To avoid the ethical issues of embryonic stem cells, genome engineering has focused on inducible pluripotent stem cells, which can develop into all 3 germ layers. The ability to detect methylation patterns in these cells allows research into pluripotency markers. The recently developed CRISPR system has allowed widespread application of genome engineering techniques. The CRISPR-Cas9 system, a potent system for genome editing, can be used for gene knockout or knock-in genome manipulations through substitution of a target genetic sequence with a desired donor sequence. Two types of genome engineering can be initiated: homologous or nonhomologous DNA repair by the Cas9 nuclease. Delivery of the CRISPR-Cas9 and target donor vectors in human pluripotent stem cells can be accomplished via viral and nonviral delivery methods. Nonviral delivery includes lipid-mediated transfection and electroporation. It has become the most common and efficient in vitro delivery method for human pluripotent stem cells. The CRISPR-Cas9 system can be combined with inducible pluripotent stem cells to generate single or multiple gene knockouts, correct mutations, or insert reporter transgenes. Knockouts can also be utilized to investigate epigenetic roles and targets, such as investigation of DNA methylation. CRISPR could be combined with human pluripotent stem cells to explore genetic determinants of lineage choice, differentiation, and stem cell fate, allowing investigators to study how various genes or noncoding elements contribute to specific processes and pathways. The CRISPR-Cas9 system can also be used to create null or nuclease-dead Cas9, which has no enzymatic activity but has been utilized through fusion with other functional protein domains. In conclusion, RNA-guided genome targeting will have broad implications for synthetic biology, direct perturbation of gene networks, and targeted ex vivo and in vivo gene therapy.
Key words : CRISPR-Cas9 system, Gene therapy, Inducible pluripotent stem cells
Genome engineering focuses on gene editing by gene insertion, deletion, or DNA replacement in a specific cell or cells. Identifying the target DNA sequence of interest is necessary to correct the injured gene, to produce antigens, to manipulate cytokine and tumor suppressor genes, to stimulate or suppress growth factor genes, to replicate inhibitor genes, and to produce suicide genes. In studies of genetic disorders, researchers have focussed on how damaged DNA could be replaced by new healthy DNA. In this regard, the ability to cleave the specific region of DNA by nuclease factors for DNA breaks is needed.
With homologous recombination, nucleotide sequences are exchanged between 2 similar or identical molecules of DNA, and the use of nuclease factors to find DNA breaks can be done by 3 types of nucleases1: (1) zinc finger nucleases (ZFNs), which comprise the first generation of programmable nucleases, (2) transcription activator-like effector nucleases (TALENs), and (3) second-generation programmable nucleases and third-generation pro-grammable nucleases obtained through the CRISPR-Cas9 system. The 3 types of nucleases, ZFN, TALEN, and CRISPR-Cas9-generated nucleases, consist of sequence-specific DNA-binding domains attached to nonspecific DNA nucleases. The ZFNs consist of a zinc finger DNA-binding domain and a nuclease domain derived from the restriction enzyme FokI. The TALENs also contain the same FokI nuclease domain but use a different DNA-binding domain derived from transcription activator-like effectors.
The CRISPR-Cas9 system-generated nucleases are composed of a DNA recognition part of a single-guide RNA and a nuclease (Cas9). The nuclease (FokI or Cas9) cuts the target DNA sequence and generates a double-strand break (DSB) that is repaired by either nonhomologous end-joining (NHEJ) repair or homology-directed repair (HDR). With the NHEJ pathway, random insertions or deletions occur at the site of editing, often leading to gene disruption. A donor template that includes sequences homologous to the DSB flanking regions can be incorporated into the genomic DNA, resulting in gene correction. The emerging technology of genome editing, also known as genome engineering, seeks to meet this need by providing the ability to more efficiently introduce a variety of genetic alterations, ranging from single-nucleotide modi-fications to whole gene addition or deletion, all with a high degree of target specificity.2 The key features of the most widely used genome-editing systems, in addition to the major advantages and disadvantages of each, are described in Table 1.
In the first stage, engineered endonucleases are used to make DNA DSBs at the speciﬁc sites in the genome. As a consequence, intrinsic cellular DNA damage repair pathways are activated to repair the DSBs and to sometimes introduce insertions and/or deletions at the target sites. The ZFNs, TALENs, and clustered regularly interspaced short palindromic repeats (CRISPR)-CRISPR-associated protein (Cas) systems have a high degree of targeting ﬂexibility and can efﬁciently make DSBs in targeted genome loci. Regardless, there are signiﬁcant differences among these 3 systems.
The ZFNs and TALENs are engineered proteins that target speciﬁc DNA sequences through protein-DNA interactions. As a result, ZFNs and TALENs are usually difﬁcult to design and construct. In contrast, the CRISPR-Cas system relies on a simple design and synthesis of an RNA molecule complementary to the DNA target sites. Of the 2 main mechanisms for DSB repair, repair with NHEJ does not require donor DNA, and often insertions or deletions are introduced at the target sites. Repair with HDR requires a donor DNA to facilitate precise genome modiﬁcation at the cleavage sites.3 The predominant and error-prone NHEJ pathway often results in small nucleotide insertions or deletions that can be used to construct knockout alleles. Alternatively, HDR activity can result in precise modifications incorporating exogenous DNA fragments into the cut site. However, genetic recombination in mammalian systems through the HDR pathway is an inefficient process and requires cumbersome laboratory methods to identify the desired accurate insertion events. This is further compromised by the activity of the competing DNA repair pathway, NHEJ, which repairs most nuclease-induced DNA DSBs and is also responsible for mutagenic insertion and deletion events at off-target locations throughout the genome. Various methodologies have been developed to increase the efficiency of designer nuclease-based HDR-mediated gene editing.4
Repurposing of engineered nucleases can also enable inactivation of their endonuclease activity while retaining the DNA-binding capabilities. When fused with functional moieties (eg, epigenetic modiﬁcation enzymes), these tools can be harnessed for epigenetic modiﬁcation at target genome sites. As another example, when fused with a base editor, these tools can convert a speciﬁc nucleotide to another without introducing a DSB.
The cleavage enzymatic systems have gained widespread popularity in applications ranging from gene editing, transcriptional and epigenetic mani-pulations, agricultural science, human embryo editing, and disease therapies. Therapeutic effects of Cas9 for genome engineering include genome repair for genetically mutated genes, genome disruption for invading pathogen genomes, genome repression for oncogenic pathogens, and tumor genome activation in the case of tumor suppression genes.5 The latter will be discussed in more detail regarding their application in stem cells.
Stem cells in genome engineering
The combination of genome editing with stem cells is promising for medical sciences by creating similar diseases in animals and developing new therapeutic models. It might be possible to engineer pluripotent stem cell-derived grafts, with the usual caveats concerning activated oncogenes so that they would be immunologically inert and identifiable by an array of imaging strategies. The target stem cells could be embryonic stem cells (ESC) or adult stem cells. The latter cells are mostly induced pluripotent stem cells (iPSC) and rarely hematopoietic stem cells (HSCs) (Table 2).
Genome editing has been used to generate or correct stem cell-based disease models. It allows a disease-causing target gene mutation to be inserted into healthy control stem cells to produce diseased stem cells for research or vice versa to allow correction of the diseased genome for treatment of a congenital disease. One of the most promising applications of somatic cell reprogramming is the production of customized pluripotent stem cells followed by gene correction, differentiation into adult stem cells, and autotransplant with intent to cure any one of the many different inherited genetic disorders of the blood-forming system.
Shinya Yamanaka of Kyoto University6 and James Thomson of the University of Wisconsin-Madison7 both published papers on their separate discoveries of iPSCs. These pluripotent cells were created from skin cells that had 4 genes inserted into them with viruses. This procedure resulted in the skin cells acquiring properties similar to ESCs. These iPSC were reprogrammed with 4 transcription factors (OCT4, SOX2, KLF4, and MYC) from somatic cells.8 These reprogrammed cells could be differentiated into a variety of cells of almost any type. Embryonic stem cells are derived from the early-stage preimplantation embryo and can also be differentiated into almost any type of cell, and HSCs can be derived from bone marrow and can be differentiated into all types of blood cells.
Successful cell transplant requires optimization of the best cell type and site for engraftment, prevention of limitations regarding cell migration, and mini-mization of immunologic reactivity. Either in vivo or ex vivo models can be used for genome editing (see illustration in Clement and associates9). With the in vivo approach, or direct delivery-based in vivo editing, the CRISPR-Cas9 system components are directly delivered into the patient using either viral or nonviral vectors for in situ gene editing.9,10 With the ex vivo approach, genes are edited in patient-derived cells in vitro. Once either the iPSCs are reprogrammed or the somatic stem/progenitor cells are expanded, they are transplanted back into the same patient after correction of the target genome. Although the second approach could select and analyze the target cells more precisely, it requires cell expansion in culture, which can lead to additional unwanted genomic alterations.
Role of plasmid transport in genome engineering
Plasmid is a small DNA molecule within a cell that is separated from a chromosomal DNA and can replicate independently. Plasmids are widely used as vectors in molecular cloning, serving to drive the replication of recombinant DNA sequences within host organisms. It can transfer information from cell to cell. There are 2 types of plasmids that can be used within host bacteria: (1) nonintegrated plasmid replication, which is separate from bacterial DNA and could be expressed transiently and (2) episomes, which can integrate into the host chromosome and could then be permanently expressed on the DNA.
There are currently 3 methods in which the necessary components can be introduced into the cell11: (1) transient transfection of 2 plasmids, one containing the coding sequence for Cas9 and the other expressing the guide RNA, (2) viral integration of the Cas9 coding sequence and transient expression of the guide RNA, and (3) viral integration of both Cas9 and guide RNA sequences. Genome-integrating viral vectors are mostly highly efficient and allow long-term transgene expression. However, there are risks associated with viral vector-based methods (ie, mutagenesis, aberrant transgene expression, and immunogenicity). With nonintegrated viruses, although highly efficient in transduction and safer, they offer only a limited duration of transgene expression.
The application of viral vectors for gene delivery is based on their natural ability to infect cells. Their use as a relatively safe molecular biology tool was made possible by modification of the viral genome by deletion of some critical coding sequences to prevent spontaneous replication in target cells.12 Currently, viruses are commonly used as vehicles in both in vitro and in vivo delivery paradigms. Retroviruses, lentiviruses, and adeno-associated viruses are the 3 main integrating virus types that are used for transduction of mammalian cells for long-term transgene expression. Each virus has different characteristics with respect to their tendency to transfect DNA or mRNA and their efficiency in different types of stem cells.
Nonviral-based vectors for gene manipulation can also transfect stem cells by plasmid and mRNA without the necessary viral particles and with different efficiencies (Table 3). Methods for this type of gene manipulation could be chemically or physically based. Chemical-based methods use calcium phosphate, cationic lipids, cationic polymers, cationic peptides, cationic polysaccharides, or inorganic nanomaterials. Physical techniques used to transfer plasmid and associated nucleases include sono-transfection (which uses ultrasonographic frequencies), electroporation (which uses electrical currents), microporation targeting of any individual cell, nucleofection (which transports the complex directly to the nuclei), electromagnetic poration (which uses the electromagnetic field), and microinjection (manually injects the complex into the nuclei). Although low-immunogenicity and unlimited sizes of transgenes could be transmitted, these methods have the disadvantage of a relatively low efficiency compared with viral transduction. Established nonviral methods of transfection of plasmid delivery of nucleic acids to human pluripotent stem cells are limited using standard lipofection reagents; therefore, most studies utilize electroporation or nucleofection.10
Clinical applications of engineered stem cells
Transplantation of gene-edited autologous cells to treat monogenic diseases is a multistep process. Somatic cells are harvested and converted into iPSCs, which are treated with targeted nucleases to correct the mutation, differentiated into the relevant cell type, and returned to the patient.13 The explosion of preclinical and clinical studies in this field suggests that CRISPR-Cas9 (and older ZFNs and TALENs) will become the dominant genome-editing platforms for cell therapy (Table 4).
Several mutation types have been corrected in beta thalassemia patient-derived iPSCs using wild-type or humanized SpCas9. Sickle cell anemia, another beta-globin disease, is caused by a homozygous missense point mutation in the HBB gene. CRISPR-Cas9 has corrected the mutation in one allele in patient iPSCs and detected beta-globin expression in erythrocytes differentiated from the corrected cell.14 The targeted iPSC clones can find the corrected one and also disrupt the sickle cell alleles through erythroid-differentiation assays. Erythrocytes from either the corrected or its parental (uncorrected) iPSC line have been generated with similar efficiencies. Up to 10% of differentiated erythrocytes have been shown to lack nuclei, which is characteristic of further matured erythrocytes called reticulocytes. A new approach to treat both beta thalassemia and sickle cell anemia is to upregulate expression of gamma globin, which is normally silenced after birth. This approach for erythroid precursor cells has been used by CRISPER-CAS9 to knock out expression of a transcriptional regulator that is responsible for gamma globin silencing.15
Genome editing with the use of CRISPR-Cas9 has also been applied to iPSCs from patients with other hematologic diseases, such as hemophilia or poly-cythemia vera. Corrected iPSCs could rescue hemophilia by switching of endothelial cells to produce factor 8.
Correction of the cystic fibrosis gene has also been conducted with the CRISPER-Cas9 system.16 Firth and associates used the CRISPR approach to correct the mutated gene with efficiency of 16.7%, which corrected the epithelium airway. Exon knock-in was an effective method for Duchenne disease, restoring expression of full-length dystrophin proteins.
There are many other disorders in which patient iPSCs have been corrected by CRISPR-Cas9, including chronic granulomatous disease, limb girdle muscular dystrophy, alpha-1-antitrypsin deficiency, retinitis pigmentosa, Alzheimer disease, and congenital hearing loss.17 For the latter disorder, the authors showed that cationic lipid-mediated in vivo delivery of Cas9-guide RNA complexes can ameliorate hearing loss in a mouse model of human genetic deafness. Disruption of the CCR5 genome in iPSCs by CRISPR-Cas9 has been used successfully to generate human immunodeficiency virus-resistant cells.18 Mismatch of human leukocyte antigens (HLA) adversely impacts outcomes of patients after allogeneic HSC transplant. Elimination of HLA-A expression in HSC was achieved using artificial ZFN designed to target HLA-A alleles.19 These engineered HSCs maintained their ability to engraft and reconstitute hematopoiesis in immunocompromised mice. This introduced loss of HLA-A expression, decreasing the need to recruit large numbers of donors to match with potential recipients, with particular importance in patients whose HLA repertoire is underrepresented in the current donor pool. Furthermore, genetic engineering of stem cells has provided a translational approach to HLA matching of a limited number of third-party donors with a wide number of recipients.
Transgene strategies with stem cells have been used for tumor therapy. In one study, CD34-positive human umbilical cord blood stem cells (UCBSCs) were engineered to express interleukin 21 (IL-21) and then were transplanted into A2780 ovarian cancer xenograft-bearing Balb/c nude mice.20 The therapeutic efficacy of this procedure on ovarian cancer indicated that UCBSCs did not form gross or histologic teratomas until up to 70 days postinjection. The CD34-positive UCBSC-IL-21 therapy showed a consistent effect in treatment of mice with ovarian cancer, delaying tumor appearance, reducing tumor size, and extending life expectancy. The efficacy was attributable to keeping CD34-positive UCBSC-IL-21 in neoplastic tissues for more than 21 days. The secreted IL-21 not only increased the quantity of CD11a-positive and CD56-positive natural killer cells but also increased natural killer cell cytotoxicities to YAC-1 cells and A2780 cells. The efficacy was also associated with enhanced levels of interferon-gamma, interleukin 4, and tumor necrosis factor alpha in mice and high expression levels of NKG2D and MIC A/B molecules in tumor tissue. Another hypothetical effect of antitumor efficacy of edited iPSC is delivery of proapoptotic proteins via stem cells, allowing engineered HSCs to express tumor antigens and produce a suicide gene.
Human mesenchymal stem cells have also been engineered to release interferon-beta against glioma.21 In this study, human bone marrow-derived mesen-chymal stem cells were applied to treat glioma successfully. Recently, partial reprogramming of iPSCs derived from other organs and changed into other organs was achieved.22 With this method, it is possible to switch cardiomyocyte stem cells directly into neurons.
The ability to pass pluripotency and directly reprogram readily accessible human tissues, such as skin, into neural cells, offers a fast and efficient approach to study neurologic disorders.23 Although direct neuronal conversion may offer unique benefits, this approach is currently limited to a small number of protocols to specify neuronal subtypes using postnatal or adult human samples. Advantages of this direct method for iPSC engineering is predicted to be an unlaborious process, has short turnaround time and cost, eliminates the possibility of teratoma formation, and permits a smaller subset of somatic cell types necessary for this procedure.
The construction of a cell model of disease in vitro is an urgent task and is the key to discovering the pathogenesis of the disease. Autosomal dominant polycystic kidney disease (ADPKD) is a common life-threatening inherited renal disorder, characterized by the progressive formation of renal cysts and resulting in severe destruction of normal renal parenchyma and eventually leading to renal failure. Genetic defects in 2 genes named PKD1 (polycystin-1; PC1) and PKD2 (polycystin-2; PC2) are associated with ADPKD. Mutations of these 2 PKD genes account for approximately 91% of the pathogenesis of the disease. Because of an absence of credible human cell models, the pathogenesis of ADPKD has not been thoroughly investigated. In one study,24 iPSCs from ADPKD patients without PKD1/PKD2 mutations were generated. These iPSCs were indistinguishable from human ESCs with respect to colony morphology, passaging, surface and pluripotent markers, normal karyotype, DNA methylation, and differentiation potential. They also differentiated to ADPKD directly. Results from the study24 revealed that special ADPKD-iPSCs without PKD1/PKD2 gene mutations can be generated and induced to differentiate into functional kidney-like cells using a modified differentiation protocol.
In the case of α1-antitrypsin deficiency, a point mutation in α1-antitrypsin was corrected. The corrected human iPSCs were differentiated into hepatocyte-like cells and transplanted into mice, thereby restoring the structure and function of α1-antitrypsin by secreted human albumin.25
Genome editing by CRISPR-Cas9 technology has also been applied to evaluate colon cancer by intestinal stem cells. Briefly, a colorectal cancer model was created with intestinal stem cells by mutating the cancer-related genes.26 The use of this model allowed the researchers to analyze the involvement of each gene in colon cancer pathways.
An engineered genome for cardiovascular diseases has also shown benefits.27 These techniques can improve the isolation, selection, and differentiation of stem cells before implantation and can allow the development of strategies to promote the retention, mobilization, survival, integration, and tracking of stem cells after implantation. In one study, human MSCs genetically engineered to express myocardin, a dynamic cardiomyogenic transcription factor, attained a cardiac phenotype with an efficiency of 90% to 100% when implanted into an infarcted mouse heart. As a result, left ventricular function was increased and detrimental ventricular remodeling was reduced compared with the use of control nonengineered MSCs.28
Stem cell transplant for production of beta cells may represent an unlimited source of insulin-producing cells.29 Although pancreas transplant and the infusion of cadaveric islets are currently implemented clinically, the procedures have many adverse effects, including over-immunosuppression over the lifetime of the recipient. There are also limited numbers of donors. A diabetic, immuno-competent animal model was corrected by using human stem cell-derived beta cells. These cells were encapsulated with alginate derivatives capable of mitigating foreign body responses in vivo and implanted into the intraperitoneal space.30
Genome engineering with stem cells translated to the clinic has recently had a significant upsurge as a novel therapy. The discovery and characterization of pro-grammable nucleases using the ZFN, TALEN, and CRISPR-Cas9 systems have shown great promise for the development of therapeutic strategies to treat human genetic diseases. These technologies have unique opportunities for significant impact when used together with stem cells. In particular, various iPSCs derived from patients with different genetic diseases have been corrected with genome editing, providing hope to many patients. The CRISPR-Cas9 system has advantages over ZFNs and TALENs given that Cas9 is extremely easy to use and has many variants that can be used in a variety of medical research areas. Despite the many advantages of this system, there are several reports indicating that CRISPR-Cas9 had high off-target effects, limiting its use in clinical settings. Many obstacles should be resolved before the use of genome engineering as routine therapy, such as recurrent autoimmunity, which would require immunosuppression for some diseases, and resolving the optimized site for delivery of engineered iPSC and the method for transplant of the engineered cells. Because of safety and ethical concerns, CRISPR-Cas9-based genome editing studies in human zygotes are not legally approved. The resolution of all of these issues through close collaboration between scientists, clinicians, and industry is critical.
Volume : 17
Issue : 1
Pages : 31 - 37
DOI : 10.6002/ect.MESOT2018.L34
From the Shahidbeheshti University of Medical Sciences, Tehran, Iran
Acknowledgements: The author has no sources of funding for this study and has no conflicts of interest to declare.
Corresponding author: Hassan Argani, Shahidbeheshti University of Medical Sciences, Tehran, Iran
Phone: +91 20303274
Table 1. Nucleases Used for Gene Editing Systems
Table 2. Types of Stem Cells Used for Genome Engineering
Table 3. Nonviral-Based Vectors for Genetic Manipulation That Can Also Transfect Stem Cells by Plasmid and mRNA
Table 4.Mutations in Patient-Derived Cells That Have Been Corrected Using CRISPR-Cas9