 5
Losartan
Negative control 6 Fluvoxamine
Sertraline 0.00
0.02 0.04 0.06 0.08 0.10 0.15 0.20 0.25
*
**** ***
ns
**
ns ***
**** ****
Treatments
EROD activity (pmol/filament/min)
Figure 7. EROD activity (pmol/filament tip/min, mean ± SD, n=6) in rainbow trout (Oncorhynchus mykiss) gill filament tips exposed in vitro to different pharmaceuticals at a concentration of 50 µM (oxazepam, miconazole, ciprofloxacine (10 µM due to low solubility), chlorpromazine, haloperidol, fenofibrate, losartan, fluvoxamine and sertraline). The exposure lasted for 6 h at 19°C. Significant differences were examined by a paired t-test or a Mann-Whitney test (* P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001).
19
Effects of indigo on EROD activity and gene expression
Two screening experiments were performed to analyse the EROD induction potency of indigo using the in vitro assay. In the first experiment, 0.1, 1 and 10 nM concentrations of indigo were tested. Exposure to 0.1 and 1 nM did not cause any statistically significant difference compared to the negative control (figure 8). A concentration of 10 nM indigo induced an EROD activity more than three times higher than that of NC 1 (0.080 versus 0.029 pmol/filament tip/min). In another experiment 50 nM indigo showed similar induction: 0.081 versus 0.033 pmol/filament tip/min (NC 2).
Indigo concentrations (nM)
EROD activity (pmol/filament/min)
Negative control 0.1nM
1nM 10nM
Negative control (2) 10nM
(2) 50nM 0.00
0.05 0.10 0.15
ns ns
*** *** ***
Figure 8. EROD activity (pmol/filament tip/min, mean ± SD, n=6) in rainbow trout (Oncorhynchus mykiss) gill filament tips exposed in vitro to different concentrations of indigo (0.1 nM, 1 nM, 10 nM and 50 nM). The exposure lasted for 6 h at 19°C. Significant differences between negative control and test concentrations within the same study were examined by one-way ANOVA and Dunnett’s post test (*** P < 0.001).
20 The expression of mRNA was determined for three reference genes and six test genes. The qPCR profile for the suggested reference gene Salmon18S was very strange indicating a potential method error. Both RbEF1 and Rbβactin have been shown to be useful as reference genes but RbEF1 was used for the calculation because it is known to be the most stable one. The mixture of 1 nM indigo and 50 µM miconazole caused highest induction of the three CYP1C genes (around 6-fold compared to the unexposed control) and a statistically significant lower induction of the two CYP1A genes (about 2-fold) (figure 9). It did not show any statistically significant induction of CYP1B1 mRNA, but the high variation made the observation not conclusive. Indigo at 10 nM only caused statistically significant induction of CYP1A3 mRNA and CYP1C3 mRNA (about 1.5-fold and 2-fold respectively) (figure 9).
However, since the standard deviation of CYP1B1 was extremely big it was not possible to exclude induction of this gene as well (figure 9).
CYP1A1
CYP1A3
CYP1B1
CYP1C1
CYP1C2
CYP1C3
EF1 0
2 4 6 8 10
Control
Indigo 1nM + Miconazole 50 M Indigo 10nM
*ns **
*
*
**
ns ns
ns
ns
**
***
CYP1A genes
fold induction
Figure 9. Relative transcript level of RbCYP1A1, RbCYP1A3, RbCYP1B1, RbCYP1C1, RbCYP1C2 and RbCYP1C3 in gills of rainbow trout (Oncorhynchus mykiss) exposed in vitro to either a mixture of 1 nM indigo and 50 µM miconazole or to 10 nM indigo. Calculations were made using EF1 as the reference gene.
Statistically significant differences compared with the unexposed control were examined by Kruskal-Wallis test followed by Dunns post test (*P < 0.05, **P < 0.01 and ***P < 0.001), n=6.
Discussion
Pharmaceutical products are a rising threat to aquatic environments. Consumption keeps increasing while specific removal treatments in wastewater treatment plants still do not exist.
Most of the pharmaceuticals are hydrophilic (Fent et al., 2005) and are therefore found in water where they can diffuse into aquatic organisms for example by uptake via the gills. It is therefore important to developed analytical methods to assess the extent of the potential toxicity of these chemicals to organisms.
The new method
Different exposure times and temperatures were first tested to optimize the in vitro exposure method. The goal of the screening experiment was to determine the optimal experimental conditions to get an EROD activity similar to the one obtained after using the ex vivo assay developed by Jönsson et al. (2002). When using this method, exposure to 1 µM BNF has resulted in EROD activities of 0.3 pmol resorufin/filament tip/min in rainbow trout (Jönsson et al., 2002). In my experiments the activities after exposure to 1 µM BNF in vivo were 0.17, 0.094, 0.13 and 0.045 pmol resorufin/filament tip/min (unpublished results).
After 4.5 h of in vitro exposure at 8°C, EROD activity in BNF-exposed gills was similar to the activity of non-exposed gills (table 2). This indicates that at this low temperature, synthesis of mRNA and/or protein was too low to induce any reaction. At 16°C, EROD activity was increasing with increasing exposure time until reaching a similar activity between 9 h and 17 h of exposure (table 2). Since the time between 9 and 17 h is long, a similar EROD activity at these time points might also mean that the activity peak is between 9 and 17 h. The EROD activity in relation to time of exposure at 22°C showed a clear decrease after 17 h of exposure, probably as a result of degradation of CYP1A (table 2). A temperature of 19°C was finally selected as the exposure temperature for the in vitro assay giving the highest EROD activity. In order to combine practical reasons and optimization of the method, 6 h of exposure was selected as the exposure time.
In the ex vivo assay developed by Jönsson et al. (2002), BNF was dissolved in acetone.
Therefore negative controls were prepared by diluting acetone in HC buffer to determine the background activity caused by the solvent. However while doing this during the screening study of the in vitro method, the negative control activities were found to be around ten times higher than in ex vivo studies. In Jönsson et al. (2002) the negative control showed a mean activity of 0.004 pmol resorufin/filament tip/min, and in Jönsson et al. (2006) the activity of the control was 0.0011 pmol resorufin/filament tip/min. The activity of a negative control in the ex vivo assay from my own experiments was 0.0031 pmol resorufin/filament tip/min (unpublished results). In the screening experiment the lowest negative control activity found was 0.018 ± 0.002 pmol resorufin/filament tip/min and the highest was 0.138 ± 0.003 pmol resorufin/filament tip/min (table 2). Jönsson et al. (2010) found that different solvents do not have any effect on the CYP1 mRNA expression. To reduce negative control activities in the
22 present study, different solvents were tested to replace acetone. DMSO proved to be the one causing the lowest background activity of all - similar to HC buffer without added solvent (figure 5). However the negative control activity was still around ten times higher than that of the negative controls of ex vivo studies. The final concentration of the solvent was adjusted in the in vitro assay to be the same as the concentration in the ex vivo assay, and therefore the negative control activity was expected to be similar. Differences could be explained by the fact that in the ex vivo study the control fish are exposed to the solvent for 24 h in a large volume of fresh water while in the in vitro assay, gill filament tips are directly exposed to the solvent for 6 h in the wells. However gill EROD activities of BNF-exposed fish were similar when the fish were exposed in vivo and when gill filaments were exposed in vitro.
EROD inhibition/induction by the pharmaceuticals
In the inhibition study miconazole showed the strongest effect on the EROD activity with the activity reduced by 30 times from the positive control value (figure 6). Beijer et al. (2010) found similar EROD inhibition with miconazole in the three-spined stickleback, and at a concentration of 1 µ M the activity decreased from 0.045 pmol resorufin/filament/min for the positive control to an activity of 0.013 pmol resorufin/filament/min for the exposed group.
Very strong inhibition was also observed after exposure to chlorpromazine in the present study and strong inhibitions after exposure to (increasing potency): ciprofloxacin <
haloperidol sertraline fluvoxamine (figure 6).
Apart from oxazepam, none of the tested pharmaceuticals showed a significant induction of EROD activity (figure 7). Oxazepam induced CYP2B1 and B2 in rats and mice, increasing liver size and inducing hepatic tumors in a 2-year exposure study (McKim Jr et al., 1999). In this study we have shown that in rainbow trout it induces CYP1A. However the induction was so low that further studies have to be performed to confirm the result. Moreover, in this study negative controls were high, showing actvities up to 0.04 pmol resorufin/filament tip/min.
Most of the test pharmaceuticals were actually showing significantly decreased EROD activities compared with their own negative control. That is the case for miconazole, chlorpromazine, fenofibrate, haloperidol, fluvoxamine and sertraline (figure 7). Those pharmaceuticals are also the ones showing strongest inhibitions in the ex vivo EROD inhibition study, except for fenofibrate which was not showing any inhibition effect in that study (figure 6). After the 6 h of exposure to the pharmaceutical, the buffer was exchanged a couple of times by reaction buffer. Still some of the compound may be present in the gill tissue, possibly binding irreversibly to CYP1A, resulting in inhibition of basal EROD activity.
If this theory is true, then it is possible to say that ciprofloxacin does not bind to gill tissues.
In fact it was shown to be a weak inhibitor in the inhibition study (figure 6) but it did not show any activity in the induction study (figure 7). This induction study can therefore be used to confirm the inhibition potency of the pharmaceutical demonstrated earlier. However, it is hard to explain why fenofibrate seems to be a CYP inhibitor within the induction study (figure 7), while it was not shown to be one in the inhibition study (figure 6).
23
The indigo/miconazole interaction
As explained earlier in this report, CYP enzymes in the gills of the fish can catalyze first-pass metabolism, ensuring a decrease in systemic bioavailability of rapidly metabolised waterborne pollutants (Jönsson et al,. 2006). Jönsson et al. (2006) demonstrated that this is the case for indigo, which is a very potent CYP inducer. Presence of waterborne CYP inhibiting pharmaceuticals (and pollutants in general) may increase its bioavailability and thus increase its potential toxic effects. Miconazole has been proved to be a very strong CYP inhibitor and it was therefore used in this study combined with indigo (figure 6). In this experiment our hypothesis was that co-exposure of indigo and miconazole will potentiate the inductive activity of the dyeing agent.
In the EROD induction study with different concentrations of indigo, 10 nM indigo was shown to cause maximal induction and 1 nM indigo showed a low EROD induction (figure 8). The EROD activity after in vitro exposure to 1 nM indigo is similar to that obtained by Gao et al. (2011) after in vivo exposure in three-spined stickleback. For the mRNA study, concentrations of 10 nM indigo and 1 nM indigo/50 µM miconazole were therefore selected.
The hypothesis seems to have been confirmed by the results concerning the induction of the three CYP1C genes: 4-fold to more than 6-fold higher activity was found for the mixture compared to the unexposed control (figure 9). However, unlike what was expected, at a ten times higher concentration, indigo alone did not significantly induce transcription of any CYP genes, except for approximately two-fold induction of CYP1A3 and CYP1C3. It is possible that 6 h exposure time is too long and already leads to the CYP-induced metabolisation of the compound. However, Gao et al. (2011) found that after 6 h of in vivo exposure to 1 nM indigo in three-spined stickleback the CYP1A gene was very highly induced in the gills (200-fold compared to the unexposed control). This difference in induction of mRNA expression between the two studies might be the result of species differences as well as a result of the different routes of exposure in vivo and in vitro.
24
Conclusion
In this thesis we established a new in vitro experimental method to study the induction by pharmaceuticals of the catalytic activity of CYP1 in gill tissue in rainbow trout. The method gives slightly less sensitive results compared to the in vivo assay developed by Jönsson et al.
(2002). However there are several advantages of the new method - reduction of cost since very little of the test compound is used, reduction of waste, and lower number of test animals needed. More pharmaceuticals have to be tested with both in vivo and in vitro methods in order to establish a comparison scale.
Among the nine chemicals tested regarding their inhibition and induction properties, six showed to be inhibitors. Among these, ciprofloxacine as a broad-spectrum antibiotic is found in very high concentration, sometimes even above the human therapeutic plasma level, in effluent water (Larsson et al., 2007; Brown et al., 2006). The study on co-exposure of a well- proved CYP1 inhibitor, miconazole, and a well-known CYP1 inducer that is readily metabolised, indigo, showed that the hypothesis concerning the potentiation of the effect of indigo by an inhibitor may be right. However, it is intriguing that the CYP1C mRNAs were much more strongly induced than the CYP1A mRNAs by the mixture. Even though more experimental studies have to be performed to confirm this results, this illustrates that CYP1 inhibitors, like certain pharmaceuticals, may influence the disposition and effects of other contaminants in fish in the recipients.
25
Acknowledgements
First of all I would like to thank my supervisor Björn Brunström for his help, advices and patience. Thank you very much for being there for the presentation of this thesis. I’d also like to give a huge tack så mycket to Tina Beijer for being the best PhD supervisor ever as well as a very good friend. Thank you also to Ingvar Brandt for welcoming me in your research group, and for your advices. Thank you Jan Örberg for your support when I decided to stay in Uppsala to do my Master degree. Thank you Maria and Kai for your constant attention and patience to answer my questions and for sharing your fish with me. Greetings to “Maria- Greece”: I hope this report will help you writing yours!! Finally I would like to thank everybody at the Department for being in turn great classmates, colleagues, teachers and even the three at the same time!
Mes derniers mots vont à ma famille: merci pour votre aide et soutien quand l’hiver se faisait bien long.
