Here we provide a detailed protocol to carry out in vivo cardiac gene editing in mice using recombinant Adeno-Associated Virus(rAAV)-mediated delivery of CRISPR. This protocol offers a promising therapeutic strategy to treat dystrophic cardiomyopathy in Duchenne muscular dystrophy and can be used to generate cardiac-specific knockout in postnatal mice.
The clustered, regularly interspaced, short, palindromic repeat (CRISPR) system has greatly facilitated genome engineering in both cultured cells and living organisms from a wide variety of species. The CRISPR technology has also been explored as novel therapeutics for a number of human diseases. Proof-of-concept data are highly encouraging as exemplified by recent studies that demonstrate the feasibility and efficacy of gene editing-based therapeutic approach for Duchenne muscular dystrophy (DMD) using a murine model. In particular, intravenous and intraperitoneal injection of the recombinant adeno-associated virus (rAAV) serotype rh.74 (rAAVrh.74) has enabled efficient cardiac delivery of the Staphylococcus aureus CRISPR-associated protein 9 (SaCas9) and two guide RNAs (gRNA) to delete a genomic region with a mutant codon in exon 23 of mouse Dmd gene. This same approach can also be used to knock out the gene-of-interest and study their cardiac function in postnatal mice when the gRNA is designed to target the coding region of the gene. In this protocol, we show in detail how to engineer rAAVrh.74-CRISPR vector and how to achieve highly efficient cardiac delivery in neonatal mice.
The clustered, regularly interspaced, short palindromic repeat (CRISPR) technology represents the most recent advancement in genome engineering and has revolutionized the current practice of genetics in cells and organisms. CRISPR-based genome editing utilizes a single guide RNA (gRNA) to direct a CRISPR-associated endonuclease such as CRISPR-associated protein 9 (Cas9) and Cpf1 to a genomic DNA target that is complementary in sequence to the protospacer encoded by the gRNA with a specific protospacer adjacent motif (PAM)1,2. The PAM varies depending on the bacterial species of the Cas9 or Cpf1 gene. The most commonly used Cas9 nuclease derived from Streptococcus pyogenes (SpCas9) recognizes a PAM sequence of NGG that is found directly downstream of the target sequence in the genomic DNA, on the non-target strand. At the target site, the Cas9 and Cpf1 proteins generate a double-strand break (DSB), which is then repaired via intrinsic cellular DNA repair mechanisms including non-homologous end joining (NHEJ) and homology-directed repair (HDR) pathways3,4. In the absence of a DNA template for homologous recombination, the DSB is primarily repaired by the error-prone NHEJ pathway, which can result in insertions or deletions (indels) of nucleotides causing frameshift mutations if the DSB were created in the coding region of a gene5,6. Thus, the CRISPR system has been widely used to engineer loss-of-function mutations in various cellular models and whole organisms, in order to investigate the function of a gene. Moreover, the catalytically inactive Cas9 and Cpf1 have been developed to transcriptionally and epigenetically modulate the expression of target genes7,8.
More recently, both SpCas9 and SaCas9 (derived from Staphylococcus aureus) have been packaged into recombinant adeno-associated virus (rAAV) vectors for postnatal gene correction in the animal model of human diseases, particularly, Duchenne muscular dystrophy (DMD)9,10,11,12,13,14,15. DMD is a fatal X-linked recessive disease caused by mutations in the DMD gene, which encodes dystrophin protein16,17, affecting approximately one in 3500 male births according to newborn screening18,19. It is characterized by progressive muscle weakness and wasting. The DMD patients usually lose the ability to walk between 10 and 12 years of age and die of respiratory and/or cardiac failure by the age of 20-30 years20. Notably, cardiomyopathy develops in >90% of DMD patients and represents the leading cause of death in DMD patients21,22. Although Deflazacort (an anti-inflammation drug) and Eteplirsen (an exon skipping medicine for skipping exon 51) have recently been approved for DMD by Food and Drug Administration (FDA)23,24, none of these treatments correct the genetic defect at the genomic DNA level. The mdx mouse, which carries a point mutation at exon 23 of the dystrophin gene, has been widely used as a DMD model25. Furthermore, we previously demonstrated that the functional dystrophin expression was restored by in-frame deletion of the genomic DNA covering exons 21, 22 and 23 in skeletal muscles of mdx mice using intramuscular Ad-SpCas9/gRNA delivery9, and in the heart muscles of mdx/utrophin+/- mice using intravenous and intraperitoneal rAAVrh.74-SaCas9/gRNA delivery26.
In this protocol, we describe in detail every step in postnatal cardiac gene editing to restore dystrophin in mdx/utrophin+/- mice using rAAVrh.74-SaCas9/gRNA vectors, from gRNA design to dystrophin analysis in the heart muscle sections.
The animals used in this protocol were maintained at The Ohio State University Laboratory Animal Resources in accordance with animal use guidelines. All animal studies were authorized by the Institutional Animal Care, Use, and Review Committee of the Ohio State University.
1. Design and Cloning of gRNAs into the CRISPR Vector
2. rAAV Production
3. rAAV Purification and Concentration
4. rAAV Titering
5. Intravenous and Intraperitoneal Injection of rAAVrh.74-CRISPR into Neonatal mdx/utrophin+/- Mice
6. Analysis of Target Gene Editing Outcomes
7. Off-target analysis by T7E1 assay
The efficacy of the gRNAs to induce the deletion of the target genomic DNA region should be evaluated in cell cultures before packaging into rAAV for in vivo studies. To this end, the gRNAs and SaCas9-expressing constructs were electroporated into C2C12 cells and the genomic DNA was analyzed by PCR with the primers flanking the target sites (Figure 1). A PCR product of ~500 bp indicates successful deletion of the target genomic DNA resulting from gene editing while the control cells should not yield a band due to the large size of the genomic DNA.
Once the efficacy of the gRNAs has been confirmed in cell cultures, they can be packaged into rAAVrh.74 or other serotypes (such as AAV6, 8 or 9, all showing robust cardiac gene delivery) by using the three plasmids co-transfection into AAV293 cells. We found that it is not necessary to make two individual rAAV preparation for the two gRNAs, instead, we mixed the two gRNA constructs in an equal molar ratio and produced a single rAAV preparation. The rAAV viral particles were purified using density gradient ultracentrifugation and the purity was analyzed by SDS-PAGE. The highly purified rAAV preparation would yield only three bands corresponding to the capsid proteins VP1, VP2, and VP3, respectively, on SDS-PAGE as shown in Figure 2.
The purified rAAVrh.74 viral particles were injected into day 1-3 neonates of mdx/utrophin+/- mice via retro-orbital or intraperitoneal injection. We found both methods worked well for cardiac delivery of the rAAVrh.74-CRISPR. The injected mice can be analyzed for dystrophin expression by immunofluorescence staining (Figure 3) at 10 weeks of age.
Figure 1: Validation of the gRNAs for SaCas9-mediated gene editing in C2C12 cells. The representative agarose gel electrophoresis showed the PCR products of genomic DNA extracted from electroporated C2C12 with or without SaCas9/gRNAs. The arrow indicates the PCR product resulted from gene editing. Gapdh served as a reference. Please click here to view a larger version of this figure.
Figure 2: Analysis of rAAVrh.74 particles by SDS-PAGE. Purified rAAVrh.74 viral particles run on SDS-PAGE showed three bands, corresponding to the three capsid proteins of AAV, VP1 (87 kDa), VP2 (72 kDa) and VP3 (62 kDa), respectively. The presence of only these 3 bands indicates the high purity of the rAAV preparation without contaminating other cellular proteins. Please click here to view a larger version of this figure.
Figure 3: Immunofluorescence staining of the heart sections from mdx/utrophin+/- mice at 10 weeks following AAV-SaCas9/gRNAs treatment. The representative immunofluorescence images showed the expression of dystrophin (red) and caveolin-3 (green) in the heart sections of WT, mdx/utrophin+/- and mdx/utrophin+/- mice treated with 1 x 1012 vg AAV-SaCas9/gRNA systemically. DAPI was used to label the nuclei (blue). Scale bar: 100 µm. Please click here to view a larger version of this figure.
In this protocol, we detail all the steps necessary to achieve in vivo cardiac gene editing in postnatal mice using rAAVrh.74-mediated delivery of SaCas9 and two gRNAs. As previously noted, this approach can be used to restore dystrophin expression in dystrophic mice as described in our work27, but it could also enable rapid reverse genetics studies of cardiac-related genes in mice without the lengthy process of the traditional knockout approach to generate and breed knockout mouse models.
In order to achieve high gene editing efficiency in the target tissue, there are several critical steps to keep in mind: 1) the efficacy of the SaCas9/gRNA to induce the deletion of targeted genomic DNA needs to be validated using appropriate cell lines; 2) it is very crucial to select a proper AAV serotype that can mediate a high level of gene transduction into the target tissue and 3) determine the optimal delivery route of AAV vectors for gene transfer to the desired tissue.
In the current protocol, the designed SaCas9/gRNAs were packaged into AAVrh.74 rather than previously reported AAV8 and AAV9 to edit Dmd gene12,28. AAVrh.74 exhibits about 93% homology to AAV829 and has been shown to be non-pathogenic and effective in transducing skeletal muscle following isolated limb delivery29,30,31. In this protocol, we use retro-orbital and intraperitoneal injection of rh.74 into neonatal mice for cardiac gene transfer. We found that the neonate administration of rAAVrh.74 is highly efficient for the restoration of dystrophin expression in about 30-40% of cardiomyocytes in the mice receiving a high dosage of rAAVrh.74 (1 x 1012 vg). Interestingly, we found that the skeletal muscle dystrophin expression was not effectively rescued. This is likely due to the low skeletal muscle tropism of rh.74 serotype when administered via intravenous or intraperitoneal injection, although previous studies showed that isolated limb delivery of rAAVrh.74 can effectively transduce mouse and monkey skeletal muscle29,30,31. Also, when the rAAVrh.74 was administered into adult mice, we observed very low efficiency in restoring dystrophin expression. It would be interesting to determine the minimum dosage which is required to achieve the highest dystrophin rescue. Furthermore, other serotypes and delivery methods can be tested and compared for achieving optimal cardiac gene editing.
Successful deletion of a large genomic DNA piece requires simultaneous delivery of two gRNAs. We found it is not necessary to make two separate rAAV preparations for delivering two gRNAs. In this protocol, we simply mixed the two gRNA plasmids in an equal amount during transfection for rAAV production, and this combined production of rAAV was found to be effective in delivering both gRNAs. This combined approach is, therefore, cost- and labor-saving and recommended when simultaneous delivery of two gRNAs are required.
In many cases, it may be necessary to analyze the gene editing efficiency by genetic approaches when a good antibody is not available for the target gene product. However, it is technically challenging to accurately quantify the gene editing efficacy with two gRNAs targeting the same locus since the gene editing products would include target sequence deletion, target sequence inversion, and small indels at only one target site. Instead, it is more meaningful and feasible to quantify the efficiency of the desired targeted sequence deletion as in our case where the reduction of the transcripts containing exons 21-23 can be quantified by real-time RT-PCR.
In this protocol, we also describe the use of T7E1 assay and computational prediction to analyze the potential off-target activity. For the assay to work well, it is necessary to use a high-fidelity DNA polymerase. Also, it is always a good idea to prepare for a negative control experiment (e.g. control DNA products amplified from the cells/tissues without gene editing, and treated the same way as the test samples) and a positive control experiment (e.g. a known sample with well-characterized gene editing). However, it is of note that the detection limit of the T7E1 assay is about 5%32, and thus it may not be detectable by the T7E1 assay if the off-target activity is below 5%. GUIDE-seq33 can be used to unbiasedly analyze the off-target activity of the SaCas9/gRNA-mediated gene editing. The identified off-target sites from T7E1 assay, computational prediction and/or GUIDE-seq approaches can further be quantified for indel frequencies following in vivo gene editing by high-throughput sequencing.
Taken together, we have provided an optimized cardiac gene editing approach using rAAVrh.74. This established protocol can be further tested for therapeutic applications and reverse genetic studies.
The authors have nothing to disclose.
R.H. is supported by US National Institutes of Health grants (R01HL116546 and R01 AR064241).
Alexa Fluor 555 goat anti-rabbit IgG | Thermo Fisher Scientific | A-21428 | |
Benzonase Nuclease | Sigma-Aldrich | E1014 | |
Bio Safety Cabinets | Thermo Fisher Scientific | 1347 | 1300 Series A2 Class II, Type A2 |
BsaI | New England BioLabs Inc | RO535S | |
Competent cells | Agilent Technologies | 200314 | |
Cell culture dishes, 150mm | Nest Scientific USA | 715001 | |
Cryostat | Leica Biosystems | Leica CM3050S | |
DMEM | Thermo Fisher Scientific | MT10017CV | Corning cellgro |
Dystrophin antibody | Spring Bioscience | E2660 | |
Fast-Ion Midi Plasmdi Kits | IBI Scientific | IB47111 | |
FBS | Thermo Fisher Scientific | 26-140-079 | |
Filter Unit, 0.45μm | Thermo Fisher Scientific | 166-0045 | |
GoTaq Master Mixes | Promega | M7122 | |
High-speed plasmid mini kit | IBI Scientific | IB47102 | |
Horse serum | Abcam | ab139501 | |
Isotemp 2239 Water Bath | Thermo Fisher Scientific | 15-460-11Q | |
Isotemp Heat Block | Thermo Fisher Scientific | 11-718 | |
Inverted confocal microscope | Carl Zeiss Microscopy, LLC | LSM 780 | |
LightCycler 480 instrument | Roche | 5015243001 | LightCycler 480 Instrument II |
Large Capacity Centrifuge | Thermo Fisher Scientific | 46-910 | Thermo Scientific Sorvall RC 6 Plus |
Microcentrifuge | Thermo Fisher Scientific | 75002445 | Sorvall Legend Micro 21R |
Nanodrop spectrophotometers | Thermo Scientific | ND2000CLAPTOP | NanoDrop™ 2000/2000c |
Oligos for i20-F | Integrated DNA Technologies | CACCgGGCGTTGAAATTCTCATTAC | |
Oligos for i20-R | Integrated DNA Technologies | AAAC GTAATGAGAATTTCAACGCCc; | |
Oligos for i23-F | Integrated DNA Technologies | CACCgCACCGATGAGAGGGAAAGGTC | |
Oligos for i23-R | Integrated DNA Technologies | GACCTTTCCCTCTCATCGGTGc | |
Oligos | Integrated DNA Technologies | ||
OptiPrep Density Gradient Medium | Sigma-Aldrich | D1556-250ML | |
OptiMEM | Thermo Fisher Scientific | 31985070 | |
PEG8000 | Sigma-Aldrich | 89510-1kg-F | |
Polyethylenimine | Polysciences | 23966 | |
Proteinase K | Sigma-Aldrich | P2308 | |
Quick-Seal centrifuge tube | Beckman Coulter Inc | 342414 | |
QIAquick Gel Extraction kit | Qiagen | 28704 | |
Radiant SYBR Green Hi-ROX qPCR Kits | Alkali Scientific | QS2005 | |
RevertAid RT Reverse Transcription Kit | Thermo Fisher Scientific | K1691 | |
Rotor | Beckman Coulter Inc | 337922 | Type 70Ti |
T4 DNA ligase | New England BioLabs Inc | M0202S | |
Thermal Cycler | Bio-rad | 1861096 | |
TRIzol Reagent | Thermo Fisher Scientific | 15596026 | |
Ultracentrifuge | Beckman Coulter Inc | 8043-30-1192 | Optima le-80k |
Ultra centrifugal filter unit | MilliporeSigma | UFC510096 | |
VECTASHIELD Mounting Medium with DAPI | Vector Laboratories, Inc | H-1200 |