Research ArticleGENETICS

CRISPR-Cpf1 correction of muscular dystrophy mutations in human cardiomyocytes and mice

See allHide authors and affiliations

Science Advances  12 Apr 2017:
Vol. 3, no. 4, e1602814
DOI: 10.1126/sciadv.1602814
  • Fig. 1 Correction of DMD mutations by Cpf1-mediated genome editing.

    (A) A DMD deletion of exons 48 to 50 results in splicing of exons 47 to 51, generating an out-of-frame mutation of dystrophin. Two strategies were used for the restoration of dystrophin expression by Cpf1. In the “reframing” strategy, small INDELs in exon 51 restore the protein reading frame of dystrophin. The “exon skipping” strategy is achieved by disruption of the splice acceptor of exon 51, which results in splicing of exons 47 to 52 and restoration of the protein reading frame. (B) The 3′ end of an intron is T-rich, which generates Cpf1 PAM sequences, enabling genome cleavage by Cpf1. (C) Illustration of Cpf1 gRNA targeting DMD exon 51. The T-rich PAM (red line) is located upstream of exon 51 near the splice acceptor site. The sequence of the Cpf1 g1 gRNA targeting exon 51 is shown, highlighting the complementary nucleotides in blue. Cpf1 cleavage produces a staggered end distal to the PAM site (demarcated by red arrowheads). The 5′ region of exon 51 is shaded in light blue. Exon sequence is in uppercase letters. Intron sequence is in lowercase letters. (D) Illustration of a plasmid encoding human codon-optimized Cpf1 (hCpf1) with a nuclear localization signal (NLS) and 2A-GFP, driven by a hybrid form of cytomegalovirus and chicken β-actin promoters (CBh). The plasmid also encodes a Cpf1 gRNA driven by the U6 promoter. Cells transfected with this plasmid express GFP, allowing for selection of Cpf1-expressing cells by FACS. (E) T7E1 assays using human 293T cells or DMD iPSCs (Riken51) transfected with plasmid expressing LbCpf1 or AsCpf1, gRNA, and GFP show genome cleavage at DMD exon 51. Red arrowheads point to cleavage products. M, marker; bp, base pair.

  • Fig. 2 DMD iPSC-derived cardiomyocytes express dystrophin after Cpf1-mediated genome editing by reframing.

    (A) DMD skin fibroblast-derived iPSCs were edited by Cpf1 using gRNA (corrected DMD iPSCs) and then differentiated into cardiomyocytes (corrected cardiomyocytes) for analysis of genetic correction of the DMD mutation. (B) A DMD deletion of exons 48 to 50 results in splicing of exon 47 to 51, generating an out-of-frame mutation of dystrophin. Forward primer (F) targeting exon 47 and reverse primer (R) targeting exon 52 were used in RT-PCR to confirm the reframing strategy by Cpf1-meditated genome editing in cardiomyocytes. Uncorrected cardiomyocytes lack exons 48 to 50. In contrast, after reframing, exon 51 is placed back in frame with exon 47. (C) Sequencing of representative RT-PCR products shows that uncorrected DMD iPSC-derived cardiomyocytes have a premature stop codon in exon 51, which creates a nonsense mutation. After Cpf1-mediated reframing, the ORF of dystrophin is restored. Dashed red line denotes exon boundary. (D) Western blot analysis shows dystrophin expression in a mixture of DMD iPSC-derived cardiomyocytes edited by reframing with LbCpf1 or AsCpf1 and g1 gRNA. Even without clonal selection, Cpf1-mediated reframing is efficient and sufficient to restore dystrophin expression in the cardiomyocyte mixture. α-Myosin heavy chain (αMHC) is loading control. (E) Immunocytochemistry shows dystrophin expression in iPSC-derived cardiomyocyte (CM) mixtures following LbCpf1- or AsCpf1-mediated reframing. Red, dystrophin staining; green, troponin I staining. Scale bar, 100 μm. (F) Western blot analysis shows dystrophin expression in single clones (#2 and #5) of iPSC-derived cardiomyocytes following clonal selection after LbCpf1-mediated reframing. αMHC is loading control. (G) Immunocytochemistry shows dystrophin expression in clone #2 LbCpf1-edited iPSC-derived cardiomyocytes. Scale bar, 100 μm. (H) Quantification of mtDNA copy number in single clones (#2 and #5) of LbCpf1-edited iPSC-derived cardiomyocytes. Data are means ± SEM (n = 3). &P < 0.01 and #P < 0.005. ns, not significant. (I) Basal OCR of single clones (#2 and #5) of LbCpf1-edited iPSC-derived cardiomyocytes, and OCR in response to oligomycin, FCCP, rotenone, and antimycin A, normalized to cell number. Data are means ± SEM (n = 5). *P < 0.05, &P < 0.01, and #P < 0.005.

  • Fig. 3 DMD iPSC-derived cardiomyocytes express dystrophin after Cpf1-mediated exon skipping.

    (A) Two gRNAs [either gRNA (g2 or g3), which target intron 50, and the other (g1), which targets exon 51] were used to direct Cpf1-mediated removal of the exon 51 splice acceptor site. (B) T7E1 assay using 293T cells transfected with LbCpf1 and gRNA2 (g2) or gRNA3 (g3) shows cleavage of the DMD locus at intron 50. Red arrowheads denote cleavage products. M, marker; Ctrl, control. (C) PCR products of genomic DNA isolated from DMD-iPSCs transfected with a plasmid expressing LbCpf1, g1 + g2, and GFP. The lower band (red arrowhead) indicates removal of the exon 51 splice acceptor site. (D) Sequence of the lower PCR band from (C) shows a 200-bp deletion, spanning from the 3′ end of intron 50 to the 5′ end of exon 51. This confirms removal of the “ag” splice acceptor of exon 51. The sequence of the uncorrected allele is shown above that of the LbCpf1-edited allele. (E) RT-PCR of iPSC-derived cardiomyocytes using primer sets described in Fig. 2B. The 700-bp band in the WT lane is the dystrophin transcript from exons 47 to 52; the 300-bp band in the uncorrected lane is the dystrophin transcript from exons 47 to 52 with exon 48 to 50 deletion; and the lower band in the g1 + g2 mixture lane (edited by LbCpf1) shows exon 51 skipping. (F)Sequence of the lower band from (E) (g1 + g2 mixture lane) confirms skipping of exon 51, which reframed the DMD ORF. (G) Western blot analysis shows dystrophin protein expression in iPSC-derived cardiomyocyte mixtures after exon 51 skipping by LbCpf1 with g1 + g2. αMHC is loading control. (H) Immunocytochemistry shows dystrophin expression in iPSC-derived cardiomyocyte mixtures following Cpf1-mediated exon skipping with g1 + g2 gRNA compared to WT and uncorrected cardiomyocyte mixtures. Red, dystrophin staining; green, troponin I staining. Scale bar, 100 μm.

  • Fig. 4 CRISPR-Cpf1–mediated editing of exon 23 of the mouse Dmd gene.

    (A) Illustration of mouse Dmd locus highlighting the mutation at exon 23. Sequence shows the nonsense mutation caused by C-to-T transition, which creates a premature stop codon. (B) Illustration showing the targeting location of gRNAs (g1, g2, and g3) (in light blue) in exon 23 of the Dmd gene. Red line represents LbCpf1 PAM. (C) T7E1 assay using mouse 10T1/2 cells transfected with LbCpf1 or AsCpf1 with different gRNAs (g1, g2, or g3) targeting exon 23 shows that LbCpf1 and AsCpf1 have different cleavage efficiency at the Dmd exon 23 locus. Red arrowheads show cleavage products of genome editing. M, marker. (D) Illustration of LbCpf1-mediated gRNA (g2) targeting of Dmd exon 23. Red arrowheads indicate the cleavage site. The ssODN HDR template contains the mdx correction, four silent mutations (green), and a Tse I restriction site (underlined).

  • Fig. 5 CRISPR-LbCpf1–mediated Dmd correction in mdx mice.

    (A) Strategy of gene correction in mdx mice by LbCpf1-mediated germline editing. Zygotes from intercrosses of mdx parents were injected with gene editing components (LbCpf1 mRNA, g2 gRNA, and ssODN) and reimplanted into pseudopregnant mothers, which gave rise to pups with gene correction (mdx-C). (B) Illustration showing LbCpf1 correction of mdx allele by HDR or NHEJ. (C) Genotyping results of LbCpf1-edited mdx mice. Top: T7E1 assay. Blue arrowhead denotes uncleaved DNA, and red arrowhead shows T7E1-cleaved DNA. Bottom: Tse I RFLP assay. Blue arrowhead denotes uncorrected DNA, and red arrowhead points to Tse I cleavage, indicating HDR correction. mdx-C1 to mdx-C5 denote LbCpf1-edited mdx mice. (D) Top: Sequence of WT Dmd exon 23. Middle: Sequence of mdx Dmd exon 23 with C-to-T mutation, which generates a stop codon. Bottom: Sequence of Dmd exon 23 with HDR correction by LbCpf1-mediated editing. Black arrows point to silent mutations introduced by the ssODN HDR template. (E) H&E staining of tibialis anterior (TA) and gastrocnemius/plantaris (G/P) muscles from WT, mdx, and LbCpf1-edited mice (mdx-C). Scale bar, 100 μm. (F) Immunohistochemistry of tibialis anterior and gastrocnemius/plantaris muscles from WT, mdx, and LbCpf1-edited mice (mdx-C) using an antibody to dystrophin (red). mdx muscle showed fibrosis and inflammatory infiltration, whereas mdx-C muscle showed normal muscle structure.

  • Table 1 Serum CK measurement and forelimb grip strength of WT, mdx, and LbCpf1-corrected mdx-C mice.

    M, male; F, female.

    Mouse no.Percent correction by HDRSexWeight (g)CK (U/liter)Forelimb grip strength (grams of force)
    Trial 1Trial 2Trial 3Trial 4Trial 5Trial 6Average ± SD
    WT-1M16.64551031069771888291.2 ± 13.4
    WT-2M19.522087957373747880.0 ± 9.1
    WT-3M18.730679977478848482.7 ± 8.0
    WT-4F15.518486978583858887.3 ± 5.0
    WT-5F15.4175888510096888690.5 ± 6.1
    WT-6F12.615776757378646171.2 ± 7.0
    mdx-10M18.8857976928633322958.0 ± 29.9
    mdx-20M21.0944062585445454751.8 ± 7.3
    mdx-30M21.5593677546957566162.3 ± 8.9
    mdx-40F16.1630669636961676064.8 ± 4.0
    mdx-50F15.8634983888559555470.7 ± 16.2
    mdx-60F16.2416869715359595761.3 ± 7.1
    mdx-C116%F21.71233112111112115101109110.0 ± 4.8
    mdx-C28%F19.74920119109108958685100.3 ± 13.8
    mdx-C350%F20.2248110115114112112104111.2 ± 3.9
    mdx-C48%M22.7660749454230252135.3 ± 11.5
    mdx-C525%F17.732399211096911109599.0 ± 8.7

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/3/4/e1602814/DC1

    fig. S1. Genome editing of DMD exon 51 and Dmd exon 23 by LbCpf1 or AsCpf1.

    fig. S2. Genomic PCR and T7E1 assay of off-target sites.

    fig. S3. Capillary electrophoresis and fragment analysis of DMD g1 off-target sites.

    fig. S4. Capillary electrophoresis and fragment analysis of DMD g2 off-target sites.

    fig. S5. Histological analysis of muscles from WT, mdx, and LbCpf1-edited mice (mdx-C).

    fig. S6. Immunohistochemistry of skeletal muscles, heart, and brain from WT, mdx, and LbCpf1-edited mice (mdx-C).

    fig. S7. H&E staining of skeletal muscles, heart, and brain from WT, mdx, and LbCpf1-edited mice (mdx-C).

    fig. S8. Western blot analysis of skeletal muscles, heart, and brain from WT, mdx, and LbCpf1-edited mice (mdx-C).

    fig. S9. T7E1 and Tse I RFLP analysis of germ cells from LbCpf1-edited mice (mdx-C) and uncorrected mdx mice.

    table S1. Sequence of the on-target site and 10 potential off-target sites.

    table S2. Primers used in the off-target site analysis.

    table S3. Comparison of CRISPR-Cas9– and CRISPR-Cpf1–mediated HDR correction in mdx mice.

    table S4. Sequence of primers.

  • Supplementary Materials

    This PDF file includes:

    • fig. S1. Genome editing of DMD exon 51 and Dmd exon 23 by LbCpf1 or AsCpf1.
    • fig. S2. Genomic PCR and T7E1 assay of off-target sites.
    • fig. S3. Capillary electrophoresis and fragment analysis of DMD g1 off-target sites.
    • fig. S4. Capillary electrophoresis and fragment analysis of DMD g2 off-target sites.
    • fig. S5. Histological analysis of muscles from WT, mdx, and LbCpf1-edited mice (mdx-C).
    • fig. S6. Immunohistochemistry of skeletal muscles, heart, and brain from WT, mdx, and LbCpf1-edited mice (mdx-C).
    • fig. S7. H&E staining of skeletal muscles, heart, and brain from WT, mdx, and LbCpf1-edited mice (mdx-C).
    • fig. S8. Western blot analysis of skeletal muscles, heart, and brain from WT, mdx, and LbCpf1-edited mice (mdx-C).
    • fig. S9. T7E1 and Tse I RFLP analysis of germ cells from LbCpf1-edited mice (mdx-C) and uncorrected mdx mice.
    • table S1. Sequence of the on-target site and 10 potential off-target sites.
    • table S2. Primers used in the off-target site analysis.
    • table S3. Comparison of CRISPR-Cas9– and CRISPR-Cpf1–mediated HDR correction in mdx mice.
    • table S4. Sequence of primers.

    Download PDF

    Files in this Data Supplement: