Sei sulla pagina 1di 8
Biotechnology Advances 31 (2013) 994 – 1001 Contents lists available at ScienceDirect Biotechnology Advances

Contents lists available at ScienceDirect

Biotechnology Advances

journal homepage: www.elsevier.com/locate/biotechadv

journal homepage: www.elsevier.com/locate/biotechadv Research review paper Advances in genetic modi fi cation

Research review paper

Advances in genetic modication of pluripotent stem cells

Andrew Fontes, Uma Lakshmipathy

Primary and Stem Cell Systems, Life Technologies, 5781 Van Allen Way, Carlsbad, CA 92008, USA

Technologies, 5781 Van Allen Way, Carlsbad, CA 92008, USA article info Available online 12 July 2013

article info

Available online 12 July 2013

Keywords:

Stem cells Embryonic stem cells Induced pluripotent stem cells Genetic modi cation

abstract

Genetically engineered stem cells aid in dissecting basic cell function and are valuable tools for drug discovery, in vivo cell tracking, and gene therapy. Gene transfer into pluripotent stem cells has been a challenge due to their intrinsic feature of growing in clusters and hence not amenable to common gene delivery methods. Several advances have been made in the rapid assembly of DNA elements, optimization of culture conditions, and DNA delivery methods. This has lead to the development of viral and non-viral methods for transient or stable modication of cells, albeit with varying efciencies. Most methods require selection and clonal expansion that demand prolonged culture and are not suited for cells with limited proliferative potential. Choosing the right platform based on preferred length, strength, and context of transgene expression is a critical step. Random integration of the transgene into the genome can be complicated due to silencing or altered regu- lation of expression due to genomic effects. An alternative to this are site-specic methods that target transgenes followed by screening to identify the genomic loci that support long-term expression with stem cell proliferation and differentiation. A highly precise and accurate editing of the genome driven by homology can be achieved using traditional methods as well as the newer technologies such as zinc nger nuclease, TAL effector nucleases and CRISPR. In this review, we summarize the different genetic engineering methods that have been successfully used to create modied embryonic and induced pluripotent stem cells. © 2013 Elsevier Inc. All rights reserved.

Contents

Introduction

1. .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

995

2. Gene insertion

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

996

2.1. Random genomic integration

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

996

2.1.1.

Naked plasmid

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

996

2.1.2. Lentivirus .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

996

2.1.3. Transposon .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

997

2.2. Non-integrating technology

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

997

2.2.1. Episomal vectors

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

997

2.2.2. Adenovirus .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

997

2.2.3. Minicircle .

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

997

2.2.4. BacMam

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

997

2.3. Site-speci c genomic integration

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

998

2.3.1. Adeno-associate virus (AAV)

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

998

2.3.2. PhiC31 integrase

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

998

3. Gene targeting and editing

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

998

3.1. Homologous recombination

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

998

3.2. Zinc nger nucleases

 

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

.

999

3.3. TAL effectors

1000

3.4. Crispr/CAS

1000

References

1000

Corresponding author at: Primary and Stem Cell Systems, 5781 Van Allen Way, Carlsbad, CA 92008, USA. Tel.: +1 760 268 7465. E-mail address: uma.lakshmipathy@lifetech.com (U. Lakshmipathy). URL: http://www.lifetechnologies.com (U. Lakshmipathy).

0734-9750/$ see front matter © 2013 Elsevier Inc. All rights reserved.

A. Fontes, U. Lakshmipathy / Biotechnology Advances 31 (2013) 9941001

995

1. Introduction

The ability of pluripotent stem cells to indenitely proliferate in culture and differentiate into multiple cell types under the right cues provides an ideal source for genetic modication for various down- stream applications. This enables scientists to dissect basic biology and explore the potential use of pluripotent cells in regenerative medi- cine and drug discovery. However, a key challenge lies in identifying the ideal platform suited for the intended application. Various viral and non-viral platforms have been utilized for expression of exogenous genes and for targeting endogenous DNA in human embryonic stem cells (hESC) and induced pluripotent stem cells (iPSC), albeit with varying efciencies. Methods that have been widely used for murine embryonic stem cells (mESC) have largely been suboptimal for hESC thus necessitating further optimization of culture conditions (Braam et al., 2008). The fundamental unit for gene delivery into cells is a plasmid DNA or vector carrying the transgenes of choice. The primary architecture of such plasmid DNA comprises a transgene of interest driven by a promoter of choice. Promoter choice is dependent on the type of expression needed. Constitutive promoter is expressed in all cell types, an inducible promoter can be activated or inactivated in the

presence of small molecules, and a lineage specic promoter is active in specic cell types. Traditional restriction endonuclease-mediated cloning processes are rapidly being replaced by recombination- mediated cloning methods such as Multisite Gateway® or Lego that enable speedy assembly of multiple DNA fragments. The base plasmid also carries a drug resistance gene which can be utilized for screening cells that harbor the plasmid and a bacterial origin of replication and an antibiotic resistance gene important for propagating the plasmid in bacteria. An additional factor that is critical for successful gene modication is efcient gene delivery methods to introduce the DNA fragments into pluripotent stem cells. Chemical-based reagents such as Lipofectamine 2000, Fugene HD, Gene Jammer, etc. offer the advantage of direct addi- tion to the culture media but suffer from poor efciencies. Advances to traditional electroporation devices such as Amaxa Nucleofector and Neon electroporation system have allowed higher efciency of transfec- tion of hESC (Lakshmipathy et al., 2004; Liu et al., 2009). In cases where chemical and electroporation methods pose a challenge, viral delivery systems have been utilized. A modied Lentivirus system has been reported to achieve rapid generation of stable clones in both murine as well as human ESC (Suter et al., 2006). The combination of modular cloning methods, optimal vector design and efcient gene delivery

optimal vector design and ef fi cient gene delivery Fig. 1. (A) High throughput ef fi

Fig. 1. (A) High throughput ef cient cloning systems that can rapidly assemble plasmids of choice and (B) deliver to cells via ef cient nontoxic methods, are critical for successful gene modi cation for gene insertion or targeted gene editing of stem cells.

996

A. Fontes, U. Lakshmipathy / Biotechnology Advances 31 (2013) 9941001

into target cells is critical for modifying hard-to-transfect ESCs (Fig. 1). In addition, choices of culture conditions and media systems are equally important factors. Although ESCs and iPSCs can be cultured under feeder-free media systems prior to transfection, recovery and selection of feeders has been the most traditional method used to maintain and clone out the genetically modied stem cells. Gene modication of cells can be broadly classied into two major categories based on application, (1) gene insertion and (2) gene targeting. Gene insertion methods are most widely used to deliver DNA fragments such as cDNA or shRN A for overexpression or knock- down of speci c genes, respectively. The platform of choice is large- ly dependent on length and strength of expression required. The wide range of gene modi cation approaches offers unique advan- tages that can be utilized to robustly express exogenous genes at levels signicant enough to alter cellular function. Here, we review the various methods that have been successfully utilized to alter hESCs and iPSCs.

2. Gene insertion

2.1. Random genomic integration

Randomly integrating technologies enable users to create stable systems leading to lasting expression with or without the use of anti- biotic selection. These platforms result in the random insertion of se- lected DNA fragments into the host genome without the use of DNA homology. Random genomic integration provides a valuable tool for long-term expression in human ES cells despite their rapid and in - nite dividing capabilities. However, consequences of random inser- tion leading to variable copy number per cell are inconsistent integration sites and unpredictable expression patterns. In addition, the locus of insertion can result in partial or complete silencing in human ES cells, which can occur during routine culture and maintenance as well as throughout differentiation. A major risk with these methods

is insertional mutagenesis resulting in genome instability and toxicity (Baum et al., 2006).

2.1.1. Naked plasmid

Randomly integrating platforms have progressed with the optimiza- tion of transfection system for ES cells. Using solely plasmid DNA pro- vides a relatively simple integrating engineering platform since they require no additional recombination or preparation prior to transfec- tion. As an integrating system, plasmids provide stable expression of complex cassettes with or without selection. The use of plasmid for random integration has the potential to be used in ES cells for a variety of applications including the overexpression of exogenous cassettes (Wobus and Boheler, 2005). In human ES cells, plasmid insertion re- mains limited by efcient electroporation methods, which will progres- sively decrease in efciency with increased vector size (Moore et al., 2010). Furthermore, the recovery of colonies after single-cell transfec- tion remains a complex factor for ES cell transfection. Despite these hur- dles, scientists have been able to identify locations within the genome where limited silencing occurs using random integration of plasmid DNA (Costa et al., 2005). One site, deemed the Envy locus, has shown to remain un-silenced by expression of GFP using a constitutive pro- moter at all levels of differentiation in the absence of selection in mouse ES cells. Although cumbersome to identify, this site offers a plat- form for high expression of reporters or complex gene cassettes.

2.1.2. Lentivirus

To circumvent electroporation roadblocks, viral platforms offer an attractive alternative to plasmid transfection. Recombinant Lentivirus for gene delivery is replication incompetent with the ability to infect a wide variety of cell types including human embryonic stem cells. The mechanism of entry requires attachment to common glycoprotein on the cell exterior to facilitate cell membrane fusion. Lentivirus has been used as a gene modication tool in human ES cells since Dr. James Thomson's lab showed expression of GFP in ESC with minimal silencing of upon differentiation to hematopoietic precursors (Ma et al., 2003).

to hematopoietic precursors ( Ma et al., 2003 ). Fig. 2. Integrational methods are most commonly

Fig. 2. Integrational methods are most commonly used for the creation of transient or stably modi ed human ESC. (A) Lentiviruses can be used to create PGK-GFP H9 ESC where single or multiple copies of this gene can randomly integrate into any part of the host chromosome. (B) Episomal vectors, a non integrating system with the plasmids maintained extra chromosomally, were used to create a stable H9 ESC dual reporter expressing GFP and Tag RFP both driven by EF1a promoters. Site-speci c integration method using PhiC31 integrase was used to insert Oct4-eGFP into a placed target site on chromosome 13. (C) Site-speci c integration method using PhiC31 integrase was used to insert Oct4-eGFP into a placed target site on chromosome 13.

A. Fontes, U. Lakshmipathy / Biotechnology Advances 31 (2013) 9941001

997

With improved ease of use, Lentivirus typically requires only one over- night transfection. This allows for minimal manipulation of ESCs grown in adherent cell culture systems and cells cultured on feeder layers. However, murine embryonic broblast feeder layers show more robust transduction compared to the variable transduction and expression observed within the ESC colony (Fig. 2A). With the ability to transfect dividing or non-dividing cells, Lentivirus provides a tool for pluripotent cells as well as differentiation inter- mediates and expression of markers by lineage speci c promoters. Overexpression using Lentivirus can provide a selection tool for pluripotency as well as an editing tool, as shown with microRNAs intro- duced into ESC via Lentivirus, to identify and monitor the repression and expression of key pluripotency genes (Xu et al., 2009). Along with high transfection efciency, there is an increased risk of insertional mutagen- esis and multiple proviral integrations that can cause alternative splicing and aberrant transcripts (Moiani et al., 2012). Additionally, differential expression can result due to variable copy number without selecting for individual clones. Lentivirus load capacity is relatively small with decreased transfection efciency as well as recombination in direct correlation with higher payload.

2.1.3. Transposon

Non-viral vectors, such as Sleeping Beauty (SB) transposon, achieve efcient delivery and integration into the host genome. The SB transpo- son system is comprised of a catalytic component and a cargo-containing component. The cargo-containing component is anked by inverted terminal repeats that encode non-identical direct repeats that are recog- nized by transposons to facilitate transposition (Ivics et al., 1997). When this construct is co-delivered with a construct expressing SB transposon, the cargo containing the gene of interest anked by the direct repeats is excised from the donor plasmid and inserted randomly into the host genome. This method has been used to generate modied human embryonic stem cells (Wilber et al., 2007) and modied ESCs used to fur- ther study differentiation (Orban et al., 2009). Although the transposon systems such as Sleeping Beauty, Tol2 and PiggyBac result in random integration (Huang et al., 2010), it is thought to less likely integrate into transcribed genes or regulatory regions observed with viral systems (Mitchell et al., 2004).

2.2. Non-integrating technology

Human embryonic stem cells have the potential to become valuable clinical tools for therapeutics, clinical research, and diagnostics. However, the majority of cell engineering tools utilize integrating platforms that can cause genomic effects removing the possibility of downstream appli- cations. To reassure clinical relevancy it is crucial to have non-integrating engineering platforms available for early research.

2.2.1. Episomal vectors

The EpsteinBarr Nuclear Antigen 1 (EBNA) region has been utilized for stable expression in mammals for a variety of applications for decades (Yates et al., 1985). The trans acting EBNA1 element requires an origin of replication (OriP) for plasmid replication to occur. The vector containing OriP replicates once per cell cycle with the binding of multiple EBNA homodimers to the exogenous OriP. However, in pri- mate and human cells, the stable expression of EBNA expressing vectors previously required the creation of a cell line expressing EBNA. Upon transfection of a plasmid containing the OriP, the vector would replicate with the genomic DNA. In the past ten years, scientists have optimized this combination for stable expression in human ES cells (Ren et al., 2006). Expression by a two-step integrating system, however, negates the use of a non-integrating system for human ES cells, which are prone to continued silencing and differential expression. In 2009, a vector containing both the EBNA and OriP was introduced for the creation of stable expressing cell lines in human ESC (Thyagarajan et al., 2009). Expression remained stable in the presence of antibiotic selection

as well as during random differentiation of embryoid bodies. The vector system offers an efcient method to create reporter lines as shown via GFP expression driven by the POU5F1 (Oct4) promoter (Fig. 2B). These vectors, although limited by the identical electroporation issues detailed above, face additional size constraints. To contain both the EBNA and OriP genes with a selection cassette, the vector backbone is typically above 10 kB in size. For stable expression, the inclusion of a selection cassette is required. Although the plasmid replicates once per cell cycle, the vectors localize in the cytoplasm during cell division. This results in an uneven distribution of plasmid copy number between the daughter cells. Furthermore, uneven distribution can lead to heteroge- neity in exogenous gene expression within the pooled population of cells.

2.2.2. Adenovirus

Adenovirus vectors can transduce both dividing and non-dividing cells but are limited in the payload they can deliver ( Russell, 2000). Gutlessadenoviruses can carry a larger payload, but are cumbersome requiring the co-infection of a helper virus making the subsequent pu- rication processes hard (Alba et al., 2005). Further efciency of trans- duction is dependent on the expression of coxsackie and adenovirus receptor on cells. While this method has been used to achieve efcient gene transfer in mouse embryonic stem cells (Kawabata et al., 2005), in the case of human cells, it has also been successfully used to transduce neural stem cells derived from human ESCs.

2.2.3. Minicircle

Regular plasmid vectors contain bacterial elements such as antibiotic resistance genes and origins of replication. These elements are excised out of the vector via an intramolecular recombination catalyzed by the PhiC31 integrase to generate a minicircle DNA (Jia et al., 2010). The resulting mammalian expression minicircle, which is comprised of pro- moters and genes, has shown to have higher and longer expression but eventually turns over thus providing an ideal non-integrating footprint free system (Chen et al., 2003). This method has been used for the gen- eration of iPSCs from somatic cells (Jia et al., 2010) and for modication

of adult stem cells such as neural stem cells (Madeira et al., 2013), but has limited use for generating gene modied ESCs.

2.2.4. BacMam

Additional viral vectors are utilized in non-integrating gene modica- tion platforms. The baculovirus is a non-integrating DNA insect virus with the ability to infect mammalian and insect cells with high efciency. The virus historically has been utilized for recombinant protein produc- tion (Kost et al., 2005). In mammalian systems, however, the baculovirus does not retain the ability to replicate and become infectious, offering a valuable transient over-expression tool. The incorporation of a mamma- lian promoter or a mammalian expression system within the baculovirus genome denotes the viral platform BacMam (Baculoviral Mammalian expression). The BacMam virus has become a standard gene transfer technique due to its safety and ease of transfection and virtually zero cytotoxic effects on human cells (Gao et al., 2002; Kost et al., 2005). The baculovirus is relatively large compared to typical viral capsid and is a double-stranded DNA virus. Viral entry is proposed to occur via the G64 glycoprotein causing endocytosis and eventual migration to the nucleus. With advances in stem cell research scientists require increasingly more complex expression cassettes. The baculoviral system has the capacity for extremely large DNA cassettes (N 35 kB) providing a exi- ble system for the introduction of complex engineering fragments (Fornwald et al., 2007). Transient expression provides a fast and ef- cient system for the identication of specic lineages during differenti- ation. BacMam can be used to overexpress markers using lineage specic promoters enabling ES cell scientists to create more throughput differentiation protocols. Additionally, baculovirus has been used to overexpress proteins to directly drive differentiation and trigger osteo- genesis (Chuang et al., 2007).

998

A. Fontes, U. Lakshmipathy / Biotechnology Advances 31 (2013) 9941001

2.3. Site-specic genomic integration

Site-specic modication is an ideal method to avoid variable expression patterns and copy number variation as a result of random integration. These tools enable the user to locate specic sites within the chromosome where silencing is minimized. In order to identify spe- cic locations within the genome these platforms rely on methods that have preferential insertion into the mammalian genome. Additionally, selection is recommended to identify single copy insertions of the gene of interest and eliminate false positives. Despite the added selec- tion, these platforms require a screening step for multiple copy variants.

2.3.1. Adeno-associate virus (AAV) AAV allows for site-specic modication utilizing a single-stranded DNA genome. The recombination event for insertion occurs with a rela- tively high efciency (1%), an improvement from solely homologous recombination of plasmid vectors (Hirata et al., 2002). Additional advantages of using AAV include the low level of innate cell immune response from the presence of the virus in humans. This establishes AAV as a more relevant platform for downstream clinical applications compared to other viral models. The AAV virus is limited in payload capacity, typically carrying no more than a 1 kB cassette. Payload chal- lenge is increasingly more relevant due to the need for more complex cassettes containing polycistronic sets or enhancing elements. With the increased recombination efciency compared to DNA targeting site identication, AAV offers a relatively efcient gene editing tool with less cumbersome assembly required by Zinc Finer Nucleases (ZFN). In pluripotent stem cells, AAV vectors have shown the ability to modify genes for the creation and correction of mutations at the HPRT loci (Khan et al., 2010).

2.3.2. PhiC31 integrase Additional site-specic technologies rely on integrase-mediated recombination. PhiC31 integrase is a bacteriophage integrase that nds native attachment P(attP) sites within bacteria. Human cells contain a number of pseudo attP sites which allow for localized recom- bination into a specic attP site in human cells. Although there are a variety of pseudo attP sites within the genome there are specic hotspots with a higher likelihood of recombination. The location identied on chromosome 13 is a known intronic region unaffected by chromatin remodeling during differentiation (Chalberg et al., 2006; Thyagarajan et al., 2008). Targeting the chromosome 13 locus results in a similar expression and characteristics of the ROSA locus (Irion et al., 2007). Although the site has the capability to accept a high payload, the limiting factor remains to be the electroporation of large DNA vectors. As men- tioned above, the most efcient method for transfecting DNA into ES cells remains to be electroporation. This requires single-cell suspension and clone generation that can cause genomic defects such as karyotypic abnormalities when not kept in the correct conditions. For this reason, additional means of insertion for the PhiC31 integrase, such as viral, have been explored. The integrase is introduced into human ESCs with a plasmid DNA cotransfected with a target plasmid carrying the gene of interest. PhiC31 integrases mediate the insertion of the target plasmid into pseudo sites in the mammalian genome. The resulting drug selected stable clones show robust expression and are less prone to silencing with extended culture or with differentiation (Fig. 2C). One of the PhiC31 hotspots in hESC is located on chromosome 13 and genes inserted into this site show sustained expression. This feature has been utilized to place a R4 integrase target site to create a target ESC line that can be rapidly retargeted to generate modied ESC expressing genes of interest (Liu et al., 2009; Macarthur et al., 2012; Thyagarajan et al., 2008). Active expression at a single genomic locus in human ES cells is a valuable tool for differentiation studies, gain or loss of function assays, as well as disease model creation.

3. Gene targeting and editing

All cells have endogenous repair mechanism to repair DNA double strand breaks either via non-homologous end joining (NHEJ) or homol- ogous recombination (HR). The relatively inaccurate NHEJ double- strand break repair machinery has been utilized to create gene disrup- tion for the generation of gene knock-outs. The most accurate method is HR which corrects the damaged chromosome using the undamaged sister chromatid as a template (Fig. 3A). Targeted gene editing to correct single base pair mutations of gene disruption by insertional mutagene- sis utilizes this repair machinery to replace or modify specic chromo- somal regions with extra chromosomal donor DNA containing the modication or gene replacement of interest (Capecchi, 1989). This method has also been used to insert reporters and genes in specic pro- moter locations and safe harbor sites for sustained gene expression. Targeted gene editing is therefore the method of choice to modify human ESCs and iPSCs to generate correct disease phenotypes to create cell models, and insert or delete genes to create knock-in or knock-out cell lines to dissect basic developmental biology. The basic methods to achieve targeted gene editing are outlined in Fig. 3B.

3.1. Homologous recombination

Gene targeting using traditional methods via homologous recom- bination has been extensively used to specically alter genes in the case of animal models such as the mouse. This method involves the introduction of a targeting construct homologous to the target gene sequence with the desired mutation and oxed drug selectable markers into a germ-line competent embryonic stem cell line (Capecchi, 1989; Thomas and Capecchi, 1987). The combination of positive and negative drug selection is then utilized to exclude clones with randomly inserted targeted constructs and positively select for mutant embryonic stem cell. Following the subsequent removal of the drug selection cassette via Cre recombinase (Gu et al., 1993; Hasty et al., 1991) mutant cells are injected into a normal blastocyst to produce a heterozygous chime- ric mouse and in-bred to generate homozygous mutant mice. Homozy- gous mutant cells can also be directly created by inactivation of both alleles of the gene (Mortensen et al., 1992). Despite this method being of low efciency and laborious due to the requirement of targeting con- structs with long homology arms and multiple rounds of drug selection to isolate desired clones, it has been used to generate several knock-in and knock-down cell lines and transgenic mice. Since the rst successful establishment of human embryonic stem cells by Thompson in 1998 (Thomson et al., 1998), this technology was used to knock down genes (Di Domenico et al., 2008; Irion et al., 2007; Urbach et al., 2004; Zwaka and Thomson, 2003). It has also been used to successfully used to knock in reporter genes into specic endogenous promoters to create lineage reporter ESC lines (Davis et al., 2008; Elliott et al., 2011; Goulburn et al., 2011; Xue et al., 2009; Zwaka and Thomson, 2003). Here, the targeting construct comprises of a core region with the reporter gene and a drug selection cassette anked with homology arms. Following transfection into cells, positive clones are identied based on drug selection and true targeted clones identied by PCR analysis. Finally, the drug selection cassette is oxed out to eliminate the potential of any interference it may cause to the reporter gene or other genes in the host cell genome (Davis et al., 2009). Conventional gene targeting via HR has also been used to insert RFP at human homo- log of the mouse ROSA26 locus in hESC and expression has shown to be sustained both in undifferentiated and differentiated cells thus suggesting this region to be a safe harborloci (Irion et al., 2007). Spontaneous gene targeting occurs at a very low frequency in mam- malian cells with an efciency of 1 in a million cells but the presence of a double-strand break is recombinogenic and increases the HR frequency by several thousand fold (Jasin, 1996). This feature has been harnessed for the development of novel and efcient targeting methods.

A. Fontes, U. Lakshmipathy / Biotechnology Advances 31 (2013) 9941001

999

/ Biotechnology Advances 31 (2013) 994 – 1001 999 Fig. 3. (A) The highly recombinogenic double

Fig. 3. (A) The highly recombinogenic double strand break recruits an endogenous double-strand DNA repair machinery that can be utilized to achieved deletion, and insertion of precise editing of the genomic loci. (B) Targeted gene editing can be carried out using traditional homologous recombination or via ZFN, TALENS, and CRISPR that rely on molecular scissors to precisely create double-strand breaks at speci c genomic sites.

3.2. Zinc nger nucleases

Zinc nger (ZF) motifs are articial DNA-binding proteins made up of approximately 30 amino acids with conserved Cys2His2 residues that chelate to zinc ion thus stabilizing the tertiary structure for the alpha helix of the motif to bind to the major grove of the DNA double helix (Pavletich and Pabo, 1991). Key amino acids at the start of the alpha helix of each ZF motif can be changed to generate different triplet sequences to confer specicity to the DNA recognition site; and multiple ZF motifs are linked in tandem to facilitate specic binding to longer DNA sequences to generate Zinc nger proteins or ZFP (Tan et al., 2003). This platform technology can then be linked to additional

functionality to activate or repress genes and also create non-specic double-strand breaks using FolkI nuclease to create Zinc nger nucleases that can recognize and cleave specic target sequences (Kim et al., 1996). Since dimerization of FokI domain is required for its activity, ZFN pairs are designed to bind to the DNA region of interest in the oppo- site orientation (Vanamee et al., 2001). The utility of this technology is best served when the ZFN-induced double-strand break is restricted to the target site. In order to increase specicity, the original ZFN design has been signicantly modied to facilitate a unique and specic targeting at essentially any locus (Miller et al., 2007). Knock-out of PIG-A, a disease related gene mutated in hematopoietic stem cells from patients with paroxysmal nocturnal hemoglobinuria,

1000

A. Fontes, U. Lakshmipathy / Biotechnology Advances 31 (2013) 9941001

was achieved in ESC and iPSC both via ZFN alone to create site-specic break followed by error-prone NHEJ and with a donor DNA via inser- tional mutagenesis (Zou et al., 2009). Knock-in of eGFP reporter was also achieved in the previously targeted Oct4 locus in human ESC and the resulting Oct4-eGFP cells retained pluripotency and differentiation potential (Hockemeyer et al., 2009). The same study also reported the targeting of one ESC line and two iPSC lines at the transcriptionally inac- tive PITX3 locus that is not expressed in pluripotent state but turned on after differentiation. In addition to the creation of knock-out and knock- in cell lines, ZFN technology has been used for gene and shRNA insertion into the safe harbor locus AAVS1 on chromosome 19 for over expression of knock-down of specic genes (DeKelver et al., 2010; Hockemeyer et al., 2009). The knockout of CCR5 gene in CD34+ cells conferring resis- tance to HIV infection demonstrates the potential of ZFN in potential clinical applications (Holt et al., 2010). However, its widespread use is hindered due to the challenges associated with designing DNA sequences for nger design that confer sequence specicity thus elimi- nating off target effects.

3.3. TAL effectors

Transcription activator-like effectors (TALE), rst identied in the plant pathogen Xanthamonas, recognize DNA in a modular manner ( Boch et al., 2009; Moscou and Bogdanove, 2009) and hence have recently gained popularity for DNA targeting because it confers higher specicity. Specic DNA sequence recognition is achieved by a central repetitive region consisting of varying numbers of identical repeat units of typically 33 35 amino acids with two variable amino acids termed the repeat-variable diresidues (RVD). It is thought that these RVDs speci cally contact base pairs for target recognition, albeit with a minor mismatch. The code that constitutes most frequent asso- ciations can be used to predict TALE binding sites for custom design spe- cic to target DNA sequences of choice. Similar to the ZFN technology, the combination of the TALE and FokI nuclease is designed in pairs to bind opposing DNA target sites separated by spacer sites that are used to design TAL effector nucleases, or TALENs (Bogdanove and Voytas, 2011), and to successfully target endogenous genes in human cell lines (Cermak et al., 2011; Miller et al., 2011). Knock-in of eGFP at the endogenous Oct4 and PITX3 promoters and gene insertion of constitu- tive eGFP at the safe-harbor AAVS1 sites were targeted in ESCs and iPSCs using TALENs and the targeting efciencies were found to be comparable (Hockemeyer et al., 2011) to the frequencies previously observed for these sites with ZFN (Hockemeyer et al., 2009). Given the powerful combination of high targeting efciency and simpler recognition code for effective design of sequence specic TALENs, this method holds a huge potential for the modication of pluripotent stem cells for various downstream applications.

3.4. Crispr/CAS

Recently, a new class of genome engineering tools that harnesses the ability of RNA to program sequence-specic DNA cleavage was reported. Based on the prokaryotic CRISPR (clustered regularly interspaced short palindromic repeats), this minimal 3-component platform comprises of Cas9 (codon optimized and attached to nuclear localization signals), and two noncoding RNAs: pre-crRNA (array containing nuclease guide sequences or spacers interspaced by identical direct repeats), and an 89-nucleotide tracrRNA. Fig. 3A outlines the basic schematic of the CRISPR cleavage. Directed by the two noncoding short RNAs, Cas9 nucle- ases have been shown to precisely and efciently cleave endogenous genomic loci in human and mouse cells (Cong et al., 2013). Further, this system was used to target the AAVS1 locus in induced pluripotent stem cells with targeting rates of 24% (Mali et al., 2013). Recently, this tool was utilized for the disruption of ve genes simultaneously in mouse ES cells as well as insertion into mouse zygotes for the generation of mice with multiple point mutations at specic targets (Wang et al.,

2013). The advantage of this method over all the previous gene- targeting systems is that although specicity and efciency maybe similar or better, this robust method allows simultaneous editing of multiple target loci in the mammalian genome. The methods outlined above for the creation of genetically engineered embryonic stem cells and induced pluripotent stem cells aid in uncovering key aspects of basic stem cell biology and understanding human disease. Modied stem cells also serve as a relevant in vitro cell model to develop safe drugs and toxicity screening methods. As advances in genetic modication of stem cells continue, we approach the realiza- tion of their unparalleled potential in regenerative medicine and drug discovery.

References

Cermak T, Doyle EL, Christian M, Wang L, Zhang Y, Schmidt C, et al. Ef cient design and assembly of custom TALEN and other TAL effector-based constructs for DNA targeting. Nucleic Acids Res 2011;39:e82. Chalberg TW, Portlock JL, Olivares EC, Thyagarajan B, Kirby PJ, Hillman RT, et al. Inte- gration speci city of phage phiC31 integrase in the human genome. J Mol Biol 2006;357:28 48. Chen ZY, He CY, Ehrhardt A, Kay MA. Minicircle DNA vectors devoid of bacterial DNA result in persistent and high-level transgene expression in vivo. Mol Ther 2003;8(3):495 500. Chuang CK, Sung LY, Hwang SM, Lo WH, Chen HC, Hu YC. Baculovirus as a new gene delivery vector for stem cell engineering and bone tissue engineering. Gene Ther 2007;14:1417 24. Cong L, Ann Ran F, Cox D, Lin S, Barretto R, Habib N, et al. Multiplex genome engineering using CRISPR/CAS systems. Science 2013;339(6121):81923. http://dx.doi.org/10.

1126/science.1231143. [Feb 15, Epub 2013]. Costa M, Dottori M, Ng E, Hawes SM, Sourris K, Jamshidi P, et al. The hESC line Envy ex-

Elliott DA, Braam SR, Koutsis K, Ng ES, Jenny R, Lagerqvist EL, et al. NKX2-5(eGFP/w)

A. Fontes, U. Lakshmipathy / Biotechnology Advances 31 (2013) 9941001

1001

Ivics Z, Hackett PB, Plasterk RH, Izsvak Z. Molecular reconstruction of Sleeping Beauty, a

Khan IF, Hirata RK, Wang PR, Li Y, Kho J, Nelson A, et al. Engineering of human plurip- otent stem cells by AAV-mediated gene targeting. Mol Ther 2010;18:1192 9. Kim YG, Cha J, Chandrasegaran S. Hybrid restriction enzymes: zinc nger fusions to Fok I cleavage domain. Proc Natl Acad Sci U S A 1996;93:1156 60. Kost TA, Condreay JP, Jarvis DL. Baculovirus as versatile vectors for protein expression in insect and mammalian cells. Nat Biotechnol 2005;23:567 75. Lakshmipathy U, Pelacho B, Sudo K, Linehan JL, Coucouvanis E, Kaufman DS, et al. Ef- cient transfection of embryonic and adult stem cells. Stem Cells 2004;22:531 43. Liu Y, Thyagarajan B, Lakshmipathy U, Xue H, Lieu P, Fontes A, et al. Generation of plat- form human embryonic stem cell lines that allow ef cient targeting at a predetermined genomic location. Stem Cells Dev 2009;18:145972. Ma Y, Ramezani A, Lewis R, Hawley RG, Thomson JA. High-level sustained transgene expression in human embryonic stem cells using lentiviral vectors. Stem Cells 2003;21:111 7. Macarthur CC, Xue H, Van Hoof D, Lieu PT, Dudas M, Fontes A, et al. Chromatin insulator elements block transgene silencing in engineered human embryonic stem cell lines at a de ned chromosome 13 locus. Stem Cells Dev 2012;21:191 205. Madeira C, Rodrigues CA, Reis MS, Ferreira FF, Correia RE, Diogo MM, et al. Nonviral gene delivery to neural stem cells with minicircles by microporation. Biomacromolecules 2013;14(5):1379 87. Mali P, Yang L, Esvelt KM, Aach J, Guell M, DiCarlo JE, et al. RNA-guided human genome engineering via Cas9. Science 2013;339(6121):823 6. http://dx.doi.org/10.1126/ science.1232033 . [Feb 15, Epub 2013 Jan 3]. Miller JC, Holmes MC, Wang J, Guschin DY, Lee YL, Rupniewski I, et al. An improved zinc- nger nuclease architecture for highly speci c genome editing. Nat Biotechnol 2007;25:778 85. Miller JC, Tan S, Qiao G, Barlow KA, Wang J, Xia DF, et al. A TALE nuclease architecture for ef cient genome editing. Nat Biotechnol 2011;29:143 8. Mitchell RS, Beitzel BF, Schroder AR, Shinn P, Chen H, Berry CC, et al. Retroviral DNA in- tegration: ASLV, HIV, and MLV show distinct target site preferences. PLoS Biol

Mortensen RM, Conner DA, Chao S, Geisterfer-Lowrance AA, Seidman JG. Production of homozygous mutant ES cells with a single targeting construct. Mol Cell Biol 1992;12:2391 5. Moscou MJ, Bogdanove AJ. A simple cipher governs DNA recognition by TAL effectors. Science 2009;326:1501. Orban TI, Apati A, Nemeth A, Varga N, Krizsik V, Schamberger A, et al. Applying a double-feature promoter to identify cardiomyocytes differentiated from human embryonic stem cells following transposon-based gene delivery. Stem Cells 2009;27:1077 87. Pavletich NP, Pabo CO. Zinc nger-DNA recognition: crystal structure of a Zif268-DNA complex at 2.1 A. Science 1991;252:809 17. Ren C, Zhao M, Yang X, Li D, Jiang X, Wang L, et al. Establishment and applications of epstein-barr virus-based episomal vectors in human embryonic stem cells. Stem Cells 2006;24:1338 47. Russell WC. Update on adenovirus and its vectors. J Gen Virol 2000;81(11):2573 604. Suter DM, Cartier L, Bettiol E, Tirefort D, Jaconi ME, Dubois-Dauphin M, et al. Rapid gen- eration of stable transgenic embryonic stem cell lines using modular lentivectors. Stem Cells 2006;24:615 23. Tan S, Guschin D, Davalos A, Lee YL, Snowden AW, Jouvenot Y, et al. Zinc-nger protein-targeted gene regulation: genomewide single-gene speci city. Proc Natl Acad Sci U S A 2003;100:11997 2002. Thomas KR, Capecchi MR. Site-directed mutagenesis by gene targeting in mouse embryo-derived stem cells. Cell 1987;51:503 12. Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ, Marshall VS, et al. Em- bryonic stem cell lines derived from human blastocysts. Science 1998;282:11457. Thyagarajan B, Liu Y, Shin S, Lakshmipathy U, Scheyhing K, Xue H, et al. Creation of engineered human embryonic stem cell lines using phiC31 integrase. Stem Cells 2008;26:119 26. Thyagarajan B, Scheyhing K, Xue H, Fontes A, Chesnut J, Rao M, et al. A single EBV-based vector for stable episomal maintenance and expression of GFP in human embryonic stem cells. Regen Med 2009;4:23950. Urbach A, Schuldiner M, Benvenisty N. Modeling for Lesch-Nyhan disease by gene targeting in human embryonic stem cells. Stem Cells 2004;22:635 41. Vanamee ES, Santagata S, Aggarwal AK. FokI requires two speci c DNA sites for cleav- age. J Mol Biol 2001;309:69 78. Wang H, Yang H, Shivalila C, Dawlaty M, Cheng A, Zhang F, et al. One-step generation of mice carrying mutations in multiple genes by CRISPR/Cas-mediated genome engi- neering. Cell 2013;4(153):910 8. http://dx.doi.org/10.1016/j.cell.2013.04.025 . [May 02]. Wilber A, Linehan JL, Tian X, Woll PS, Morris JK, Belur LR, et al. Ef cient and stable transgene expression in human embryonic stem cells using transposon-mediated gene transfer. Stem Cells 2007;25:291927. Wobus AM, Boheler KR. Embryonic stem cells: prospects for developmental biology and cell therapy. Physiol Rev 2005;85:635 78. Xu N, Papagiannakopoulos T, Pan G, Thomson JA, Kosik KS. MicroRNA-145 regulates OCT4, SOX2, and KLF4 and represses pluripotency in human embryonic stem cells. Cell 2009;137:647 58. Xue H, Wu S, Papadeas ST, Spusta S, Swistowska AM, MacArthur CC, et al. A targeted neuroglial reporter line generated by homologous recombination in human em- bryonic stem cells. Stem Cells 2009;27:1836 46. Yates JL, Warren N, Sugden B. Stable replication of plasmids derived from Epstein-Barr virus in various mammalian cells. Nature 1985;313:812 5. Zou J, Maeder ML, Mali P, Pruett-Miller SM, Thibodeau-Beganny S, Chou BK, et al. Gene targeting of a disease-related gene in human induced pluripotent stem and embry- onic stem cells. Cell Stem Cell 2009;5:97 110. Zwaka TP, Thomson JA. Homologous recombination in human embryonic stem cells. Nat Biotechnol 2003;21:319 21.