The protocol describes how to generate knock-out myoblasts using CRISPR/Cas9, starting from the design of guide-RNAs to the cellular cloning and characterization of the knock-out clones.
One important application of clustered regulatory interspaced short palindromic repeats (CRISPR)/Cas 9 is the development of knock-out cell lines, specifically to study the function of new genes/proteins associated with a disease, identified during the genetic diagnosis. For the development of such cell lines, two major issues have to be untangled: insertion of the CRISPR tools (the Cas9 and the guide RNA) with high efficiency into the chosen cells, and restriction of the Cas9 activity to the specific deletion of the chosen gene. The protocol described here is dedicated to the insertion of the CRISPR tools in difficult to transfect cells, such as muscle cells. This protocol is based on the use of lentiviruses, produced with plasmids publicly available, for which all the cloning steps are described to target a gene of interest. The control of Cas9 activity has been performed using an adaptation of a previously described system called KamiCas9, in which the transduction of the cells with a lentivirus encoding a guide RNA targeting the Cas9 allows the progressive abolition of Cas9 expression. This protocol has been applied to the development of a RYR1-knock out human muscle cell line, which has been further characterized at the protein and functional level, to confirm the knockout of this important calcium channel involved in muscle intracellular calcium release and in excitation-contraction coupling. The procedure described here can easily be applied to other genes in muscle cells or in other difficult to transfect cells and produce valuable tools to study these genes in human cells.
With the progress of gene sequencing and the identification of mutations in genes of unknown functions in a specific tissue, the development of relevant cellular models to understand the function of a new target gene and confirm its involvement in the related pathophysiological mechanisms constitutes an essential tool. In addition, these models are of major importance for future therapeutic developments1,2, and constitute an interesting alternative to the development of knock-out animal models in straight line with the international recommendations for reduction in the use of animals in experimentation. Gene editing using CRISPR/Cas9 is among the most powerful tools currently available, which has allowed the development of many knock-out/knock-in models, and targeted gene validation using CRISPR/Cas9 is among the most widely used application of CRISPR/Cas93. The success of gene editing relies on the ability to introduce the CRISPR tools (the guide RNAs and the nuclease Cas9) in the target cell model, which can be a challenge in many difficult-to-transfect cells, such as muscle cells4. This challenge can be overcome with the use of virus, usually lentivirus, which has the great advantage to transduce efficiently many cell types and deliver its transgene. But its major drawback is the integration of the transgene in the host cell genome, leading to potential alteration of genes localized at the integration site and to the permanent expression of the transgene, which in the case of the nuclease Cas9 would result in damaging consequences5. A smart solution has been proposed by Merienne and colleagues6, which consists of introduction into the cells of a guide-RNA targeting the Cas9 gene itself, leading to Cas9 inactivation. An adaptation of this strategy is presented here as a user friendly and versatile protocol allowing to knock-out virtually any gene in difficult-to-transfect cells.
The goal of the protocol presented here is to induce the inactivation of a gene of interest in immortalized muscle cells. It can be used to knock-out any gene of interest, in different types of immortalized cells. The protocol described here contains steps to design the guide RNAs and their cloning into lentiviral plasmids, to produce the CRISPR tools in lentiviral vectors, to transduce the cells with the different lentiviruses, and to clone the cells to produce a homogenous edited cell line.
Using this protocol, immortalized human skeletal muscle cells have been developed with deletion of the type 1 ryanodine receptor (RyR1), an essential calcium channel involved in intracellular calcium release and muscle contraction7. The knock-out (KO) of the gene has been confirmed at the protein level using Western blot, and at the functional level using calcium imaging.
Muscle biopsies were obtained from the Bank of Tissues for Research (Myobank, a partner in the EU network EuroBioBank, Paris, France) in accordance with European recommendations and French legislation. Written informed consent was obtained from all individuals. Immortalized myoblasts were kindly produced by Dr. V. Mouly (Myology Institute, Paris, France), and the protocols were approved by Myology Institute ethic committee (MESRI, n AC-2019-3502).
1. CRISPR guide design
2. Plasmid cloning
NOTE: In this step, the gRNAs will be inserted in the plasmid backbone for lentivirus production. A cassette encoding the two gRNAs is first produced by successive polymerase chain reactions (PCR), using the overlapping primers. The new cassette is then inserted into the lentiviral backbone plasmid #87919.
3. Lentivirus production
4. Lentivirus titration
NOTE: The virus titration is performed on HEK293 cells. The titration is important to incorporate in the subsequent steps a precise number of lentiviruses per cell (whatever the batch of lentivirus), for the cells of interest. The number of viral particles that efficiently transduce a cell is called the multiplicity of infection (MOI): MOI 10 thus correspond to the introduction of 10 viral particles per cell. As the freezing/thawing cycle affects the viability of the lentivirus, the titration is performed with a frozen lentivirus aliquot, and each subsequent experiment will be performed with a new aliquot of the same pool. One titration method is described here, but other methods can be used.
5. Myoblast transduction
NOTE: The immortalized myoblasts are successively transduced with the three lentiviruses previously produced. They are maintained at a density below 50% in a proliferation medium composed of Ham's F10 supplemented with 20% FBS, 2% penicillin/streptomycin, 2% Ultroser G, and cultured at 37 °C, 5% CO2.
6. Cellular cloning
NOTE: As myoblast transduction is difficult and never reaches 100% efficiency, even when using lentivirus, cellular cloning is required in order to get a fully corrected cell line. This is only possible with immortalized cells, or cells that can be cultured and amplified during few weeks/months.
7. Clone selection
NOTE: This step is performed to identify which of the growing clones have been appropriately modified.
8. Characterization of edited clones
NOTE: Once a few clones have been picked and confirmed by DNA sequencing, the deletion of the targeted gene can be confirmed at the protein level using Western blot, and at the functional level if a functional cellular assay is available for this gene. In the case of RYR1-KO, as RyR1 is a calcium channel, the functional characterization has been performed using calcium imaging on cultured cells.
This protocol was applied to immortalized myoblasts from a healthy subject15 (so-called HM cells, for human myoblasts), in which the RyR1 has been previously characterized16, in order to knock out the RYR1 gene encoding the RyR1 protein. The design of the guides RNA was made to delete the sequence encompassing part of exon 101 and intron 101 of the gene. Deletion of part of exon 101 is foreseen to result in disruption of the reading frame. In addition, exon 101 encodes the pore of the protein and is thus required to produce a functional calcium channel. Many patients affected with a severe RyR1-related myopathy have a mutation in this region of the gene7. The best guides predicted with Crispor software were selected in order to have as few off-targets as possible, while keeping the best efficiency. We chose to use two guides at the same time in the same viral vector, to make sure that a major part of the cells will be knock-out, and to ease the detection of the deletion in edited clones using PCR. The two selected guides (see Table 2 and Figure 1) are localized, respectively, in the end of exon 101 and in the beginning of intron 101, resulting in a deletion of 326 bp and a subsequent frameshift. This ensures that the edited cells will be RYR1-KO.
The gRNA targeting the SpCas9 was the one already designed by Merienne and colleagues6 and present in the plasmid #87919 under the weak promoter 7SK. As its efficiency to target the SpCas9 sequence has been already demonstrated in their study6, it has been used to create the plasmid p_Killer, but under the control of the strong promoter H1. The second guide in the p_Killer plasmid is designed to target the mCherry sequence under the promoter U6. This will allow the deletion of mCherry, which could be disturbing in some subsequent experiments with the newly created cell line.
Plasmids p_guides and p_killer were created and all the required lentiviruses were produced in the dedicated BSL2 culture room. After myoblasts transduction, the cells were cultured for 2 weeks until growth recovery and analyzed using PCR to confirm the presence of edited cells. For cellular cloning, two 96-well plates were seeded with treated cells at 1 cell per well. Cell growth was observed in 32% of cultured wells. Genomic DNA was then extracted from the clones and PCR analysis was done until two corrected clones were identified (so-called Cl-KOR-1 and Cl-KOR-2), see Figure 3. The deletion of the targeted portion of RYR1 was confirmed by Sanger sequencing. Additional control cells have been used, such as the initial HM cells, or HM cells treated with the very same procedure but non-edited as confirmed by PCR and sequencing (Cl-CTRL). Selected clones were then amplified and further characterized at the protein level and functional level.
To validate the complete extinction of RyR1 and of Cas9 at the protein level, Western blot analysis was performed. The results are presented on Figure 4. The Cas9 protein, absent in the HM cells, was detected 5 days after the transduction of the cells with LV-Cas9 (HM+LV-Cas9), and as expected, its expression was abolished in the Cl-CTRL and in the edited Cl-KOR-1 and Cl-KOR-2, although a small amount of Cas9 was still detected in Cl-KOR-2 (Figure 3A). As the gRNA targeting Cas9 is still expressed, the amount of remaining Cas9 is expected to progressively disappear. The expression of other important proteins was also assessed using Western blot: RyR1, to confirm the full deletion of the protein, the alpha1 subunit of DHPR, which is the functional partner of RyR1 involved in excitation-contraction coupling, and the myosin heavy chain as a marker of differentiation, because RyR1 and DHPR are expressed only in differentiated myotubes (Figure 3B). DHPR and MYHC are not supposed to be modified. Although the Western blot confirmed the complete deletion of RyR1 in the two RyR1-KO clones (Cl-KOR-1 and Cl-KOR-2), additional modifications-resulting from Cas9 or from the cloning procedure-may have occurred in the Cl-KOR-2, which presented in addition an absence of DHPR and a reduction in myosin heavy chain (MHC). As both proteins are expressed only in differentiated myotubes, their absence most probably reflects alteration in differentiation, which could possibly stem from the long single cell cloning procedure associated with a local overgrowth of the cells leading to the lost in their myogenic potential. Thus, only Cl-KOR-1 has been further characterized at the functional level. This demonstrates that selection of numerous edited clones is important to be able to work with the best one(s).
The functional characterization was performed using calcium imaging. The mean fluorescence variation as a function of time is represented on Figure 5, on at least 180 myotubes of each genotype, using either a direct RyR1 stimulation by 4-CMC or stimulation of the calcium release complex (DHPR associated with RyR1) by KCl membrane depolarization.
These experiments showed that the non-modified Cl-CTRL had a behavior comparable to the initial cell population (HM cells) and responded to direct RyR1 stimulation by its activator 4-CmC and by stimulation of the calcium release complex by membrane depolarization. In contrast, the edited Cl-KOR-1 was unable to respond to either stimulation (4-CmC and KCl depolarization), as expected for a RyR1-KO.
Altogether, the Western blot and the functional characterization confirmed that the Cl-KOR-1 cell line is a RyR1-KO muscle cell line. This protocol can be applied to any other gene expressed in skeletal muscle, or other cell type (such as induced Pluripotent Stem Cells – iPSC). The complete protocol is illustrated in Figure 6.
Figure 1: Localization and design of the guides RNA. (A) Schematic localization of the guides designed for the creation of the RYR1-KO cell line. Guide 1 targets the end of exon 101 in the RYR1 gene, Guide 2 targets the intron 101, creating a deletion of 326 bp and a frameshift. Guide Killer targets the initiation codon in the sequence of the SpCas9, Guide mCherry targets the end of the mCherry sequence. (B) Localization of the guides in the genomic sequences for the previously described Guide 1, Guide 2, Guide killer, and Guide mCherry. The sequence of each guide (20 bp in length) is presented in color and the PAM for each guide is bolded and underlined. Please click here to view a larger version of this figure.
Figure 2: Cloning procedure for production of the cassette. Schematic presentation of the different PCR performed to produce the cassettes with the guides, to be inserted into the lentivirus backbone plasmid. PCR A, PCR B, PCR C, and PCR D are realized with program 1 illustrated on the right top, and PCR final with the PCR program 2, illustrated on the right bottom. Please click here to view a larger version of this figure.
Figure 3: Selection of the clones using PCR. (A) Schematic presentation of the PCR amplification of a 1,200 bp region encompassing the guides, with forward and reverse primers represented by the gray arrows. DNA cleavage at both guides is expected to result in reduction of 326 bp in the size of the PCR fragment, as depicted by the dotted green vertical lines below the scissors. (B) PCR products produced from HM cells, Cl-CTRL, Cl-KOR-1, and Cl-KOR-2 were loaded on a 1% agarose gel. The DNA of HM and Cl-CTRL appears non modified (full length), whereas the DNA of Cl-KOR-1 and Cl-KOR-2 is shorter than the control, as expected (cleaved). Please click here to view a larger version of this figure.
Figure 4: Characterization of the cells at protein level. (A) Western blot analysis for the presence of Cas 9 protein using the V5-tag in HM cells, HM cells transduced with LV-Cas9, Cl-CTRL, Cl-KOR-1, and Cl-KOR-2. Cas9 is clearly detected in the HM cells transduced with LV-Cas9, whereas its expression has been abolished completely (in Cl-CTRL and Cl-KOR-1) or almost completely (in Cl-KOR-2) by LV-Killer. (B) Expression of RyR1, Myosin heavy chain (MYHC), alpha1 subunit of DHPR and GAPDH as a loading control in Cl-CTRL, Cl-KOR-1, and Cl-KOR-2. Please click here to view a larger version of this figure.
Figure 5: Functional characterization of the cells. Fluo-4 calcium imaging performed on myotubes produced from HM cells, Cl-CTRL, and Cl-KOR-1. The curves represent the fluorescence variation in HM myotubes (black curves), Cl-CTRL myotubes (gray curves), or Cl-KOR-1 (green curves). All values are presented as mean ± standard error of mean (SEM) of n myotubes. In each condition, n = 180 to 256 myotubes have been analyzed, from at least three different experiments (exact number indicated for each curve). 4 CmC (500 µM) was used to directly stimulate RyR1 in the presence of 2 mM external calcium. KCl depolarization (140 mM) was induced in the presence of 2 mM external calcium. Please click here to view a larger version of this figure.
Figure 6: Schematic presentation of the whole protocol. The different steps of the protocol are presented, from the in silico design to the in vitro molecular cloning and the final cell selection. Please click here to view a larger version of this figure.
Name | Sequence 5'-3' | Purpose/Use | ||
Primer_Guide1F | Guide1 + gttttagagctagaaatagc | Plasmid cloning | ||
Primer_Guide1R | Guide 1-RC +AGATCTGTGGTCTCATACAG | |||
Primer_Guide 2F | Guide 2 + gttttagagctagaaatagc | |||
Primer_Guide 2R | Guide 2-RC + GTTTCGTCCTTTCCACAAG | |||
Primer_XmaI F | TAGTGGATCCCCCGGGAAAATTCAAAATTTT | Plasmids cloning and colonies control | ||
Primer_BlpI R | CGCTTCCATTGCTCAGCTGGCCGCTGCCCC | |||
Primer_KillerF | AATGGAGTACTTCTTGTCCAgttttagagctagaaatagc | Cloning plasmid p_Killer | ||
Primer_KillerR | TGGACAAGAAGTACTCCATTAGATCTGTGGTCTCATACAG | |||
Primer_mCherryF | GAACAGTACGAACGCGCCGAgttttagagctagaaatagc | |||
Primer_mCherryR | TCGGCGCGTTCGTACTGTTCGTTTCGTCCTTTCCACAAG | |||
Primer_RYR1exonF | GCTCGTATTCGTCACCCGCGgttttagagctagaaatagc | Cloning plasmid p_guides_RyR1 | ||
Primer_RYR1exonR | CGCGGGTGACGAATACGAGCAGATCTGTGGTCTCATACAG | |||
Primer_RYR1intronF | TAAGTCAGTTCATAGGCGCTgttttagagctagaaatagc | |||
Primer_RYR1intronR | AGCGCCTATGAACTGACTTAGTTTCGTCCTTTCCACAAG | |||
Primer_BeforeRYR1cut | AGTCGTTACCATGTCTTCAGCCCT | Clone selection for RYR1 KO | ||
Primer_AfterRYR1cut | CTGCCGATTCCACAGATGAAGCAC |
Table 1: List of primers. Primers used for plasmid cloning and colony control, as well as the ones for the clone selection.
Name | Guide sequence | PAM | Reverse Complement without PAM | ||
Guide 1: RYR1exon | GCTCGTATTCGTCACCCGCG | GGG | CGCGGGTGACGAATACGAGC | ||
Guide 2: RYR1intron | TAAGTCAGTTCATAGGCGCT | CGG | AGCGCCTATGAACTGACTTA | ||
Guide Killer Cas9 | TGGACAAGAAGTACTCCATT | GGG | AATGGAGTACTTCTTGTCCA | ||
Guide mCherry killer | GAACAGTACGAACGCGCCGA | GGG | TCGGCGCGTTCGTACTGTTC |
Table 2: Guides used to create the plasmid p_guides and the p_Killer. The sequences used for the development of the RyR1-KO clones are presented in this table.
A major step on the way to the characterization of genes of unknown function involved in pathologies is the development of relevant cellular models to study the function of these genes. The use of gene editing using CRISPR/Cas9 is an exponentially growing field of research, and the development of knock-out models as presented here is among its most widely used applications. In this context, we propose here a versatile protocol to develop a human cell line knock-out in any gene of interest, allowing the characterization of this gene in a relevant human cell model. The protocol presented here can be used in immortalized myoblasts as well as in iPSC, and thus could virtually result in the production of any human cell model KO in the gene of interest.
The limitation is that this protocol should be applied to immortalized cells, as many weeks of culturing as well as a cellular cloning are required along with numerous successive amplifications. Alternatively cells, which can be amplified and maintained for few weeks or months should be used. Using this procedure, human immortalized muscle cell line KO for the gene of interest (here the RYR1 gene) has been produced in about 3-4 months (excluding the functional characterization of the cells).
This strategy is suited for difficult to transfect cells, such as muscle cells, as it relies on the use of lentivirus to introduce all the tools into the target cells, and it is thus more powerful than other methods such as transfection or electroporation. Its major drawback is integration in the host genome, and a possible alternative to avoid damaging integration of lentivirus into the host genome is the use on non-integrative lentivirus17, but the efficiency of these non-integrative lentiviral particles should be at least equivalent to the regular lentivirus.
The integration site could probably influence the expression level of Cas9, but this most probably is not responsible for the small amount of Cas9 in Cl-KOR 2. The small amount of Cas9 is most probably related to the transduction of this specific clone with the LV-Killer, which could have been lower than in the other clone, or alternatively the number of LV-Cas9 could have been higher.
It is not recommended to express Cas9 alone without gRNA for a long time (such as for the development of a Cas9 clone), as this has been described to increase the off-targets, and it may also increase the mortality of the cells.
We have chosen to use the SpCas9 as nuclease, because it is widely used, it has a short PAM, and therefore offers several options for the choice of guide RNAs. In addition, its size is compatible with the production of lentiviral vectors. Other nucleases can be used if a specific PAM has to be targeted (such as SaCas9 or Cpf1)18.
The different lentiviruses can be produced in a regular BSL2 culture room19 or purchased from a virus production facility or a company. The requirement to implement this protocol is expertise in molecular biology and molecular cloning, and in culturing of the chosen cell model.
The lentiviruses encoding the different guides also encode the fluorescent protein, mCherry. This fluorescent protein can be useful, allowing automated sorting of the cells and cloning by FACS instead of single cell cloning, and in this case, the expression of mCherry should not be suppressed with the LV-killer. If mCherry is to be maintained, then the second guide against mCherry should not be cloned into the p-killer (step 2.7). For the RYR1 gene, as the functional characterization makes use of fluorescent probes, and the future use of this cell line could also require immunolabeling with fluorescent antibodies, we have chosen to suppress mCherry expression. In the LV-killer, the gRNA targeting Cas9 has been placed under a strong promotor H1 instead of the weak promotor 7SK initially present in the original plasmid, to ensure a similar efficiency of the nuclease on all the selected gRNAs, and to increase the Cas9 cleavage once the LV-killer is added to the cells.
The longest and most strenuous step is single cell cloning. In our procedure, the muscle cells displayed a reduced growth after lentiviral transduction with Cas9 and recovered after a few weeks. If the single cell cloning is performed too early after the viral transductions, probably because of massive cell death, only a few wells of a 96-well plate will contain growing cells. Thus, it is better to amplify the cells before cloning, and wait for at least 2-3 splitting. It is also possible to clone the cells at 10 cells/well if they don't grow well as single cells. In addition, the myoblasts should always be cultured at a confluency below 50% in order to maintain a good differentiation ability.
The authors have nothing to disclose.
This work was funded by grants from Association Française contre les myopathies (AFM-Téléthon) and from Auvergne-Rhône Alpes Région (AURA).
Anti-CACNA1S antibody | Sigma-Aldrich | HPA048892 | Primary antibody |
Blp I | NE BioLabs | R0585S | Restriction enzyme |
CalPhos Mammalian Transfection Kit | Takara | 631312 | Transfection kit |
Easy blot anti Mouse IgG | GeneTex | GTX221667-01 | HRP secondary antibody |
Easy blot anti Rabbit IgG | GeneTex | GTX221666 | HRP secondary antibody |
Fluo-4 direct | Molecular Probes | F10472 | Calcium imaging |
GAPDH(14C10) Rabbit mAb | Cell Signaling Technology | #2118 | Primary antibody |
HindIII | Fermentas | ER0501 | Restriction enzyme |
InFusion HD Precision Plus | Takara | 638920 | Ligation kit |
MasterMix Phusion High Fidelity with GC | ThermoFisher Scientific | F532L | Mix for PCR reaction with High fidelity Taq polymerase and dNTPs |
Myosin Heavy Chain antibody | DHSB | MF20 | Primary antibody |
NucleoBond Xtra Maxi EF | Macherey-Nagel | REF 740424 | Maxipreparation kit for purification of plasmids |
NucleoSpin Gel and PCR Clean-up | Macherey-Nagel | 740609 | DNA purification |
NucleoSpin Tissue | Macherey-Nagel | 740952 | Kit for DNA extraction from cell |
One Shot Stbl3 Chemically Competent E. coli | ThermoFisher Scientific | C737303 | Chemically competent cells |
Plasmid #87904 | Addgene | 87904 | Lentiviral plasmid encoding the SpCas9 (for LV-Cas9) |
Plasmid #87919 | Addgene | 87919 | Lentiviral backbone for insertion of cassette with guides (for LV-guide-target) |
Plasmid #12260 | Addgene | 12260 | Lentiviral plasmid encoding lentiviral packaging GAG POL |
Plasmid #8454 | Addgene | 8454 | Lentiviral plasmid encoding envelope protein for producing lentiviral and MuLV retroviral particles |
V5 Tag Monoclonal Antibody | Invitrogene | R96025 | Primary antibody |
XL10-Gold Ultracompetent Cells | Agilent | 200317 | Chemically competent cells |
Xma I | NE BioLabs | R0180S | Restriction enzyme |