A pipeline for precise and efficient genome editing by sgRNA-Cas9 RNPs in Drosophila

Abstract

Genome editing via homology-directed repair (HDR) has made possible precise and deliberate modifications to gene sequences. CRISPR/Cas9-mediated HDR is the simplest means to carry this out. However, technical challenges remain to improve efficiency and broaden applicability to any genetic background of Drosophila melanogaster as well as to other Drosophila species. To address these issues, we developed a two-stage marker-assisted strategy in which embryos are injected with RNPs and pre-screened using T7EI. Using sgRNA in complex with recombinant Cas9 protein, we assayed each sgRNA for genome-cutting efficiency. We then conducted HDR using sgRNAs that efficiently cut target genes and the application of a transformation marker that generates RNAi against eyes absent. This allows for screening based on eye morphology rather than colour. These new tools can be used to make a single change or a series of allelic substitutions in a region of interest, or to create additional genetic tools such as balancer chromosomes.

Keywords: CRISPR; Drosophila; gene editing; homology dependent repair.

Figure 1.

Sources of Drosophila CRISPR reagents and their individual pros and cons. (a) Sources of Cas9. Source shaded grey highlights the novel source used in this study. (b) Sources of sgRNA. Since some sgRNAs are inactive for inducing DSBs in vivo, a quality control (QC) test is preferable. Region shaded grey highlights the novel prescreening method to identify active sgRNAs. (c) Sources of donor plasmids to act as repair templates for HDR. This panel only shows plasmids related to the novel plasmid used in this study (shaded grey)

Figure 2.

Screening sgRNAs for cleavage activity in vivo. (a) Schematic of the screening assay. Individual embryos are injected with RNPs composed of a particular sgRNA. Genomic DNA from each embryo is PCR-amplified, and amplicons are denatured and re-annealed. Heteroduplexes with mismatches due to indels in embryonic DNA are cleaved by T7EI enzyme. Gel electrophoresis identifies embryos with detectable cleavage events. (b) PCR products of a target site in the forked gene 892 bp in length were digested by T7EI as indicated. Shown are two representative embryos out of the nine assayed that were injected with forked RNPs. Also shown are two out of the six embryos that were uninjected. The predicted T7EI digest products are 393 and 436 bp (arrows). Although a minority of heteroduplexes derived from an embryo are T7EI-sensitive, they can be detected by this assay. A total of 6 out of 9 injected embryos showed evidence of T7EI sensitivity, whereas 0 out of 6 control embryos showed evidence. This difference is statistically significant (p = 0.0168, Fishers exact test). (c) A T7EI assay was performed using a different sgRNA that targets non-coding DNA. T7EI digestion products of sizes 295 and 502 bp (arrows) would be detected if indels are significantly generated. Since a background band runs at 500 bp, only detection of a band at 295 bp would be diagnostic for the presence of indels. None of the RNP-treated or uninjected control samples exhibit a detectable 295 bp digestion product. In sum, no 295 bp product was detected for 12/12 embryos treated with RNPs and 6/6 control embryos (p > 0.05, Fishers exact test)

Figure 3.

Identifying precise HDR edits by eye colour. (a) Shown is the transgenic marker for counterselection of imprecise HDR events. The GMR element contains 5 tandem binding sites for the transcription factor Glass fused to the Hsp70 minimal promoter. The transcript contains a shRNA stem-loop followed by an intron from the ftz gene to facilitate transcript stability. After the shRNA is processed by Drosha and Dicer, the guide RNA strand is loaded into RISC. The guide RNA is perfectly complementary to all mRNA isoforms of eya. Shown only is isoform C, and the location of the RNAi target is indicated. (b-g) Eye phenotypes of adults that had been injected with RNPs and the forked HDR donor plasmid. (b,d,f) Eyes visualized with brightfield illumination. (c,e,g) Same eyes visualized for DsRed fluorescence. All eyes are oriented anterior left and dorsal top. Listed above panels are the number of G0 lines that produced G1s with the same phenotypes as the ones shown. (b,c) Adult without apparent HDR event. (d,e) Adult with DsRed expression and no eya RNAi phenotype. (f,g) Adult with DsRed expression and an eya RNAi phenotype

Figure 4.

Workflow for two-step genome editing. (1a) Target sites flanking the area to be edited are identified (red, blue, green, purple) using online tools searching for optimal targets and with minimal off-target cleavage. (1b) Sequences from the selected target sites are transcribed in vitro to generate sgRNAs. (1c) Cas9 protein is incubated with sgRNAs before injection into embryos. (1d) Active sgRNAs that cleave embryo DNA are identified by T7 endonuclease I reactions. (1e) One of the active sgRNAs is chosen for genome editing in Step 2. (2a) Homology arms flanking the region of interest are cloned into the pBS-GMR-eya(shRNA) donor plasmid. In this example, the CRISPR target site (red triangle) is 5ʹ to the bases to be edited (black bar). (2b) Embryos are injected with the repair template plasmid and RNPs composed of sgRNA and Cas9 protein. (2c) Adult flies that develop from injected embryos are crossed back to the parental line. G1 progeny are screened for the DsRed marker. Positive G1 animals may have small eyes due to eya(shRNA) but these are not selected (green circle). Only positive G1 animals with normal eyes are selected (red circle). (2d) These are crossed to make purebred lines and molecularly analysed to determine if they contain the desired editing events. (2e) PiggyBac transposase is expressed in the germline, either by a single cross to a transgenic line, or in this example, by embryo injection of a plasmid expressing the transposase. (2f) Since the DsRed marker is dominant, adult flies developing from injected embryos that do not have red fluorescent eyes are then crossed and analysed with molecular tests to determine whether they have precisely excised the marker gene. Only the intended genomic edit remains