Maternal Rest/Nrsf Regulates Zebrafish Behavior through snap25a/b

Abstract

During embryonic development, regulation of gene expression is key to creating the many subtypes of cells that an organism needs throughout its lifetime. Recent work has shown that maternal genetics and environmental factors have lifelong consequences on diverse processes ranging from immune function to stress responses. The RE1-silencing transcription factor (Rest) is a transcriptional repressor that interacts with chromatin-modifying complexes to repress transcription of neural-specific genes during early development. Here we show that in zebrafish, maternally supplied rest regulates expression of target genes during larval development and has lifelong impacts on behavior. Larvae deprived of maternal rest are hyperactive and show atypical spatial preferences. Adult male fish deprived of maternal rest present with atypical spatial preferences in a novel environment assay. Transcriptome sequencing revealed 158 genes that are repressed by maternal rest in blastula stage embryos. Furthermore, we found that maternal rest is required for target gene repression until at least 6 dpf. Importantly, disruption of the RE1 sites in either snap25a or snap25b resulted in behaviors that recapitulate the hyperactivity phenotype caused by absence of maternal rest Both maternal rest mutants and snap25a RE1 site mutants have altered primary motor neuron architecture that may account for the enhanced locomotor activity. These results demonstrate that maternal rest represses snap25a/b to modulate larval behavior and that early Rest activity has lifelong behavioral impacts.

Significance statement: Maternal factors deposited in the oocyte have well-established roles during embryonic development. We show that, in zebrafish, maternal rest (RE1-silencing transcription factor) regulates expression of target genes during larval development and has lifelong impacts on behavior. The Rest transcriptional repressor interacts with chromatin-modifying complexes to limit transcription of neural genes. We identify several synaptic genes that are repressed by maternal Rest and demonstrate that snap25a/b are key targets of maternal rest that modulate larval locomotor activity. These results reveal that zygotic rest is unable to compensate for deficits in maternally supplied rest and uncovers novel temporal requirements for Rest activity, which has implications for the broad roles of Rest-mediated repression during neural development and in disease states.

Keywords: Rest/Nrsf; locomotor behavior; maternal effect; snap25; zebrafish.

Figure 1.

Transcriptome comparison of Mrest^sbu29/+ and MZrest^sbu29/sbu29. A, Venn diagram showing the overlap of upregulated and downregulated genes in Mrest^sbu29/+ and MZrest^sbu29/sbu29 blastula. The number of genes with a predicted RE1 site near them is indicated. B, GO analysis showing the significant biological processes that are enriched in the upregulated genes.

Figure 2.

RE1-containing genes are upregulated in Mrest^sbu29/+ embryos. qPCR analysis showing fold differences relative to the Mrest^sbu29/+ transcript levels (defined as 1). Significance: *p < 0.05 (Student's t test). All markers shown are upregulated at blastula stage in Mrest^sbu29/+ embryos. snap25a (A), snap25b (B), and gpr27 (C) are upregulated at the 8 somite stage (11.5 hpf). npas4a (D) and amph (E) are upregulated at 6 dpf. ND, Not detectable.

Figure 3.

Rest target genes are inappropriately expressed in Mrest^sbu29/+ embryos. RNA Whole-mount in situ hybridization at 24 h and 6 dpf for Rest target genes for Mrest^sbu29/+ and Zrest^sbu29/+. Ectopic expression (white bracket or arrow) is observed with probes for snap25a (A,B), snap25b (E,F), and syt4 (I,J) but not nfsa (M,N) or amph (Q,R) in the hindbrain of Mrest^sbu29/+ embryos at 24 hpf. Increase expression in Mrest^sbu29/+ observed at 6 dpf in with nsfa (O,P) and amph (S,T) but not snap25a (C,D), snap25b, or syt4 probes. OV, Otic vesicle; MB, midbrain.

Figure 4.

Larvae lacking maternal rest are hyperactive at 6 dpf. A–F, Mrest^sbu29/+ (N = 71) exceed Zrest^sbu29/+ (n = 72) in total movements (A,B), total distance (C,D), and total duration (E,F) over 15 min. G–L, Similarly, MZrest^sbu29/sbu29 (n = 48) exceed related WT controls (n = 72) in total movements (G,H), total distance (I,J), and total duration (K,L) over 15 min. All graphs represent mean; error bars indicate SE. Significance: Student's t test for the entire testing periods and two-way repeated-measures ANOVAs, with genotype serving as the independent factor and time serving as the repeated measure for the 1 min analysis. #Genotype (p < 0.05).

Figure 5.

Larvae lacking maternal rest show an atypical spatial preference at 6 dpf. A, C, Representative locomotion diagrams of movement in 1 min. Mrest^sbu29/+ (A) and MZrest^sbu29/sbu29 (C) larvae display a preference for the outer well. Green represents small velocity movements. Red represents large velocity movements during a spontaneous locomotion assay in the light. B, Quantification of percentage of time spent in the outer well over 15 min shows that Mrest^sbu29/+ (n = 71) larvae spend significantly more time in the outer well compared with Zrest^sbu29/+ (n = 72) (p = 0.002). D, Quantification of percentage of time spent in the outer wall over 15 min reveals that MZrest^sbu29/sbu29 (n = 48) larvae spend more time in the outer well compared with related WT controls (n = 72) (p = 0. 0307). Significance was defined using a Student's t test. #p < 0.05.

Figure 6.

Mrest^sbu29/+ males, but not females, showed increased wall preference in the novel environment assay. A, Locomotion diagrams for individual fish over 5 min showing the Mrest^sbu29/+ male wall preference. B, During the assay, Mrest^sbu29/+ (N = 21) males spent more time near the wall compared with Zrest^sbu29/+ (N = 20) controls. C, D, Analysis of percentage of time spent near the walls for females (Mrest^sbu29/+, N = 18; Zrest^sbu29/+, N = 20) (C) and males (D) in 1 min intervals reveals that Mrest^sbu29/+ males, but not females, tend to swim near the side of the tank over the entire assay. Significance was defined using a multivariate ANOVA to identify main effects of sex and/or genotype and significant interactions between the two over the testing period. A two-way repeated-measures ANOVA was also used to compare within-sex data collected in 1 min bins across the 15 min testing period. #Genotype (p < 0.05).

Figure 7.

CRISPR-CAS9 targeting of RE1 sites. A, B, A sequence alignment of WT and snap25 RE1 site mutations (A) snap25a^sbu82 or (B) snap25b^sbu83. Black represents the genomic sequence surrounding the RE1 site. Red represents RE1 site. C, E, qPCR analysis showing fold differences relative to the RE1 mutant transcript levels (defined as 1). Significance: p < 0.05 (Student's t test). D, F, RNA whole-mount in situ hybridization with (D) snap25a probe on a Snap25a RE1 site^sbu82/+ inx or (F) snap25b probe on a snap25b RE1 site^sbu83/+ inx. White bracket or white arrow indicates ectopic expression. OV, Otic vesicle; MB, midbrain. #Genotype (p < 0.05).

Figure 8.

snap25a and snap25b RE1 site mutants are hyperactive at 6 dpf. A–F, The snap25a RE1^sbu82/sbu82 mutants (N = 24) exceeded sibling WT (N = 30) controls and snap25a RE1^sbu82/+ heterozygotes (N = 74) in (A,B) number of movements, (C,D) distance, and (E,F) duration at 6 dpf. G–L, The snap25b RE1^sbu83/sbu83 mutant (N = 44) exceeded sibling WT (N = 37) and the snap25b RE1^sbu83/+ heterozygotes (N = 62) in (G,H) number of movements, (I,J) distance, and (K,L) duration at 6 dpf. Significance was defined using a one-way ANOVA over the entire test period; when the data were compared on per minute bases, the data were compared using two-way ANOVAs with repeated-measures designs, with genotype serving as the independent factor and time serving as the repeated measure. #Genotype (p < 0.05).

Figure 9.

Rest regulates primary motor neuron development. znp-1 immunohistochemistry on whole-mount zebrafish embryos at 56 hpf to label primary motor neurons. Confocal images were acquired from the truck using the yolk extension as a landmark (10 μm stacks). A, C, Changes in the primary motor neuron architecture are apparent in Mrest^sbu29/+ and snap25a RE1 site mutant embryos compared with controls. B, D, Significant increase in fluorescence was observed in the Mrest^sbu29/+ and snap25a RE1 site mutant. Significance was defined using the Student's t test, and control was set to 1. No changes are apparent in primary motor neuron architecture; fluorescence was observed in snap25b RE1 site mutant embryos (E,F). #Genotype (p < 0.05).