Reelin immunohistochemistry. Anti-Reelin antibody labeling of coronal and sagittal sections of the adult zebrafish brain. In wild-type anti-Reelin antibody labeling is seen in the (a) medial part of the dorsal telencephalon (Dm) and in the ventral nucleus of the ventral telencephalon (Vm), in the (b) ventral thalamic nuclei (Th) and (c) the ventral hypothalamic nuclei (Hy). Labeling is also detected in the (d) torus longitudinalis (TL), (e) stratum fibrosum marginale (sfm) and the stratum opticum (so) of the optic tectum. (f) A few labeled cells are observed in the interpeduncular nucleus (nin). In the hindbrain, Reelin protein localizes to the (g) corpus cerebelli (CCe), the lobus caudalis cerebelli (LCa), crista cerebellaris (CC) and (h) the intermediate and inferior part of the reticular formation (RF). (i,j) Anti-Reelin antibody labeling is not detected in the brain of reln^–/– mutants. Black arrowheads show position of Reelin-positive areas in (b,c,f).

Social preference test. (A) Social preference. WT, reln^+/– and reln^–/– show a significant preference to spend time near a group of unfamiliar fish [1st strangers; p < 0.0001 for all genotypes; two-way ANOVA followed by Sidak’s post hoc, genotype factor: F(2,62) = 1.818, p = 0.17, stranger factor: F(1,62) = 2392, p < 0.0001, interaction genotype × stranger: F(2,62) = 4.418, p = 0.02]. (B) Preference for social novelty. Both WT and reln^+/– spend an equal amount of time near both groups of unfamiliar fish (1st and 2nd strangers; p = 0.99 for WT and p = 0.38 for reln^+/–). reln^–/– do not switch preference to the second group of unfamiliar fish [2nd strangers; p < 0.0001; two-way ANOVA followed by Sidak’s post hoc, genotype factor: F(2,62) = 0.5184, p = 0.60, stranger factor: F(1,62) = 21.49, p < 0.0001, interaction genotype × stranger: F(2,62) = 6.886, p = 0.002]. n = 12 wild-type, n = 12 reln^+/– and n = 10 reln^–/–. ^****p < 0.0001. Mean ± SEM. (C,D) Photographs showing the experimental setup for social preference (C) and preference for social novelty (D).

Behavior of reln^+/– and reln^–/– zebrafish. (A–C) Both genotypes shoal normally. (A) Inter-individual distance (Kruskal–Wallis test with Dunn’s multiple comparisons test: chi-square = 3.71, p > 0.99 for reln^+/– and p = 0.36 for reln^–/–); (B) Polarization (Kruskal–Wallis test with Dunn’s multiple comparisons (test: chi-square = 0, p > 0.99 for both genotypes) and (C) velocity [one-way ANOVA Dunnett’s multiple comparisons test: F(2,3) = 5.498, p = 0.09 for WT, p = 0.31 for reln^+/– and p = 0.07 for reln^–/–]. n = 2 groups of 5 WT, n = 2 groups of 5 reln^+/– and n = 2 groups of 5 reln^–/–. (D–F) Both reln^+/– and reln^–/– behave similarly to WT in the open field test. (D) Time at the side of the tank [thigmotaxis, one-way ANOVA Dunnett’s multiple comparisons test: F(2,33) = 1,687, p = 0.20, p = 0.13 for WT vs. reln^+/– and p = 0.50 for WT vs. reln^–/–]; (E) time spent in the center of the tank [one-way ANOVA Dunnett’s multiple comparisons test: F(2,33) = 1,431, p = 0.25 for WT, p = 0.25 for reln^+/– and p = 0.99 for reln^–/–]. (F) Locomotion is not affected in the open field test [one-way ANOVA Dunnett’s multiple comparisons test: F(2,33) = 1,625, p = 0.21, p = 0.14 for reln^+/– and p = 0.61 for reln^–/–]. n = 12 WT, n = 12 reln^+/–, n = 12 reln^–/–. (G,H) Both genotypes exhibit normal anxiety-like behavior. (G) Time at the bottom of a novel tank (Kruskal–Wallis test with Dunn’s multiple comparisons test: chi-square = 0,5705, p = 0.93 for reln^+/– and p > 0.99 for reln^–/–); (H) locomotion in a novel tank [one-way ANOVA Dunnett’s multiple comparisons test: F(2,35) = 1,54, p = 0.23 for WT, p = 0.75 for reln^+/– and p = 0.43 for reln^–/–) n = 13 wild-type, n = 13 reln^+/–, n = 12 reln^–/–. (I) No difference in aggression levels between WT, reln^+/– and reln^–/– (Kruskal–Wallis test with Dunn’s multiple comparisons test: chi-square = 3,372, p = 0.97 for reln^+/– and p = 0.13 for reln^–/–) n = 12 wild-type, n = 12 reln^+/–, n = 8 reln^–/–. Mean ± SEM.)

Behavior of dab1a^–/– zebrafish. (A,B) Social preference test. (A) Both genotypes prefer to spend time near a group of unfamiliar fish [1st strangers; p = 0.0005 and p < 0.0001 respectively; two-way ANOVA followed by Sidak’s post hoc, genotype factor: F(1,30) = 0.9841, p = 0.3291, stranger factor: F(1,30) = 72.81, p < 0.0001, interaction genotype × stranger: F(1,30) = 7.716, p = 0.0093]. (B) WT and dab1a^–/– spend equal time near both groups when a second unfamiliar group is added [p = 0.61 for WT and p = 0.44 for dab1a^–/–; two-way ANOVA followed by Sidak’s post hoc, genotype factor: F(1,30) = 0.004, p = 0.95, stranger factor: F(1,30) = 0.1016, p = 0.75, interaction genotype × stranger: F(1,30) = 2.175, p = 0.15. n = 10 wild-type and n = 7 dab1a^–/–]. (C–E)dab1a^–/– exhibit normal behavior in the open field test. (C) Time at side of the tank [thigmotaxis, t-test (Welch): t_(0.3262) = 9.971, p = 0.75] and (D) time in center of the tank [t-test (Welch): t_(0.1684) = 14.64, p = 0.87] (E) locomotion [t-test (Welch): t_(0.5277) = 13.05, p = 0.61. n = 10 wild-type and n = 7 dab1a^–/–]. (F)dab1a^–/– exhibit normal anxiety-like behavior. Time at the bottom of novel tank (Mann–Whitney test: U = 17, p = 0.06). (G)dab1a^–/– fish are more active than WT in the NTT [t-test (Welch): t_(3.234) = 7.409, p = 0.0133. n = 10 wild-type and n = 7 dab1a^–/–]. (H)dab1a^–/– display heightened aggression levels compared to WT (Mann–Whitney test: U = 11.5, p = 0.0445. n = 10 wild-type and n = 7 dab1a^–/–). ^∗p < 0.05, ^∗∗∗p < 0.001, ^****p < 0.0001. Mean ± SEM.

Behavior of vldlr^–/– zebrafish. (A,B) Social preference test. (A) Both genotypes prefer to spend time near a group of unfamiliar fish [1st strangers; p < 0.0001 for both genotypes; two-way ANOVA followed by Sidak’s post hoc, genotype factor: F(1,24) = 0.0039, p = 0.95, stranger factor: F(1,24) = 120.1, p < 0.0001, interaction genotype × stranger: F(1,24) = 0.4912, p = 0.49]. (B) WT and vldlr^–/– spend equal time near both groups when a second unfamiliar group is added [p = 0.58 for WT and p = 0.73 for vldlr^–/–; two-way ANOVA followed by Sidak’s post hoc, genotype factor: F(1,24) = 1.353, p = 0.26, stranger factor: F(1,24) = 0.005, p = 0.94, interaction genotype × stranger: F(1,24) = 0.02779, p = 0.87. n = 8 wild-type and n = 7 vldlr^–/–]. (C–E)vldlr^–/– behavior in the open field test. (C)vldlr^–/– spent the same time at side of the tank [t-test (Welch): t_(1.581) = 9.068, p = 0.15] compared to WT but more (D) time in center of the tank [t-test (Welch): t_(2.595) = 9.47, p = 0.03]. (E) Locomotion [t-test (Welch): t_(0.3779) = 12.84, p = 0.71. n = 8 wild-type and n = 7 vldlr^–/–]. (F)vldlr^–/– exhibit normal anxiety-like behavior. Time at the bottom of novel tank (Mann Whitney test: U = 16, p = 0.32), (G) locomotion in a novel tank [t-test (Welch): t_(0.1803) = 10.92, p = 0.86. n = 7 wild-type and n = 7 vldlr^–/–]. (H) No difference in aggression levels between WT and vldlr^–/– [t-test (Welch): t_(1.87) = 11.41, p = 0.09. n = 8 WT and n = 7 vldlr^–/–]. ^∗p < 0.05, ^****p < 0.0001.

High pressure liquid chromatography. 5-HT levels are increased in the hindbrain of reln^–/– compared to WT [t-test: t_(3.694) = 70, p = 0.0021, multiple t-tests with Holm–Sidak multiple comparisons correction. n = 8 WT, n = 8 reln^–/–]. (A) Telencephalon, (B) diencephalon, (C) optic tectum and (D) hindbrain. No differences in the breakdown of (E) DOPAC/DA, (F) HVA/DA, (G) 5-HIAA/-5HT. DA, dopamine; Di, diencephalon; DOPAC, 3,4-Dihydroxyphenylacetic acid; Hb, Hindbrain; HVA, homovanillic acid; Tel, telencephalon; TeO, optic tectum; 5-HIAA, 5-hydroxyindoleacetic acid; 5-HT, 5-hydroxytryptamine. ^∗∗p < 0.01. Mean ± SEM.

Buspirone increase the social preference for novelty in WT but not reln^–/–. (A) WT treated with saline spend equal time near the two groups of unfamiliar fish, whereas WT treated with buspirone switch preference and spend more time near the new group of unfamiliar fish. (B) Buspirone treatment does not completely rescue the impaired preference for social novelty of reln^–/– zebrafish (WT with saline p = 0.51; WT with buspirone p = 0.0039; reln^–/– with saline p < 0.0001; reln^–/– with buspirone p = 0.012). Three-way ANOVA followed by Sidak’s post hoc, genotype factor: F(1,56) = 0.2488, p = 0.62, stranger factor: F(1,56) = 2.11, p = 0.16, treatment factor: F(1,56) = 0.0124, p = 0.91, interaction genotype × stranger: F(1,56) = 40.28, p < 0.0001, interaction genotype × treatment: F(1,56) = 0.01879, p = 0.89, interaction stranger × treatment: F(1,56) = 3.362, p = 0.07, interaction stranger × genotype × treatment: F(1,56) = 0.0566, p = 0.81). n = 8 WT and n = 8 reln^–/–. ^∗p < 0.05, ^∗∗p < 0.01, ^****p < 0.0001. Mean ± SEM.

Oxytocin and risperidone do not rescue preference for social novelty. (A) WT treated with saline (p = 0.57), oxytocin (p = 0.53), and risperidone (p = 0.14) spend equal time near the two groups of unfamiliar fish. (B) Neither oxytocin (p = 0.048) nor risperidone (p = 0.0023) treatment rescues the impaired preference for social novelty of reln^–/– zebrafish (saline control, p = 0.014). Three-way ANOVA followed by Sidak’s post hoc for oxytocin, genotype factor: F(1,64) = 0.3928, p = 0.53, stranger factor: F(1,64) = 2.133, p = 0.15, treatment factor: F(1,64) = 0.0143, p = 0.90, interaction genotype × stranger: F(1,64) = 17.44, p < 0.0001, interaction genotype × treatment: F(1,64) = 0.0282, p = 0.87, interaction stranger × treatment: F(1,64) = 0.0707, p = 0.79, interaction stranger × genotype × treatment: F(1,64) = 0.04759, p = 0.83. Three-way ANOVA followed by Sidak’s post hoc for risperidone, genotype factor: F(1,64) = 1.073, p = 0.30, stranger factor: F(1,64) = 2.573, p = 0.11, treatment factor: F(1,64) = 0.8058, p = 0.37, interaction genotype × stranger: F(1,64) = 25.65, p < 0.0001, interaction genotype × treatment: F(1,64) = 0.3363, p = 0.56, interaction stranger × treatment: F(1,64) = 0.01244, p = 0.91, interaction stranger × genotype × treatment: F(1,64) = 0.4869, p = 0.49. n = 9 WT and n = 9 reln^–/–. ^∗p < 0.05, ^∗∗p < 0.01. Mean ± SEM.