A consolidated AAV system for single-cut CRISPR correction of a common Duchenne muscular dystrophy mutation.

Duchenne muscular dystrophy (DMD), caused by mutations in the X-linked dystrophin gene, is a lethal neuromuscular disease. Correction of DMD mutations in animal models has been achieved by CRISPR/Cas9 genome editing using Streptococcus pyogenes Cas9 (SpCas9) delivered by adeno-associated virus (AAV). However, due to the limited viral packaging capacity of AAV, two AAV vectors are required to deliver the SpCas9 nuclease and its single guide RNA (sgRNA), impeding its therapeutic application. We devised an efficient single-cut gene-editing method using a compact Staphylococcus aureus Cas9 (SaCas9) to restore the open reading frame of exon 51, the most commonly affected out-of-frame exon in DMD. Editing of exon 51 in cardiomyocytes derived from human induced pluripotent stem cells revealed a strong preference for exon reframing via a two-nucleotide deletion. We adapted this system to express SaCas9 and sgRNA from a single AAV9 vector. Systemic delivery of this All-In-One AAV9 system restored dystrophin expression and improved muscle contractility in a mouse model of DMD with exon 50 deletion. These findings demonstrate the effectiveness of CRISPR/SaCas9 delivered by a consolidated AAV delivery system in the correction of DMD in vivo, representing a promising therapeutic approach to correct the genetic causes of DMD.

Introduction

Duchenne muscular dystrophy (DMD) is a lethal muscle disorder caused by mutations in the DMD gene residing on the X chromosome., The DMD gene encodes the dystrophin protein, which is a large cytoskeletal protein essential for tethering the intracellular actin cytoskeleton and extracellular laminin., Absence of dystrophin protein in striated muscles causes skeletal muscle degeneration and myocardial fibrosis, and ultimately progresses to fatal respiratory and cardiac failure. With no transformative treatment available, there is an urgent need to develop new therapeutic approaches for DMD.

Genome editing by clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR-associated proteins (CRISPR-Cas) represents a promising technology to correct disease-causing mutations in the genome.5, 6, 7 With this approach, Cas9 nuclease is directed by a sequence-specific single guide RNA (sgRNA) to the genome, where it can induce double-stranded breaks (DSBs). In the absence of a repair template, DNA DSBs are repaired by two distinct repair pathways, which are nonhomologous end joining (NHEJ) when there is no sequence microhomology present at the breakage point or microhomology-mediated end joining (MMEJ) when there are 2–25 base pairs (bp) of microhomology on each side of the DSB.,

Recent studies by our group and others explored the potential of CRISPR-Cas9 gene editing and the NHEJ DNA repair pathway as a means of correcting diverse DMD mutations in vivo.10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21 In mice, sustained dystrophin expression and functional improvement can be observed for at least 12–18 months after systemic delivery of CRISPR-Cas9 genome-editing components by AAV., Nevertheless, challenges remain for therapeutic adaptation of CRISPR-Cas9-mediated gene editing for correction of DMD. For example, the limited packaging capacity of AAV requires a dual system consisting of two AAV vectors to separately package Streptococcus pyogenes Cas9 (SpCas9) and sgRNA. In contrast to SpCas9, the Cas9 ortholog from Staphylococcus aureus (SaCas9) is small enough to be co-packaged with sgRNA into a single AAV vector. However, all current SaCas9-based genome-editing systems have used a pair of sgRNAs to induce two DNA DSBs flanking the mutated dystrophin exon.,,17, 18, 19 This “double-cut” strategy has been reported to introduce additional unwanted genomic modifications, including inversions and AAV integration., Moreover, if one DNA DSB is rejoined by NHEJ repair before the initiation of the second DNA DSB, the mutant exon cannot be excised, rendering this double-cut strategy ineffective.

In this study, we explored the potential of CRISPR-E782K/N968K/R1015H (CRISPR-KKH) SaCas9-mediated single-cut gene editing as a means of correcting an exon 50 DMD deletion mutation. KKH SaCas9 is a SaCas9 variant carrying three amino acid substitutions in the protospacer-adjacent motif (PAM)-interacting domain that enable strong genome-editing activities at target sites with a 5′-NNNRRT-3′ PAM. We first performed single-cut gene editing with KKH SaCas9 in cardiomyocytes derived from human DMD induced pluripotent stem cells (iPSCs) harboring a deletion of exons 48–50 (ΔEx50), the most common “hotspot” region for DMD exon deletions. High frequency of a two-nucleotide deletion was observed after KKH SaCas9-mediated single-cut gene editing, which restored the open reading frame (ORF) of the dystrophin gene. Next, we packaged KKH SaCas9 and sgRNA into a single AAV9 vector and performed in vivo genome editing of exon 51 in mice with a deletion of Dmd exon 50. Systemic delivery of the consolidated CRISPR-KKH SaCas9 AAV9 vector showed efficient restoration of dystrophin expression in skeletal muscle and heart and improved muscle contractility. These findings show that delivery of KKH-SaCas9 with a single sgRNA in a single vector system is effective in correcting DMD in vivo, representing an important advancement toward potential therapeutic translation.

Results

Strategies for CRISPR-KKH SaCas9-mediated genome editing of human DMD exon 51

The majority of DMD deletion mutations are clustered in hotspot regions, comprised of exons 2–20 and exons 45–55, that disrupt the continuity of the ORF with downstream exons. Exon deletions immediately preceding exon 51, which disrupt the reading frame of this exon, represent the most common type of human DMD mutation. Our ultimate goal was to develop a consolidated AAV expression system encoding SaCas9 and an optimal sgRNA for reframing of exon 51 so as to enable in vivo DMD correction with an All-In-One vector. To optimize this gene editing strategy, we used iPSCs generated from DMD patients harboring a deletion of exons 48 to 50 (ΔEx48–50) in the DMD gene. This deletion results in splicing of exon 47 to exon 51, which introduces a premature stop codon in exon 51 (Figure 1A).

Figure 1

Strategies for CRISPR KKH SaCas9-mediated gene editing of human DMD exon 51

(A) An out-of-frame deletion of human DMD exons 48 to 50 (ΔEx48–50) results in splicing of exons 47 to 51, generating a premature termination codon in exon 51. A single-cut editing strategy was designed to enable CRISPR-KKH SaCas9 DNA cutting to restore the open reading frame of the DMD gene. Small insertions and deletions (indels) with two nucleotide deletions (3n-2) can reframe exon 51. Large indels disrupting the 5′-AG-3′ splice acceptor sequence cause exon 51 skipping, resulting in splicing of exon 47 to exon 52. (B) Illustration of the sgRNA targeting human DMD exon 51. This sgRNA recognizes a 5′-AACAGT-3′ PAM in exon 51 and generates a DSB 4 bp upstream of the 5′-TGA-3′ premature termination sequence (indicated in red). The 5′-AG-3′ splice acceptor sequence is indicated in yellow. (C) Illustration of a plasmid-encoding KKH SaCas9 with 2A-GFP, driven by a hybrid form of cytomegalovirus and chicken beta-actin promoter (CBh). The plasmid also encodes a sgRNA driven by the U6 promoter. Cells transfected with this plasmid express GFP, allowing for selection of KKH SaCas9-expressing cells by FACS. (D) Analysis of genomic indels in KKH SaCas9-edited DMD ΔEx48–50 iPSCs. Productive editing is defined as indels with 3n-2 deletion, which are capable of reframing or skipping exon 51. Data are represented as mean ± SEM. Unpaired two-tailed Student’s t tests was performed. ∗p < 0.05 (n = 3).

We did not identify efficient sgRNAs for wild-type (WT) SaCas9 capable of reframing human DMD exon 51, so we considered various SaCas9 mutants with amino acid substitutions in the PAM-interacting domain that expand the range of DNA cutting by relaxing PAM specificity. From this analysis, we identified a sgRNA for the KKH variant of SaCas9 that recognizes a 5′-AACAGT-3′ PAM in exon 51 and generates a DNA DSB 4-bp upstream of the premature termination codon (Figure 1B). Depending on the repair outcome, two types of insertions and deletions (indels) could restore the exon 51 ORF. Exon 51 could potentially be reframed through indels that delete two nucleotides (3n-2), or exon 51 could be skipped if the indel is large enough to delete the 5′-AG-3′ splice acceptor (Figure 1A).

The gene-editing efficiency of KKH SaCas9 was tested by transfecting DMD ΔEx48–50 iPSCs with a plasmid expressing KKH SaCas9 and sgRNA, and gene-edited cells were enriched through fluorescence-activated cell sorting (FACS) (Figure 1C). We performed tracking of indels by decomposition (TIDE) analysis to assess sgRNA cutting efficiency and indel patterns. We found that this sgRNA enabled high editing activity of DMD exon 51, generating over 65% of total indels (Figure 1D). More than 55% of indels allowed productive editing (3n-2), capable of restoring the DMD exon 51 ORF (Figures 1D and S1).

Interestingly, 45% of KKH SaCas9-induced indels had a deletion of a 5′-CT-3′ dinucleotide, which allows reframing of the DMD exon 51 ORF (Figure S1). The sgRNA designed in this study enables the KKH SaCas9 nuclease to induce a DNA DSB between the 5′-CTCT-3′ tetranucleotide, generating a 2-nt 5′-CT-3′ microhomology on each side of the breakage site (Figure S1), leading to high frequency of precise deletion of the 5′-CT-3′ dinucleotide. These data demonstrate that KKH SaCas9-mediated single-cut gene editing is an efficient and practicable strategy to restore the dystrophin ORF in DMD exon 51, caused by deletion of preceding exons.

CRISPR-KKH SaCas9-mediated single-cut gene editing restores dystrophin expression in DMD ΔEx48–50 iPSC-derived cardiomyocytes

Human iPSCs generated from a ΔEx48–50 DMD patient were corrected by KKH SaCas9 and sgRNA using the single-cut gene-editing approach and then differentiated to cardiomyocytes (iPSC-CMs) (Figure 2A). The mutation in uncorrected DMD ΔEx48–50 iPSC-CMs results in a premature termination codon following the first eight amino acids encoded by exon 51 (Figure S2). Correction of DMD ΔEx48–50 iPSC-CMs was accomplished by reframing of the DMD gene, as assessed by reverse transcription PCR (RT-PCR) and sequencing using a forward primer targeting exon 47 and a reverse primer targeting exon 52. Corrected DMD ΔEx48–50 iPSC-CMs had a deletion of the 5′-CT-3′ dinucleotide, which reframed exon 51 (Figure S2). We confirmed restoration of dystrophin protein expression by immunocytochemistry and western blot analysis (Figures 2B and 2C). Even without clonal selection and expansion, cardiomyocytes differentiated from KKH SaCas9-edited DMD ΔEx48–50 iPSC mixtures showed a high level of dystrophin protein comparable to healthy control iPSC-CMs (Figures 2B and 2C).

Figure 2

Restoration of dystrophin expression in DMD ΔEx48–50 cardiomyocytes after CRISPR-KKH SaCas9-mediated single-cut gene editing

(A) DMD ΔEx48–50 iPSCs were edited by KKH SaCas9 (corrected DMD iPSCs) and then differentiated into corrected cardiomyocytes (CMs) for downstream analysis. (B) Immunocytochemistry shows dystrophin restoration in mixtures of DMD ΔEx48–50 CMs following KKH SaCas9-mediated single-cut gene editing. Red, dystrophin staining; green, troponin I staining. Scale bar, 100 μm. (C) Western blot shows dystrophin restoration in mixtures of DMD ΔEx48–50 CMs following KKH SaCas9-mediated single-cut gene editing. Dilutions of protein extract from healthy control CMs were used to standardize dystrophin protein expression. Vinculin was used as the loading control. (D) Representative traces of spontaneous calcium activity of iPSC-derived CMs cultured with calcium indicator Fluo-4AM. Traces show change in fluorescence intensity (F) in relationship to resting fluorescence intensity (Fo). (E) Quantification of calcium release phase of contraction, as measured by time to peak, in iPSC-derived CMs. Data are represented as mean ± SEM. One-way ANOVA was performed with post hoc Tukey’s multiple-comparisons test. ∗∗∗∗p < 0.0001 (n = 40). (F) Quantification of calcium reuptake phase of contraction, as measured by tau, in iPSC-derived CMs. Data are represented as mean ± SEM. One-way ANOVA was performed with post hoc Tukey’s multiple-comparisons test. ∗∗∗∗p < 0.0001 (n = 40).

Dysregulation of calcium handling is a common pathogenic phenotype seen in DMD cardiomyocytes. To assess the consequences of the DMD ΔEx48–50 mutation and the effect of gene editing by the KKH SaCas9-mediated single-cut strategy, we analyzed spontaneous calcium activity in healthy control and DMD ΔEx48–50 iPSC-CMs (Figure 2D). Calcium transient kinetics, including time to peak and decay rate, were abnormally elevated in uncorrected DMD ΔEx48–50 iPSC-CMs (Figures 2E and 2F). After KKH SaCas9 gene editing, DMD ΔEx48–50 iPSC-CMs displayed normal calcium transient kinetics similar to healthy control iPSC-CMs (Figures 2E and 2F), indicating restoration of calcium release and reuptake. Next, we performed genotoxicity analysis in KKH SaCas9-edited DMD ΔEx48–50 iPSC-CMs. We did not observe significant genomic editing at the top eight predicted off-target sites (Figure S3). Therefore, KKH SaCas9-mediated single-cut gene editing represents an efficient and safe strategy to restore the ORF of human DMD exon 51 caused by exon 50 deletion, thereby allowing functional restoration in gene-edited DMD iPSC-CMs.

Systemic delivery of All-In-One AAV-packaged CRISPR-KKH SaCas9 restores dystrophin expression in ΔEx50 mice

To further evaluate the efficacy of CRISPR-KKH SaCas9 gene editing in vivo, we packaged the KKH SaCas9 nuclease and its sgRNA in one AAV vector (Figure 3A). In this All-In-One AAV system, KKH SaCas9 expression was driven by a muscle-specific Creatine Kinase 8 (CK8) promoter, restricting its expression to skeletal muscles and heart. Because the sgRNA is rate limiting for in vivo gene editing of DMD mouse models,, we included two copies of an expression cassette encoding the same sgRNA (targeting mouse Dmd exon 51) driven by two RNA polymerase III promoters, 7SK and U6, in this All-In-One AAV system (Figure 3A).

Figure 3

Systemic delivery of All-In-One AAV-packaged KKH SaCas9 restores dystrophin expression in ΔEx50 mice

(A) Illustration of the All-In-One AAV vector used to deliver KKH SaCas9 gene-editing components. KKH SaCas9 expression is driven by a muscle-specific CK8 promoter. Two copies of the same sgRNA targeting mouse Dmd exon 51 are driven by two RNA polymerase III promoters, 7SK and U6. (B) Illustration of systemic delivery of All-In-One AAV vectors in ΔEx50 mice. Postnatal day 4 ΔEx50 mice were injected intraperitoneally with 2 × 1014 or 4 × 1014 vg/kg of All-In-One AAV vectors. Four weeks after systemic delivery, ΔEx50 mice and WT littermates were dissected for analysis. (C) Immunohistochemistry shows restoration of dystrophin in the tibialis anterior (TA), triceps, diaphragm, and heart of ΔEx50 mice 4 weeks after systemic delivery of AAV-packaged KKH SaCas9 and sgRNA. Dystrophin is shown in green. n = 6 for each muscle group. Scale bars, 100 μm.

Postnatal day 4 (P4) DMD mice with exon 50 deletion (ΔEx50) were injected intraperitoneally (i.p.) with All-In-One AAV-packaged KKH SaCas9 at two different doses, 2 × 1014 vector genomes (vg)/kg (low dose) and 4 × 1014 vg/kg (high dose) (Figure 3B). Four weeks after systemic AAV delivery, the skeletal muscles and heart of KKH SaCas9-edited ΔEx50 mice were harvested for analysis. Assessment by immunohistochemistry showed that dystrophin restoration in skeletal muscles was dose dependent (Figures 3C, S4, and S5). ΔEx50 mice receiving the low-dose All-In-One AAV treatment displayed 36% and 52% dystrophin-positive myofibers in tibialis anterior (TA) and triceps muscles, respectively (Figure S5). With low-dose All-In-One AAV treatment, the diaphragm showed higher percentages of dystrophin-positive myofibers, reaching 79% (Figure S5). When the dose of All-In-One AAV was increased to 4 × 1014 vg/kg, a substantial increase of dystrophin-positive myofibers in TA and triceps was observed (Figure S5).

Next, we performed western blot analysis to quantitatively assess dystrophin restoration in skeletal muscles and heart after systematic delivery of All-In-One AAV-packaged KKH SaCas9 and sgRNA. ΔEx50 mice receiving the low-dose All-In-One AAV restored 12% and 26% of dystrophin protein in TA and triceps, respectively (Figures 4A and 4B). When the dose of All-In-One AAV was increased to 4 × 1014 vg/kg, dystrophin protein restoration in TA and triceps was over 27%. Dystrophin protein expression in the diaphragm and heart exceeded 45% and 38%, respectively, even at the low dose of All-In-One AAV treatment (Figures 4A and 4B), indicating that dystrophin protein restoration in the diaphragm and heart is greater than in TA and triceps.

Figure 4

Western blot and genomic analysis of skeletal muscles and heart of ΔEx50 mice receiving systemic All-In-One AAV delivery of KKH SaCas9 gene-editing components

(A) Western blot analysis shows restoration of dystrophin expression in the TA, triceps, diaphragm, and heart of ΔEx50 mice 4 weeks after systemic delivery of All-In-One AAV-packaged KKH SaCas9 and sgRNA. The dose of the AAV vector is indicated. Dilutions of protein extract from WT mice were used to standardize dystrophin protein expression. Vinculin was used as the loading control (n = 3 for each dose). (B) Quantification of dystrophin expression in the TA, triceps, diaphragm, and heart. Relative dystrophin intensity was calibrated with vinculin internal control before normalizing to the WT control. (C) Genomic indel quantification by deep-sequencing analysis of the TA, triceps, diaphragm, and heart of ΔEx50 mice 4 weeks after systemic delivery of All-In-One AAV-packaged KKH SaCas9 and sgRNA. Data are represented as mean ± SEM.

To quantify in vivo gene-editing efficiency in ΔEx50 mice, we performed deep-sequencing analysis to determine the indel frequency and pattern at the genomic level. ΔEx50 mice treated with low-dose All-In-One AAV had an average of 4%–10% of total indels in skeletal muscles and heart; the total indels in the high-dose group increased to 8%–12% (Figure 4C). Notably, a −2-nt deletion, which is capable of reframing the Dmd exon 51 ORF, was the predominant indel in the All-In-One AAV-treated ΔEx50 mice. These findings indicate that KKH SaCas9-mediated single-cut gene editing coupled with the single vector delivery system can effectively correct DMD mutations in vivo.

Systemic delivery of All-In-One AAV-packaged CRISPR-KKH SaCas9 restores muscle integrity and improves muscle function in ΔEx50 mice

To evaluate whether systemic delivery of All-In-One AAV-packaged KKH SaCas9 was able to rescue pathological phenotypes seen in dystrophic mice, we performed hematoxylin and eosin (H&E) (Figures S6 and S7) and Masson’s trichrome staining (Figures S9 and S10) of skeletal muscles and heart isolated from ΔEx50 mice 4 weeks after KKH SaCas9-mediated gene editing. Skeletal muscles from ΔEx50 mice without gene editing displayed necrosis and inflammatory infiltration (Figures S6 and S7). The percentage of regenerating myofibers with central nuclei in untreated ΔEx50 mice was between 25% and 35% across different skeletal muscle groups (Figures S8A–S8C). After All-In-One AAV treatment, the percentage of centrally nucleated myofibers declined substantially (Figures S8A–S8C). Distribution of myofiber cross-sectional area also showed an improvement in the TA muscle after delivery of All-In-One AAV at both doses (Figure S8D). Masson’s trichrome staining showed substantial fibrosis and necrosis in untreated ΔEx50 mice (Figures S9 and S10), ranging between 10% and 15% across different skeletal muscle groups (Figures S11A–S11C). After All-In-One AAV treatment, the percentage of fibrotic and necrotic area dramatically declined (Figures S9–S11).

To examine the effect of gene editing on muscle function, we performed grip-strength analysis on ΔEx50 mice at 4 weeks after systemic All-In-One AAV delivery. ΔEx50 mice without gene editing showed a 56% and 45% reduction of grip strength in forelimb and hindlimb compared to the WT littermates, respectively (Figure S12). Forelimb and hindlimb grip strength of ΔEx50 mice receiving low-dose All-In-One AAV treatment showed a trend toward improvement (Figure S12). Moreover, ΔEx50 mice receiving high-dose All-In-One AAV treatment showed a dramatic improvement of forelimb and hindlimb grip strength by 86% and 67%, respectively, compared to the untreated ΔEx50 littermates (Figure S12). In addition, we also performed electrophysiological analysis on soleus and extensor digitorum longus (EDL) muscles isolated from ΔEx50 mice at 4 weeks after receiving the high-dose All-In-One AAV treatment. We observed rescue of specific force and maximal tetanic force in the soleus and EDL muscle of the corrected ΔEx50 mice (Figures 5A–5D). Without KKH SaCas9 gene editing, muscle force was reduced by 30% in slow-twitch soleus muscle and 69% in fast-twitch EDL muscle compared to the WT littermates (Figures 5A and 5B). After systemic delivery of All-In-One AAV-packaged KKH SaCas9, muscle-specific force of the soleus and EDL was increased by 51% and 78%, respectively, compared to the untreated ΔEx50 littermates (Figures 5A and 5B). The maximal tetanic force of the soleus and EDL also followed a similar pattern as seen for specific force (Figures 5C and 5D).

Figure 5

Systemic delivery of All-In-One AAV-packaged CRISPR-KKH SaCas9 improves muscle function in ΔEx50 mice

(A and B) Specific force (mN/mm2) of the soleus (A) and extensor digitorum longus (EDL) (B) in WT, ΔEx50 mice untreated, and ΔEx50 mice treated with All-In-One AAV-packaged KKH SaCas9. Data are represented as mean ± SEM. Brown-Forsythe and Welch ANOVA test was performed. ∗p < 0.05, ∗∗p < 0.005 (n = 6). (C and D) Maximal tetanic force of the soleus (C) and EDL (D) in WT, ΔEx50 mice untreated, and ΔEx50 mice treated with All-In-One AAV-packaged KKH SaCas9. (E and F) Fatigue resistance analysis of soleus (E) and EDL (F) in WT, ΔEx50 mice untreated, and ΔEx50 mice treated with All-In-One AAV-packaged KKH SaCas9. Data are represented as mean ± SEM. Nonlinear regression with extra sum-of-squares F test was performed (n = 6).

Next, we performed fatigue analysis in WT and ΔEx50 mice. Without KKH SaCas9 gene editing, ΔEx50 mice exhibited faster force reduction in soleus and EDL. The average time for 50% force reduction in soleus and EDL from ΔEx50 mice was reduced by 38% and 29%, respectively, compared to the WT littermates (Figures 5E and 5F). After KKH SaCas9-mediated gene editing, the force reduction rate of soleus and EDL from ΔEx50 mice was restored to the WT level (Figures 5E and 5F), indicating enhanced fatigue resistance. Improvement of muscle function correlated with increased dystrophin expression and decreased muscle degeneration (Figure S13).

Elevated serum creatine kinase (CK) is a pathological hallmark of DMD. After receiving the low- and high-dose All-In-One AAV treatment, CK levels in the ΔEx50 mice were reduced by 66% and 81%, respectively, compared to the untreated ΔEx50 littermates (Figure S14). Together, these findings demonstrate that KKH SaCas9-mediated single-cut gene editing improves muscle integrity and provides functional benefit to DMD ΔEx50 mice.

Discussion

Despite intense efforts to develop therapeutic strategies to restore dystrophin expression in DMD patients through oligonucleotide-mediated exon skipping and gene therapy with truncated forms of dystrophin, there remains a major unmet need for approaches to restore maximal portions of the dystrophin gene in patients with different DMD deletions.26, 27, 28 Exon deletions that disrupt the continuity of the dystrophin ORF in exon 51 represent the most predominant cause of DMD. Skipping or reframing exon 51, in principle, can provide therapeutic benefit to ~13% of the DMD population. To date, there is no report of using SaCas9-mediated single-cut gene editing to correct DMD mutations. Previously published studies employed two sgRNAs to direct SaCas9 to induce two DNA DSBs flanking an out-of-frame exon.,,17, 18, 19,

In this study, we developed a single-cut gene-editing strategy in which KKH SaCas9 introduces a single DNA DSB within exon 51 to reframe the dystrophin ORF in human cardiomyocytes lacking exons 48–50 and in mouse muscles lacking exon 50. Cardiomyocytes derived from human iPSCs with the ΔEx48–50 mutation and corrected by editing with KKH SaCas9 restored dystrophin expression and showed improved calcium transient kinetics. We also packaged KKH SaCas9 and its sgRNA into a single AAV vector and performed in vivo gene editing. DMD ΔEx50 mice receiving systemic All-In-One AAV treatment restored dystrophin expression with consequent improvement in muscle contractility and force. This study represents the first application of KKH SaCas9-mediated single-cut gene editing for the treatment of DMD.

SpCas9-mediated single-cut gene editing has been widely used for correcting diverse DMD mutations with high efficiency, especially for mutations that can be reframed by a 1-bp insertion.,,,,, In contrast, prior studies of SaCas9-mediated correction of DMD mutations relied on two sgRNAs to completely excise the out-of-frame exon.,,17, 18, 19, These different approaches are dependent on the topological distinctions between the DNA DSBs induced by Sp- and SaCas9. Studies using molecular dynamics simulations suggest that an SpCas9-induced DNA DSB generates a staggered cut, producing a single-nucleotide 5′ overhang, leading to a high frequency of a 1-bp insertion after NHEJ-mediated repair., In contrast, the DSB generated by SaCas9 cutting is blunt ended, with indels with varying lengths. Therefore, NHEJ-mediated 1-bp insertion appears to be an SpCas9-specific phenomenon, which does not apply to SaCas9. This distinction poses limitations to SaCas9-mediated single-cut gene editing as a general strategy for reframing out-of-frame exons. In order to address this issue, we screened for sgRNAs capable of directing KKH SaCas9 to induce a DNA DSB between a microhomology sequence. Studies have demonstrated that DNA DSBs around regions of microhomology tend to generate deletions with predictable length., As expected, with SaCas9 cutting, we observed a majority of productive editing events containing a precise deletion of the 5′-CT-3′ dinucleotide.

Although CRISPR correction of DMD has shown promise in pre-clinical studies, several questions and challenges remain to be addressed. The first concern is durability of CRISPR gene editing in muscle cells. Skeletal muscle has resident stem cells (satellite cells) capable of regenerating or fusing to myofibers. Although there is increasing evidence that AAV9 delivery of CRISPR-Cas9 components can transduce and edit satellite cells,,, the efficiency of viral transduction and gene editing in satellite cells remains low. Whether unedited satellite cells will gradually dilute out corrected nuclei in regenerating myofibers remains unknown. Engineering novel AAV serotypes with strong tropism to satellite cells may offer a potential solution to this issue. Another concern is the AAV dose administered in gene editing. In pre-clinical studies, the average AAV dose used in in vivo gene editing of DMD animal models varies between 1.6 × 1014 and 1.8 × 1015 vg/kg,10, 11, 12, 13, 14,16, 17, 18, 19, 20, 21, 22 which becomes an obvious burden for industrial production and clinical translation. Similarly, AAV dose of 2.0 × 1014 vg/kg or higher has been necessary to direct therapeutically beneficial levels of micro-dystrophin in early-stage clinical trials.

Self-complementary AAV has been shown to be superior to single-stranded AAV in viral transduction and CRISPR gene editing., When self-complementary AAV is used for CRISPR sgRNA delivery, the viral dose can be reduced to 8 × 1013 vg/kg. However, SaCas9 used in this study is too large to be packaged into self-complementary AAV. Potential solutions to address these concerns include (1) screening more compact CRISPR/Cas systems to bypass the packaging limit of self-complementary AAV and (2) dividing SaCas9 into two parts to accommodate self-complementary AAV packaging and using the split-intein system to reconstitute the full-length Cas9 after AAV delivery.,

Although complete restoration of normal levels of dystrophin is not achievable for therapeutic gene editing because AAV viral transduction of skeletal muscles is not 100%, studies in patients with Becker muscular dystrophy have estimated that ~15% of normal levels of dystrophin protein could provide therapeutic benefits. Our in vivo data demonstrate that All-In-One SaCas9-mediated single-cut gene editing has high efficiency in dystrophin restoration, capable of restoring 30%–50% of dystrophin protein levels in multiple skeletal muscles and 50% in the heart within 4 weeks of administration. Therefore, All-In-One SaCas9-mediated single-cut gene editing developed in this study shows strong potential for therapeutic translation and represents a promising therapy for permanent correction of DMD.

Finally, we have recently reported the effectiveness of base editing as a strategy for exon skipping in DMD via splice-site modification. Whether this approach might be adapted to an All-In-One strategy is under investigation. Together, these various approaches add to the expanding toolbox of gene-editing strategies that may ultimately be applied to different DMD mutations.

Materials and methods

Study design

This study was designed with the primary aim of investigating the feasibility of using CRISPR/SaCas9-mediated single-cut gene editing for the correction of DMD mutations. The secondary objective was to design an All-In-One AAV packaging system to deliver CRISPR/SaCas9 and sgRNAs for in vivo therapeutic gene editing. We did not use exclusion, randomization, or blind approaches to assign the animals for the experiments. Grip-strength tests, histology validation, immunostaining analysis, CK analysis, and muscle electrophysiology were performed as blinded experiments. For each experiment, sample size reflects the number of independent biological replicates and was provided in the figure legends.

KKH SaCas9 vector cloning and AAV vector production

WT SaCas9 complementary DNA (cDNA) was cut from pX601 plasmid, a gift from F. Zhang (Addgene plasmid #61591), using AgeI-HF and BamHI-HF, and subcloned into pLbCpf1-2A-GFP plasmid by replacing LbCpf1, generating the pSaCas9-2A-GFP plasmid. Modified SaCas9 sgRNA scaffold and KKH SaCas9 C terminus cDNA (E782K/N968K/R1015H) were synthesized as gBlocks (Integrated DNA Technologies) and subcloned into pSaCas9-2A-GFP plasmid using an In-Fusion cloning kit (Takara Bio), generating the pKKH-SaCas9-2A-GFP plasmid. The sgRNAs targeting human DMD exon 51 or mouse Dmd exon 51 were subcloned into the newly generated pKKH-SaCas9-2A-GFP plasmid using BbsI digestion and T4 ligation. KKH SaCas9, 7SK, and U6 sgRNA expression cassettes were subcloned into the pSSV9 single-stranded AAV plasmid using an In-Fusion cloning kit (Takara Bio). Cloning primer sequences are listed in Table S1. AAV viral plasmid was column purified and digested with SmaI and AhdI to check inverted terminal repeat (ITR) integrity. AAV was packaged by Boston Children’s Hospital Viral Core, and serotype 9 was chosen for capsid assembly. AAV titer was determined by quantitative real-time PCR assay.

Human iPSC maintenance, nucleofection, and differentiation

DMD ΔEx48–50 iPSCs (RBRC-HPS0164) were purchased from Cell Bank RIKEN BioResource Center. Stem cell work described in this manuscript has been conducted under the oversight of the UT Southwestern Stem Cell Research Oversight (SCRO) Committee. Human iPSCs were cultured in mTeSR plus medium (STEMCELL Technologies) and passaged approximately every 4 days (1:18 split ratio). One hour before nucleofection, iPSCs were treated with 10 μM ROCK inhibitor (Y-27632) and dissociated into single cells using Accutase (Innovative Cell Technologies). iPSCs (1 × 106) were mixed with 5 μg of the pKKH-SaCas9-2A-GFP plasmid. The P3 Primary Cell 4D-Nucleofector X kit (Lonza) was used for nucleofection according to the manufacturer’s protocol. After nucleofection, iPSCs were cultured in mTeSR plus medium supplemented with 10 μM ROCK inhibitor, and Primocin (100 μg/mL; InvivoGen). Three days after nucleofection, GFP+ cells were sorted by FACS and subjected to TIDE analysis. KKH SaCas9-edited iPSC mixtures and single clones were differentiated into cardiomyocytes, as previously described.

Calcium imaging

Calcium imaging was performed as previously described. iPSC-derived cardiomyocytes were replated on glass surfaces at single-cell density and loaded with the fluorescent calcium indicator Fluo-4 AM (Thermo Fisher) at 2 μM. Spontaneous calcium transients of beating iPSC-derived cardiomyocytes were imaged at 37°C using a Nikon A1R+ confocal system. Calcium transients were processed using Fiji software and analyzed using Microsoft Excel and Clampfit 10.7 software (Axon Instruments). The calcium-release phase was represented with time to peak, which was calculated as the time from baseline to maximal point of the transient. The calcium-reuptake phase was represented with the time constant tau by fitting the decay phase of calcium transients with a first-order exponential function.

In vivo AAV delivery into ΔEx50 mice

The ΔEx50 DMD mouse model was developed by deleting the mouse Dmd exon 50 using CRISPR/Cas9-mediated mutagenesis. P4 ΔEx50 mice were injected intraperitoneally with 80 μL of AAV9 containing 2 × 1014 (low dose) or 4 × 1014 vg/kg (high dose) of All-In-One AAV9-KKH-SaCas9-sgRNAs using an ultrafine BD insulin syringe (Becton Dickinson). Four weeks after systemic delivery, ΔEx50 mice and WT littermates were dissected for physiological, biochemical, and histological analysis. Animal work described in this manuscript has been approved and conducted under the oversight of the University of Texas Southwestern Institutional Animal Care and Use Committee.

Genomic DNA and RNA isolation, cDNA synthesis, and PCR amplification

Genomic DNA of DMD ΔEx48–50 iPSCs, skeletal muscles and hearts of ΔEx50 mice was isolated using DirectPCR (cell) lysis reagent (Viagen Biotech) according to the manufacturer’s protocol. Total RNA of skeletal muscles and heart of ΔEx50 mice was isolated using miRNeasy (QIAGEN) according to the manufacturer’s protocol. cDNA was reverse-transcribed from total RNA using iScript Reverse Transcription Supermix (Bio-Rad Laboratories) according to the manufacturer’s protocol. Genomic DNA and cDNA was PCR amplified using LongAmp Taq DNA polymerase (New England BioLabs) PCR products were sequenced and analyzed by TIDE analysis. Primer sequences are listed in Table S1.

Amplicon deep-sequencing analysis of genomic DNA

PCR of genomic DNA was performed using primers designed against the human DMD exon 51, off-target sites, and mouse Dmd exon 51. A second round of PCR was performed to add Illumina flow cell binding sequence and barcodes. All primer sequences are listed in Table S1. Deep sequencing and data analysis were performed as previously described.

Dystrophin immunocytochemistry and immunohistochemistry

Dystrophin immunocytochemistry was performed as previously described. Primary antibodies used in immunocytochemistry were mouse anti-dystrophin antibody (MANDYS8, Sigma-Aldrich, D8168) and rabbit anti-troponin I antibody (H170, Santa Cruz Biotechnology). Secondary antibodies used in immunocytochemistry were biotinylated horse anti-mouse immunoglobulin G (IgG) (BMK-2202, Vector Laboratories) and fluorescein-conjugated donkey anti-rabbit IgG (Jackson ImmunoResearch). Skeletal muscles and heart were cryosectioned into 8-μm transverse sections. Immunohistochemistry was performed as previously described. Antibodies used in immunohistochemistry were mouse anti-dystrophin antibody (MANDYS8, Sigma-Aldrich, D8168) and mouse on mouse biotinylated anti-mouse IgG (BMK-2202, Vector Laboratories).

Dystrophin western blot analysis

For western blot of iPSC-derived cardiomyocytes, 4 × 106 cells were lysed in lysis buffer (10% SDS, 62.5 mM tris [pH 6.8], 1 mM EDTA, and protease inhibitor). Heart and skeletal muscles were crushed into fine powder using a liquid-nitrogen-frozen crushing apparatus and lysed in the same lysis buffer as iPSC-derived cardiomyocytes. A total 50 μg of protein was loaded onto 4%–20% Criterion TGX Precast Midi Protein Gel (Bio-Rad Laboratories). Details of western blot running, transferring, and developing were previously described. Primary antibodies used in western blot were mouse anti-dystrophin antibody (MANDYS8, Sigma-Aldrich, D8168) and mouse anti-vinculin antibody (Sigma-Aldrich, V9131). Secondary antibody used in western blot was goat anti-mouse horseradish peroxidase (HRP) antibody (Bio-Rad Laboratories).

Electrophysiological analysis of isolated EDL and soleus muscles

Four weeks after systemic All-In-One AAV9-KKH-SaCas9-sgRNAs gene editing, soleus, and EDL muscles of ΔEx50 mice and WT littermates were isolated for electrophysiological analysis. In brief, soleus and EDL muscles were surgically isolated from 4-week-old ΔEx50 mice, mounted on Grass FT03.C force transducers, bathed in physiological salt solution at 37°C, and gassed continuously with 95% O2–5% CO2. After calibration, muscles were adjusted to initial length at which the passive force was 0.5 g and then stimulated with two platinum-wire electrodes to establish optimal length (Lo) for obtaining maximal isometric tetanic tension step by step following the protocol (at 150 Hz for 2 s). Specific force (mN/mm2) was calculated by normalizing contraction force to muscle cross-sectional area.

Statistics

All data are shown as means ± SEM. Unpaired two-tailed Student’s t tests was performed to analyze Figure 1D and Figures S13C and S13D; two-way analysis of variance (ANOVA) with post hoc Tukey’s multiple-comparisons test was performed to analyze Figure 4B; Brown-Forsythe and Welch ANOVA test was performed to analyze Figures 5A and 5B; nonlinear regression with extra sum-of-squares F test was performed to analyze Figures 5E and 5F; one-way ANOVA with post hoc Tukey’s multiple-comparisons test was performed to analyze the rest of data. A p < 0.05 value was considered statistically significant. Data analyses were performed with GraphPad Prism software.

Data and materials availability

All data needed to evaluate the conclusions in the paper are present in the paper and/or the supplemental information. Additional data related to this paper may be requested from the authors.