Skip to main content
NCBI home page
Molecular Therapy logoLink to Molecular Therapy
. 2014 Dec 16;23(3):570–577. doi: 10.1038/mt.2014.226

Efficient and Allele-Specific Genome Editing of Disease Loci in Human iPSCs

Cory Smith 1,2,3, Leire Abalde-Atristain 1,2,4, Chaoxia He 1,2, Brett R Brodsky 1, Evan M Braunstein 1, Pooja Chaudhari 4,5, Yoon-Young Jang 2,4,5, Linzhao Cheng 1,2,3,4,*, Zhaohui Ye 1,2,*
PMCID: PMC4351458  PMID: 25418680

Abstract

Efficient and precise genome editing is crucial for realizing the full research and therapeutic potential of human induced pluripotent stem cells (iPSCs). Engineered nucleases including CRISPR/Cas9 and transcription activator like effector nucleases (TALENs) provide powerful tools for enhancing gene-targeting efficiency. In this study, we investigated the relative efficiencies of CRISPR/Cas9 and TALENs in human iPSC lines for inducing both homologous donor-based precise genome editing and nonhomologous end joining (NHEJ)-mediated gene disruption. Significantly higher frequencies of NHEJ-mediated insertions/deletions were detected at several endogenous loci using CRISPR/Cas9 than using TALENs, especially at nonexpressed targets in iPSCs. In contrast, comparable efficiencies of inducing homologous donor-based genome editing were observed at disease-associated loci in iPSCs. In addition, we investigated the specificity of guide RNAs used in the CRISPR/Cas9 system in targeting disease-associated point mutations in patient-specific iPSCs. Using myeloproliferative neoplasm patient-derived iPSCs that carry an acquired JAK2-V617F point mutation and α1-antitrypsin (AAT) deficiency patient-derived iPSCs that carry an inherited Z-AAT point mutation, we demonstrate that Cas9 can specifically target either the mutant or the wild-type allele with little disruption at the other allele differing by a single nucleotide. Overall, our results demonstrate the advantages of the CRISPR/Cas9 system in allele-specific genome targeting and in NHEJ-mediated gene disruption.

Introduction

Techniques to edit genomic DNA at a precise locus in human induced pluripotent stem cells (iPSCs) present an unprecedented potential for regenerative medicine as well as disease modeling of genetic variants. The efficiency of multiple published techniques in human iPSCs has been extremely low until the development of engineered nucleases such as zinc finger nucleases (ZFNs), transcription activator like effector nucleases (TALENs), and Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)/Cas systems.1,2,3,4,5,6 ZFNs and TALENs are fusion proteins, in which Fok1 nuclease domain is fused to a DNA binding domain that can bind to and cleave a specific DNA sequence of interest. Once a targeted double strand break (DSB) has been introduced, the DNA is repaired by the cell's endogenous DNA repair machinery through one of two pathways. The error-prone nonhomologous end joining (NHEJ) pathway often results in small insertions or deletions (indels), while the Homology-directed Repair (HDR) pathway results in precise repair with a homologous chromosome or an exogenous donor template. These engineered proteins acting as designer nucleases proved to be an efficient means to target and manipulate the genome for both gene knock out (KO) and knock in (KI) experiments. Compared to ZFNs, design of a pair of TALENs is more feasible for most laboratories and favored by many investigators. However, a pair of TALENs, each has ~15 peptide modules of 33-amino-acid units, still takes time to synthesize and test to ensure its efficiency as well as specificity.

In comparison to a pair of TALENs, the CRISPR-Cas type II is more user-friendly as the protein component (Cas9) remains the same while the short RNA components for one or multiple targets can be rapidly designed and synthesized. The system was originally identified to have three essential components: (i) Cas9 endonuclease, (ii) CRISPR RNA (crRNA) to bind the complementary DNA target, and (iii) trans-activating RNA (traRNA) to associate crRNA to Cas9. This was further reduced to two components by fusing the traRNA and crRNA to a single guide RNA (gRNA).7 Cas9 has been adapted for better expression in mammalian cells and was shown to be an efficient and adaptable tool in human cell lines including iPSCs.4,5

In order to gain a better understanding of the advantages and disadvantages of CRISPR/Cas9 and TALENs technologies in human iPSCs, we compared the efficiency of Cas9-gRNAs versus TALENs in targeting disease-associated loci. We investigated their efficiencies by measuring both NHEJ-mediated indel mutations and homologous donor-based precise gene editing. The model disease genes we used include JAK2, in which an acquired somatic point mutation (JAK2-V617F) occurs in approximately 95% of patients with polycythemia vera (PV),8 and the SERPINA1 gene, in which an inherited point mutation (AAT Z-mutation) causes α1–antitrypsin (AAT) deficiency.9 We also included the previously validated Cas9-gRNA and TALEN designer nucleases targeting the AAVS1 locus commonly used as a “safe harbor” in the human genome for stable transgene expression.2,10,11,12,13

Concerns were raised over the specificity of the CRISPR/Cas9 system, when reports using human cell lines detected NHEJ-mediated off-target mutations at loci with up to five mismatches to the gRNA.14,15 As this was further investigated in other biologically relevant samples such as mouse and nonhuman primate embryos and human adult stem cells, off-target effects were not detected or minimal.16,17,18,19,20,21,22 To further investigate genome stability, whole genome sequencing (WGS) was performed on either TALEN- or CRISPR-edited iPSC clones revealing ~100–400 mutations in each modified clone. However, none of these mutations were similar to the gRNA target or recurrent among clones sequenced, suggesting that these variations were not the direct result of nuclease-mediated gene targeting.23,24,25 As a complementary and more quantitative approach to WGS of selected iPSC clones, here we investigated the specificity of Cas9-gRNAs in human iPSCs in stimulating both NHEJ-mediated indel induction and donor-based genome editing by targeted deep sequencing of whole iPSC populations treated with Cas9-gRNAs.

The high specificity of Cas9-gRNAs also prompted us to investigate whether this technology can facilitate allele-specific targeting at point mutations in patient-specific and normal iPSCs. For this purpose, we used a variety of integration-free human iPSC lines including PV-iPSC lines that carry the JAK2-V617F mutation, AAT deficiency-iPSC lines that carry the Z-AAT mutation, and control BC1 iPSC line whose genomic integrity has been characterized in detail by next generation sequencing26,27 (Table 1). We designed gRNAs targeting either the mutant or the wild-type allele and examined their efficiency in disrupting or correcting the intended target allele or the other allele differing by a single nucleotide.

Table 1. Human iPSC lines used in this study.

graphic file with name mt2014226t1.jpg

Open in a new tab

Results

Cas9-gRNAs are superior to TALENs in inducing NHEJ-mediated indel mutations

We evaluated the efficiency of Cas9-gRNA versus TALEN in targeting multiple endogenous loci in human iPSCs. We first determined, by MiSeq deep sequencing, the frequencies of small indels resulting from NHEJ-mediated DNA repair of DSBs caused by either Cas9 or TALENs in the absence of an exogenous donor template (Figure 1). Cas9 and the guide RNA (gR-JAK2-F) designed to target the JAK2-V617F mutation were compared to a pair of JAK2-TALENs designed to target the same JAK2-V617F region. Cas9 and the guide RNA (gR-AAT-Z) designed to target the Z-AAT mutation were compared to a pair of AAT-TALENs that previously have been shown to efficiently target the AAT Z-mutation region28 (Figure 1 and Supplementary Figure S1). In addition to these designer nucleases targeting disease-associated mutations, previously reported Cas9-gRNA (gR-AAVS1-T2) and TALENs targeting the AAVS1 safe harbor locus were also used in this study.4,29 These Cas9-gRNA or TALEN expression vectors were first tested in 293T cells using a previously reported GFP reporter system to validate their functionality1,30 (Supplementary Figure S1). They were then transfected into three human iPSC lines with appropriate target sequences (Table 1). Three days after transfection, the genomic DNA was amplified by high-fidelity PCR using primers flanking the common targeted DNA region shared by both designer nucleases for each locus. MiSeq analysis of these PCR-amplified regions revealed significantly higher indel rates at all three endogenous loci after Cas9 targeting than TALENs (Figure 2). At all three loci, Cas9-gRNAs induced between 10- and 100-fold more indels than did TALENs in human iPSCs, reaching the level of 0.7–2.5% mutation rates.

Figure 1.

Figure 1

Open in a new tab

Genomic loci targeted by Cas9-gRNAs and transcription activator like effector nucleases (TALENs) in human induced pluripotent stem cells (iPSCs). (a) The genomic structure and nucleotide sequence around JAK2-V617F mutation in JAK2 gene are shown. Exon sequence is shown in uppercase. The G to T point mutation in exon 14 is indicated. A diagram of the donor vector used in HR-based JAK2 targeting (Tables 2 and 3) is shown. The vertical arrow indicates the G>T point mutation in JAK2-V617F. The PGK-puroΔtk dual selection cassette was flanked by piggyBac (PB) 5′ and 3′ inverted terminal repeats to facilitate potential footprint-free genome editing after PB-transposase-mediated excision. The vectors contain JAK2 homology arms flanking the putative Cas9-gRNA and TALEN cutting sites. (b) The genomic structure and nucleotide sequence around Z-AAT mutation in SERPINA1 gene. The G to A variant in exon 5 is indicated. The diagram of a previously reported donor vector used in AAT targeting is shown. (c) The genomic structure and nucleotide sequence in intron 1 of the PPP1R12C gene. The diagram of a previously reported donor vector used in AAT targeting is shown. Recognition sequences of Cas9-gRNA (boxed, PAM sequences shown in green) and TALENs (underlined) are shown at each locus. The relative efficiencies of each pair of Cas9-gRNA and TALENs in an EGIP reporter assay in 293T cells are shown in Supplementary Figure S1. PCR primers for identifying targeted integration events (red arrows) and or un-targeted allele (black arrows) are shown. Detailed information on PCR screening of targeted integration events are shown in Supplementary Figures S2 and S3. HA-L, left homology arm; HA-R, right homology arm.

Figure 2.

Figure 2

Open in a new tab

Efficiencies of NHEJ-mediated gene disruption at endogenous loci induced by Cas9-gRNAs and transcription activator like effector nucleases (TALENs) in human induced pluripotent stem cells (iPSCs). Frequencies of small indels at indicated locus after Cas9-gRNA or TALEN transfection in human iPSCs. PVB1.4 (homozygous JAK2-V617F mutation), iAAT3 (homozygous AAT Z-mutation) and BC1 (normal control) (Table 1) were used in experiments targeting JAK2, SERPINA1, and AAVS1 loci, respectively. Genomic DNA from un-transfected parental iPSCs was used as background controls (red) compared to those transfected with TALENs or Cas9-gRNA (blue). Error bars indicate 95% Wilson score intervals. N.S., not significant (P > 0.05); **P < 2 × 10–5; ***P < 2.2 × 10–16. Data represent single transfection and MiSeq experiments. The details of the data are presented in Supplementary Table S1.

Comparable efficiencies between Cas9-gRNAs and TALENs in facilitating genome editing by homologous donors

One common approach in precise genome editing is to introduce a homology donor together with the engineered endonucleases into the target cells. To assess the ability of Cas9-gRNAs and TALENs in promoting homology-directed repair, we conducted gene targeting experiments with homologous donors at JAK2, SERPINA1, and AAVS1 loci in the same human iPSCs as we did in the MiSeq/NHEJ experiments. We designed an HR donor template with homology arms near the Cas9-gRNA and TALEN cutting sites at the JAK2 locus (Figure 1a). Previously reported HR donors shown to successfully target the AAT Z-mutation by ZFN and TALEN technologies28,31 and the AAVS1 locus29 were also used (Figure 1b,c). In each targeting experiment, 2 × 106 iPSCs were cotransfected with a homology donor vector and either the Cas9-gRNA or the TALENs. iPSC colony numbers were counted after puromycin selection, and iPSC clones were randomly picked and expanded. Genomic DNA isolated from each clone was used for PCR screening for targeted integration (TI) events. The absolute targeting efficiencies were calculated as “targeted integration events per million input cells” based on (i) the percentage of randomly selected iPSC clones that are positive for TI; (ii) the total colonies after puromycin-selection; and (iii) the total input cells (Table 2). In contrast to what was observed in the NHEJ experiments, transfection of Cas9-gRNA and TALENs designed to target each locus resulted in comparable efficiencies; the largest difference was observed at the AAVS1 site targeting where the Cas9-gRNA showed about twofold advantage over TALENs (Table 2).

Table 2. HDR efficiencies of TALEN and hCas9 at three loci in human iPSCs.

graphic file with name mt2014226t2.jpg

Open in a new tab

High specificity of CRISPR/Cas9 in human iPSCs revealed by genome-wide targeted deep sequencing

We next determined the specificity of the gRNAs that together with Cas9 generate indels in human iPSCs at much higher efficiency than TALENs. We also used MiSeq deep sequencing to determine indel frequencies at potential off-target loci with DNA sequences similar to the intended locus. The potential off-target loci were predicted using a previously described bioinformatics program,14 and the top 14–15 loci (Supplementary Table S1) were measured together with the intended loci (JAK2 and AAVS1) in the same experiments (Figure 3). In addition to human iPSCs, we also measured indel frequencies induced by these same Cas9-gRNAs in the widely used human 293T cell line using the same Miseq strategy (Figure 3). After transfection with Cas9 and gRNAs, genomic DNA was isolated from the entire transfected population. Upon successful PCR amplification, these loci were analyzed by MiSeq in comparison to those in their parental (un-transfected) cells. Among the 29 most likely off-targets of the two gRNAs, statistically significant indels were detected in 22 of them in 293T cells (Figure 3 and Supplementary Table S1). Two of the most significant AAVS1 off-targets each had >10% of absolute indel rates and were >100-fold higher than observed in the controls (OT-5: 9683/63422 (15.27%) versus 130/401234 (0.03%), P < 2.2 × 10-16; OT-8: 14261/121156 (11.77%) versus 40/59427 (0.07%), P < 2.2 × 10-16). In comparison, overall NHEJ rates in iPSCs were ~50-fold lower than those in transfected 293T cells (Figure 3). AAVS1 OT-5, with less than 5% of the on-target efficiency, is the only site that showed statistically significant indels in the targeted iPSCs (Figure 3 and Supplementary Table S2). These results demonstrated the high specificity of Cas9-gRNAs in human iPSCs as compared to 293T cells that have an overall higher level of NHEJ activity and also higher off-target activities by some gRNAs.

Figure 3.

Figure 3

Open in a new tab

Targeted genome-wide deep sequencing of potential off-targets reveals high specificity of CRISPR/Cas9 in human induced pluripotent stem cells (iPSCs). Summarized results of targeted deep sequencing of the 29 bioinformatically predicted most likely off-targets in the human genome. For each Cas9-gRNA, the indel frequencies (x axis) of their intended target and each potential off-target (OT) in the transfected cells (shown in blue) are shown in comparison to the background (DNA from parental cells, shown in red). The mismatched nucleotides of each off-target are shown. The PAM sequence of the target and putative off-targets are shown in green. PVB1.4 (homozygous JAK2-V617F mutation) and BC1 (normal control) iPSC lines were used for JAK2 and AAVS1 targeting, respectively. Note that the guide RNA gR-JAK2-F used in this study was designed to target the JAK2-V617F mutation; therefore a single nucleotide mismatch is present in the 293T JAK2 locus (see Figure 3a). *Sites with statistically significant (P < 0.05) indels above background level. Data represent single transfection and MiSeq experiments. Detailed sequence and analysis information of each locus are shown in Supplementary Table S2.

Allele-specific gene targeting at point mutation loci in human iPSCs using Cas9-gRNAs

Having demonstrated high specificity of Cas9-gRNAs in genome-wide studies, we then investigated the efficiency of targeting specific base pairs in human iPSCs. For this purpose, we again chose to target the somatic JAK2-V617F (rs77375493, G>T) mutation and the inherited α1-antitrypsin (AAT) Z mutation (rs28929474, G>A) in the SERPINA1 gene. Two panels of healthy donor and patient-derived iPSCs carrying either a homozygous wild-type target allele, heterozygous mutant allele, or homozygous mutant allele were used in these experiments26,28,32,33,34 (Table 1). Two gRNAs, gRNA-JAK2-F, and gRNA-JAK2-V, were designed with a single nucleotide difference to specifically recognize the V617F mutant allele (T) or the wild-type allele (G) of JAK2 (Figure 4a). Another pair of gRNAs, gRNA-AAT-Z and gRNA-AAT-M, were designed to specifically recognize the AAT Z-allele (A) or the wild-type M-allele (G) (Figure 4b). We examined the efficiency and specificity of the two pairs of gRNAs in targeting the point mutations at endogenous loci by cotransfection with Cas9 (Figure 4). Indel frequency analyses of the JAK2 gene revealed that only the gRNA-JAK2-F successfully targeted the JAK2 locus in the JAK2-V617F homozygous PVB2.7 and PVB1.4 iPSCs. Likewise, only gRNA-JAK2-V successfully targeted the wild-type BC1 and PVB1.11 iPSCs (Figure 4c). The single-nucleotide mismatch between gRNAs and target genomic sequence prevented any major indel mutagenesis by Cas9. Similar results were observed in AAT targeting. The gRNA-AAT-Z, designed to target the Z-allele, had no detectable effect on the wild-type AAT allele, while significant indels were observed after transfection into iPSCs that carry the homozygous Z-mutation. Although a statistically significant (compared to the un-transfected control) indel rate was observed at the Z-alleles in iAAT3 (homozygous mutant) after gRNA-AAT-M targeting, this indel rate (0.055%) is 40-fold lower than that observed after gRNA-AAT-Z transfection (Figure 4d).

Figure 4.

Figure 4

Open in a new tab

Allele-specific gene targeting of JAK2-V617F and Z-AAT mutations in patient-specific induced pluripotent stem cells (iPSCs). (a) Guide RNAs gR-JAK2-F and gR-JAK2-V were designed to specifically target the mutant allele (with nucleotide T) and wild-type allele (with nucleotide G), respectively. (b) Guide RNAs gR-AAT-Z and gR-AAT-M were designed to specifically target the mutant allele (with nucleotide A) and wild-type allele (with nucleotide G), respectively. (c) The specificity of gRNA-JAK2-F and gRNA-JAK2-V that differ by a single nucleotide were evaluated in iPSC lines with V/V or F/F genotype. Human iPSCs with WT JAK2 (BC1 and PVB1.11) or mutant JAK2 (PVB1.4 and PVB2.7, both are homozygous for JAK2-V617F mutation) were cotransfected with Cas9 and either one of the gRNA constructs. The indel frequency at the JAK2 locus in each sample is shown. Guide RNA used in each transfection is indicated underneath the indel frequency. (d) The specificity of gRNA-AAT-Z and gRNA-AAT-M that differ by a single nucleotide were evaluated in iPSC lines with M/M (control BC1) or Z/Z (AAT deficiency patient-specific iAAT3, homozygous AAT Z-mutation) genotype. Major indel events (>0.1%) were only observed when an iPSC line was transfected with the gRNA designed for its genotype. Error bars in c and d indicate 95% Wilson score intervals. N.S., not significant (P > 0.05); *P < 0.001; **P < 2 × 10–5; ***P < 2.2 × 10–16. The details of the data are also presented in Supplementary Table S1.

For gene/cell therapy and disease modeling, it is often highly desirable to achieve precise genome editing based on homologous donors. We therefore evaluated the specificity of Cas9-gRNAs in this setting at the endogenous JAK2 and SERPINA1 loci, using either the PV-iPSC line iPV183 heterozygous for the JAK2-V617F point mutation32 or the iPSC line iAAT5 heterozygous for the Z-AAT point mutation.28,33 To investigate the allelic specificity of inducing integration events by Cas9-gRNAs, we cotransfected iPV183 with a homology donor template and either Cas9/gRNA-JAK2-F or Cas9/gRNA-JAK2-V, and iAAT5 with donor template and either Cas9/gRNA-AAT-Z or Cas9/gRNA-AAT-M. Candidate targeted iPSC clones were randomly picked after cotransfection and puromycin selection expanded and screened by PCR and sequencing to confirm targeted integration at the targeted allele as well as the sequence integrity of the nontargeted allele (Supplementary Figures S2 and S3). Among the expanded PV-iPSC clones, 24/25 and 25/29 had targeted integration by Cas9/gRNA-JAK2-F and Cas9/gRNA-JAK2-V, respectively. Among the expanded AAT-iPSC clones, 15/15 and 14/15 had targeted integration by Cas9/gRNA-AAT-Z and Cas9/gRNA-AAT-M, respectively. Strikingly, in all 49 clones with targeted integration at JAK2 locus, the integration events occurred only at the JAK2 allele specified by the gRNAs (Table 3). Similarly there was only one clone from the Cas9/gRNA-AAT-Z experiment that had targeted integration in both alleles; in all other clones the integration events occurred only at the SERPINA1 allele specified by the gRNAs (Table 3). In addition, sequencing of each targeted clone showed the absence of NHEJ-mediated mutations on the other nontargeted allele, further demonstrating the targeting specificity (Table 3). Taken together, our results from these patient-specific iPSCs strongly suggest the feasibility of specifically targeting a single-nucleotide mutation or variant by CRISPR/Cas9 in human iPSCs.

Table 3. Allele-specific gene targeting at point mutation loci by homologous donors in heterozygous human iPSCs.

graphic file with name mt2014226t3.jpg

Open in a new tab

Discussion

Our quantitative comparison of Cas9-gRNAs and TALENs at three endogenous loci in human iPSCs suggests a higher efficiency of gene disruption by the Cas9 endonuclease, corroborating with recent studies.4,21,35 Although not statistically significant, the current study also suggests an inverse correlation between gene expression level and the advantage of Cas9 over TALEN; the lowest differential indel frequency was observed at AAVS1, which has the highest level of expression in human iPSCs among the three genes. Conversely, the most advantageous indel induction by Cas9 was observed at the SERPINA1 locus, which has the lowest expression level (based on RNA-Seq data from ENCODE/Caltech, GSE33480).36 These data demonstrate that the CRISPR/Cas9 system can efficiently target both expressed and nonexpressed loci in human iPSCs, and may have a particular advantage at nonexpressed loci.

In our previous studies, we have observed high efficiency gene targeting at both AAT and AAVS1 loci in human iPSCs using TALENs.28,29 Therefore, the significantly lower targeting efficiency assayed by indel frequency observed in this study was unexpected. We further examined whether the higher rates of NHEJ-mediated gene disruption correlate with more efficient gene-editing using donor templates with homology arms. We chose to conduct such experiments using the conventional drug selection-based donors as we did in previous studies.28,29 Although detecting NHEJ and HDR in the same sets of experiments would have been more quantitative, practically this can only be done with a short donor template (such as oligonucleotide donors) in order for both events to be detected by MiSeq sequencing. The current efficiency of targeting by an oligonucleotide donor, however, is still extremely low in human iPSCs on average.21,37 Without drug selection, it is technically difficult to identify targeted clones before breakthroughs in technology are made. Therefore, we focus on the efficiency of the conventional selection-based donor in the current study because it remains a major genome-editing tool for most investigators engaged in research using human iPSCs.

Our quantitative investigation of Cas9 and TALENs in facilitating targeted integration events revealed comparable efficiencies of the two technologies (Table 2), even though their ability to induce small indels at the same endogenous loci varied dramatically (Figure 2). There are multiple potential explanations for this. First, the difference in binding/releasing kinetics of TALE or Cas9 to genomic DNA (or other molecules interacting with either Cas9 or TALENs), may differentially affect the subsequent recruitment of the components required for either NHEJ or HDR. As a result, TALENs have a relative preference for HDR over NHEJ at induced DSBs. Alternatively, TALENs generate a higher percentage of single-strand DNA breaks between their binding sites while Cas9-gRNA binding results in more DSBs. Studies using engineered DNA nickases have shown that the single-strand DNA breaks can stimulate efficient HDR without inducing the error-prone NHEJ pathway.38,39,40 Finally, TALENs typically generate DSBs with single-strand overhangs in the space between the two TALE-binding sites while Cas9 has been reported to make blunt-end DSBs.7,41 Different types of DSBs generated by TALENs and Cas9 may result in different preference for repair pathway. We anticipate that future studies at the molecular level of nuclease-DNA interaction and in DNA repair pathways may shed light on the exact mechanism, which can aid to further improve this gene targeting technology.

We further examined the specificity of CRISPR/Cas9 system in human iPSCs using targeted deep sequencing, which provides the most accurate measurement of small indels at endogenous loci.4,5,42 Our previous study using whole genome sequencing (WGS) technology demonstrated the feasibility of generating isogenic iPSC clones through Cas9-mediated gene targeting.23 While the WGS approach offers broad genome coverage with an emphasis on the quality of individual clones, the approach used in this study allowed for a more quantitative analysis by examining predicted off-targets in significantly greater depth in the entire cell population that has undergone genome editing. Each approach has its own advantages and disadvantages. Importantly, we have now demonstrated the high specificity of CRISPR/Cas9 in human iPSCs using these complementary methodologies.

The observed high specificity and efficiency provide a basis for us to conduct allele-specific gene targeting of point mutations in patient-specific iPSCs. Targeting two disease-associated point mutations, the inherited AAT Z-mutation and the acquired JAK2-V617F mutation, in five human iPSC lines, we have shown that each Cas9-gRNA almost exclusively targets its intended sequence and generates only background levels of indels at the other allele that differs by a single nucleotide alone. Using iPSC lines that are heterozygous for either mutation, we have also shown allele-specific targeting in a setting that is more critical for making precise editing for molecular therapy purposes. An additional advantage of using heterozygous cell lines in this study is that it provides an internal control for each Cas9-gRNA and eliminates the experimental variations such as transfection efficiency that may compromise the results.

We anticipate that allele-specific gene targeting can be widely applied to many other loci in human iPSCs using the CRISPR/Cas9 technology. It should be noted that one limitation of the current technique is that it will require a PAM sequence close enough to the variant of interest. In this study, each point mutation is located at the 5th nucleotide 5′ of their respective PAM sequence. For other genetic variants that locate further from PAM sequences, the specificity of Cas9-gRNAs may decrease. However, for certain mutations such as the JAK2-V617F and Z-AAT reported here, it offers a unique targeting specificity that was not easily achieved by previous technologies. Additionally, as new PAM sequences become targetable by adapting Cas9 from other species and/or through protein engineering, we anticipate that this technique will be applicable to more loci in the near future.

Materials and Methods

The experiments using human iPSCs were approved by the Internal Review Board and Stem Cell Research Oversight committees in the Johns Hopkins University.

Maintenance and expansion of human iPSCs. Human iPSCs were cultured with E8 medium (Life Technologies, Carlsbad, CA) on tissue culture plates coated with Matrigel (BD Biosciences, San Jose, CA) or Vitronectin (Life Technologies) as previously described.43,44 For routine passaging, iPSCs were digested with Accutase (Sigma, St. Louis, MO) for 5 minutes and washed with PBS by centrifugation at 200g for 5 minutes. Digested iPSCs were then plated at a density of 2 × 104 per cm2 with E8 medium supplemented with 10 μmol/l ROCK Inhibitor Y-27632 (Stemgent, Cambridge, MA) for the first 24 hours.

Expression vectors used in CRISPR/Cas9 and TALEN experiments. We used the CRIPSR/Cas9 system previously described,4 for which an expression vector encoding humanized (h) Cas9 protein was obtained from Addgene.org (Plasmid #41815). 455-bp guide RNA (gRNA) expression cassettes including 20-bp target-specific sequence for each locus were synthesized (Integrated DNA Technology, Coralville, IA). The JAK2 and SERPINA1 gRNAs were synthesized as Gene Blocks and cloned into Zero-blunt TOPO vector (Life Technologies). The AAVS1 gRNA-T2 was obtained from Addgene.org (Plasmid #41818).

TALEN constructs targeting the AAVS1 and AAT loci have been described in previous publication.28,29 The JAK2 TALENs were constructed with the Joung Lab's REAL Assembly TALEN Kit (Addgene #1000000017) following published protocol.45 The Cas9-gRNA and TALENs targeting each locus were first validated using a GFP reporter system in 293T cells (Supplementary Figure S1).

MiSeq deep sequencing of endogenous loci after transient expression of engineered endonucleases. Human iPSCs were digested with Accutase for 5 minutes and the single cells were washed once with PBS. 2 × 106 iPSCs were then resuspended in 100 μl of P3 Primary Cell Solution (Lonza, Frederick, MD) supplemented with 2.5 μg of hCas9 plasmid and 2.5 μg of gRNA plasmid, and then nucleofected in 4D-Nucleofector (Lonza) using the hES H9 program. In TALEN experiments, 2.5 μg of each TALEN expression vector were used. The nucleofected iPSCs were then plated onto Matrigel-coated plates in E8 medium supplemented with 10 μmol/l Y-27632. Three days after the transfection, all the cells were harvested and the genomic DNAs were isolated using DNeasy Blood and Tissue Kit (Qiagen, Hilden, Germany). The genomic regions of interest were PCR amplified using Phusion DNA polymerase (New England Biolabs, Ipswich, MA). PCR products purified by PCR Purification Kit (Qiagen) were deep sequenced by MiSeq Personal Sequencer (Illumina, San Diego, CA) and demultiplexed using ea-utils program FastqMultx only allowing reads in which both paired ends agreed yielding an average coverage of >300,000 reads per sample. Reads for each sample were aligned using bowtie2 and the indel percentage was calculated by the number of reads with indels around 20-bp of the putative cutting site divided by the total number of aligned reads. The reference sequences for each alignment are shown in Supplementary Table S3. Sequencing data was deposited at the Short Read Archive (SRA) with the accession number SRP042279.

Amplification and sequencing of putative off-target binding sites of gRNAs. To bioinformatically predict the off-target binding sites of three gRNAs (gR-AAVS1-T2, gR-JAK2-F and gR-PEAR1) a list of all sites in hg19 within five mismatches to the gRNA that were followed by a PAM sequence (NRG) was generated using the EMBOSS tool fuzznuc. The lists were ranked based on likelihood of cleavage predicted by an experimentally determined weighting algorithm.14 Using Primer3, oligos were designed to amplify the top 15 putative off-target loci for each gRNAs which were independently transfected with Cas9 into iPSCs and 293T cells. After 96 hours, cells were harvested and gDNA was extracted with DNeasy Blood & Tissue Kit (Qiagen) and the intended targets (4) and predicted off-targets (60) were PCR amplified using Phusion DNA polymerase. Amplicons were pooled into four conditions: (i) parental iPSCs; (ii) iPSC Cas9/gRNA targeted; (iii) HEK293T parental; and (iv) HEK293T Cas9/gRNA targeted which underwent Kapa Biosystems–High-Throughput Library Prep Kit Product # KK8234 for the MiSeq 500 cycle sequencing chemistry. Reads were aligned using bowtie2 and indel mutation rates for each of the amplicons was calculated as previously described.14

Homologous donor-based gene targeting in human iPSCs. In homologous donor-based experiments targeting JAK2 or SERPINA1 loci, 2 × 106 patient-specific iPSCs cells were resuspended in 100 μl P3 Primary Cell Solution supplemented with either (i) 2.5 μg hCas9 plasmid, 2.5 μg of guide RNA plasmid and 5 μg HR donor vector; or (ii) 2.5 μg plasmid coding TALEN-left, 2.5 μg of TALEN-right and 5 μg HR donor vector. In experiments targeting AAVS1 locus, 2 million BC1 iPSCs were nucleofected with either (i) 5 μg AAV-CAGGS-EGFP (addgene # 22212), 3 μg each of a heterodimeric TALEN pair targeting the AAVS129 or (ii) 5 μg AAV-CAGGS-EGFP (Addgene # 22212), 3 μg hCas9 (Addgene #41815), and 3 μg AAVS1-T2 gRNA (Addgene #41815). The nucleofected iPSCs were plated onto Matrigel-coated six-well plates (nine wells per sample in AAT targeting experiments; 12 wells per sample in JAK2 and AAVS1 targeting experiments) immediately following nucleofection. 10 μmol/l ROCK Inhibitor Y-027632 was added in the E8 medium for the first 24 hour. There is no cell passaging between the initial plating and colony picking. Medium was replaced on a daily basis. Starting at 96 hours after nucleofection, puromycin (0.5 μg/ml) was added to the medium for the selection of targeted events. After puromycin selection, colonies were manually picked from different wells (e.g., two colonies were picked from each individual well in the AAT experiments) and expanded. We also tried to avoid picking colonies that grew close to each other. Genomic DNAs were isolated from the expanded clones and analyzed for targeted integration (Supplementary Figures S2 and S3).

Statistical analyses. MiSeq sequencing data analysis was performed as previously described.14 Statistical significance for targeted amplicon sequencing was determined using the upper tail test of population proportion comparing each targeted sample to the untransfected control (α = 0.05). To adjust for multiple comparisons in Figure 2 the bonferroni correction was used (P = 0.05, n = 29) yielding an experiment-wide significance level of 1.72 × 10–3.

SUPPLEMENTARY MATERIAL Figure S1. Relative efficiencies of engineered endonucleases in mediating HR in 293T cells harboring a chromosomally integrated EG*IP reporter. Figure S2. HR-based gene targeting in iPV183 cell line using Cas9-gRNAs and homology donors. Figure S3. HR-based gene targeting in iAAT5 cell line using Cas9-gRNAs and homology donor. Table S1. MiSeq analysis data presented in Figures 2 and 4. Table S2. MiSeq analysis of the putative off-target sites in human iPSCs and 293T cells. Table S3. Reference sequences for MiSeq analysis of indels at AAVS1, JAK2, and SERPINA1 loci.

Acknowledgments

This work was supported in part by grants from Maryland Stem Cell Research Fund (2011-MSCRFE-0087, 2009-MSCRFII-0047, 2011-MSCRFII-0088, 2010-MSCRFII-0101, and 2013-MSCRFII-0170) and by NIH (2R01-HL073781 and U01-HL107446). L.A.-A. received a fellowship from La Caixa Foundation (Spain). L.C. is also supported by Edythe Harris Lucas and Clara Lucas Lynn Chair in Hematology in the Johns Hopkins Medicine. The authors declare no conflict of interest.

Supplementary Material

Supplementary Figures and Tables
Click here for additional data file. (403.6KB, zip)

References

  1. Zou J, Maeder ML, Mali P, Pruett-Miller SM, Thibodeau-Beganny S, Chou BK.et al. (2009Gene targeting of a disease-related gene in human induced pluripotent stem and embryonic stem cells Cell Stem Cell 597–110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Hockemeyer D, Soldner F, Beard C, Gao Q, Mitalipova M, DeKelver RC.et al. (2009Efficient targeting of expressed and silent genes in human ESCs and iPSCs using zinc-finger nucleases Nat Biotechnol 27851–857. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Hockemeyer D, Wang H, Kiani S, Lai CS, Gao Q, Cassady JP.et al. (2011Genetic engineering of human pluripotent cells using TALE nucleases Nat Biotechnol 29731–734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Mali P, Yang L, Esvelt KM, Aach J, Guell M, DiCarlo JE.et al. (2013RNA-guided human genome engineering via Cas9 Science 339823–826. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cong L, Ran FA, Cox D, Lin S, Barretto R, Habib N.et al. (2013Multiplex genome engineering using CRISPR/Cas systems Science 339819–823. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Damian M, Porteus MH. A crisper look at genome editing: RNA-guided genome modification. Mol Ther. 2013;21:720–722. doi: 10.1038/mt.2013.46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Jinek M, Chylinski K, Fonfara I, Hauer M, Doudna JA, Charpentier E. A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science. 2012;337:816–821. doi: 10.1126/science.1225829. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Levine RL, Gilliland DG. Myeloproliferative disorders. Blood. 2008;112:2190–2198. doi: 10.1182/blood-2008-03-077966. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Carrell RW, Lomas DA. Alpha1-antitrypsin deficiency–a model for conformational diseases. N Engl J Med. 2002;346:45–53. doi: 10.1056/NEJMra010772. [DOI] [PubMed] [Google Scholar]
  10. Smith JR, Maguire S, Davis LA, Alexander M, Yang F, Chandran S.et al. (2008Robust, persistent transgene expression in human embryonic stem cells is achieved with AAVS1-targeted integration Stem Cells 26496–504. [DOI] [PubMed] [Google Scholar]
  11. Lombardo A, Cesana D, Genovese P, Di Stefano B, Provasi E, Colombo DF.et al. (2011Site-specific integration and tailoring of cassette design for sustainable gene transfer Nat Methods 8861–869. [DOI] [PubMed] [Google Scholar]
  12. Zou J, Sweeney CL, Chou BK, Choi U, Pan J, Wang H.et al. (2011Oxidase-deficient neutrophils from X-linked chronic granulomatous disease iPS cells: functional correction by zinc finger nuclease-mediated safe harbor targeting Blood 1175561–5572. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Zou C, Chou BK, Dowey SN, Tsang K, Huang X, Liu CF.et al. (2012Efficient derivation and genetic modifications of human pluripotent stem cells on engineered human feeder cell lines Stem Cells Dev 212298–2311. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hsu PD, Scott DA, Weinstein JA, Ran FA, Konermann S, Agarwala V.et al. (2013DNA targeting specificity of RNA-guided Cas9 nucleases Nat Biotechnol 31827–832. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Fu Y, Foden JA, Khayter C, Maeder ML, Reyon D, Joung JK.et al. (2013High-frequency off-target mutagenesis induced by CRISPR-Cas nucleases in human cells Nat Biotechnol 31822–826. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Niu Y, Shen B, Cui Y, Chen Y, Wang J, Wang L.et al. (2014Generation of gene-modified cynomolgus monkey via Cas9/RNA-mediated gene targeting in one-cell embryos Cell 156836–843. [DOI] [PubMed] [Google Scholar]
  17. Wu Y, Liang D, Wang Y, Bai M, Tang W, Bao S.et al. (2013Correction of a genetic disease in mouse via use of CRISPR-Cas9 Cell Stem Cell 13659–662. [DOI] [PubMed] [Google Scholar]
  18. Schwank G, Koo BK, Sasselli V, Dekkers JF, Heo I, Demircan T.et al. (2013Functional repair of CFTR by CRISPR/Cas9 in intestinal stem cell organoids of cystic fibrosis patients Cell Stem Cell 13653–658. [DOI] [PubMed] [Google Scholar]
  19. Wu X, Scott DA, Kriz AJ, Chiu AC, Hsu PD, Dadon DB.et al. (2014Genome-wide binding of the CRISPR endonuclease Cas9 in mammalian cells Nat Biotechnol 32670–676. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Yang H, Wang H, Shivalila CS, Cheng AW, Shi L, Jaenisch R. One-step generation of mice carrying reporter and conditional alleles by CRISPR/Cas-mediated genome engineering. Cell. 2013;154:1370–1379. doi: 10.1016/j.cell.2013.08.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Yang L, Guell M, Byrne S, Yang JL, De Los Angeles A, Mali P.et al. (2013Optimization of scarless human stem cell genome editing Nucleic Acids Res 419049–9061. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Yin H, Xue W, Chen S, Bogorad RL, Benedetti E, Grompe M.et al. (2014Genome editing with Cas9 in adult mice corrects a disease mutation and phenotype Nat Biotechnol 32551–553. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Smith C, Gore A, Yan W, Abalde-Atristain L, Li Z, He C.et al. (2014Whole-genome sequencing analysis reveals high specificity of CRISPR/Cas9 and TALEN-based genome editing in human iPSCs Cell Stem Cell 1512–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Veres A, Gosis BS, Ding Q, Collins R, Ragavendran A, Brand H.et al. (2014Low incidence of off-target mutations in individual CRISPR-Cas9 and TALEN targeted human stem cell clones detected by whole-genome sequencing Cell Stem Cell 1527–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Suzuki K, Yu C, Qu J, Li M, Yao X, Yuan T.et al. (2014Targeted gene correction minimally impacts whole-genome mutational load in human-disease-specific induced pluripotent stem cell clones Cell Stem Cell 1531–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Chou BK, Mali P, Huang X, Ye Z, Dowey SN, Resar LM.et al. (2011Efficient human iPS cell derivation by a non-integrating plasmid from blood cells with unique epigenetic and gene expression signatures Cell Res 21518–529. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Cheng L, Hansen NF, Zhao L, Du Y, Zou C, Donovan FX.et al.; NISC Comparative Sequencing Program 2012Low incidence of DNA sequence variation in human induced pluripotent stem cells generated by nonintegrating plasmid expression Cell Stem Cell 10337–344. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Choi SM, Kim Y, Shim JS, Park JT, Wang RH, Leach SD.et al. (2013Efficient drug screening and gene correction for treating liver disease using patient-specific stem cells Hepatology 572458–2468. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Yan W, Smith C, Cheng L. Expanded activity of dimer nucleases by combining ZFN and TALEN for genome editing. Sci Rep. 2013;3:2376. doi: 10.1038/srep02376. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Porteus MH. Mammalian gene targeting with designed zinc finger nucleases. Mol Ther. 2006;13:438–446. doi: 10.1016/j.ymthe.2005.08.003. [DOI] [PubMed] [Google Scholar]
  31. Yusa K, Rashid ST, Strick-Marchand H, Varela I, Liu PQ, Paschon DE.et al. (2011Targeted gene correction of a1-antitrypsin deficiency in induced pluripotent stem cells Nature 478391–394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Ye Z, Zhan H, Mali P, Dowey S, Williams DM, Jang YY.et al. (2009Human-induced pluripotent stem cells from blood cells of healthy donors and patients with acquired blood disorders Blood 1145473–5480. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Choi SM, Liu H, Chaudhari P, Kim Y, Cheng L, Feng J.et al. (2011Reprogramming of EBV-immortalized B-lymphocyte cell lines into induced pluripotent stem cells Blood 1181801–1805. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Ye Z, Liu CF, Lanikova L, Dowey SN, He C, Huang X.et al. (2014Differential sensitivity to JAK inhibitory drugs by isogenic human erythroblasts and hematopoietic progenitors generated from patient-specific induced pluripotent stem cells Stem Cells 32269–278. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Ding Q, Regan SN, Xia Y, Oostrom LA, Cowan CA, Musunuru K. Enhanced efficiency of human pluripotent stem cell genome editing through replacing TALENs with CRISPRs. Cell Stem Cell. 2013;12:393–394. doi: 10.1016/j.stem.2013.03.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Djebali S, Davis CA, Merkel A, Dobin A, Lassmann T, Mortazavi A.et al. (2012Landscape of transcription in human cells Nature 489101–108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Soldner F, Laganière J, Cheng AW, Hockemeyer D, Gao Q, Alagappan R.et al. (2011Generation of isogenic pluripotent stem cells differing exclusively at two early onset Parkinson point mutations Cell 146318–331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Ramirez CL, Certo MT, Mussolino C, Goodwin MJ, Cradick TJ, McCaffrey AP.et al. (2012Engineered zinc finger nickases induce homology-directed repair with reduced mutagenic effects Nucleic Acids Res 405560–5568. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Ran FA, Hsu PD, Lin CY, Gootenberg JS, Konermann S, Trevino AE.et al. (2013Double nicking by RNA-guided CRISPR Cas9 for enhanced genome editing specificity Cell 1541380–1389. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Shen B, Zhang W, Zhang J, Zhou J, Wang J, Chen L.et al. (2014Efficient genome modification by CRISPR-Cas9 nickase with minimal off-target effects Nat Methods 11399–402. [DOI] [PubMed] [Google Scholar]
  41. Maresca M, Lin VG, Guo N, Yang Y. Obligate ligation-gated recombination (ObLiGaRe): custom-designed nuclease-mediated targeted integration through nonhomologous end joining. Genome Res. 2013;23:539–546. doi: 10.1101/gr.145441.112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Hendel A, Kildebeck EJ, Fine EJ, Clark JT, Punjya N, Sebastiano V.et al. (2014Quantifying genome-editing outcomes at endogenous loci with SMRT sequencing Cell Rep 7293–305. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Chen G, Gulbranson DR, Hou Z, Bolin JM, Ruotti V, Probasco MD.et al. (2011Chemically defined conditions for human iPSC derivation and culture Nat Methods 8424–429. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Wang Y, Chou BK, Dowey S, He C, Gerecht S, Cheng L. Scalable expansion of human induced pluripotent stem cells in the defined xeno-free E8 medium under adherent and suspension culture conditions. Stem Cell Res. 2013;11:1103–1116. doi: 10.1016/j.scr.2013.07.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Reyon D, Khayter C, Regan MR, Joung JK, Sander JD. Engineering designer transcription activator-like effector nucleases (TALENs) by REAL or REAL-Fast assembly. Curr Protoc Mol Biol. 2012;Chapter 12:Unit 12.15. doi: 10.1002/0471142727.mb1215s100. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Figures and Tables
Click here for additional data file. (403.6KB, zip)

Articles from Molecular Therapy are provided here courtesy of The American Society of Gene & Cell Therapy

RESOURCES