CRISPR-Cas9 Genome Editing of Rat Embryos using Adeno-Associated Virus (AAV) and 2-Cell Embryo Electroporation

Summary

This protocol presents an optimized approach for producing genetically modified rat models. Adeno-associated virus (AAV) is used to deliver a DNA repair template, and electroporation is used to deliver CRISPR-Cas9 reagents to complete the genome editing process in 2-cell embryo.

Abstract

Genome editing technology is widely used to produce genetically modified animals, including rats. Cytoplasmic or pronuclear injection of DNA repair templates and CRISPR-Cas reagents is the most common delivery method into embryos. However, this type of micromanipulation necessitates access to specialized equipment, is laborious, and requires a certain level of technical skill. Moreover, microinjection techniques often result in lower embryo survival due to the mechanical stress on the embryo. In this protocol, we developed an optimized method to deliver large DNA repair templates to work in conjunction with CRISPR-Cas9 genome editing without the need for microinjection. This protocol combines AAV-mediated DNA delivery of single-stranded DNA donor templates along with the delivery of CRISPR-Cas9 ribonucleoprotein (RNP) by electroporation to modify 2-cell embryos. Using this novel strategy, we have successfully produced targeted knock-in rat models carrying insertion of DNA sequences from 1.2 to 3.0 kb in size with efficiencies between 42% and 90%.

Introduction

The development of CRISPR-based genome editing tools has accelerated our ability to efficiently generate new and more sophisticated genetically engineered rat models. Single-guide RNA, along with Cas9 nuclease, is combined to form Ribonucleoprotein (RNP) complexes that target DNA sequences of interest within the genome and result in double-stranded DNA breaks. Because cellular DNA repair mechanisms are error-prone, insertions and deletions (INDELs) are introduced during the repair process that can disrupt a target gene's function. When there is co-delivery of a desired engineered DNA sequence (repair template) along with genome editing reagents, insertion of the repair template in the region containing the double-stranded DNA break occurs through a process called homology-directed repair (HDR). This is an effective strategy for generating animal models with targeted DNA insertions/substitutions (knock-ins). One limitation is that knock-in sequences are often large in size, which has been shown to reduce gene editing efficiency, thus making generating the desired model more difficult¹. Strategies to increase knock-in efficiencies have included linearization of both double-stranded DNA (dsDNA) and single-stranded DNA (ssDNA) repair templates and chemical modification of DNA repair templates^2,3,4. In addition, pronuclear microinjection along with HDR stimulating compounds, application of electrical pulses in conjunction with microinjection, and timed microinjection into 2-cell embryos have all been attempted^5,6,7. Despite the success of some of these approaches, the incorporation of DNA sequences larger than 1.0 kb remains technically challenging.

Electroporation, which is a common method for introducing reagents into cultured cell lines, offers an alternative to microinjection for delivering CRISPR-Cas9 components into embryos. Embryo electroporation, first demonstrated in rat embryos⁸, has since been successfully used as a delivery method in mice^{9,10,11,12,13}, pigs^14,15, and other animal model organisms^16,17,18. Embryos, suspended in the medium containing CRISPR-Cas9 reagents, are placed into a cuvette or onto a glass slide between two electrodes and subjected to direct pulses of electrical currents. This creates transient openings in the zona pellucida and embryo plasma membrane through which the CRISPR-Cas9 components enter the embryos. Typically, mid-level electrical "poring" pulses are used to create the temporary openings followed by lower level electrical "transfer pulses" that facilitate movement of the negatively charged genome editing components. Embryo electroporation is efficient, has a high throughput, and is easy to perform. However, while embryo electroporation has been shown to be highly successful for the introduction of small (^<200 bp) ssDNA repair templates, there are few reports of successful electroporation of larger (^>1.0 kb) repair templates^13,19. This size restriction represents a major limitation of embryo electroporation for generating knock-in animal models requiring large insertions.

In the context of gene therapy, adeno-associated viruses (AAVs) have long been used as vehicles to deliver genetic material due to their efficient in vivo infectivity of both dividing and non-dividing cells, lack of pathogenicity, and rare genomic integration^20,21. Recently, more studies have combined AAVs with CRISPR-Cas9 technology to introduce DNA repair templates and CRISPR reagents^22,23,24. This approach allows delivery of larger DNA repair templates without the need for microinjection techniques.

The HDR pathway is more active in the late S and G2 phases of the cell cycle^25,26. In studies performed in vitro, significant increases in knock-in efficiency were achieved by delivery of CRISPR-Cas9 RNPs and DNA repair templates into G2-synchronized cells or by restriction of the presence of Cas9 protein to late S and G2 phases using a Cas9-Geminin fusion protein². Moreover, there is a major zygotic genome activation (ZGA) event which occurs during the extended G2 phase of the 2-cell stage embryo, and this is associated with an open chromatin state. It is speculated that this provides the CRISPR-Cas9 RNPs and repair templates with greater accessibility to the genomic DNA.

Our goal was to build on all these observations, by combining the AAV approach with embryo electroporation to introduce CRISPR-Cas9 RNPs at the 2-cell stage of embryo development. This strategy takes advantage of the larger DNA repair template delivery capacity of AAV, the technical ease of electroporation and the more optimal 2-cell time point for genomic accessibility during embryo development to create an efficient method for targeted genetic engineering of DNA insertions. As highlighted in this protocol, our optimized method allows for the production of targeted knock-in rat models carrying insertions of DNA sequences from 1.2 to 3.0 kb in size without the need for microinjection techniques.

Protocol

All experimental procedures were approved by the University of Missouri's Institutional Animal Care and Use Committee (ACUC protocol #25580) and were performed according to the guidelines set forth in the Guide for the Use and Care of Laboratory Animals.

1. AAV-mediated DNA repair template delivery

1. Design the DNA construct to contain the desired knock-in sequence in between two homology arms. Typical homology arm lengths are 300-500 bp in length. Ensure that the entire DNA repair template (homology arms + desired knock-in sequence) is flanked by inverted terminal repeats (ITRs) by cloning into an appropriate plasmid backbone and packaging into recombinant ssAAV1 or ssAAV6 (serotypes 1 or 6).
   NOTE: If the sgRNA target sequence is present in the DNA repair template, it is important to engineer silent mutations to disrupt this sequence thus preventing CRISPR-Cas9 mediated editing of the correctly targeted allele.
   CAUTION: Recombinant AAV viral vectors are non-integrating and classified as BSL-1. Standard aseptic techniques are used for pipetting and handling embryos that have been exposed to virus.
2. Add engineered ssAAV1 or ssAAV6 (packaged with DNA repair template) to 30 µL of KSOM-R media^27,28 to a final concentration of 3 × 10⁷ genome copies (GC)/µL under mineral oil in 35 mm Petri dish.
3. Place zygotes with visible pronuclei into the KSOM-R media containing engineered ssAAV.
   NOTE: Using this protocol there is no need to thin the zona pellucida as AAV serotypes 1 and 6 are able to readily pass through for efficient DNA repair template delivery.
4. Incubate at 37 °C with 5% CO₂ and maximal humidity for 18-20 h to allow for sufficient viral transduction.

2. Electroporation preparation

1. On the following day, make an electroporation mixture containing 100 ng/µL of sgRNA and 100 ng/µL of Cas9 protein diluted in Opti-MEM media in a sterile RNase-free tube. Gently mix.
2. Incubate at room temperature for 10 min to allow for RNP formation.
3. During the incubation time, turn on the electroporator and connect leads to a 1.5 mm gap glass slide electrode.
4. Set electroporation parameters as described in Table 1.

3. 2-cell embryo electroporation

1. Gently pipette 5 µL of the CRISPR electroporation mixture between the two electrodes on the glass slide taking care not to introduce air bubbles to the solution.
2. Remove the 2-cell stage embryo from the AAV solution. Line up 2-cell stage embryos in the mixture without allowing them to touch the sides of the electrodes. This is done to avoid lysis.
3. Press the Ohm button to measure the impedance. Check to ensure impedance is between 0.100 and 0.300 kΩ. The reaction volume will alter the impedance (add volume to decrease impedance and subtract volume to increase impedance).
4. Press Start. The process takes few seconds to complete.
5. Remove embryos immediately following the electroporation and wash 3 times in fresh 30 µL drops of KSOM-R media.
6. Place embryos back in 37 °C incubator with 5% CO₂ and maximal humidity for in vitro culture in 500 µL drops of KSOM-R media under mineral oil. Alternatively, surgically transfer embryos to a 1.5-day post coitum (dpc) pseudopregnant female rat to produce live offspring.

Figure 1: Schematic of AAV-mediated DNA delivery and 2-cell embryo electroporation pipeline for CRISPR-Cas9 genome editing. Please click here to view a larger version of this figure.

Results

Following the protocol, there is effective AAV-mediated delivery of the DNA repair template allowing for highly efficient HDR after 2-cell embryos are electroporated with CRISPR-Cas9 RNPs. As shown in the video, a successful electroporation process results in bubbles forming on each electrode (Figure 2C) and the impedance remaining within the range of 0.100 and 0.300 kΩ. After electroporation it is not uncommon for embryos to swell filling the perivitelline space inside the zona pellucida (Figure 3). This swelling should subside after a few hours in culture. It is also not uncommon to see a small percentage (up to 20%) of cellular fusion events after electroporating 2-cell embryos (Figure 3). Cellular fusion can potentially be avoided by careful horizontal axis alignment rather than vertical axis alignment between the electrodes (Figure 2B). However, we typically just discard any fused embryos after the electroporation process.

Our AAV-mediated DNA delivery with 2-cell embryo CRISPR-Cas9 RNP electroporation technique has proven to be highly effective for generating knock-in rat models with targeted insertions. Testing our method in rat embryos and screening cultured blastocysts for an insertion of an engineered short artificial intron containing loxP sites, we noted over 60% of blastocysts contained the correctly targeted knock-in allele based on Next Generation Sequence (NGS) analysis (Figure 4). Further, testing our method to produce live knock-in rats, we designed a project to engineer a Cre recombinase coding sequence targeted to the ATG start site of the rat Drd2 gene. The AAV-packaged DNA donor template consisted of ITR sequences required for AAV packaging, a 5' homology arm (422bp in length), a Cre-pA sequence (1,339bp in length), and a 3' homology arm (477bp in length) (Figure 5A). Of the 11 offspring born, 10 were positive by PCR analysis for the correctly inserted Cre-pA sequence into the ATG start site of rat Drd2 (Figure 5B-D). The targeted insertion was later confirmed by DNA sequence analysis. As a summary over 7 different projects, we have achieved a 76% average knock-in success rate in founder animals as determined by PCR analysis across homology arm junctions and DNA sequence verification (Table 2).

	Voltage [V]	Pulse Length [msec]	Pulse Interval [msec]	# of Pulses	Decay Rate [%]	Polarity
Poring Pulse	40	3.5	50	4	10	+
Transfer Pulse	5	50	50	5	40	+/-

Table 1: Electroporation parameters. Ensure that the Current Limit is ON during the electroporation.

Figure 2: 2-cell embryo electroporation process. (A) Glass slide electrode. (B) Embryos suspended in CRISPR-Cas9 genome editing reagents and loaded into the glass slide electrode prior to electroporation. (C) Air bubble formation on each electrode during electroporation. Please click here to view a larger version of this figure.

Figure 3: Embryo appearance after electroporation. Cellular fusion is noted in up to 20% of 2-cell embryos after electroporation. Embryonic cells can also swell inside of the zona pellucida filling the perivitelline space. Please click here to view a larger version of this figure.

Figure 4: AAV-mediated DNA delivery and 2-cell embryo electroporation knock-in efficiency in culture rat blastocysts. Data presented as percentage of blastocysts positive for correct knock-in allele detect by Next Generation Sequencing (NGS). N=28 blastocysts (culture-only control group), N=37 blastocysts (electroporation (EP) only control group), and N=35 blastocysts (AAV+EP group). Please click here to view a larger version of this figure.

Figure 5: Creation of a Drd2-Cre knock-in rat model. (A) Targeting of the rat Drd2 gene. The RNP complex was designed to target Exon 2 which contains the coding region start site. The AAV donor template contained viral inverted terminal repeat sequences (ITR), 5' and 3' Drd2 gene homology arms (HA) and the Cre gene with a polyA sequence. Successful integration of the donor template knocks in the Cre gene 3' to the start site within Exon 2. This puts the Cre gene under control of the Drd2 promoter while simultaneously knocking out the Drd2 gene. The location of PCR primers used for confirmation of successful knock-in are indicated by purple arrows. (B) PCR analysis of potentially genome edited animals (Lanes B10-C8) with primers spanning across the 5' homology arm. Lane C9: wild type (negative control); Lane C10: no template (negative control). Expected amplicon size for knock-in positive animals is 601bp. (C) PCR analysis of potentially genome edited animals (Lanes D8-E6) with primers spanning across the 3' homology arm. Lane E7: wild type; Lane E8: no template control. Expected amplicon size for knock-in positive animals is 676bp. (D) Cre gene detection. Expected amplicon size is 381bp. Lanes C7-D5: potential founders; Lane D6: wild type; Lane D7: no template control. PCR reactions were analyzed by capillary electrophoresis with 15bp-3kb alignment markers (indicated in green). The QX DNA size marker is shown to the left of each image with peak size corresponding to PCR amplicon size in base pairs (bp). This figure is modified with permission from reference²⁴. Please click here to view a larger version of this figure.

Project description	Offspring positive for knock-in (%)
Oprm1-P2A-Flp (1.7 kb insertion)	5/5 (100%)
Drd2-Cre (1.3 kb insertion)	10/11 (91%)
Triml2-P2A-Cre (1.3 kb insertion)	9/16 (56%)
Cxcl15-iCre (1.3 kb insertion)	6/6 (100%)
Opn4-P2A-Cre (1.2 kb insertion)	6/8 (75%)
Vipr1-FLEX-EGFP (1.4 kb insertion)	5/12 (42%)
Rosa26-FLEX-MTS-KillerRed (3.1 kb insertion)	5/7 (71%)

Table 2: Knock-in model success rates

Discussion

The CRISPR-Cas genome editing system has revolutionized the field of genetic engineering by allowing the efficient production of both straightforward and complex, customized genetic modifications in a variety of animal species. Frequent refinements and improvements in techniques associated with genome editing further its versatility. Here we have described a new approach using ssAAV-mediated DNA delivery along with 2-cell embryo electroporation of CRISPR-Cas9 reagents into 2-cell rodent embryos. We have demonstrated that this is an effective way to produce targeted knock-in animal models.

A major advantage of using AAV is that it enables delivery of larger DNA repair templates into embryos without the need for microinjection techniques. AAV-mediated delivery also helps somewhat circumvent the DNA template size restrictions of electroporation alone. However, there are some important considerations when using AAV. The zona pellucida, the hardened glycoprotein matrix surrounding the fertilized egg serves as a physical barrier that often complicates delivery of nucleic acids. Thinning of the zona pellucida is a common technique that is used for better access to the embryo. However, some studies have reported that embryo development is hindered by infection with higher doses (1 × 10⁷ GC/µL to 1 × 10⁹ GC/µL) AAV in embryos in which the zona pellucida was thinned prior to AAV infection²². There are, however, some serotypes of AAV that can diffuse across the zona pellucida²⁹. Our protocol does not involve thinning of the zona pellucida and uses AAV with serotypes 1 and 6 at a concentration of 3 × 10⁷ GC/µl to achieve knock-in rates higher than previously published reports. We did not note decreased embryo development under these conditions.

The high knock-in efficiency achieved by electroporation of 2-cell embryos is most likely due to the extended G2 phase in embryos at this stage of development, making this an optimal development stage in which to perform electroporation. However, one potential complication is the observed cellular fusion seen in some 2-cell embryos during the electroporation process. This observation is not surprising given that electrofusion procedures are commonly used for fusing mammalian cells and fusion can potentially be avoided by careful horizontal rather than vertical axis alignment between the electrodes^30,31. Furthermore, when genome editing occurs at a later stage in development such as 2-cell rather than 1-cell, there is greater potential for increased mosaicism. Using our 2-cell electroporation approach, we observed a slight increase in mosaicism compared to pronuclear injection (86% for AAV approach and 67% for PNI approach) at a 1-cell stage but were still able to achieve germline transmission of the modified allele in all founder animals tested. Another potential concern is the possibility of tandem concatemer integration of AAV-derived DNA templates. While we did not perform long-read sequencing for the models described in this protocol, we did confirm single integration of the DNA templates for all models using droplet-digital PCR (ddPCR) copy number variation.

Finally, while our method involved AAV delivery of DNA sequences of interest, it should be possible to use other viral and non-viral mediated DNA delivery approaches in conjunction with embryo electroporation to successfully genetically manipulate embryos. In conclusion, combining ssAAV-mediated DNA delivery and electroporation of CRISPR-Cas9 RNPs into 2-cell stage embryos is an efficient and easily adaptable method for efficient generating knock-in rat models.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Leads with alligator clips for electrodes	NepaGene	C117
Mineral oil	MilliporeSigma	M5310
NepaGene21 Super Electroporator	NepaGene	NEPA21
Platinum plate electrodes on slide glass	NepaGene	CUY501P1-1.5
PURedit Cas9 Protein	MilliporeSigma	PECAS9
sgRNA (chemically-modified)	Synthego	N/A
ssAAV1 or ssAAV6 packaged DNA repair template	Vectorbuilder	N/A