This protocol provides a method to facilitate the generation of defined heterozygous or homozygous nucleotide changes using CRISPR-CAS9 in human pluripotent stem cells.
Human pluripotent stem cells offer a powerful system to study gene function and model specific mutations relevant to disease. The generation of precise heterozygous genetic modifications is challenging due to CRISPR-CAS9 mediated indel formation in the second allele. Here, we demonstrate a protocol to help overcome this difficulty by using two repair templates in which only one expresses the desired sequence change, while both templates contain silent mutations to prevent re-cutting and indel formation. This methodology is most advantageous for gene editing coding regions of DNA to generate isogenic control and mutant human stem cell lines for studying human disease and biology. In addition, optimization of transfection and screening methodologies have been performed to reduce labor and cost of a gene editing experiment. Overall, this protocol is widely applicable to many genome editing projects utilizing the human pluripotent stem cell model.
Human embryonic stem cells (hESCs) and induced pluripotent stem cells (iPSCs) are valuable tools for modeling human disease due to their capacity for renewal, while maintaining the ability to generate cell types of different lineages1,2,3,4. These models open the possibility to interrogate gene function, and understand how specific mutations and phenotypes are related to various diseases5,6. However, to understand how a specific alteration is linked to a particular phenotype, the use of a paired isogenic control and mutant cell lines is important to control for line to line variability7,8. Transcription activator-like effector nucleases (TALENs) and zinc finger nucleases have been used to generate insertion or deletion (indels) mutations in diverse genetic models, including primary cells; but these nucleases can be cumbersome to use and expensive9,10,11,12,13,14. The discovery of the clustered regularly interspaced short palindromic repeat (CRISPR)-CAS9 nuclease has revolutionized the field due to efficiency in indel formation in virtually any region of the genome, simplicity of use, and reduction in cost15,16,17,18,19.
A challenge in using the CRISPR-CAS9 based genome editing technology has been the generation or correction of specific mutations in one allele without creating an indel mutation in the second allele20. The major goal of this protocol is to overcome this challenge by using two single-stranded oligonucleotide (ssODN) repair templates to reduce indel formation in the second allele. Both ssODNs are designed to contain silent mutations to prevent re-cutting by the CAS9 nuclease, but only one contains the alteration of interest. This method increases the efficiency of generating a specific heterozygous genetic modification without inducing indel formation in the second allele. Using this protocol, gene editing experiments in six independent genomic locations demonstrate the precise introduction of the desired genomic change in one allele without indel formation in the second allele and occurs with an overall efficiency of ~10%. The described protocol has been adapted from Maguire et al.21.
1. Design and construction of guide RNA (gRNA)
NOTE: Each gRNA is made up of two 60 base pair (bp) oligonucleotides that are annealed to generate a 100 bp double stranded (ds) oligonucleotide (Figure 1A-C). The timeline for gRNA design, generation, and testing cutting efficiency is approximately 2 weeks (Figure 2).
2. Design of PCR primers for screening
3. Preparation of gRNA_cloning vector plasmid
4. Assembly of gRNA vector
5. Test gRNA cutting efficiency
6. Clone screening
7. Precise genome editing in pluripotent stem cells using single strand oligo DNA (ssODNs)
8. Transfection setup
9. Checking for mutations in single colonies
Generation of gRNAs and screening for indels
Each gRNA will be cloned into a plasmid vector and expressed using the U6 promoter. The AflII restriction enzyme is used to linearize the plasmid (addgene #41824) and is located after the U6 promoter. The 100 bp band generated after annealing the two 60 bp oligos is cloned into the gRNA expression vector using the DNA assembly. Once the gRNA plasmids are generated, they are transfected into hESCs or iPSCs along with a CRISPS-CAS9 GFP plasmid (addgene #44719). The GFP+ cells are sorted after 2 days to enrich for transfected cells and plated (see section 5). After 10-14 days, single cell derived colonies are picked and used to isolate DNA to screen for indel formation generated for each gRNA. A PCR amplification with primers spanning the gRNA site is used to visualize indel formation using a 2.5% (w/v) agarose gel with EtBr, electrophoresed at 70-90 V for 1 h.
Generation of 100 bp ssODN to introduce specific mutations
Two ssODN oligos are designed around the gRNA with the most efficient cutting. Each gRNA is 100 bp and contains silent mutations, preferably at the PAM sequence, to avoid re-cutting (Figure 4A). A silent mutation in the PAM sequence can generate a unique restriction site, which helps to screen for successful integration into one or two alleles (Figure 4B).
ssODN recombination outcomes
Using this protocol, the frequency of targeting events for six different genes using the two ssODN approach is shown in Figure 5. By examining genomic modification at the gRNA site in both alleles, different outcomes were expected including recombination of the wildtype (WT) ssODN, the mutant ssODN and indel formation. The clones in which only one allele had undergone recombination, the most common outcome was indel formation in the second allele (10-42%). The clones that had integration of the ssODNs in both alleles led to three different outcomes: 1) integration of the WT ssODN in both alleles (0-8%), 2) integration of the mutant ssODN in both alleles (0-25%), and 3) integration of both the WT and mutant ssODN WT (8-21%).
Figure 1: Overview of gRNA generation. (A) In the region of interest, design four gRNAs ending in GG that are not more than 20 bp apart. (B) Each gRNA of 23 bp has a PAM sequence of NGG at the 3' end. (C) The 60 bp primers are comprised of a 40 bp sequence complementary to the gRNA backbone plasmid and the 20 bp gRNA sequence without the PAM sequence. Please click here to view a larger version of this figure.
Figure 2: Protocol timeline for gRNA generation and clone screening. The timeline for gRNA generation and screening clones is shown. Designing and cloning gRNA expression plasmids takes approximately one week. Approximately 48 hours after transfection into human PSCs, GFP+ cells are sorted and plated at limiting dilution. Colonies should be visible between 7-10 days and at approximately day 20, clones can be picked for screening, expansion, and sequence confirmation. Please click here to view a larger version of this figure.
Figure 3: gRNA test for cutting efficiency. (A) For gRNA construction, a 100 bp band is excised from a 1.5% agarose gel. (B) Following cell transfection, validation of gRNA cutting is visualized using a 2.5% agarose gel. A 180 bp PCR product is used to check for indel formation in which an uncut control and different clones are analyzed for band shifts indicative of indel formation. (C) Different gRNAs may have different cutting efficiencies. Please click here to view a larger version of this figure.
Figure 4: Generation and screening of mutations using two ssODNs. (A) To avoid CRISPR-CAS9 re-cutting of the edited alleles, the PAM sequence can be modified using a G to A silent mutation. This modification adds a unique EcoRI restriction site that can be used for screening. In addition to this silent mutation, the mutant ssODN contains the desired base change. (B) Screening for oligonucleotide recombination using EcoRI restriction enzyme digestion can result in three possible outcomes: no cutting represented by a WT band at ~650 bp, recombination in one allele represented by an uncut band of 650 bp and two smaller bands or insertion into both alleles represented by only smaller bands. Please click here to view a larger version of this figure.
Figure 5: ssODN integration efficiency and outcomes. Integration efficiency and outcomes of the two ssODN approach were determined by analyzing six different genes. If only one ssODN integrated, indel formation was usually detected in the other allele. If two ssODNs integrated, three possible outcomes were detected, as shown. Please click here to view a larger version of this figure.
Reagent | Volume (μL) |
dNTPs (10 mM) | 0.4 |
Taq polymerase | 0.2 |
PCR buffer | 4.0 |
Annealed oligos (10 μM) | 10.0 |
ddH2O | 5.4 |
Table 1: gRNA cloning PCR conditions.
PCR program | |
Step 1 | 98 °C for 30 s |
Step 2 | 98 °C for 10 s |
Step 3 | 55 °C for 20 s |
Step 4 | 72 °C for 30 s |
Step 5 | Repeat steps 2−4 for 30 cycles |
Step 6 | 72 °C for 5 min |
Step 7 | Hold at 4 °C |
Table 2: PCR cycling parameters.
Reagent | Volume (μL) |
linearized gRNA vector using AflII (i.e., 50 ng/μL) | 1.0 |
100 bp DNA (i.e., 250 ng/μL) | 1.0 |
Master mix 2x | 10.0 |
ddH2O | q.s. to 20 |
Table 3: gRNA assembly reaction conditions.
Reagent | Amount |
DMEM/F12 | 50.0 μL |
pCas9_GFP vector (addgene plasmid 44719) | 0.5 μg |
gRNA plasmid | 0.5 μg |
lipid transfection reagent | 3.0 μL |
Table 4: Cell transfection master mixture.
hESC medium | |
Reagent | Final concentration |
DMEM/F12 | |
Knockout serum replacement (KDR) | 15% (v/v) |
Nonessential amino acids | 100 µM |
Sodium pyruvate | 1 mM |
Glutamine | 2 mM |
β-mercaptoethanol | 0.1 mM |
bFGF human (basic fibroblast growth factor) | 10 ng/mL |
10x Proteinase K buffer | |
Tris.HCl pH:7.4 | 50 mM |
Ammonium sulfate pH: 9.3 | 15 mM |
MgCl2 | 2.5 mM |
Tween 20 | 0.1% (v/v) |
Proteinase K | 100 µg/mL |
Table 5: Cell culture medium and proteinase K digestion buffer.
Reagent | Amount |
DMEM/F12 | 50.0 μL |
pCas9_GFP vector (addgene plasmid 44719) | 0.5 μg |
gRNA plasmid | 0.5 μg |
ssODN (0.5 μg of each ssODN) | 1.0 μg |
lipid transfection reagent | 3.0 μL |
Table 6: ssODN cell transfection master mixture.
In this protocol, the use of CRISPR-CAS9 along with two ssODN repair templates to generate specific heterozygous or homozygous genome changes is demonstrated in human pluripotent stem cells. This method resulted in the successful generation of isogenic cell lines expressing heterozygous genomic changes with an efficiency close to 10%. This protocol has been optimized for both human ESCs and iPSCs grown on irradiated MEFs which support cell growth and survival after culturing cells at low density after cell sorting. Cell death can be minimized by maintaining cells with 10 ng/mL bFGF and Y-27632 dihydrochloride. It is possible that this protocol may be adapted to feeder free culture systems, but further optimization may be necessary.
Transfection efficiency can be variable from cell line to cell line, but the use of 3 µg of DNA and 3 µL of a lipid transfection reagent generally gave the best results of between 0.5-2% transfection efficiency. However, if transfection efficiency is lower, optimization of the amount of DNA and lipid reagent can be performed. Other transfection reagents can also be tested by the investigator.
The efficiency of indel generation will vary with different gRNAs and can be influenced by genomic location. Within the vast majority of cases, if four gRNAs are designed, at least one and many times 2 to 3, will work efficiently. Additionally, prior to testing gRNAs, optimization of the PCR strategy to visualize indels with a clean single band DNA product is important. For screening ssODN based genome editing, it is best to add a restriction enzyme site when introducing a silent mutation(s), but if not possible, deletion of a restriction enzyme site can be used as well. In addition, homology directed repair requires actively cycling cells so the cell density of cultures prior to transfection is critical to enhance the frequency of clones repaired using the introduced ssODN template.
Some of the limitations of this protocol are related to the position in the genome that will be edited. When coding regions are modified, ssODN carrying silent mutations prevent the Cas9 from re-cutting the edited site and the silent mutations do not alter the protein product. However, editing in regulatory or non-coding regions using this methodology becomes more difficult as silent mutations are not possible. If the base to be edited is part of a PAM sequence, this can be done successfully but only homozygous changes can be generated efficiently.
The protocol described here is useful to generate or correct heterozygous coding mutations without the addition of unintended indels in the second allele. This protocol will facilitate the use of human PSCs to study a wide range of topics from developmental biology to modeling of genetic diseases. The use of isogenic lines is critical to define functions of a given coding mutation without confounding effects due to differing genetic backgrounds.
The authors have nothing to disclose.
This research was supported by funding from the National Heart, Lung, and Blood Institute (NHLBI), National Institutes of Health through grants U01HL099656 (P.G. and D.L.F.) and U01HL134696 (P.G. and D.L.F.).
5-ml polystyrene round-bottom tube with cell-strainer cap | Corning | 352235 | |
6-well polystyrene tissue culture dishes | Corning | 353046 | |
AflII restriction endonuclease | New England Biolabs | R0520 | |
Agarose | VWR | N605 | |
DMEM/F12 medium | ThermoFisher | 11320033 | |
dNTPs | Roche | 11969064001 | |
Fluorescence-activated cell sorter (FACS) apparatus | |||
Gel extraction kit | Macherey-Nagel | 740609 | |
Gibson Assembly Kit | New England Biolabs | E2611 | |
gRNA_Cloning Vector | Addgene | 41824 | |
LB agar plates containing 50 μg/ml kanamycin | |||
Lipofectamine Stem Reagent | ThermoFisher | (STEM00001) | |
Matrigel Growth Factor Reduced (GFR) | Corning | 354230 | |
Murine embryonic fibroblasts (MEFs) | |||
Nucleospin Gel Extraction and PCR Clean-up Kit | Macherey-Nagel | 740609 | |
Orbital shaking incubator | |||
pCas9_GFP vector | Addgene | 44719 | |
PCR strip tubes | USA Scientific | 1402-2900 | |
Phusion High Fidelity DNA Polymerase and 5× Phusion buffer | New England Biolabs | M0530 | |
PurelinkTM Quick Plasmid Miniprep Kit | Invitrogen | K210011 | |
Proteinase K | Qiagen | Qiagen 19133 | |
StellarTM electrocompetent Escherichia coli cells | Takara | 636763 | |
SOC medium | New England Biolabs | B9020S | |
TrypLE Express Enzyme | ThermoFisher | 12605036 | |
Y-27632 dihydrochloride/ROCK inhibitor (ROCKi) | Tocris | 1254 |