This protocol describes the conversion of skin fibroblasts into myoblasts and their differentiation into myotubes. The cell lines are derived from patients with neuromuscular disorders and can be used to investigate pathological mechanisms and to test therapeutic strategies.
Investigations into both the pathophysiology and therapeutic targets in muscular dystrophies have been hampered by the limited proliferative capacity of human myoblasts. Several mouse models have been created but they either do not truly represent the human physiopathology of the disease or are not representative of the broad spectrum of mutations found in humans. The immortalization of human primary myoblasts is an alternative to this limitation; however, it is still dependent on muscle biopsies, which are invasive and not easily available. In contrast, skin biopsies are easier to obtain and less invasive to patients. Fibroblasts derived from skin biopsies can be immortalized and transdifferentiated into myoblasts, providing a source of cells with excellent myogenic potential. Here, we describe a fast and direct reprogramming method of fibroblast into a myogenic lineage. Fibroblasts are transduced with two lentiviruses: hTERT to immortalize the primary culture and a tet-inducible MYOD, which upon the addition of doxycycline, induces the conversion of fibroblasts into myoblasts and then mature myotubes, which express late differentiation markers. This quick transdifferentiation protocol represents a powerful tool to investigate pathological mechanisms and to investigate innovative gene-based or pharmacological biotherapies for neuromuscular disorders.
Cellular models obtained directly from human tissues are useful to model many human genetic disorders, with the advantage of having the original genomic context and, in many cases, reproducing the same molecular and cellular hallmarks observed in the patients. In the field of neuromuscular disorders, muscle biopsies have been a great source of human myoblasts and have helped in the elucidation of pathological mechanisms. Additionally, they are an important tool for in vivo testing of drugs and gene therapies. On one hand, the derivation of myoblasts from muscle fragments is relatively easy. On the other hand, the culture and maintenance of primary myoblasts are challenging, because of their limited proliferation rate and replicative senescence in vitro1. An alternative for these limitations is to immortalize myoblasts with the insertion of the human telomerase (hTERT) and/or cyclin-dependent kinase 4 (CDK4) genes2,3, with preservation of skeletal muscle features4. Nevertheless, the obtention of primary myoblasts is still dependent on muscle biopsy, a surgical procedure with disadvantages to the patients, which, in many cases, have their muscles in advanced degeneration. Thus, the muscle of these patients is composed of a significant proportion of fibrotic and/or adipose tissue and yields fewer muscle cells, requiring the purification of the cells previously to the immortalization.
In contrast to muscle biopsies, skin biopsies are more accessible and are less harmful to patients. Primary fibroblasts can be derived from skin fragments in vitro. Although fibroblasts are not primarily affected by mutations causing neuromuscular disorders, they can be transdifferentiated into myoblasts. This can be achieved by the insertion of the Myod gene, a myogenic regulatory transcription factor5. In this manuscript, we describe the protocol to obtain transdifferentiated myoblasts, from the establishment of fibroblasts cultures to the obtention of differentiated myotubes (a representative summary of the method is depicted in Figure 1).
Pre-clinical testing of therapeutic strategies is dependent on cellular and animal models carrying mutations similar to the mutations found in human patients. Although the development of animal models has become more feasible with the advance of gene-editing technologies such as CRISPR/Cas96, it is still challenging and costly. Thus, patient-derived cell lines are an accessible option to have models, covering the large spectrum of mutations of disease such as Duchenne muscular dystrophy (DMD). Obtention and creation of cell models are crucial to the development of personalized therapies for such pathologies.
Among personalized therapies that have been investigated, exon skipping strategies is one of the promising ones for different muscular dystrophies7,8. This strategy consists of producing a shorter but functional protein. This is performed by hiding the exon definition to the spliceosome, therefore excluding the mutated exon from the final messenger. This is a very promising technology that has been approved by the FDA for DMD. Thus, we also describe in this protocol, methods to transfect myoblasts with two different exon skipping related technologies: antisense oligonucleotides (AON) and U7snRNA-adeno-associated virus (AAV). AON transfection is a good tool for the initial screening of several sequences designed to promote exon skipping9. However, the activity of AONs is transient. To obtain a sustained expression of antisense sequences, we also explored small nuclear RNAs (snRNAs) combined with AAV, allowing nuclear localization and inclusion in the splicing machinery10. U7 is an snRNA involved in the processing of histone mRNA that can be engineered to bind proteins that will redirect it to the spliceosome and deliver antisense sequences11. The use of modified U7 snRNAs in combination with AAV vectors overcomes limitations of AONs resulting in a continued expression of the AONs and better transduction of tissues of interest12. We use cells derived from DMD patients for this protocol to illustrate the exon-skipping strategy.
All experiments and biopsies were carried out following the ethical rules of the institutions involved under the approval of the Nationwide Children's Hospital Institutional Review Board.
1. Initiation of dermal fibroblasts culture
2. Establishment of FibroMyoblasts (FM) cell line
3. Transdifferentiation protocol
4. Immunostaining of differentiated myotubes
NOTE: For immunostaining, grow the cells in glass coverslips or chamber slides as noted above.
5. Antisense oligonucleotide transfection
NOTE: The protocol below is for transfection of a 6-well plate. Adjust volumes accordingly for smaller or bigger plates. The transfection is done in 100% confluent myoblasts when the cells are ready for the differentiation step.
6. AAV1-U7 transduction
NOTE: This protocol was optimized for 6-well plates. Adjust the volumes proportionally to the culture surface area. The transduction is done in 100% confluent myoblasts when the cells are ready for the differentiation step. AAV1 is the AAV serotype with the best transduction capacity of cultured myoblasts.
7. RNA extraction
NOTE: All material used during this step should be RNase free.
8. RT-PCR analysis
NOTE: In this step, we present a suggestion of reagents to detect the expression of dystrophin mRNA, but it can be easily adapted to other reagents of choice.
9. Detection of dystrophin expression by Western Blotting
NOTE: This protocol is optimized for dystrophin, a large membrane protein. Specific conditions may be needed for different proteins.
This protocol shows how to establish human skin-derived fibroblast cultures and convert them into myoblasts and then into differentiated myotubes. This type of cell line is extremely useful for the study of neuromuscular disorders and in vitro testing of potential therapies.
A schematic representation of the fibroblast conversion is shown in Figure 1. Figure 2A shows a fragment of skin and the fibroblasts emerging from it. The fibroblasts should be passed to a new dish when confluence is reached (Figure 2B). Figure 3A shows the ideal confluence of fibroblasts before changing to myoblast growth medium supplemented with doxycycline. The cells should be around 70% confluent because they still proliferate during the conversion process. If cells are above 80% confluent, the differentiation may be compromised. The conversion into myoblasts takes two to four days, and it is confirmed by observation of the morphology. The cells become elongated and parallelly oriented, as shown in Figure 3B. After the addition of the differentiation medium, the myoblasts stop dividing and start to fuse to form multinucleated myotubes (Figure 3C). When the myotubes borders look white and bright, they are about to detach (Figure 4). At this point, collect or fix the cells.
The differentiation success will vary between different cell lines/mutations. Immunostaining of muscle proteins expressed by mature myotubes confirms the myogenic potential of converted fibroblasts (Figure 5). RNA-Seq analysis comparing FM myotubes and skeletal muscle showed high-level expression of transcripts from the embryonic (MYH3) and neonatal (MYH8) myosin chain genes and good overall transcriptome-wide correlation with muscle (Figure 6). Transcripts for the giant sarcomeric proteins titin (TTN), nebulin (NEB), and obscurin (OBSCN) are also expressed by FM myotubes, indicating upregulation of these large transcripts involved in myofibrillogenesis. Thus, FM cells have a muscle-specific expression profile, demonstrating that they are a useful and reliable surrogate for muscle-derived cell lines.
To illustrate exon skipping, we used this protocol in one of the most frequent exon duplications in the DMD gene. Duplication of exon 2 leads to disruption of the DMD reading frame, thus the restoration of the reading frame following exon skipping should lead to the expression of the full-length dystrophin. However, it is also possible that skipping of exon 2 is very efficient resulting in an out-of-frame transcript. Nevertheless, in this case, skipping of both copies of exon 2 induces the utilization of an alternative internal ribosome entry site (IRES) present in exon 5, thereby producing functional N-truncated dystrophin that was identified in patients still ambulant in their 70s12. Figure 7A shows representative results of RT-PCR of FM cells with exon 2 duplication. FM cells were treated either with AON or AAV1-U7 carrying an antisense sequence to skip exon 2. In Figure 7B, an immunoblot shows the detection of the N-truncated dystrophin in FM cells treated with AAV1-U7. In vitro treatment of FM cells serves as proof of concept for exon-skipping strategies.
Figure 1: Schematic representation of fibroblasts conversion into myogenic cells. A skin biopsy is obtained from human subjects. Skin fragments are placed on culture dishes. Within one week, fibroblasts start to emerge. Fibroblasts are first transduced with the hTERT gene, and then with the Myod gene, using lentiviral vectors. After antibiotic selection of infected cells, the conversion into myoblasts is induced by the addition of doxycycline to the myoblast growth medium. Within two to four days, the cells become elongated and parallelly oriented. After switching to differentiation medium, the myoblast fuse with each other and form multinucleated myotubes. Please click here to view a larger version of this figure.
Figure 2: Skin biopsy fragments in culture. (A) First fibroblasts emerging from skin fragment. (B) Confluent fibroblasts emerged from the skin fragment. Scale bar: 50 µm. Please click here to view a larger version of this figure.
Figure 3: Fibroblasts transdifferentiation. (A) Representative image of 70% confluent fibroblasts. (B) Converted myoblasts have elongated morphology and are parallelly organized. (C) Myotubes were differentiated for 7 days. Scale bar: 50 µm. Please click here to view a larger version of this figure.
Figure 4: Representative image of detaching myotubes. The arrows indicate the white and bright edges of myotubes starting to detach. Scale bar: 50 µm. Please click here to view a larger version of this figure.
Figure 5: Immunofluorescence of differentiated myotubes. Immunostaining of myosin heavy chain in myotubes derived from a healthy (A) individual and patients with neuromuscular disorders (B and C). In B are shown cells from myotonic dystrophy type 1 (DM1) carrying 230 CTG repeats, and in C are DM1 cells with 900 CTG repeats. Scale bar: 100 µm. Please click here to view a larger version of this figure.
Figure 6: Transcriptome pattern of FM myotubes compared to skeletal muscle. Transcriptome pattern of FM myotubes compared to skeletal muscle. The read counts per million mapped reads for 12,134 transcripts are shown for Illumina RNA-Seq libraries prepared from FM myotubes and a human skeletal muscle biopsy. Transcript levels between the two libraries had a Pearson correlation of 0.71 and a Spearman rank correlation of 0.73. Transcripts for the developmental myosin heavy chains and the large sarcomeric proteins are highlighted in red. Please click here to view a larger version of this figure.
Figure 7: Representative RT-PCR and Western blot showing DMD exon skipping in FM cells. (A) Expression DMD by RT-PCR. Fibroblasts from a patient harboring a duplication of DMD exon 2 were converted into FM cells. RNA extracted from muscle biopsy was used as the control, showing that FM untreated cells express the same duplicated transcript. FM cells treated with AON have a partial skipping of exon 2 duplication, while AAV1-U7 treated cells showed a predominance of transcripts with exon 2 duplication skipped. (B) Representative immunoblot of FM cells treated with AAV1-U7. Smaller N-truncated dystrophin was detected 14 days after treatment (indicated by the arrow). Data previously published in Wein et al. Translation from a DMD exon 5 IRES results in a functional dystrophin isoform that attenuates dystrophinopathy in humans and mice. Nature Medicine. 2014. 2020 Springer Nature Limited. Please click here to view a larger version of this figure.
Fibroblast growth medium | DMEM with 20% FBS, 1% antibiotic-antimicotic |
Freezing medium | 10% DMSO, 90% fibroblast medium |
Doxycycline stock solution 1000X | 8 mg of doxycycline in 1 mL ultra-pure water. Filter in 0.22 µm syringe filter. Aliquot in PCR tubes. Store at -20 °C, protected from light. |
Myoblast medium | Skeletal muscle cell growth medium (see list above) with supplements, 8 µg/mL doxycycline. For example: 100 µL of 1000X stock solution in 100 mL. |
Differentiation medium | Skeletal muscle cell differentiation medium with supplements (see list above), 8 µg/mL doxycycline. For example: 100 µL of 1000X stock solution in 100 mL. |
Blocking solution for IF staining | 10% goat serum (or serum of animal in which secondary antibody was raised) in 1X PBS |
Base buffer for protein extraction | NaCl 150 mM, Tris 50 mM, 0.05 % NP-40. Adjust pH to 7.4. Store at 4 °C. |
Table 1: Medium recipes
To obtain FM cell lines with good quality, some steps are critical. The sooner the skin biopsy is processed, the greater the chances are to obtain healthy fibroblasts. The passage number of fibroblasts cultures is also important. Primary cells have limited proliferative capacity and after many passages, they enter in replicative senescence. Thus, it is better to have a stock of fibroblasts at a low passage number and transform cells at the earliest passage as possible.
Another important step is also to have viral production that has maximum purity and accurate quantification. For example, viral genome quantification using qPCR provides reasonable measurements, but quantification by ddPCR (digital droplet PCR) is more accurate.
In addition, the adequate confluence of fibroblasts for myoblast conversion is critical. If the cells are below 70% or above 80% confluent, the myogenic differentiation may be impaired. If cells are too confluent, there will be the superposition of layers of myotubes, which interfere with staining and imaging. The concentration of doxycycline is crucial for correct activation and sustained expression of the Myod gene. It is very critical to always add the doxycycline to the medium right before doing media changes, as it degrades quickly after diluted in medium and stored at 4 °C. The stock should be stored at -20 °C at a concentration of 1000X and protected from light. Do not re-freeze thawed aliquots. It is very important to follow these details to ensure reproducible experiments and discriminate an impaired differentiation due to a genetic mutation from technical issues. Nevertheless, depending on the mutation or the type of disease, a good differentiation may not be possible. To ensure trustful results, it is very important to replicate experiments at similar passage numbers.
In our experience, the differentiation capacity persists at least up to passage 25-27, especially in wild-type controls. The same may be valid for some diseased cell lines, but it depends on the cell line. Some DMD cell lines still retain the myogenic potential above P20. In opposition, a myotonic dystrophy type 1 (DM1) cell line presented reduced myogenicity after P8. However, in the case of DM1, this is not surprising as it has been demonstrated that mutations in DM1 indirectly play a role in muscle differentiation16. The retention of the myogenic differentiation capacity should be addressed individually, but generally, most of the cell lines retain it up to P20-25.
In summary, the conversion of fibroblasts into myoblasts is a powerful tool to study and test therapeutic strategies for neuromuscular disorders. It facilitates access to human cell models by avoiding the complicated obtention of muscle biopsies and reduce the inconvenience of a muscle biopsy for the patients.
The authors have nothing to disclose.
We would like to thank Dr. Vincent Mouly for sharing his knowledge in the past regarding the model. This work has been supported by the US National Institutes of Health National Institute of Neurological Disorders and Stroke (R01 NS043264 (K.M.F., and R.B.W.)), the US National Institutes of Health National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS) (P50 AR070604-01 (K.M.F., K.M., R.N., and N.W.). N.W. has received fellowship support from the Ohio State University/Nationwide Children's Hospital Muscle Group and the Philippe Foundation. This work was also supported by internal discretionary funds and part of the exon 2 skipping work has been supported also by CureDuchenne (K.M.F.) and Association Francaise Contre Les Myopathies. IRB number: IRB #: IRB10-00358/ CR00005138 and IBCSC#: IBS00000123.
100 mm dish | Corning | 430167 | |
0.25% Trypsin-EDTA, phenol red | Thermo Fisher | 2500056 | |
10X Phosphate buffered saline (PBS) | Fisher Scientific | BP3994 | |
12-well plate | Corning | 3513 | |
20X Transfer buffer | Thermo Fisher | NP00061 | |
20X Tris-acetate SDS running buffer | Thermo Fisher | LA0041 | |
3-8% Tris-Acetate gel | Thermo Fisher | EA0378BOX | |
75 cm2 flask | Corning | 430641U | |
Antibiotic-Antimicotic 100X | Thermo Fisher | 15240062 | |
Anti-myosin heavy chain, sarcomere antibody | Developmental Studies Hybridome Bank | MF20 supernatant | Dilution 1:50 |
Antioxidant | Thermo Fisher | NP0005 | |
BCA Protein Assay | Thermo Fisher | 23227 | |
Chloroform | Sigma-Aldrich | C2432 | |
DAPI | Thermo Fisher | D3571 | Dilution 1:1000 |
Digitonin | Millipore Sigma | 300410250MG | |
Dimethyl sulfoxide | Sigma-Aldrich | D2438 | |
DMEM, High glucose, GlutaMAX supplement, Pyruvate | Thermo Fisher | 10569044 | |
DNAse I set (250U) | Zymo Research Corporation | E1010 | |
Doxycycline Hydrochloride | Fisher Scientific | BP2653-5 | |
Dup2 human primers | Fw_5' GCTGCTGAAGTTTGTTGG TTTCTC 3' |
Rv_5' CTTTTGGCAGTTTTTGCC CTGT 3' |
|
Dystrophin antibody | Abcam | ab15277 | Dilution 1:200 |
Fetal bovine serum | Thermo Fisher | 16000 | |
Glycine | Sigma-Aldrich | G8898 | |
Goat anti-mouse, Alexa Fluor 488 | Thermo Fisher | A11001 | Dilution 1:1000 |
Halt Protease inhibitor cocktail 100X | Thermo Fisher | 78430 | |
Hemocytometer | Hausser Scientific | 3100 | |
Hygromycin B | Thermo Fisher | 10687010 | |
IRDye 680RD goat anti-Rabbit IgG (H+L) | Li-Cor | 926-68071 | Dilution 1:5000 |
Lab-Tek II CC2 chamber slide system | Thermo Fisher | 15852 | |
Laemmli | Bioworld | 105700201 | |
Lipofectamine 3000 Transfection Reagent | Thermo Fisher | L3000008 | |
Matrigel GFR membrane matrix | Corning | 354230 | |
Methanol | Fisher Scientific | A412P-4 | |
Mr. Frosty Freezing Container | Thermo Fisher | 51000001 | |
Nitrocellulose membrane 0.45 µm | GE Healthcare Life Sciences | 10600002 | |
Normal Goat serum control | Thermo Fisher | 10000C | |
Odyssey Blocking Buffer (PBS) | Li-Cor | 927-40003 | Blocking buffer for Western blot |
Opti-MEM I Reduced Serum Medium | Thermo Fisher | 11058021 | |
Paraformaldehyde | Sigma-Aldrich | 158127 | |
PCR master mix | Thermo Fisher | K0172 | |
Phosphatase inhibitor | Thermo Fisher | A32957 | |
Precision Plus Protein Dual Color Standards | Bio Rad | 1610374 | |
Puromycin | Thermo Fisher | A1113803 | |
Revert 700 Total Protein Stain for Western Blot Normalization | Li-Cor | 926-11021 | |
RevertAid kit | Thermo Fisher | K1691 | |
RNA Clean & Concentrator-25 | Zymo Research Corporation | R1018 | |
Scalpels | Aspen Surgical | 372611 | |
Skeletal Muscle Cell Differentiation medium | Promocell | C23061 | |
Skeletal Muscle Cell Growth medium | Promocell | C23060 | |
Triton X-100 | Acros Organics | 215682500 | |
TRIzol reagent | Thermo Fisher | 15596026 | |
Tween 20 | Fisher Scientific | BP337500 | |
Ultra low temperature freezer | Thermo Scientific | 7402 | |
UltraPure 0.5M EDTA, pH 8.0 | Thermo Fisher | 15575020 | |
Vectashield antifade mounting medium | Vector Labs | H1000 |