Methods for gene repair in primary human muscle stem cells (satellite cells) in vitro and gene-corrected human muscle stem cells

Abstract

The present invention relates to a method for gene repair in primary human muscle stem cells (satellite cells) in vitro, which comprises the steps of: providing a sample of isolated muscle fiber-containing tissue sample taken from at least one patient with a monogenic muscle disease, said monogenic muscle disease being caused by at least one mutation in at least one gene encoding at least one muscle protein; isolating and culturing primary stem cells from said muscle fiber-containing tissue sample; and correcting said at least one mutation in at least one gene encoding at least one muscle protein in said cultured primary stem cells by targeted modification of said at least one mutation by gene editing using CRISPR / Cas-based tools.The present invention also relates to the gene-repaired human muscle stem cells obtained in such a method.

Description

[Technical field]

[0001] The present invention relates to a method for in vitro / ex vivo gene repair in primary human muscle stem cells (satellite cells) and to gene-corrected human muscle stem cells obtained using said method. [Background technology]

[0002] Myofibers are syncytial structures with postmitotic nuclei that are formed by the fusion of muscle precursor cells called myoblasts during prenatal and postnatal development. Skeletal muscles can be regenerated from muscle stem cells (MuSCs), also called satellite cells, which are a pool of tissue-specific stem cells located between the basement membrane and the sarcolemma that surrounds all fibers (Non-Patent Document 1). In healthy muscles, satellite cells are quiescent or undergo a slow cell cycle. When activated in response to severe injury, satellite cells proliferate over a wide area and give rise to numerous myoblasts that fuse with damaged myofibers and with each other to generate new myofibers (Non-Patent Document 2). Skeletal muscle regeneration cannot occur without satellite cells. Muscular dystrophy (MD) patients suffer from constant tissue degeneration, which causes constant activation of satellite cells, leading to satellite cell exhaustion, failure to regenerate, and replacement of muscle with fat and connective tissue (Non-Patent Document 3).

[0003] Although cell replacement therapy using well-defined, highly myogenic cell populations could be a safe and long-term therapeutic avenue for MD patients, MuSCs are scarce and traditionally difficult to manipulate ex vivo. In addition, skeletal muscle is the most abundant tissue in the body, making it extremely challenging to develop cell replacement therapy for MD patients.

[0004] Various methods have been developed to generate new muscle cells for use in cell replacement therapy for MD. Allogeneic transplantation of myoblasts or mesoangioblasts has so far failed to produce clinical utility. Induced pluripotent stem cells (iPSCs) are an attractive and unlimited source of healthy cells, and several protocols exist to differentiate these into transplantable muscle precursor-like cells. However, iPSC-derived muscle cells have yet to meet quality and safety standards for transplantation into patients. Overall, allogeneic transplantation requires immunosuppression and cloaking of transplantable cells, adding an additional level of uncertainty that has yet to be demonstrated for iPSC-derived muscle cells.

[0005] Autologous transplantation of primary muscle stem cells (MuSCs) may be more predictable.

[0006] It has been previously shown that MuSCs can be isolated and expanded substantially from human muscle biopsy specimens, and that MuSCs maintain regenerative potential in vivo in xenotransplantation models (Non-Patent Document 4). To use MuSCs in an autologous setting in MD patients, the genetic defect must be corrected before retransplantation. Genetically corrected muscle stem cells will be very similar to, but not identical to, healthy ("wild-type") muscle stem cells. Formally, this has been demonstrated in base-edited human fibroblasts carrying a mutation in LMNA (Non-Patent Document 5). Since there are no healthy MuSCs taken from patients (because everyone has a genetic defect), gene repair is necessary in a therapeutic sense.

[0007] Muscular dystrophies (MD) are more than 40 monogenic diseases that lead to severe muscle weakness. Currently, there is no curative treatment. Skeletal muscle can be regenerated from muscle stem cells and their progeny, myoblasts, which are the progeny of muscle precursor cells.

[0008] There is currently no treatment or cure for MD. However, there are many efforts to change this. Companies and research groups are working intensively on solutions. An exon skipping strategy, a gene replacement approach using cDNA packaged in an AAV vector, is in clinical trials for some MD. Cell-based therapies for MD are feasible and will probably be available soon (Non-Patent Document 6). Cell-based therapeutic approaches could rely either on "universal donor cells" that lack individual immunological markers, or on autologous cells that are "genetically corrected" and transplanted into the patient.

[0009] The CRISPR / Cas system can make it relatively easy to target defined regions of the genome with unprecedented precision. Since the initial description of Cas9 as an RNA-guided endonuclease, numerous directed evolution and mutagenesis approaches have produced Cas enzymes with enhanced specificity and increased safety in terms of on-target vs. off-target editing profiles. In addition, engineered Cas9-fusion protein-like base editors and prime editors have enabled precise rewriting of genomes without relying on cellular DNA repair pathways, cell cycle phases, or exogenous DNA templates. One example is the adenine base editor (ABE), which converts A nucleotides to G nucleotides in targeted genomic loci by combining the RNA-guided DNA binding ability of Cas9 with the enzymatic activity of evolved tRNA adenosine deaminase.

[0010] Chemmello et al. (Non-Patent Document 7) outline the application of the CRISPR / Cas system to correct muscular dystrophies. Various strategies have been described: A) Exon deletion by double-cut myoediting: Here, the mutated exon is excised to obtain a truncated but functional dystrophin. This strategy is effective to delete exons that code for regions of dystrophin that tolerate the deletion. However, exon deletion can lead to diverse and unpredictable genome modifications, including the integration of foreign DNA or aberrant splicing at the cut site. B) Exon skipping and exon reframing by single-cut myoediting to restore the correct ORF: Only one sgRNA is used to generate a single cut in the genome. The sgRNA is designed to target genomic sequences in close proximity to the intron-exon region of an out-of-frame exon and produce one DSB. Another strategy exists to induce exon skipping with antisense oligonucleotides without modifying the genome and may be safe. However, this strategy can only be used to treat a subset of patients with defined types of mutations in genes such as DMD, and although removing one or a few exons is theoretically advantageous, it requires regular administration. C) Nucleotide muscle editing to correct point mutations: This allows site-specific single-base substitutions.

[0011] Chemmello et al. also outline possible in vivo delivery methods for the CRISPR gene editing system, including engineered nanoparticles, muscle precursor engraftment, and AAV delivery.

[0012] In vivo gene replacement therapy using adeno-associated viral vectors (AAV) may be feasible for genes up to a certain size. However, exogenously provided coding sequences are not physiologically controlled in terms of splicing or spatiotemporal expression, and there is a risk of insertional mutagenesis. AAV-mediated delivery of CRISPR / Cas9 directly into muscle allows highly efficient in vivo gene editing in Duchenne muscular dystrophy (DMD) mice and large animal models, potentially reaching and permanently repairing a very large proportion of myonuclei. In vivo adenine base editing (ABE) in muscle by AAV delivery has been achieved in a DMD mouse model. However, the disease process may not be reversible by any treatment unless performed before substantial degeneration, fatty fiber replacement, and muscle stem cell depletion occur.

[0013] In addition, many individuals have pre-existing antibodies against AAV, Cas9, or other Cas proteins that prevent in vivo delivery of any of these. Even in patients without pre-existing immunity, repeated administration of AAV (for gene replacement therapy) or AAV-Cas9 (for in vivo gene editing) would be impossible due to adaptive immunity to the viral or bacterial Cas9 protein triggered by the first administration. This may be particularly crucial for gene replacement therapy using AAV vectors, because skeletal muscle tissue has a constant turnover (much faster than in dystrophic muscle), preventing long-term transgene expression by gradual loss of the AAV genome. Toxicity of AAV systemic administration has also caused the deaths of three patients from liver-related adverse events in a clinical trial for myotubular myopathy. In two other trials for spinal muscular atrophy (SMA) and DMD, adverse events arose from immune responses to AAV.

[0014] Indeed, persistent Cas9 expression after AAV-mediated CRISPR / Cas9 delivery has been shown to induce muscle inflammation and anti-Cas9 humoral and cellular immune responses in dogs, which appear to be specific to Cas9 and may represent a significant barrier to AAV-CRISPR therapy in large mammals (Non-Patent Document 8).

[0015] Non-targeted insertion of therapeutic transgenes, e.g., via transposons or integrating viral vectors, carries the risk of insertional mutagenesis due to uncontrolled insertion sites and the addition of foreign sequences before and after the transgene that are necessary for the genome integration process.

[0016] Although in vivo gene editing interventions have been successful in repairing genetic defects in postmitotic myonuclei in muscle fibers of mice and dogs (Non-Patent Document 9) and pigs (Non-Patent Document 10), reaching and repairing MuSCs remains a major challenge for approaches that deliver gene editing machinery in vivo. Furthermore, it is well known that in muscular dystrophies, skeletal muscle is gradually replaced by fat and connective tissue and loses regenerative capacity, especially due to the depletion of MuSCs. Therefore, in vivo gene editing may not provide long-term therapeutic effects, especially in muscular dystrophies, due to the fast turnover of skeletal muscle. Therefore, a pool of healthy MuSCs is required for sustained healthy muscle homeostasis. [Prior art documents] [Patent documents]

[0017] [Patent Document 1] WO2016 / 030371A1 [Non-patent literature]

[0018] [Non-Patent Document 1] Mauro A. SATELLITE CELL OF SKELETAL MUSCLE FIBERS. J Biophys Biochem Cytol. 1961;9(2):493-495 [Non-licensed document 2] Yin H, Price F, Rudnicki MA. Satellite Cells and the Muscle Stem Cell Niche. Physiol Rev. 2013;93(1):23-67 [Non-licensed document 3] Blau HM, Webster C, Pavlath GK. Defective myoblasts identified in Duchenne muscular dystrophy. Proc Natl Acad Sci USA. 1983;80(15):4856-4860

Non-licensed Document 4

Non-licensed Document 5

Non-licensed Document 6

Non-licensed Document 7

Non-licensed Document 8

Non-licensed literature 9

Non-licensed literature 10

Non-licensed Document 11

Non-licensed Document 12

Non-licensed Document 13

Non-licensed Document 14

Non-licensed Document 15

Non-licensed Document 16

Non-licensed Document 17

[0019] As can be seen, currently available approaches for genetic repair of muscle cells and tissues remain limited.

[0020] Therefore, there is a need for alternative approaches for genetic repair of muscle cells that can be used for cell replacement therapy without the drawbacks mentioned above. It would be particularly advantageous to provide an effective method for genetic repair of autologous primary muscle stem cells of patients with muscular dystrophy. [Means for solving the problem]

[0021] This object is solved by a method having the features of claim 1.

[0022] Thus, there is provided an ex vivo (in vitro) method for gene repair in primary human muscle stem cells (satellite cells), said method comprising: providing a sample of an isolated muscle fiber-containing tissue specimen taken from at least one patient having a monogenic muscle disease, said monogenic muscle disease being caused by at least one mutation in at least one gene encoding at least one muscle protein; isolating and culturing primary muscle stem cells from the muscle fiber-containing tissue sample; correcting said at least one mutation in said at least one gene encoding at least one muscle protein in said cultured primary muscle stem cells by targeted modification of said at least one mutation by gene editing using a CRISPR / Cas based tool; Including, The CRISPR / Cas-based gene editing tool is delivered to the cultured primary muscle stem cells by at least one DNA and / or RNA-based carrier, in particular a plasmid, a non-integrating viral vector, an mRNA, or a protein. method.

[0023] This targeted therapy method using CRISPR / Cas-based, particularly CRISPR / Cas9-based gene / base editing technology, uses primary human myoblasts with regenerative potential, which are isolated and harvested according to a previously described method (Patent Document 1). The primary human myoblasts harvested by this method are more than 98% pure and therefore available for gene editing.

[0024] Previous gene delivery studies performed on primary human myoblasts using a transposase system (Patent Document 1) also showed that isolated primary human muscle stem cells are generally amenable to genetic manipulation. However, transposase-mediated gene insertion is non-targeted and therefore not a preferred delivery tool for site-specific mutations.

[0025] In contrast, by using current CRISPR / Cas-based tools, specific mutations can be corrected, providing gene-edited primary human myoblasts with regenerative potential.

[0026] This was unexpected. In particular, in the past, it was generally believed that human muscle stem cells could not be transfected and / or could not be genetically engineered very efficiently. This would hinder approaches to using human muscle stem cells for therapy, because, unlike iPS cells, primary MuSCs cannot be expanded from one gene-corrected clone to produce a sufficient number of cells, for example for an autograft. This means that it is essential to obtain a high enough percentage of cells with the desired edit at the population level. For these reasons, gene editing in primary human MuSCs has not been attempted so far.

[0027] Marg et al. (Non-Patent Document 11) showed that only a few green muscle cells were inefficiently transfected with lipofectamine and a plasmid vector encoding GFP. These results were unexpected, as targeted gene correction could correct primary ex vivo expanded human MuSCs with efficiency of over 90% for several targeted loci / mutations.

[0028] Chemello et al. (Non-Patent Document 12) mention that muscle progenitor cells were modified by CRISPR genome editing and then engrafted into the muscle of mdx mice by intramuscular injection, as shown by Zhu et al. (Non-Patent Document 13). Here, Cas9 was delivered ex vivo to MuSCs from dystrophic mice using adenovirus. However, the editing efficiency of mouse MuSCs appears to be low (Figure 2E), and precise quantification was not shown. Furthermore, the approach includes low engraftment efficiency and failure to generate muscle stem cells with long-term reconstitution capacity.

[0029] In contrast, here we demonstrate for the first time highly efficient targeted genome editing in primary human MuSCs at the population level using plasmid-based transfection. Our method yields >90% edited MuSCs with a FACS-based enrichment step to select transfected cells and 40-60% efficient editing without this enrichment step. We show that primary human MuSCs can be gene edited with very high efficiency using various methods to deliver the gene editing machinery ex vivo (mRNA nucleofection). This method is much more advanced in the clinical translation sense and does not require the use of selection markers to enrich for gene-edited cells.

[0030] Transplanted autologous myoblasts, whose phenotypic defects have been therapeutically addressed through gene editing in a site-specific manner, can build new muscle tissue in which the pathological phenotype is restored in whole or in part.

[0031] Moreover, these cells can reconstitute recipient muscles with muscle stem cells that sustain muscle homeostasis and regeneration in the long term.It was not clear that primary human MuSCs that have undergone such targeted genetic manipulation ex vivo could retain their regenerative properties and reconstitute the stem cell compartment as well as build new muscle fibers in vivo after transplantation.

[0032] This method has several advantages: the cells can be applied in an autologous setting, where they are unlikely to trigger immune rejection or adverse immune responses; because the edited cells maintain the ability to produce new muscle fibers and reconstitute the muscle stem cell pool in vivo, a single administration is likely sufficient to obtain a long-term sustained therapeutic effect; skeletal muscle tissue has a very low tendency to develop tumors, as do muscle stem cells and myoblasts; the edited cells can be thoroughly checked for off-target events and biosafety profile before reimplantation; therefore, these cells can be transplanted in an autologous setting with a calculable small risk.

[0033] In one embodiment of the present invention, the monogenic muscle disease includes one of the following: muscular dystrophies (MD), including all types of limb-girdle muscular dystrophies (LGMD), especially LGMD1 / D, LGMD2 / R (Non-Patent Document 14), all X-linked muscular dystrophies (Emery-Dreifuss MD, Duchenne MD, Becker MD), and all MD caused by repeat expansion (i.e., myotonic dystrophy types 1 and 2) or repeat deletion (facioscapulohumeral muscular dystrophy) mutations. Also included are the very rare diseases Pax7 myopathy (Non-Patent Document 15) and VCP myopathy.

[0034] It should be understood that the at least one genetic mutation in at least one gene encoding a muscle protein can be a deletion, an insertion, a point mutation, a repeat expansion, or a repeat deletion, particularly a deletion or a point mutation. In a preferred embodiment, only monogenic diseases are considered to have one mutation in each allele. The mutations can be the same (homozygous) or different (compound heterozygous).

[0035] As mentioned above, precise and efficient gene repair in primary somatic stem and progenitor cells ex vivo is becoming increasingly feasible due to the rapid development of CRISPR / Cas-based tools for gene editing, some of which are independent of choice of cellular DNA repair pathways.

[0036] The CRISPR / Cas-based tools, in particular CRISPR / Cas9-based tools, may be used for at least one of the following gene editing approaches: base editing, in particular adenine base editing (ABE), cytidine base editing (CBE), C-to-G base editing (CGBE), glycosylase base editing (GBE), prime editing, non-homologous end joining (NHEJ), microhomology-mediated end joining (MMEJ), and / or homology-directed repair (HDR).

[0037] Base editors are RNA-programmable deaminases that allow for precise single-base changes in DNA or RNA. Because base editors do not create double-stranded DNA breaks, they minimize the formation of undesired editing by-products, including insertions, deletions, translocations, and other large-scale chromosomal rearrangements.

[0038] Adenine base editing (ABE) allows precise targeted conversion of adenine to guanine nucleotides without inducing DNA double-strand breaks (Non-Patent Document 16). ABE consists of a catalytically impaired Cas9 fused to an adenine deaminase enzyme (TadA) that converts adenine to inosine on a single-stranded DNA bubble generated by Cas9 binding to the target site. The inosine is then replaced by guanine. To be accessible to the deaminase, the target adenine must be located at a defined distance from the protospacer adjacent motif (PAM), the so-called ABE activity window. Due to its predictable outcome, high precision, and reduced off-target effects (20,21), ABE is a safe gene editing tool.

[0039] Cytosine base editors (CBEs or BEs) allow efficient and programmable reversion of TA to CG point mutations in the human genome. BEs consist of a cytidine deaminase fused to a disordered form of Cas9 (D10A nickase) tethered to one (BE3) or two (BE4) monomers of a uracil glycosylase inhibitor (UGI). This BE structure allows the conversion of CG base pairs in human genomic DNA to TA base pairs through the formation of a uracil intermediate.

[0040] More recently, a base editor (Pin-point™) has been developed in which the effector domain (deaminase) is not fused to Cas9. Instead, the effector is fused to a cognate ligand of an RNA aptamer contained in the gRNA, facilitating recruitment of the effector to the Cas9 / gRNA complex (Non-Patent Document 17).

[0041] Non-homologous end joining (NHEJ) repairs double-stranded breaks (DSBs) in DNA (e.g., caused by Cas9 cleavage) and often introduces small insertions and / or deletions (indels) at the repair site.

[0042] The homology directed repair (HDR) pathway uses a provided DNA template that has regions of homology to the target site, thus allowing the production of the desired alteration.

[0043] In more specific embodiments, the at least one mutation is located in at least one of the following genes: LMNA, which encodes lamin A / C; CAPN3, which encodes calpain 3; DYSF, which encodes dysferlin; SGCA, which encodes α-sarcoglycan; VCP, which encodes valosin-containing protein; PAX7, which encodes paired box 7; NCAM1, which encodes neural cell adhesion molecule 1; DMD, which encodes dystrophin.

[0044] α-Sarcoglycan is a 50 kDa transmembrane protein that is part of the sarcoglycan complex and the dystrophin-associated protein complex (DAPC). DAPC protects muscle fibers from mechanical stress, and its dysfunction leads to various forms of muscular dystrophy (MD).

[0045] Loss-of-function mutations in SGCA, which encodes α-sarcoglycan, cause limb-girdle muscular dystrophy 2D / 3R, an early-onset, severe, and rapidly progressive muscular dystrophy that affects girls and boys equally. Patients suffer from muscle degeneration and atrophy affecting the limbs, respiratory muscles, and heart.

[0046] SGCA has nine coding exons and is expressed in striated muscles (2,3). Disease-causing mutations are spread throughout the entire length of the gene, without defined mutation hotspots (4,5). However, some mutations, such as c.157G>A, have been reported more frequently (Non-Patent Document 18).

[0047] In one embodiment, SGCA mutations, particularly the c.157G>A mutation, are reversed or repaired by adenine base editing (ABE).

[0048] Classic laminopathies refer to diseases caused by mutations in the LMNA gene, which encodes the nuclear lamina proteins lamin A / C.

[0049] In one embodiment, the LMNA mutation, particularly the c.1366A>G mutation, is reversed or repaired by cytosine base editing (CBE or BE).

[0050] Mutations in the DYSF gene, which codes for dysferlin, cause weakness primarily in the hip and thigh muscles, calf pseudohypertrophy, and shoulder muscles. One of the mutations in DYSF is based on a deletion of G at position 4782 in the DYSF coding sequence, which induces a frameshift and a premature stop codon. The other G>C substitution is located four positions downstream.

[0051] In one embodiment, the DYSF mutation, particularly the G deletion, is reversed or repaired by CRISPR / Cas9-induced non-homologous end joining (NHEJ). Specifically, a +1A insertion restores the DYSF reading frame and results in the removal of the premature stop codon.

[0052] Calpain 3, the protein encoded by CAPN3, is a cysteine ​​protease that is predominantly expressed in skeletal muscle. Mutations in CAPN3 cause LGMD2A, a progressive skeletal muscle disorder with no cure and the most common type of LGMD worldwide. The deletion CAPN3 c.550delA causes a frameshift in exon 4, which creates a premature stop codon.

[0053] In one embodiment, CAPN3 mutation, especially A deletion, is reversed or repaired by CRISPR / Cas9-induced non-homologous end joining (NHEJ).Specifically, +1A insertion restores reading frame and results in removal of premature stop codon.+1 insertion in exon 4 of CAPN3 to reverse CAPN3 c.550delA is exempted if repair is plasmid-based.All other approaches, such as mRNA and protein-based repair approaches, are included.

[0054] As described above, the CRISPR / Cas-based gene editing tools are delivered to the human primary muscle stem cells by at least one DNA and / or RNA and / or protein-based carrier, in particular using a plasmid, a non-integrating viral vector, a recombinant protein, or an mRNA as a delivery system. A DNA or RNA delivery system carrying the components required for the CRISPR / Cas-based genetic modification or the information for the synthesis of a Cas-based protein is introduced into the primary human stem cells having a genetic defect.

[0055] Thus, the different modes of delivery may consist of one or more of the following: For Cas enzymes / base editors / prime editors: DNA (e.g., plasmid DNA, minicircle DNA), mRNA, protein, viral vectors; For single guide RNA: DNA (e.g., plasmid DNA, minicircle DNA), RNA, viral vectors; For homology template (in case of HDR): DNA (can be single stranded like ssODN or double stranded like minicircle plasmid), viral vector (e.g. AAV).

[0056] In one preferred embodiment for delivery, plasmid is used as a delivery system.In this case, plasmid can comprise DNA encoding Cas9 and promoter-driven sgRNA expression cassette.For example, ABE plasmid can be used, which comprises codon-optimized SpCas9 fused to deaminase domain, T2A-Venus cassette, and further human U6 promoter-driven sgRNA cassette.

[0057] Plasmids containing DNA encoding Cas-based proteins, particularly base editors, and sgRNAs may be transfected into primary human stem cells carrying a genetic defect. In one embodiment, transfection is performed at 40,000-90,000 cells / 9.5 cm. 2 Within the range of 55,000 to 75,000 cells / 9.5 cm 2 50,000 to 80,000 cells / 9.5cm 2 The cell densities required for primary human stem cells are surprisingly low compared to the cell densities required for transfection of induced pluripotent stem cells (iPSCs), for example, at approximately 300,000 cells / 9.5 cm. 2 A cell density of 100% is required.

[0058] Using plasmid-based ex vivo delivery, we achieved >90% correction of a G>A MD-causing mutation (SGCA c.157G>A) in the gene encoding α-sarcoglycan in primary human MuSCs. Corrected cells displayed full myogenic and regenerative capacity in xenografts, including reconstitution of the muscle stem cell compartment.

[0059] However, plasmid-based delivery carries the risk of transgene integration and results in relatively prolonged exposure to gene-modifying enzymes.

[0060] Therefore, in another preferred embodiment, mRNA is used as a delivery system.Transgene delivery via mRNA eliminates the risk of transgene integration, while resulting in transient gene expression.Transient expression of gene modification enzyme reduces the risk of potential off-target mutagenesis.This is an important aspect for the therapeutic development of gene editing strategy.

[0061] Nucleofection allows delivery of mRNA to nearly 100% of cells, resulting in highly uniform levels of transgene expression and eliminating the need for reporter genes or enrichment steps to obtain a homogenous population of gene-edited primary muscle stem cells.

[0062] In one embodiment, nearly 100% transfection of mRNA into MuSCs was achieved using nucleofection.

[0063] As mentioned above, in a further preferred embodiment, Cas-based recombination proteins are used as the delivery mode for CRISPR / Cas-based gene editing tools.

[0064] When delivering recombinant proteins, no synthesis in the cell is required, since the protein has already been produced in vitro (usually in bacteria) and introduced into the cell. Cas-based proteins (like ABE8e in the example shown below) are usually pre-incubated with sgRNA or crRNA:tracrRNA heteroduplexes to form so-called ribonucleoprotein (RNP) complexes. The RNPs are then delivered to the cell. However, both components can also be delivered separately to the cell and allowed to form a complex inside the cell. Nucleofection is applied in protein-mediated gene editing.

[0065] In one embodiment, a ribonucleoprotein (RNP) complex was constructed consisting of ABE8e recombinant protein and sgRNA designed to mutate the splice donor site of NCAM1 exon 7. ABE8e protein and each sgRNA were mixed to form an RNP complex.

[0066] The RNP complex is then transfected into human muscle stem cells using nucleofection.

[0067] As mentioned above, a method for providing a sample of an isolated muscle fiber-containing tissue sample taken from at least one patient with muscular dystrophy has been previously described (US Pat. No. 5,399,633).

[0068] Specifically, primary stem cells from the muscle fiber-containing tissue sample taken from a muscular dystrophy patient are cultured under hypothermic conditions with a defined temperature and a defined atmosphere in the absence of oxygen supply, with the temperature not exceeding 15°C and the atmosphere having an oxygen content not exceeding 21% (v / v), for a period of 4 days to 4 weeks.

[0069] Culturing under such conditions enriches the sample for stem cells, such that after the first period of time, approximately 70-100% of all viable cells in the sample are cultured stem cells or derivatives of cultured stem cells. The culturing is carried out using a medium suitable or adapted for the stem cells being cultured.

[0070] In one embodiment the temperature does not exceed 14° C., in particular does not exceed 13° C., in particular does not exceed 12° C., in particular does not exceed 11° C., in particular does not exceed 10° C., in particular does not exceed 9° C., in particular does not exceed 8° C., in particular does not exceed 7° C., in particular does not exceed 6° C., in particular does not exceed 5° C., in particular does not exceed 4° C., in particular does not exceed 3° C., in particular does not exceed 2° C., in particular does not exceed 1° C., in particular does not exceed 0° C. In one embodiment the temperature is in the range of 0° C. to 15° C., in particular 1° C. to 14° C., in particular 2° C. to 13° C., in particular 3° C. to 12° C., in particular 4° C. to 11° C., in particular 5° C. to 10° C., in particular 6° C. to 9° C., in particular 7° C. to 8° C.

[0071] In one embodiment, the atmosphere contains no more than 20% by volume, in particular no more than 19% by volume (v / v), in particular no more than 18% by volume (v / v), in particular no more than 17% by volume (v / v), in particular no more than 16% by volume (v / v), in particular no more than 15% by volume (v / v), in particular no more than 14% by volume (v / v), in particular no more than 13% by volume (v / v), in particular no more than 12% by volume (v / v), in particular no more than 11% by volume (v / v). In particular, the atmosphere has an oxygen content not exceeding 10% (v / v), in particular not exceeding 9% (v / v), in particular not exceeding 8% (v / v), in particular not exceeding 7% (v / v), in particular not exceeding 6% (v / v), in particular not exceeding 5% (v / v), in particular not exceeding 4% (v / v), in particular not exceeding 3% (v / v), in particular not exceeding 2% (v / v), in particular not exceeding 1% (v / v), in particular not exceeding any of the oxygen contents mentioned above. Atmospheres with an oxygen content of less than 20% (v / v) are also often referred to as hypoxic conditions.

[0072] In one embodiment, the atmosphere has an oxygen content in the range of 1% (v / v) to 21% (v / v), in particular 2% (v / v) to 20% (v / v), in particular 3% (v / v) to 19% (v / v), in particular 4% (v / v) to 18% (v / v), in particular 5% (v / v) to 17% (v / v), in particular 6% (v / v) to 16% (v / v), in particular 7% (v / v) to 15% (v / v), in particular 8% (v / v) to 14% (v / v), in particular 9% (v / v) to 13% (v / v), in particular 10% (v / v) to 12% (v / v), in particular 3% (v / v) to 11% (v / v).

[0073] In an alternative embodiment, the atmosphere has an oxygen content of not more than 30% by volume, in particular not more than 29% (v / v), in particular not more than 28% (v / v), in particular not more than 27% (v / v), in particular not more than 26% by volume (v / v), in particular not more than 25% by volume (v / v), in particular not more than 24% by volume (v / v), in particular not more than 23% by volume (v / v), in particular not more than 22% by volume (v / v), in particular not more than 21% by volume (v / v).

[0074] In one embodiment, the temperature is in the range of 0°C to 10°C and the oxygen content is in the range of 0% (v / v) to 8% (v / v). In one embodiment, the temperature is in the range of 2°C to 5°C and the oxygen content is in the range of 2% (v / v) to 5% (v / v). In one embodiment, the temperature is in the range of 3°C to 4°C and the oxygen content is in the range of 3% (v / v) to 4% (v / v). In one embodiment, the temperature does not exceed 10°C and the oxygen content does not exceed 8% (v / v).

[0075] In one embodiment, growth factors are added to the medium in which the stem cells are cultured. In one embodiment, growth factors are added only if the stem cells are cultured for more than 2 weeks, in particular for more than 2 weeks at 4°C. In particular, when using HMFF as sample and culturing satellite cells, a medium with a low serum content is suitable. A suitable medium is an optimized serum-reduced basal medium, such as OptiMEM available from Life Technologies. It has been found that the length of the first period is particularly suitable for culturing and enriching stem cells from 5 days to 2 weeks, in particular 6 days to 1 week, in particular 1 week.

[0076] In one embodiment, the sample is an isolated tissue sample. The sample can be isolated from a patient by standard methods, such as biopsy. In one embodiment, the sample is an isolated muscle fiber fragment. Human muscle fiber fragments (HMFFs) are particularly suitable and can be readily obtained from muscle biopsy specimens.

[0077] In one embodiment, stem cells are cultured in an integrated cell structure. This can be best achieved by a support structure that allows for an integrated cell structure. In one embodiment, this support structure is the natural structure that cells grow in the body (such as HMFF, which is very suitable for culturing satellite cells in the context of the present invention). In another embodiment, this support structure is an artificial structure that mimics or closely resembles the natural structure that cells grow in the body.

[0078] After culturing a sufficient amount of primary stem cells, the gene editing methods described above using CRISPR / Cas-based tools are applied to the primary stem cells to correct the mutations in muscle genes.

[0079] After gene repair, the human muscle cells are screened for the desired gene repair / editing event. A selection step (e.g., by FACS) may be performed. In particular, fluorescent molecules expressed by the transfected / edited cells may be used, or selection may be based on live cell staining of extracellular epitopes.

[0080] The genetically modified primary stem cells are further cultured.

[0081] In one embodiment, the modified primary stem cells are fused into multinucleated myotubes in vitro.

[0082] In a further aspect, the present invention also relates to genetically repaired or modified human muscle stem cells obtained after carrying out the above method.

[0083] In one embodiment, the genetically repaired human muscle stem cells comprise at least one gene encoding at least one muscle protein, which at least one gene has been subjected to targeted modification of at least one mutation in said gene, e.g., by NHEJ, base editing, or prime editing using CRISPR / Cas-based tools, as described in detail above.

[0084] As shown above, the gene-corrected muscle stem cells obtained by this method are different from healthy ("wild type") muscle stem cells and are not identical to wild type. This has been demonstrated in base-edited human fibroblasts with LMNA mutations (Non-Patent Document 5). It should be emphasized that "wild type" MuSCs from muscular dystrophy patients do not naturally exist.

[0085] In particular, site-specific insertion and deletion cause frame shift. Such reframing often results in a DNA sequence that is different from the wild type. Site-specific insertion may restore the reading frame, but the protein encoded by such a restored reading frame may be different from the wild type, as shown for the DYSF or CAPN3 genes (see further examples below).

[0086] In a preferred embodiment, the genetically repaired human muscle stem cells contain at least one modified gene encoding at least one of the following muscle proteins: LMNA encoding lamin A / C, CAPN3 encoding calpain 3, DYSF encoding dysferlin, SGCA encoding α-sarcoglycan, VCP encoding valosin-containing protein, PAX7 encoding paired box 7, NCAM1 encoding neural cell adhesion molecule 1, DMD encoding dystrophin.

[0087] In a further embodiment, the present invention also relates to the medical use of gene-repaired or modified muscle stem cells. Thereby, the stem cells are intended to be used in cell replacement therapy for muscular dystrophies (MD), including all types of limb-girdle muscular dystrophies (LGMD), in particular LGMD1 / D, LGMD2 / R (Non-Patent Document 14), and all X-linked muscular dystrophies, in particular Emery-Dreifuss MD, Duchenne MD, Becker MD, and all MD caused by repeat expansion (i.e., myotonic dystrophy types 1 and 2) or repeat deletion (facioscapulohumeral muscular dystrophy) mutations. The stem cells can also be used to treat Pax7 myopathy (Non-Patent Document 15) or VCP myopathy.

[0088] Thereby, a method of transplanting cultured stem cells into a subject or patient in need thereof is disclosed, which comprises administering to the patient genetically repaired or modified stem cells cultured as outlined above, particularly by autologous transplantation.

[0089] Thereby, a method of treating a patient suffering from muscular dystrophy is disclosed, the method comprising administering genetically modified muscle stem cells to the patient. In one embodiment, the muscle stem cells are satellite cells. In one embodiment, the genetically modified stem cells are administered as a cell suspension. They can also be delivered in the form of a tissue support system.

[0090] The invention will now be explained in more detail with reference to exemplary embodiments and figures. [Brief description of the drawings]

[0091] [Figure 1A] ABE repairs the SGCA c.157G>A mutation in primary muscle stem cells of patients and carriers without detectable off-target editing. [Figure 1B] ABE repairs the SGCA c.157G>A mutation in primary muscle stem cells of patients and carriers without detectable off-target editing. [Figure 1C] ABE repairs the SGCA c.157G>A mutation in primary muscle stem cells of patients and carriers without detectable off-target editing. [Figure 1D] ABE repairs the SGCA c.157G>A mutation in primary muscle stem cells of patients and carriers without detectable off-target editing. [Figure 1E] ABE repairs the SGCA c.157G>A mutation in primary muscle stem cells of patients and carriers without detectable off-target editing. [Diagram 2] CBE repairs the LMNA c.1366A>G mutation in patient-derived cells. [Diagram 3] DYSF editing in primary patient MuSCs / myoblasts. [Figure 4A] mRNA-mediated delivery of ABE7.10 disrupts nearly 100% of the NCAM1 splice donor in human primary MuSCs. [Figure 4B]mRNA-mediated delivery of ABE7.10 disrupts nearly 100% of the NCAM1 splice donor in human primary MuSCs. [Figure 4C] mRNA-mediated delivery of ABE7.10 disrupts nearly 100% of the NCAM1 splice donor in human primary MuSCs. [Figure 4D] mRNA-mediated delivery of ABE7.10 disrupts nearly 100% of the NCAM1 splice donor in human primary MuSCs. [Figure 5A] mRNA-mediated delivery of ABE7.10 repairs the SGCA c.157G>A mutation in primary human MuSCs. [Figure 5B] mRNA-mediated delivery of ABE7.10 repairs the SGCA c.157G>A mutation in primary human MuSCs. [Figure 5C] mRNA-mediated delivery of ABE7.10 repairs the SGCA c.157G>A mutation in primary human MuSCs. [Figure 6A] Optimization of GFP mRNA nucleofection in human primary MuSCs: transfection efficiency and cell viability with different nucleofection programs. [Figure 6B] Optimization of GFP mRNA nucleofection in human primary MuSCs: transfection efficiency and cell viability with different nucleofection programs. [Figure 7] Comparison of the efficiency of ABE7.10 mRNA-mediated repair of the SGCA c.157G>A mutation in human primary MuSCs under two different nucleofection programs. [Figure 8A] ABE8e protein-mediated disruption of the NCAM1 splice donor in human primary MuSCs. [Figure 8B] ABE8e protein-mediated disruption of the NCAM1 splice donor in human primary MuSCs. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0092] Example 1: Gene editing of cells derived from patients with SGCA-associated muscular dystrophy

[0093] a) Isolation of primary MuSCs from patients and carriers with the compound heterozygous SGCA c.157G>A mutation.

[0094] Primary MuSCs were isolated and characterized from muscle biopsy specimens obtained from a 10-year-old male LGMD2D patient with a compound heterozygous SGCA c.157G>A mutation and a related carrier.

[0095] Primary MuSC isolation and culture. Immediately after biopsy, muscle specimens were transferred to transport solution A (30 mM HEPES, 130 mM NaCl, 3 mM KCl, 10 mM D-glucose, and 3.2 μM phenol red, pH 7.6). Fresh muscle specimens were manually dissected, and fragments were cryo-treated at 4-6 °C for 2-7 days prior to downstream processing for MuSC isolation. Oligoclonal MuSC colonies were obtained after mechanical dissection as described (Non-Patent Document 19). Growing colonies were expanded to passage 4 and characterized before cryopreservation. To increase the probability of usable MuSCs in difficult-to-handle biopsy specimens, classical purification was performed in parallel (Non-Patent Document 3). All cell populations used in this study were >95% Desmin positive. To induce myoblast-myotube fusion, the medium was switched to Opti-MEM I Reduced Serum Media (Thermo Fisher Scientific) once cells reached confluence.

[0096] Primary MuSC cultures from patients and carriers were 95–100% Desmin+ and expressed the myogenic markers Pax7, MyoD, Myf5 and the proliferation marker Ki-67. The c.157G>A mutation affects the last coding nucleotide of exon 2.

[0097] b) ABE corrects SGCA c.157G>A by >90% in primary human MuSCs without detectable off-target editing.

[0098] We found that c.157G>A was an ideal ABE target because it is located 15 bp upstream of the -NGG PAM (corresponding to protospacer position 6 and therefore in the center of the ABE activity window). No other adenines are located within the ABE activity window, making unwanted bystander editing unlikely. We first assessed whether ABE could be used to repair the c.157G>A mutation in patient iPSCs.

[0099] MuSCs from patients were transfected with various amounts of a plasmid encoding ABE7.10_4.1 (Figure 1A), an ABE7.10-based vector (16) that contains a codon-optimized Cas9(D10A) nickase N-terminally fused to a TadA heterodimer, followed by a T2A-Venus cassette under the control of a CAG promoter, and a U6 promoter-driven sgRNA expression cassette.

[0100] This was enriched for Venus-positive cell detection (FIG. 1A) via FACS, and ABE efficiency was assessed using EditR (Non-Patent Document 20).

[0101] Transfection and sorting of human primary MuSCs. Human primary MuSCs were cultured in Skeletal Muscle Cell Growth Medium (SMCGM, Provitro) at 55,000 cells / 9.5 cm 1 day prior to transfection. 2Cells were plated at a density of 1000 x 1000 and transfected using Lipofectamine® 3000 (Thermo Fisher Scientific) according to the manufacturer's instructions. SMCGM was replaced after 1 day. Two days after transfection, cells were collected for FACS sorting in PBS containing 50% SMCGM, 0.05 mM EDTA, and 100 μg / ml Primocin™. Venus-positive cells were sorted using a FACSAria Fusion cell sorter (BD Biosciences) and cultured in SMCGM. 100 μg / ml Primocin™ was added to the medium for 2 days.

[0102] As analyzed by EditR, the c.157G nucleotide rate was >99% in patient and carrier MuSCs at all vector concentrations (Figure 1B). Subsequent analysis by Crispresso2 (26) for amplicon sequencing confirmed the high c.157G nucleotide rate of >90% in patient MuSCs and >85% in carrier MuSCs (Figure 1C). A bystander A>G edit at protospacer position 10 was detected in a very low percentage (0.2–2%) of reads. In two samples, 1.1% and 0.3% of reads contained indels. Omission of gRNA did not result in A>G edits or indels (Figure 1C).

[0103] Detection bias was ruled out, as both alleles were equally represented in the amplicon sequencing data ( Fig. 1D ).

[0104] No Cas9-dependent off-target editing events were detected at the top predicted (by CRISPOR) off-target loci containing A nucleotides in the ABE activity windows with the lowest or highest vector concentrations ( Figure 1E ).

[0105] We found that ABE7.10_4.1, when combined with an appropriate gRNA (gRNA#1), induced an efficient c.157A>G conversion with only minimal (0.2–2%) bystander A>G editing detected by amplicon sequencing and Crispresso2 analysis.

[0106] Therefore, it was concluded that the SGCA c.157G>A mutation could be repaired with high efficiency and specificity via ABE in human primary MuSCs. The repaired SGCA c.157G>A is hereafter referred to as SGCA c.157Grep.

[0107] c) SGCA c.157Grep primary MuSCs show normal α-sarcoglycan mRNA and protein expression.

[0108] To assess the functional outcome of ABE, we analyzed α-sarcoglycan mRNA and protein expression in SGCA c.157Grep myotubes. We found that the splicing defect was restored, as shown by the fact that α-sarcoglycan transcripts containing exon 2 in SGCA c.157Grep were increased compared to unedited patient and carrier myotubes and reached levels similar to those of control 3 (het. c.748-2A>G carrier) in patient myotubes. Furthermore, the overall level of SGCA mRNA was increased in patient myotubes after ABE, presumably because the co-skipping of exons 2+3 (but not exon 2 alone) induces a frameshift leading to a premature stop codon, which presumably leads to nonsense-mediated mRNA decay (NMD). Western blot and immunostaining analyses revealed that α-sarcoglycan protein was restored in SGCA c.157Grep patient cells.

[0109] d) MuSCs from SGCA c.157Grep patients are viable, proliferative, and myogenic.

[0110] Primary cells are particularly susceptible to stress induced by extensive manipulations. Primary MuSCs derived from MD patients with mutations in genes governing membrane integrity are particularly vulnerable. Compared to untransfected patient MuSCs, reduced cell proliferation was observed during the first few days after transfection and sorting. However, Venus-positive cells (>48% of the source cell population) expanded extensively after sorting and were further expanded for at least 2–3 passages before cryopreservation. SGCA c.157Grep primary MuSCs could readily fuse into multinucleated myotubes in vitro. Furthermore, the pattern of α-sarcoglycan localization was indistinguishable from control myotubes.

[0111] e) SGCA c.157Grep primary MuSCs regenerate muscle and reconstitute the satellite cell niche in vivo.

[0112] We transplanted primary SGCA c.157Grep MuSCs into irradiated tibialis anterior muscles of compromised NSG mice. We found that SGCA c.157Grep patient MuSCs gave rise to abundant human myofibers. Furthermore, the satellite cell niche between the sarcolemma and the basement membrane contained numerous Pax7+ cells of human origin. Taken together, SGCA c.157Grep patient MuSCs are capable of both myofiber regeneration and reconstitution of the satellite cell compartment in vivo.

[0113] Human MuSC transplantation SGCA c.157Grep patient MuSCs that were 99% Desmin+, 27% Pax7+, 25% Ki-67+, 66% MyoD+, and 40% Myf5+ were used for transplantation. Six-week-old male NOD.Cg-Prkdc scid Il2rg tm1WjI / SzJ(NSG) mice were purchased from Charles River Laboratories 1 week prior to the experiment. Animal housing and hygiene monitoring followed the recommendations of FELASA. Local irradiation of the recipient hind limbs was performed 2 days prior to cell transplantation as described (17,18). 2.5 × 10 4Two injections of 5.5 μl containing 10 cells were made along a parallel trajectory into the medial part of the TA muscle (a total of 5 × 10 cells per graft). 4 Mice were sacrificed 19 days after cell transplantation. TA muscles were cryopreserved in liquid nitrogen-chilled isopentane, mounted in gum tragacanth, and stored at −80°C.

[0114] Example 2: Gene editing in cells from patients with LMNA-associated muscular dystrophy

[0115] Classical laminopathies refer to diseases caused by mutations in the gene LMNA, which encodes the nuclear lamina protein lamin A / C. State-of-the-art gene editing tools offer the possibility to correct mutations at the genome level, in particular powerful base editors that correct single base mutations without DNA double-strand breaks.

[0116] An 8-year-old girl was diagnosed with muscular dystrophy carrying the dominant LMNA c.1366A>G mutation. Base editing was first performed using patient-derived induced pluripotent stem cells (iPSCs), an unlimited cell source for testing editing efficiency.

[0117] Initial testing was performed by co-transfecting two vectors into iPSCs in mTeSR Plus stem cell medium using the transfection reagent Lipofectamine® 3000. One DNA vector carries CBE4max, an enhanced version of the first reported cytidine base editor, and SpRY, a novel near-PAMless Cas9 that can edit mutations previously uneditable by classical Cas9, while the other vector expresses an sgRNA.

[0118] Initial editing results showed a G to A conversion, although the editing efficiency was low due to low transfection efficiency (Figure 2).

[0119] Based on the preliminary results of the ABE editing project, transfection with synthetic Cas9 mRNA instead of DNA vectors showed higher editing efficiency. Therefore, the editing efficiency can be improved by transfecting synthetic mRNA containing CBE4max and SpRY sequences together with synthetic sgRNA.

[0120] Following optimized CBE editing in iPSCs, patient-derived muscle stem cells could be edited using a similar protocol, and ultimately, the corrected muscle stem cells could be used in transplantation therapy to improve muscle function in patients.

[0121] Example 3: DYSF gene editing in primary patient MuSCs / myoblasts

[0122] a) Human primary MuSC culture

[0123] Human MuSCs were grown in skeletal muscle cell growth medium (SMCGM) (Provitro) supplemented with fetal calf serum (FCS) supplement mix (Provitro) and 2.72 mM glutamine (GlutaMAX™, Thermo Fisher Scientific) on 10 cm plastic dishes (Corning) in a humidified atmosphere containing 5% CO2 at 37°C. For cell passaging, human MuSCs were washed with Dulbecco's phosphate buffered saline (DPBS) (Thermo Fisher Scientific) and treated with 0.25% Trypsin / EDTA (Thermo Fisher Scientific) for 5 min at 37°C.

[0124] Detached cells were collected in SMCGM+supplement, diluted 1:10, and centrifuged at 200g for 5 min at room temperature (RT). The pellet was resuspended in an appropriate volume of SMCGM+supplement and plated at 1–2*10 cells / well in a 10 cm plate. 4 cells / cm 2 Cells were passaged every 2-3 days depending on growth rate / confluence.

[0125] b) CRISPR / Cas9-based gene editing

[0126] [Table 1]

[0127] c) Lipotransfection of Cas9 / sgRNA complexes

[0128] For CRISPR / Cas9 experiments, human MuSCs were seeded at a density of 75,000 cells / well per well of a 6-well plate one day before transfection. 1 μg of SpCas9::Venus plasmid DNA (with or without sgRNA) was transfected using Lipofectamine® 3000 transfection reagent (Invitrogen, Germany) according to the manufacturer's instructions. 48 hours after transfection, Venus-positive cells were sorted using a FACSAria Fusion (BD). Sorted cells were re-plated and expanded for genomic DNA isolation and dysferlin protein analysis via flow cytometry or immunofluorescence staining, respectively.

[0129] For dysferlin immunostaining, human MuSCs were seeded in SMCGM+supplement in 8-well ibidi μ-Slides (IBIDI GmbH Martinsried, Cat. No. 80826) and grown until they reached 70% / 80% confluence. Myoblast fusion was induced by switching the medium to OptiMEM (Thermo Fisher Scientific). After 4 days, cells were fixed and stained with an antibody against the N-terminal portion of dysferlin (ab124684, Abcam).

[0130] Figure 3 shows the results obtained with DYSF gene editing. Genomic DNA was isolated from a control (top panel) and a patient (middle panel) carrying the homozygous DYSF c.4782_4786delinsCCC mutation in exon 44. The c.4782_4786delinsCCC mutation is a deletion of a G nucleotide at position c.4782 in the DYSF coding sequence (causing a frameshift) and a G>C substitution four downstream positions.

[0131] The +1A insertion restores the DYSF reading frame, resulting in the removal of a premature stop codon while differing from the wild-type protein sequence by four amino acids (shown in blue) (lower panel). Dysferlin protein expression is restored in patient MuSCs and derived myotubes after restoring the reading frame with the +1A insertion. Dysferlin localization in +1A-reframed patient myotubes is similar to that in control myotubes.

[0132] Example 4: CAPN3 gene editing in patient primary myoblasts

[0133] Calpain 3, the protein encoded by CAPN3, is a cysteine ​​protease predominantly expressed in skeletal muscle. Mutations in CAPN3 cause limb-girdle MD2A (LGMD2A), a progressive skeletal muscle disorder with no cure and the most common type of LGMD worldwide.

[0134] We isolated and expanded human primary muscle stem cells from 35 patients with 37 different CAPN3 mutations. Twenty percent of patients harbored the well-known founder mutation CAPN3 c.550delA, which causes a frameshift in exon 4, resulting in the formation of a premature stop codon. In most cases, patients harbored compound heterozygous CAPN3 c.550delA mutations, but two patients with homozygous c.550delA mutations were also included in the cohort.

[0135] Primary MuSCs from homozygous patients were isolated, expanded, and transfected with a plasmid carrying the mutation-specific sgRNA and SpCas9::Venus. After cell sorting and expansion, detailed sequence analysis of the CAPN3 c.550 DNA region showed insertions and deletions of bases (indels) at the targeted CAPN3 locus with up to 60% efficiency. One of the sgRNAs preferentially inserted +1 bp at the position of the mutation, indicating indel signature bias of the specific sgRNA and reframing of the open reading frame (ORF).

[0136] Effects on protein levels were analyzed using a custom-made anti-calpain 3 monoclonal antibody suitable for immunostaining.

[0137] Example 5: mRNA-mediated delivery of Cas9 gene editing tools into human primary muscle stem cells

[0138] a) mRNA nucleofection results in near 100% transfection efficiency of primary human MuSCs with minimal toxicity.

[0139] Primary MuSCs were harvested using TrypLE Express, spun down at 200g for 5 min, and washed once with DPBS. After a second centrifugation, the supernatant was removed and cells were seeded at 7.5x10 in P5 Primary Cell Nucleofector Solution (Lonza, Basel, Switzerland) premixed with mRNA (SpCas9mRNA, Aldevron, ND, USA; ABE7.10mRNA, AmpTec GmbH, Hamburg, Germany; GFP mRNA, Aldevron) or sgRNA (IDT / Synthego, CA, USA). 6The cells were resuspended at a concentration of 1000 cells / ml. For 3 μg of gene editing molecule encoding mRNA, 2 μg of 5' / 3' end modified sgRNA (1:0.67 ratio) was added to a 20 μl reaction. The reaction can be linearly scaled up. The cells were electroporated with an Amaxa 4D Nucleofector (Lonza, Basel, Switzerland) using an X Unit with a 16-well nucleofection cuvette. After the cells were nucleofected, 80 μl of pre-warmed SMCGM was added and the cells were transferred to one well of a 6-well plate containing 2 ml of pre-warmed SMCGM. The cell medium was changed the next day.

[0140] If edited cells are intended to be used for transplantation therapy, cell fitness and viability are crucial. To determine the parameters that result in high transfection efficiency and minimal cytotoxicity as measured in GFP+ cells, we compared different nucleofection programs. Nucleofection was optimized by gradually decreasing the intensity of the pulse (i.e., nucleofection program). Pulse codes EY-100, EX-100, EP-100, EO-100, EH-100, EE-100, DU-00, DI-100, DH-100, DA-100, CY-100, CX-100, and CO-100 resulted in transfection efficiencies of 95% or higher (Figure 6A). Cell viability after nucleofection gradually increased from the strongest pulse code EY-100 to the weaker pulse codes CY-100 and then CX-100 (Figure 6B). To confirm that the editing efficiency remained unchanged after optimizing the pulse code, we compared the editing results using strong pulse (EY-100) and weak pulse (CY-100) parameters (Figure 7). No significant differences were observed.

[0141] b) mRNA-mediated delivery of SpCas9 results in highly efficient gene editing in MuSCs from multiple donors.

[0142] To develop and systematically evaluate a pipeline for mRNA-mediated delivery of gene editing tools into primary human MuSCs, a universal readout system relevant for MuSCs from all donors was established. For this purpose, a strategy was designed to target the gene encoding neural cell adhesion molecule 1 (NCAM1), a membrane protein expressed by all human MuSCs and myoblasts that has an extracellular domain that is easy to detect in living cells. MuSCs from six donors of different ages and genders were delivered with mRNA encoding S. pyogenes Cas9 (SpCas9). Transfection of various concentrations of SpCas9 mRNA and a fixed ratio of sgRNA targeting NCAM1 exon 3 resulted in efficient indel formation, with the highest editing rate observed with 2 μg of SpCas9 mRNA. Gene editing resulted in a similar percentage of NCAM1 protein knockout assessed by fluorescent immunostaining and flow cytometry. An increase in the percentage of edited MuSCs was observed in all donors between days 2 and 8 post-nucleofection, reaching a maximum indel rate of >90% at day 8. Consistently, NCAM1-positive cells decreased between days 4 and 6 post-nucleofection and remained constant thereafter.

[0143] c) mRNA-mediated delivery of ABE7.10 enables highly efficient, selection-free base editing of human MuSCs.

[0144] We investigated mRNA-based delivery of the base editor ABE7.10. We designed an sgRNA to mutate the splice donor site in NCAM1 exon 7 (Figure 4A). Amplicon sequencing confirmed that mRNA delivery of ABE7.10 and the corresponding sgRNA resulted in >90% A to G conversion of the targeted adenine located in the center of the ABE activity window, at protospacer position 5 (A5) (Figure 4B-4D). The adjacent adenine at protospacer position 8 (A8) was co-edited in up to 22% of sequencing reads. In contrast, plasmid-based experiments using the same splice donor targeting strategy had lower average editing efficiencies and lower A5 vs. A8 to G conversion rates (not shown). The described nucleotide changes at A5 and A8 did not result in clear NCAM1 protein knockout.

[0145] d) Human MuSCs retain their myogenic and proliferative properties after mRNA-mediated gene and base editing.

[0146] To determine whether mRNA-mediated delivery of gene editing tools or knockout of NCAM1 alters the myogenic or proliferative properties of human MuSCs, we analyzed the expression profiles of myogenic and proliferation markers in passage-matched unedited and edited cells from the same donor. The purity of the MuSC populations was consistently >95%, as determined by counterstaining with the fibroblast marker TE7 and myogenic marker Desmin (DES). Myogenic transcription factors PAX7, MYF5, and proliferation marker KI-67 were similarly expressed in edited and unedited MuSCs. KI-67-positive proliferating MuSCs varied between cell populations, ranging from 30% to 60% pre- and post-editing (not shown). We next assessed the in vitro differentiation potential of edited MuSCs in a myoblast fusion assay. All edited and unedited MuSC populations gave rise to multinucleated myotubes with a typical striated pattern (not shown). Fusion indices were consistent between donor- and passage-matched untransfected cells and cells edited with SpCas9 or ABE7.10 mRNA and the respective sgRNAs (not shown).

[0147] e) mRNA-based ABE delivery efficiently corrects the SGCA c.157G>A muscular dystrophy-causing mutation in human MuSCs.

[0148] Human MuSCs carrying a heterozygous SGCA c.157G>A mutation were transfected with ABE7.10 mRNA and the corresponding sgRNA (Figure 5A). mRNA-mediated base editing efficiently repaired the mutation, resulting in a c.157G nucleotide rate of over 80%. Bystander editing of the adenine located at protospacer position 10 was found in a very low percentage of reads, less than 0.4%, and no indels specific to the mRNA-transfected or edited samples were detected (Figure 5B, 5C).

[0149] Example 6. Nucleofection of ABE8e recombinant protein and sgRNA results in disruption of the NCAM1 exon 7 splice donor site in human primary MuSCs.

[0150] First, we constructed a ribonucleoprotein (RNP) complex consisting of ABE8e recombinant protein and sgRNA designed to mutate the splice donor site of NCAM1 exon 7 (ABE8e recombinant protein was produced at the Max-Delbruck Center Protein Core Facility, and sgRNA was purchased from Integrated DNA Technologies, IDT). A target adenine that fits the consensus NCAM1 exon 7 splice donor site is located at position 5 of the protospacer. An additional adenine is present at position 8 of the protospacer, within the ABE editing window (Figure 8-target adenine: A5; bystander editing: A8). ABE8e protein and the respective sgRNA were gently mixed in a molar ratio of 1:1.2 (ABE8e:sgRNA) in sterile PBS, and the mixture was incubated at room temperature for 20 min to 1 h to form the RNP complex. The RNP was then stored at 4 °C until nucleofection. The final concentration of ABE8e protein in the RNPs was 2 μg / μl.

[0151] Cultured primary MuSCs were harvested using TrypLE Express and centrifuged at 200g for 5 min. They were resuspended in PBS and counted. The appropriate number of cells was centrifuged again at 200g. The supernatant was aspirated and the cell pellet was resuspended in P5 Primary Cell Nucleofector Solution containing RNP (Lonza, Basel, Switzerland). 150,000 cells were nucleofected in a 20 μl reaction using an Amaxa 4D Nucleofector, specifically an X Unit with a 16-well nucleofection cuvette (Lonza, Basel, Switzerland) using pulse code CY-100 or EY-100. After nucleofection, cells were harvested by adding 80 μl of pre-warmed Skeletal Muscle Cell Growth Medium + Supplement (SMCGM, Provitro) to each cuvette and transferred to one well (if MuSCs were nucleofected with pulse code CY-100) or two wells (if MuSCs were nucleofected with pulse code EY-100) of a 6-well plate containing 2 ml of pre-warmed SMCGM. The medium was changed every other day, and cells were allowed to grow and then harvested for genomic DNA isolation 7 days after nucleofection. Base editing was analyzed by PCR amplification of the targeted locus followed by Sanger sequencing. The % peak area corresponding to A or G was calculated using the online tool EditR (Non-Patent Document 21). Edited MuSCs retained myogenic and proliferative properties as determined by microscopic determination of the percentage of cells staining positive for the myogenic markers Desmin, PAX7, MYF5, MYOD, and the proliferation marker KI-67 and negative for the fibroblast marker TE7 (not shown).

Claims

1. 1. An ex vivo method for gene repair in primary human muscle stem cells (satellite cells), comprising: providing an isolated muscle fiber-containing tissue sample from at least one patient having a monogenic muscle disease, said monogenic muscle disease being caused by at least one mutation in at least one gene encoding at least one muscle protein; isolating and culturing primary stem cells from the muscle fiber-containing tissue sample; correcting said at least one mutation in at least one gene encoding said at least one muscle protein in said cultured primary stem cells by targeted modification of said at least one mutation by gene editing using a CRISPR / Cas based tool; Including, The CRISPR / Cas-based gene editing tool is delivered to the cultured primary muscle stem cells by at least one DNA and / or RNA and / or protein-based carrier, in particular a plasmid, a non-integrating viral vector, an mRNA, or a protein. method.

2. 10. The method of claim 1 , The monogenic muscle disease comprises one of the following: muscular dystrophies, including all types of limb-girdle muscular dystrophies (LGMD), particularly LGMD1 / D, LGMD2 / R, all X-linked muscular dystrophies (Emery-Dreifuss MD, Duchenne MD, Becker MD), all MD caused by repeat expansion (i.e. myotonic dystrophy types 1 and 2) or repeat deletion (facioscapulohumeral muscular dystrophy) mutations, Pax7 myopathy, or VCP myopathy; A method comprising:

3. The method according to any one of claims 1 to 2, The at least one genetic mutation may be a deletion, an insertion, a point mutation, a repeat expansion, or a repeat deletion, in particular a deletion or a point mutation. A method comprising:

4. The method according to any one of claims 1 to 3, The at least one mutation is located in at least one of the following genes: LMNA, which encodes lamin A / C; CAPN3, which encodes calpain 3; DYSF, which encodes dysferlin; SGCA, which encodes alpha sarcoglycan; VCP, which encodes valosin-containing protein; PAX7, which encodes paired box 7; NCAM1, which encodes neural cell adhesion molecule 1; or DMD, which encodes dystrophin. A method comprising:

5. The method according to any one of claims 1 to 4, The gene editing using the CRISPR / Cas-based tool includes at least one of the following: adenine base editing (ABE), cytidine base editing (CBE / BE), C-to-G base editing (CGBE), glycosylase base editing (GBE), prime editing, non-homologous end joining (NHEJ), microhomology-mediated end joining (MMEJ), and / or homology-directed repair (HDR); A method comprising:

6. The method according to any one of claims 1 to 5, The CRISPR / Cas-based gene editing tools are delivered to the cultured primary muscle stem cells using a plasmid as a delivery system. A method comprising:

7. 7. The method of claim 6, Transfection of plasmids as delivery systems for the CRISPR / Cas-based gene editing tools was performed at 40,000-90,000 cells / 9.5 cm 2 in the range of 55,000 to 75,000 cells / 9.5 cm 250,000 to 80,000 cells / 9.5 cm 2 The results were performed using human primary muscle cells at cell densities ranging from A method comprising:

8. The method according to any one of claims 1 to 5, The CRISPR / Cas-based gene editing tool is delivered to the cultured primary muscle stem cells using mRNA as a delivery system. A method comprising:

9. 9. The method of claim 8, Transfection of mRNA as a delivery system for the CRISPR / Cas-based gene editing tool was performed using electroporation (nucleofection). A method comprising:

10. The method according to any one of claims 1 to 5, The CRISPR / Cas-based gene editing tools are delivered to the cultured primary muscle stem cells using recombinant proteins as a delivery system. A method comprising:

11. 11. The method of claim 10, Transfection of recombinant proteins as a delivery system for the CRISPR / Cas-based gene editing tools was performed using electroporation (nucleofection). A method comprising:

12. The method according to any one of claims 1 to 11, The primary stem cells from the muscle fiber-containing tissue sample are cultured under hypothermic conditions having a defined temperature and a defined atmosphere by treatment without oxygen supply; said temperature not exceeding 15° C. and said atmosphere having an oxygen content not exceeding 21% (v / v); The first period is from 4 days to 4 weeks. A method comprising:

13. 13. The method of claim 12, The temperature does not exceed 10° C. and the oxygen content does not exceed 10% (v / v). A method comprising:

14. The method according to any one of claims 1 to 13, The genetically modified primary stem cells are then further cultured. A method comprising:

15. Genetically corrected human muscle stem cells obtained after carrying out the method according to any one of claims 1 to 14.

16. 1. A genetically corrected human muscle stem cell comprising: at least one gene encoding at least one muscle protein; The at least one gene has been modified to target at least one mutation in the gene. A genetically corrected human muscle stem cell.

17. 17. The gene-corrected human muscle stem cell of claim 16, The at least one modified gene encodes at least one of the following muscle proteins: LMNA encoding lamin A / C, CAPN3 encoding calpain 3, DYSF encoding dysferlin, SGCA encoding alpha sarcoglycan, VCP encoding valosin-containing protein, PAX7 encoding paired box 7, NCAM1 encoding neural cell adhesion molecule 1, or DMD encoding dystrophin. A genetically corrected human muscle stem cell.

18. Genetically corrected human muscle stem cells for use in cell replacement therapy for muscular dystrophies, particularly all types of limb-girdle muscular dystrophies (LGMD), in particular LGMD1 / D, LGMD2 / R, all X-linked muscular dystrophies (Emery-Dreifuss MD, Duchenne MD, Becker MD), all MD caused by repeat expansion (i.e. myotonic dystrophy types 1 and 2) or repeat deletion (facioscapulohumeral muscular dystrophy) mutations, and Pax7 or VCP myopathies.

Images