A high-throughput functional genomics workflow based on CRISPR/Cas9-mediated targeted mutagenesis in zebrafish

Abstract

The zebrafish is a popular model organism for studying development and disease, and genetically modified zebrafish provide an essential tool for functional genomic studies. Numerous publications have demonstrated the efficacy of gene targeting in zebrafish using CRISPR/Cas9, and they have included descriptions of a variety of tools and methods for guide RNA synthesis and mutant identification. However, most of the published techniques are not readily scalable to increase throughput. We recently described a CRISPR/Cas9-based high-throughput mutagenesis and phenotyping pipeline in zebrafish. Here, we present a complete workflow for this pipeline, including target selection; cloning-free single-guide RNA (sgRNA) synthesis; microinjection; validation of the target-specific activity of the sgRNAs; founder screening to identify germline-transmitting mutations by fluorescence PCR; determination of the exact lesion by Sanger or next-generation sequencing (including software for analysis); and genotyping in the F₁ or subsequent generations. Using these methods, sgRNAs can be evaluated in 3 d, zebrafish germline-transmitting mutations can be identified within 3 months and stable lines can be established within 6 months. Realistically, two researchers can target tens to hundreds of genes per year using this protocol.

Figure 1

Overview of CRISPR/Cas9 mutagenesis pipeline.

Figure 2

Schematic of sgRNA synthesis and 3-primer PCR strategies for fluorescent PCR or next-gen sequencing. a) sgRNAs are synthesized by annealing two oligos. The top strand is target-specific and the bottom strand is a generic oligo (Ultramer), both oligos are annealed and extended with a DNA polymerase and used as template for in vitro transcription. In this schematic, the T7 or SP6 promoter sequence is depicted in purple, 18-20 bp target sequence in orange (do not include NGG PAM site in this sequence), and remaining sgRNA sequence in green. b) Genotyping strategy by fluorescent PCR using an M13F-FAM primer or an M13F-barcoded primer and next-gen sequencing to determine lesions. Gene-specific primers are designed flanking the target site (200-325bp region). The M13F sequences are added to the forward primer, and the pigtail sequences are added to the reverse primer. Depending on the application (fluorescent PCR vs Sequencing), the third primer is added to the PCR reactions to generate fluorescent or barcoded amplicons.

Figure 3

Agarose gel electrophoresis of Cas9 mRNA, target assembly and sgRNA. a) Restriction digest of pT3SnCas9n plasmid and in vitro transcription of Cas9 mRNA on a 1% (wt/vol) DNA agarose gel; lane M: Marker (1kb plus), lane 1: undigested plasmid, lane 2: digested plasmid, lane 3: Cas9 mRNA. A complete DNA digestion should give a single band at 7.3kb while a successful Cas9 mRNA synthesis should produce a band at ∼1.6kb. b) Oligo assembly of sgRNA oligos. Target-specific oligos are annealed and extended with the generic gRNA oligo, and the assembled products are run on a 2.5% (wt/vol) agarose gel. Correctly assembled oligos should run at 117-120bp (lane 1, 2). Bottom strand oligo (oligo2) or top strand oligo will run at much smaller sizes (lane 3, 4) . c) The assembled oligos were used as a template for in vitro transcription reaction. The sgRNAs are run on a 2.5% (wt/vol) DNA agarose gel. A successful sgRNA synthesis will produce 2 bands between 100bp and 200bp. The higher size band is due to secondary structure of the sgRNA (lane 2) and will disappear upon heating at 65°C (lane1), incorrect oligo assembly will not yield any RNA (lane 3 and 4).

Figure 4

Schematic using the diameter of a sphere to calculate the dose for injection into the embryos. When 10 pulses of the injector are equivalent to 0.3mm then the volume of 1 pulse is equal to 1.4nl.

Figure 5

Evaluation of sgRNA activity by pigmentation phenotype and CRISPR-STAT plots. Comparison of uninjected (a, e) tyr sgRNA injected into the yolk (b, d) or cell (c, f) of embryos. Examples of sgRNA activity determined using CRISPR-STAT showing a highly active target (g), a target with low activity (h) and a target with no activity (i). CRISPR-STAT plots for an uninjected embryo for each sgRNA are shown in the left panels and for an injected embryo are shown in the right panels. The X-axis represents the size (in nucleotides) of the peaks and the red star denotes the position of the WT allele. The Y-axis shows the peak height (intensity) of the fluorescent PCR product. Data shown is for peaks within 20 nucleotides of the WT allele. All animal experiments were performed under an approved animal study protocol by the National Human Genome Research Institute Animal Care and Use Committee at the National Institutes of Health.

Figure 6

Fluorescent PCR plots of F1 embryos from an injected founder fish crossed with a WT fish. Wildtype samples show a single peak (top panel) whereas a heterozygous embryo shows 2 peaks (middle and bottom panels). The difference in the size of WT and mutant peaks indicates the size of the lesion. The X-axis represents the size of the peaks and the red star denotes the position of the WT allele. The Y-axis shows the peak height (intensity) of the fluorescent PCR product. All animal experiments were performed under an approved animal study protocol by the National Human Genome Research Institute Animal Care and Use Committee at the National Institutes of Health.

Figure 7

Fin amputation of an adult zebrafish caudal fin for genomic DNA extraction. Fin amputation before (a) and after (b) the cut to show roughly the amount typically removed for genomic DNA extraction. All animal experiments were performed under an approved animal study protocol by the National Human Genome Research Institute Animal Care and Use Committee at the National Institutes of Health.

Figure 8

Chromatograms of a WT and heterozygous sample. After removal of the WT nucleotides from the heterozygous chromatogram the remaining sequence reveals the mutant allele sequence. In this example the mutation is a deletion of 2bp (GT).