Experimental model and BDE-209 exposure
Healthy embryos, collected from copulatory adult zebrafish (Tubingen strain) soon after light stimulation, were chosen for the following exposure from 3-5 hours post fertilization (hpf) to 7 dpf. One control group and two BDE-209 exposure groups were employed in the assay. The vehicle of the control group was 10% Hanks’ solution with 0.1 % DMSO. BDE-209 stock solution of 5.0 × 103 mg/L was prepared by dissolving 5 mg BDE-209 (98.3%, AccuStandard, USA) in 1 mL DMSO completely and then diluted 1000 times to 5 mg/L using sterile water. The nominal concentration of two BDE-209 exposure groups was 5 μg/L and 50 μg/L respectively with the same vehicle. The nominal concentrations were selected according to previous studies . Every group had 50 embryos in a glass petri dish with 30 mL exposure solution, which was half renewed every day in the exposure duration. All experiments were performed in triplicates.
The BDE-209 standard was purchased from AccuStandard. 5 mL exposure solution (1 and 3 dpf), collected from two treatment groups, was subjected to liquid−liquid extraction with 5 ml toluene (HPLC grade, Merck) for three repetitions. Then the extracts were combined and dried under a gentle stream of nitrogen and reconstituted with 200 μL of toluene for the later analysis. The analysis was performed using a GC-MS with ECNI mode (Agilent, 7890A-5875C) . A concentrated extract (1 μL) was injected onto the column (DB-5HT, 15 m × 0.25 mm, 0.1 µm). The initial temperature was 100 °C and then ramping at 30 °C/min to 320 °C followed by isothermal elution at 320 °C for 6 min. The SSL injector temperature was 280 °C and the MS transfer line temperature was isothermal at 320 °C. The carrier gas was helium at 1 mL/min.
Neurobehavioral test protocol
Neurobehavioral tests were performed on a ZebraBox platform (ViewPoint, France) according to our previous methods [20, 21], which can realize automatic tracking and high-throughput screening of larvae. Locomotion, path angle, and two-fish social activity of the same batch of larvae were adopted to evaluate the neurobehavioral effects induced by BDE-209 on exposed larvae at 5, 6, and 7 dpf respectively with light stimulus. The protocol of light stimulus included an initial 10 min of light adaption, followed by three repeated cycles with 10 min of dark period and 10 min of light period.
In the locomotion and path angle tests, larvae were transferred into a 48-well microplate from the glass petri dish at 5 dpf. Each well accommodated one larva with 1 mL exposure solution and each group had 16 larvae. The system can record the swimming distance of larvae as the output of locomotion test. For path angle test, three angle classes, namely straight motion (-10° ~ 0°, 0° ~ +10°), average turn (-10° ~ -90°, +10° ~ +90°) and responsive turn (-180°~ -90°, +90°~ +180°), were defined in this test. The system can record the frequency of different classes during the whole test.
The two-fish social activity test was performed with 6-well microplates, where two larvae were placed in each well of the plate. A valid social contact was recorded by the system when the distance of two larvae was less than 0.5 cm. The system can record the frequency of valid contacts and the duration of each contact in the total test period.
Quantitative real-time PCR (qRT-PCR) analysis
At 7 dpf, the expression of five opsin-coding genes (zfgr1, zfblue, zfuv, zfred, and rho) after BDE-209 exposure was investigated using the qRT-PCR experiment. Trizol (Invitrogen, USA) was used to extract total RNA, and cDNA reverse transcription was performed according to the instructor of High Capacity cDNA Reverse Transcription Kits (ABI, USA). The threshold cycle values for selected genes and housekeeping β-actin were used to calculate the relative RNA amounts. The analysis was performed on a 7500 Real-Time PCR System (Applied Biosystems, USA). Fold changes of genes were defined as the ratio of RNA amounts in the treatments versus the control. More details including the sequences of primers were given in Table S1 in the Supporting Information.
Matlab R2019a (Mathworks, USA) was used to pre-process the raw data of larval behaviors from the Zebrabox system, and formed a processed file for statistical analysis of different behaviors. The 2-ΔΔt method was used for the relative quantification analysis of the qRT-PCR data. The software GraphPad Prism 7 was used for statistical analysis and graph drawing. The statistical differences between the treatment groups and the control group were analysed using one-way ANOVA followed by Dunnett’s test and the differences were considered significant when the p value was < 0.05. Error bars in the figures represent the standard errors of mean (SEM).