Stable gene silencing in zebrafish with spatiotemporally targetable RNA interference

Abstract

The ability to regulate gene activity in a spatiotemporally controllable manner is vital for biological discovery that will impact disease diagnosis and treatment. While conditional gene silencing is possible in other genetic model organisms, this technology is largely unavailable in zebrafish, an important vertebrate model organism for functional gene discovery. Here, using short hairpin RNAs (shRNAs) designed in the microRNA-30 backbone, which have been shown to mimic natural microRNA primary transcripts and be more effective than simple shRNAs, we report stable RNA interference-mediated gene silencing in zebrafish employing the yeast Gal4-UAS system. Using this approach, we reveal at single-cell resolution the role of atypical protein kinase Cλ (aPKCλ) in regulating neural progenitor/stem cell division. We also show effective silencing of the one-eyed-pinhead and no-tail/brachyury genes. Furthermore, we demonstrate stable integration and germ-line transmission of the UAS-miR-shRNAs for aPKCλ, the expressivity of which is controllable by the strength and expression of Gal4. This technology shall significantly advance the utility of zebrafish for understanding fundamental vertebrate biology and for the identification and evaluation of important therapeutic targets.

Figure 1

Functional validation of miR-shRNAs by transient in vivo transgenesis. (A) Diagram of the conditional miR-shRNA expression system. The guide stand (bottom) is highlighted in light blue. (B) Mosaic expression of miR-shRNA5^apkcλ (left), miR-shRNA4^oep (top right), and miR-shRNA4^ntla (bottom right) causes morphological defects similar to those observed in respective mutants, which can be rescued by delivery of shRNA-resistant wild-type mRNAs. The miR-30e vector-injected embryos serve as controls. (C) Quantitative RT–PCR analysis shows a gene-specific reduction of endogenous mRNA levels by shRNAs cognate to their target genes. The miR-30e vector-injected embryos serve as controls. (D) Fluorescent immunostaining of aPKCλ (green) shows knockdown effects at the protein level by different miR-shRNAs^apkcλ. In a single miR-shRNA^apkcλ-expressing cell, as indicated by the tdTomato fluorescence (red), miR-shRNA1^apkcλ (middle panel) and miR-shRNA5^apkcλ (left panel) lead to significant knockdown of aPKCλ, whereas the ineffective miR-shRNA7^apkcλ does not show obvious knockdown effect (right panel). (E) Mosaic expression of miR-shRNA5^apkcλ causes defects in mitotic division orientation of radial glia progenitors in the developing zebrafish forebrain. Left: examples of mitotic division orientation in the miR-30e vector-injected (control) and miR-shRNA5^apkcλ-expressing radial glia progenitors. Right: Quantification of division orientation in miR-30e vector control and miR-shRNA5^apkcλ-expressing radial glia progenitors (n = 27 for control, and n = 24 for miR-shRNA5^apkcλ). The nuclei of radial glia progenitors were labeled with 3NLS:EGFP (green) and the individual radial glia progenitors expressing UAS-miR30e vector (control) or UAS-miR-shRNA5^apkcλ were highlighted by tdTomato (red).

Figure 2

Functional validation of miR-shRNAs by stable in vivo transgenesis. (A) Schematic shows the stable transgenesis of miR-shRNAs^apkcλ in zebrafish using the Tol2 transposon system. (B) F₁ embryos derived from a cross between the UAS-TdTom-miR-shRNA1^apkcλ founder A and Ubi-GFF transgenic line. TdTom-negative F₁ embryos are normal (top panels). In contrast, ∼30% of the TdTom-positive F₁ embryos display a heart defect with cardiac edema (middle panels). This heart defect is rescued by the delivery of shRNA-resistant apkcλ mRNA (bottom panels). (C) Quantification of heart defect and rescue in B. (D) Embryos (60 hpf) derived from a cross between the UAS-TdTom-miR-shRNA1^apkcλ founder A and hsp70-Gal4 transgenic animal. Heat shock was performed at 8 hpf at 37 °C for 1 hr. The TdTom-positive embryo (right) displayed a heart defect with cardiac edema, whereas the heat-shocked control sibling (left) did not. (E) Droplet digital PCR analyses of endogenous apkcλ mRNA levels in stable UAS-TdTom-miR-shRNAs^apkcλ transgenic lines at different conditions of Gal4 induction. The representative images of 48-hpf embryos for each condition were shown on top of the bar graph. *P < 0.05, **P < 0.01, ***P < 0.001, vs. control sibling.