Adult wild-type zebrafish were obtained from Exopet (Madrid, Spain) and maintained in fish water [reverse-osmosis purified water containing 90 μg/ml of Instant Ocean (Aquarium Systems, Sarrebourg, France) and 0.58 mM CaSO₄.2H₂O] at 28 ± 1°C in the Research and Development Centre of the Spanish Research Council (CID-CSIC) facilities under standard conditions. Embryos were obtained by natural mating, collected using a mesh strainer and maintained in renewed fish water (pH 7.8 ± 0.2 and a conductivity of 800 ± 55 µS) in a thermostatic chamber (POL-EKO APARATURA Climatic chamber KK350, Poland) at 28.5°C on a 12L:12D photoperiod at one embryo/mL ratio. Zebrafish embryos under these conditions begin hatching from 48 to 72hpf, after which they are considered larvae (Kimmel et al., 1995). Embryo/larvae were not fed during the experimental period to avoid interference with exposure concentrations of SER. All procedures were approved by the Institutional Animal Care and Use Committees at the CID-CSIC and conducted following the institutional guidelines under a license from the local government (agreement number 11336).

Sertraline Hydrochloride (CAS: 79559-97-0, ≥98%) (SER) was obtained from (Merck, United States ). Experimental concentrations in fish water of the four nominal concentrations used in this study (0.01 μg/L, 0.1 μg/L, 1 μg/L, and 10 μg/L), with three replicates for each concentration, were determined using ultra-high-performance liquid chromatography with triple quadrupole detector (UHPLC-MS/MS). Furthermore, the stability of SER in fish water was assessed by preparing triplicate solutions of the four selected concentrations, which were maintained under the exposure conditions (28 °C and 12L:12D photoperiod). Aliquots of these solutions were analyzed at times 0, 24, and 48 h.

Stock solutions of SER were prepared in dimethyl sulfoxide (DMSO), and then diluted in fish water to a final carrier concentration of 0.1% in all treatments. The four selected SER concentrations are within the range of those found in surface waters and those detected in fish plasma (Valenti et al., 2012).

Two independent experiments were carried out. All exposures were performed at 28.5°C (POL-EKO APARATURA Climatic chamber KK350, Poland) with 12L:12D photoperiod. For each assay, samples for each analysis were collected from at least two independent experiments. Zebrafish embryos were previously sorted to the same developmental of the blastula period stage (approximately the oblong and sphere stage). Following this, embryos were then exposed to four environmental relevant concentrations of SER (0.01, 0.1, 1, and 10 μg/L) from four hpf (hours post fertilization) to eight dpf (days post fertilization). Exposure was conducted in 6-well plates with 10 embryos per well containing 10 ml of medium, and a total of 12 wells for each treatment. The medium was changed every 48 h. Survival and hatching were recorded every 24 h starting from the beginning of the exposure, according to the OECD 212, “Test Guideline 212”, 1998 (Test No. 212: Fish, Short-term Toxicity Test on Embryo and Sac-Fry Stages, 1998) standards, along with any obvious morphological abnormalities. Total body length and truck thickness was measured in 12–20 larvae at the end of the exposure (8 dpf). Larvae selected for behavioral tests were transferred 24 h before trials to 48-well plates where one larva was placed in each well containing 1 ml of exposure medium.

Zebrafish larvae were collected and fixed overnight at 4 °C in phosphate buffered saline 1× solution (PBS) with 4% PFA (paraformaldehyde) as previously described (Garcia-Reyero et al., 2014). For detailed morphological observation and larvae measurements, fixed larvae were then washed several times with PBS and gradually transferred to 90% glycerol in PBS for long-term preservation and positioning under the stereomicroscope.

All individuals were observed and photographed under a Nikon SMZ1500 (NIKON Instruments INC. New York, United States ) dissecting microscope fitted with Nikon Digital Sight DSRi1 camera and NIS Elements AR software (version 3.0) (NIKON Instruments INC. New York, United States ) and saved as high resolution (3840 × 3005 pixels) tagged image file format (TIFF). All images were taken on the left side of each larva. The total body length (anterior-most part of the snout to posterior-most point of the tail, mm) and trunk thickness (mm) were obtained using a measuring tool from the NIS Elements software.

Experiments were conducted in 48-well plates at one larva per well, maintained under dark before transferring them to a behavioral testing chamber equipped with a temperature control unit set to 28°C (DanioVision, Noldus Information Technology, Leesburg, VA). Once inside the chamber, larvae were left to acclimate in the dark for a further 10 min before initiating video recording.

The vibrational startle response assay was performed under near-infrared light as described in Faria et al., 2019) (n = 79–82). 50 consecutive tapping stimuli selected at the highest intensity (intensity level: 8) were delivered every second. Videos were recorded at 30 frames per second and the vibrational startle response (VSR or Startle) was analyzed for each larva by leveling basal movement to Y = 0 and then measuring the distance moved (cm) over the 1 s period following the first stimulus, while the habituation of the startle responses was measured as the area under the curve (AUC) of plots of distance moved relative to the response of the first stimulus (n = 76–82). Video tracking and measurement of the startle response were analyzed using the EthoVision XT 13 software (Noldus, Wageningen, Netherlands).

Basal locomotor activity (BLM) and visual motor response (VMR) analyses of eight dpf zebrafish larvae were performed also using a DanioVision system running an Ethovision XT 13 software essentially as described by Faria et al., 2015). The BLM (n = 76–83), represents the total distance (cm) traveled by each larva during 10min under dark conditions, while the VMR (n = 68–84) reports larvae responses following a transition of light to dark and is represented as the difference of the total distance (cm) traveled during 2 minutes after and before the transition. Video tracking conditions consisted of a 40 min cycle including a 15 min dark period, followed by a 10 min light period and then a second 15 min dark cycle. The position of each larva was recorded using an IR digital video camera Basler acA1300-60 g m (Basler Inc., Exton, PA) and an EthoVision XT 13 video tracking system. All measurements were performed between 10:00 and 16:00 h. Tracks were analyzed in terms of total distance moved (cm) calculated for each dark or light period. All microplates were analyzed at 28 ± 0.5 °C with the same detection and acquisition settings.

Gene expression analysis was performed as previously reported (Prats et al., 2017). Total RNA was extracted from a total of six pools of four larvae (8 dpf) using the Trizol Reagent (Invitrogen Life Technologies, Carlsbad, CA). RNA concentration was measured by spectrophotometric absorption at 260 nm in a NanoDrop™ ND-8000 spectrophotometer (Fisher Scientific) and the quality was checked in an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA). RIN (RNA Integrity Number) values ranged between 9 and 10. After DNase I treatment (Ambion, Austin, TX), 1 µg of total RNA was used to synthesize the first strand of cDNA with the First Strand cDNA synthesis Kit (Roche Diagnostics, Mannheim, Germany) using oligo (dT), following the manufacturer’s instructions.

Real-Time PCR was performed in a LightCycler®480 Realtime PCR System using SYBR Green PCR Master Mix (Roche Diagnostics, Mannheim, Germany). Cycling parameters were 95°C for 15 min followed by 45 cycles of 95°C, 10 s and 60°C, 30 s. For each experimental condition, qPCR analyses were performed from two independent experiments, with six biological replicates on each experiment and three technical replicates for each sample. Primer sequences of the seven selected genes related to the monoaminergic system (tph1a, sert (or slc6a4a) and vmat2 (or slc18a2), dat (or slc6a3), dbh, th1 and th2) are reported in Table 1. The housekeeping gene ppiaa was used as a reference gene for normalization purposes (Prats et al., 2017). Primers were synthesized by Sigma-Aldrich (Steinheim, Germany). The efficiency and specificity of all primers were checked before the analyses. Results were normalized to ppiaa and the relative abundance of mRNA was calculated following the ΔΔCt method (Livak and Schmittgen, 2001) deriving fold-change ratios from these values.

Crystalline solid standards and internal labeled standards (IS) of the selected monoaminergic neurochemicals (Table 2) were obtained from Toronto Research Chemicals (TRC, Toronto, Canada), Merck (Darmstadt, Germany), and Sigma-Aldrich (St. Louis, MO, United States ), detailed information can be found in a previous study (Mayol-Cabré et al., 2020). Monoaminergic neurochemicals were extracted from four to six pools of 15 larvae heads following an extraction procedure adapted from a previous study (Bellot et al., 2021). Briefly, larvae were euthanized by chilling prior to decapitation, larvae were then positioned on the lateral side under a dissecting microscope. Heads were sectioned immediately from caudal to the otic vesicle and cranial to the anterior intestine, pooled and collected into a 1.5 ml tube. The excess medium was removed and heads were immediately frozen in dry ice and stored at -80°C. The extraction process was based on the use of a solvent of polarity similar enough to the neurotransmitters to be able to extract them from the sample. Next, 300 μL of cold extractant solvent (ACN:H₂O (90:10) + 1% formic acid) were added to pools of larvae heads. Each sample was spiked with 50 ng of an internal standard mixture (ISM), except for 5-HT-d₄ which was spiked at 20 ng. Three stainless steel beads (3 mm diameter) were placed in each pool and samples were homogenized using a bead mill homogenizer (TissueLyser LT, Quiagen, Hilden, Germany) at 50 osc/s for 90s and centrifuged for 20 min at 15870 g at 4°C. Finally, the supernatant was filtered using 0.20 μm PTFE filters (DISMIC -13 JP, Advantec^®) and kept at -80°C until the LC-MS/MS analysis. Temperature is a key element throughout the extraction process because neurotransmitters are very unstable and high temperatures can degrade these compounds. As commented before, the analysis was performed by ultra-high-performance liquid chromatography (Acquity UPLC H-Class Waters, Milford, MA, United States ) coupled to a triple quadrupole mass spectrometer equipped with an electrospray (ESI) source (Xevo TQ-S micro, Waters, United States ) (Mayol-Cabré et al., 2020; Bellot et al., 2021). Neurotransmitter levels were then normalized by total protein determined using Bradford Method (Bradford, 1976), and represented as ng/mg of protein.

Highly pure (>99%) SER analytical standard and sertraline-d₃ internal standard were used to define its stability. Both standards (native and labeled) were purchased from Sigma-Aldrich (St Louis, Missouri, U.S.). Stock solution of SER was prepared in 100% dimethyl sulfoxide (DMSO, Merck Darmstadt, Germany). Individual IS was directly obtained as a solution (1 mg/ml) in acetonitrile (ACN). Work solution of IS was prepared at a concentration of 2 mg/L. For the chromatographic separation and the MS analysis, high purity mobile phase solutions were prepared using ACN and HPLC water (Optima™ LC-MS Grade) purchased from Fisher Chemical (Fisher Scientific SL, Madrid, Spain). For SER stability, serial dilutions were prepared from a stock of 1490 mg/L. The spiked fish water was prepared with SER concentrations from 0.01 ng/ml to 10 ng/ml and its stability was evaluated at 0, 24, and 48 h in triplicate. Aliquots of 1 ml of spiked water were added to a 2-ml vial in which 5 µL of 2 mg/L sertraline-d₃ solution in ACN were previously evaporated (Final IS concentration: 10 ng/ml). Vials were accurately vortexed for 2 min using a BenchMixer XLQ QuEChERS Vortexer (Benchmark Scientific, Sayreville NJ, U.S.).

Data were analyzed with IBM SPSS v25 (Statistical Package 2010; Chicago, IL), and plotted with GraphPad Prism 8.31 for Windows (GraphPad Software Inc, La Jolla, CA) or with Microsoft Excel 2016. Normality was assessed using Kolmogorov-Smirnov and Shapiro-Wilk tests. Multiple comparison tests were used to determine differences between normally distributed groups. One-way ANOVA followed by Tukey’s post hoc was applied for groups meeting parametric requirements, whereas the Kruskal–Wallis test followed by Dunn’s multiple comparison test was used for those groups who did not meet parametric assumptions. Water samples were analyzed by pairwise Students’ t-test. Significance was set at p < 0.05.

Nominal concentrations of SER in fish water are presented in Table 3. The lowest concentration, 0.01 μg/L, was below detection limits. The actual concentration corresponding with the nominal concentration of 0.1 μg/L was somewhat higher, this can be because it was not possible to eliminate or effectively compensate for the phenomena of suppression/enhancement of the matrix (assuming using matrix-matched calibration combined with internal standards) due to the high background noise due to its proximity to the detection limit. The remaining experimental concentrations ranged from 102 to 104% of nominal concentrations. SER concentration in fish water following 24 and 48 h remained stable for all conditions except for concentrations 0.1 and 10 μg/L, were it decreased 19.8%–22.0% for 0.1 μg/L, and 12.0%–10.5% for 10 μg/L at 24 and 48 h respectively.

Zebrafish embryos were exposed to four environmentally relevant concentrations of SER, from four hpf to eight dpf. Mortality was recorded throughout the whole exposure period and no significance was found for any condition relative to the control (Figure 1B). The lowest accumulated mortality was that of the control registering 6.5 ± 6.2%, while the highest was that of 10 μg/L registering 13.3 ± 9.8% (p = 0.121). The hatching rate was also unaffected by SER exposure. In general, at 48 hpf 88% of the surviving embryos had already hatched, reaching 100% at 72 hpf (Figure 1A). At the end of the developmental exposure, zebrafish larvae were fixed and conserved in 90% glycerol for more detailed morphological observations and size measurement. As Figure 2 shows, no changes in larvae morphology was observed across all experimental conditions. Furthermore, larvae’ total length and truck thickness were unaffected by SER developmental exposure (Table 4).

Four larvae behaviors were analyzed to assess the effects of early life exposure to SER: 1) the escape response evoked by a vibrational acoustic stimulus, also known as the startle response, 2) habituation to repetitive acoustic stimulation which addresses non-associative learning, 3) basal locomotor activity (BLM), and 4) response to visual stimuli (VMR) (Figure 3). The startle response was the only behavioral parameter that passed the normality test (p = 0.160, Shapiro-Wilk) and therefore parametric statistical analysis was applied. One-way ANOVA analysis revealed statistical differences between groups (F _(4.397) = 2.998, p = 0.019). The post hoc test showed a very mild effect in larvae startle, where it was found significantly decreased at only 0.1 μg/L (p = 0.045), while the remaining concentrations did not alter this response. Kruskal–Wallis non parametric test divulged significant changes for larvae habituation (H (4) = 19.685, p = 0.001) and BLM (H (4) = 16.457, p = 0.002) following SER exposure. A hormetic pattern of response was observed for habituation. A pairwise comparison test showed that the time needed for habituation significantly increased at 0.1 (p = 0.004) and 1 μg/L (p < 0.001) and then recovered back to control levels at 10 μg/L. On the other hand, larvae showed a significant decrease in locomotor activity from 0.1 μg/L and upwards (p < 0.01). Finally, Figure 3 shows that SER does not affect VMR (full plot is available in Supplementary Material S1).

The expression of four genes playing a key role in the serotonergic system was analyzed (Figure 4A). No variation was observed for the reference gene (ppiaa) across all treatments (F _(4.25) = 2.655, p = 0.0656, one-way ANOVA). One-way ANOVA analysis showed that only tph1a (gene encoding for tryptophan hydroxylase 1a) was significantly affected by developmental exposure to SER (F _(4.25) = 4.449, p = 0.007). The observed difference was mainly due to the downregulation of its expression at the highest exposure concentration (p = 0.017). Similarly, SER also showed a very mild effect on the expression of genes related to the dopaminergic system (Figure 5). One-way ANOVA analysis revealed that only th2 (tyrosine hydroxylase 2) gene gave differences between groups (F _(4.24) = 5.734, p = 0.002), which contributed to the strong downregulation caused by SER 10 μg/L (p = 0.038) (Figure 5).

Sertraline’s effect on serotonergic (Figure 4B) and dopaminergic (Figure 6) signaling was then assessed. In general, both pathways were significantly altered due to exposure to the pharmaceutical during early life. For the serotonergic system, levels of both serotonin (5-HT) (F_(4,23) = 4.036, p = 0.013) and its metabolite (5-HIAA) (F_(4,20) = 7.935, p = 0.001) increased with SER exposure. These differences were related to the significant increase in 5-HT levels observed in larvae exposed at the two highest concentrations (p < 0.030). In addition, an increase of 5-HIAA levels was also dose-dependent, where the highest reported level was reported for the highest concentration of SER (Figure 4B). Interestingly, the major effect of SER on the dopaminergic neurochemical profile was the significant increase in all three metabolites of dopamine: DOPAC (F_(4.21) = 10.757, p < 0.001), 3-MT (F_(4.21) = 3.663, p = 0.021) and HVA (F_(4.22) = 3.357, p = 0.027) (Figure 6). Levels of norepinephrine, a catecholamine synthesized from dopamine, were also found to be increased by SER (F_(4.22) = 9.965, p < 0.001). The observed increase in dopamine metabolite levels was overall dose-dependent. Effects over DOPAC and 3-MT occurred already at the lowest exposure concentration, especially for DOPAC (p = 0.011) (Figure 6). The highest level of 3-MT was reported at 0.1 μg/L (p = 0.007) and then remained relatively stable at 1 and 10 μg/L. On the other hand, for DOPAC, the highest reported levels were observed for 1 and 10 μg/L (p < 0.001, for both). In the same line as that observed for DOPAC and 3-MT, HVA levels showed a dose-dependent increase, reaching the highest levels at 1 (p = 0.019) and 10 μg/L (p = 0.022) (Figure 6). Very similar to the above neurochemicals, norepinephrine levels also increased notably at 1 and 10 μg/L (p < 0.001, for both). Curiously, despite the observed augmented levels of its transformation products and metabolites, dopamine levels in larvae heads remained unchanged.