Fig. 1. In vivo anti-seizure pipeline to test curcumin analogs (A). Illustration of the structural modifications in two notable analogue families used in this study in comparison to the base compound curcumin. (B) Our pipeline follows a funnel approach that whittled down an initial 68 analogues to two promising candidates using zebrafish larval behavioural and neuronal hyperactivity assays.

Fig. 2. Anti-seizure property of curcumin in vivo (A) Timeline of the experimental procedure corresponding to our PTZ behavioural hyperactivity assay. (B & C) Quantification of the total distance travelled (m) and maximum acceleration (m/s²) of 5 dpf zebrafish larvae treated or not with PTZ 12 mM and pre-incubated or not with curcumin at different concentrations (0 – 15 µM) The * ** * marker denotes significance at P < 0.0001 respectively as compared to the 12 mM PTZ only group using one-way Analysis of Variance (ANOVA) followed by Dunnett's post hoc test. (D) Timeline of the experimental procedure corresponding to our behavioural hyperactivity assay using gabra1-/- and gabrg2-/- genetic seizure models. (E, F) Quantification of the maximum acceleration (m/s²) of gabra1 + /+ and -/- (E) or gabrg2 + /+ and -/- (F) larvae. Larvae were incubated with different concentration of curcumin (10 – 15 µM). The * ** * marker denotes significance at P < 0.0001 respectively as compared to the homozygous (-/-) group using one-way Analysis of Variance (ANOVA) followed by Dunnett's post hoc test. VHC (vehicle) treatments correspond to 0.1% DMSO.

Fig. 3. In vivo screening of 68 curcumin analogs on PTZ-induced acute seizures (A) Timeline of the experimental procedure corresponding to our PTZ behavioural hyperactivity assay. (B & C) Quantification of the total distance travelled (m) and maximum acceleration (m/s²) respectively of 5 dpf zebrafish larvae treated or not with PTZ 12 mM and pre-incubated or not with analogues. The analogues were applied at a concentration of 5 µM except for several analogues that were applied at 4 µM (MS 65), 3 µM (MS 72 C and MS 40 B), 2 µM (CA 22, CA 98(3), MS 55, MS 4 and MS 33 A), 1 µM (MS 52) and 0.5 µM (MS 33 C) due to their toxicity to zebrafish larvae at higher concentrations. Green indicates significant changes with the * , * *, * ** and * ** * markers denoting significance at P < 0.05, P < 0.01, P < 0.001 and P < 0.0001 respectively as compared to the 12 mM PTZ only group using one-way ANOVA followed by Dunnett's post hoc test.

Fig. 4. A limited number of analogues are active against genetic seizure models. (A) Timeline of the experimental procedure corresponding to our behavioural hyperactivity assay using gabra1-/- and gabrg2-/- genetic seizure models. (B) Quantification of the maximum acceleration (m/s²) of gabra1 + /+ and -/- (B) or gabrg2 + /+ and -/- (C) larvae. Larvae have been pre-incubated with vehicle (VHC) or different analogues (2 – 5 µM). Green indicates significant changes with the * , * *, * ** and * ** * markers denoting significance at P < 0.05, P < 0.01, P < 0.001 and P < 0.0001 using one-way Analysis of Variance (ANOVA) followed by the Šídák multiple comparison test in a plate wise manner.

Fig. 5. CA 80(1) and CA 20(1) are preventing brain seizures in gabrg2-/- larvae. (A) Timeline of the experimental procedure corresponding to our brain activity assay using gabra1-/- and gabrg2-/- genetic seizure models that have been crossed with the [neurod:GCaMP6f] transgenic line [1]. (B, C) Quantification of the fluorescence intensity upon LASER excitation (averaged normalised mean grey value for 2 s) of gabra1 and gabrg2 5 dpf larvae that have been pre-incubated with vehicle (VHC) or analogues at 2–5 µM. The * ** * marker denotes significance at P < 0.0001 between the indicated groups using two-way Analysis of Variance (ANOVA). (D & E) Single confocal images corresponding to the first 400 ms after LASER light exposure showing neuronal activity of gabra1 and gabrg2 mutants pre-incubated with vehicle (VHC) or specific analogues.