Cyclosporin A, doxorubicin hydrochloride, phenanthrene, rhodamine B, verapamil hydrochloride and vincristine sulfate were from Sigma-Aldrich (Schnelldorf, Germany). Calcein-acetoxymethylester (calcein-am), MK571 and vinblastine sulfate were from Biozol (Eiching, Germany) and bodipy-vinblastine was from Invitrogen (Karlsruhe, Germany). PSC833 was a kind gift from Novartis (Basel, Switzerland). Galaxolide (73% of the total are diasterimeric isomers of 1,3,4,6,7,8-hexahydro-4,6,6,7,8,8-hexamethyl-cyclopenta-γ--benzopyran) was a kind gift from International Flavors & Fragrances Inc. (IFF; Union Beach, NJ, USA) and Tonalide (7-acetyl-1,1,3,4,4,6-hexamethyl-1,2,3,4-tetrahydronaphthalene; purity: 98%) was obtained from Bush Boake Allen Inc. (Jacksonville, FL, USA). RhB was dissolved in MilliQ water; stock solutions of all other chemicals were prepared in dimethyl sulfoxide (DMSO, Sigma-Aldrich). Final DMSO solutions in exposure media did not exceed 0.2%.
Culture of zebrafish, collection of eggs and culture of embryos
Adult zebrafish from the WIK wildtype strain were maintained and bred according to standard protocols . Collection of eggs and culturing of the embryos were performed as described .
RNA extraction and reverse transcription
Total RNA was extracted from 30 to 50 embryos from 1, 6, 12, 24 and 48 hpf zebrafish embryos using TRIzol Reagent (Invitrogen) according to the manufacturer’s instructions. Genomic DNA contaminations were removed with DNAse I (Roche, Grenzach, Germany) treatment. cDNA was synthesized from total RNA using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Darmstadt, Germany).
Cloning of zebrafish abcb4 and abcb5 cDNAs and phylogenetic and synteny analysis
Zebrafish abcb4 (predicted sequence accession no. ENSDARG00000010936) and abcb5 (predicted ENSDARG00000021787) sequences were obtained using reverse transcription polymerase chain reaction (RT-PCR) with primer pairs designed either based on the predicted sequences or based on sequences obtained with rapid amplification of cDNA ends (RACE) (Clontech, Palo Alto, USA). For PCR, Advantage 2 (Clontech) and Phusion (Finnzymes, Thermofisher, Schwerte, Germany) polymerases were used. PCR products were gel purified, cloned and sequenced on an ABI 3100 sequencer (Applied Biosystems) using standard cycle sequencing protocols. Sequences were edited and assembled using Sequencher 4.9 (Genecodes, Ann Arbor, USA) and analyzed using the National Center for Biotechnology Information (NCBI) basic alignment search tool (BLAST) and the Expert Protein Analysis System (ExPASy, Swiss Institute of Bioinformatics, Lausanne, Switzerland) . The zebrafish abcb4 and abcb5 sequences were submitted to GenBank (NCBI). Accession numbers are listed in Additional file 1: Table S2.
Identity rates of zebrafish transporter nucleotide/amino acid sequences with vertebrate orthologs were determined with ClustalX2 (Conway Institute UCD, Dublin, Ireland). Phylogentic trees were generated with MEGA5 (Center for Evolutionary Medicine and Informatics, Tempe, USA) using the neighbor-joining method with percentage concordance based on 1,000 bootstrap iterations. To establish syntenic relationships between vertebrate genomes within the chromosomal regions of interest, we made use of ortholog predictions in the Ensembl database .
Quantification of mRNA expression levels in zebrafish embryos
mRNA expression levels of abcb4 and abcb5 in 1, 6, 12, 24 and 48 hpf zebrafish embryos were quantified with quantitative RT-PCR (qPCR) using the SYBR Green PCR Master Mix (Quantace, Berlin, Germany). with an iCycler Real-Time PCR Detection System (BioRad, Munich, Germany).
Primers for housekeeping and zebrafish ABC transporter genes (Additional file 1: Table S5) were designed against available mRNA sequences from Ensembl  and self-obtained sequences using Beacon Designer (Primer Biosoft, Palo Alto, USA). Samples were run in triplicate in optically clear 96-well plates (Biozym, Hessisch Oldendorf, Germany). PCR was performed with RNA extracts from three different zebrafish embryo batches. qPCR results were calculated relative to the housekeeping gene, 18S (for selection of the housekeeping gene refer to Additional file 1: Figure S2), according to the normalization procedure of the Q-Gene Core Module[47–49], which takes varying PCR amplification efficiencies into account (Additional file 1: Table S6). All qPCR experiments were performed according to the MIQE (Minimum Information for Publication of Quantitative Real-Time PCR Experiments) guidelines . A MIQE checklist is found in Additional file 1.
Whole-mount in situ hybridization
For whole-mount in situ hybridization (WISH), abcb4 cDNA fragments were amplified (for primers refer to Additional file 1: Table S7), cloned into pCRII (Invitrogen) and verified by sequencing. WISH with 18, 38 and 120 hpf (functional proof of the used probes, specific staining in the intestine, see Additional file 1: Figure S3) zebrafish embryos was performed as described previously . WISH staining was analyzed with a stereomicroscope (MZ16F, Leica, Wetzlar, Germany).
Procedure for measuring efflux transporter protein activity in zebrafish embryos with fluorescent dyes
Fluorescent dyes, rhodamine B [22, 52], calcein-am [53, 54] and bodipy-vinblastine , served as proxies for efflux transporter activity in the fish embryos. When this activity is absent due to pharmacologic transporter inhibition or the absence of functional protein due to morpholino knock-down, accumulation of dye in the embryo tissue increases, resulting in a stronger fluorescence signal. Solutions for exposures were prepared in zebrafish embryo culture water with 0.5 μM rhodamine B, 1 μM calcein-am and 1 μM bodipy-vinblastine and inhibitors cyclosporin A (CsA), PSC833 and MK571, respectively. Up to 10 embryos per mL were incubated in the test solutions for one hour at 26°C in the dark, rinsed three times with clean culture water to remove dye from the chorion and subsequently photographed with a fluorescence microscope DMI 4000B (Leica) and DFC 350 FX camera (Leica). For quantification of RhB dye uptake, 10 embryos per treatment were sonicated in 200 μL of a hypotonic lysis buffer (10 mM KCl, 1.5 mM MgCl2, 10 mM Tris HCl, pH 7.4), the sonicates were briefly centrifuged, 150 μL of the supernatant were transferred to a black 96-well microplate (Nunc, Sigma-Aldrich, Schnelldorf, Germany) and the rhodamine B fluorescence was measured at 595 nm (emission)/530 nm (excitation) in a GENios plus fluorescence plate reader (Tecan, Männedorf, Switzerland). This assay enabled parallel examination of multiple treatments. Triplicates of five treatments along with a solvent control were run per experiment. Each experiment was repeated with embryos from three different egg batches laid on different days. The amount of rhodamine B accumulated in zebrafish embryos was quantified with a rhodamine B standard curve (Additional file 1: Figure S4).
Embryo toxicity experiments
For determining toxicities of vinblastine, vincristine and doxorubicin, 20 embryos were incubated in glass petri dishes with 10 mL test solutions and two to three replicates per treatment. Exposures to phenanthrene were set up in tightly closed glass vials containing 2 mL solution with four embryos per vial according to Schreiber et al.  to prevent volatilization of phenanthrene from the test solutions. Per tested treatment, five vials were set up in parallel. Exposures were started with 4- to 16-cell stage embryos to assure successful fertilization and terminated after 48 hours. Exposure experiments were repeated with at least three batches of embryos from different days. During exposure, embryos were regularly examined using a stereo microscope and dead embryos were removed and recorded. A final mortality count was performed at 48 hours and embryos were declared as dead if at least one of the following criteria applied: i) coagulation of eggs, ii) no heart beat, iii) no blood circulation, iv) no somites, v) tail not detached . Initially, toxic concentration ranges of vinblastine, vincristine, doxorubicin and phenanthrene and the transporter inhibitors CsA and PSC833 were identified. Subsequently, effects of transporter inhibitors at non-toxic concentrations and, in the case of vinblastine, of morpholino knock-down of Abcb4 and Abcb5 on the sensitivity of the embryos to toxic test compounds were determined. Controls contained i) 0.2% DMSO used as solvent, ii) inhibitors only or iii) morpholino knock-down embryos only. Mean mortality percentages and standard deviations at 48 hpf were calculated from all experimental replicates and the paired t-test was applied to determine whether inhibitors or morpholino knock-down significantly modulated vinblastine sensitivity of embryos.
Production of recombinant zebrafish Abcb4 protein with the baculovirus expression system and ATPase activity measurements
The zebrafish abcb4 cDNA was sub-cloned into pFastBac1 (Invitrogen) and sequenced for confirmation. Abcb4 baculovirus was generated using the Bac-to-Bac Baculovirus Expression System (Invitrogen). Sf9 cells cultured in Sf-900 II SFM (Invitrogen) were used for virus amplification and protein expression. Crude Sf9 membranes with Abcb4 protein were prepared 60 hours after infection according to  and then stored at −80°C until use. Total protein in the membrane preparations was quantified using the bicinchoninic acid assay (BCA) and bovine serum albumin (BSA) as standard and the presence of Abcb4 protein in the membranes was checked in 1 μg of total protein by Western blotting using the anti-MDR1 antibody C219 (Additional file 1: Figure S5) as described previously .
ATPase assays were performed as described in  with minor modifications. Instead of 37°C, incubations of the membranes with the test compounds were performed for 40 minutes at 27°C, which is within the physiological temperature range of zebrafish. A total of 20 μg of protein was used for each reaction. The ATPase stimulating effect of verapamil, a classical stimulating agent of the mammalian Abcb1 ATPase, was also detected for zebrafish Abcb4 (Figures 5A) and 40 μM verapamil were used as positive control in ATPase stimulation assays and as ATPase stimulating agent in the ATPase inhibition assays. DMSO was used as the solvent for all compounds. The final DMSO concentration in the reaction was 2%, which did not affect ATPase activities.
Abcb4 and Abcb5 knock-down by morpholino (MO) microinjection
Embryos were injected (FemtoJet, Eppendorf, Hauppauge, USA) through the chorion into the yolk compartment at the two-cell stage. Injection needles were pulled from borosilicate glass capillary tubes with filament (Warner Instruments, Hamden, USA) using a micropipette puller (Narishige, Tokyo, Japan).
Morpholinos (Gene Tools, Philomath, USA) were dissolved in MilliQ water and injected at the following concentration ranges: 0.5 mM to 2 mM Abcb4-MO-SP (SP = splice-blocking) (5′-AAT CTA ACT GCA TGA CGT ACT CTG T-3′); 0.0625 mM to 1 mM Abcb4-ATG-MO (translation-blocking) (5′-GCA AAC ATG GGC AAG AAA TCC AAA C-3′), 0.5 mM to 2 mM Abcb5-MO-SP (5′- GCA ACA GGT ACA TTC ATG TCT TTC T-3′) and 1 mM to 2 mM control splice morpholino (ctrl-MO) (MO against human beta-globin) (5′- CCT CTT ACC TCA GTT ACA ATT TAT A-3′). Functionality of Abcb4-SP-MO and Abcb5-SP-MO morpholinos were proven by means of RT-PCR (Additional file 1: Figure S6) and Abcb4-ATG-MO morpholino with co-injected Abcb4-ATG-GFP mRNA (Additional file 1: Figure S6). Injected embryos were cultured for ≤48 h and transporter activity dye assays and toxicity experiments were done as described above.