Susceptibility of larval zebrafish to the seizurogenic activity of GABA type A receptor antagonists

Abstract

Previous studies demonstrated that pentylenetetrazole (PTZ), a GABA type A receptor (GABA_AR) antagonist, elicits seizure-like phenotypes in larval zebrafish (Danio rerio). Here, we determined whether the GABA_AR antagonists, tetramethylenedisulfotetramine (TETS) and picrotoxin (PTX), both listed as credible chemical threat agents, similarly trigger seizures in zebrafish larvae. Larvae of three, routinely used laboratory zebrafish lines, Tropical 5D, NHGRI and Tupfel long fin, were exposed to varying concentrations of PTZ (used as a positive control), PTX or TETS for 20 min at 5 days post fertilization (dpf). Acute exposure to PTZ, PTX or TETS triggered seizure behavior in the absence of morbidity or mortality. While the concentration-effect relationship for seizure behavior was similar across zebrafish lines for each GABA_AR antagonist, significantly less TETS was required to trigger seizures relative to PTX or PTZ. Recordings of extracellular field potentials in the optic tectum of 5 dpf Tropical 5D zebrafish confirmed that all three GABA_AR antagonists elicited extracellular spiking patterns consistent with seizure activity, although the pattern varied between chemicals. Post-exposure treatment with the GABA_AR positive allosteric modulators (PAMs), diazepam, midazolam or allopregnanolone, attenuated seizure behavior and activity but did not completely normalize electrical field recordings in the optic tectum. These data are consistent with observations of seizure responses in mammalian models exposed to these same GABA_AR antagonists and PAMs, further validating larval zebrafish as a higher throughput-screening platform for antiseizure therapeutics, and demonstrating its appropriateness for identifying improved countermeasures for TETS and other convulsant chemical threat agents that trigger seizures via GABA_AR antagonism.

Keywords: Chemical threat agents; Picrotoxin; Seizures; TETS; Tetramethylenedisulfotetramine; Zebrafish.

Figure 1.. GABA_AR antagonists cause concentration-dependent morphological defects and lethality.

At 5 dpf, Tropical 5D larvae were exposed to convulsant chemicals for 1 or 24 h. At the end of the exposure period, larvae were evaluated for morphological defects and lethality. (A) Representative examples of morphologies evaluated using the following grading system to score morbidity and mortality: Grade 0 - no overt morphological abnormality; Grade 1 - mild defects in gross morphology, with slight axial deformation; Grade 2 - pronounced axial deformation (high curvature) and/or decreased trunk length; Grade 3 – marked axial deformation and/or death. Quantitative analyses of the incidence of morphological defects after 1 h (B) or 24 h (C) continuous exposure to GABA_AR antagonists (n=24 per group from ≥ 3 independent spawning events).

Figure 2.. GABA_AR antagonists trigger seizures across three different lines of wildtype zebrafish larvae.

(A) Larvae from three different commonly used lines of wildtype zebrafish (Tropical 5D, NHGRI and Tupfel long fin) were placed in 96-well plates at 24 hpf. At 5 dpf, larvae were exposed to increasing concentrations of PTZ, PTX or TETS for 20 min. Seizure behavior was scored throughout the 20 min exposure period using the scoring system described in Table 1. (B) Seizure scores in zebrafish larvae exposed to PTZ (left panel), PTX (middle panel), and TETS (right panel) reveal similar concentration-response curves between wildtype fish lines for each GABA_AR antagonist tested. Data are presented as mean ± S.D. (n = 24 per group from ≥ 3 independent spawning events).

Figure 3.. TETS is a potent chemical convulsant in zebrafish larvae.

(A) Tropical 5D, NHGRI and Tupfel long fin zebrafish larvae were placed in 96-well plates at 24 hpf. At 5 dpf, larval zebrafish were exposed to increasing concentrations of PTZ, PTX and TETS for 20 min. (B) Percentage of animals within each zebrafish line that exhibited a seizure score of 2 following exposure to increasing concentrations of each GABA_AR antagonist (n = 16–28 per group from ≥ 3 independent spawning events). (C) CC₅₀ and CC₉₀, GABA_AR antagonist concentrations at which 50% or 90% of 5 dpf zebrafish exhibit a seizure score of 2. Data presented as the estimated concentration and corresponding 95% confidence interval.

Figure 4.. The latency to a seizure score of 2 decreases with increasing concentrations of GABA_AR antagonists.

Latency to the first sign of behavior consistent with a seizure score of 2 that persisted at least 0.2 seconds was recorded in Tropical 5D, NHGRI and Tupfel long fin zebrafish larvae exposed to varying concentrations of PTZ, PTX or TETS. For all three convulsant chemicals, latency was significantly reduced at higher concentrations. Data are presented as box plots, in which dots represent individual data points; the horizontal line in the box, the median; the ends of the box, the interquartile range (IQR), and the vertical lines extending to the last observation within 1.5 IQR of the ends of the box (n = 5–32 per group from ≥ 3 independent spawning events). Although we emphasize the groups with the largest difference from vehicle control in the figure, a few other differences remained significant after FDR correction. Tropical 5D larvae exposed to 0.1 mM PTX (p=0.006) or 1 μM TETS (p=0.02) had lower latency than vehicle controls as did NHGRI larvae exposed to 0.4 mM PTX (p=0.009) or 1 μM TETS (p=0.01) and Tupfel long fin larvae exposed to 0.4 mM PTX (p=0.007). ***Significantly different from vehicle control (0 μM) at p < 0.001.

Figure 5.. The duration of GABA_AR antagonist-induced seizure behavior scored as 2 exhibits a non-monotonic concentration-response relationship.

Total time that fish exhibited behavior consistent with a seizure score of 2 during 20 min of exposure to varying concentrations of PTZ, PTX or TETS was recorded for 5 dpf Tropical 5D, NHGRI and Tupfel long fin zebrafish. Data are presented as box plots, in which dots represent individual data points; the horizontal line in the box, the median; the ends of the box, the interquartile range, and the vertical lines extending to the last observation within 1.5 IQR of the ends of the box (n = 6–47 per group from ≥ 3 independent spawning events). Although we emphasize the groups with the largest difference from vehicle control in the figure, a few other differences remained significant after FDR correction, all for the NHGRI larvae. In particular, the NHGRI larvae exposed to 1 mM of PTZ (p=0.007 ), 0.1 mM of PTX (p=0.002), or 0.4 μM TETS (p=0.01 ) had longer seizures than vehicle control. ***Significantly different from vehicle control (0 μM) at p < 0.001.

Figure 6.. GABA_AR antagonists evoke epileptiform-like electrographic activity in larval zebrafish.

(A) For electrophysiological assessments, 5 dpf Tropical 5D zebrafish larvae were individually exposed to D-tubocurarine (D-tub) for 10 min then embedded in 1.5% agar. Field recordings from the forebrain were collected for 20 min in the presence of GABA_AR antagonists. (B) Representative tectal field recordings from vehicle control zebrafish larvae (no seizure development) and from zebrafish larvae exposed to a chemical convulsant. (C) Burst frequency was averaged for the last 10 min of recording (n = 5–8 per group from ≥ 3 independent spawning events). Data are presented as mean ± SD. ***Significantly different from vehicle control (0 μM) at p < 0.001. #Significantly different from PTX (0.4 NM) at p < 0.001.

Figure 7.. Post-exposure exposure to benzodiazepines attenuates PTX and TETS, but not PTX-induced epileptic activity in vivo.

(A) For electrophysiological assessments, 5 dpf Tropical 5D zebrafish larvae were exposed to D-tubocurarine (D-tub) and then embedded in 1.5% low melting agarose. After recording baseline tectal field potentials for 5 min, larvae were exposed to 10 mM PTZ, 0.4 mM PTX or 4 μM TETS for 20 min prior to adding diazepam (DZP), midazolam (MDZ) or allopregnanolone (ALLO) and recording for an additional 20 min. (B) Representative electrophysiological tracings after application of antiseizure agents (1 μM DZP, MDZ and ALLO for larvae exposed to PTZ or TETS; 100 μM DZP/MDZ pr 10 μM ALLO for larvae exposed to PTX). (C) Burst frequency was averaged for the last 10 min of each 20 min recording period. *Significantly different from GABA_AR antagonist-induced electrographic activity in the absence of antiseizure agent (black bars) at p < 0.05; **p < 0.01. Data are presented as mean ± S.D. (n = 3–4 per experimental condition from ≥ 3 independent spawning events).

Figure 8.. Chemical convulsant-induced locomotor activity exhibits a non-monotonic concentration-effect relationship.

(A) For automated assessments of locomotor behavior, Tropical 5D, NHGRI and Tupfel long fin zebrafish larvae were placed in 96-well plates at 24 hpf. At 5 dpf, larval zebrafish were exposed to varying concentrations of PTZ, PTX or TETS and locomotor behavior recorded for 20 min. (B) Locomotor behavior was quantified as the total distance moved over the 20 min recording period. Data are presented as box plots, in which dots represent individual data points; the horizontal line in the box, the median; the ends of the box, the interquartile range (IQR), and the vertical lines extending to the last observation within 1.5 IQR of the ends of the box (n = 6–47 per group from ≥ 3 independent spawning events). Although we emphasize the groups with the largest difference from vehicle control in the figure, a few other differences remained significant after FDR correction. In particular, Tropical 5D larvae exposed to 0.04 mM PTX (p=0.02) or 0.4 μM TETS (p=0.005) had more locomotor activity than vehicle controls as did NHGRI larvae exposed to 1 mM PTZ (p=0.006) or 0.1 mM PTX (p=0.01) and Tupfel long fin larvae exposed to 0.1 (p=0.02) or 0.4 (p=0.002) μM TETS. ***Significantly different from vehicle control (0 μM) at p < 0.001.

Figure 9.. Allopregnanolone suppresses GABA_AR antagonist-induced behavioral seizures for up to 1 h.

(A) For automated assessments of locomotor behavior, Tropical 5D zebrafish larvae were placed in 96-well plates at 24 hpf. At 5 dpf, larval zebrafish were exposed to 10 mM PTZ, 0.4 mM PTX or 4 μM TETS for 20 min prior to adding midazolam (MDZ), diazepam (DZP) or allopregnanolone (ALLO) and recording locomotor activity for an additional 60 min. (B) Locomotor behavior was quantified as the distance moved during the first, second and third 20 min blocks of the 60 min exposure to antiseizure agents. The black line in each graph indicates distance moved by fish exposed to vehicle (DMSO) in the absence of any chemical convulsant, while the red line indicates distance moved by fish exposed only to the chemical convulsant. Dashed lines in the ALLO plots indicate ALLO concentrations that are significantly different from the corresponding concentration in the other treatments. MDZ, DZP and ALLO significantly reduced the locomotor activity induced by exposure to PTZ, PTX, and TETS at most concentrations across the time period of observation (p<0.001); most reductions resulted in levels similar to or lower than that of vehicle controls (DMSO), particularly for ALLO. The figure emphasizes the largest differences, relative to vehicle controls, but additional comparisons survived the FDR correction for multiple comparisons including larvae exposed to PTZ treated with 0.3 μM (period 2) and 10 μM (periods 1 and 2) of MDZ (p<0.01), treated with DZP at all concentrations other than 10 μM during period 2 (p<0.02), and treated with 0.1 μM of ALLO in the second period (p=0.005); there were also differences in TETS exposed larvae treated with 0.3 μM (periods 1 and 2) or 3 μM (period 2) MDZ (p<0.03), 1 or 10 μM (period 2) DZP (p<0.01) or 3 μM ALLO (periods 1 and 2; p<0.01). ALLO is significantly more effective in reducing distance moved (p<0.001) in larvae exposed to PTZ compared to DZP and MDZ across all time blocks and all but the lowest concentration (0.1 μM). Similarly, all concentrations of ALLO are significantly more effective than DZP and MDZ in reducing locomotor activity induced by PTX (p<0.001). ALLO also was significantly more effective than DZP and MDZ in reducing locomotor activity triggered by TETS at all concentrations except 0.1 μM (p<0.001); ALLO caused a greater decrease in distance moved in the first period compared to DZP (p<0.001) and compared to DZP (p=0.002) or MDZ (p=0.02) in the third period. * Significantly different from control larvae exposed only to the chemical convulsant (p<0.001). + Significantly different from vehicle control larvae exposed only to DMSO (p<0.001).