Figure 7. Chemical disruption of the habenula reduces hunting behavior in prey-experienced fish.

(A) Representative image of 7 dpf Tg(gng8:Gal4;UAS:NTR-mCherry) fish, dorsal view. NTR-mCherry is depicted in red, image is a maximum intensity projection of a Z-stack. Additional example image is shown in Figure 7—figure supplement 1C). Scale bar is 100 μm. (B) Representative images of NTR-mCherry expression (red) in the habenula in NTR+ fish after 19 hr in either 0.2% DMSO (-MTZ) (top) or 5 mM MTZ in 0.2% DMSO (+MTZ) (bottom). Images are maximum intensity projections of 5 µm/slice Z-stacks. Scale bar is 40 μm. Additional example images are shown in Figure 7—figure supplement 2A. (C) Volume of NTR-mCherry fluorescence is reduced in NTR-mCherry expressing fish treated with MTZ, p = 0.003, N = 10 animals per group. Differences in signal likely reflect variable transgene expression. Dependence of volume measurement on total fluorescence differs between +MTZ and -MTZ (Figure 7—figure supplement 2B). Symbols indicate individual fish. Box plot shows median, 25th and 75th percentiles. (D, E) Cumulative average number of eye convergences after addition of paramecia in 7 dpf control (NTR-) fish (D) and Tg(gng8:Gal4;UAS:NTR-mCherry) (NTR+) siblings (E), pretreated for 19 hours with either 0.2% DMSO alone (-MTZ) or 0.2% DMSO containing 5 mM metronidazole (+MTZ) and tested for 20 minutes one hour after washout of the drug. Eye convergence rate is reduced in MTZ-treated enzyme expressing fish (NTR+/+MTZ) compared to untreated (NTR+/-MTZ) animals (two-way repeated measures ANOVA shows an effect of treatment with p = 0.033), while we observe no significant difference for NTR- animals (two-way repeated measures ANOVA, effect of treatment p = 0.24). (n) = number of fish in each group. Error bars show SEM. (F) Percent time of eyes converged over 20 min of recording period in NTR+ fish is significantly lower in NTR+/+MTZ fish than in NTR+/-MTZ (p = 0.004. Symbols indicate individual fish. Box plot shows median and 25th and 75th percentiles. A permutation test is used for all pairwise comparisons unless otherwise specified (see Materials and methods, Behavioral data analysis and statistics). Data tables for panels C, D, E, and F in Figure 7—source data 1.

Figure 7—figure supplement 1. Expression pattern of Tg(gng8:Gal4;UAS:NTR-mCherry) fish, paramecia consumption and swim behavior.

(A–B) Paramecia consumption over 20 min. (A) NTR- fish showed no significant difference between +MTZ or -MTZ-treated groups (two-way repeated measures ANOVA, effect of treatmet p = 0.33). Error bars are SEM. (B) Nitroreductase expressing NTR+/+MTZ fish consume paramecia more slowly than NTR+/-MTZ fish (minute 10: p = 0.036; minute 15: p = 0.040; minute 20: p = 0.051; two-way repeated measures ANOVA, effect of treatment: p = 0.073). (C) Expression of NTR-mCherry in Tg(gng8:Gal4;UAS:NTR-mCherry) fish is specific to the habenula. Image is a maximum intensity projection of a Z-stack, NTR expression is shown in red. Scale bar is 200 µm. (D) The number of paramecia initially added to the wells was not significantly different between the +MTZ and -MTZ-treated groups, for either the NTR+ fish (p = 0.69, left side) or NTR- fish (p = 0.74, right side). Every dot represents one fish, boxplots show median and 25^th and 75^th percentiles. (E–F) No significant difference in average swim speed between +MTZ and -MTZ fish in NTR+ animals during the 10 min acclimation period before addition of paramecia (E) (p = 0.46) or during the 20 min prey capture period after addition of paramecia (F) (p = 0.29). Dots represent individual fish, boxplots show median and 25^th and 75^th percentiles. (G) Habenula ablation does not affect paramecia consumption in non-experienced fish. Two-way repeated measures ANOVA shows no effect of treatment, p = 0.37. Error bars are SEM. A permutation test was used for all pairwise comparisons unless specified otherwise (see Materials and methods, Behavioral data analysis and statistics). Data tables for panels A, B, D, E, and F are in Figure 7—figure supplement 1—source data 1.

Figure 7—figure supplement 2. Differences in mCherry signal intensity and correlation with expression volume.

(A) Examples of mCherry pattern in healthy control habenulae of NTR+ fish treated with vehicle only (-MTZ) and metronidazole-treated (+MTZ). Although brightness levels vary, the -MTZ controls show a typical diffuse pattern, whereas the +MTZ-treated animals have a blotchy distribution. Scale bar represents 40 μm. (B) Relation between total fluorescent intensity in maximum projection of Z-stacks and calculated expression volume. Symbols represent individual fish, a permutation test was used for all pairwise comparisons unless specified otherwise (see Materials and methods, Behavioral data analysis and statistics). Data table for panel B is in (Figure 7—figure supplement 2—source data 1).