Albendazole causes stage-dependent developmental toxicity and is deactivated by a mammalian metabolization system in a modified zebrafish embryotoxicity test

Reproductive Toxicology 34 (2012) 31–42

Contents lists available at SciVerse ScienceDirect

Reproductive Toxicology journal homepage: www.elsevier.com/locate/reprotox

Albendazole causes stage-dependent developmental toxicity and is deactivated by a mammalian metabolization system in a modified zebrafish embryotoxicity test Anna Mattsson a,∗ , Erik Ullerås b , Johan Patring c , Agneta Oskarsson a a

Swedish University of Agricultural Sciences, Department of Biomedicine and Veterinary Public Health, Box 7028, SE-75007, Uppsala, Sweden Department of Immunology, Genetics and Pathology, Science for Life Laboratory, Uppsala University, Rudbeck laboratory, SE-75185 Uppsala, Sweden c Swedish University of Agricultural Sciences, Department of Aquatic Sciences and Assessment, Box 7082, SE-75007 Uppsala, Sweden b

a r t i c l e

i n f o

Article history: Received 1 December 2011 Received in revised form 24 January 2012 Accepted 25 February 2012 Available online 5 March 2012 Keywords: Zebrafish Metabolism Developmental toxicity Albendazole Malformations

a b s t r a c t The zebrafish embryotoxicity test has previously been combined with an external metabolic activation system (MAS) to assess developmental toxicity of metabolites produced by maternal metabolism. Due to toxicity of MAS the exposure was limited to one early and short period. We have modified the method and included additional testing time points with extended exposure durations. Using the anthelmintic drug albendazole as a model substance, we demonstrated stage-dependent toxic effects at three windows of zebrafish embryo development, i.e. 2–3, 12–14 and 24–28 h post fertilization, and showed that MAS, by metabolic deactivation, reduced the toxicity of albendazole at all time points. Chemical analysis confirmed that albendazole was efficiently metabolized by MAS to the corresponding sulfoxide and sulfone, which are non-toxic to zebrafish embryos. To conclude, the modified zebrafish embryotoxicity test with MAS can be expanded for assessment of metabolites at different developmental stages. © 2012 Elsevier Inc. All rights reserved.

1. Introduction Development is a delicately coordinated process that may be disrupted by exposure to chemical substances. An increasingly popular model organism for assessing developmental toxicity is the zebrafish (Danio rerio) embryo, which has the convenience of cell-culture studies (e.g. fast, cost-effective assays amendable to high-throughput screening) but contrary to cell cultures it has the advantages of animal studies as it is a complete vertebrate embryo that can be studied regarding numerous relevant endpoints. Zebrafish embryos are transparent and develop externally, which greatly facilitates chemical exposure, manipulation and evaluation of numerous morphological, developmental and behavioral endpoints in the intact living embryo. The zebrafish

Abbreviations: ABZ, albendazole; ABZSO, albendazole sulfoxide; ABZSO2 , albendazole sulfone; ABZSO2 NH2 , albendazole-2-aminosulfone; dpf, days post fertilization; hpf, hours post fertilization; MAS, metabolic activation system; mDarT, zebrafish embryotoxicity test with MAS; NADPH, reduced ␤-nicotinamide adenine dinucleotide phosphate; ␤-NF, ␤-naphtoflavone; DMSO, dimethyl sulfoxide; PB, phenobarbital. ∗ Corresponding author. Tel.: +46 18 671702; fax: +46 18 673532; mobile: +46 735105796. E-mail addresses: [email protected] (A. Mattsson), [email protected] (E. Ullerås), [email protected] (J. Patring), [email protected] (A. Oskarsson). 0890-6238/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.reprotox.2012.02.007

embryo model has been evaluated regarding its ability to predict the teratogenic potential of chemicals in mammals. Using a 48-h D. rerio toxicity/teratogenicity test (DarT), 88% of 41 tested compounds were found to be in agreement with findings from developmental toxicity tests in mammals [1,2]. A similarly high concordance was obtained in two more recent studies where 87% of 31 tested compounds in a 5-day assay [3] and 81% of 27 tested compounds in a 6-day assay [4] were correctly categorized concerning the embryotoxic potential. In contrast, Van den Bulck and colleagues recently reported a concordance of only 60% among 15 tested compounds. Of the misclassified compounds, four were false positives and two were false negatives [5]. The overall concordance in a study may depend on factors such as study design, bioavailability (placenta, chorion, etc.), the physico-chemical properties of the test compounds and the mammalian tests to which the results were compared. Furthermore, kinetic factors such as metabolism of the test compound may in certain cases explain the discordance between species. In the zebrafish embryotoxicity model, the exposure occurs directly via the ambient medium, and consequently the maternal metabolism of the test substance is not taken into account [6]. Endogenous metabolism by the zebrafish embryo itself is not much investigated and may be insignificant in a test situation and/or qualitatively different from that of humans. The zebrafish embryo test may thus underestimate the teratogenic potency of a substance that would be metabolically activated in a pregnant woman. Conversely, the potency may be overestimated

32

A. Mattsson et al. / Reproductive Toxicology 34 (2012) 31–42

Fig. 1. Albendazole (ABZ) and its major metabolites. Albendazole is sequentially metabolized to albendazole sulfoxide (ABZSO), albendazole sulfone (ABZSO2 ) and albendazole-2-aminosulfone (ABZSO2 NH2 ) [16]. FMO: flavin-containing monooxygenase, CYP: cyotochrome P450 monooxygenase.

for a substance that would undergo maternal metabolic deactivation in humans. Busquet et al. recently presented a zebrafish embryotoxicity model, called mDarT, in which they incorporated an external mammalian metabolic activation system (MAS) that would mimic maternal metabolism [7]. The MAS was based on liver microsomes from rats and contains various xenobiotica-metabolizing flavin-containing monooxygenases (FMO) and cytochrome P450 (CYP) monooxygenases. The authors demonstrated metabolic activation of the pro-teratogenic compounds ethanol, cyclophosphamide and acetaminophen in the mDarT assay [7,8]. Zebrafish embryos were exposed together in a vial containing the test substance and MAS for 1 h starting at 2 h post fertilization (hpf). The short exposure was chosen to avoid the toxicity of MAS itself. A major drawback of such a limited exposure window is that susceptible developmental processes that occur during other windows of development will not be considered, resulting in false negative results. One of Wilson’s six principles of teratology stresses that susceptibility to developmental toxicants varies with the developmental stage at the time of exposure [9], and thus it would be a great advantage to enable use of the metabolizing system at several time points during development. Albendazole (ABZ) is a benzimidazole methylcarbamate, which is used as an anthelmintic drug to control gastrointestinal parasites in humans and domestic animals. Administration of ABZ during gestation has been shown to cause embryotoxic effects in cattle, rat, rabbit and sheep [10]. Observed effects include increase of resorptions, decreased fetal weight and increase of teratogenic effects, such as vascular, craniofacial, skeletal and external malformations [11–15]. The metabolism of ABZ is similar in humans, rat, mice, cattle and sheep [10]. In mammals, ABZ undergoes a rapid and extensive first-pass oxidation to the pharmacologically active metabolite albendazole sulfoxide (ABZSO), which is sequentially metabolized to the inactive albendazole sulfone (ABZSO2 ) and albendazole-2

-aminosulfone (ABZSO2 NH2 ) [16] (see Fig. 1). The formation of ABZSO has been shown to depend on both the CYP and the FMO system in the studied species [17–19]. CYP2C6, CYP3A1/2 and CYP2A1 are suggested to be involved in sulfoxidation of ABZ in rat liver microsomes [18]. There is conflicting evidence of the embryotoxic potential of ABZ. Due to the rapid metabolism, the unmetabolized form of ABZ has either not been identified or only found at low concentrations in plasma after treatment of animals or humans. Thus, the in vivo pharmacological and embryotoxic action of ABZ in mammals are generally assumed to be related to ABZSO [10]. ABZSO, but no other ABZ metabolite, produced the embryotoxic effects related to ABZ treatment in rats [20–22]. The association between developmental toxicity and metabolite disposition has been studied in rats 12 h following oral administration of the ABZ-forming pro-drug netobimin on gestation day 10 [13]. The authors found a significant correlation between rate of malformed embryos and concentrations of ABZ and ABZSO, but not ABZSO2 , in embryonic tissue. The lowest teratogenic dose tested resulted in ABZSO concentrations of ∼1.5 ␮g/mL in maternal plasma and ∼1.5 ␮g/g in embryo tissue at the termination of the experiment. Although the corresponding ABZ concentrations were substantially lower (∼0.06 ␮g/g in embryo tissue) it could not be excluded that ABZ contributed to the observed embryotoxicity. However, co-administration of ABZ and an inhibitor of oxidative metabolism from day 8–15 of gestation nearly eliminated the toxic effects of ABZ in rat [20], suggesting that the metabolite rather than the parent compound is the cause of developmental toxicity. In contrast to the conclusions from the above mentioned in vivo studies, there are results from in vitro models suggesting that ABZ is a more potent toxicant than ABZSO, e.g. regarding inhibition of cell proliferation in micromass cell cultures of rat embryo midbrain and limb bud cells [23] and in a human hepatoma cell line lacking ABZmetabolizing capacity [24]. The in vitro studies are in accordance with results from our laboratory, which show that only ABZ and not the metabolites ABZSO, ABZSO2 and ABZSO2 NH2 is embryotoxic in zebrafish embryos after continuous exposure from 2 hpf [25]. The reason for the discrepancy between the mammalian in vivo studies on the one hand and the zebrafish studies together with the in vitro studies on the other hand remains to be clarified. Nevertheless, it is clear that only ABZ, and not its mammalian metabolites, shows toxicity in zebrafish embryos and ABZ may thus be a useful model substance to study effects of metabolic transformation by the external mammalian metabolization system utilized in the mDarT method. The aim of the current work was to modify and optimize the mDarT conditions to enable exposure of zebrafish embryos together with a mammalian metabolizing system for extended durations and at several developmental stages, i.e. at 2–3, 12–14 and 24–28 hpf. To improve the possibility to detect late and more subtle effects, we included additional endpoints of developmental toxicity and assessed these at several time points up to six days post fertilization. The exposure was also changed from grouped exposure to individual exposure in a microplate format, which renders each embryo a replicate. The modified mDarT method was evaluated using ABZ as a model substance which, according to expectations, was efficiently metabolized and showed reduced toxicity in zebrafish embryos when tested together with the metabolization system. Formation of metabolites was confirmed by chemical analysis of the exposure medium. 2. Material and methods 2.1. Materials Albendazole (ABZ, CAS# 54965-21-8, purity 99.0%) was obtained from Dr. Ehrensdorfer GmbH, Germany. Dimethyl sulfoxide (DMSO) and reduced ␤nicotinamide adenine dinucleotide phosphate (NADPH) were purchased from

A. Mattsson et al. / Reproductive Toxicology 34 (2012) 31–42 Sigma–Aldrich AB, Sweden. Tris was purchased from Merck KGaA, Darmstadt, Germany. Tricaine methane sulfonate powder (MS222) was from Pharmaq Ltd., Fordingbridge, United Kingdom. Rat liver microsomes were purchased from XenoTech LLC, Lenexa, KS, USA (lot no. 0310131). The microsomes were from rats treated with ␤-naphtoflavone (␤-NF) and phenobarbital (PB), which induce various CYPs. 2.2. Preparation of solutions Embryo medium (standardized water, ISO 1996) was prepared fresh by dissolving salts in deionized water (CaCl2 × 2H2 O, 117.6 mg/L; MgSO4 × 7H2 O, 49.3 mg/L; NaHCO3 , 25.9 mg/L; KCL, 2.3 mg/L) [26], and was aerated for at least half an hour. Tris buffer was prepared by dissolving Tris in embryo medium to a final concentration of 0.1 M and adjusting the pH to 7.6 with HCl at 26 ◦ C (results in a pH of ∼7.4 at 32 ◦ C). Stock solutions of ABZ were prepared in DMSO at concentrations 1000 times the final exposure concentrations and were stored at −18 ◦ C until use. Final concentration of DMSO in exposure solutions was 0.1%, which is well tolerated by the zebrafish embryos in this test system. Stock solutions of 4 mg/mL Tris-buffered tricaine (pH 7) was prepared in advance and stored in aliquots at −18 ◦ C. 2.3. Test animals and egg collection A breeding stock of adult wild-type AB strain zebrafish (D. rerio; obtained from Department of Medical Biochemistry and Biophysics at Karolinska Institute, Stockholm, Sweden) was housed in 60 L-aquariums with a flow-through system of charcoal-filtered tap water. Fish were kept in an environmentally controlled room, with a 12:12 h light:dark cycle and a temperature of 26 ◦ C, and were fed twice a day with Sera® Vipan flakes supplemented with frozen artemia and blood worms. Fertilized eggs were obtained from group spawnings in cages of stainless steel mesh. The eggs fell through the mesh to the bottom of the aquarium and were thus protected from being predated by the fish. Half an hour after the onset of light in the morning, the eggs were collected and rinsed from debris using a tea strainer. Fertilized eggs of good quality were transferred to embryo medium for subsequent experiments. Embryos exposed from 12 hpf were obtained from a breeding stock kept on a reverse day/night cycle, i.e. spawning occurred when the lights were turned on in the evening. The embryo experiments were approved by the local Ethics Committee for Animal Research, Uppsala, Sweden (permission number C84/11). 2.4. Metabolic activation system (MAS) MAS consisted of Tris buffer (0.1 M, pH 7.4 at 32 ◦ C), NADPH (1 mM) and liver microsomes at a final microsomal protein concentration of 0.7 mg/mL, as described previously [7]. The liver microsomes were from rats treated with the CYP-inducing substances ␤-NF and PB. The microsomes contain a variety of microsomal CYPs and have, according to the product specification, catalytic activity associated with CYP1A and CYP2B enzymes. The microsomes were obtained from rat, rather than zebrafish, because the metabolic profile can be assumed to be more conserved between humans and rodents than between humans and fish.

33

and each exposure was performed once. However, a few exposures were repeated together with appropriate controls to verify the results, i.e. 0.3 ␮M ABZ at 2–3 hpf, 3 ␮M ABZ + MAS at 2–3 hpf and 0.3 ␮M ABZ at 12–14 hpf. The MAS components were kept on ice and mixed immediately before the experiments. Exposure solutions were prepared in polypropylene vials by diluting ABZ stock solutions to the final exposure concentrations in either Tris buffer or MAS. Exposure solutions were pre-incubated for 25 min in an oscillating water bath at 32 ◦ C in order to start the metabolism before exposure. Eggs or embryos at the desired stage were aspired with a wide-bore glass pipette and were placed individually in wells of 96-well microplates. The embryos were quickly dispensed one by one, without dispensing embryo medium, by gently touching the bottom of the empty well with the tip of the pipette. The microplates were then filled with 100 ␮L/well of pre-incubated exposure solution using a multi-pipette and subsequently incubated at 32 ◦ C under gentle agitation for 1, 2 or 4 h. The incubation temperature (32 ◦ C) was a compromise of what is tolerated by zebrafish embryos (up to 33 ◦ C) and what is optimal for mammalian microsomes (37 ◦ C) [7]. After exposure, the exposure solution was collected, pooled from all wells in the same group, and stored at −18 ◦ C for subsequent chemical analysis. The embryos were rinsed several times in embryo medium, and were then reared individually at 26 ◦ C on 96-well microplates containing 250 ␮L embryo medium per well. 2.7. Evaluation of viability and developmental effects Embryos/larvae were observed and scored for developmental effects using an inverted microscope (CKX41, Olympus AB, Solna, Sweden). Developmental toxicity may be manifested as death (here defined as degraded/coagulated embryo), malformations, altered growth/development and functional deficits. Recorded morphological and functional endpoints were mainly as suggested by Brannen et al. [3] and are shown in Table 1. Each observed anatomical structure was judged as normal or showing either mild or severe abnormality. However, the final data compilation was based on classification of the endpoints as normal or abnormal. Embryos exposed at 2 hpf were observed at 8 hpf and at 1, 2 and 3 days post fertilization (dpf), embryos exposed at 12 hpf were observed at 1, 1.5, 2.5 and 6 dpf, and embryos exposed at 24 hpf were observed at 1.5, 2, 3 and 6 dpf. Embryos exposed at 2 hpf were not subjected to final scoring at 6 dpf since they either died early on or, at lower concentrations, were still normal at 3 dpf. Developmental staging was made according to Kimmel et al. [27]. Heart rate was analyzed at 2.5 dpf by manually counting the number of beats per minute. Touch response was tested 6 dpf by gently touching the tail with a pipette tip. Absence of avoidance behavior was recorded as abnormal. The embryos were euthanized at the end of the experiment by adding buffered tricaine (MS-222) to the embryo medium (≥0.3 mg/mL). 2.8. Effect of omitting NADPH from MAS The mDarT assay was repeated at 2 hpf as described above, except that an additional ABZ + MAS group was included for which no NADPH was added to the MAS mixture. An ABZ concentration of 1 ␮M was chosen since this caused 100% mortality in previous experiments (Fig. 3). The groups were: vehicle control, ABZ, MAS, ABZ + MAS, and ABZ + MAS without NADPH. Twenty embryos were used per group.

2.5. CYP activity in MAS 2.9. ABZ transformation in the absence and presence of embryos The batch of rat liver microsomes subsequently used in the zebrafish embryo experiments was analyzed regarding CYP activity under the experimental conditions used in the mDarT method. CYP2C6 is together with CYP3A1/2, CYP2A1 and FMOs suggested to catalyze sulfoxidation of ABZ in rat liver microsomes [18] and was therefore chosen as a representative CYP enzyme. The microsomal CYP2C6 activity was analyzed using the P450-GloTM CYP2C6 Assay (Promega, Madison, USA), which is based on a reaction where a luminogenic substrate is converted by CYP2C6 to a luciferin product that generates light in a second reaction. The assay was performed according to the manufacturer’s recommendations with some modifications. Briefly, the MAS mixture was prepared as described in section 2.4. The reactions were carried out in 96-well plates in a total volume of 50 ␮L and a temperature of 32 ◦ C. The light signal was detected in a microplate reader. A Michaelis–Menten saturation curve was generated by assaying serial dilutions of the substrate, and the Vmax and Km were calculated from the curve. A final substrate concentration of 150 ␮M was chosen based on a Km -value of 110 ␮M. The linearity of the reaction was checked by analyzing samples at different incubation times during 1 h. In the final analysis four replicates were used. In addition, the activity was measured in four replicates where NADPH was omitted from the MAS mixture. 2.6. Assessment of effects of ABZ in the mDarT assay The mDarT assay was modified from Busquet et al. [7]. An overview is presented in Fig. 2. Embryos were exposed from the 32–64 cell stage (approximately 2 hpf), the 5-somite stage (approximately 12 hpf) and 24 hpf (prim-5 stage). Staging was done according to Kimmel et al. [27]. Hpf and dpf denote time in hours and days, respectively, passed since the onset of light which triggered the spawning. The concentrations of ABZ tested at 2 hpf ranged from 0.1 to 1 ␮M without MAS and from 1 to 27 ␮M with MAS. At 12 and 24 hpf ABZ was tested at 0.3, 2 and 6 ␮M, both with and without MAS. Typically, 20 embryos were used per exposure group

Twenty embryos per replicate were incubated together in a vial containing 2 mL of ABZ. A low, non-toxic, concentration (0.1 ␮M) was used in order to avoid lethality which otherwise may compromise the ability of the embryos to metabolize the compound. The embryos were incubated for 10 h at 2–12, 12–22 or 24–34 hpf. The temperature was 26 ◦ C and the samples were gently oscillated during incubation. A sample of the exposure solution was taken at the end of the incubation and was stored at −18 ◦ C for subsequent chemical analyses. Three reference samples were used; one without ABZ, one ABZ-sample that was not incubated and one that was incubated without embryos. 2.10. Chemical analysis The concentration of ABZ and its metabolites ABZSO, ABZSO2 and ABZSO2 NH2 was analyzed in the exposure solution, which was collected at termination of exposure as described in section 2.6. The analytical method was based on an automated on-line solid phase extraction-liquid chromatography–tandem mass spectrometry (SPE-LC–MS/MS) procedure, described by Jansson and Kreuger [28]. Limit of detection was 0.001 ␮g/L for ABZ and 0.002 ␮g/L for the three metabolites. Details concerning the method, recovery and precision are described elsewhere [25]. 2.11. Statistical analysis The statistical analysis was compiled using GraphPad Prism (GraphPad Software Inc., San Diego, CA, USA). Frequency of abnormal and/or dead embryos was analyzed for each treatment group versus control using Fisher’s exact test followed by Bonferroni correction of p-values for multiple comparisons. Heart rate was tested for equal variance between the groups and was then analyzed using one-way analysis of variance (ANOVA) with Dunnett’s multiple comparison post hoc test.

34

A. Mattsson et al. / Reproductive Toxicology 34 (2012) 31–42

Fig. 2. Schematic illustration of the modified mDarT assay. Dilution series of ABZ were pre-incubated for 25 min± a metabolic activation system (MAS) in vials placed in an oscillating water bath. Embryos at the selected developmental stages (2, 12 and 24 hpf) were exposed to the pre-incubated test solution in 96-well plates under moderate agitation. Following 1, 2 or 4 h exposure, the incubation medium was pooled and later analyzed regarding concentration of ABZ and its metabolites. The embryos were thoroughly rinsed in embryo medium and were then cultured in embryo medium in 96-well microplates. Embryos/larvae were evaluated for viability and developmental parameters at three or four developmental stages, depending on stage at exposure.

3. Results

ultracentrifugation following the pre-incubation/metabolization step, i.e. before the exposure solution was added to the embryos. However, removal of the microsomes before exposure did not sufficiently reduce the toxicity to permit continuous incubation. In fact, microsomes and NADPH were toxic when tested both alone and in combination. Thereafter, various incubation durations were tested, starting at 2, 12 and 24 hpf. The longest tested tolerable incubation times were: 1 h from 2 hpf, 2 h from 12 hpf and 4 h from 24 hpf (data not shown). In addition, we chose to expose the embryos in 96-well plates instead of in vials. The 96-well plate format offers the advantages that embryos are exposed individually and will not influence each other. The final mDarT experimental procedure used in this work was based on these optimization results and is described in sections 2.4, 2.6 and 2.8 and is illustrated in Fig. 2.

3.1. Sensitivity of embryos to MAS Since extended incubation periods with MAS cause toxicity to zebrafish embryos, exposures with MAS have previously been restricted to a very limited period in zebrafish development, i.e. 1 h at 2 hpf [7,8]. We therefore investigated whether the method could be optimized to allow exposure for longer periods and at other developmental stages. Continuous incubation with MAS alone from 2 hpf caused developmental delay, which was noted already after 5 h, and embryonic death, which was noted after two days of exposure (microsomes 0.7 mg/mL; NADPH 10 mM). Reducing the concentration of MAS in steps down to 1.4% of the original concentration did not sufficiently reduce MAS toxicity to allow continuous exposure for days, starting at 2 hpf. After 24 h of exposure all embryos were developmentally delayed, even at the lowest concentration tested (n = 10). It was also tested whether the toxicity could be reduced by removing the microsomes by

3.2. CYP activity in MAS The in vitro assay of CYP2C6 activity showed a substrate conversion rate of 1.10 ± 0.06 ␮M/h in MAS. The activity was constant

Table 1 Endpoints assessed at the designated time points in zebrafish embryos. Endpoint

Dead/coagulated Development/growth General body shape Transparency Pigmentation (eyes and body) Head and face morphology Eye morphology Otic vesicle morphology Notochord morphology Somite/myotome morphology Tail morphology Fin morphology Yolk morphology Heart morphology Heart rate Blood circulation Edema Motility Touch response Hatched a b c

Method of assessment

% incidence in group Normal/abnormal + description Normal/abnormal + description Normal/abnormal + description Normal/abnormal + description Normal/abnormal + description Normal/abnormal + description Normal/abnormal + description Normal/abnormal + description Normal/abnormal + description Normal/abnormal + description Normal/abnormal + description Normal/abnormal + description Normal/abnormal + description Beats per minute Normal/abnormal + description Yes/no + location Yes/no Yes/no Yes/no

Time point for observation (dpf) 0.3a

1a,b

1.5b,c

2a,c

2.5b,c

3a,c

6b,c

× ×

× × × ×

× × ×

× × ×

× × ×

× × ×

× × × × × × ×

× × ×

× × ×

× × ×

× × ×

× × × × × × × × × × ×

× × ×

×

×

× ×

× × × × × ×

×

× × ×

× × ×

× × × × × ×

× × × × × × × × ×

× × × × ×

Embryos exposed 2–3 hpf were observed at 8 hpf (0.3 dpf) and at 1, 2 and 3 dpf. Embryos exposed 12–14 hpf were observed at 1, 1.5, 2.5 and 6 dpf. Embryos exposed 24–28 hpf were observed at 1.5, 2, 3 and 6 dpf.

× ×

×

A. Mattsson et al. / Reproductive Toxicology 34 (2012) 31–42

35

Fig. 3. Frequency (%) of coagulated and abnormally developed embryos observed at the designated time points. Embryos were exposed to ABZ ± MAS at 2–3, 12–14 and 24–28 hpf. N = 16–45.

during at least 1 h of incubation. In the samples where NADPH was omitted the activity was 0.0006 ± 0.0004 ␮M/h. 3.3. ABZ in the mDarT assay 3.3.1. Effects The frequency of coagulated (dead) and abnormally developed zebrafish embryos is shown in Fig. 3. The frequency of the individual endpoints is found in Tables 2 and 3. For simplicity, the developmental effects of exposure are shown for only a few of the observation time-points. No effects were found at the excluded time-points that could not be found at any of the other examined stages. A few exposures were repeated with appropriate controls, since they represented, or were close to, the LOAEL. These were 0.3 ␮M ABZ at 2–3 hpf, 3 ␮M ABZ + MAS at 2–3 hpf and 0.3 ␮M ABZ

at 12–14 hpf. The repeated experiments produced similar results as the previous ones and therefore only the combined results are shown in Fig. 3 and Tables 2 and 3. At 2–3 hpf, zebrafish embryos were particularly sensitive to ABZ. The graphs in Fig. 3 (first column) indicate a very steep concentration–response curve with a lowest effect concentration of 0.3 ␮M. Effects could be seen already at the termination of the exposure, as the blastomeres had failed to divide and were irregular in shape. These eggs failed to enter the gastrulation stage and were all coagulated by 1 dpf. Representative microphotographs of a vehicle-exposed and an ABZ-exposed egg as they appeared at 8 hpf are shown in Fig. 4. A few embryos exposed to 0.3 ␮M ABZ showed normal gastrulation but later developed an abnormal head and eye morphology (see Table 2). The heads were small and appeared condensed and non-transparent. The eyes were small and irregular in

36

A. Mattsson et al. / Reproductive Toxicology 34 (2012) 31–42

Table 2 Frequency (%) of effects observed at the designated time points in embryos exposed from 2 to 3 hpf. -MAS

+MAS

Concentration (␮M)

0

0.1

0.2

0.3

0.6

1.0

0

1

2

3

9

27

No. of embryos

40

20

20

40

20

20

20

20

20

40

20

20

8 hpf Coagulated Delay and/or irregular cells ˙ coagulated or abnormal

0 0 0

0 0 0

0 0 0

2.5 53 55*

15 85 100*

15 80 95*

0 0 0

0 0 0

0 0 0

0 0 0

1 dpf Coagulated Developmental delay Low transparency in head a Underdeveloped head Small, irregular eyes Malformed notochord Bent tail ˙ coagulated or abnormal

0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0

63 5 38 28 28 2.5 15 100*

100 0 0 0 0 0 0 100*

100 0 0 0 0 0 0 100*

0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0

3 dpf Coagulated Developmental delay Underdeveloped head Small eyes Bent tail Elongated or swollen heart Low blood circulation ˙ coagulated or abnormal

2.5 0 0 0 0 2.5 2.5 5

0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0

65 2.5 5 5 13 10 2.5 77*

100 0 0 0 0 0 0 100*

100 0 0 0 0 0 0 100*

0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0

10 85 95*

0 100 100*

0 0 15 7.5 0 0 0 15

90 5 10 5 5 5 0 100*

100 0 0 0 0 0 0 100*

0 0 0 0 0 0 0 0

95 5 5 5 5 0 5 100*

100 0 0 0 0 0 0 100*

Note: Frequencies >50% are shown in bold. a Low or no transparency, fuzzy texture of the tissue and diffuse brain boundaries. * Significantly increased frequency of ˙ coagulated or abnormal (p < 0.001, Fisher’s exact test with Bonferroni correction). Table 3 Frequency (%) of effects observed at the designated time points in embryos exposed 12–14 and 24–28 hpf. Exposure 12–14 hpf

Exposure 24–28 hpf

-MAS

+MAS

Concentration (␮M)

0

0.3

2

Number of embryos

6

+MAS

2

6

0

0.3

22

20

21

16

19

19

0 0 0

0 0 0

0 0 0

– – –

– – –

0 100 100 100 14 5 33 29 62 100*

0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0

0 90 19 90 62 0 5 5 10 5 0 14 0 95*

0 0 0 0 0 0 0 0 0 0 0 9 0 9

0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 5 0 0 0 0 14 0 14

0 0 0 0 0 0 0 0 0 0 0 0 0 0

43

45

21

21

1 dpf Coagulated Low transparency in head a ˙ coagulated or abnormal

0 0 0

2 0 2

5 5 10

0 38 38*

1.5 dpf Coagulated Low transparency in head a Underdeveloped head Small, irregular eyes Malformed notochord Bent tail Yolk sac edema Low heart rate Low blood circulation ˙ coagulated or abnormal

0 0 0 0 0 0 0 0 0 0

2 0 2 0 0 0 2 0 0 4

5 5 14 0 0 0 0 5 0 24#

6 dpf Coagulated Developmental delay Malformed head and/or eye Curved body and short tail Wavy or bent notochord Malformed myotomes Pericardial/yolk sac edema Malformed yolk sac Elongated or swollen heart No or low blood circulation Hemorrhage in head Unhatched No touch response ˙ coagulated or abnormal

0 0 0 0 0 0 0 0 0 0 0 0 0 0

2 0 0 2 0 0 0 0 2 0 0 2 0 10

5 0 5 5 5 0 5 0 5 5 0 5 0 14

-MAS

0

2

0

2

6

20

19

19

19

– – –

– – –

– – –

– – –

– – –

0 0 0 0 0 0 0 0 0 0

0 5 0 0 0 0 0 0 11 11

0 80 0 0 0 0 0 0 80 80*

0 0 0 0 0 0 0 0 5 5

0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 100 100 100 84 0 16 10 0 0 0 0 37 100*

10 0 90 90 65 50 75 30 55 30 15 0 70 100*

0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0

Note: Frequencies > 50% are shown in bold; –, not applicable. a Low/no transparency, fuzzy texture of the tissue and diffuse brain boundaries. * Significantly increased frequency of ˙ coagulated or abnormal (p < 0.001, Fisher’s exact test with Bonferroni correction). # Significantly increased frequency of ˙ coagulated or abnormal (p < 0.05, Fisher’s exact test with Bonferroni correction).

6

A. Mattsson et al. / Reproductive Toxicology 34 (2012) 31–42

Fig. 4. Inverted microscope images of zebrafish eggs at 8 hpf after exposure from 2 to 3 hpf. (A) Control and (B) exposed to 1 ␮M ABZ. The control has reached the germ ring stage which denotes the onset of gastrulation, whereas the ABZ exposed egg shows irregular blastomeres that ceased to divide early on.

shape. This phenotype is shown in Fig. 5B. Similar effects were seen when MAS was included but only at substantially higher concentrations (Table 2), i.e. MAS drastically reduced the embryotoxicity exerted by ABZ. At 12–14 and 24–28 hpf, the embryos were less sensitive to ABZ-exposure and showed malformations rather than mortality (Fig. 3 and Table 3). Effects were seen at 2 and 6 ␮M. The earliest appearing endpoint following exposure at 12–14 and 24–28 hpf was abnormal head morphology, which was observed already 8 h after exposure started. The effect on head morphology was similar to that found following exposure from 2 to 3 hpf, as described above. See microphotographs in Fig. 5. Abnormal head morphology was also induced by exposure from 24 to 28 hpf. However, 12 h after exposure started (1.5 dpf) the head phenotype was less abnormal compared to that seen after exposure at 2–3 and 12–14 hpf. On day six, the head phenotypes had become more severely abnormal in embryos exposed from 24 to 28 hpf whereas those exposed from 12 to 14 hpf had partly recovered.

37

Fig. 5. Inverted microscope images of zebrafish embryos at 1.5 dpf (36 hpf) after exposure from 12 to 14 hpf. (A) Control. (B) Exposed to 6 ␮M ABZ. The control embryo is transparent and shows clear brain boundaries and pigmented, well-defined, eyes. The exposed embryo shows an underdeveloped head region with condensed tissue, small, irregular, eyes and diffuse brain boundaries. Hb: hindbrain; c: cerebellum; mb: midbrain; fb: forebrain; e: eye.

Exposure at 2–3 and 12–14 hpf caused deformation of the body axis, which was evident from the second day of incubation and became more apparent on day six, when most embryos were hatched. Typically, the body was severely curved, the notochord distorted (often “wavy”) and the tail was short (Fig. 6). A consistent difference in body curvature between embryos exposed 12–14 and 24–28 hpf was that the former embryos displayed scoliosis (lateral deviation) whereas the latter displayed lordosis (dorsal deviation). Compare image C, E and G in Fig. 6. Embryos with body curvature displayed circular swimming movements. The majority of embryos exposed to the highest concentration from 24 to 28 hpf failed to show avoidance behavior in response to touch and some embryos trembled severely. Cardiovascular defects were readily inducible by exposure from 24 to 28 hpf and to some extent by exposure from 12 to 14 hpf (Table 3). This was first seen at 1.5 dpf as a transient reduction of blood flow to tail and/or head region and at later stages as pericardial and/or yolk sac edema, altered heart morphology, bradycardia and hemorrhages in the head region. The heart rate assessed at 2.5 dpf was slightly, but significantly, reduced by exposure to 2 and 6 ␮M ABZ from 24 hpf (Fig. 7). The heart rate reduction was reversed

38

A. Mattsson et al. / Reproductive Toxicology 34 (2012) 31–42

Fig. 6. Inverted microscope images of zebrafish larvae at 6 dpf following exposure between 12 and 14 hpf (A–D) or between 24 and 28 hpf (E–H). Larvae were exposed with (right panel) or without (left panel) addition of MAS. (A) and (B) Ventral view of a vehicle and MAS control larva, respectively. (C) and (D) Exposure to 6 ␮M ABZ between 12 and 14 hpf. (E) and (F) Exposure to 2 ␮M ABZ between 24 and 28 hpf. (G) and (H) Exposure to 6 ␮M ABZ between 24 and 28 hpf. Malformations of tail and head were typically seen in ABZ exposed larvae (C, E and G) and these were frequently accompanied by pericardial oedema, malformed heart and excess yolk remnant in embryos exposed to 6 ␮M ABZ between 24 and 28 hpf (G). Also note the shortened snout with underdeveloped and malformed upper and lower jaw (E) and (G). Exposure together with MAS offered complete protection against these effects (right panel). All images are shown with the same magnification. nc: notochord; t: tail; e: eye; lj: lower jaw; ht: heart: pe: pericardial edema; y: yolk sac.

by MAS. There was no significant change in heart rate compared to vehicle control in any other group. Gross heart morphology was not affected in embryos exposed to 2 ␮M ABZ from 24 hpf, suggesting that heart rate was a more sensitive endpoint than abnormal heart morphology. At the higher concentration (6 ␮M) however, elongated or swollen heart was a common endpoint (>50%). In summary, ABZ alone caused lethality and malformations at all three tested developmental stages and the toxicity was markedly reduced by MAS. 3.3.2. Chemical analysis of exposure solution The chemical analysis of the exposure solution after exposure to ABZ alone showed that ABZ mainly remained in its original form with some (10–30%) transformation to the first metabolite ABZSO (Table 4). When incubated with MAS, a minor fraction remained as ABZ whereas the major fractions consisted of ABZSO and the second metabolite ABZSO2 . The third metabolite, ABZSO2 NH2 , was found at low concentrations in some MAS-treated samples. It should also be noted that the total recovery, i.e. sum of ABZ and metabolites as fraction of nominal ABZ concentration, was near 100% at the lowest nominal concentrations but substantially lower at the higher nominal concentrations. We have no explanation for the low recovery at high concentrations, but binding to plastics and embryos or precipitation during the incubation may have reduced the concentration

in the exposure solution. Overall, the chemical analyses demonstrate that ABZ without MAS remained relatively stable while it was extensively metabolized by MAS to ABZSO, ABZSO2 and, to some extent, to ABZSO2 NH2 , when incubated for up to four and a half hour. 3.4. Effect of omitting NADPH from MAS The test aimed to elucidate whether the reducing co-factor NADPH was required for the protection against ABZ-toxicity obtained by MAS-treatment in the previous experiments. Again, exposure of zebrafish embryos to 1 ␮M ABZ 2–3 hpf caused 100% mortality, which could be completely prevented by MAS-treatment of the exposure solution (Fig. 8). Treatment with MAS without NADPH, i.e. buffer and microsomes alone, failed to protect against ABZ-toxicity. Thus, NADPH was required for the protecting effect of MAS. 3.5. ABZ transformation in the absence and presence of embryos With this experiment we aimed to elucidate whether zebrafish embryos at early developmental stages are able to metabolize ABZ and whether ABZ is spontaneously oxidized in exposure medium. The concentration of ABZ and its metabolites, ABZSO, ABZSO2 and

A. Mattsson et al. / Reproductive Toxicology 34 (2012) 31–42

39

Fig. 9. Measured concentration of ABZ and its major metabolite ABZSO in exposure solution after incubation for 10 h in the presence of 10 zebrafish embryos/mL or without embryos. Samples that were incubated for 10 h showed larger fractions of ABZSO than the un-incubated sample, regardless if they were incubated in the presence or absence of embryos. The square (#) denotes that no embryos were incubated with the sample. The figure shows mean concentration and range.

Fig. 7. Heart rate (beats/min) in zebrafish embryos at 2.5 dpf (60 hpf). Embryos were exposed to ABZ with or without MAS from 12 to 14 hpf (A) and from 24 to 28 hpf (B). Embryos exposed from 2 to 3 hpf were not analyzed. All groups were compared with the corresponding vehicle control (0 ␮M in figure) using one-way analysis of variance (ANOVA) followed by Dunnett’s multiple comparison post hoc test. *p < 0.05, **p < 0.01. Bars show group mean and standard deviation. N equals 10 in A and 16–20 in B.

Fig. 8. Effect of omitting NADPH from MAS. Embryos were exposed from 2 to 3 hpf to vehicle (C) or 1 ␮M ABZ, with or without MAS. The group shown to the right in graph was exposed to ABZ together with MAS lacking the reducing co-factor NADPH. Bars show frequency of coagulated (dead) embryos one day after exposure. All surviving embryos developed normally. N = 20 in all groups.

ABZSO2 NH2 , in exposure solution was studied after incubation of ABZ for 10 h with and without embryos. A non-toxic concentration of ABZ was used (0.1 ␮M). As shown in Fig. 9, none of the analyzed substances were detected in the vehicle controls where no ABZ had been added. An ABZ sample that had not been incubated at all contained mainly ABZ and a minor fraction of ABZSO (∼5%). Samples that were incubated for 10 h showed larger fractions (∼20–50%) of ABZSO than the un-incubated sample, regardless if they were incubated with embryos or not. The other two metabolites, ABZSO2 and ABZSO2 NH2 , were not detected in any of the samples. Because of high spontaneous transformation to ABZSO after 10 h and a high variability it was not possible to say whether the embryos may have contributed or not. The transformation of ABZ did not seem to be higher at the later developmental stages. To summarize, there was a spontaneous transformation of ABZ to ABZSO but we could not see any obvious contribution from metabolism by the embryos. 4. Discussion In toxicity testing, both qualitative and quantitative aspects of the metabolic competence of the test system are of crucial importance. The mDarT test combines the zebrafish embryotoxicity test with an external mammalian metabolic activation system and thus enables studies of developmental toxicity associated with potential maternal metabolism of the tested compound [7]. We modified the mDarT experimental procedure and succeeded to include additional testing time points and to extend the exposure durations. Using the modified method, we demonstrated stage-dependent developmental toxicity elicited by ABZ and showed that ABZ could be metabolically deactivated by MAS. Due to inherent toxicity of the MAS components, exposures with MAS have previously been confined to only 1 h duration, starting from 2 hpf [7,8]. In the present study we showed that zebrafish embryos could be exposed together with MAS at two additional developmental time points, namely at 12 and 24 hpf. The embryos were considerably less vulnerable to MAS at these stages, and the exposure duration could therefore be extended to up to 2 and 4 h starting at 12 and 24 hpf, respectively, without any signs of impact on development. The earliest exposure period used in this study, 2–3 hpf, corresponds approximately to the period between the 32cell stage and the 1000-cell stage. The exposure period from 12 to 14 hpf represents the beginning of the segmentation period, which is a period between approximately 10 and 24 hpf when many important developmental processes occur. By 12 hpf five somites

40

A. Mattsson et al. / Reproductive Toxicology 34 (2012) 31–42

Table 4 Measured concentration of ABZ, ABZSO, ABZSO2 and ABZSO2 NH2 at termination of exposure in the experiments where embryos were exposed 2–3, 12–14 and 24–28 hpf. The analyses were made on pooled samples. ˙: Sum of measured ABZ and metabolites in ␮M and as percent of nominal ABZ concentration within parentheses. Treatment (␮M ABZ)

Measured concentration (␮M)



ABZ

ABZSO

ABZSO2

ABZSO2 NH2

Exposure 2–3 hpf 0 0.1 0.2 0.3 0.6 1 0 + MAS 1 + MAS 2 + MAS 3 + MAS 9 + MAS 27 + MAS

n.d. 0.070 0.16 0.25 ± 0.01 0.45 0.82 n.d. 0.002 m.v. 0.013 ± 0.005 0.038 0.47

n.d. 0.024 0.06 0.04 ± 0.04 0.063 0.098 n.d. 0.089 m.v. 1.20 ± 0.01 6.8 7.7

n.d. n.d. n.d. n.d. n.d. n.d. n.d. 0.29 m.v. 0.6 ± 0.1 0.83 0.64

n.d. n.d. n.d. n.d. n.d. n.d. n.d. 0.028 m.v. 0.006 ± 0.008 n.d. n.d.

– 0.094 (94) 0.22 (108) 0.29 ± 0.03 (96 ± 10)a 0.51 (86) 0.92 (92) – 0.38 (38) m.v. 1.8 ± 0.2 (61 ± 7)a 7.7 (85) 8.8 (33)

Exposure 12–14 hpf 0 0.3 2 6 0 + MAS 2 + MAS 6 + MAS

n.d. 0.26 0.70 1.06 n.d. 0.01 0.07

n.d. 0.04 0.06 0.08 n.d. 0.5 1.5

n.d. n.d. n.d. n.d. n.d. 0.5 0.4

n.d. n.d. n.d. n.d. n.d. 0.03 n.d.

– 0.3 (100) 0.8 (38) 1.1 (19) – 1.1 (53) 1.9 (32)

Exposure 24–28 hpf 0 0.3 2 6 0 + MAS 2 + MAS 6 + MAS

n.d. 0.24 0.32 1.1 n.d. 0.01 0.22

n.d. 0.04 0.04 0.1 n.d. 0.51 3.0

n.d. n.d. n.d. n.d. n.d. 0.43 0.65

n.d. n.d. n.d. n.d. n.d. n.d. 0.02

– 0.29 (96) 0.36 (18) 1.2 (20) – 0.94 (47) 3.9 (65)

a

Samples from two experiments; m.v.: missing value; n.d.: non detected. Limit of detection was 0.001 ␮g/L for ABZ and 0.002 ␮g/L for the three metabolites.

have formed and during the following 12 h the remaining somites develop, neurulation occurs, the rudiments of the primary organs become visible and the first cells differentiate morphologically [27]. By 24 hpf the zebrafish embryo has entered the pharyngula stage, which is the phylotypic stage at which vertebrate embryos of different species show the highest similarities. The zebrafish embryo now displays intense spontaneous body movements. It is well recognized that there is an association between type of developmental perturbations caused by a chemical and the developmental stage at the time of exposure. By exposing for extended periods or at several stages, the possibility to detect the potential of a chemical to elicit specific developmental perturbations can therefore be substantially improved. The developmental toxicity of ABZ showed a steep concentration–response relationship, with a small increase in concentration causing a large increase in response. Many of the embryotoxic effects seen in mammals following in utero exposure to ABZ were also seen in zebrafish embryos in the present study, for instance cardiovascular defects, eye- and head malformations, dysmorphic somites, short tail and axial deviations [11–15]. The embryotoxic potential of ABZ in zebrafish has been related to the unmetabolized form of the molecule [25], whereas in mammals ABZSO alone is sufficient to induce the developmental effects caused by ABZ, and the embryotoxic potential of ABZ in mammals has therefore been attributed to this metabolite [10]. After administration of teratogenic doses of the ABZ-forming drug netobimin to pregnant rat, the resulting embryo tissue concentrations were much higher for ABZSO (∼1.5 ␮g/g) than for ABZ (∼0.06 ␮g/g) [13]. Despite the lower concentration of ABZ, compared to its metabolite, it could not be excluded that ABZ contributed to the toxicity observed in rat embryos. The lowest observed effect concentration of ABZ in the present study was 0.3 ␮M (0.08 ␮g/mL embryo medium), which is in a similar concentration range as in

the rat embryos mentioned above (∼0.06 ␮g/g) [13] and the effect concentration in cultures of differentiating rat embryonic cells (0.3 ␮M) [23]. We found both quantitative and qualitative differences in embryotoxicity depending on stage at exposure to ABZ, which highlights the importance of the timing of exposure. Exposure from 2 to 3 hpf resulted in early mortality, whereas exposure from 12 to 14 hpf and from 24 to 28 hpf resulted in various teratogenic and functional effects. The embryos were more sensitive at 2–3 hpf than at 12–14 and 24–28 hpf. It should be kept in mind that the exposure duration was only 1 h at 2 hpf compared to two respectively 4 h at 12 and 24 hpf. The observed embryotoxic potential of ABZ at these different stages is therefore not directly comparable. The most prominent teratogenic effect was an aberrant body curvature which was frequently accompanied by notochordal distortions and a short tail. Interestingly, exposure from 12 hpf induced scoliosis whereas exposure from 24 hpf induced lordosis. Another difference was lack of touch-response in embryos exposed to the highest concentration at 24–28 hpf which was not seen after exposure 12–14 hpf. Exposure 2–3 hpf appeared to arrest blastomere cleavage and resulted in irregular cells and subsequently mortality. The pharmacological mechanism of ABZ is binding to ␤-tubulin of parasites and inhibiting the polymerization to microtubules [29]. As microtubules are essential for a variety of cellular processes, including mitotic spindle assembly and mitotic cell divisions, microtubular perturbation could be the mechanism behind the observed early embryotoxic/cytotoxic effect. Exposure 12–14 and 24–28 hpf caused cardiovascular effects, including heart malformations, bradycardia, low blood circulation, hemorrhage and edema. These effects were more frequently observed after exposure 24–28 hpf. ABZ also induces vascular malformations in sheep and rat exposed in utero to the ABZ-forming pro-drug netobimin [11,12] and affect

A. Mattsson et al. / Reproductive Toxicology 34 (2012) 31–42

retinal vasculature in zebrafish [30]. A possible mechanism is inhibition of vascular endothelial growth factor (VEGF) since ABZ has been shown to reduce plasma concentration of VEGF in a clinical trial where ABZ was evaluated as a potential anticancer agent [31]. The mDarT method was modified to individual exposure in 96well plates. Individual exposure has the advantage over grouped exposure in that the embryos will not influence each other and each embryo can thus be considered an independent biological replicate. With a given number of embryos, individual exposure offers better statistical power than grouped exposures where each experiment is regarded as a replicate. We also increased the number of time points for effect assessment in order to improve the sensitivity of the assay. Some effects became more severe later in development, e.g. the axial defects, whereas the embryos seemed to recover from other effects, e.g. impaired blood circulation. Thus, effect assessment at a single time point may not pick up all types of effects. ABZ has previously been shown to be embryotoxic in zebrafish with a potency at least 100-fold higher than that of its main metabolite ABZSO [25]. Dechorination of zebrafish embryos did not influence the embryotoxic potential of ABZSO, suggesting that bioavailability was not a limiting factor [25]. Metabolism of ABZ thus represents a deactivation step concerning zebrafish embryotoxicity which is in accordance with the protecting effect provided by MAS co-incubation in the present study. ABZ was tested with and without MAS at 2–3, 12–14 and 24–28 hpf and the results clearly demonstrate that the embryotoxic effects caused by ABZ at these stages could be completely inhibited or drastically decreased by co-incubation with MAS. The protective effect was completely abolished by omitting the reducing co-factor NADPH from MAS, suggesting that the protection was caused by an enzyme dependent metabolic deactivation, for instance by CYP or FMO enzymes, rather than by unspecific events such as protein binding. Using an in vitro assay measuring the catalytic activity of CYP2C6, we confirmed that MAS exhibited CYP activity under the conditions used in the test and that this activity was NADPH dependent. CYP2C6 may be involved in the sulfoxidation of ABZ that occur in the presence of rat liver microsomes [18]. Furthermore, chemical analysis of the exposure medium confirmed that ABZ was efficiently metabolized by MAS to ABZSO, ABZSO2 and, to some extent, to ABZSO2 NH2 . The chemical analysis thus provided important information regarding the actual exposure situation in the presence and absence of MAS although we are aware that uptake and concentration inside the embryo has not been addressed. It should also be noted that metabolism is taking place during both the 25-min preincubation and during the subsequent exposure of embryos, and that the concentrations of ABZ and its metabolites were measured at the termination of exposure only. The maximal and mean concentration of ABZ is therefore expected to be higher than what was found at the end of the exposure. Most zebrafish CYP genes are expressed as mRNA during at least some point of zebrafish embryonic development and may thus enable endogenous metabolism of xenobiotica [32]. In a recent study, ten different compounds that are known or suspected to be metabolically activated were tested on zebrafish embryos from 2.5 to 72 hpf, and nine of these induced malformations [33]. It may be tempting to conclude that the compounds were bioactivated in the zebrafish embryo, but it remains to be shown whether toxicity was caused by the parent compounds or by the metabolites. In the present study, zebrafish embryos were incubated for 10 h with a nontoxic concentration of ABZ at 2–12, 12–22 and 24–34 hpf to determine whether they were able to metabolize ABZ at these stages. There was no obvious endogenous metabolism of ABZ by the zebrafish embryos at these stages, but because of unexpectedly high spontaneous transformation of ABZ to ABZSO a contribution from the embryos could have been masked. In one of our previous studies, as much as 92–94% of the ABZ measured in the

41

exposure medium at initiation of exposure (at 24 hpf) had been biotransformed to ABZSO at the termination of the experiment (at 144 hpf), whereas nearly 100% remained as ABZ in medium that was incubated without embryos [25]. Thus, although ABZ metabolism may be inefficient at the early stages studied in the present work, it is drastically improved later in development. Alderton and co-workers investigated uptake and metabolism of 15 different compounds and revealed that zebrafish larvae at 7 dpf have the ability to perform both phase 1 (oxidation, N-demethylation, O-de-ethylation, and N-dealkylation) and phase 2 (sulfation and glucuronidation) metabolic reactions. However, only a small fraction (0.1–0.6%) of most of the compounds was found as metabolites in the larvae and the metabolic pathways were only partially conserved with those of mammals [34]. Taken together, the overall ability of the zebrafish embryo to biotransform xenobiotica appears to be inefficient and somewhat qualitatively different from that of mammals. Quantitative and qualitative aspects of endogenous metabolism in the developing zebrafish embryo need to be more thoroughly investigated in order to better understand the zebrafish embryo model and in order to decide at what stages it would be warranted to supplement the model with MAS. The mDarT method shows promising results, but it needs to be further evaluated regarding both metabolic deactivation and metabolic activation using a larger set of test compounds. 5. Conclusions We have developed the mDarT method further, mainly by including additional testing time points, extending the exposure duration and applying individual exposure. We demonstrated that the model substance ABZ was efficiently metabolized by MAS and verified that the metabolism was accompanied by the expected reduction in embryotoxic potential. The results also showed stagedependent toxic effects of ABZ during three windows of zebrafish embryonic development, i.e. at 2–3, 12–14 and 24–28 hpf. Our results emphasize, a) that metabolism may have a great impact on the developmental toxicity of a chemical and b) the importance of studying exposure at various developmental stages. Conflict of interest The authors declare that there is no conflict of interest that would prejudice the impartiality of this scientific work. Funding Financial support was provided by the Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning. Acknowledgments We want to acknowledge Dr. Franc¸ois Busquet and Dr Stefan Weigt for introducing us to the mDarT method. References [1] Nagel R. DarT: the embryo test with the Zebrafish Danio rerio – a general model in ecotoxicology and toxicology. ALTEX 2002;19(Suppl. 1):38–48. [2] Bachmann J. Entwicklung eines Teratogenitäts-Screening-Tests mit Embryonen des Zebrabärblings Danio rerio. In: Technische Universität. Dresden; 2002. [3] Brannen KC, Panzica-Kelly JM, Danberry TL, Augustine-Rauch KA. Development of a zebrafish embryo teratogenicity assay and quantitative prediction model. Birth Defects Res B Dev Reprod Toxicol 2010;89:66–77. [4] Selderslaghs IWT, Blust R, Witters HE. Feasibility study of the zebrafish assay as an alternative method to screen for developmental toxicity and embryotoxicity using a training set of 27 compounds. Reprod Toxicol., 2011; doi:10.1016/j.reprotox.2011.08.003.

42

A. Mattsson et al. / Reproductive Toxicology 34 (2012) 31–42

[5] Van den Bulck K, Hill A, Mesens N, Diekman H, De Schaepdrijver L, Lammens L. Zebrafish developmental toxicity assay: a fishy solution to reproductive toxicity screening, or just a red herring. Reprod Toxicol 2011;32:213–9. [6] Spielmann H, Seiler A, Bremer S, Hareng L, Hartung T, Ahr H, et al. The practical application of three validated in vitro embryotoxicity tests The report and recommendations of an ECVAM/ZEBET workshop (ECVAM workshop 57). Altern Lab Anim 2006;34:527–38. [7] Busquet F, Nagel R, von Landenberg F, Mueller SO, Huebler N, Broschard TH. Development of a new screening assay to identify proteratogenic substances using zebrafish Danio rerio embryo combined with an exogenous mammalian metabolic activation system (mDarT). Toxicol Sci 2008;104:177–88. [8] Weigt S, Huebler N, Braunbeck T, von Landenberg F, Broschard TH. Zebrafish teratogenicity test with metabolic activation (mDarT): Effects of phase I activation of acetaminophen on zebrafish Danio rerio embryos. Toxicology 2010;275:36–49. [9] Wilson JW. Environment and birth defects (environmental science series). London: Academic Press; 1973. [10] JECFA. Albendazole. In WHO food additives series 25. Prepared by the thirtyfourth meeting of the Joint FAO/WHO Expert Committee on Food Additives (JECFA). Geneva: World Health Organization; 1990. [11] Navarro M, Canut L, Carretero A, Cristofol C, Pérez-Aparicio FJ, Arboix M, et al. Developmental toxicity in rat fetuses exposed to the benzimidazole netobimin. Reprod Toxicol 1999;13:295–302. [12] Navarro M, Cristofol C, Carretero A, Arboix M, Ruberte J. Anthelmintic induced congenital malformations in sheep embryos using netobimin. Vet Rec 1998;142:86–90. [13] Cristòfol C, Navarro M, Franquelo C, Valladares JE, Carretero A, Ruberte J, et al. Disposition of netobimin, albendazole, and its metabolites in the pregnant rat: Developmental toxicity. Toxicol Appl Pharmacol 1997;144:56–61. [14] Delatour P, Garnier F, Benoit E, Longin CH. A correlation of toxicity of albendazole and oxfendazole with their free metabolites and bound residues. J Vet Pharmacol Ther 1984;7:139–45. [15] Mantovani A, Ricciardi C, Stazi AV, Macrì C. Effects observed on gestational day 13 in rat embryos exposed to albendazole. Reprod Toxicol 1995;9:265–73. [16] Gottschall DW, Theodorides VJ, Wang R. The metabolism of benzimidazole anthelmintics. Parasitol Today 1990;6:115–24. [17] Rawden HC, Kokwaro GO, Ward SA, Edwards G. Relative contribution of cytochromes P-450 and flavin-containing monoxygenases to the metabolism of albendazole by human liver microsomes. Br J Clin Pharmacol 2000;49:313–22. [18] Moroni P, Buronfosse T, Longin-Sauvageon C, Delatour P, Benoit E. Chiral sulfoxidation of albendazole by the flavin adenine dinucleotide-containing and cytochrome P450-dependent monooxygenases from rat liver microsomes. Drug Metab Dispos 1995;23:160–5. [19] Capece BPS, Virkel GL, Lanusse CE. Enantiomeric behaviour of albendazole and fenbendazole sulfoxides in domestic animals: Pharmacological implications. Vet J 2009;181:241–50.

[20] Delatour P, Parish RC, Gyurik RJ. Albendazole: a comparison of relay embryotoxicity with embryotoxicity of individual metabolites. Ann Rech Vet 1981;12:159–67. [21] Capece BPS, Navarro M, Arcalis T, Castells G, Toribio L, Perez F, et al. Albendazole sulphoxide enantiomers in pregnant rats embryo concentrations and developmental toxicity. Vet J 2003;165:266–75. [22] Teruel MT, Felipe AE, Solana HD, Sallovitz JM, Lanusse CE. Placental and fetal toxicity of albendazole sulphoxide in Wistar rats. Vet Hum Toxicol 2003;45:131–6. [23] Whittaker SG, Faustman EM. Effects of albendazole and albendazole sulfoxide on cultures of differentiating rodent embryonic cells. Toxicol Appl Pharmacol 1991;109:73–84. [24] Rolin S, Amri HS-E, Batt A-M, Levy M, Bagrel D, Siest G. Study of the in vitro bioactivation of albendazole in human liver microsomes and hepatoma cell lines. Cell Biol Toxicol 1989;5:1–14. [25] Carlsson G, Patring J, Ullerås E, Oskarsson A. Developmental toxicity of albendazole and its three main metabolites in zebrafish embryos. Reprod Toxicol 2011;32:129–37. Organization for Standardization (ISO). Water [26] International quality–determination of the acute lethal toxicity of substances to a freshwater fish [Brachydanio rerio Hamilton-Buchanan (teleostei, Cyprinidae)]. Part 1. Static method. ISO 7346-1; 1996. [27] Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, Schilling TF. Stages of embryonic development of the zebrafish. Dev Dyn 1995;203:253–310. [28] Jansson C, Kreuger J. Multiresidue analysis of 95 pesticides at low nanogram/liter levels in surface waters using online preconcentration and high performance liquid chromatography/tandem mass spectrometry. J AOAC Int 2010;93:1732–47. [29] Lacey E. Mode of action of benzimidazoles. Parasitol Today 1990;6: 112–5. [30] Kitambi SS, McCulloch KJ, Peterson RT, Malicki JJ. Small molecule screen for compounds that affect vascular development in the zebrafish retina. Mech Dev 2009;126:464–77. [31] Pourgholami M, Szwajcer M, Chin M, Liauw W, Seef J, Galettis P, et al. Phase I clinical trial to determine maximum tolerated dose of oral albendazole in patients with advanced cancer. Cancer Chemother Pharmacol 2010;65:597–605. [32] Goldstone J, McArthur A, Kubota A, Zanette J, Parente T, Jonsson M, et al. Identification and developmental expression of the full complement of Cytochrome P450 genes in Zebrafish. BMC Genomics 2010;11:643. [33] Weigt S, Huebler N, Strecker R, Braunbeck T, Zebrafish Broschard TH. (Danio rerio) embryos as a model for testing proteratogens. Toxicology 2011;281:25–36. [34] Alderton W, Berghmans S, Butler P, Chassaing H, Fleming A, Golder Z, et al. Accumulation and metabolism of drugs and CYP probe substrates in zebrafish larvae. Xenobiotica 2010;40:547–57.