AP-4 loss in CRISPR-edited zebrafish affects early embryo development

Abstract

Mutations in the heterotetrametric adaptor protein 4 (AP-4; ε/β4/μ4/σ4 subunits) membrane trafficking coat complex lead to complex neurological disorders characterized by spastic paraplegia, microcephaly, and intellectual disabilities. Understanding molecular mechanisms underlying these disorders continues to emerge with recent identification of an essential autophagy protein, ATG9A, as an AP-4 cargo. Significant progress has been made uncovering AP-4 function in cell culture and patient-derived cell lines, and ATG9A trafficking by AP-4 is considered a potential target for gene therapy approaches. In contrast, understanding how AP-4 trafficking affects development and function at the organismal level has long been hindered by loss of conserved AP-4 genes in key model systems (S. cerevisiae, C. elegans, D. melanogaster). However, zebrafish (Danio rerio) have retained AP-4 and can serve as an important model system for studying both the nervous system and overall development. We undertook gene editing in zebrafish using a CRISPR-ExoCas9 knockout system to determine how loss of single AP-4, or its accessory protein tepsin, genes affect embryo development 24 h post-fertilization (hpf). Single gene-edited embryos display abnormal head morphology and neural necrosis. We further conducted the first exploration of how AP-4 single gene knockouts in zebrafish embryos affect expression levels and patterns of two autophagy genes, atg9a and map1lc3b. This work suggests zebrafish may be further adapted and developed as a tool to uncover AP-4 function in membrane trafficking and autophagy in the context of a model organism.

Keywords: CRISPR; Coat proteins; Gene editing; Membrane trafficking; Zebrafish.

Fig. 1.. AP-4 or tepsin gene loss causes irregular head size and neural necrosis in zebrafish.

Representative intermediate knockout phenotypes of each AP-4 subunit or tepsin gene at 24 h post-fertilization (hpf). Rows depict gene target, and columns depict injection condition, Knockout embryos display irregularly small heads with neural necrosis compared to the uninjected (wild-type) control. This phenotype is consistent across all guide RNAs for all five KO genes (N values; Table 1). Uninjected 24 hpf embryos are from the pair-matched clutch for each of the knockout embryos.

Fig. 2.. Sequencing validates successful ap4b1 and enthd2 single gene knockout zebrafish embryos.

Embryos displaying intermediate phenotypes (Fig. 1; Fig. S2) were analyzed by Amplicon EZ deep sequencing (Azenta). Percentage of reads containing insertions (green) or deletions (blue) in the sgRNA target site (gray box) are displayed for representative ap4b1 (A) and enthd2 (C) KO embryos. Sequencing data were obtained from a single representative 24 hpf embryo in the intermediate phenotype category. (B, D) Three different representative reads (R1, R2, R3) from sequencing of the ap4b1 (B) and enthd2 (D) KO embryos are shown together with the WT sequence (underlined target site highlighted in gray with nucleotide numbers marked). Both point mutations (yellow) and deletions (dashes) are observed.

Fig. 3.. Autophagy gene expression levels in ap4b1 and enthd2 single gene knockout models.

Autophagy gene expression at 24 hpf was assayed by RT-qPCR of pooled wild-type, ap4b1 or enthd2 KO embryos. (A) Expression levels of atg9a were generally elevated in enthd2 knockout embryos though not statistically significant. (B) Map1lc3b expression levels were not significantly different from WT. Each data point represent a pool of approximately 40 embryos from 3 independent injections. Normalized expression values (determined in Bio-Rad Maestro) were calculated using a control gene (elfa). Data are presented as the mean ± SEM with p-value from one-way ANOVA with Dunnett’s multiple comparison post-hoc analysis.

Fig. 4.. Insight into AP-4 loss using knockout zebrafish models.

(A) AP-4 and enthd2 single gene knockout embryos display abnormal head development, enlarged yolk sacs, and shortened tails containing fewer somites. Autophagy gene expression patterns (represented in blue by atg9a) exhibit darker staining patterns in the zebrafish brain, notochord, and tail. (B) At the cellular level, AP-4 mediates ATG9A export from the TGN. ATG9A is required for generation and maintenance of autophagosomes, suggests links between AP-4 trafficking and autophagy. CRISPR-edited AP-4 single gene knockout zebrafish models suggest AP-4 loss may be linked to trafficking or autophagy defect early in development. (Created using Biorender).