Targeted genome editing in the model system Danio rerio (zebrafish) has been greatly facilitated by the emergence of CRISPR-based approaches. Herein, we describe a streamlined, robust protocol for generation and identification of CRISPR-derived nonsense alleles that incorporates the heteroduplex mobility assay and identification of mutations using next-generation sequencing.
Characterization of the clustered, regularly interspaced, short, palindromic repeat (CRISPR) system of Streptococcus pyogenes has enabled the development of a customizable platform to rapidly generate gene modifications in a wide variety of organisms, including zebrafish. CRISPR-based genome editing uses a single guide RNA (sgRNA) to target a CRISPR-associated (Cas) endonuclease to a genomic DNA (gDNA) target of interest, where the Cas endonuclease generates a double-strand break (DSB). Repair of DSBs by error-prone mechanisms lead to insertions and/or deletions (indels). This can cause frameshift mutations that often introduce a premature stop codon within the coding sequence, thus creating a protein-null allele. CRISPR-based genome engineering requires only a few molecular components and is easily introduced into zebrafish embryos by microinjection. This protocol describes the methods used to generate CRISPR reagents for zebrafish microinjection and to identify fish exhibiting germline transmission of CRISPR-modified genes. These methods include in vitro transcription of sgRNAs, microinjection of CRISPR reagents, identification of indels induced at the target site using a PCR-based method called a heteroduplex mobility assay (HMA), and characterization of the indels using both a low throughput and a powerful next-generation sequencing (NGS)-based approach that can analyze multiple PCR products collected from heterozygous fish. This protocol is streamlined to minimize both the number of fish required and the types of equipment needed to perform the analyses. Furthermore, this protocol is designed to be amenable for use by laboratory personal of all levels of experience including undergraduates, enabling this powerful tool to be economically employed by any research group interested in performing CRISPR-based genomic modification in zebrafish.
The conservation of molecular machinery across eukaryotes underlies the power of using model organisms for research. Many of these model systems facilitate the use of reverse-genetic approaches such as targeted gene knockouts to characterize the contribution of a gene product to a biological or disease process of interest. Gene disruption techniques in organisms such as zebrafish have historically relied on targeted introduction of frameshift mutations that result from imprecise repair of DSBs1,2. When a DSB is introduced into the genome, the DNA lesion is repaired through one of two pathways that are universally present in nearly all cell types and organisms: non-homologous end joining (NHEJ) and homology-directed repair (HDR)3,4. The imprecise nature of the NHEJ machinery frequently produces indels of various lengths5,6,7,8,9. Introduction of frameshift mutations in the coding sequence of a gene can produce a premature stop codon, which often renders the gene nonfunctional.
Early genome engineering strategies in zebrafish to promote indels included meganucleases, zinc-finger nucleases, and transcription activator-like effector nucleases, all of which utilized DNA-protein interactions to target a nuclease to a specific genomic target where it introduced a DSB10,11,12,13,14,15. However, these technologies are often difficult to apply due to the laborious and complex engineering needed to generate a nuclease that targets the DNA sequence of interest. Unlike previous strategies, CRISPR-based gene editing does not rely on protein-DNA interactions for targeting. Instead, the CRISPR-associated (Cas) endonuclease is directed via an RNA guide that uses nucleotide base pairing interactions to target a genomic site of interest16,17,18,19,20,21. Due to the simplicity of designing a RNA guide with the desired base pairing interactions for targeting it is relatively easy to target the Cas endonuclease to the desired locus. The type II CRISPR system in particular has been widely developed for genome editing applications due to several advantageous features including use of a single multidomain Cas nuclease (Cas9) that requires interaction with DNA to stimulate endonuclease activity and use of a single guide RNA (sgRNA) to target it to the cognate DNA sequence18. The sequence requirements necessary for targeting of the cognate sgRNA are well understood19, and the desired sgRNA is easily generated by in vitro transcription. The simplicity and robustness of the CRISPR/Cas9 approach greatly facilitates targeted genetic modification in zebrafish and a wide variety of other organisms.
The enhanced ability to undertake targeted genome editing in zebrafish as a result of developing CRISPR-based reagents has significantly increased the opportunity to study processes emblematic of vertebrate organisms such as development of the central nervous system. The zebrafish genome contains orthologs of 70% of the protein-coding genes found in the human genome as well as 84% of genes associated with diseases in humans22. Zebrafish development exhibits several key qualities that enhance its use in reverse genetic studies: the embryos are laid in large clutches, develop externally from the mother making them amenable to genetic manipulation by microinjection, and adult zebrafish sexually mature by 3 months of age, allowing for rapid propagation of desired lines23.
Numerous protocols are available that describe a variety of approaches to generate and identify CRISPR-derived indels in zebrafish24,25,26,27,28,29,30,31. However, many of these procedures are time intensive, require access to expensive equipment, and can be challenging for labs with limited expertise. The steps described herein provide a simple, robust, and economical CRISPR/Cas9-strategy to engineer zebrafish knockout lines. This protocol describes the use of a highly efficient kit to synthesize sgRNAs using DNA oligonucleotides (oligos), similar to other approaches that have been previously described32. The described protocol includes two steps in particular that greatly simplify analysis of CRISPR-mutated lines: step-by-step use of the PCR-based HMA33 to easily determine the presence of genome modifications, and sequencing analysis of heterozygous zebrafish to rapidly and easily determine the nature of multiple indels in an economical fashion. In addition, step-by-step instructions are included for robust selection, reliable production, and injection of guide RNAs. The steps provided here exemplify a robust, relatively inexpensive protocol that enables laboratory personnel with a range of expertise to contribute to the identification of gene knockouts in zebrafish.
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Purdue Animal Care and Use Committee (PACUC number 08-031-11).
1. Design of Template-specific Oligos for Guide RNA Production
2. Preparation of CRISPR-reagents for Embryo Microinjection
3. Microinjection of CRISPR-components into Zebrafish Embryos
4. Analysis of Efficiency of Indel Formation Using an HMA
5. Identification and Propagation of Knock-out Lines
The experimental approaches described in this protocol allow for efficient, cost-effective production of zebrafish knock-out lines using CRISPR/Cas9 technology. The following figures have been included in this article to facilitate interpretation and troubleshooting of the results obtained using this protocol. Following successful production and microinjection of CRISPR-reagents, the zebrafish embryos can be analyzed for overt phenotypes and for indel formation using HMA. A helpful control to visualize the success of the CRISPR-experiment is the use of the sgRNA described in step 1.5 to target the pigment-producing gene tyrosinase. Cas9-induced indel formation at tyrosinase results in loss of pigmentation and is easily scored by 48 hpf (Figure 1). Another helpful control to ensure that preparation of the CRISPR-reagents for injection has been successful, is to verify that full-length (120 nt) sgRNA has been synthesized using a denaturing polyacrylamide gel (Figure 2, Lane 1 and 2). If the RNA has been degraded it may appear as a smear, for example Lane 3 (Figure 2) shows degraded RNA that is not suitable for injection.
To analyze the indel formation frequency of genes targeted by CRISPR-Cas9 that do not result in overt phenotypes such as tyrosinase, HMA analysis is a simple and reliable method. sgRNA/Cas9 injected embryos analyzed using HMA results in the formation of heteroduplex bands, and reduction of the intensity of the homoduplex band (Figure 4). The presence of heteroduplex bands is further utilized in this protocol to identify potential founder fish from the microinjected embryos and as adults (Figure 4 and Figure 5), to analyze the germline transmission efficiency of a founder (Figure 6), and to verify presence of an indel in a heterozygous F1 fish (Figure 7). The heterozygous fish that contain an indel are candidates for NGS to identify the nature of the indel and to determine if a premature stop codon is present in the coding region of the target gene.
Figure 1: Zebrafish embryos exhibit a pigment defect when injected with a sgRNA targeting tyrosinase at the one-cell stage. (A) Wild-type, uninjected embryo at 48 hpf and (B) injected embryo at 48 hpf. Please click here to view a larger version of this figure.
Figure 2: In vitro transcription of sgRNA using synthesis kit. Oligos were synthesized using in vitro transcription according to the sgRNA synthesis kit instructions. 500 ng of RNA was run on a urea/PAGE gel as described. sgRNA loaded in lanes 1 and 2 shows a band corresponding to the full length, intact 120 nt RNA. The sgRNA in lane 3 shows a degraded RNA sample that is not suitable for injection.
Figure 3: Comparison of the health of 24 hpf injected embryos. A living embryo (A) developed to 24 hpf, is easily distinguished from an embryo that has aborted development (B). Embryos that resemble (B) or have drastically altered features to (A), such as spinal curvature or altered head and eye development should be removed from dish. Please click here to view a larger version of this figure.
Figure 4: Heteroduplex mobility assay of sgRNA-Cas9 microinjected zebrafish embryos. Pools of 5 embryos per sample were collected at 72 hpf and gDNA was extracted. Heteroduplex analysis was performed as described, samples were loaded equally with 500 ng of DNA. Lanes: M = 100 bp marker; 1 = uninjected control; 2 = injection sample 1; 3 = injection sample 2. Expected band size = 98 bp.
Figure 5: Heteroduplex mobility assay of gDNA extracted from the tail of an adult CRISPR-injected zebrafish. Embryos that were injected with an sgRNA and Cas9 protein were grown to adulthood (3 months). Fish B and C exhibit heteroduplex bands and were subsequently bred to identify germline transmitted indels; fish A was not used in subsequent analysis because it does not exhibit a positive heteroduplex band. Lanes: 1 = wild-type control; 2 = Fish A; 3 = Fish B; 4 = Fish C. Expected band size = 98 bp. Please click here to view a larger version of this figure.
Figure 6: Heteroduplex mobility assay of single embryos generated by breeding a F0 CRISPR-injected zebrafish to a wild-type fish to identify germline transmitted indels. Zebrafish were mated, and the F1 embryos grown for 72 h. Single embryos were collected and heteroduplex analysis performed as described. Lanes: 1 = wild-type control; 2-10 = a single F1 embryo per lane. This gel shows that 7 out of 10 embryos show a positive heteroduplex band, indicating a germline transmission rate of 70% of the indel. Expected band size = 98 bp. Please click here to view a larger version of this figure.
Figure 7: Heteroduplex mobility assay of adult F1 zebrafish tail clips. Adult F1 fish were scored by HMA to identify indels. Fish that exhibited a positive heteroduplex band were PCR amplified and submitted for wide sequencing analysis to determine the nature of the mutation. (A) Lanes: 1 = wild-type control; 2 = fish A (4 bp deletion, 1 bp mismatch). (B) These F1 fish were identified from the same founder, and yet show different heteroduplex patterning, indicating germline transmission of multiple modified alleles from a single founder. Lanes: 1 = wild-type control; 2 = fish B (4 bp insertion, 7 bp mismatch), 3 = fish C (4 bp deletion, 4 bp mismatch). Each of these indels created a premature stop codon in the coding sequence of the target gene, as determined by NGS. Please click here to view a larger version of this figure.
This protocol describes the production of gene knockouts in the zebrafish vertebrate model system using CRISPR-Cas9 technology. A number of protocols have previously been described to undertake CRISPR-mediated genome engineering in zebrafish15,25,26,50,51,52. This protocol builds on previous efforts by combining a number of simple yet reproducibly consistent experimental techniques, in particular HMA and NGS of multiple heterozygous fish, to create a straightforward, economical, and experimentally robust protocol for CRISPR-mediated mutagenesis in zebrafish that is appropriate for labs staffed with personnel with a range of training and experience, as well as teaching labs.
Recommendations for design and synthesis of guide RNAs are included in this protocol. A major consideration in guide RNA design is the minimization of off-target effects. Several prediction algorithms have been developed to allow CRISPR-users to access computation tools with user-friendly graphical interfaces that predict both the activity of the on-target guide and the chance of off-target effects34,35,36. A specific advantage of the zebrafish system is lowered rates of off-target effects because the Cas9 is injected into the embryos and therefore expression is transient, which has been shown in mice to result in decreased off-target effects53. Nevertheless, off-target effects have been demonstrated to occur in zebrafish54. One way to control for off-target effects is to phenotype founder zebrafish that have been generated by two independent guide RNAs that target the same gene, as these guides would be very likely to affect different off-target sites. An alternative method to minimize off-target effects that is not described in this protocol is the use of a mutated Cas9 that generates single strand breaks at the target DNA, which are repaired with high efficiency. Pairing DNA nicks within proximity of one another that are complementary to the opposite strands results in effective indel formation at the desired locus and minimizes off-target effects55,56.
In addition to having different rates of off-target effects, different sgRNAs can have different rates of mutagenesis of the desired target57,58,59. This protocol uses HMA to analyze the efficiency of mutagenesis of a given sgRNA using heteroduplex band formation33,40. Heteroduplex bands are created by hybridization of PCR-generated DNA strands that contain mismatches, and can be easily resolved using gel electrophoresis. Unlike other methods commonly used to measure indel formation, such as the T7 endonuclease assay or high resolution melt analysis25,26, HMA does not require an expensive enzyme to cut mismatched DNA, and does not require complicated analysis of PCR melt curves. Importantly, using HMA to verify high rates of indel formation in the injected population also enables the investigator to minimize the number of fish needed for subsequent production of knock-out lines, which reduces the cost of identifying a mutation with the desired characteristics.
The relative ease of generating CRISPR-based indels enables creation of multiple alleles of multiple genes at once. Web-based software is available for analysis of single mutations from heterozygous fish using Sanger sequencing of PCR products49. In the case where three or more CRISPR-mutated alleles are analyzed, NGS to characterize the nature of the indel is likely to be more cost effective to characterize the nature of the indel as this approach allows a pool of up to 50 different alleles to be characterized at once (see Table of Materials)60,61,62. Such economy of scale would likely be particularly useful in an undergraduate laboratory setting.
In summary, this protocol provides step-by-step directions for reproducibly generating high quality CRISPR-reagents (in particular, sgRNA) such that fewer adult fish need to be created and analyzed to successfully identify the mutant alleles of interest, which also reduces the time and cost of generating the desired lines. Importantly, this protocol has been designed such that it can be applied by laboratories with limited resources to produce mutant zebrafish in an affordable manner. Furthermore, we have found that this approach is suitable for undergraduates and thus expands the opportunities for education and training of undergraduate students interested in hands-on experience in CRISPR-based genome editing.
The authors have nothing to disclose.
This work was supported by the National Institutes of Health (R21CA182197 to J.O.), the Ralph W. and Grace M. Showalter Research Trust, and by the Purdue Agricultural Science and Extension for Economic Development (AgSEED) program. Sanger sequencing data were acquired by the Purdue Genomics Core facility supported by P30 CA023168. Mary Witucki was supported by the Purdue University Center for Cancer Research Summer Undergraduate Research Program supported by the Carroll County Cancer Association. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. We thank Benjamin Carter, Ellen Denning, and Taylor Sabato for providing critical feedback on the manuscript. We thank the Department of Biochemistry for support of this work, and the Center for Zebrafish Research at Purdue University for their dedication in the care and welfare of our zebrafish colony. Finally, we thank the Purdue Genomics Core Facility, and contributions of Phillip San Miguel regarding the NGS services.
CRISPOR | sgRNA analysis software. http://crispor.tefor.net/ | ||
Breaking-Cas | sgRNA analysis software. https://omictools.com/breaking-cas-tool | ||
ChopChop | sgRNA analysis software. https://chopchop.rc.fas.harvard.edu/ | ||
Primer 3 | PCR primer design. http://bioinfo.ut.ee/primer3-0.4.0/ | ||
Rnase free technique | http://genomics.no/oslo/uploads/PDFs/workingwithrna.pdf | ||
RNase-free water | Any brand | Synthesis of guide RNAs. Thermo Fisher and Qiagen both carry suitable water. | |
Rnase-free ethanol | Any brand | Molecular biology grade ethanol is Rnase free. Fischer, VWR sell suitable ethanol. | |
Cas9 Protein | PNA Bio | CP01 | Cas9 protein with nuclear localization signal. |
Target specific oligos | Integrated DNA Technologies (IDT) | Synthesize guide RNAs. | |
PCR primers flanking guide RNA cut site | Integrated DNA Technologies (IDT) | Use for heteroduplex analysis, 100-300 bp product. | |
PCR primers flanking guide RNA cut site | Integrated DNA Technologies (IDT) | Use for sequencing CRISPR-alleles,300-600 bp product recommended. | |
EnGen sgRNA Synthesis Kit | New England BioLabs | E3322 | In vitro transccribe guide RNAs. |
10X TBE | Thermo Fisher | B52 | RNase-free buffer for electrophoresis of nucleic acids. |
6X Load Dye | New England BioLabs | B7025S | Loading DNA for electropheresis. |
5M Ammonium acetate | Thermo Fisher | AM9071 | Clean up reagent for purification of guide RNAs. |
Ethanol (200 proof, nuclease-free) | Any brand | Clean up reagent for purification of guide RNAs. Molecular grade materials are RNase free. | |
Nuclease-free Water | Any brand | For synthesis and purification of guide RNAs. | |
RNase away | Thermo Fisher | 10328011 | Eliminates RNase contamination from surfaces, equipment, glassware. |
DNA Ladder, 100 bp | New England BioLabs | N0467S | Molecular weight standards for gel electrophoresis of DNA. |
DNA Ladder, 1 kb | New England BioLabs | N0552S | Molecular weight standards for gel electrophoresis of DNA. |
RNA Ladder | New England BioLabs | N0364S | Single strand RNA marker to verify that full length sgRNA has been synthesized |
TEMED | Thermo Fisher | 17919 | Casting polyacrylamide gel. |
Ammonium persulfate (APS) | Sigma-Aldrich | A3678 | Casting polyacrylamide gel. |
40% Polyacrylamide (19:1) | BioRad | 161-0154 | Casting denaturing polyacrylamide gel. |
Urea | Sigma-Aldrich | U5378-100G | Casting denaturing polyacrylamide gel. |
RNA Gel Loading Dye (2x) | Thermo Fisher | R0641 | Loading guide RNA onto denaturing gel for electrophoresis. |
Ethidium bromide | Sigma-Aldrich | E1510-10ML | Visualization of nucleic acids. |
SYBER Green | Thermo Fisher | S7563 | Alternate method to ethidium bromide for detection of dsDNA in agarose or polyacrylamide gels. |
Microcentrifuge | Eppendorf | 5424 | RNA clean up, PCR purification. |
MyTaq Polymerase | Bioline | BIO-21105 | For amplification of DNA using PCR. |
Thermocycler | Any brand | PCR amplification. | |
Spectrophotometer | Any brand | NanoDrop or related product to quantify the amout of DNA/RNA. | |
E3 embryo media | Made in house | Media for raising embryos and making injeciton plate. Recipe: http://cshprotocols.cshlp.org/content/2011/10/pdb.rec66449 | |
Methylene Blue | Sigma-Aldrich | M9140 | Added to 1x E3 media to prevent fungal growth on embryos. |
Petri dish | Thermo Fisher | FB0875711Z | Store embryos, cast injection plate. |
Agarose | Denville | CA3510-8 | Casting injection plate, agarose gels. |
Microinection mold | Adaptive Science Tools | TU-1 | To create wells to hold embryos during injection. |
Phenol Red | Sigma-Aldrich | P0290 | Dye for visualization of injection. |
Mineral Oil | Sigma-Aldrich | M5904-5ML | To calibrate needle injection volume. |
Transfer pipettes | Any brand | Moving embryos. | |
Razor blade | Thermo Fisher | 11295-10 | Cutting injection needle, tail clipping adult fish. |
Incubator | Any brand | Maintaining embryos at 28.5 oC. | |
Verticle pipette puller | David Kopf Instruments | 700C | Geneate needles for injection. Additional needle pulling instructions: http://cshprotocols.cshlp.org/content/2006/7/pdb.prot4651.long |
Capillary tubes | Sutter Instruments | BF100-58-10 | Geneate needles for injection. |
Microloader tips | Eppendorf | 930001007 | Load solution into injection needles. |
Microinjector | World Precision Instruments | PV 820 | Injecting embryos. |
Disecting microscope | Leica | Injecting embryos. | |
30% Polyacylamide (29:1) | BioRad | 161-0156 | Heteroduplex gel casting. |
MS-222 (Tricaine) | Sigma-Aldrich | A-5040 | Anesthetize zebrafish for tail clipping and gDNA extraction from embryos. Recipe: https://wiki.zfin.org/display/prot/TRICAINE |
Microwave | Any brand | Casting injection plate, agarose gels. | |
Scale | Any brand | Casting injection plate, agarose gels. | |
Gloves | Any brand | For all aspects of the protocol. | |
N2 | Any brand | To expell liquid from the capillary for embryo injection. | |
1,000 µL tips | Any brand | For all aspects of the protocol. | |
200 µL tips | Any brand | For all aspects of the protocol. | |
10 µL tips | Any brand | For all aspects of the protocol. | |
PCR strip tubes | Any brand | For all aspects of the protocol. | |
Micropipettes | Any brand | For all aspects of the protocol. | |
NGS Sequencing platform: Wideseq | If your institution does not offer this servicce visit: https://www.purdue.edu/hla/sites/genomics/wideseq-2/ (External cost: $35). | ||
Software for sequence analysis | For Sanger sequencing of heterozygous fish. Visit: http://yosttools.genetics.utah.edu/PolyPeakParser/ | ||
Software for sequence analysis | SnapGene Viewer, Sequence Scanner for analysis of NGS sequencing. |