Fast, multiplexable and efficient somatic gene deletions in adult mouse skeletal muscle fibers using AAV-CRISPR/Cas9

Molecular screens comparing different disease states to identify candidate genes rely on the availability of fast, reliable and multiplexable systems to interrogate genes of interest. CRISPR/Cas9-based reverse genetics is a promising method to eventually achieve this. However, such methods are sorely lacking for multi-nucleated muscle fibers, since highly efficient nuclei editing is a requisite to robustly inactive candidate genes. Here, we couple Cre-mediated skeletal muscle fiber-specific Cas9 expression with myotropic adeno-associated virus-mediated sgRNA delivery to establish a system for highly effective somatic gene deletions in mice. Using well-characterized genes, we show that local or systemic inactivation of these genes copy the phenotype of traditional gene-knockout mouse models. Thus, this proof-of-principle study establishes a method to unravel the function of individual genes or entire signaling pathways in adult skeletal muscle fibers without the cumbersome requirement of generating knockout mice.


Figure S1 :
Figure S1: Expression of Cas9 in skeletal muscle fibers does not affect muscle physiology.a Cross sections of tibialis anterior (TA) muscle stained for type IIB (blue), type IIA (yellow) fibers and laminin (red) from control and Cas9mKI mice.b Quantification of the TA fiber-type composition.c Minimal fiber feret distribution (left) and mean minimal fiber feret (right) of TA muscle of control (grey) and Cas9mKI (green) mice according to their fiber type.d Whole-mount view of neuromuscular junctions of the extensor digitorum longus (EDL) muscle stained for the motor nerve using a mixture of antibodies directed against synaptic vesicle glycoprotein 2A (SV2; green) and neurofilament (2H3; green), and for postsynaptic AChRs with α-bungarotoxin (αBTX; magenta).e Ex-vivo twitch response of isolated soleus (SOL) muscle from Cas9mKI and control mice.Peak twitch (Pt), time-to-peak twitch (TPT) and half-relaxation time (1/2RT) are indicated.f Quantification of ex-vivo twitch response parameters (TPT, Pt, 1/2RT) of isolated SOL muscle from Cas9mKI and control mice.g Force-frequency curve (left) and fatigue response to multiple stimulations (right) of SOL muscle from control and Cas9mKI mice.h Changes in muscle mass of TA muscle of control or Cas9mKI mice 21 days post cardiotoxin (CTX) injury.While there is no significant difference between control and Cas9mKI mice, muscles are significantly heavier after regeneration (P < 0.05).i Cross sections of TA, stained for hematoxylin and eosin, of uninjured (vehicle) and injured (CTX) from control and Cas9mKI mice, 21 days post injection.Data are means ± SEM.For b and c, n = 4 mice.For e -g, n = 3 mice.For h, n = 5 (control) and 4 (Cas9mKI) mice.None of the data in b, c, e -g are not significantly different between control and Cas9mKI mice (P > 0.05) using unpaired student's two-sided t-test.For h, statistical significance is based on two-way ANOVA with Tukey's post-hoc test.*P < 0.05.Source data and precise p-values are provided as a source data file.

Figure S3 :
Figure S3: Efficiency testing of sgRNAs in cultured C2C12 myotubes.a Schematic of the expression plasmid used for transfection of C2C12 myoblasts.For abbreviations, see legend to Figure 1.U6: human U6 promoter; EFS: eukaryotic translation elongation factor 1 α short promoter; Pac: puromycin N-acetyltransferase. b Western blot analysis for FLAG-tagged Cas9 and PKCα using lysates from C2C12 myotubes after transfection with the indicated constructs and puromycin selection.c Quantification of protein abundance.d Tracking of Indels by Decomposition (TIDE) analysis of Cas9/sgPKCα-1expressing cells.e Frequency distribution of the DNA editing events around the binding site of sgPKCα-1.Data are means ± SEM with n = 3 wells for each sgRNA.Significance was determined using one-way ANOVA with Fishers LSD post-hoc test.*P < 0.05, **P< 0.01, ***P < 0.001.Source data and precise p-values are provided as a source data file.
Figure S5: AVVMYO-mediated sgRNA delivery into TA muscle induces a strong reduction of PKCα in nearby muscles.a Representative images of cross sections of extensor digitorum longus (EDL) and gastrocnemius (GAS) muscle stained for tdTomato (magenta) and DAPI (blue).b Distribution of AAVs in TA, EDL, GAS, heart and liver upon intramuscular injection of AAV9-3sgPKCα-1 (light blue) or AAVMYO-3sgPKCα-1 (cyan) into Cas9mKI mice.c, f Western blot analysis for PKCα and tdTomato of the indicated muscles and conditions.d, g Quantification of Western blots shown in c and f for the indicated proteins.Results for PKCα were normalized to the levels in PBS-injected muscles (grey).For tdTomato, levels of AAV9-3sgPKCα-1-injected muscle were set to 1.e, h TIDE analysis for sgPKCα-1 at its target site.Data are means ± SEM. n = 4 mice.Significance was determined using one-way ANOVA with Fishers LSD post-hoc test(d, Figure S7: Efficiency of sgRNA delivery using the liver-detargeted AAVMYO2 and AAVMYO3 compared to AAVMYO.a Representative images of dissected mice and cross sections of tibialis anterior (TA), extensor digitorum longus (EDL), soleus (SOL), triceps brachii (TRI) or diaphragm (DIA) muscle stained for tdTomato (magenta) and DAPI (blue), 6 weeks post-intravenous injection of PBS or AAV (1.0 x 10 14 vg/kg) into Cas9mKI mice.For comparison, images of AAVMYO-3sgPKCα-1-injected mice are included, same as on Fig. 4b.b Distribution of AAVs in TA, EDL, SOL, TRI and DIA upon intravenous injection of AAVMYO2-3sgPKCα-1 (light green) or AAVMYO3-3sgPKCα-1 (dark green) into Cas9mKI mice.c TIDE analysis of the sgPKCα-1-targeted Prkca locus.

Figure S11 :
Figure S11: Analysis of Musk transcripts upon AAVMYO-7sgMusk injection.a Fragment analyzer gel (top) and electropherogram (bottom) of PCR fragments for Musk cDNA of AAVMYO-7sgMusk-injected (red colors) tibialis anterior (TA) muscle of Cas9mKI mice and controls (grey).Primers used for PCR amplification are indicated.Dotted lines indicate size of major PCR products from controls (815 bp: all exons included; 785 bp: skip exon 6; 754 bp: skip exon 6 and 7; 642 bp: skip of exon 5 and 6).b Sashimi plot indicating NGS read junctions in control or AAVMYO-7sgMusk-injected Cas9mKI mice.Read junctions observed in control mice are indicated in grey; dotted grey lines are reads from alternatively spliced Musk transcripts.Reads junctions occurring only in AAVMYO-7sgMusk-injected mice are indicated in red colors corresponding to the amount of virus used.Height of black squares and numbers represents number of reads of this particular variant.Localization of each sgMusk is highlighted in brown.Exons are drawn to scale.Data represents 1 mouse per condition.Source data are provided as a source data file.
Figure S12: Intramuscular AAV9-7sgMusk injection does not spread systemically but results in the local loss of NMJs. a Schematic presentation of the experimental procedure using two doses of sgRNA-delivering AAV9.b Body weight progression of controls (grey) and AAV9-7sgMusk-injected Cas9mKI mice (red colors) at the indicated doses.Mass changes in muscles of the injected (c) and the contralateral leg (d).Muscles are tibialis anterior (TA), extensor digitorum longus (EDL), soleus (SOL) and gastrocnemius (GAS).e Relative mRNA expression of Musk in AAV9-7sgMusk-injected TA muscle of Cas9mKI mice.f Representative whole-mount images of extensor digitorum longus (EDL) muscles of controls and AAV9-7sgMusk-injected Cas9mKI mice.
Figure S13: On/off-target editing by individual sgAcvr2a-(1-7) and sgAcvr2b-(1-7).a Schematic representation of amplicon design for each sgAcvr2a-(1-7) and sgAcvr2b-(1-7).Location and orientation of each sgRNA, and forward (FW) and reverse (RV) primers for each amplicon are indicated.b, c CRISPResso2 analysis for each sgAcvr2a (b) and for each sgAcvr2b (c) at its target site.d, e CRISPResso2 analysis for each sgAcvr2a (d) and for each sgAcvr2b (e) at its primary off-target site.Note that sgAcvr2b-4 has a strong editing activity at the off-target site.DNA was isolated from tibialis anterior (TA) muscle from PBS-(control; grey) or AAVMYO-7sgAcvr2a/b-injected Cas9mKI mice (orange).n = 3 mice.Note that there were only n = 2 for the control sample of sgAcvr2a-6 in b.Statistical significance is based on unpaired student's two-sided t-test.*P < 0.05, **P < 0.01, ***P < 0.001.Source data and precise p-values are provided as a source data file.