Hannover Medical School Institute of Experimental Hematology

Establishing a zinc finger nuclease platform for targeted genome editing in human cell lines, primary keratinocyte stem cells and cardiomyocytes in vivo

INAUGURAL - DISSERTATION

For the degree of Doctor of Natural Sciences

- Doctor rerum naturalium –

(Dr. rer. nat)

Submitted by

Kafaitullah Khan

Place of birth - Zhob, Pakistan

Hannover 2012

Angenommen vom Senat der Medizinischen Hochschule Hannover am 23.05.2012

Gedruckt mit Genehmigung der Medizinischen Hochschule Hannover

Präsident: Prof. Dr. med. Dieter Bitter-Suermann

Betreuer: Prof. Dr. phil. Toni Cathomen

Kobetreuer: PD Dr. med Albert Heim

1. Gutachter: Prof. Dr. phil. Toni Cathomen

2. Gutachter: PD Dr. med Albert Heim

3. Gutachter: Prof. Dr. rer. nat. Jürgen Alves

Tag der mündlichen Prüfung vor der Prüfungskommission: 23.05.2012

Prof. Dr. rer. nat. Jürgen Alves

Prof. Dr. phil. Toni Cathomen

PD Dr. med Albert Heim

Prof. Dr. rer. nat. Jürgen Alves

Dedicated to my grandmother

Table of contents 3

Table of Contents

1. Introduction...... 4 1.1. Genome editing ...... 4 1.1.1. Gene addition ...... 5 1.1.2. Targeted gene addition...... 5 1.1.3. Targeted gene correction...... 6 1.1.4. Targeted gene knockout or disruption ...... 7 1.1.5. Chromosomal rearrangement...... 8 1.2. Nucleic acid delivery systems...... 9 1.2.1. Non-viral delivery systems...... 9 1.2.1.1. Physical methods of nucleic acid delivery ...... 9 1.2.1.2. Chemical methods of nucleic acid delivery ...... 11 1.2.2. Viral delivery systems ...... 12 1.2.2.1. Retroviruses...... 13 1.2.2.2. Lentiviruses...... 14 1.2.2.3. Adenovirus...... 15 1.2.2.4. Adeno-associated virus...... 16 1.2.2.5. Other viral vectors...... 19 1.3. Zinc finger nuclease...... 20 2. Aims of the thesis ...... 22 3. Publication 1 ...... 23 4. Publication 2 ...... 33 5. Summary of results and discussion ...... 43 6. Summary ...... 53 7. Zusammenfassung ...... 54 8. References ...... 55 9. Appendix...... 75 9.1. List of own publication ...... 75 9.2. Curriculum Vitae ...... 76 9.3. Acknowledgment ...... 77 9.4. List of abbreviation...... 78 9.5. Declaration ...... 80

Introduction 4

1. Introduction

Living organisms have a certain set of genes which are their unit of heredity. Genes can be defined as “a locatable region of genomic sequence, corresponding to a unit of inheritance, which is associated with regulatory regions, transcribed regions and/or other functional sequence regions” [1]. Genes have information for certain type of structural or functional proteins. These proteins have a wide range of functions which includes defence against pathogen, homeostasis, maintenance and regulation of different cellular process. Most of diseases have a genetic factor which is either entirely or partially responsible for that disease. Disorders like thalassaemia, Sickle cell anaemia, and Gaucher’s have either completely lack or has mutation in a particular gene while diseases like hypertension and hypercholesterolemia involve interaction of genetic and environmental factors. In some other disorders, like Angelman syndrome, the causative agent is the loss of particular gene activity in specific cells. Even the infectious disease have genetic basis which involve the genes of the infectious agent and the host.

Some of these genetic/infectious diseases have no treatment while in the case where treatment is present, majority of them are prophylactic rather than curative, by minimizing or reducing the severity of the diseases instead of providing permanent solution. One example of it is treatment of Gaucher's disease. Gaucher's disease is lysosomal storage disease due to accumulation of fatty substances in cells and certain organs [2]. The causative agent of this disease is a recessive mutation in a glucocerebrosidase gene which leads to the deficiency of the glucosylceramidase, an enzyme that degrades fatty acid glucosylceramide into D-glucose and N-acylsphingosine. Current treatment for Gaucher's disease includes enzyme replacement therapy with intravenous introduction of recombinant glucosylceramidase but this treatment is very expensive and should be continued for whole life. So to improve the life standard of these patients, there is a need of more effective and convenient long lasting therapy like gene based therapy or simply gene therapy.

Gene-based therapy, which can be achieved by genome editing, are broadly define as “the introduction, using a vector, of nucleic acids into cells with the intention of altering gene expression to prevent, halt or reverse a pathological process” [3].

1.1 Genome editing

Depending upon the scientific or medical need editing of the genome can be performed by different ways including gene addition, targeted gene addition, targeted gene correction, targeted gene disruption and chromosomal rearrangements.

Introduction 5

1.1.1 Gene addition

Gene addition is the insertion of a functional copy of a mutated native gene. The first type of viruses used for gene addition therapy was retroviruses as their normal life cycle favour them to be used for permanent gene delivery in cells. The biggest barrier for the use of gene addition therapy is its inability to specify the insertion point of the transgene. This "stone throwing" approach of random insertion of a therapeutic gene, despite the successes of transferring it, may result in malignant transformation from the insertional activation of oncogenes [4]. In addition, randomly inserted genes are also subjected to silencing and position-effect variegation which make their expression unreliable and unpredictable [5, 6]. vice versa the randomly integrated transgene may affect endogenous genes and chromatin with the potential to change the cell behaviour and induce its transformation [7]. With some setbacks [8], gene addition has been used to successfully treat many diseases [9-12].

1.1.2 Targeted gene addition

Gene targeting is the method which allows for the precise and efficient modification of the genome. This method holds great potential not only as a tool for studying function of particular genes but also in biotechnology and human gene therapy [13]. Gene targeting or gene replacement strategy can be defined as “replacement of the endogenous DNA segment, with exogenously introduced DNA fragments, at predetermined genomic location by harnessing homologous recombination”. Targeted gene addition can be permanent or conditional. Conditional means that the transgene may express at specific time during the development of the organism or its expression may be specific to certain tissue. The expression of transgene, for example, can be made specific to a tissue or cells by using tissue specific promoter [14, 15] or detargeting by employing tissue specific microRNA (mir) target sites [16].

Targeted gene addition has three important advantages over other methods for gene editing. First it is possible to choose the genetic locus for gene editing as the addition takes place at predetermined genomic location. Second, the investigator can use full advantage of all the resources provided by the known sequences of the organism, especially for the mouse and human genomes. Third, investigators have control over the degree of modulation of the chosen genomic location [17]. Specifically the last advantage provides the investigator with the ability to modify the chosen genomic location according to pursuing biological questions. This includes the creation of null mutations or hypomorphic mutations as well as introduction of reporter genes to determine gene expression profiles, cell lineage or cell fate in transplantation studies, to restrict the manipulation to specific cell groups or organs (spatial restriction) or to any particular period of the development of an organism (temporal restriction) [13].

Introduction 6

The location of introducing a new gene in this method is of prime importance both in terms of safety and stability of the gene. As its position cannot only create problems for the integrity of the genome and normal surveillance of the organism/cell but may also affect the degree of expression of the newly introduced gene. So the best choice for targeted integrations are genomic safe harbours which must be potentially dispensable or, alternatively, extragenic regions. Genomic safe harbours can be defined as “Intragenic or extragenic regions of the genome that are able to accommodate the predictable expression of newly integrated DNA without adverse effects on the host cell or organism” [18]. The benefit of genomic safe harbours includes sufficient/uniform expression of transgene and the inability of the integrated transgene to transform the cell into a malignant form or alter its normal functions. In brief, based on prior knowledge, the outcome of integration can be predicted. To date the following three sites in the human genome have been used for targeted gene addition: the adeno-associated virus integration site1 (AAVS1) [19-25], chemokine (C-C motif) receptor 5 locus [26] and ROSA26 locus [27] which is the human orthologue of the mouse ROSA26 locus [18].

As discussed above, targeted gene addition has great potential in developing the systems for studying gene function, biotechnology as well as human gene therapy. The same method can be used to produce mutant transgenic mice, for example to create models of human genetic disease [17, 28, 29]. Different variations of this technique have been applied since its discovery some 31 years ago [30, 31]. In addition targeted gene addition has enormous potential in human gene therapy and numbers of investigator employed this method in attempts to establish treatment for human monogenic diseases, such as sickle cell anemia [32] and haemophilia [33].

1.1.3 Targeted gene correction

Gene correction is “the in situ replacement of mutated endogenous gene with exogenously introduced (donor) normal/wild type copy, by gene targeting, potentially without perturbing rest of the genome, with the aim of restoring normal gene function”.

The method and potential of gene correction to treat genetically inherited diseases can be appreciated in one of most recent published study by Yusa et al. [34]. In their study they successfully corrected a point mutation in alpha1-antitrypsin (A1AT, also known as SERPINA1) by gene correction method emplyoing induced Pluripotent Stem Cells (iPSC), Zinc Finger Nuclease (ZFN) and transposon (PiggyBac) technologies. The authors tried to correct human iPSC derived from individuals with alpha1-antitrypsin deficiency (A1ATD) [35]. A1ATD is the most common genetically inherited, autosomal recessive, metabolic disorder of the liver found in North European descent [36, 37]. The only available current therapy for this disorder is liver transplantation [34]. In their study they achieved biallelic correction of point mutation in these iPSC by introducing the donor with the help of ZFN and removing the unwanted sequences (e.g. puromycin selection cassette) by transposon piggyBac. The differentiation of these corrected iPSC into hepatocyte-like cells and Introduction 7 subsequent functional studies showed that they share key functional attributes of hepatocytes. Further, they also performed the transplantation of these hepatocyte-like cells into mice and showed the restoration of alpha1-antitrypsin in those cells. In brief, this proof of principle study clearly shows not only the correction of mutated gene, but also the integrity of genome and later the functionality of the corrected gene.

1.1.4 Targeted gene knockout or disruption

Targeted gene disruption or Knockout is a gene manipulation technique which can be defined as “a technique to altered one or more genes in order to make them inoperative or change the product, either by producing mutations in them or their complete deletion”. Depending upon the experimental limitations and the desired outcome, gene disruption can be achieve by several routes, including the induction of frame shift mutations to create nonsense transcripts, deletion of critical amino acids and/or the disruption of splicing.

The deletions or insertions of nucleotides in the coding sequence, that is not a multiple of three, will alter the reading frame resulting in a different transcript than wild type. This shift in reading frame or in other words frame shift mutation, generally results in a nonsense transcript due to exposure to premature in frame stop codon. This nonsense transcript is then degraded by the Nonsense-Mediated Decay (NMD) pathway. NMD is a cellular mechanism of mRNA surveillance that detects nonsense mutation of truncated or “wrong” protein [38].

A gene can also be disrupted by introducing in-frame deletions or insertions. These deletions or insertions can be programmed to delete critical amino acids in the translated protein. Deleting critical amino acids in a protein will disrupt its function or/and stability. For example, we know that enzymes have active sites where substrates bind to undergo a chemical reaction. Deleting critical amino acids in those active sites will render the enzymes catalytically inactive.

In addition, a gene can be disrupted by manipulating the splice junctions. Splicing is a eukaryotic cellular mechanism of maturing nascent primary mRNA transcript by removing introns and joining exons. Mutations in splice junctions may cause loss of function of that site, resulting in the introduction of premature stop codon or/and inclusion of introns or loss of exons. This may also lead to variations in splice locations, resulting in deletions or insertions which may cause a shift in reading frame. In brief, the disruption of splice junctions may disrupt the transcript/peptide integrity, including its degradation by control mechanism such as NMD.

Gene knockout has been used extensively to produced knockout model organisms and cell lines. An endogenous gene can be disrupted in cell lines as well as cells isolated from different organisms including human [39, 40]. To perform gene disruption in vitro and ex vivo, the gene of interest can either be mutated directly [41] or by using a target vector Introduction 8 employing homologous recombination. The vectors normally have reporter and/or selection marker genes flanked by homologous arm, which replace the endogenous gene. The manipulated cell clones are than picked, expanded and tested for the mutant genotype along with confirmation by sequencing. In addition, functional tests are performed to assess the mutant phenotype.

Model organisms are non-human in vivo models for studying specific biological phenomena such as genetic disorders, gene function, mechanism of infection by pathogen, behavioural responses, as well as to develop new technologies in therapeutics, pharmacology, biotechnology and industry. They range from as small as viruses to as large as mammals. Based upon the specific goals, for human genome based studies usually animal models having taxonomic equivalency as well as physiological, anatomical and genomic resemblance to humans are used.

Among animal models the mouse is closely related to humans and is used as a perfect small animal model of human diseases. This is also supported by the comparative genomic studies, as they show high homology between human and mouse genomes [42]. Many different strains of mice are used for knockout of different genes in order to create the model for studying that particular gene or DNA sequence. Knockout mice are generally produced by two main methods namely gene targeting and gene trapping.

Gene targeting is based on more precise homologous recombination and embryonic stem cell culturing [43-45]. After twenty years of developing this method, the Nobel prize in medicine or physiology was awarded in 2007 to Martin Evans, Mario Capecchi and Oliver Smithies for their discoveries of “principles of introducing specific gene modification in mice by the use of embryonic stem cells” [46]. The standard concept of generating knockout mouse is straight forward but are lengthy and the combination of low-probability events [47]. The gene trapping method uses random insertion property of trapping vectors in mammalian genome. The principle element of these vectors is the gene trapping cassette which usually consist of a promoterless reporter gene and/or a selectable marker flanked by an upstream splice acceptor and a 3` transcription termination signal (polyadenylation signal) [48].

The knockout technologies have a wide range of application including biomedical research, biotechnology, pharmacology, genetics, functional genomics and proteomics etc.

1.1.4 Chromosomal rearrangements

In humans chromosomal structure can alter either spontaneously or after exposure to some DNA damaging agents [49]. These alterations are of significant biological consequences and beside some advantages they may cause adverse effects not only to the individual in which they take place but also to the next generation as they might be inherited [50, 51]. These abnormalities in human chromosomes are the major cause of Introduction 9 fetal loss and developmental disorders [52]. In addition translocation in human chromosomes may cause tumour [53].

Owing to physiological, anatomical and genomic similarities to human as well as the ease with which its genome can be manipulate, mice have been used as a model for chromosomal rearrangements studies [49]. Several chromosomal rearrangement events like deletion [54], duplication [55], inversion [56] and translocations [57] can be induced in mice. To achieve these chromosomal rearrangements a variety of different techniques, including recombinase systems like FLP/FRT [58] and Cre/loxP [59], bacterial artificial chromosomes (BAC) [60], endonucleases like I-SceI [61, 62] and ZFNs [63-65] have been employed. Some of these rearrangement techniques have been used in mice for modelling chromosomal rearrangement disorders like DiGeorge syndrome [66], Smith-Magenis syndrome [67, 68] and Down syndrome [69].

1.2 Nucleic acid delivery system

The main concept of genome editing is based on the introduction of exogenous genetic material into target cells to alter the endogenous genome. To achieve this, technologies which are capable of transferring exogenous nucleic acid to multiple targets are required. Producing such a delivery systems, which is safe and efficient, is one of the biggest barrier to genome editing. Gene delivery systems are broadly divided into non-viral and viral.

1.2.1 Non viral delivery systems

Non viral delivery system have certain advantages over viral vectors, including the simplicity of production and the low host immune response but are less efficient in delivering nucleic acids. They are divided into physical and chemical delivery system.

1.2.1.1 Physical methods of nucleic acid delivery

Physical methods include mechanical methods such as microinjection, high pressure, particle bombardment, magnetofection, and sonoporation, electrical methods, like electroporation, as well as photonic tools like laser.

Microinjection is the direct injection of uncomplexed, naked nucleic acid into the cell. It is a simple and easy method for nucleic acid delivery. Specific cells, for example, embryonic stem cells, can be targeted by this method. However the fact that only single cells can be targeted by this approach not only make it slow and laborious but inappropriate for research where large number of cells are desired for targeting as well as for in vivo application [70].

High pressure is also used to mediate the delivery of naked oligonucleotide. Transfection of cardiovascular tissue has been achieved with an efficiency of 50% [71]. Naked DNA is Introduction 10 also delivered to hepatocytes by similar hydrodynamic force (rapid injection by tail vein) [72, 73]. In addition the same hydrodynamic based transfection was used to deliver nucleic acid to different organs of post-natal mice [74]. Those particles or molecules which normally are unable to traverse the plasma membrane can be delivered using the same high pressure approach. However this method is unspecific as it targets the cells without specification.

Particle bombardment or biolistic particle delivery by gene gun is used to deliver nucleic acid to many cells simultaneously. In this method exogenous nucleic acid is coated with metal (such as gold or tungsten) nanoparticles which are than accelerated to high velocity to enter the cell [75]. Gene delivery to brain [76], skin, mammary gland and liver [77] as well as tumour cells [78] has been achieved by using this method. However its in vivo use is mainly limited to superficial cells due to difficulty in controlling the path of the particles and poor tissue penetration [79, 80].

Magnetofection is the delivery of nucleic acid to the cells by the use of magnetic field which can be defined as “nucleic acid delivery guided and mediated by magnetic force acting as an associate of magnetic particles and nucleic acids” [81, 82]. In this approach magnetic particles complexed with nucleic acid are concentrated on cells by using magnetic field. Its benefits include rapid transfection kinetics, improved dose-response relationship while drawback includes difficulty to apply in vivo [83].

Sonoporation is the use of the sound (generally ultrasound) for nucleic acid delivery. The main principle of this approach is acoustic cavitation (i.e. rupturing of liquids and associated effects) of microbubble which release energy that in turn temporarily increase the internalization of exogenous nucleic acid into the cells by changing the permeability of the plasma membrane. Sonoporation has been used for in vitro [84, 85] as well as in vivo delivery of nucleic acid including skeletal muscle [86, 87], blood vessel [88], skin [89], heart [90], and pancreas [91].

Electroporation is the delivery of nucleic acid into the cell by applying electric pulse. High voltage electric pulses transiently permeabilize plasma membrane by opening the pores in it and allowing the cellular uptake of the macromolecules including nucleic acid. This technique is simple and has the benefit of efficient delivery to a variety of cell types [92, 93]. Beside this the disadvantages of electroporation are the need of a large amount of nucleic acid, the need of optimization the procedure for every cell line, and high mortality of the cells. In addition, although it has been used for in vivo delivery of nucleic acids to skin [94], corneal endothelium [95], muscle [92], melanoma [96] and liver [97] the in vivo application remains difficult.

Optical transfection is the introduction of foreign nucleic acids into the cell by the use of light. In this method exogenous nucleic acid get entry into the cells through transient pores in the plasma membrane, which are generated after exposure to highly focused laser for Introduction 11 brief time. Since the delivery of plasmid DNA for the first time [98] in 1984 it has been used for many cell types [99, 100]. Beside in vitro this technique is also used for in vivo delivery [101, 102].

1.2.1.2 Chemical methods of nucleic acid delivery

In chemical methods nucleic acids are normally complexed with cationic molecules which are attracted to negatively charge plasma membrane by electrostatic interaction. Nucleic acid being anionic is condensed by these cationic carriers so that the complex as a whole maintains a positive charge. These complexes enter into the cells by the process of endocytosis after binding to negatively charged plasma membrane. After entering into the cells swelling of the endosome take place. The presence of cations in endosomes create a charge gradient between the endosome and the surrounding cytosol, causing water to enter the endosome that start to swell and then rupture to release the nucleic acid complex into the cytosol. After endosomal escape, the nucleic acid in the case of DNA must reach to the nucleus. This intracellular trafficking of the DNA to the nucleus occurs by diffusion and cytoskeleton pathways. Once it reaches the perinuclear space the DNA enter the nucleus through the nuclear pore complex by passive diffusion, active transport or calcium regulated transport depending upon the size. After nuclear localization, the DNA is separated from the complex and transcribed.

The chemical methods or non viral vectors are subdivided into organic and inorganic. Organic includes lipoplexes and polyplexes, depending upon the nature of the organic substance used.

Lipoplexes are complexes of cationic lipids with nucleic acid. Owing to the positive charge cationic lipids condense and encapsulate the anionic nucleic acid into liposome. These complexes bind to the negatively charged plasma membrane by non-specific electrostatic interaction [103] which facilitates the intracellular nucleic acid entry by endocytosis. They are easy and cheap to synthesise as well as non-pathogenic. Since their development in 1987 [104] they have been successfully used for in vitro [105, 106] and in vivo delivery, including the lung [107], brain [108], tumours [109, 110] and skin [111]. Although used in clinical trials [112, 113] problems of low efficiency and toxicity are associated with the method [114].

Polyplexes are complexes between nucleic acids and a cationic polymer. In polyplexes, like lipoplexes, anionic nucleic acid is condensed into stable complexes by cationic polymers due to electrostatic interactions [115]. The cationic surface charge of the complex is used to bind the plasma membrane by interacting with anionic proteoglycan [103, 116, 117] followed by endocytosis [118] of the complex. Polylysine (PLL) [119, 120] and polyethyleneimine (PEI) [121, 122] are the most widely used polymers for polyplexes, with PLL being the first polymer used for nucleic acid delivery [123]. They have several benefits such as easy and cheap production, targeting of specific cells and tissue, low Introduction 12 immunotoxicity and freedom with the size of nucleic acid [124]. However they may exert toxic effect [114].

The inorganic substances capable of nucleic acid delivery are gold nanoparticles, silica, iron oxide, magnesium phosphate and calcium phosphate.

Iron-based nanoparticles have improved storage stability and can be designed for different application. They have been shown to transfect mouse cells with the application of magnetic field [125]. Their transfection efficiency has been improved by surface modification, for example, by minimizing coagulation of particles [126]. However they are difficult to translate in vivo [127].

The surface charge of silica particles is negative. In order to interact with a negatively charged plasma membrane and enter the cells they are normally modified with cationic surface molecules [128-130]. Modification of silica particles by organic substances protect the nucleic acid from enzymatic degradation and improved the transfection [131]. The prime advantage of these organosilicates is their low toxicity to cells, in contrast to many organic materials, while disadvantages include the unknown particle fate [127].

Gold nanoparticles (GNPs) are easy for surface modification and can be used together with gene gun technology [75, 132]. They also have the advantage of photothermal properties and are locally exothermic after irradiation with the light. This property of GNPs is used for specific destruction of cancer cells where they release macromolecules after excitation with laser [133]. Their drawbacks includes non degradable nature and accumulation in the liver [127].

Calcium phosphate has been used for in vitro and in vivo transfection including human cells [134, 135]. Calcium is a strong cation and forms complexes with anionic nucleic acid, providing the positive surface charge which is necessary to enter the cells. The surface modification of calcium phosphate complexes can attribute different properties to them such as longer blood circulation and targeting to specific tissue. They also have the advantage of being biodegradable and biocompatible but it is difficult to control their synthesis [127].

Magnesium phosphate is also shown to transfect cells [136]. Magnesium is similar to calcium as both are divalent ion and are highly biocompatible but is relatively unexplored and more research is needed to prove their efficiency for transfecting different cell types.

1.2.2 Viral delivery systems

Viruses are obligate intracellular parasites that need the host cell machinery for their replication. They have efficient strategies to infect host cell and deliver their genetic information to the nucleus where it remains either latently as part of the host genome or Introduction 13 episomally as autonomous genetic entity. Viruses can be manipulated to be used as vector to deliver exogenous nucleic acid into the cells. Viruses used for vector generation can be classified by different ways including enveloped and non-enveloped, integrating and non- integrating, lytic and lysogenic as well as RNA and DNA viruses.

RNA viruses are a large and diverse group that infect a broad spectrum of cells from prokaryotes to eukaryotes [137]. Among them the Retroviridae family (retroviruses) based vectors has been used extensively as nucleic acid delivery vehicles. Retroviruses are generally categorized into simple (gammaretroviruses) and complex including lentiviruses such as human immunodeficiency virus 1 (HIV1), and spumaviruses such as chimpanzee foamy virus [138].

1.2.2.1 Retroviruses

Gammaretroviruses have an envelope that encapsulates the cone shaped protein capsid, which further contains the viral genome [139]. Their envelope is a lipid bilayer which originates from host cell membrane and contains virally encoded surface and transmembrane glycoproteins [140].

Their genome consist of two 7-12 kb long linear, nonsegmented, single stranded RNAs that contain the three major coding segments gag, pol and env. The gag encodes virus structural protein which forms the capsid, matrix and nucleoprotein complex, while pol encodes the viral enzymes, such as reverse transcriptase and integrase whereas env contains information for encoding surface and transmembrane glycoproteins [138, 139].

The life cycle of retroviruses is well studied. They attach to specific host cell receptors by their envelope's glycoprotein which is followed by membrane fusion and release of the viral capsid into the host cell cytoplasm. After internalization linear double stranded DNA (dsDNA) is generated by reverse transcription from viral genomic RNA using its reverse transcriptase [141]. Viral dsDNA is unable to cross nuclear membrane and integrate into host genome (to form provirus) unless cells undergo mitosis (division of cells) resulting in nuclear membrane disintegration [142]. The integrated provirus uses cis-acting elements which are mainly present in long terminal repeats (LTRs). To express the viral proteins they assemble the host cell cytoplasm and escape by budding out from the plasma membrane [143].

Retroviral vectors are derived from different gammaretroviruses including murine leukemia virus (MLV), spleen necrosis virus, Rous sarcoma virus and avian leukosis virus [144]. The majority of the retroviral vectors are derived from MLV and they were among the first to be used for human gene therapy [145]. They are replication defective by replacing all trans- acting elements (gag, pol and env) with a transgene of interest, while only leaving the packaging signals (psi), LTRs, attachment sites and sequences necessary for transgene Introduction 14 expression [146, 147]. Replication and packaging of vector is achieved in packaging cells that express gag/pol and env from separate constructs.

Retroviral vectors have several advantages which includes their stable integration into host genome that lead to long lasting transgene expression, easy production of vector titres, high enough for nucleic acid delivery, infectivity of a broad range of target cell types and their flexibility regarding the size of transgene. Their disadvantages includes the instability of some vectors, insertional mutagenesis due to random integration of viral vectors in the host genome and limitation of delivering transgenes only to dividing cells, due to their mitosis dependency [148].

1.2.2.2 Lentiviruses

Lentiviruses are complex retroviruses of the family Retroviridae. They have different serogroups depending on the host they are associated with. The well known and common example of lentivirus is HIV1.

The genome of lentiviruses is similar to that of gammaretroviruses with the addition of two regulatory (tat and rev) and four auxiliary (vpr, vif, vpu and nef) genes which encode for the proteins needed for viral binding, infection, replication and release [149, 150].

The life cycle of lentiviruses is related to that of gammaretroviruses, including binding, internalization and reverse transcription, however, there are some important differences between the two: First lentiviral gene expression takes place in two separate phases namely early and late phases and the transition from early to late phase is induced by viral Rev protein [151]. Second, due to viral protein Vpr they are also able to infect non-dividing cells [152].

The common lentiviral vectors are derived from HIV1 but those derived from HIV2 [153], feline immunodeficiency virus (FIV) [154], equine infectious anaemia virus (EIA) [155], simian immunodeficiency virus (SIV) [156] and maedi-visna virus [157] are also present. The lentiviral vectors contain only the essential cis elements, the packaging sequences (psi), a Rev response element and the transgene cassette [158, 159]. Lentiviral vectors are produced generally by transfecting HEK293T cells with vector plasmid and two helper plasmids, which provide trans elements for assembling of vector particles [160]. Vectors with HIV1 envelope glycoprotein infect only cells with CD4 and appropriate co-receptors, which restrict the range of targeted cells. Thus it was replaced with vesicular stomatitis virus glycoprotein (VSV-G) which not only widens vector tropism [158, 161] but also facilitate the concentration of the vector by ultracentrifugation [162, 163].

Vectors based on lentiviruses have several advantages including the ability of infecting mouse and rat embryo with tissue specific transgene expression [164], relatively large capacity for a gene of interest [139], ability of infecting also non-dividing cells and stable Introduction 15 transgene expression [148, 165]. However, the fact that these vectors are based on severely pathogenic viruses put them on disadvantage due to the possibility of virus mobilization and generation [166]. In addition, immunotoxicity [167], oncogenesis [168], and the inability of targeting specific cells [166] are the drawbacks of this vector.

Lentiviral vectors are one of the best choice for the investigators for ex vivo approaches and have been used to deliver nucleic acids to a variety of cells including terminally differentiated cells such as those of nervous and cardiac origin [165]. They have been employed in gene based therapies for diseases like adrenoleukodystrophy (ALD) [10], Parkinson’s disease [169], sickle cell anemia and ß-thalassemia [170], HIV [171] and cancer [172].

1.2.2.3 Adenovirus

Among the DNA based vectors the most widely used and studied ones are those based on adenoviruses and adeno-associated viruses. The basic principle of attachment to host cells and internalization are the same as those of RNA viruses

Adenovirus (Ad), named after its first discovery in 1953 from human adenoids [173], has been isolated from large number of species [174, 175]. The human Ad family consists of more than 50 serotypes and, based upon haemagglutination properties, are classified into six species (A-F) which infect different organs including the eye, respiratory tract, liver, gastrointestinal tract and urinary bladder [176]. Among human Ads, type 2 (Ad2) and type 5 (Ad5) of species C are the most intensively characterized types [148].

The capsid of the Ad is a nonenveloped icosahedral shaped protein shell that contains the viral genome. Ads contain fiber proteins extending from a pentameric penton base protein, which is located at each vertex of the capsid. These fiber proteins have terminal globular knob domain [177].

Their genome comprises a 26-40 kb long linear, double-stranded DNA (dsDNA) with two major transcription regions, known as early and late region [178, 179], which are flanked by 103 base pairs (pb) inverted terminal repeat (ITR) on both sides [177].

The life cycle of Ad is divided into early phase and late phase. Early phase begins with the contact of the virus with the host cell and ends with the start of genome replication. The attachment takes place by binding of the globular knob domain of the viral capsid to coxsackievirus and adenovirus receptor (CAR) of the host cell surface [180]. In addition to this binding, efficient internalization occurs by the interaction of viral penton base protein with host cell secondary receptor, integrin α-V [181]. After receptor mediated endocytosis and endosome escape of the virion, the genome is release from the protein capsid and enters the host cell nucleus where it persist episomally. After nuclear localization the replication of viral DNA starts, which mark the end of early phase and begin of late phase. Introduction 16

After DNA replication and assembly of viral progeny, the disruption of the host cell takes place, resulting in the release of virus particles. Initial infection with Ad causes nonspecific host response (synthesis of cytokines) followed by a cytotoxic T lymphocyte response against infected cells, in addition to the humoral response [175].

The Ad vectors are most commonly used vectors [177]. In the initial vectors based on Ad2 and Ad5, E1 and E3 transcription units of early region in viral genome were replaced with a transgene [182, 183], while the “gutted” or “gutless” vectors are devote of any viral sequences except the ITRs and cis-acting packaging sequences [184].

The advantages of Ad vectors include their ability of targeting dividing as well as quiescent cells, a broad host range [185, 186], the stability of recombinant vectors, the relatively large insert capacity for foreign nucleic acid, the non-oncogenicity and the possibility of producing high titers. These advantages are, however, accompany with disadvantages, including difficulty in limiting tissue tropism due to the widely expressed nature of CAR [186], transit transgene expression owing to episomal nature of vector genome, and humoral as well as cellular host immune response which limit the repeated application of the vector thus hindering its application in the treatment of hereditary diseases [187].

Wild type Ad is released from the cells after disrupting them which is the basis of the oncolytic characteristic of replication competent vectors based on them. These oncolytic vectors are used for treatment of several tumours including malignant glioma, head and neck tumors and non-small cell lung cancers [188]. Beside the treatment of the cancer, Ad vectors have been used to treat ornithine transcarbamylase (OTC) deficiency (a urea cycle disorder) [189], cardiovascular diseases [190], pulmonary tuberculosis [191], AIDS [192], diabetes mellitus [193], and stem cell differentiation [194].

1.2.2.4 Adeno-associated virus

Adeno-associated viruses (AAV) belong to genus Dependovirus of the Parvoviridae viral family. The prototype AAV serotype2 was discovered in 1965 as contaminant in a preparation of Simian adenovirus [195]. They are replication defective viruses which depend (and hence Dependovirus) on co-infection of helper viruses, such as adenovirus or herpes virus, or cellular stress, such as irradiation or treatment with genotoxic agents, to complete replication in the nucleus of infected cell [196, 197].

Several serotypes and more than hundred variants of AAV have been isolated from adenovirus stocks or human and non-human primates [198, 199]. A new serotype is “a newly isolated virus that doesn’t efficiently cross-react with neutralizing sera specific for all other existing and characterized serotypes” [200]. AAV1, AAV2, AAV3, AAV4 and AAV6 were isolated as contaminants in adenovirus stocks including simian adenovirus type 15 stock (SV15), adenovirus type 12 stocks and adenovirus type 7 stocks [201-203]. AAV5 was isolated from human penile condylomatous warts [204]. AAV7 to AAV11 were isolated as DNA sequences, by a novel PCR strategy, rather than live viruses [205]. AAV7 and Introduction 17

AAV8 were isolated from rhesus monkeys, while AAV9 from humans [205]. Serotypes AAV10 and AAV11 were isolated from cynomolgus monkeys [198]. AAV genome have also been isolated from other animals beside primates such as lizard [206], snake [207], chicken [208], goat [209, 210], cow [211] and horse [212]. All AAV serotypes display different tissue tropism which, in part, is determined by their specific capsid.

The capsid of AAV is a nonenveloped, icosahedral protein shell which encapsidates the viral genome and comprises the viral proteins (VP) known as VP1, VP2 and VP3. The AAV capsid is made by self assembly of 60 monomers of these three VP in icosahedral symmetry. After the assembly of empty capsid the viral genome is packaged inside the capsid [213]. Capsid assembly and integrity is independent of full-length viral genome replication [214]. The helicase activity of the viral replication protein Rep52 inserts the viral genome into preformed empty capsids [215]. Investigators have shown that VP1 and VP3 together are sufficient for producing infectious virus particles [216, 217] as deletion of VP2 did not affect the infectivity of virus particles [218]. Each serotype of AAV has a characteristic capsid, with the motifs on the surface, which have specific affinity to particular host cell receptors leading to specific transduction efficiencies of each serotype in different tissues. To understand tissue tropism knowledge of AAV binding to specific host cell receptors are of prime importance. AAV2 utilizes heparin sulfate proteoglycan (HSPG) as primary receptor and at least three different co-receptors including αVβ5 integrin, human fibroblast growth factors receptor-1 (FGFR-1), or hepatocyte growth factor (c-met), to facilitate the entry into the cell [219, 220]. AAV3 binds to heparin, heparin sulfate and FGFR-1 [221]. AAV4 binds to α2-3 O-linked sialic acid while AAV5 utilize platelet-derived growth factor receptor (PDGFR) as a primary and α2-3 N-linked sialic acid as coreceptors [222-224]. Closely related AAV1 and AAV6 used α2-3 linked as well as α2- 6 linked sialic acid as primary receptors [225]. Studies also showed that the 37/36-kDa laminin receptor (LamR) is important in binding and transduction of AAV8, AAV9 and AAV2 [226]. Receptors for AAV7, AAV10 and AAV11 are unknown.

The AAV genome is a linear, single-stranded DNA molecule with two open reading frames (ORF), encoding for non-structural replication proteins (Rep) and structural proteins (Cap), flanked by 145 bp long ITRs sequence on each side [227]. It has three promoters at map positions 5, 19 and 40. The 5`ORF (Rep) contains four overlapping Rep genes which encode for the four multifunctional replication proteins Rep78, Rep68, Rep52 and Rep40, named according to their molecular weight in kilodaltons (kDa) [228]. Rep78 and Rep68 are translated from p5 promoter transcript while Rep52 and Rep40 from p19 promoter transcript by spliced and unspliced transcripts variants. The 3`ORF contains three overlapping genes which encodes for three capsid proteins, namely VP1, VP2 and VP3 from spliced and unspliced transcript of the single p40 promoter with molecular weights of 87, 72 and 62 kDa, respectively [229].

The 145bp ITRs at each end of the AAV genome have palindromic regions which fold into a T-shaped secondary structure through self-basepairing. ITRs are divided into four sub- Introduction 18 regions denoted as A, B, C, and D. Among these, B and C form 44bp asymmetric small internal palindromes that make the arm of the T strucure, whereas A is a symmetric palindrome that flanks B and C to make the stem of the T structure. The D sub-region is present in one copy and thus remains single-stranded. The 3` ends of the ITR serves as primer and allow primase independent synthesis of the second strand of the viral genome. In addition, ITRs also serve as origin for the replication of DNA. The two large Rep proteins (Rep78 and Rep68) bind to the tetranucleotide tandem “GAGC” repeats within the A palindromic subregion of ITR, called Rep binding element (RBE), which is also referred to as Rep recognition sequence (RRS) [230, 231]. This Rep-ITR complex is further stabilized by binding of Rep proteins to the internal palindrome of ITR known as RBE´ [230, 232, 233]. For the complete conversion of single-stranded viral genomes into double-stranded genomes, resolution of the hairpin structure at the 3´ITR is necessary. This resolution, known as terminal resolution, is achieved by Rep mediated ATP-dependent isomerization of the palindromic A subregion at the junction of the A and D subregions that lead to the exposition of terminal resolution site (trs). Rep protein then produce a sequence-specific and strand-specific nick at the trs, on one strand, unwind the 3´ hairpin structure which leads to completion of replication. Although in addition to the ITR another cis element, known as cis-acting replication element (CARE), which augments replication [234] is present in the AAV genome. Unlike the large Rep proteins, the two small Rep proteins (Rep52 and Rep40) are not directly involved in replication but help in accumulation of the single stranded progeny genomes and their encapsidation in preformed empty viral capsids [235].

The life cycle of the AAV2 has been much studied and it begins with the attachment of viral capsid to specific host receptor and co-receptor. The virus than enter the cells by receptor mediated endocytosis [236] which is followed by endosomal processing, an important step for nuclear transport of the virus [236]. AAV is then released from endosomes before it traffics to the nucleus [236]. The mode of AAV movement includes free and anomalous diffusion of the virus/endosome in the cytoplasm and directed motion to nuclei by motor proteins in the cytoplasm and nucleus [237]. Nuclear transportation of the AAV particles takes place prior to uncoating [238]. However nuclear transport appears to be a slow and inefficient process [236]. Once inside the nucleus replication of AAV and assembly of new virus takes place followed by lysis of the cell and release of virus. However, in non- permissive conditions (such as absence of helper virus) AAV DNA either integrates into the host genome at a specific locus to establish a latent proviral state or remain episomally by forming concatemers [239]. In human cells, AAV2 integrates in the AAVS1 locus on chromosome 19 by binding of the AAV Rep protein [230] simultaneously to AAVS1 locus and the AAV ITR sequence [240].

Recombinant vectors based on adeno-associated virus (rAAV) are attractive tools for gene transfer. In these “gutted” rAAV vectors all viral genes are replaced with a transgene cassette including promoter and polyadenylation signal [241]. In the absence of Rep protein rAAV remain largely episomally in the nucleus. In first generation rAAV packaging Introduction 19 systems, the vector construct was co-transfected with a packaging plasmid, which encode the viral rep and cap genes, into helper virus infected cells [242]. But, as in this procedure helper virus is also produced, it leads to the contamination with helper virus and need more laborious purification and inactivation steps. Later, helper free packaging systems were developed by replacing helper with an adenovirus helper plasmid containing only the adenovirus genes necessary for helper function [243]. This three plasmid system needs the simultaneous transfection of helper plasmid, packaging plasmid and vector plasmid. The most recent packaging system involves the combination of helper and packaging function in one single plasmid and hence need the co-transfection of only two plasmid, packaging/helper and vector, for production of AAV [244].

The traditional purification protocol of AAV, which was based on ammonium sulfate precipitation of AAV particle and multiple round of density ultracentrifugation in cesium chloride gradient [245], has been replaced by iodixanol. This new method gives at least 100-fold more purification of the crude cell lysates [246]. For more pure grade rAAV, ligand affinity chromatography purification technique has been developed which is based on the use of cellular receptor or its commercial analogues [246-248]. This ligand affinity chromatography together with iodixanol produces over 99% pure rAAV vectors with good particle-to-infectivity ratio (100:1) [246].

The rAAV vectors have a number of features that make them favourable to use as gene delivery vehicle. These include their broad host range as well as restricted cell/tissue type tropism, transduction of dividing as well as quiescent cells, non-integrating episomal nature of vector genome in host cell nucleus [139, 249], maintenance of transgene expression over long period of time (years), apparent lack of pathogenicity by parental wild type AAV and low immunogenicity. Beside these advantages they also have certain limitations which include the relative small capacity for gene of interest, the slow onset of transgene expression, the production which, especially for high titers, is labour-intensive and costly, and the concerns about potential integration (although at low frequencies) and possible malignancy.

Owing to its relative safety and selective tropism, rAAV is on of the best choice for in vivo gene delivery and has been used to deliver transgene to a variety of organs [200]. In rAAV mediated gene therapy trials the main focus remains monogenic diseases (53%), such as haemophilia, followed by cancer (23%) [250].

1.2.2.5 Other viral vectors

Beside above mentioned viral vectors, vectors based on viruses like baculoviruses [251], herpes simplex virus (HSV) [252] and poxviruses [253] are used. Also several hybrid vectors based on different viruses are being employed such as HSV/AAV [254, 255] and AAV/Ad [256] hybrid vectors. In addition, hybrid vectors combining merits of non-viral and viral systems by incorporating viruses or their components into non-viral vector systems, Introduction 20 such as virus–liposome–DNA complexes [257] and lipid–DNA mixed with the VSV glycoprotein [258] have been developed.

1.3 Zinc finger nuclease

Different targeting reagents can be used to achieve or/and enhance any of the different modes of genome editing. These targeting reagents include small interfering RNA (siRNA), Sleeping Beauty transposon, meganucleases, ZFNs, and Transcription Activator-Like Effector Nucleases (TALENs).

Among the well characterized targeting reagents, ZFNs have proved to be the most versatile and efficient. They are synthesised chimeric proteins that consist of a separate specific DNA binding domain and an unspecific cleavage domain. The DNA binding domain is based on sequence-specific zinc fingers. Arrays of these zinc fingers are linked with non specific endonuclease domain of the restriction enzyme FokI.

The term zinc fingers (ZFs) are used broadly to identify any compact domain that is stabilized by a zinc ion [259]. They are a structurally and functionally diverse group that are involved in several fundamental cellular processes, including replication and repair, transcription and translation, signalling, cell proliferation and apoptosis. ZFs typically are interaction modules and binds to nucleic acids or proteins. They can be classified into eight groups based on the structural properties in the vicinity of the zinc binding site. These groups include Cys2His2-like (such as the classical C2H2 zinc finger motif), gag knuckle, treble clef, zinc ribbon, Zn2/Cys6, TAZ2 domain like, short zinc binding loops and metallothioneins [260].

The ZFs used as sequence-specific DNA binding domain in ZFNs are based on the C2H2 zinc finger motif. These ZFs contain 20-30 amino acids including two cysteines and two histidine residues which bind the zinc atom [259]. Individual zinc finger consist of two antiparallel β sheets, which contain the two cysteines residue, and an α helix, which harbours the two histidine residues. The zinc ion held’s the two structural units together whereas the α helix makes sequence-specific DNA contacts in the major groove through hydrogen bonding. Each finger recognizes about three nucleotides (triplets) by a one-to- one interaction of single amino acids with a specific single DNA base [261]. The individual ZFN generally contain three to six zinc fingers, each one recognizing three nucleotides. Engineered zinc finger which bind to desired sequences can be developed by different strategies, including modular assembly [262], the selection based oligomerized pool engineering (OPEN) system [40, 263], and context-dependent assembly (CoDA) [264].

Introduction 21

Figure 1: Architecture and application of ZFNs. A ZFN designed to create a DNA double-strand break (DSB) in the target locus is composed of two monomer subunits. Each subunit encompasses three zinc-fingers (orange, 1-2-3), which recognize 9 base pairs within the full target site, and the FokI endonuclease domain (green). A short linker (grey) connects the two domains. After dimerization the nuclease is activated and cuts the DNA in the spacer sequence, separating the two target half-sites (L) and (R). (Adopted from Cathomen & Joung, 2008)

The FokI is a bacterial type IIS restriction endonuclease naturally found in Flavobacterium okeanokoites. This enzyme consist of two functionally distinct domains the N-terminal DNA binding and the C-terminal DNA cleavage domain [265]. The binding domain is specific and binds to a particular sequence of the DNA while the cleavage domain is unspecific and cuts DNA phosphodiester bonds. The cleavage domain is similar to a BamHI monomer and contains a single catalytic centre [266]. In the unbound state, the binding domain sequesters the cleavage domain through protein-protein interactions until its activity is required. Through a conformational change, the cleavage domain swings over to the major groove for DNA cleavage [266]. The cleavage domain can be separated from the intrinsic binding domain and fused to an unrelated DNA binding domain, such as zinc finger arrays [267]. The unspecific nuclease domain of FokI requires the dimerization of two subunits to achieved cleavage of DNA [266, 268]. Efficient cleavage of DNA thus needs a pair of two ZFN monomers binding to neighbouring sequences. Upon binding to recognition sequences ZFs facilitate dimerization of FokI and subsequent cleavage of DNA.

The functionality of the ZFN was initially evaluated in vitro [269] with artificial DNA substrates and later in Xenopus eggs [270]. With the advancement of the ZFN technology it has been successfully applied for editing of endogenous genomes from a variety of organisms, including maize [271], tobacco [272, 273], petunia [272], fruit fly [274], nematode [275], silkworm [276], zebrafish [277], sea urchin [278], frog [279], rat [280], mouse [281], hamster [282], pig [283], and human [284].

ZFNs has been used for different strategies in genome editing including targeted gene addition [20, 285], targeted gene correction [32], targeted gene disruption [286] and the creation of chromosomal rearrangements [63, 65]. Although targeted genome editing by ZFNs is achieve by a specific DNA binding domain but the binding domain may recognize unspecific sequences which lead to undesired “off-target” cleavage that not only reduce the “on-target” genome modification but also exert toxicity [287]. These toxicities can be reduced by optimising reagents being delivered to the cells. Aims of the thesis 22

2. Aims of the thesis

There are many human inherited and acquired chronic diseases, which still lack permanent treatment besides supportive care. Inherited genetic disorders are responsible for a large number of admissions in children's [288] as well as adult hospitals [289]. The most obvious treatment to these diseases is the editing of the mutated genome in order to correct the phenotype. ZFNs are designer targeting reagents which help to achieve this goal by allowing researchers to specifically edit the genome by producing DNA double strand breaks and harnessing the activated DNA repair pathways.

The goal of achieving permanent treatment for genetically inherited diseases will rely on the genome editing in transplantable, patient-derived autologous stem cells. The first goal of this thesis was to establish ZFNs to modify the genome of keratinocyte stem cells (KSCs). For this proof-of-principle approach, I planned to optimize the architecture of EGFP-specific ZFNs to identify nucleases with high activity and low associated toxicity. While our collaboration partners assessed the impact of ZFN mediated EGFP knockout on the stem cell potential of KSCs, I have planned to do the molecular characterization of the outcome of ZFN-mediated gene disruption.

Efficient genome editing is based on efficient transfer of targeting reagents to the target cells. The second goal of this thesis was to establish a viral vector delivery system based on adeno-associated virus (AAV). In a proof-of-principle study I aimed to demonstrate the versatility of combining efficient gene transfer technology of AAV vectors with therapeutic genome editing technology of ZFNs for rational editing of the human genome and to assay platform-associated toxicity.

Keeping in view the importance of in vivo genome editing as well as the safety and efficiency record of AAV in clinical and preclinical studies, I aimed at establishing a protocol for ZFN mediated in vivo gene knockout. For this proof-of-principle project in a transgenic EGFP mouse, I designed a single construct containing both subunits of a ZFN pair and utilized the AAV-ZFN platform for in vivo genome editing of cardiomyocytes after a single viral vector administration.

In summary, this doctoral thesis aimed to improve targeted genome editing technologies based on ZFNs and achieve proof-of-principle genome editing in different experimental setups, including in vitro, ex vivo and in vivo settings, to pave the way for therapeutic applications of this technology for different human diseases.

Publication 1 23

3. Publication 1

Highly Efficient Zinc-Finger Nuclease-Mediated Disruption of an eGFP Transgene in Keratinocyte Stem Cells without Impairment of Stem Cell Properties. Stem Cell Reviews and Reports, 2011

Author’s contribution statement:

This study was completed in collaboration of Dr. Julia Reichelt research group. I constructed, optimized and characterized the zinc finger nuclease expression plasmids used in this study and performed all the genotyping for qualitative and quantitative analysis of EGFP disruption on the genome level (Figure 2 a and b). In addition, I contributed to writing manuscript.

Stem Cell Rev and Rep DOI 10.1007/s12015-011-9313-z

Highly Efficient Zinc-Finger Nuclease-Mediated Disruption of an eGFP Transgene in Keratinocyte Stem Cells without Impairment of Stem Cell Properties

Thorsten Höher & Lee Wallace & Kafaitullah Khan & Toni Cathomen & Julia Reichelt

# Springer Science+Business Media, LLC 2011

Abstract Zinc-finger nucleases (ZFNs) are sequence-specific they expressed the in vivo KSC markers K15, NFATc1 and genome engineering tools with great potential for the Sox9. Our data suggest that the stem cell potential of KSCs is development of gene therapies. The achievement of permanent not impaired by highly efficient ZFN treatment. cures through gene therapy requires targeting of stem cells but the effects and/or side effects of ZFN treatment on adult stem Keywords Keratinocytes . Stem cells . Gene targeting . cell potency are largely unknown. Keratinocyte stem cells Mice . Cytotoxicity. EGFP protein . Gene therapy (KSCs) are attractive candidates for the development of gene therapies as their isolation, culture and grafting are well established. We derived KSCs from eGFP-transgenic mice Introduction and knocked out eGFP expression by disrupting the open reading frame with specific ZFNs in cell culture. EGFP- Gene therapy is currently the most promising approach for the negative KSCs were then used as a model system to study the treatment of monogenetic diseases [1]. To achieve permanent impact of ZFN treatment on stem cell potential. We achieved cure of an affected tissue by gene therapy, targeting of organ- high gene disruption efficiencies with up to 18% eGFP- specific stem cells is advantageous. For several reasons, negative KSCs. As expected, ZFN cytotoxicity increased with genetic blistering skin diseases represent ideal targets for the rising ZFN concentrations. However, the ratio of correctly development of ex vivo gene therapy approaches involving targeted KSCs among total treated cells was similar at different stem cells. Firstly, skin is an easily accessible organ. ZFN doses. Most importantly, our in vitro assays showed that Secondly, isolation, propagation and grafting of keratinocyte ZFN-treated KSCs maintained their stem cell potential. They stem cells (KSCs) are well established and have been widely retained the capacity to both self-renew and form fully used to treat burn patients for more than 30 years [2, 3]. differentiated epidermal equivalents in culture. Moreover, they Thirdly, there is a high therapeutic demand for inherited skin were able to form spherical aggregates in suspension culture, a blistering conditions which are extremely painful, debilitat- characteristic hallmark shared with other stem cell types, and ing, often involve devastating disfigurement to the point of mutilation and are potentially life-threatening. Two major requirements for the clinical use of gene : : * T. Höher L. Wallace J. Reichelt ( ) therapies are safety and high efficiency. In their pioneering Institute of Cellular Medicine, Newcastle University, Framlington Place, Newcastle upon Tyne( NE2 4HH, UK work, Mavilio et al. [4] applied gene therapy successfully in a e-mail: [email protected] patient with junctional epidermolysis bullosa, caused by URL: http://www.ncl.ac.uk/icm/people/profile/julia.reichelt recessive mutations in the laminin 5 gene, by grafting URL: http://www.nesci.ac.uk/about/team/profile/julia.reichelt autologous KSCs which were virally transduced with laminin J. Reichelt 5 cDNA. Although highly efficient, retrovirus-mediated gene North East England Stem Cell Institute, Newcastle University, therapy bears several risks and may cause insertional Newcastle upon Tyne, UK mutagenesis resulting in carcinogenesis [5]. Even novel self- inactivating lentiviral vectors, which promise to represent a K. Khan : T. Cathomen Hannover Medical School, Institute of Experimental Hematology, safer tool, can be genotoxic by generation of viral-cellular Hannover, Germany fusion transcripts from proviral integration sites [6]. Stem Cell Rev and Rep

Hereditary blistering skin diseases are heterogeneous and GGACGGC-3′) of the eGFP open reading frame [12]into caused by a variety of different mutations. One major group pRK5 vectors [13] containing the obligate heterodimeric are dominant-negative keratin mutations which are ex- KV/EA FokIvariants[14]. tremely difficult to treat as recovery of normal tissue A KSC line (HD-KSC) was established by serial function requires ablation of the mutant keratin [7]. One cultivation of primary keratinocyte cultures [15] derived current line of research aims to employ RNAi technology to from the epidermis of RA/EGxdelCre mice. Briefly, suppress expression of dominant-negative keratin mutations epidermis was recovered from neonatal mice after over [8]. However, the delivery of siRNA and their stable night incubation on dispase (Invitrogen, Paisley, UK) expression in skin remain to be established, and transient solution (5 U/ml) at 4°C. The epidermis was then minced expression of siRNA would require repeated treatment of using scalpels and agitated for 30 min at 1000 rpm in FAD patients. In contrast to siRNA-driven suppression of medium (FAD medium: DMEM/HAM’s F12, 3.5 : 1.1 with dominant-negative mutant gene expression, zinc-finger 0.05 mM Ca2+ (Biochrom, Berlin, Germany), 10% FCS nuclease (ZFN) technology is a highly efficient tool to (FCS Gold, PAA), 0.18 mM adenine, 0.5 μg/ml hydrocor- knock out genes based on the error-prone non-homologous tisone, 5 μg/ml insulin, 10−10 M cholera toxin (all from end joining (NHEJ) pathway [9]. ZFN-mediated knockout Sigma, Gillingham, UK), 10 ng/ml EGF (Invitrogen), of chemokine receptor 5 (CCR5) was used in primary T 2 mM glutamine, 100 U/ml penicillin, 100 μg/ml strepto- cells of HIV patients to render the cells resistant to infection mycin, 250 ng/ml amphotericin B (CnT-ABM, Cellntec, with the virus [10] and a phase I clinical trial is in progress Switzerland)). Prior to addition of FCS to the medium, FCS [11]. ZFNs are engineered endonucleases which can be was treated with 2 g Chelex 100 (Bio-Rad, Hempstead, designed to recognize a specific target site within a complex UK) per 50 ml serum rotating at 4°C over night to remove genome via their zinc-finger domains. Upon binding of two Ca2+, and then sterile filtered. Keratinocytes were co- ZFN subunits in correct orientation and spacing, the unspe- cultured with inactivated 3T3-J2 fibroblast feeder cells for cific FokI endonuclease domains dimerize and generate a the first 8 passages. Culture dishes (TPP, Trasadingen, DNA double-strand break in the target gene. In the absence Switzerland) were coated with 50 μg rat tail collagen I (BD of homologous DNA, the double-strand break is repaired by Biosciences, Oxford, UK)/ml 0.02 M acetic acid for 1 h and NHEJ, which frequently causes frameshift mutations, result- washed twice with PBS (Lonza, Tewkesbury, UK). Kerati- ing in functional deletion of gene expression. This mecha- nocytes were cultivated at 32°C and 5% CO2. Medium was nism can be harnessed to inactivate dominant-negative changed twice weekly and cultures were split at 1:1 for the mutant genes such as disease-causing mutant keratin alleles. first 3–4 passages, and thereafter at 1:2 or 1:3. For splitting, A great advantage of the ZFN technology, in contrast to keratinocytes were washed with PBS, incubated with 0.02% virus-based gene therapy, is that ZFNs do not persist in the EDTA in PBS for 5 min and then for 8–12 min with 0.05% treated cells. Following ex vivo treatment, ZFN expression trypsin/0.02% EDTA in PBS (Invitrogen) at 37°C. will therefore recede before targeted cells are grafted. 3T3-J2 fibroblast feeder cells were grown in DMEM Furthermore, ZFN-induced modifications are permanent medium (Invitrogen) with 10% FCS and antibiotic/anti- and passed on to daughter cells further supporting the mycotic solution. Prior to co-cultivation with keratinocytes, therapeutic potential of this approach. feeder cells were inactivated with 0.4 μg/ml mitomycin C A prerequisite for the application of ZFN technology for (Sigma) for 2 h at 37°C. the treatment of KSCs from patients with genodermatoses is that the stem cell potential is maintained following ZFN Transfection of Keratinocytes treatment. Using transgenic eGFP-expressing murine KSCs, we provide proof-of-principle that ex vivo ZFN-mediated For determination of ZFN efficiency and cytotoxicity functional deletion of targeted genes is highly efficient and, (Figs. 1 and 2): HD-KSCs, passage (P) 27 were grown in most importantly, our results indicate that the characteristic 12-well plates and transfected at 50% confluence using stem cell properties of self-renewal and differentiation are lipofectamine 2000 (Invitrogen) with 0.4, 0.6, 0.8, 1.0 and not impaired by ZFN treatment. 1.2 μg of eGFP-specific ZFN expression plasmid/ml/well. After 48 h, cells were transferred to 6-well plates and further grown for 6–8 days to allow for eGFP protein Materials and Methods degradation. The increase in eGFP-negative keratinocytes was measured by flow cytometry. Plasmids and Cell Culture For determination of stem cell potential after ZFN treatment (Figs. 3 and 4): HD-KSCs (P27) were seeded in ZFN expression vectors were generated by subcloning a ZFN 6-well plates and transfected at 50% confluence with 0.6 or pair specific for position 502 (5′-ATCCGCCACnnnnnnGA 1.0 μg/ml total ZFN plasmid using lipofectamine 2000. Stem Cell Rev and Rep

RFig. 1 a Side scatters (left column), and cell numbers (right column) from flow cytometry analysis were plotted against eGFP intensities. Mock-transfected controls showed a background of 0.9% eGFP- negative cells which was subtracted from the measurements of ZFN- transfected samples to yield the percentage of eGFP-negative KSCs (ΔGFP-). For mock-transfected HD-KSCs, and HD-KSCs transfected with 0.4–1.0 μg ZFN DNA/ml/12well, the total number of counted cells was 105. Note that for HD-KSCs transfected with 1.2 μg ZFN DNA/ml/12-well only 2×104 cells were counted due to high cell loss caused by ZFN toxicity. b Results from several independent flow cytometry experiments showed that ZFN efficiency increased in a linear manner with the amount of transfected ZFN DNA. Data points represent means +/− SEM, n≥4; R2, coefficient of determination

Transfected KSC cultures were termed ZFN600-KSC and ZFN1000-KSC, respectively. After 48 h, cells were trans- ferred to 25 cm2, later to 75 cm2 flasks and further grown for 6–8 days.

Flow Cytometry and Fluorescence-Activated Cell Sorting (FACS)

Flow cytometry was used to determine ZFN efficiency. ZFN-treated HD-KSCs were trypsinised, resuspended in PBS, and DAPI was added to identify dead cells. DAPI- positive cells were excluded from counting. EGFP-negative and eGFP-positive cells were quantified using a FACS- Canto II (BD Biosciences). FACS was used for isolation and subsequent cultivation of eGFP-negative and eGFP-positive populations of ZFN- treated HD-KSCs. ZFN600-KSCs and ZFN1000-KSCs were trypsinised and resuspended in FAD medium contain- ing 0.8% FCS (Chelex-treated). DAPI was used to exclude dead cells from sorting. The eGFP-negative population was separated from the eGFP-positive population using a FACSAria II (BD Biosciences). Sorted eGFP-negative ZFN600-KSCs and ZFN1000-KSCs were immediately plated and expanded in culture for further analysis. EGFP fluorescence was measured using a BP 530/30 filter. BD FACSDiva™ software version 6.1.3 was used for analysis.

Genotyping

The T7 endonuclease I (T7E1) assay was performed as previously described [16] using genomic DNA from sorted eGFP-negative ZFN1000-KSCs and from HD-KSCs (con- trol) as templates. In short, 100 ng of the purified 544-base pair PCR amplicon, which contains part of the ZFN binding site of the eGFP gene (primers #76 5′-taaacggccacaagtt cagcgt and #185 5′-gtgctcaggtagtggttgtcg), was melted at 95°C and reannealed at room temperature to allow for the formation of heteroduplex DNA. The reannealed DNA was treated with 5 U of T7E1 (New England Biolabs, Frankfurt, Germany) for 20 min at 37°C, and fragments separated on a 2% agarose gel. Stem Cell Rev and Rep

Part of the eGFP gene containing the ZFN binding site was amplified by PCR using 100 ng of genomic DNA from sorted eGFP-negative ZFN1000-KSCs as template, along with 10 μM of each primer (#850 5′-atcgacttcaaggaggacggc and #185 5′-gtgctcaggtagtggttgtcg), 10 mM dNTPs, and 0.5 U of Phusion High-Fidelity DNA Polymerase (Fisher Scientific, Schwerte, Germany) in 1x reaction buffer for 30 cycles. The 227-base pair PCR amplicon was purified and cloned into pCR4Blunt-TOPO vector using Zero Blunt TOPO PCR Cloning Kit for Sequencing (Invitrogen, Darmstadt, Germany). DNA from 12 single clones was then sequenced using M13 forward primer provided with the kit.

Cytotoxicity Assay

Cell viability was assessed 24 h after ZFN plasmid transfection by determining keratinocyte numbers using a hemocytometer. HD-KSCs (P27) were seeded in 12-well plates and transfected as described above. As a control to test for DNA cytotoxicity, HD-KSCs were transfected with the same amounts of DNA (0.4–1.2 μg/ml/12-well) of an unrelated plasmid (pEGFP-K10). All experiments were performed in duplicate and Student’s t-test was used for statistical analysis.

KSC Aggregate Formation Assay

Keratinocyte aggregates were grown in Aggrewell-400 plates (Stemcell Technologies, Grenoble, France). In order to generate aggregates of a size of 100 cells/aggregate, the two eGFP-negative KSC populations of ZFN600-KSCs and ZFN1000-KSCs, obtained by FACS, as well as untreated HD-KSCs (all P29) were seeded at a density of 1.2×105 cells/well in high calcium (1.2 mM Ca2+) FAD medium in Aggrewell plates. 48 h after seeding, aggregates were harvested into a 50 ml reaction tube and left for 5–10 min prior to removal of the supernatant and transfer into a Fig. 2 a T7E1-based genotyping confirmed the presence of sequence 1.5 ml reaction tube in 1 ml medium. After aggregates had modifications in the targeted eGFP sequence. Genomic DNA was settled, excess medium was removed, aggregates were re- isolated from cells transfected with E502-specific ZFN expression – μ plasmids (ZFN1000-KSCs; lanes 1, 2) or mock-transfected cells (HD- suspended in 50 60 l of 1.2% alginate (Sigma) in PBS, KSCs, lanes 3, 4). PCR amplicons encompassing the ZFN targeted site added to 50 ml 102 mM CaCl2 and incubated for 10 min to were then digested with T7E1 or left untreated (+ or -, as indicated at set. Once set, using forceps the aggregate-containing the bottom). The percentage of modified alleles is indicated at the droplets were placed into a drop of OCT-compound (Fisher bottom. b Sequence analysis of cloned PCR products from genomic DNA of sorted eGFP-negative ZFN1000-KSCs revealed several Scientific, Loughborough, UK), and frozen in isopentane, distinct mutations in the targeted sequence of the eGFP transgene. pre-cooled over liquid nitrogen. To simplify aggregate The sequences of the recognition sites of the left (E502L) and right droplet recovery, OCT-compound (Fisher Scientific) was subunit (E502R) of ZFN E502 are indicated by upper case letters, dots pre-colored with 1% neutral red solution. indicate deletions and dashes accommodate the detected insertion. c The graph shows the number of viable cells 24 h after transfection of HD-KSCs with ZFN DNA at concentrations of 0.4 - 1.2 μg/ml/12- KSC Differentiation and Generation of Epidermal well. Results are shown in means +/− SEM, n=16; ***, P<0.001. d Equivalents Although the fractions of dead cells differed strongly between KSCs μ treated with 0.6 and 1.0 g ZFN plasmid DNA/ml/12-well (white μ areas) the fractions of correctly targeted (eGFP-negative) KSCs were Filter inserts (Millicell PCF, 0.4 m, 12 mm, Millipore, similar in the two populations (black areas) Watford, UK) were coated with collagen I prior to seeding Stem Cell Rev and Rep

Fig. 3 Immunofluorescence analysis of KSC aggregates generated NFATc1 and absence of K1 in both samples. Cy3 (red) and DAPI from untreated HD-KSCs and from sorted eGFP-negative ZFN600- (blue) were used for visualization of specific antigens and nuclei, KSCs showed similar expression patterns for K5, K15, Sox9, and respectively; scale bar, 100 μm

2× 105 keratinocytes per insert in FAD medium. Inserts Immunofluorescence Analysis were placed in 6 cm dishes and media levels of dishes were adapted to those of the inserts. The two eGFP-negative Cryosections (5 μm) of epidermal equivalents and KSC populations of ZFN600-KSCs and ZFN1000-KSCs, isolated aggregates were fixed for 10 min with acetone at −20°C, or by FACS, as well as untreated HD-KSCs (all P30) were grown (for detection of Sox9 and NFATc1) with 4% formaldehyde/ submerged for 2 days until confluence was reached. The PBS. Antigen retrieval after formaldehyde fixation was medium was changed to high calcium (1.2 mM Ca2+)FAD performed using 10 mM citrate buffer and a microwave oven medium including 50 μg/ml vitamin C and cultures were at sub-boiling temperature for 20 min. Immunofluorescence further grown submerged for 16 h. The medium was then analysis was performed as described before [17]. Antibodies aspirated and keratinocytes were grown for 12 days at the air- were diluted as follows: Ki-67 (Dianova, Hamburg, liquid interphase with medium levels in the 6 cm dishes kept Germany), 1:50; K10 (DE-K10, Progen, Heidelberg, Germany) up to the filter level. Medium was changed every 1–2days. 1:10; filaggrin (PRB-417P, Covance, Berkeley, CA), 1:1000; Filters with epidermal equivalents were excised, embedded in K5 (PRB-160P, Covance), 1:1000; K1 (PRB-165P, Covance), OCT-compound (Fisher Scientific, Loughborough, UK), 1:1000; K15 (GP-15.1, Progen), 1: 200; Sox9 (sc-20095, Santa frozen in liquid nitrogen and stored at −80°C. Cruz, Heidelberg, Germany), 1:50; NFATc1 (sc-7294, Santa Stem Cell Rev and Rep

average 18% eGFP-negative keratinocytes among the sur- viving cell populations (Fig. 1a and b). NHEJ-induced ZFN-mediated target site modifications in sorted eGFP-negative ZFN1000-KSCs was verified using the mismatch-sensitive endonuclease T7E1 (Fig. 2a). The deter- mined amount of 68% of mutated alleles is likely an underestimation as quantifications above 50% are not accurate. Furthermore, we have sequenced parts of the targeted eGFP transgene derived from genomic DNA from sorted eGFP-negative ZFN1000-KSCs and detected dele- tions of 1, 2 and 24 base pairs as well as an insertion of 2 base pairs (Fig. 2b). The genotyping data confirmed that the loss of eGFP expression in eGFP-negative ZFN1000-KSCs resulted from ZFN action. ZFN are known to exhibit cytotoxicity [19]. In our experimental setting, ZFN plasmid amounts ≤0.6 μg/ml/12- well did not cause significant cytotoxicity (Fig. 2c). At Fig. 4 HD-KSCs and sorted eGFP-negative ZFN600-KSCs showed ≥ μ differentiation and formation of epidermal equivalents in high calcium plasmid concentrations 0.8 g/ml/12-well, cytotoxicity medium at the air-liquid interphase. Immunofluorescence analysis increased significantly with rising plasmid concentrations. revealed that proliferating, i.e. Ki-67-positive, keratinocytes were The observed ZFN-mediated cytotoxicity at DNA concen- confined to the basal layers, K10 was expressed throughout suprabasal trations ≥0.8 μg/ml/12-well was unrelated to the amount of layers and filaggrin expression was restricted to terminally differen- tiated keratinocytes in the upper suprabasal layers. Cy3 (red) and exogenous DNA used for transfection as in control experi- DAPI (blue) were used for visualization of specific antigens; dashed ments using HD-KSCs, the same amounts of DNA of an lines indicate basement membranes; scale bar, 100 μm unrelated plasmid showed no detectable cytotoxicity (data not shown). Together, the results from flow cytometry and cytotoxicity assay revealed a range of lower ZFN plasmid Cruz), 1:50. Secondary antibodies were species-specific Cy3- concentrations (<0.8 μg/ml/12-well) which resulted in high conjugated goat antibodies (Molecular Probes, Eugene, OR). gene disruption efficiencies of at least 7.6% (on average at Nuclei were stained with DAPI (Invitrogen) and microscopical 0.6 μg/ml/12-well) without causing significant cytotoxicity. analysis performed using an Axio Imager operated through The numbers of eGFP-negative HD-KSCs obtained from Axiovision software (Zeiss, Hertfordshire, UK). flow cytometry (Fig. 1b) were projected on the total number of cells used for ZFN transfection employing the ZFN cytotoxicity data determined 24 h after transfection of Results HD-KSCs (Fig. 2c), allowing us to compare the absolute fractions of eGFP-negative KSCs at ZFN plasmid concen- EGFP-expressing KSCs were derived from heterozygous trations below (0.6 μg/ml/12-well) and above (1.0 μg/ml/ RA/EGxdelCre mice carrying one copy of the eGFP 12-well) the determined cytotoxicity threshold (Fig. 2d). transgene per cell [18]. The resulting KSC line, termed Although ZFN cytotoxicity-induced cell loss upon treatment HD-KSC, was used to study the efficiency of ZFN- with plasmid concentrations of 0.6 μg/ml/12-well (5.3%, not mediated functional deletion of the eGFP transgene and significant) and 1.0 μg/ml/12-well (61.5%, significant with its consequences on stem cell potential. We used plasmid P<0.001) differed strongly, the absolute fractions of eGFP- concentrations ranging from 0.4 to 1.2 μg/ml/12-well and negative KSCs were similar in the two populations, at 7.1% show that the efficiency of eGFP gene disruption increased and 6.1% respectively (Fig. 2d). with the amount of transfected ZFN plasmid DNA. The results We demonstrate that both ZFN cytotoxicity and ZFN from a representative flow cytotometry analysis are shown in targeting efficiency of single genes rise in KSCs with Fig. 1a. The experiment was repeated (n≥4) for individual increasing ZFN plasmid concentrations. Our results reveal DNA concentrations and the resulting coefficient of deter- that (i) the relative knockout frequency increases with mination (R2) described a strong linear correlation between increasing ZFN plasmid amounts while (ii) the absolute DNA concentration and targeting efficiency (Fig. 1b, fraction of eGFP-negative cells in the transfected wells R2=0.98). At plasmid concentrations of 0.4 μg/ml/12-well, remains the same. This raises the question whether eGFP- on average 2.6% of the treated KSCs were eGFP-negative. negative KSCs obtained via highly efficient, high ZFN dose Although we observed high cytotoxicity in samples treated treatment accompanied by high cytotoxicity show impaired with 1.2 μg plasmid DNA/ml/12-well, we also detected on stem cell capacity when compared to eGFP-negative KSCs Stem Cell Rev and Rep arising from lower efficiency treatment, without involve- thus far. Gene therapy studies involving adult stem cells in ment of significant cytotoxicity. combination with ZFN technology have until now been We addressed this question and used FACS to isolate restricted to hematopoietic stem cells (HSCs) used for HIV eGFP-negative ZFN600-KSCs (resulting from transfection treatment [10, 22] and mesenchymal stem cells (MSCs) with 0.6 μg/ml ZFN DNA, a dose which did not cause [23]. Beside HSCs and MSCs, KSCs are probably the best cytotoxicity), and eGFP-negative ZFN1000-KSCs (result- investigated, most accessible and transplantable adult stem ing from treatment with 1.0 μg/ml ZFN DNA, a dose which cell type, which make genetic skin diseases and KSCs an caused significant cytotoxicity), and propagated them in ideal target for the development of ZFN-mediated gene culture to test their stem cell specific capacity to self-renew therapies. Our study, using eGFP-transgenic murine KSCs and differentiate. Proliferation of both populations of eGFP- as a model system, is the first to show successful ex vivo negative KSCs was comparable to that of untreated HD- ZFN-mediated disruption of a specific gene in KSCs. Using KSCs, indicating that self-renewal was not affected by lipofection-based gene transfer of ZFN plasmids, we ZFN-treatment (data not shown). Furthermore, both pop- observed a high knockout efficiency of the targeted eGFP ulations of eGFP-negative KSCs were able to cluster and allele in cultured KSCs, consistent with previously pub- form aggregates in suspension culture (Fig. 3), a character- lished gene disruption efficiencies achieved in primary istic hallmark of several types of stem cells, including porcine cells, which were electroporated to allow uptake of KSCs [20]. KSC aggregates generated from both eGFP- ZFN-encoding mRNAs [24]. negative ZFN600-KSCs and eGFP-negative ZFN1000- In our experimental setting cytotoxicity was detected at KSCs showed expression of keratin (K) 15, Sox9 and higher concentrations of ZFN DNA (≥0.8 μg/ml) whereas NFATc1 (Fig. 3), characteristic in vivo markers of multi- lower concentrations, which were still efficient at targeting, potent KSCs residing in the hair follicle bulge region [21]. had no influence on cell viability. ZFNs are known to be Furthermore, KSC aggregates showed expression of K5, genotoxic resulting in high cell death rates among treated which is a marker for undifferentiated keratinocytes, cells. The reasons for this toxicity are not entirely clear but whereas the epidermal differentiation marker K1 was absent off-target nuclease cleavage sites, which are generated (Fig. 3). As eGFP-negative ZFN600-KSCs and ZFN1000- during the short period of time when ZFNs are expressed KSCs showed identical results in the aggregate formation in transfected cells, are one likely cause. Since cytotoxicity assay, data for eGFP-negative ZFN1000-KSCs are not is inversely correlated with ZFN specificity [25] several shown. strategies have been devised to reduce ZFN toxicity, such as In order to test for the differentiation potential of KSCs increasing ZFN specificity and redesigning the FokI after ZFN treatment, we used an established in vitro assay nuclease dimer interface [26]. In order to improve the and generated epidermal equivalents. Upon growth in high safety of ZFN technology for the development of gene calcium medium at the air-liquid interphase both eGFP- therapies involving autologous stem cell grafting, optimi- negative ZFN600- and ZFN1000-KSCs were equally able to zation of ZFN engineering is essential to avoid long-term stratify and form well-differentiated epidermal equivalents side effects. Another prerequisite for using ZFNs in cellular (Fig. 4; eGFP-negative ZFN1000-KSCs are not shown). The gene therapies is that stem cell multi- or pluripotency are equivalents showed proliferation of keratinocytes in the basal not impaired by the treatment. layer, as assessed by Ki-67 staining, and differentiation of Pluripotent stem cells such as embryonic stem cells (ESCs) suprabasal keratinocytes as indicated by expression of the and induced pluripotent stem cells (iPSCs) hold great potential early epidermal differentiation marker K10 and the late for regenerative medicine, and several publications describe differentiation marker filaggrin (Fig. 4). successful genetic manipulation of these stem cells by ZFNs The results from our in vitro assays indicate that highly without affecting pluripotency [22, 27, 28]. Although iPSCs efficient ZFN treatment does not affect stem cell properties are very similar to ESCs [29] and their use as therapeutic or impair stem cell potential of KSCs. tools would circumvent ethical issues associated with ESCs, robustness of epigenetic reprogramming during iPSC gener- ation must be ensured before they can be regarded as safe for Discussion human gene therapies [30, 31]. Adult tissue-specific multi- potent stem cells, such as the well accessible KSCs, represent Permanent cure of genetic diseases via gene therapy will, an excellent alternative to pluripotent stem cells as targets for for most affected organs, rely on targeting of tissue-specific ex vivo gene therapies. stem cells. ZFN technology promises to be an extraordinary We used two in vitro assays to assess the effects of ZFN safe and effective tool for both gene disruption and gene treatment on stem cell potency of KSCs. The assays were correction therapy. The effect of ZFN-treatment on adult applied to two correctly targeted KSC populations (FACS- multipotent stem cell potential has been poorly investigated sorted eGFP-negative KSCs) which differed in the amounts Stem Cell Rev and Rep of ZFN plasmid DNA they had been transfected with. The KSCs is a highly promising approach for the treatment of compared ZFN plasmid amounts were either below genodermatoses caused by dominant-negative mutations in (ZFN600-KSCs) or above (ZFN1000-KSCs) the previously epidermal genes. determined threshold of significant cytotoxicity. Interest- ingly, the potential of surviving stem cells was unaffected Acknowledgements We are grateful to Prof. Bernd Arnold for by the ZFN treatment and both populations showed kindly providing skin from neonatal RA/EGxdelCre mice and Mrs. Sabine Schmitt for excellent technical assistance (DKFZ, Heidelberg, comparable self-renewal and differentiation capacities. The Germany). We thank Newcastle University Cytometry Core Facility unaffected stem cell potential may be explained by highly staff for help with flow cytometry and FACS. Furthermore, we thank efficient repair processes ensuring KSC survival from ZFN Prof. Nick Europe-Finner and Prof. Nick Reynolds for critically side effects and/or efficient elimination of damaged KSCs. reading the manuscript. Our work was funded by the British Skin Foundation, the German Research Foundation (DFG; RE 2673/1-1), Another explanation is that the sorted eGFP-negative the Institute of Cellular Medicine and the Medical Faculty of ZFN1000-KSCs represented only that fraction of the total Newcastle University (JR), and the European Commission’s 7th population of targeted ZFN1000-KSCs which expressed Framework Programme (PERSIST–222878 to TC). KK is funded by ZFNs at levels similar to the amount expressed in the a fellowship from the German Academic Exchange Service (DAAD). ZFN600-KSC population. KSCs expressing higher ZFN Conflict of interest The authors declare no potential conflicts of doses in the population of targeted ZFN1000-KSCs may interest. have died and were lost before sorting. In this study, we targeted and disrupted a transgenic eGFP-allele in murine KSCs by employing ZFN-induced References endogenous NHEJ. It was recently shown that murine KSCs in their in vivo hair follicle bulge niche are able to 1. Wong, G. K., & Chiu, A. T. (2011). Gene therapy, gene targeting repair DNA damage in response to ionizing irradiation and induced pluripotent stem cells: applications in monogenic disease treatment. Biotechnology Advances, 29(1), 1–10. much faster than other, more differentiated, epidermal 2. Gallico, G. G., 3rd, O’Connor, N. E., Compton, C. C., Kehinde, keratinocytes due to highly efficient NHEJ [32]. This rapid O., & Green, H. (1984). Permanent coverage of large burn DNA lesion repair in KSCs was shown to rely upon their wounds with autologous cultured human epithelium. The New increased DNA-PK activity, a key enzyme in the NHEJ England Journal of Medicine, 311(7), 448–451. 3. Rheinwald, J. G., & Green, H. (1975). Serial cultivation of strains pathway [32]. Increased NHEJ in response to ionizing of human epidermal keratinocytes: the formation of keratinizing irradiation was also reported for murine HSCs [33] but not colonies from single cells. Cell, 6(3), 331–343. for human HSCs [34]. The choice between the error-prone 4. Mavilio, F., Pellegrini, G., Ferrari, S., et al. (2006). Correction of DNA damage repair via NHEJ and the high-fidelity junctional epidermolysis bullosa by transplantation of genetically modified epidermal stem cells. Nature Medicine, 12(12), homology-directed repair system seems to depend upon 1397–1402. the cell cycle stage where NHEJ predominates in quiescent 5. Howe, S. J., Mansour, M. R., Schwarzwaelder, K., et al. (2008). cells whereas cycling cells tend to use high-fidelity Insertional mutagenesis combined with acquired somatic mutations homology-directed DNA repair mechanisms in the S/G2 causes leukemogenesis following gene therapy of SCID-X1 patients. The Journal of Clinical Investigation, 118(9), 3143–3150. phase of the cell cycle [33]. If these findings also apply to 6. Almarza D, Bussadori G, Navarro M, Mavilio F, Larcher F, cultured human KSCs, the cell cycle stage of KSCs during Murillas R. (2011). Risk assessment in skin gene therapy: viral- ex vivo ZFN treatment will influence targeting efficiency cellular fusion transcripts generated by proviral transcriptional and should therefore be considered in the development of read-through in keratinocytes transduced with self-inactivating lentiviral vectors. Gene Ther, [Epub ahead of print]. gene therapy treatments. The consequences of an increased 7. Aumailley, M., Has, C., Tunggal, L., & Bruckner-Tuderman, L. NHEJ pathway activity for ZFN-mediated functional (2006). Molecular basis of inherited skin-blistering disorders, and deletion of specific alleles in KSCs are yet unknown. It is therapeutic implications. Expert Reviews in Molecular Medicine, possible that increased NHEJ activity may positively 8(24), 1–21. 8. Hickerson, R. P., Smith, F. J., Reeves, R. E., et al. (2008). Single- influence ZFN efficiency in KSCs, but it may also bear a nucleotide-specific siRNA targeting in a dominant-negative skin higher risk for DNA damage accrual [35]. model. The Journal of Investigative Dermatology, 128(3), In this study we show that ZFNs provide a highly 594–605. efficient tool for the deletion of specific alleles in KSCs and 9. Cathomen, T., & Joung, J. K. (2008). Zinc-finger nucleases: the next generation emerges. Molecular Therapy, 16(7), 1200–1207. that ZFN toxicity does not impair the stem cell potential of 10. Holt, N., Wang, J., Kim, K., et al. (2010). Human hematopoietic KSCs. Further studies will be necessary to investigate long- stem/progenitor cells modified by zinc-finger nucleases targeted to term in vivo effects of ZFN treatment of KSCs, originating CCR5 control HIV-1 in vivo. Nature Biotechnology, 28(8), from possible DNA damage, including carcinogenesis or 839–847. 11. Cannon, P., & June, C. (2011). Chemokine receptor 5 knockout accelerated aging. In combination with the well-established strategies. Current Opinion in HIVand AIDS, 6(1), 74–79. method for isolation, expansion and grafting of autologous 12. Maeder, M. L., Thibodeau-Beganny, S., Osiak, A., et al. (2008). human KSCs 2, ZFN-mediated ex vivo gene therapy of Rapid “open-source” engineering of customized zinc-finger Stem Cell Rev and Rep

nucleases for highly efficient gene modification. Molecular Cell, 24. Watanabe, M., Umeyama, K., Matsunari, H., et al. (2010). 31(2), 294–301. Knockout of exogenous EGFP gene in porcine somatic cells 13. Alwin, S., Gere, M. B., Guhl, E., et al. (2005). Custom zinc-finger using zinc-finger nucleases. Biochemical and Biophysical Research nucleases for use in human cells. Molecular Therapy, 12(4), Communications, 402(1), 14–18. 610–617. 25. Cornu, T. I., Thibodeau-Beganny, S., Guhl, E., et al. (2008). DNA- 14. Szczepek, M., Brondani, V., Buchel, J., Serrano, L., Segal, D. J., binding specificity is a major determinant of the activity and & Cathomen, T. (2007). Structure-based redesign of the dimer- toxicity of zinc-finger nucleases. Molecular Therapy, 16(2), ization interface reduces the toxicity of zinc-finger nucleases. 352–358. Nature Biotechnology, 25(7), 786–793. 26. Handel, E. M., & Cathomen, T. (2011). Zinc-finger nuclease based 15. Reichelt, J., & Haase, I. (2010). Establishment of spontaneously genome surgery: it’s all about specificity. Current Gene Therapy, immortalized keratinocyte lines from wild-type and mutant mice. 11 (1), 28–37. Methods in Molecular Biology, 585,59–69. 27. Hockemeyer, D., Soldner, F., Beard, C., et al. (2009). Efficient 16. Mussolino C, Morbitzer R, Lütge F, Dannemann N, Lahaye T, targeting of expressed and silent genes in human ESCs and iPSCs Cathomen T. (2011). A novel TALE nuclease scaffold enables high using zinc-finger nucleases. Nature Biotechnology, 27(9), genome editing activity in combination with low toxicity. Nucleic 851–857. Acids Res, [Epub ahead of print]. 28. Zou, J., Maeder, M. L., Mali, P., et al. (2009). Gene targeting of a 17. Reichelt, J., Doering, T., Schnetz, E., Fartasch, M., Sandhoff, K., disease-related gene in human induced pluripotent stem and & Magin, A. M. (1999). Normal ultrastructure, but altered stratum embryonic stem cells. Cell Stem Cell, 5(1), 97–110. corneum lipid and protein composition in a mouse model for 29. Guenther, M. G., Frampton, G. M., Soldner, F., et al. (2010). epidermolytic hyperkeratosis. The Journal of Investigative Derma- Chromatin structure and gene expression programs of human tology, 113(3), 329–334. embryonic and induced pluripotent stem cells. Cell Stem Cell, 7 18. Constien, R., Forde, A., Liliensiek, B., et al. (2001). Character- (2), 249–257. ization of a novel EGFP reporter mouse to monitor Cre 30. Sun, N., Longaker, M. T., & Wu, J. C. (2010). Human iPS cell- recombination as demonstrated by a Tie2 Cre mouse line. Genesis, based therapy: considerations before clinical applications. Cell 30(1), 36–44. Cycle, 9(5), 880–885. 19. Bohne, J., & Cathomen, T. (2008). Genotoxicity in gene therapy: 31. Yamanaka, S., & Blau, H. M. (2010). Nuclear reprogramming to a an account of vector integration and designer nucleases. Current pluripotent state by three approaches. Nature, 465(7299), Opinion in Molecular Therapeutics, 10(3), 214–223. 704–712. 20. Vollmers, A., Wallace, L., Fullard, N., et al. (2011). Two- and 32. Sotiropoulou, P.A., Candi, A., Mascre, G., et al. (2010). Bcl-2 and three-dimensional culture of keratinocyte stem and precursor cells accelerated DNA repair mediates resistance of hair follicle bulge derived from primary murine epidermal cultures. Stem Cell stem cells to DNA-damage-induced cell death. Nature Cell Reviews and Reports. doi:10.1007/s12015-011-9314-y Biology, 12(6), 572–582. 21. Tumbar, T., Guasch, G., Greco, V., et al. (2004). Defining the 33. Mohrin, M., Bourke, E., Alexander, D., et al. (2010). Hematopoi- epithelial stem cell niche in skin. Science, 303(5656), 359–363. etic stem cell quiescence promotes error-prone DNA repair and 22. Lombardo, A., Genovese, P., Beausejour, C. M., et al. (2007). mutagenesis. Cell Stem Cell, 7(2), 174–185. Gene editing in human stem cells using zinc finger nucleases and 34. Milyavsky, M., Gan, O. I., Trottier, M., et al. (2010). A distinctive integrase-defective lentiviral vector delivery. Nature Biotechnology, DNA damage response in human hematopoietic stem cells reveals 25(11), 1298–1306. an apoptosis-independent role for p53 in self-renewal. Cell Stem 23. Benabdallah, B. F., Allard, E., Yao, S., et al. (2010). Targeted gene Cell, 7(2), 186–197. addition to human mesenchymal stromal cells as a cell-based 35. Mandal, P. K., Blanpain, C., & Rossi, D. J. (2011). DNA damage plasma-soluble protein delivery platform. Cytotherapy, 12, response in adult stem cells: pathways and consequences. Nature 394–399. Reviews. Molecular Cell Biology, 12(3), 198–202. Publication 2 33

4. Publication 2

Versatile and Efficient Genome Editing in Human Cells by Combining Zinc- Finger Nucleases With Adeno-Associated Viral Vectors.

Human Gene Therapy, 2012

Author’s contribution statement:

This study was conducted in collaboration with Prof. Dr. Robert Kotin and Prof. Dr. Regine Heilbronn. I generated and characterized the viral plasmids containing EGFP specific zinc finger nucleases (ZFNs). In addition, I performed the EGFP disruption and deletion experiments in cells (Figure 1 A and C), determined viral vector integration in the target site (Figure 1 F), performed the genotyping (Figure 1 B), assessed the AAV expression kinetics experiments (Figure 3 A and B), performed cell cycle profiling experiments and assessed platform associated toxicity by measuring apoptosis (Figure 4 A and B). Furthermore, I contributed to the writing of the manuscript.

HUMAN GENE THERAPY 23:321–329 (March 2012) ª Mary Ann Liebert, Inc. DOI: 10.1089/hum.2011.140

Versatile and Efficient Genome Editing in Human Cells by Combining Zinc-Finger Nucleases With Adeno-Associated Viral Vectors

Eva-Maria Ha¨ndel,1,2,* Katharina Gellhaus,2,*,{ Kafaitullah Khan,1,2,* Christien Bednarski,1,2 Tatjana I. Cornu,2,{ Felix Mu¨ller-Lerch,2 Robert M. Kotin,3 Regine Heilbronn,2 and Toni Cathomen1,2

Abstract Zinc-finger nucleases (ZFNs) have become a valuable tool for targeted genome engineering. Based on the enzyme’s ability to create a site-specific DNA double-strand break, ZFNs promote genome editing by activat- ing the cellular DNA damage response, including homology-directed repair (HDR) and nonhomologous end- joining. The goal of this study was (i) to demonstrate the versatility of combining the ZFN technology with a vector platform based on adeno-associated virus (AAV), and (ii) to assess the toxicity evoked by this platform. To this end, human cell lines that harbor enhanced green fluorescence protein (EGFP) reporters were generated to easily quantify the frequencies of gene deletion, gene disruption, and gene correction. We demonstrated that ZFN-encoding AAV expression vectors can be employed to induce large chromosomal deletions or to disrupt genes in up to 32% of transduced cells. In combination with AAV vectors that served as HDR donors, the AAV- ZFN platform was utilized to correct a mutation in EGFP in up to 6% of cells. Genome editing on the DNA level was confirmed by genotyping. Although cell cycle profiling revealed a modest G2/M arrest at high AAV-ZFN vector doses, platform-induced apoptosis could not be detected. In conclusion, the combined AAV-ZFN vector technology is a useful tool to edit the human genome with high efficiency. Because AAV vectors can transduce many cell types relevant for gene therapy, the ex vivo and in vivo delivery of ZFNs via AAV vectors will be of great interest for the treatment of inherited disorders.

Introduction the two target half-sites (Smith et al., 2000). ZFNs have been shown to cleave target DNA with high specificity and effi- n the last two decades various platforms have been ciency in a variety of human cell lines and primary cells, in- Ideveloped that allow for efficient and precise modification cluding T cells (Perez et al., 2008), hematopoietic stem cells of the human genome. A very promising approach is based on (Lombardo et al., 2007; Holt et al., 2010), mesenchymal stromal zinc-finger nuclease (ZFN) technology (Urnov et al., 2010; cells (Benabdallah et al., 2010), embryonic stem cells (Lom- Rahman et al., 2011). ZFNs are designer nucleases that consist bardo et al., 2007; Hockemeyer et al., 2009; Zou et al., 2009), and of two functional domains: a customized zinc-finger array induced pluripotent stem cells (Hockemeyer et al., 2009; Zou that specifies DNA binding and the endonuclease domain of et al., 2009). the restriction enzyme FokI that contains the catalytic activity A major issue for therapeutic applications of designer (Smith et al., 2000). Each zinc-finger array in a ZFN subunit nucleases is the balance between ZFN activity and nuclease- confers binding to the respective target half-site, and upon associated toxicity (Ha¨ndel and Cathomen, 2011). ZFNs have dimerization of the two subunits in correct spacing and ori- undergone several critical improvements in design in the last entation, the nuclease domain introduces a DNA double- decade, including the development of novel platforms to strand break (DSB) within the spacer sequence that separates generate specific zinc-finger arrays (Maeder et al., 2008;

1Institute of Experimental Hematology, Hannover Medical School, 30625 Hannover, Germany. 2Institute of Virology, Campus Benjamin Franklin, Charite´ Medical School, 12203 Berlin, Germany. 3Molecular Virology and Gene Delivery Section, Laboratory of Biochemical Genetics, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland 20892. *These authors contributed equally. {Present address: Epiontis GmbH, 12489 Berlin, Germany.

321 322 HA¨ NDEL ET AL.

Urnov et al., 2010; Kim et al., 2011; Sander et al., 2011), im- vectors are shown in Figs. 1 and 2. ZFN pairs EB1/BA1 (here proved architecture of the dimer interface to prevent termed EB/BA) (Cornu et al., 2008), E292 and E502 (Maeder homodimerization of ZFN subunits (Miller et al., 2007; et al., 2008), and their respective recognition sites have been Szczepek et al., 2007; So¨llu¨ et al., 2010; Doyon et al., 2011; described before. For this study, the corresponding zinc- Ramalingam et al., 2011), and customized interdomain link- finger arrays were first transferred into a pRK5 plasmid ers that connect the DNA-binding domain with the nuclease backbone (Alwin et al., 2005) that contained either of the domain (Bibikova et al., 2001; Ha¨ndel et al., 2009). After ZFN- obligate heterodimeric FokI variants KV or EA under control mediated DNA cleavage of the target site, the resulting DSB of a cytomegalovirus (CMV) promoter (Szczepek et al., 2007; leads to activation of the cellular DNA damage response, So¨llu¨ et al., 2010), and then subcloned into the AAV backbone involving nonhomologous end-joining (NHEJ) and homology- plasmids pFB-GFPR (Urabe et al., 2002) or pTR-UF (gift of directed repair (HDR) (Shrivastav et al., 2008). NHEJ is an Nicholas Muzyczka, Gainsville, FL). Plasmid pAV.LHA- error-prone repair pathway that can lead to small insertions Sce1D encodes a CMV-driven, C-terminally truncated ver- or deletions (indels) at the junction site. If the DSB occurs sion of I-SceI that served as a negative control. The AAV within coding sequences, indels can result in frame-shift or donor vGFPiNwpre and the lentiviral vector encoding the nonsense mutations. HDR, on the other hand, is based on mGFPiNwpre target locus (Fig. 2) have been described be- homologous recombination and is the basis of ZFN-induced fore (Gellhaus et al., 2010; So¨llu¨ et al., 2010). The lentiviral gene targeting. Although NHEJ is the more prominent repair vector encoding target locus dsEGFP (Fig. 1) was generated pathway in mammalian cells, when high numbers of donor by adding two PEST domains at the C-terminus of EGFP of DNA are delivered to the nucleus, the HDR pathway be- vector pLV-CMV.EGFPiNwpre (Cornu and Cathomen, 2007) comes more prominent for DSB repair (Lombardo et al., 2007; and inserting the EB/BA target site in the DU3 region of the Gellhaus et al., 2010). 3¢ long terminal repeat (LTR). Maps and sequences of all An important aspect in ZFN-based genome engineering is plasmids are available upon request. vectorization of the nuclease and the donor. Ideally, ZFNs as well as the donor DNA are present only transiently at high AAV-mediated genome engineering concentrations. Episomal DNA-based ZFN expression sys- All experiments were performed with recombinant AAV tems, such as plasmid DNA or viral vectors, usually harbor type 2 vectors. With the exception of AAV.ZFN and strong promoters to ensure high ZFN levels. Thus, the du- EB AAV.ZFN , all AAV vectors were produced and purified as ration of ZFN expression is limited through rapid dilution BA previously described (Gellhaus et al., 2010). Vectors during cell division in mitotic cells. In addition to plasmid AAV.ZFN and AAV.ZFN were generated using the DNA, transient nuclease expression in human cells has been EB BA baculovirus system, as described in detail elsewhere (Smith reported from integrase-deficient lentiviral vectors (Cornu et al., 2009). AAV vector titers were determined by quanti- and Cathomen, 2007; Lombardo et al., 2007), adenoviral tative real-time PCR (LightCycler, Roche) using SYBR Green vectors (Perez et al., 2008), and vectors based on adeno- (DyNAmoTM Capillary SYBR Green, Finnzymes) and ap- associated virus (AAV) (Porteus et al., 2003; Gellhaus et al., propriate primers, as previously described (Gellhaus et al., 2010). The use of AAV vectors is of particular interest be- 2010). For AAV-mediated genome editing, 5 · 104 target cells cause these vectors function efficiently as a substrate for were infected with vector doses ranging from 103 to 105 ge- HDR, even if the target locus contains only a DNA single- nome copies (gc)/cell in 500–2000 lL of standard medium. strand nick (Metzger et al., 2011) or no DNA damage at all To analyze AAV-mediated EGFP expression kinetics, 105 (Hendrie and Russell, 2005). In proof-of-concept studies us- U2OS cells were transduced with either 50 or 500 gc/cell of ing AAV as a donor for HDR and an I-SceI–induced DSB in AAV2.EGFP (Schwartz et al., 2007). At the indicated time the target locus to stimulate gene targeting, we and others points, cells were analyzed by flow cytometry (FACSCalibur, showed that gene targeting at a chromosomally integrated BD Biosciences). U2OS-based cell lines were cultured in marker gene can be achieved in up to 65% of cultured human Dulbecco’s modified Eagle’s medium supplemented with cells (Miller et al., 2003; Porteus et al., 2003; Gellhaus et al., 10% fetal bovine serum and penicillin/streptomycin (In- 2010; Hirsch et al., 2010). Moreover, in a recent study ZFN- vitrogen). The polyclonal target cells were generated by stimulated in vivo gene targeting was reported after AAV- lentiviral transduction (Cornu et al., 2008), followed by se- based gene transfer in a mouse model for factor IX deficiency lection with 0.4 mg/mL geneticin (Ha¨ndel et al., 2009). (Li et al., 2011). The goals of this study were to demonstrate the versatility Genotyping of combining AAV vectors with the ZFN platform for rational editing of the human genome and to assay platform- Genomic DNA from target cells was extracted with the associated toxicity. We show that ZFN encoding AAV ex- Blood MiniKit (Qiagen). To amplify part of the proviral DNA pression vectors can be employed (i) to delete a lentiviral containing the EGFP locus, 100 ng of genomic DNA was provirus, (ii) to disrupt an expression cassette, or (iii) to used as a template, along with 10 lM of each primer (#1270 correct a mutational insertion. 5¢-tacatcaatgggcgtggata and #598 5¢-gaactccagcaggaccatgt), 10 mM dNTP, and 0.125 U of Phusion High-Fidelity DNA Materials and Methods Polymerase (Finnzymes) in 1 · reaction buffer for 25 cycles. A 150-bp amplicon of the PTBP2 allele (primers #1274 5¢- Plasmids tctccattccctatgttcatgc and #1275 5¢-gttcccgcagaatggtgaggtg) Plasmids were assembled by polymerase chain reaction served as a control. T7 endonuclease I (T7E1) assays were (PCR)-directed cloning and standard molecular biology carried out as described previously (Hoher et al., 2011; procedures. Relevant parts of the plasmids and the resulting Mussolino et al., 2011). In brief, 100 ng of the purified 544-bp GENOME ENGINEERING WITH AAV AND ZFN 323

FIG. 1. Adeno-associated virus (AAV)-mediated ge- nome editing by nonhomol- ogous end-joining (NHEJ). (A) Targeted deletion of len- tiviral provirus. U2OS.693 cells carry an integrated copy of a destabilized enhanced green fluorescence protein (dsEGFP) expression cassette that was introduced by len- tiviral transduction (target). The lentiviral long terminal repeats harbor a target site for zinc-finger nuclease (ZFN) pair EB/BA (fat ar- rows). Expression of both ZFN subunits (ZFNEB and ZFNBA) upon infection of cells with 3 · 103 (gray) or 105 genome copies (gc)/cell (black) of the respective AAV vectors leads to excision of the expression cassette (product). The graph shows the percentage of EGFP- negative cells, as assessed by flow cytometry; * indicates statistically significant in- crease in EGFP-negative cells as compared to cells infected with AAV-ZFNEB alone ( p < 0.05). (B) Polymerase chain reaction (PCR)-based molecular characterization. Genomic DNA of single EGFP-negative clones (lanes 1–8) was extracted and used as a template to amplify a 1-kb fragment contained in the

provirus. The positions of the primers are indicated as arrows in (A). Amplification of a 150-bp fragment in the PTBP2 gene served as a control. NT, nontransduced U2OS.693 cells; U, U2OS parental cell line; H2O, PCR control. (C) Targeted gene disruption. U2OS.693 were infected with 103,104,or105 gc/cell of AAV vectors that express ZFNs targeting either position 292 or 502 (fat arrows) of the EGFP reading frame, respectively. Error-prone repair of the DNA double-strand break leads to disruption of the coding sequence (product). The graph shows the fraction of EGFP-negative cells, as assayed 5 days post- infection by flow cytometry; * indicates statistically significant increase in EGFP-negative cells, as compared to cells infected with a control vector expressing a nonfunctional nuclease ( p < 0.02). (D) T7 endonuclease I (T7E1)-based genotyping. Genomic DNA of cells transduced with E502-specific ZFN expression vectors was used as a template to amplify a 544-bp fragment containing the target site. The amplicon was subjected to digestion with mismatch-sensitive T7E1 to verify the presence of insertions/deletions at E502. The position of the expected 446-bp band is indicated. (E) PCR strategy to detect AAV-ZFN integration into E502 site in antisense orientation. The positions of the PCR primers to detect the 5¢- and 3¢- junctions, respectively, are shown. (F) Qualitative assessment. Genomic DNA of U2OS.693 cells transduced with E502- specific ZFN expression vectors was subjected to PCR analysis. DNA from nontransduced (NT) cells served as a control. The position of the expected *800-bp band is indicated. H2O, PCR control.

PCR amplicon containing part of the EGFP gene (primers #76 of 20 lL) was used as a template for a second amplification 5¢-taaacggccacaagttcagcgt and #185 5¢-gtgctcaggtagtggttgtcg) round with primers #361 5¢-gaggagctgttcaccggg and #559 for was melted and re-annealed to allow the formation of het- 25 cycles. eroduplex DNA, treated with 5 U of T7E1 (New England BioLabs) for 20 min at 37C, and separated on a 2% agarose Analysis of ZFN expression gel. To detect AAV vector integration into the E502 site, primers recognizing the EGFP gene and the AAV-ZFN ex- Western blot analysis to detect ZFN expression was per- pression vector were used for the 5¢-junction primers #76 formed as previously (Shimizu et al., 2011). To assess ZFN and #1401 (5¢-accatggcccaacttgttta; SV40 pA) and for the 3¢- expression at the RNA level, 105 U2OS cells were transduced 5 junction primers #185 and #14 (5¢-aatggggcggagttgttacgac; with 10 gc/cell of AAV.ZFNEB and AAV.ZFNBA each, and CMV). Correction of the EGFP target locus was verified us- harvested in TRIzol reagent (Invitrogen) at indicated time ing a nested PCR. For the first PCR, 100 ng of genomic DNA points. The cDNA was generated using QuantiTect reverse was used as a template, along with 300 lM of each primer transcription kit (Qiagen) and used as a template to amplify (#136 5¢-caagggcgaggagctggt and #559 5¢-ctcggcgcgggtcttg a 162-bp fragment of the FokI nuclease using primers #1436 tag), 200 mM dNTP, 0.125 U of Taq polymerase (PEQLAB) in 5¢-agtcaagagcgagctggaag and #1437 5¢-cggtagccgtacaccttcat. 1 · reaction buffer for 13 cycles. PCR products were purified The PCR products were separated on a 2% agarose gel and using QIAquick PCR Purification Kit (Qiagen) and 1 lL (out quantified from nonsaturated gel images with Quantity One 324 HA¨ NDEL ET AL.

V.4.6.9 (BioRad) software (Hoher et al., 2011; Mussolino et al., apoptosis was quantified by determining the percentage of 2011). subG1 events by plotting the whole cell population as FL2-H versus cell counts. Cell cycle profiles and apoptosis Statistical analysis U2OS target cells were synchronized by serum starvation for 1 day. Then, 105 cells were transduced with 105 gc/cell All experiments were performed at least three times, with each of both AAV-ZFNsubunit expression vectors or treated the exception of data point E502 in Fig. 4. Error bars repre- with 100 nM vinblastine for 4 hr. After 4 days, cells were sent standard deviation. Statistical significance was deter- fixed overnight in ethanol and stained with propidium io- mined using a one-sided Student’s t-test with unequal dide (PI) staining solution (3.8 mM sodium citrate, 25 lg/mL variance. of PI, 0.2 mg/mL of RNase A in phosphate-buffered saline). Before analysis by flow cytometry (FACSCalibur), low Results molecular weight DNA was released by incubating cells in AAV-ZFN–mediated gene disruption DNA extraction buffer (200 mM Na HPO ,100mMcitric 2 4 and provirus excision acid, pH 7.8). The fractions of cells in G1, S, and G2/M phases were determined by plotting FL3-W versus FL3-A. As a paradigm for inactivating and/or deleting a retro- The number of cells containing a fragmented genome due to viral sequence in the human genome, we generated a re- porter line based on human osteosarcoma U2OS cells that expresses a destabilized EGFP (U2OS.693) by lentiviral transduction. Quantitative PCR established that U2OS.693 cells contain between three and seven copies of the provirus (data not shown). The lentiviral LTRs of the lentivector used contain a target site for the previously characterized ZFN pair EB/BA (Cornu et al., 2008). Upon expression of both ZFN subunits the entire provirus should be excised, leaving just a single LTR signature in the genome (Fig. 1A). To this end, U2OS.693 cells were infected with 3 · 103 or 1 · 105 gc of AAV particles per cell, either individually with a single ZFN subunit or in combination, and the percentage of EGFP- expressing cells was determined 5 days later. As expected, successful excision of the provirus was dependent on the expression of both ZFN subunits and on the vector dose.

After co-transduction with 105 gc/cell of AAV vectors ‰ FIG. 2. AAV-mediated genome editing by homology- directed repair (HDR). (A) Schematic of gene correction in U2OS.893 cells. The target locus consists of a mutated EGFP (mGFP) gene under control of a cytomegalovirus (CMV) promoter, followed by an ires-NeoR-wpre cassette. The mu- tation in mGFP is based on a 43-bp insertion that includes three in-frame stop codons and a recognition site for ZFN pair EB/BA (fat arrow). The AAV donor vector contains a 5¢- truncated EGFP gene (vGFP) followed by the ires-NeoR-wpre cassette. HDR (indicated by two crosses) between target lo- cus and AAV donor leads to expression of a functional EGFP gene (product). (B) AAV-ZFN mediated gene targeting. U2OS.893 cells were transduced with the indicated vector dose of AAV donor and ZFN expression vectors, and as- sessed by flow cytometry 6 days post-transduction. The graph displays the average percentage of EGFP-positive cells; * and ** indicate statistically significant increase in EGFP-positive cells as compared to mock infected cells ( p < 0.01 and p < 0.002, respectively). (C) AAV-mediated ZFN expression. U2OS cells were transduced separately with 4 5 AAV-ZFNEB, AAV-ZFNBA, or a control vector at 10 or 10 gc/cell, harvested after 72 hr, and ZFN expression levels detected using an HA tag specific antibody. (D) Genotyping. Genomic DNA of U2OS.696 cells infected with 104 gc/cell of AAV donor or AAV-ZFN, as indicated, was isolated 30 days post-transduction. The DNA was subjected to nested PCR analysis (primer positions shown in panel A) to detect the corrected target locus; 104 and 105 copies of an EGFP plasmid served as positive controls (cto1, cto2). GENOME ENGINEERING WITH AAV AND ZFN 325 encoding the EB and BA subunits, 11% of U2OS.693 cells creased to 105 gc/cell, suggesting that the ratio between turned EGFP-negative. To verify ZFN-mediated excision of donor and ZFN expression vector is crucial. Immunoblot the provirus on the genome level, single clones were ex- analysis confirmed equal expression of both ZFN subunits panded and the genomic DNA of EGFP-negative cells iso- upon transduction (Fig. 2C). lated. PCR-based amplification of a region spanning most of To verify that EGFP-positive cells were the result of a the EGFP gene and part of the CMV promoter revealed that genuine HR event between the target locus and the AAV in the majority of clones all proviral genomes were excised donor vector, genomic DNA from U2OS.893 cells was ex- (Fig. 1B). In three out of the eight clones at least one provirus tracted 30 days post-transduction. The amplification products copy remained untouched. Since all clones were EGFP neg- of an allele-specific PCR, in which primers that specifically ative, the respective expression cassettes must have been recognize the corrected target locus but neither the AAV silenced before or after AAV transduction. donor DNA nor the mutated target locus were employed As a proof of principle that demonstrates AAV-ZFN me- (Gellhaus et al., 2010), confirmed that efficient gene targeting diated inactivation of functional gene expression, ZFN pairs was dependent on the presence of all three vectors (Fig. 2D). targeting the EGFP marker gene at positions 292 (E292) and 502 (E502), respectively (Maeder et al., 2008), were employed. Expression kinetics U2OS.693 cells were infected with 103,104,or105 gc/cell of Transient expression of ZFNs is crucial for many appli- each ZFN expression vector and evaluated after 5 days. The cations because prolonged expression of the designer nu- percentage of EGFP-negative cells was dependent on both the cleases may be toxic to the cell. To characterize expression vector dose and the ZFN pair used (Fig. 1C). Whereas 9% of kinetics of AAV-based transgene expression both at the cells lost functional EGFP expression upon infection with 105 protein and RNA levels, we transduced U2OS cells either gc/cell of the AAV vectors encoding the left and right subunits with an AAV2-based EGFP expression vector or the AAV- of the E292 ZFN pair, 32% of cells transduced with 105 gc/cell ZFN vectors. U2OS cells transduced with 50 or 500 gc/cell of of the E502 encoding AAV vectors turned EGFP-negative. AAV.EGFP, respectively, were assessed by flow cytometry at Molecular characterization of genomic DNA from pooled cells 2, 5, and 8 days post-transduction (Fig. 3A). EGFP fluores- using the mismatch-sensitive T7E1 confirmed the presence of cence dropped more than threefold between days 5 and 8, indels at the E502 site after expression of the respective ZFNs suggesting that the vector is rapidly diluted out in these (Fig. 1D). To assess whether the AAV-ZFN expression vectors can integrate into a ZFN-induced DSB, a PCR strategy to de- tect vector–host genome junctions at the E502 site was applied (Fig. 1E). Both PCR amplifications produced the expected

*800-bp fragments (Fig. 1F), confirming AAV integration into nuclease-induced DSBs at low frequencies as previously shown (Gellhaus et al., 2010). Together these experiments demonstrated that AAV vec- tors encoding ZFNs can be efficiently employed to modify the human genome by harnessing the cellular NHEJ path- way for DNA repair. The AAV-ZFN platform was success- fully used to either disrupt a gene or to delete an entire expression cassette, as illustrated by the excision of a lenti- virus vector genome.

AAV-ZFN–mediated gene correction To demonstrate AAV-mediated gene correction by HR, we established an U2OS-based cell line harboring a promoter- proximal 43-bp insertion that contains the recognition site for the ZFN pair EB/BA at position 327 of the EGFP marker gene (U2OS.893). Because promoter-free donors are advan- tageous, such a target configuration requires the generation of AAV donors with asymmetric homology arms. The AAV donor contained a 5¢-truncated EGFP (vGFP) and a homol- FIG. 3. AAV expression kinetics. (A) Protein expression. ogous 3¢-sequence stretch (ires-NeoR-wpre). Of note, the 5¢- U2OS cells were transduced with an AAV-EGFP vector at 50 homology arm between the donor and target was only or 500 gc/cell, respectively, and subjected to flow cytometric 268 bp long (Fig. 2A). analysis at the indicated time points. The graph displays U2OS.893 cells were co-infected with different amounts of mean fluorescence intensity measured in channel FL-1. (B) AAV donor vector, AAV-ZFN , and AAV-ZFN , and the RNA expression. U2OS cells were transduced with AAV- EB BA ZFN and AAV-ZFN at 105 gc/cell. RT-PCR amplifying a percentage of EGFP-positive cells was determined 10 days EB BA fragment contained in the ZFN FokI domain was performed post-transduction (Fig. 2B). Co-infection of target cells with 4 5 on RNA extracted from transduced cells at indicated time 10 gc/cell of donor vector and 10 gc/cell of both AAV- points. The position of the expected 162-bp band is indicated. ZFNEB and AAV-ZFNBA led to correction of 6% of trans- Numbers below the gel indicate signal intensities relative to duced cells. The percentage of EGFP-positive cells dropped day 2. NT, nontransduced cells; AAV, cells transduced with significantly to 0.9% when the amount of donor was in- an AAV control vector. 326 HA¨ NDEL ET AL.

cycle, the percentage of the subG1 fraction, representing apoptotic cells, did not increase above background for all samples analyzed (Fig. 4B). Also Annexin V staining of these cells did not reveal any significant signs of apoptosis (data not shown). Together, these results suggest that AAV- mediated expression of ZFNs can induce a mild G2/M arrest but does not induce measurable levels of apoptosis.

Discussion An important safety measurement when developing platforms for therapeutic ZFN expression in human cells is to make sure that nuclease expression quickly reaches a high peak value after transfection/transduction but that expres- sion lasts only transiently. The latter point is of utmost im- portance because long-lasting high ZFN expression may increase ZFN-associated off-target activity and hence cyto- toxicity (Alwin et al., 2005; Doyon et al., 2008; Meng et al., 2008; Foley et al., 2009; Pruett-Miller et al., 2009). Nucleofec- tion of plasmid DNA and transduction with adenoviral or integrase-deficient lentiviral vectors (IDLVs), respectively, have been successfully used to achieve ZFN-mediated gene editing in primary human cells (Lombardo et al., 2007; Perez et al., 2008; Holt et al., 2010). Moreover, microinjection of ZFN-encoding mRNA into one-cell zebrafish and rat em- bryos was shown to allow for efficient gene disruption (Doyon et al., 2008; Meng et al., 2008; Geurts et al., 2009). In this study we present AAV vectors as a potentially valuable vector platform for ZFN-mediated genome engineering FIG. 4. Platform-associated toxicity. U2OS.693 cells were based on NHEJ and HDR. We demonstrated that combining 5 infected with 10 gc/cell of AAV-ZFN vectors, as indicated, the two platforms allows for efficient genome editing in fixed in ethanol 4 days after infection, stained with propi- human cells, such as the targeted deletion of an entire ex- dium iodide (PI) and analyzed by flow cytometry. Cells pression cassette, the targeted disruption of an expressed treated with 100 nM vinblastine (Vin) were assayed after 1 gene, and the correction of mutations in an open reading day and served as a control. (A) Cell cycle profile. The plots frame. show the respective cell populations in G1 (gray), S (white), The concentration and configuration of the donor DNA is and G2/M phase (black). The average percentage of cells in a crucial parameter in HDR-based genome editing (Hirata G2/M phase is indicated. (B) Apoptosis. The graph displays and Russell, 2000; Moehle et al., 2007; Gellhaus et al., 2010; the average fraction of cells in the subG1 population at day 4 post-transduction. Orlando et al., 2010). It has been reported previously that AAV vectors serve as efficient templates for HDR in human cells, both in the absence (Chamberlain et al., 2004; Khan et al., 2010) and the presence (Miller et al., 2003; Porteus et al., cultured cells. Total RNA was isolated from cells transduced 2003; Gellhaus et al., 2010; Hirsch et al., 2010) of a DSB. For with 105 of AAV-ZFN and AAV-ZFN each at days 2, 5, EB BA the experimental systems used in this and in our previous and 8 post-transduction. Reverse-transcription PCR con- studies, however, we were not able to detect AAV-based firmed rapid loss of ZFN expression over the timeframe of 1 gene targeting in the absence of a DSB-inducing enzyme. The week (Fig. 3B). Together these results confirmed the tran- reason for this discrepancy likely resides in the donor ar- sient, short-lived nature of ZFN expression in mitotic cells. chitecture, which in our case was optimized to work effi- ciently for DSB-stimulated gene targeting. AAV-ZFN–associated toxicity Co-applying the AAV donors with AAV-ZFN expression In the present study we applied three ZFN pairs that were vectors resulted in functional gene correction in up to 6% of generated by selection-based approaches (Hurt et al., 2003; cells. Although this number is therapeutically relevant, it is Maeder et al., 2008). To assess cytotoxicity associated with about four times smaller as compared with a similar setting AAV-based expression of these ZFNs, U2OS cells were co- in which U2OS target cells were co-infected with the same infected with 105 gc/cell of each vector encoding either ZFN donor and an AAV I-SceI expression vector (25% EGFP- subunit. After 2 and 4 days cells were fixed and stained with positive cells) (Gellhaus et al., 2010). Moreover, when the PI to perform a cell cycle analysis and determine the per- same cells were infected with a single AAV vector that centage of apoptotic cells by quantification of the subG1 combined donor function and I-SceI expression, 65% of cells fraction. The cell cycle profiles at day 2 (not shown) and day revealed a corrected target locus (Gellhaus et al., 2010). 4 post-transduction (Fig. 4A) revealed a modest increase in Hence, the lower gene targeting activity cannot be solely the G2/M populations for cells expressing the E292 and E502 reduced to differences in nuclease activities. This is in ZFN pairs. In agreement with this modest effect on the cell agreement with previous comparisons between I-SceI and GENOME ENGINEERING WITH AAV AND ZFN 327

ZFNs pairs, in which we did not detect more than twofold reported that a targeted DSB shifts the balance from illegit- differences in the ability of these nucleases to stimulate gene imate recombination to gene targeting (Gellhaus et al., 2010), targeting when plasmid transfection was used (Cornu et al., which strongly supports the use of designer nucleases to 2008; Szczepek et al., 2007). We therefore speculate that the promote HDR-based genome engineering. drop in promoting gene targeting is a vector-associated Quantitative assessment of ZFN-associated toxicity has been phenomenon, and reducing the number of different AAV challenging, especially since the platforms to generate ZFNs vector constructs from three to two may be crucial to further improved in terms of producing designer nucleases with increase the gene targeting frequency. Future experiments higher activity and lower toxicity (Ha¨ndel and Cathomen, will consequently focus on generating AAV vectors that 2011). In this study we applied two assays to quantify AAV- harbor a so-called two-in-one ZFN construct; i.e., a combined ZFN–induced apoptosis but did not detect signs of toxicity. expression cassette in which the two ZFN subunits are sep- Cell cycle analysis, on the other hand, demonstrated a mild arated by a 2A autoproteinase (So¨llu¨ et al., 2010). arrest of cells expressing ZFN pairs E292 and E502 in the G2/ High peak expression of ZFN is a prerequisite to reach the M phase. A G2/M phase arrest is indicative of DNA damage, critical nuclear concentration of the enzyme necessary to induce likely caused by the combined on- and off-target activities of a targeted DSB. As shown previously for IDLVs and adenoviral the respective ZFNs. Previous studies have further shown that vectors, sufficient ZFN expression can only be achieved with transduction of cells with wild-type AAV2 or UV-inactivated relatively high vector doses (Lombardo et al., 2007; Perez et al., AAV2 led to cell cycle arrest in the G2/M phase (Saudan et al., 2008). In many experiments we observed efficient NHEJ and 2000; Raj et al., 2001). Under the experimental conditions used HDR-based genome editing only when applying 105 AAV-ZFN in our study, we only observed minimal effects of AAV gc/cell. Although this is a large number, 105 copies of an ex- transduction on the cell cycle. The absence of both Rep ex- pression cassette per cell correspond to about 0.5 lgofa4-kb pression and UV irradiation of our AAV vectors could explain plasmid vector per 106 cells, which is typically used for ZFN the different experimental outcomes. In summary, this study expression vectors in nucleofection protocols. On the other emphasizes the importance of a comprehensive and thorough hand, we sometimes noticed inhibitory effects on gene targeting analysis of the activities and toxicities triggered by the designer when using very high AAV doses. Whether this is related to nucleases and/or the vector system. saturation of the cellular transport machinery that shuttles the In conclusion, this study validates AAV vectors as a capsids from the cell surface to the nucleus or saturation of the valuable platform to deliver designer nucleases to target DNA repair pathways that have to deal with high numbers of cells. Given the safety record of AAV vectors in clinical and incoming single-stranded DNA vectors, needs to be addressed preclinical applications, one can even envisage in vivo gene in future experiments. editing therapies, as recently demonstrated in the liver of

The use of self-complementary AAV (scAAV) vectors neonatal mice (Li et al., 2011). AAV vectors have also been (McCarty et al., 2003; Wang et al., 2003) may effectively re- used to efficiently transduce human iPS cells or mesenchy- duce the required ZFN vector dose. scAAVs harbor a double- mal stem cells ex vivo, opening the door for a wide range of stranded DNA genome and are therefore not dependent applications in regenerative medicine by combining cell on conversion into duplex forms, either via second strand therapy with targeted genome engineering. synthesis or re-association kinetics, leading to faster and usually higher transgene expression. In a proof-of-concept Acknowledgments approach Hirsch et al. (2010) showed that scAAV-based I- SceI expression vectors performed better in a gene targeting We thank Jessica Wenzl and Eva Guhl for technical as- assay than single-stranded AAV counterparts. On the sistance, Shamim H. Rahman and Sylwia Bobis-Wozowicz downside, the limited vector genome size of scAAV will not for help with experiments and critical discussions, and Ni- permit the packaging of a two-in-one ZFN expression cas- cholas Muzyczka for plasmids. This work was supported by sette or a combination of donor and ZFN. The high AAV grants ITCF–01GU0618 of the German Ministry of Education doses required to see biological effects in our studies suggest and Research, PERSIST–222878 of the European Commis- that high vector genome concentrations in the nucleus may sion’s 7th Framework Programme, and SFB/TR19–TPC5 of allow for pairing of complementary genomes of ( + ) and ( - ) the German Research Foundation to T.C., and a fellowship of polarity and as a result bypass the requirement of DNA the German Academic Exchange Service (DAAD) to K.K. second strand synthesis. ZFN and vector-associated toxicity is of utmost impor- Author Disclosure Statement tance when transferring the AAV-ZFN platform to the clinic. No competing financial interests exist. It is known that episomal DNA vectors can integrate into naturally occurring DSBs (Miller et al., 2004), and we and others have also shown that nuclease-induced DSB serve as a References prominent integration site for viral vectors, including IDLVs Alwin, S., Gere, M.B., Guhl, E., et al. (2005). Custom zinc-finger and AAV (Cornu and Cathomen, 2007; Gellhaus et al., 2010; nucleases for use in human cells. Mol. Ther. 12, 610–617. Petek et al., 2010). Also in this study, we detected integration Benabdallah, B.F., Allard, E., Yao, S., et al. (2010). Targeted gene of AAV vectors into the ZFN-induced DSB. However, al- addition to human mesenchymal stromal cells as a cell-based though AAV vectors can integrate into nuclease-induced plasma-soluble protein delivery platform. Cytotherapy 12, DSBs at low frequency, the majority of integrations seem to 394–399. occur outside the target locus (Gellhaus et al., 2010), most Bibikova, M., Carroll, D., Segal, D.J., et al. (2001). Stimulation of likely in naturally occurring DSBs or nuclease off-target sites homologous recombination through targeted cleavage by (Petek et al., 2010; Gabriel et al., 2011). Moreover, we also chimeric nucleases. Mol. Cell Biol. 21, 289–297. 328 HA¨ NDEL ET AL.

Chamberlain, J.R., Schwarze, U., Wang, P.R., et al. (2004). Gene Li, H., Haurigot, V., Doyon, Y., et al. (2011). In vivo genome targeting in stem cells from individuals with osteogenesis editing restores haemostasis in a mouse model of haemophi- imperfecta. Science 303, 1198–1201. lia. Nature 475, 217–221. Cornu, T.I., and Cathomen, T. (2007). Targeted genome modi- Lombardo, A., Genovese, P., Beausejour, C.M., et al. (2007). Gene fications using integrase-deficient lentiviral vectors. Mol. Ther. editing in human stem cells using zinc finger nucleases and 15, 2107–2113. integrase-defective lentiviral vector delivery. Nat. Biotechnol. Cornu, T.I., Thibodeau-Beganny, S., Guhl, E., et al. (2008). DNA- 25, 1298–1306. binding specificity is a major determinant of the activity and Maeder, M.L., Thibodeau-Beganny, S., Osiak, A., et al. (2008). toxicity of zinc-finger nucleases. Mol. Ther. 16, 352–358. Rapid ‘‘open-source’’ engineering of customized zinc-finger Doyon, Y., Mccammon, J.M., Miller, J.C., et al. (2008). Heritable nucleases for highly efficient gene modification. Mol. Cell 31, targeted gene disruption in zebrafish using designed zinc- 294–301. finger nucleases. Nat. Biotechnol. 26, 702–708. McCarty, D.M., Fu, H., Monahan, P.E., et al. (2003). Adeno- Doyon, Y., Vo, T.D., Mendel, M.C., et al. (2011). Enhancing zinc- associated virus terminal repeat (TR) mutant generates self- finger-nuclease activity with improved obligate heterodimeric complementary vectors to overcome the rate-limiting step to architectures. Nat. Methods 8, 74–79. transduction in vivo. Gene Ther. 10, 2112–2118. Foley, J.E., Yeh, J.R., Maeder, M.L., et al. (2009). Rapid mutation Meng, X., Noyes, M.B., Zhu, L.J., et al. (2008). Targeted gene of endogenous zebrafish genes using zinc finger nucleases inactivation in zebrafish using engineered zinc-finger nucle- made by Oligomerized Pool ENgineering (OPEN). PLoS One ases. Nat. Biotechnol. 26, 695–701. 4, e4348. Metzger, M.J., McConnell-Smith, A., Stoddard, B.L., and Miller, Gabriel, R., Lombardo, A., Arens, A., et al. (2011). An unbiased A.D. (2011). Single-strand nicks induce homologous recom- genome-wide analysis of zinc-finger nuclease specificity. Nat. bination with less toxicity than double-strand breaks using an Biotechnol. 29, 816–823. Aav vector template. Nucleic Acids Res. 39, 926–935. Gellhaus, K., Cornu, T.I., Heilbronn, R., and Cathomen, T. Miller, D.G., Petek, L.M., and Russell, D.W. (2003). Human gene (2010). Fate of recombinant adeno-associated viral vector ge- targeting by adeno-associated virus vectors is enhanced by nomes during DNA double-strand break-induced gene tar- DNA double-strand breaks. Mol. Cell Biol. 23, 3550–3557. geting in human cells. Hum. Gene Ther. 21, 543–553. Miller, D.G., Petek, L.M., and Russell, D.W. (2004). Adeno- Geurts, A.M., Cost, G.J., Freyvert, Y., et al. (2009). Knockout rats associated virus vectors integrate at chromosome breakage via embryo microinjection of zinc-finger nucleases. Science sites. Nat. Genet. 36, 767–773. 325, 433. Miller, J.C., Holmes, M.C., Wang, J., et al. (2007). An improved Ha¨ndel, E.M., and Cathomen, T. (2011). Zinc-finger nuclease zinc-finger nuclease architecture for highly specific genome based genome surgery: it’s all about specificity. Curr. Gene editing. Nat. Biotechnol. 25, 778–785. Ther. 11, 28–37. Moehle, E.A., Rock, J.M., Lee, Y.L., et al. (2007). Targeted gene Ha¨ndel, E.M., Alwin, S., and Cathomen, T. (2009). Expanding or addition into a specified location in the human genome using restricting the target site repertoire of zinc-finger nucleases: designed zinc finger nucleases. Proc. Natl. Acad. Sci. U. S. A. the inter-domain linker as a major determinant of target site 104, 3055–3060. selectivity. Mol. Ther. 17, 104–111. Mussolino, C., Morbitzer, R., Lu¨ tge, F., et al. (2011). A novel Hendrie, P.C., and Russell, D.W. (2005). Gene targeting with TALE nuclease scaffold enables high genome editing activity viral vectors. Mol. Ther. 12, 9–17. in combination with low toxicity. Nucleic Acids Res. [Epub Hirata, R.K., and Russell, D.W. (2000). Design and packaging of ahead of print]. adeno-associated virus gene targeting vectors. J. Virol. 74, Orlando, S.J., Santiago, Y., Dekelver, R.C., et al. (2010). Zinc- 4612–4620. finger nuclease-driven targeted integration into mammalian Hirsch, M.L., Green, L., Porteus, M.H., and Samulski RJ. (2010). genomes using donors with limited chromosomal homology. Self-complementary Aav mediates gene targeting and en- Nucleic Acids Res. 38, e152. hances endonuclease delivery for double-strand break repair. Perez, E.E., Wang, J., Miller, J.C., et al. (2008). Establishment of Gene Ther 17, 1175–1180. HIV-1 resistance in CD4 + T cells by genome editing using Hockemeyer, D., Soldner, F., Beard, C., et al. (2009). Efficient zinc-finger nucleases. Nat. Biotechnol. 26, 808–816. targeting of expressed and silent genes in human ESCs and Petek, L.M., Russell, D.W., Miller, D.G. (2010). Frequent endo- iPSCs using zinc-finger nucleases. Nat Biotechnol 27, 851–857. nuclease cleavage at off-target locations in vivo. Mol. Ther. 18, Hoher, T., Wallace, L., Khan, K., et al. (2011). Highly efficient 983–986. zinc-finger nuclease-mediated disruption of an eGFP trans- Porteus, M.H., Cathomen, T., Weitzman, M.D., and Baltimore, gene in keratinocyte stem cells without impairment of stem D. (2003). Efficient gene targeting mediated by adeno-associated cell properties. Stem Cell Rev. [Epub ahead of print]. virus and DNA double-strand breaks. Mol. Cell Biol. 23, 3558– Holt, N., Wang, J., Kim, K., et al. (2010). Human hematopoietic 3565. stem/progenitor cells modified by zinc-finger nucleases targeted Pruett-Miller, S.M., Reading, D.W., Porter, S.N., and Porteus, M.H. to CCR5 control HIV-1 in vivo. Nat. Biotechnol. 28, 839–847. (2009). Attenuation of zinc finger nuclease toxicity by small- Hurt, J.A., Thibodeau, S.A., Hirsh, A.S., et al. (2003). Highly molecule regulation of protein levels. PLoS Genet 5, e1000376. specific zinc finger proteins obtained by directed domain Rahman, S.H., Maeder, M.L., Joung, J.K., and Cathomen, T. shuffling and cell-based selection. Proc. Natl. Acad. Sci. U. S. (2011). Zinc-finger nucleases for somatic gene therapy: the A. 100, 12271–12276. next frontier. Hum Gene Ther. [Epub ahead of print]. Khan, I.F., Hirata, R.K., Wang, P.R., et al. (2010). Engineering of Raj, K., Ogston, P., and Beard P. (2001). Virus-mediated killing of human pluripotent stem cells by AAV-mediated gene target- cells that lack p53 activity. Nature 412, 914–917. ing. Mol. Ther. 18, 1192–1199. Ramalingam, S., Kandavelou, K., Rajenderan, R., and Chan- Kim, S., Lee, M.J., Kim, H., et al. (2011). Preassembled zinc-finger drasegaran, S. (2011). Creating designed zinc-finger nucleases arrays for rapid construction of ZFNs. Nat. Methods 8, 7. with minimal cytotoxicity. J. Mol. Biol. 405, 630–641. GENOME ENGINEERING WITH AAV AND ZFN 329

Sander, J.D., Dahlborg, E.J., Goodwin, M.J., et al. (2011). Urabe, M., Ding, C., and Kotin, R.M. (2002). Insect cells as a Selection-free zinc-finger-nuclease engineering by context- factory to produce adeno-associated virus type 2 vectors. dependent assembly (CoDA). Nat. Methods 8, 67–69. Hum. Gene Ther. 13, 1935–1943. Saudan, P., Vlach, J., and Beard, P. (2000). Inhibition of S-phase Urnov, F.D., Rebar, E.J., Holmes, M.C., et al. (2010). Genome progression by adeno-associated virus Rep78 protein is me- editing with engineered zinc finger nucleases. Nat. Rev. diated by hypophosphorylated pRb. EMBO J. 19, 4351–4361. Genet. 11, 636–646. Schwartz, R.A., Palacios, J.A., Cassell, G.D., et al. (2007). The Wang, Z., Ma, H.I., Li, J., et al. (2003). Rapid and highly efficient Mre11/Rad50/Nbs1 complex limits adeno-associated virus transduction by double-stranded adeno-associated virus vec- transduction and replication. J. Virol. 81, 12936–12945. tors in vitro and in vivo. Gene Ther. 10, 2105–2111. Shimizu, Y., Sollu, C., Meckler, J.F., et al. (2011). Adding fingers Zou, J., Maeder, M.L., Mali, P., et al. (2009). Gene targeting of a to an engineered zinc finger nuclease can reduce activity. disease-related gene in human induced pluripotent stem and Biochemistry 50, 5033–5041. embryonic stem cells. Cell Stem Cell 5, 97–110. Shrivastav, M., De Haro, L.P., and Nickoloff, J.A. (2008). Reg- ulation of DNA double-strand break repair pathway choice. Cell Res. 18, 134–147. Address correspondence to: et al. Smith, J., Bibikova, M., Whitby, F.G., (2000). Requirements Dr. Toni Cathomen for double-strand cleavage by chimeric restriction enzymes Institute of Experimental Hematology with zinc finger DNA-recognition domains. Nucleic Acids Hannover Medical School Res. 28, 3361–3369. Carl-Neuberg-Straße 1 Smith, R.H., Levy, J.R., and Kotin, R.M. (2009). A simplified baculovirus-AAV expression vector system coupled with one- D-30625 Hannover step affinity purification yields high-titer rAAV stocks from Germany insect cells. Mol. Ther. 17, 1888–1896. So¨llu¨ , C., Pars, K., Cornu, T.I., et al. (2010). Autonomous zinc- E-mail: [email protected] finger nuclease pairs for targeted chromosomal deletion. Nu- cleic Acids Res. 38, 8269–8276. Received for publication August 1, 2011; Szczepek, M., Brondani, V., Buchel, J., et al. (2007). Structure- accepted after revision October 5, 2011. based redesign of the dimerization interface reduces the tox- icity of zinc-finger nucleases. Nat. Biotechnol. 25, 786–793. Published online: October 7, 2011. Summary of results and discussion 43

5. Summary of results and discussion

Modifications of complex genomes hold great potential for use in functional genomics, proteomics, the study of biochemical pathways, drug discovery, therapeutic applications, biotechnology, agriculture and even material science. The treatment of human monogenic diseases, such as sickle cell anaemia, thalassemia, haemophilia, cystic fibrosis, glycogen storage diseases, Huntington disease, and Wiskott-Aldrich Syndrome (WAS), will not only saves billions of Euro in the healthcare sector but it will also improves the quality of life of the affected patients.

In the last few decades several targeting reagents have been developed that improve the genome editing process. These include siRNA [288], Sleeping Beauty transposon [289], triplex-forming oligonucleotides (TFOs) [290], DNA cutter ARCUT [291], poly-amides [292, 293], peptide-nucleic acids [294], meganucleases [295], ZFNs [296, 297] and TALENs [298].

Whereas ZFNs and TALENs are promising tools for precise genome modification, the former are well studied and characterized as compared to the latter. Although ZFNs have limitations, including the labour intense production process, off-target activity and accompanying toxicity, the overall benefits of this tool have greatly revolutionised the field of targeted genome editing. The reason behind our and most of other investigators choice to use ZFNs as targeting reagent, are the benefits over other targeting reagents, which includes: the ability to produce permanent and heritable genome modification that need only a transient expression of ZFNs, the rapid kinetics of genome editing, functionality in a variety of mammalian cell types and model organisms, editing of the genome with a single transfection/transduction round without the need of repeated introduction, rather rapid generation of modified cell lines and model organisms, the ability of generating mono and multiallelic genome editing, no prerequisite for the incorporation of non-native exogenous DNA sequences such as antibiotics resistant genes for selection, editing of large DNA segment by simultaneously involving two distant genomic sequences with either two or a single pair of ZFNs as well as the possibility of introducing ZFNs in various form such as DNA, mRNA, or protein.

The mode of action of ZFNs is rather straight-forward. Upon introduction into cells, they act like DNA scissors to create a double strand break (DSB) by hydrolysing the phosphodiester bonds of the nucleotides. Cells detect these breaks as potentially lethal damage and generally repair it by either homologous recombination (HR) or non- homologous end joining (NHEJ). HR is a precise pathway where under natural conditions the cellular repair machinery uses the sister chromatid as a template to repair the DSB. In genome engineering, this pathway also accepts exogenous donor DNA as a template for repairing the DSB. The relatively low frequencies of HR in mammalian cells can be increased significantly by a DSB introduced by ZFNs. In contrast, the NHEJ is error-prone Summary of results and discussion 44 pathway in which the overhanging ends of the DSB are either incorrectly filled or chewed back followed by ligation of the two ends. The result is deletions, insertions and/or substitutions at the break site. This approach has been used to create targeted gene disruption or knockout in a variety of systems, including cultured cell lines, primary stem cells and whole model organisms.

Based on these two major DNA repair pathways, the ZFN technology can be harnessed for various genome editing strategies including targeted gene addition, correction and disruption as well as chromosomal rearrangement.

A major factor for the successful application of ZFNs in genome editing is specificity which is linked to their activity and the associated toxicity. The ZFNs, more elaborately, consist of three domains: an artificially designed sequence-specific DNA binding domain, an unspecific DNA cleavage domain derived from the restriction enzyme FokI, and a short inter-domain linker which connects the first two domains. This modular structure of ZFNs allowed for optimizing and selecting each of the domains independently from each other. In the beginning of my thesis I screened for the best DNA binding domain in terms of activity and the most-optimized nuclease domain to reduce toxicity.

The nuclease domain is a critical component of the ZFN that affect its activity and specificity. For efficient DNA cleavage activity, the natural FokI enzyme requires the dimerization of two catalytic domains [266]. However, due to homodimerization of two FokI cleavage domains, two identical ZFN monomers could cleave at degenerated (depending on the specificity) homodimeric target sites. The crucial advancement for ZFN technology was the development of obligate heterodimeric nuclease variants with significantly reduced ZFN-associated toxicity [65, 299]. Initially I tested different variants with wild type to identify an obligate heterodimeric nuclease with comparable activity to wild type. For my particular target, I observed that the KV/EA variant [65] had comparable activity to wild type. These obligate heterodimeric FokI variants were than used for all subsequent studies.

Next, I looked for highly active DNA binding domains to target different positions in EGFP. These ZFs were developed by Maeder et al. using the OPEN method [40]. In this study the authors showed that OPEN derived ZFNs are more effective than previously published modular assembly methods. In their study they constructed 37 ZFN pairs by OPEN method, which were used to specifically edit the genome with allelic frequencies of 1% to 50%. On top of the five ZFNs to target endogenous loci, they made ZFNs against five target sites in EGFP (E223, E292, E382, E502 and E568) and among these two ZFN pairs specific for target sites 292 and 502 were shown to be the most efficient pairs with knockout activities ranging from ~19-29% and ~25-28%, respectively. For my thesis, I selected the E292 and E502 specific ZFNs for further optimization. By cloning them in different mammalian expression plasmids that contain an upstream intron, a six-nucleotide linker sequence between the ZFs and the nuclease domain, and obligate heterodimeric FokI variants, I increased the EGFP knockout efficiency of E292 and E502 to ~41% and ~54%, respectively. Summary of results and discussion 45

In manuscript 1 the knockout efficiencies at a single EGFP locus in keratinocyte stem cells (KSCs) derived from EGFP-transgenic mice was determined after introducing the optimized EGFP-specific ZFN pair E502. The EGFP negative cells were then used to evaluate the impact of ZFN mediated genome editing on the stem cell potential and to assess gene knockout quantitatively and qualitatively at the genomic level.

Depending on the amount of ZFN expression vector used, EGFP was knocked out in 2.6% to 18% of treated cells. Assessment of ZFN-induced toxicities revealed that an increase in the fraction of EGFP negative cells was associated with higher cytotoxicity. This increase in cytotoxicity is probably related to high ZFN expression levels rather than a result of an increase in the amount of foreign DNA in the cells, as transfection of the same amount of empty vector DNA showed no detectable cytotoxicity. This increase in cytotoxicity is in accord to the previous suggestion made by our group that high expression levels of ZFNs are not well tolerated by the cells, probably leading to their elimination over time through accumulation of unspecific DSBs [300] and decrease in cell survival [287]. Although the reason for this ZFN-induced cytotoxicity is not entirely clear, it is most likely associated with excessive cleavage at off-target sites, which, in turn, suggests imperfect target site recognition by the ZF DNA binding domains [301]. Albeit highly specific, ZFNs also exhibit off-target activity, as shown by Gabriel et al. [302]. Our study demonstrated that a certain number of EGFP-negative KSCs could not be exceeded. When further augmenting the amount of ZFN expression plasmids in the transfection, the relative increase in the EGFP knockout frequency was solely due to increased cell death. In this study these dose escalation experiments suggested that increasing the number of knockout KSCs can only be achieved by optimization of ZF specificity to reduce off-target activity and further redesign of the FokI nuclease dimer interface to improve the balance between activity and cytotoxicity.

I investigated how the optimized E502-specific ZFN modified the target site in the KSCs. As discussed earlier, in the absence of a donor DNA the ZFN induced DSBs are repaired by the error-prone NHEJ repair pathway that often leads to small insertions and deletions (indels) at the break site. I verified EGFP target site modification in sorted EGFP-negative KSCs using the mismatch-sensitive T7 endonuclease I (T7E1), which cuts heteroduplex DNA formed by mismatches between edited and unmodified target sites in a PCR segment produced from the target site. The assay indicated that 68% of alleles have been edited in these EGFP negative KSC population, which is likely an underestimation as quantification above 50% are not accurate in this assay. Next, I looked for more direct evidence of ZFN specific target site modification by sub-cloning the target site derived from genomic DNA of sorted EGFP-negative KSC. Sequence analysis revealed several distinct mutations, including deletions of 1, 2, and 24 base pairs as well as an insertion of 2 base pairs. These indels prove the efficiency of ZFNs to permanently disrupt the specific target genomic site.

The permanent treatment of many genetic diseases will rely on gene editing of tissue- specific stem cells. The result of our proof-of-principle study shows that the ZFN technology is a promising and safe genome editing tool to achieve specific gene disruption Summary of results and discussion 46 in adult stem cells. The effect of ZFN treatment on stemness of adult multipotent stem cell has been poorly studied thus far and was mainly restricted to hematopoietic stem cells (HSCs) and mesenchymal stem cells (MSCs) [303]. Beside these two stem cell types, KSCs are probably the best investigated most easily accessible and transplantable adult stem cells type. These properties make KSC an ideal candidate for gene based therapies of genetic skin diseases. The possibility of using KSCs for treating skin diseases was demonstrated some 24 years ago by Morgan et al. where they manipulated KSC ex vivo by introducing a recombinant human growth hormone gene using recombinant retroviruses [304]. Our proof-of-principle study showed the possibility of using ZFN-mediated ex vivo genome manipulation of KSC as it fulfils both safety and efficiency requirements for clinical use in gene based therapies.

In conclusion, we in this proof-of-principle study showed that ZFNs provide a highly efficient genome editing tool for the functional disruption or knockout of specific endogenous alleles in KSCs and that ZFN-induced toxicity doesn’t impair the stem cell potential of KSCs. These manipulated KSCs can serve as cell therapeutics in regenerative medicine as well as in vitro models for different investigations such as drug screening. However further studies are necessary to investigate long term in vivo effects of ZFN treatment on KSCs, including carcinogenesis or accelerated aging. Also transient expression of the ZFN is sufficient for ex vivo genome editing in KSCs. Owing to the high gene disruption efficiency, as evident by genotyping, the isolation of edited cells doesn’t require the selectable markers, which may hinder its downstream applications. In addition the simplicity of the technique will also allow for specific disruption of multiple alleles either sequentially or in parallel by using autonomous ZFN pairs. Furthermore, given the well established method for isolation, expansion and grafting of autologous human KSCs, our findings can be transferred easily to human system to establish ZFN-mediated ex vivo genome editing of KSCs for treatment of several genetic skin diseases.

The delivery mode of the necessary components in ZFN-mediated genome editing is an important aspect. The ZFN expression cassettes can be delivered either by non-viral or viral methods. In the first manuscript we introduced ZFN expression plasmids into KSCs by a commercially available, liposome-based transfection reagent. Whereas non-viral gene delivery methods may have advantages of low cost for production, ease of synthesis, relative long shelf life, low immunogenicity and degradability, they generally have low efficiency, especially for in vivo gene delivery. In contrast, owing to their highly evolved and specialized components viral systems are generally more effective means of DNA delivery, achieving high efficiencies for both delivery and expression [70]. Viral vectors can be designed based on different requirements, including efficient and easy production, safety aspects, sustained and/or regulated transgene expression, specific targeting of cells/organs, infection of dividing and non-dividing cells and site-specific integration [296]. Ideally, ZFNs are transiently expressed in high concentration, just long enough to create the DSB [305]. As discussed earlier, high expression of ZFN for prolonged period of time can lead to off-target activities and hence nucleases-associated cytotoxicity [277, 300, 306- 308]. Thus viral vectors like retroviral vectors and lentiviral vectors that integrate into the Summary of results and discussion 47 host genome are not a choice for the delivery of ZFN. However, non-integrating episomal viral vectors, such as IDLVs, adenoviral vectors and AAV may be ideal to deliver ZFNs.

The aim of manuscript 2 was to combine highly efficient gene transfer technology of AAV- based vectors with precise genome editing of ZFNs. Human U2OS osteosarcoma cells that stably express from an integrated lentiviral provirus either a functional EGFP for knockout studies or a mutated EGFP for gene correction approaches were used as paradigmatic model cell types. In parallel, system-associated toxicity was evaluated by determining relative cell survival rate and cell cycle profiles by flow cytometry.

To verify AAV-ZFN mediated excision of an entire provirus including the EGFP expression cassette at genome level I sorted the EGFP negative cells and extracted the genomic DNA from expanded single clones for genotyping. PCR-based amplification of the genomic DNA region confirmed the excision of the transgene cassette in the majority of the EGFP negative clones. This approach of excising an entire gene can be applied in numerous ways, including removal of undesirable exogenous sequences which are introduced into the cells during genome modification. Selectable marker genes, such as antibiotic resistance gene, are routinely used for the screening and selection of genetically modified cells. However, these marker genes have several potential disadvantages, such as the alteration of the expression of adjacent endogenous gene as well as activation of oncogenes. Our platform can be used to remove the selectable marker gene after selecting the genetically modified cells. Moreover, this approach also has potential utility in the recently emerging field of cell reprogramming. The discovery of induced pluripotent stem cells (iPSCs) has made way for unprecedented approaches for regenerative medicine, understanding human disease and drug discovery [309]. However, efficient generation of iPSC requires the integration of a reprogramming cassette consisting of three to four specific transcription factors. Integration of this reprogramming cassette is of concern, including the risk of mutation [309] and the potential risk of tumour formation by some of the transcription factors owing to their oncogenic properties [310]. Since it has been already shown that AAV and ZFNs can be used for gene targeting in iPSC without impairing their potential pluripotency [20, 24, 311], one could use AAV-ZFNs to precisely excise the reprogramming cassette.

It is already known that although integration of wild type AAV2 is targeted to the AAVS1 locus by the viral Rep proteins [240], AAV vectors persist episomally in the target cell nucleus [312]. However, it is also known that episomal DNA vectors, such as AAV based vectors, integrate into naturally occurring DSBs [313]. Several studies, including from our group, have shown that viral vectors based on IDLV and AAV prominently integrate into nuclease mediated DSBs [314-316]. To assess whether AAV-ZFN expression cassette integrated into DSBs induced by the EGFP-specific ZFNs, we used a PCR amplification strategy to detect vector-host genome junctions at the E502 target site in the EGFP locus. Albeit at low frequencies, the PCR reaction confirmed the integration of AAV vectors into the ZFN-induced DSB. Although integration of AAV vectors into nuclease-induced DSBs occurred at low frequencies it seems that a majority of the integrations occurred outside Summary of results and discussion 48 the target locus [315], most likely in nuclease mediated off-target sites and naturally occurring DSBs [302, 316]. Integration of AAV vector DNA into the cell genome is hence a double-edged sword where, at one hand, it can be used for gene targeting in cells in vitro [317, 318] and in vivo [319] by HR and, on other hand, it can lead to the tumour formation [320] if integration is random by NHEJ.

Different studies have demonstrated that concentration and configuration of the donor DNA are important parameters for HR-based genome editing [315, 321, 322] and that AAV vectors are efficient templates for HR, both in the absence [317, 323] and presence of a DSB [315, 324-326]. However using a similar setup in this study and some previous studies our laboratory was unable to observe AAV-based gene targeting in the absence of the DSB-inducing enzymes. This may be attributed to the architecture of the donor, which was designed and optimized for DSB-induced gene targeting.

Due to the limiting cargo capacity of AAV vectors, we had to apply three AAV vectors simultaneously in the gene correction approaches, one as a donor and one for each subunit of the ZFN. As opposed to an approach where we could use a single vector that contained donor sequences combined with a nuclease expression cassette [287, 315] we observed a significant drop in gene correction frequencies, which may be related to vector- associated phenomena, including saturation of receptor-mediated endocytosis and intracellular transport. The generation of a single AAV vector construct harbouring a combined 2-in-1 ZFN expression cassette by fusing the two subunits via a 2A autoproteinase is a strategy to overcome this problem.

As discussed earlier, a transient high expression of ZFNs is crucial in many cases because prolonged expression of the designer nucleases may produce toxic side effects on the treated cells. To this end, I assessed the expression kinetics of AAV-based transgene expression both at the protein and RNA level. For this purpose I transduced U2OS cells with AAV-based EGFP expression vector and AAV-ZFN vectors separately. It was observed that EGFP and ZFN expression dropped quickly in dividing cells, suggesting that episomal vector genomes are rapidly diluted in cultured cells and hence confirming the transient, short-lived nature of ZFN expression by AAV in mitotic cells.

In all experiments I observed better efficiencies of AAV-ZFN mediated genome editing with higher doses of AAV vectors. This has also been shown for IDLVs and adenoviral vectors where only relatively high vector doses express sufficient amounts of ZFN [286, 327]. However, although in many experiments I observed efficient NHEJ and HR-based genome editing only with high AAV vector copy numbers per cell, these numbers were comparable to non-viral gene deliver methods, such as nucleofection, where similar copy numbers of plasmid DNA is applied per cell. High vector dose required for sufficient ZFN expression can be reduced by the use of self-complimentary AAV (scAAV) vectors [227, 328] which harbour double-stranded DNA genome and are therefore not dependent on the conversion of single-stranded vector genome into double-stranded DNA for transgene expression. This has been shown for the model nuclease I-SceI where scAAV vectors performed better Summary of results and discussion 49 than single-stranded AAV [324]. However, the limited vector genome size of scAAV will not allow for the packaging of 2-in-1 ZFN expression cassettes or the combination of ZFN and donor in one vector.

The target site of a ZFN pair, which usually includes three or four ZFs per monomer, encompasses 18 or 24 bp, which is statistically large enough to define a unique site in a human genome [329]. However, ZFNs revealed undesirable off-target activities [302] which lead to ZFN-associated toxicity [274, 287, 299, 300, 330-332], a major concern in ZFN- mediated genome editing in therapeutic approaches. These toxicities can be theoretically subdivided into cytotoxicity, genotoxicity, immunotoxicity or teratogenicity. In the context of ZFN-associated toxicities, genotoxicity is any deleterious action that affects the integrity of the cellular genome, immunotoxicity is the immune response to the ZFN itself or a genetically modified cell, while teratogenicity is the ability of a compound to cause birth defects [305]. Cytotoxicity, on the other hand, is a simple parameter determined by measuring cell viability and overall cell integrity. For the quantitative assessment of AAV- ZFN associated cytotoxicity I performed cell cycle analysis and determined the percentage of apoptotic cells by quantification of the subG1 fraction. While not detecting any significant sign of apoptosis, the cell cycle profiling demonstrated a mild arrest of AAV-ZFN treated cells in the G2/M phase. This arrest in G2/M phase is indicative of DNA damage, which is likely caused by a combination of ZFN on- and off-target activities. It has been shown that in the event of DNA damage the p53 tumour suppressor protein prevents cell death by arresting the cell cycle in the G2 phase [333, 334]. Raj et al. reported that UV-inactivated wild type AAV selectively induces apoptosis only in the cells that lack active p53 while U2OS cells, which are wild type for p53 are not killed by AAV infection but undergo arrest in the G2 phase of the cell cycle [335]. In another study by Saudan et al. it has been shown that the cells were arrested in G2 phase by the AAV Rep protein [336]. Under my experimental conditions I observed only minimal effects of AAV-ZFN on the cell cycle. The absence of Rep expression and UV-irradiation may explain the different experimental outcomes.

In conclusion, I have validated the versatility of combining AAV vectors with the ZFN platform for rational targeted genome editing. The AAV-ZFN platform was employed successfully for gene deletion, gene disruption and gene correction with no apparent cytotoxic effect with the exception of a mild cell cycle arrest. Owing to the efficient ex vivo transduction of a variety of cells by AAV vectors, including iPSC and mesenchymal stem cells, this platform will open the door for a wide range of applications in regenerative medicine by combining cell therapy with targeted genome editing. Furthermore, given the safety record of AAV in clinical and preclinical trials AAV-ZFN platform can be employed for in vivo genome editing even for tissues comprising of post mitotic cells such as liver and cardiac tissues [337].

I further validated the AAV-ZFN platform for in vivo gene disruption in the mammalian heart. For this purpose, a transgenic C57BL/6J mouse model that contains a single copy of EGFP locus was used [338]. To this end, an AAV vector expressing both subunits of the Summary of results and discussion 50

E502-specific ZFN pair was introduced into the heart muscle of adult mice by direct myocardial injection. Before in vivo application, the AAV vector encoding plasmid was verified for activity in vitro as shown in Figure 1A and C. Expression of ZFN, in vivo , was confirmed by PCR and the absence of EGFP protein in about 30% of cells was revealed by immunohistology of the heart muscle in sections close to the injection site.

The 2-in-1 ZFN construct allows for transcription of both subunits from a single promoter as after translation the two monomers are separated from each other by the self-cleavage activity of the 2A autoproteinase (Figure 1B). To express the ZFNs I used either the cytomegalovirus (CMV) or the cardiac specific myosin light chain (MLC) promoter. Owing to the relatively weak expression by the MLC promoter [339], I used a variant which contained the CMV enhancer at the 5’-end [14]. Initially the activities of these constructs were evaluated by transfecting U2OS cells which stably express a destabilized EGFP. The Summary of results and discussion 51 results confirmed the ability of a single AAV plasmid harbouring a 2-in-1 ZFN cassette to produced EGFP negative cells (Figure 1A), paving the way to produce the viral vector for in vivo applications in the mammalian heart. The ability of delivering both ZFN subunits into the heart by a single viral vector is of prime importance as a single vector administration is expected to increase the efficiency of genome editing in the heart.

There are different serotypes of AAV vectors and each serotype has a distinct cell tropism, including post-mitotic cells in cardiac tissues [200]. Serotypes AAV1, AAV6, AAV8 and AAV9 have been identified as the most cardiotropic [340, 341]. The selection among these AAV serotypes depends upon the experimental setup and the ultimate goal of the study. Yang et al. showed that AAV9 has more than 13-fold higher gene transfer efficiency than AAV6 after intra-venous injection and that AAV9 transduced nearly 100% of cardiomyocytes in hamsters [342]. In another study Zincarelli et al. analysed AAV serotypes 1-9 for gene expression and tropism in mice. They observed that AAV9 mediated the most robust and rapid onset of transgene expression among the cardiotropic AAV serotypes [340]. In one another important study, Inagaki et al. showed that both AAV9 and AAV8 cross vascular endothelial cell barriers very efficiently, but that AAV9 was superior for cardiac gene delivery as it transduced myocardium 5-10 fold more efficiently than AAV8, resulting in over 80% of cardiomyocytes transduced in mice [343]. In brief, from all the above mentioned studies it can be concluded that AAV6 is most efficient to transduce primary cardiomyocytes in vitro while AAV9 is best suited for in vivo transduction of cardiomyocytes due to its efficiency to cross endothelial cell barrier as well as a fast onset rate and higher transgene expression levels. Based on these information I have chosen the AAV9 serotype to deliver our EGFP-specific 2-in-1 ZFN cassette to specifically disrupt EGFP in the heart of the EGFP mouse.

In our study we introduced the AAV9-ZFN (2-in-1) vector into the left ventricle of adult EGFP mice by three direct intra-myocardial injections to ensure reliable and robust expression of the ZFN in the target area. Immunohistology revealed clusters of EGFP negative cardiomyocytes (Figure 2A), suggesting the transduction of several localized areas. Such a pattern is in agreement to results shown by several studies where adjacent cardiomyocytes express the transgene after intramyocardial injection of AAV. Another interesting factor will be to evaluate the effect of the administration route of the viral vector on the efficiency of the AAV9-ZFN platform for specific disruption of EGFP in the heart. To this end I applied the AAV9-ZFN vector also intra-peritoneally. These probes will be analysed in the near future. Although the results of the intra-myocardial vector administration clearly demonstrated the absence of EGFP in several cardiomyocytes, further confirmation on the genomic level will be necessary. Genotyping will be performed similarly as described in manuscripts 1 and 2.

In our study I have sacrificed the mice 4 weeks after intra-myocardial administration of AAV9-ZFN. Generally, AAV9 mediated in vivo transgene expression in cardiomyocytes is Summary of results and discussion 52

detected after 2 weeks of systematic delivery [340, 342]. The 4-week time point was chosen to give enough time for the ZFNs to disrupt EGFP and clearance of previously expressed pre-existing EGFP by cardiomyocytes. Expression of ZFN was still detectable 4 weeks after delivery of the AAV9-ZFN vector (Figure 2B), which is in agreement with the known long-lasting AAV9-mediated transgene expression in cardiomyocytes. For instance, Chu et al., by comparing AAV and adenoviral vectors for efficiency and stability of transgene expression in mammalian hearts after intra-myocardial administration, detected transgene expression up to 7 days after adenoviral vector delivery and up to 3 months after administration of AAV vectors [344].

In brief, two major conclusions may be drawn from this study. First, both ZFN subunits can be expressed from a single expression cassette in vivo when separating the two ZFN open reading frames by a 2A autoproteinase. Second, the AAV-ZFN platform is efficient enough for disrupting a specific gene in vivo. This suggests that such an approach may be useful to achieve a local gene knockout for basic research questions. Moreover, even therapeutic applications may be feasible, such as the protection of the heart muscle from persistent infection with coxsackievirus by disrupting CAR (coxsackievirus and adenovirus receptor). Summary 53

6. Summary

This cumulative doctoral thesis introduces two proof-of-principle studies of targeted genome editing in different experimental settings based on the use of synthesised zinc- finger nucleases (ZFNs) designed to specifically target EGFP. The delivery of these targeting reagents was established with both non-viral and viral vectors. In addition to the assessment of the efficacy of ZFN mediated genome engineering, I evaluated the impact of system associated toxicities.

In the first manuscript keratinocyte stem cells (KSCs) derived from an EGFP-transgenic mouse model were used for ex vivo knockout of EGFP by using optimised EGFP specific ZFN delivered by a non-viral transfection method. The KSCs were than evaluated for impact of ZFN treatment on the stem cell properties as well as ZFN associated cytotoxicity. The results proved ZFNs to be a highly efficient genome editing tool for functional disruption of a marker gene and that ZFN-associated toxicity did not impair the stem cell potential of KSCs. In this publication the modification of KSCs by ZFN was demonstrated for the first time which have the potential to be used for therapeutic application in regenerative medicine as well as to create the model for different investigations such as drug screening

In the second manuscript, I aimed to combine the precise genome editing technology of ZFNs with the highly efficient gene transfer technology of vectors based on adeno- associated virus (AAV). In parallel, system-associated toxicities were evaluated by determining relative cell survival rate and cell cycle profiling. My results validate the versatility of using the AAV-ZFN platform for rational targeted genome editing, including gene deletion, gene disruption and gene correction, with no apparent cytotoxic side effects. These results encourage the use of the AAV-ZFN platform for further in vivo genome editing.

In hitherto unpublished work, I utilized the AAV-ZFN platform for in vivo gene disruption of EGFP in the mouse heart. After initial validation of a single viral vector construct harbouring both ZFN subunits in vitro, this 2-in-1 vector was introduced directly into the myocardium of EGFP transgenic mice. Expression of ZFN in the myocardium was confirmed after AAV9-ZFN injection while histological analysis showed functional disruption of EGFP in cardiomyocytes. However, further ongoing molecular characterization is needed to conclude this study. This proof-of-principle study encourages further in vivo application of the AAV-ZFN platform for future therapeutic applications.

In conclusion, I demonstrated ZFN-mediated genome editing in vitro, ex vivo and in vivo using both non-viral and viral delivery tools. These proof-of-principle studies revealed an acceptable balance between efficacy and platform-associated toxicity, suggesting that these technologies may be applied for basic research to get more insights into biological questions, and maybe translatable into the clinic for therapeutic applications. Zusammenfassung 54

7. Zusammenfassung

In dieser kumulativen Doktorarbeit werden zwei Grundsatzstudien zur gezielten Genomveränderung vorgestellt. Die Arbeiten basieren auf der Benutzung von synthetischen, EGFP-spezifischen Zink-Finger-Nukleasen (ZFNs), welche mittels viralen und nicht-viral Vektoren in die jeweiligen Zellen eingeschleust wurden. Zusätzlich zur Bewertung der Wirksamkeit der ZFN-vermittelten Genommodifikation habe ich den Einfluss der damit verbundenen Toxizität untersucht.

Im ersten Manuskript wurden Keratinozytenstammzellen (KSCs) eines EGFP-transgenen Mausmodells für einen Ex-vivo-Knockout des Markergens mit optimierten EGFP- spezifischen ZFN verwendet. Die KSCs wurden hinsichtlich Knockoutfrequenz, Stammzelleigenschaften nach ZFN-Behandlung sowie ZFN-assoziierter Zytotoxizität untersucht. Die Ergebnisse zeigten auf, dass ZFNs ein hocheffizientes Werkzeug zur Erzeugung eines funktionellen Knockouts sind und dass ZFN-assoziierte Toxizität das Stammzellpotential der KSCs nicht beeinträchtigt. In dieser Publikation wurde zudem zum ersten Mal die ZFN-vermittelte genetische Modifikation von KSCs demonstriert, welche aufgrund ihres Anwendungspotentials in der regenerativen Medizin bzw. als Model für Medikamentenscreening, attraktive neue biotechnologische Möglichkeiten erschafft.

Im zweiten Manuskript habe ich versucht die ZFN-Plattform mit der Gentransfertechnologie von auf Adeno-Assoziierten Viren (AAV)-beruhenden Vektoren zu vereinen. Die Ergebnisse stellten die Vielseitigkeit der AAV/ZFN-Plattform für die gezielte Genommodifikation, einschließlich Gendeletion, Genknockout und Genkorrektur, dar. Zudem zeigten die Untersuchung der plattformassoziierten Toxizität mittels Bestimmung der relativen Zellüberlebensrate und Zellzyklusanalysen keine offensichtlichen zytotoxischen Nebenwirkungen.

Diese Ergebnisse ermutigen mich die AAV/ZFN-Plattform für In-vivo-Genommodifikationen zu verwenden. In bisher unpublizierter Arbeit, habe ich diese Plattform für einen In-vivo- Knockout im Mausherzen eingesetzt. Ein sogenanntes „2-in-1“-Vektorkonstrukt, das beide ZFN-Untereinheiten in einem AAV-Vektor beherbergt, wurde nach In-vitro-Validierung direkt in das Myokardium von EGFP-transgenen Mäusen eingebracht. Die ZFN-Expression im Herzmuskel wurde mittels RT-PCR nachgewiesen und die histologische Analyse zeigte einen funktionellen EGFP-Knockout in den Kardiomyozyten. Nichtsdestotrotz ist eine weiterführende molekulare Charakterisierung zur Vervollständigung dieser Studie nötig. Diese erfolgreiche Grundsatzstudie spornt zu weiteren In-vivo-Anwendungen der AAV/ZFN-Plattform für therapeutische Ansätze an.

Zusammenfassend habe ich ZFN-vermittelte Genommodifikation in vitro, ex vivo und in vivo dargestellt. Diese Grundsatzstudien zeigten eine akzeptierbare Balance zwischen Effizienz und plattformassoziierter Toxizität und demonstrieren, dass diese Technologie sowohl den Weg in die Grundlagenforschung als auch zukünftig für therapeutische Anwendung in die Klinik finden könnte. References 55

8. References

1. Pearson, H., Genetics: what is a gene? Nature, 2006. 441(7092): p. 398-401. 2. Elstein, D., Abrahamov, A., Hadas-Halpern, I. and Zimran, A., Gaucher's disease. Lancet, 2001. 358(9278): p. 324-7. 3. Kay, M. A., State-of-the-art gene-based therapies: the road ahead. Nat Rev Genet, 2011. 12(5): p. 316-28. 4. Rivière, Isabelle, Dunbar, Cynthia E. and Sadelain, Michel, Hematopoietic stem cell engineering at a crossroads. Blood, 2011. 5. Kioussis, D. and Festenstein, R., Locus control regions: overcoming heterochromatin-induced gene inactivation in mammals. Curr Opin Genet Dev, 1997. 7(5): p. 614-9. 6. Ellis, J., Silencing and variegation of gammaretrovirus and lentivirus vectors. Hum Gene Ther, 2005. 16(11): p. 1241-6. 7. Nusse, R., van Ooyen, A., Cox, D., Fung, Y. K. and Varmus, H., Mode of proviral activation of a putative mammary oncogene (int-1) on mouse chromosome 15. Nature, 1984. 307(5947): p. 131-6. 8. Hacein-Bey-Abina, S., Von Kalle, C., Schmidt, M., McCormack, M. P., Wulffraat, N., Leboulch, P., Lim, A., Osborne, C. S., Pawliuk, R., Morillon, E., Sorensen, R., Forster, A., Fraser, P., Cohen, J. I., de Saint Basile, G., Alexander, I., Wintergerst, U., Frebourg, T., Aurias, A., Stoppa-Lyonnet, D., Romana, S., Radford-Weiss, I., Gross, F., Valensi, F., Delabesse, E., Macintyre, E., Sigaux, F., Soulier, J., Leiva, L. E., Wissler, M., Prinz, C., Rabbitts, T. H., Le Deist, F., Fischer, A. and Cavazzana- Calvo, M., LMO2-associated clonal T cell proliferation in two patients after gene therapy for SCID-X1. Science, 2003. 302(5644): p. 415-9. 9. Aiuti, A., Cassani, B., Andolfi, G., Mirolo, M., Biasco, L., Recchia, A., Urbinati, F., Valacca, C., Scaramuzza, S., Aker, M., Slavin, S., Cazzola, M., Sartori, D., Ambrosi, A., Di Serio, C., Roncarolo, M. G., Mavilio, F. and Bordignon, C., Multilineage hematopoietic reconstitution without clonal selection in ADA-SCID patients treated with stem cell gene therapy. J Clin Invest, 2007. 117(8): p. 2233- 40. 10. Cartier, N., Hacein-Bey-Abina, S., Bartholomae, C. C., Veres, G., Schmidt, M., Kutschera, I., Vidaud, M., Abel, U., Dal-Cortivo, L., Caccavelli, L., Mahlaoui, N., Kiermer, V., Mittelstaedt, D., Bellesme, C., Lahlou, N., Lefrere, F., Blanche, S., Audit, M., Payen, E., Leboulch, P., l'Homme, B., Bougneres, P., Von Kalle, C., Fischer, A., Cavazzana-Calvo, M. and Aubourg, P., Hematopoietic stem cell gene therapy with a lentiviral vector in X-linked adrenoleukodystrophy. Science, 2009. 326(5954): p. 818-23. 11. Gaspar, H. B., Cooray, S., Gilmour, K. C., Parsley, K. L., Zhang, F., Adams, S., Bjorkegren, E., Bayford, J., Brown, L., Davies, E. G., Veys, P., Fairbanks, L., Bordon, V., Petropolou, T., Kinnon, C. and Thrasher, A. J., Hematopoietic stem cell gene therapy for adenosine deaminase-deficient severe combined immunodeficiency leads to long-term immunological recovery and metabolic correction. Sci Transl Med, 2011. 3(97): p. 97ra80. 12. Mavilio, F., Pellegrini, G., Ferrari, S., Di Nunzio, F., Di Iorio, E., Recchia, A., Maruggi, G., Ferrari, G., Provasi, E., Bonini, C., Capurro, S., Conti, A., Magnoni, C., Giannetti, A. and De Luca, M., Correction of junctional epidermolysis bullosa by transplantation of genetically modified epidermal stem cells. Nat Med, 2006. 12(12): p. 1397-402. 13. Capecchi, M. R., Gene targeting in mice: functional analysis of the mammalian genome for the twenty-first century. Nat Rev Genet, 2005. 6(6): p. 507-12. 14. Muller, O. J., Leuchs, B., Pleger, S. T., Grimm, D., Franz, W. M., Katus, H. A. and Kleinschmidt, J. A., Improved cardiac gene transfer by transcriptional and References 56

transductional targeting of adeno-associated viral vectors. Cardiovasc Res, 2006. 70(1): p. 70-8. 15. Fukazawa, T., Maeda, Y., Sladek, F. M. and Owen-Schaub, L. B., Development of a cancer-targeted tissue-specific promoter system. Cancer Res, 2004. 64(1): p. 363-9. 16. Geisler, A., Jungmann, A., Kurreck, J., Poller, W., Katus, H. A., Vetter, R., Fechner, H. and Muller, O. J., microRNA122-regulated transgene expression increases specificity of cardiac gene transfer upon intravenous delivery of AAV9 vectors. Gene Ther, 2011. 18(2): p. 199-209. 17. Capecchi, M. R., Generating mice with targeted mutations. Nat Med, 2001. 7(10): p. 1086-90. 18. Sadelain, M., Papapetrou, E. P. and Bushman, F. D., Safe harbours for the integration of new DNA in the human genome. Nat Rev Cancer, 2011. 19. DeKelver, R. C., Choi, V. M., Moehle, E. A., Paschon, D. E., Hockemeyer, D., Meijsing, S. H., Sancak, Y., Cui, X., Steine, E. J., Miller, J. C., Tam, P., Bartsevich, V. V., Meng, X., Rupniewski, I., Gopalan, S. M., Sun, H. C., Pitz, K. J., Rock, J. M., Zhang, L., Davis, G. D., Rebar, E. J., Cheeseman, I. M., Yamamoto, K. R., Sabatini, D. M., Jaenisch, R., Gregory, P. D. and Urnov, F. D., Functional genomics, proteomics, and regulatory DNA analysis in isogenic settings using zinc finger nuclease-driven transgenesis into a safe harbor locus in the human genome. Genome Res, 2010. 20(8): p. 1133-42. 20. Zou, J., Sweeney, C. L., Chou, B. K., Choi, U., Pan, J., Wang, H., Dowey, S. N., Cheng, L. and Malech, H. L., Oxidase-deficient neutrophils from X-linked chronic granulomatous disease iPS cells: functional correction by zinc finger nuclease- mediated safe harbor targeting. Blood, 2011. 117(21): p. 5561-72. 21. Ramachandra, C. J., Shahbazi, M., Kwang, T. W., Choudhury, Y., Bak, X. Y., Yang, J. and Wang, S., Efficient recombinase-mediated cassette exchange at the AAVS1 locus in human embryonic stem cells using baculoviral vectors. Nucleic Acids Res, 2011. 39(16): p. e107. 22. Smith, J. R., Maguire, S., Davis, L. A., Alexander, M., Yang, F., Chandran, S., ffrench-Constant, C. and Pedersen, R. A., Robust, persistent transgene expression in human embryonic stem cells is achieved with AAVS1-targeted integration. Stem Cells, 2008. 26(2): p. 496-504. 23. Yang, L., Soonpaa, M. H., Adler, E. D., Roepke, T. K., Kattman, S. J., Kennedy, M., Henckaerts, E., Bonham, K., Abbott, G. W., Linden, R. M., Field, L. J. and Keller, G. M., Human cardiovascular progenitor cells develop from a KDR+ embryonic-stem- cell-derived population. Nature, 2008. 453(7194): p. 524-8. 24. Hockemeyer, D., Soldner, F., Beard, C., Gao, Q., Mitalipova, M., DeKelver, R. C., Katibah, G. E., Amora, R., Boydston, E. A., Zeitler, B., Meng, X., Miller, J. C., Zhang, L., Rebar, E. J., Gregory, P. D., Urnov, F. D. and Jaenisch, R., Efficient targeting of expressed and silent genes in human ESCs and iPSCs using zinc- finger nucleases. Nat Biotechnol, 2009. 27(9): p. 851-7. 25. Henckaerts, E., Dutheil, N., Zeltner, N., Kattman, S., Kohlbrenner, E., Ward, P., Clement, N., Rebollo, P., Kennedy, M., Keller, G. M. and Linden, R. M., Site- specific integration of adeno-associated virus involves partial duplication of the target locus. Proc Natl Acad Sci U S A, 2009. 106(18): p. 7571-6. 26. Lombardo, A., Cesana, D., Genovese, P., Di Stefano, B., Provasi, E., Colombo, D. F., Neri, M., Magnani, Z., Cantore, A., Lo Riso, P., Damo, M., Pello, O. M., Holmes, M. C., Gregory, P. D., Gritti, A., Broccoli, V., Bonini, C. and Naldini, L., Site-specific integration and tailoring of cassette design for sustainable gene transfer. Nat Methods, 2011. 8(10): p. 861-9. 27. Irion, S., Luche, H., Gadue, P., Fehling, H. J., Kennedy, M. and Keller, G., Identification and targeting of the ROSA26 locus in human embryonic stem cells. Nat Biotechnol, 2007. 25(12): p. 1477-82. 28. Capecchi, M. R., Altering the genome by homologous recombination. Science, 1989. 244(4910): p. 1288-92. References 57

29. Bradley, A., Ramirez-Solis, R., Zheng, H., Hasty, P. and Davis, A., Genetic manipulation of the mouse via gene targeting in embryonic stem cells. Ciba Found Symp, 1992. 165: p. 256-69; discussion 269-76. 30. Hinnen, A., Hicks, J. B. and Fink, G. R., Transformation of yeast. Proc Natl Acad Sci U S A, 1978. 75(4): p. 1929-33. 31. Bhattacharyya, S., Tian, J., Bouhassira, E. E. and Locker, J., Systematic targeted integration to study Albumin gene control elements. PLoS One, 2011. 6(8): p. e23234. 32. Sebastiano, V., Maeder, M. L., Angstman, J. F., Haddad, B., Khayter, C., Yeo, D. T., Goodwin, M. J., Hawkins, J. S., Ramirez, C. L., Batista, L. F., Artandi, S. E., Wernig, M. and Joung, J. K., In situ genetic correction of the sickle cell anemia mutation in human induced pluripotent stem cells using engineered zinc finger nucleases. Stem Cells, 2011. 29(11): p. 1717-26. 33. High, K. A., Gene therapy for haemophilia: a long and winding road. J Thromb Haemost, 2011. 9 Suppl 1: p. 2-11. 34. Yusa, K., Rashid, S. T., Strick-Marchand, H., Varela, I., Liu, P. Q., Paschon, D. E., Miranda, E., Ordonez, A., Hannan, N. R., Rouhani, F. J., Darche, S., Alexander, G., Marciniak, S. J., Fusaki, N., Hasegawa, M., Holmes, M. C., Di Santo, J. P., Lomas, D. A., Bradley, A. and Vallier, L., Targeted gene correction of alpha1-antitrypsin deficiency in induced pluripotent stem cells. Nature, 2011. 478(7369): p. 391-4. 35. Rashid, S. T., Corbineau, S., Hannan, N., Marciniak, S. J., Miranda, E., Alexander, G., Huang-Doran, I., Griffin, J., Ahrlund-Richter, L., Skepper, J., Semple, R., Weber, A., Lomas, D. A. and Vallier, L., Modeling inherited metabolic disorders of the liver using human induced pluripotent stem cells. J Clin Invest, 2010. 120(9): p. 3127-36. 36. Perlmutter, D. H., Autophagic disposal of the aggregation-prone protein that causes liver inflammation and carcinogenesis in alpha-1-antitrypsin deficiency. Cell Death Differ, 2009. 16(1): p. 39-45. 37. Gooptu, B. and Lomas, D. A., Conformational pathology of the serpins: themes, variations, and therapeutic strategies. Annu Rev Biochem, 2009. 78: p. 147-76. 38. Chang, Y. F., Imam, J. S. and Wilkinson, M. F., The nonsense-mediated decay RNA surveillance pathway. Annu Rev Biochem, 2007. 76: p. 51-74. 39. Naldini, L., Ex vivo gene transfer and correction for cell-based therapies. Nat Rev Genet, 2011. 12(5): p. 301-15. 40. Maeder, M. L., Thibodeau-Beganny, S., Osiak, A., Wright, D. A., Anthony, R. M., Eichtinger, M., Jiang, T., Foley, J. E., Winfrey, R. J., Townsend, J. A., Unger- Wallace, E., Sander, J. D., Muller-Lerch, F., Fu, F., Pearlberg, J., Gobel, C., Dassie, J. P., Pruett-Miller, S. M., Porteus, M. H., Sgroi, D. C., Iafrate, A. J., Dobbs, D., McCray, P. B., Jr., Cathomen, T., Voytas, D. F. and Joung, J. K., Rapid "open- source" engineering of customized zinc-finger nucleases for highly efficient gene modification. Mol Cell, 2008. 31(2): p. 294-301. 41. Hoher, T., Wallace, L., Khan, K., Cathomen, T. and Reichelt, J., Highly Efficient Zinc-Finger Nuclease-Mediated Disruption of an eGFP Transgene in Keratinocyte Stem Cells without Impairment of Stem Cell Properties. Stem Cell Rev, 2011. 42. Gregory, S. G., Sekhon, M., Schein, J., Zhao, S., Osoegawa, K., Scott, C. E., Evans, R. S., Burridge, P. W., Cox, T. V., Fox, C. A., Hutton, R. D., Mullenger, I. R., Phillips, K. J., Smith, J., Stalker, J., Threadgold, G. J., Birney, E., Wylie, K., Chinwalla, A., Wallis, J., Hillier, L., Carter, J., Gaige, T., Jaeger, S., Kremitzki, C., Layman, D., Maas, J., McGrane, R., Mead, K., Walker, R., Jones, S., Smith, M., Asano, J., Bosdet, I., Chan, S., Chittaranjan, S., Chiu, R., Fjell, C., Fuhrmann, D., Girn, N., Gray, C., Guin, R., Hsiao, L., Krzywinski, M., Kutsche, R., Lee, S. S., Mathewson, C., McLeavy, C., Messervier, S., Ness, S., Pandoh, P., Prabhu, A. L., Saeedi, P., Smailus, D., Spence, L., Stott, J., Taylor, S., Terpstra, W., Tsai, M., Vardy, J., Wye, N., Yang, G., Shatsman, S., Ayodeji, B., Geer, K., Tsegaye, G., Shvartsbeyn, A., Gebregeorgis, E., Krol, M., Russell, D., Overton, L., Malek, J. A., Holmes, M., Heaney, M., Shetty, J., Feldblyum, T., Nierman, W. C., Catanese, J. J., References 58

Hubbard, T., Waterston, R. H., Rogers, J., de Jong, P. J., Fraser, C. M., Marra, M., McPherson, J. D. and Bentley, D. R., A physical map of the mouse genome. Nature, 2002. 418(6899): p. 743-50. 43. Evans, M. J. and Kaufman, M. H., Establishment in culture of pluripotential cells from mouse embryos. Nature, 1981. 292(5819): p. 154-6. 44. Martin, G. R., Isolation of a pluripotent cell line from early mouse embryos cultured in medium conditioned by teratocarcinoma stem cells. Proc Natl Acad Sci U S A, 1981. 78(12): p. 7634-8. 45. Smithies, O., Gregg, R. G., Boggs, S. S., Koralewski, M. A. and Kucherlapati, R. S., Insertion of DNA sequences into the human chromosomal beta-globin locus by homologous recombination. Nature, 1985. 317(6034): p. 230-4. 46. Manis, J. P., Knock out, knock in, knock down--genetically manipulated mice and the Nobel Prize. N Engl J Med, 2007. 357(24): p. 2426-9. 47. Nelson, R. J., The use of genetic "knockout" mice in behavioral endocrinology research. Horm Behav, 1997. 31(3): p. 188-96. 48. Stanford, W. L., Cohn, J. B. and Cordes, S. P., Gene-trap mutagenesis: past, present and beyond. Nat Rev Genet, 2001. 2(10): p. 756-68. 49. Yu, Y. and Bradley, A., Engineering chromosomal rearrangements in mice. Nat Rev Genet, 2001. 2(10): p. 780-90. 50. Rodriguez-Caballero, A., Torres-Lagares, D., Rodriguez-Perez, A., Serrera-Figallo, M. A., Hernandez-Guisado, J. M. and Machuca-Portillo, G., Cri du chat syndrome: a critical review. Med Oral Patol Oral Cir Bucal, 2010. 15(3): p. e473-8. 51. Hirschhorn, K., Cooper, H. L. and Firschein, I. L., Deletion of short arms of chromosome 4-5 in a child with defects of midline fusion. Humangenetik, 1965. 1(5): p. 479-82. 52. Shaffer, L. G. and Lupski, J. R., Molecular mechanisms for constitutional chromosomal rearrangements in humans. Annu Rev Genet, 2000. 34: p. 297-329. 53. Rabbitts, T. H., Chromosomal translocations in human cancer. Nature, 1994. 372(6502): p. 143-9. 54. Russell, L. B., Hunsicker, P. R., Cacheiro, N. L., Bangham, J. W., Russell, W. L. and Shelby, M. D., Chlorambucil effectively induces deletion mutations in mouse germ cells. Proc Natl Acad Sci U S A, 1989. 86(10): p. 3704-8. 55. Davisson, M. T., Schmidt, C. and Akeson, E. C., Segmental trisomy of murine chromosome 16: a new model system for studying Down syndrome. Prog Clin Biol Res, 1990. 360: p. 263-80. 56. Roderick, T. H. and Hawes, N. L., Nineteen paracentric chromosomal inversions in mice. Genetics, 1974. 76(1): p. 109-17. 57. Stubbs, L., Carver, E. A., Cacheiro, N. L., Shelby, M. and Generoso, W., Generation and characterization of heritable reciprocal translocations in mice. Methods, 1997. 13(4): p. 397-408. 58. Ryder, E., Ashburner, M., Bautista-Llacer, R., Drummond, J., Webster, J., Johnson, G., Morley, T., Chan, Y. S., Blows, F., Coulson, D., Reuter, G., Baisch, H., Apelt, C., Kauk, A., Rudolph, T., Kube, M., Klimm, M., Nickel, C., Szidonya, J., Maroy, P., Pal, M., Rasmuson-Lestander, A., Ekstrom, K., Stocker, H., Hugentobler, C., Hafen, E., Gubb, D., Pflugfelder, G., Dorner, C., Mechler, B., Schenkel, H., Marhold, J., Serras, F., Corominas, M., Punset, A., Roote, J. and Russell, S., The DrosDel deletion collection: a Drosophila genomewide chromosomal deficiency resource. Genetics, 2007. 177(1): p. 615-29. 59. Zheng, B., Sage, M., Cai, W. W., Thompson, D. M., Tavsanli, B. C., Cheah, Y. C. and Bradley, A., Engineering a mouse balancer chromosome. Nat Genet, 1999. 22(4): p. 375-8. 60. Valenzuela, D. M., Murphy, A. J., Frendewey, D., Gale, N. W., Economides, A. N., Auerbach, W., Poueymirou, W. T., Adams, N. C., Rojas, J., Yasenchak, J., Chernomorsky, R., Boucher, M., Elsasser, A. L., Esau, L., Zheng, J., Griffiths, J. A., Wang, X., Su, H., Xue, Y., Dominguez, M. G., Noguera, I., Torres, R., Macdonald, L. E., Stewart, A. F., DeChiara, T. M. and Yancopoulos, G. D., High-throughput References 59

engineering of the mouse genome coupled with high-resolution expression analysis. Nat Biotechnol, 2003. 21(6): p. 652-9. 61. Honma, M., Sakuraba, M., Koizumi, T., Takashima, Y., Sakamoto, H. and Hayashi, M., Non-homologous end-joining for repairing I-SceI-induced DNA double strand breaks in human cells. DNA Repair 2007. 6(6): p. 781-8. 62. Guirouilh-Barbat, J., Huck, S., Bertrand, P., Pirzio, L., Desmaze, C., Sabatier, L. and Lopez, B. S., Impact of the KU80 pathway on NHEJ-induced genome rearrangements in mammalian cells. Mol Cell, 2004. 14(5): p. 611-23. 63. Lee, H. J., Kim, E. and Kim, J. S., Targeted chromosomal deletions in human cells using zinc finger nucleases. Genome Res, 2010. 20(1): p. 81-9. 64. Kim, S., Lee, H. J., Kim, E. and Kim, J. S., Analysis of targeted chromosomal deletions induced by zinc finger nucleases. Cold Spring Harb Protoc, 2010. 2010(8): p. pdb prot5477. 65. Sollu, C., Pars, K., Cornu, T. I., Thibodeau-Beganny, S., Maeder, M. L., Joung, J. K., Heilbronn, R. and Cathomen, T., Autonomous zinc-finger nuclease pairs for targeted chromosomal deletion. Nucleic Acids Res, 2010. 38(22): p. 8269-76. 66. Mertens, F., Johansson, B. and Mitelman, F., Isochromosomes in neoplasia. Genes Chromosomes Cancer, 1994. 10(4): p. 221-30. 67. Walz, K., Caratini-Rivera, S., Bi, W., Fonseca, P., Mansouri, D. L., Lynch, J., Vogel, H., Noebels, J. L., Bradley, A. and Lupski, J. R., Modeling del(17)(p11.2p11.2) and dup(17)(p11.2p11.2) contiguous gene syndromes by chromosome engineering in mice: phenotypic consequences of gene dosage imbalance. Mol Cell Biol, 2003. 23(10): p. 3646-55. 68. Yan, J., Keener, V. W., Bi, W., Walz, K., Bradley, A., Justice, M. J. and Lupski, J. R., Reduced penetrance of craniofacial anomalies as a function of deletion size and genetic background in a chromosome engineered partial mouse model for Smith- Magenis syndrome. Hum Mol Genet, 2004. 13(21): p. 2613-24. 69. Olson, L. E., Richtsmeier, J. T., Leszl, J. and Reeves, R. H., A chromosome 21 critical region does not cause specific Down syndrome phenotypes. Science, 2004. 306(5696): p. 687-90. 70. Luo, D. and Saltzman, W. M., Synthetic DNA delivery systems. Nat Biotechnol, 2000. 18(1): p. 33-7. 71. Mann, M. J., Gibbons, G. H., Hutchinson, H., Poston, R. S., Hoyt, E. G., Robbins, R. C. and Dzau, V. J., Pressure-mediated oligonucleotide transfection of rat and human cardiovascular tissues. Proc Natl Acad Sci U S A, 1999. 96(11): p. 6411-6. 72. Liu, F., Song, Y. and Liu, D., Hydrodynamics-based transfection in animals by systemic administration of plasmid DNA. Gene Ther, 1999. 6(7): p. 1258-66. 73. Zhang, G., Budker, V. and Wolff, J. A., High levels of foreign gene expression in hepatocytes after tail vein injections of naked plasmid DNA. Hum Gene Ther, 1999. 10(10): p. 1735-7. 74. Lewis, D. L., Hagstrom, J. E., Loomis, A. G., Wolff, J. A. and Herweijer, H., Efficient delivery of siRNA for inhibition of gene expression in postnatal mice. Nat Genet, 2002. 32(1): p. 107-8. 75. Yang, N. S. and Sun, W. H., Gene gun and other non-viral approaches for cancer gene therapy. Nat Med, 1995. 1(5): p. 481-3. 76. Jiao, S., Cheng, L., Wolff, J. A. and Yang, N. S., Particle bombardment-mediated gene transfer and expression in rat brain tissues. Biotechnology (N Y), 1993. 11(4): p. 497-502. 77. Yang, N. S., Burkholder, J., Roberts, B., Martinell, B. and McCabe, D., In vivo and in vitro gene transfer to mammalian somatic cells by particle bombardment. Proc Natl Acad Sci U S A, 1990. 87(24): p. 9568-72. 78. Sun, W. H., Burkholder, J. K., Sun, J., Culp, J., Turner, J., Lu, X. G., Pugh, T. D., Ershler, W. B. and Yang, N. S., In vivo cytokine gene transfer by gene gun reduces tumor growth in mice. Proc Natl Acad Sci U S A, 1995. 92(7): p. 2889-93. 79. Zelenin, A. V., Kolesnikov, V. A., Tarasenko, O. A., Shafei, R. A., Zelenina, I. A., Mikhailov, V. V., Semenova, M. L., Kovalenko, D. V., Artemyeva, O. V., References 60

Ivaschenko, T. E., Evgrafov, O. V., Dickson, G. and Baranovand, V. S., Bacterial beta-galactosidase and human dystrophin genes are expressed in mouse skeletal muscle fibers after ballistic transfection. FEBS Lett, 1997. 414(2): p. 319-22. 80. Furth, P. A., Gene transfer by biolistic process. Mol Biotechnol, 1997. 7(2): p. 139- 43. 81. Plank, C. and Rosenecker, J., Magnetofection: the use of magnetic nanoparticles for nucleic acid delivery. Cold Spring Harb Protoc, 2009. 2009(6): p. pdb prot5230. 82. Scherer, F., Anton, M., Schillinger, U., Henke, J., Bergemann, C., Kruger, A., Gansbacher, B. and Plank, C., Magnetofection: enhancing and targeting gene delivery by magnetic force in vitro and in vivo. Gene Ther, 2002. 9(2): p. 102-9. 83. Plank, C., Anton, M., Rudolph, C., Rosenecker, J. and Krotz, F., Enhancing and targeting nucleic acid delivery by magnetic force. Expert Opin Biol Ther, 2003. 3(5): p. 745-58. 84. Vandenbroucke, R. E., Lentacker, I., Demeester, J., De Smedt, S. C. and Sanders, N. N., Ultrasound assisted siRNA delivery using PEG-siPlex loaded microbubbles. J Control Release, 2008. 126(3): p. 265-73. 85. Otani, K., Yamahara, K., Ohnishi, S., Obata, H., Kitamura, S. and Nagaya, N., Nonviral delivery of siRNA into mesenchymal stem cells by a combination of ultrasound and microbubbles. J Control Release, 2009. 133(2): p. 146-53. 86. Li, Y. S., Davidson, E., Reid, C. N. and McHale, A. P., Optimising ultrasound- mediated gene transfer (sonoporation) in vitro and prolonged expression of a transgene in vivo: potential applications for gene therapy of cancer. Cancer Lett, 2009. 273(1): p. 62-9. 87. Lu, Q. L., Liang, H. D., Partridge, T. and Blomley, M. J., Microbubble ultrasound improves the efficiency of gene transduction in skeletal muscle in vivo with reduced tissue damage. Gene Ther, 2003. 10(5): p. 396-405. 88. Inagaki, H., Suzuki, J., Ogawa, M., Taniyama, Y., Morishita, R. and Isobe, M., Ultrasound-microbubble-mediated NF-kappaB decoy transfection attenuates neointimal formation after arterial injury in mice. J Vasc Res, 2006. 43(1): p. 12-8. 89. Yang, L., Shirakata, Y., Tamai, K., Dai, X., Hanakawa, Y., Tokumaru, S., Yahata, Y., Tohyama, M., Shiraishi, K., Nagai, H., Wang, X., Murakami, S., Sayama, K., Kaneda, Y. and Hashimoto, K., Microbubble-enhanced ultrasound for gene transfer into living skin equivalents. J Dermatol Sci, 2005. 40(2): p. 105-14. 90. Tsunoda, S., Mazda, O., Oda, Y., Iida, Y., Akabame, S., Kishida, T., Shin-Ya, M., Asada, H., Gojo, S., Imanishi, J., Matsubara, H. and Yoshikawa, T., Sonoporation using microbubble BR14 promotes pDNA/siRNA transduction to murine heart. Biochem Biophys Res Commun, 2005. 336(1): p. 118-27. 91. Chen, S., Ding, J. H., Bekeredjian, R., Yang, B. Z., Shohet, R. V., Johnston, S. A., Hohmeier, H. E., Newgard, C. B. and Grayburn, P. A., Efficient gene delivery to pancreatic islets with ultrasonic microbubble destruction technology. Proc Natl Acad Sci U S A, 2006. 103(22): p. 8469-74. 92. Aihara, H. and Miyazaki, J., Gene transfer into muscle by electroporation in vivo. Nat Biotechnol, 1998. 16(9): p. 867-70. 93. Mir, L. M., Bureau, M. F., Gehl, J., Rangara, R., Rouy, D., Caillaud, J. M., Delaere, P., Branellec, D., Schwartz, B. and Scherman, D., High-efficiency gene transfer into skeletal muscle mediated by electric pulses. Proc Natl Acad Sci U S A, 1999. 96(8): p. 4262-7. 94. Banga, A. K. and Prausnitz, M. R., Assessing the potential of skin electroporation for the delivery of protein- and gene-based drugs. Trends Biotechnol, 1998. 16(10): p. 408-12. 95. Oshima, Y., Sakamoto, T., Yamanaka, I., Nishi, T., Ishibashi, T. and Inomata, H., Targeted gene transfer to corneal endothelium in vivo by electric pulse. Gene Ther, 1998. 5(10): p. 1347-54. 96. Rols, M. P., Delteil, C., Golzio, M., Dumond, P., Cros, S. and Teissie, J., In vivo electrically mediated protein and gene transfer in murine melanoma. Nat Biotechnol, 1998. 16(2): p. 168-71. References 61

97. Heller, R., Jaroszeski, M., Atkin, A., Moradpour, D., Gilbert, R., Wands, J. and Nicolau, C., In vivo gene electroinjection and expression in rat liver. FEBS Lett, 1996. 389(3): p. 225-8. 98. Tsukakoshi, M., Kurata, S., Nomiya, Y., Ikawa, Y. and Kasuya, T., A Novel Method of DNA Transfection by Laser Microbeam Cell Surgery. Applied Physics B- Photophysics and Laser Chemistry, 1984. 35(3): p. 135-140. 99. Rhodes, K., Clark, I., Zatcoff, M., Eustaquio, T., Hoyte, K. L. and Koller, M. R., Cellular laserfection. Methods Cell Biol, 2007. 82: p. 309-33. 100. Baumgart, J., Humbert, L., Boulais, E., Lachaine, R., Lebrun, J. J. and Meunier, M., Off-resonance plasmonic enhanced femtosecond laser optoporation and transfection of cancer cells. Biomaterials, 2011. 101. Zeira, E., Manevitch, A., Khatchatouriants, A., Pappo, O., Hyam, E., Darash- Yahana, M., Tavor, E., Honigman, A., Lewis, A. and Galun, E., Femtosecond infrared laser-an efficient and safe in vivo gene delivery system for prolonged expression. Mol Ther, 2003. 8(2): p. 342-50. 102. Kohli, V., Robles, V., Cancela, M. L., Acker, J. P., Waskiewicz, A. J. and Elezzabi, A. Y., An alternative method for delivering exogenous material into developing zebrafish embryos. Biotechnol Bioeng, 2007. 98(6): p. 1230-41. 103. Ruponen, M., Ronkko, S., Honkakoski, P., Pelkonen, J., Tammi, M. and Urtti, A., Extracellular glycosaminoglycans modify cellular trafficking of lipoplexes and polyplexes. J Biol Chem, 2001. 276(36): p. 33875-80. 104. Felgner, P. L., Gadek, T. R., Holm, M., Roman, R., Chan, H. W., Wenz, M., Northrop, J. P., Ringold, G. M. and Danielsen, M., Lipofection: a highly efficient, lipid-mediated DNA-transfection procedure. Proc Natl Acad Sci U S A, 1987. 84(21): p. 7413-7. 105. Tagami, T., Suzuki, T., Matsunaga, M., Nakamura, K., Moriyoshi, N., Ishida, T. and Kiwada, H., Anti-angiogenic therapy via cationic liposome-mediated systemic siRNA delivery. Int J Pharm, 2012. 422(1-2): p. 280-9. 106. Mevel, M., Sainlos, M., Chatin, B., Oudrhiri, N., Hauchecorne, M., Lambert, O., Vigneron, J. P., Lehn, P., Pitard, B. and Lehn, J. M., Paromomycin and neomycin B derived cationic lipids: Synthesis and transfection studies. J Control Release, 2011. 107. Brigham, K. L., Meyrick, B., Christman, B., Magnuson, M., King, G. and Berry, L. C., Jr., In vivo transfection of murine lungs with a functioning prokaryotic gene using a liposome vehicle. Am J Med Sci, 1989. 298(4): p. 278-81. 108. Ono, T., Fujino, Y., Tsuchiya, T. and Tsuda, M., Plasmid DNAs directly injected into mouse brain with lipofectin can be incorporated and expressed by brain cells. Neurosci Lett, 1990. 117(3): p. 259-63. 109. Son, K. and Huang, L., Factors influencing the drug sensitization of human tumor cells for in situ lipofection. Gene Ther, 1996. 3(7): p. 630-4. 110. Plautz, G. E., Yang, Z. Y., Wu, B. Y., Gao, X., Huang, L. and Nabel, G. J., Immunotherapy of malignancy by in vivo gene transfer into tumors. Proc Natl Acad Sci U S A, 1993. 90(10): p. 4645-9. 111. Raz, E., Carson, D. A., Parker, S. E., Parr, T. B., Abai, A. M., Aichinger, G., Gromkowski, S. H., Singh, M., Lew, D., Yankauckas, M. A. and et al., Intradermal gene immunization: the possible role of DNA uptake in the induction of cellular immunity to viruses. Proc Natl Acad Sci U S A, 1994. 91(20): p. 9519-23. 112. Nabel, G. J., Nabel, E. G., Yang, Z. Y., Fox, B. A., Plautz, G. E., Gao, X., Huang, L., Shu, S., Gordon, D. and Chang, A. E., Direct gene transfer with DNA-liposome complexes in melanoma: expression, biologic activity, and lack of toxicity in humans. Proc Natl Acad Sci U S A, 1993. 90(23): p. 11307-11. 113. Caplen, N. J., Alton, E. W., Middleton, P. G., Dorin, J. R., Stevenson, B. J., Gao, X., Durham, S. R., Jeffery, P. K., Hodson, M. E., Coutelle, C. and et al., Liposome- mediated CFTR gene transfer to the nasal epithelium of patients with cystic fibrosis. Nat Med, 1995. 1(1): p. 39-46. 114. Lv, H., Zhang, S., Wang, B., Cui, S. and Yan, J., Toxicity of cationic lipids and cationic polymers in gene delivery. J Control Release, 2006. 114(1): p. 100-9. References 62

115. Tang, M. X. and Szoka, F. C., The influence of polymer structure on the interactions of cationic polymers with DNA and morphology of the resulting complexes. Gene Ther, 1997. 4(8): p. 823-32. 116. Ruponen, M., Yla-Herttuala, S. and Urtti, A., Interactions of polymeric and liposomal gene delivery systems with extracellular glycosaminoglycans: physicochemical and transfection studies. Biochim Biophys Acta, 1999. 1415(2): p. 331-41. 117. Kopatz, I., Remy, J. S. and Behr, J. P., A model for non-viral gene delivery: through syndecan adhesion molecules and powered by actin. J Gene Med, 2004. 6(7): p. 769-76. 118. Rejman, J., Conese, M. and Hoekstra, D., Gene transfer by means of lipo- and polyplexes: role of clathrin and caveolae-mediated endocytosis. J Liposome Res, 2006. 16(3): p. 237-47. 119. Wagner, E., Zatloukal, K., Cotten, M., Kirlappos, H., Mechtler, K., Curiel, D. T. and Birnstiel, M. L., Coupling of adenovirus to transferrin-polylysine/DNA complexes greatly enhances receptor-mediated gene delivery and expression of transfected genes. Proc Natl Acad Sci U S A, 1992. 89(13): p. 6099-103. 120. Wolfert, M. A., Dash, P. R., Nazarova, O., Oupicky, D., Seymour, L. W., Smart, S., Strohalm, J. and Ulbrich, K., Polyelectrolyte vectors for gene delivery: influence of cationic polymer on biophysical properties of complexes formed with DNA. Bioconjug Chem, 1999. 10(6): p. 993-1004. 121. Kircheis, R., Wightman, L. and Wagner, E., Design and gene delivery activity of modified polyethylenimines. Adv Drug Deliv Rev, 2001. 53(3): p. 341-58. 122. Anderson, D. G., Lynn, D. M. and Langer, R., Semi-automated synthesis and screening of a large library of degradable cationic polymers for gene delivery. Angew Chem Int Ed Engl, 2003. 42(27): p. 3153-8. 123. Wu, G. Y. and Wu, C. H., Receptor-mediated in vitro gene transformation by a soluble DNA carrier system. J Biol Chem, 1987. 262(10): p. 4429-32. 124. He, C. X., Tabata, Y. and Gao, J. Q., Non-viral gene delivery carrier and its three- dimensional transfection system. Int J Pharm, 2010. 386(1-2): p. 232-42. 125. Yang, S. Y., Sun, J. S., Liu, C. H., Tsuang, Y. H., Chen, L. T., Hong, C. Y., Yang, H. C. and Horng, H. E., Ex vivo magnetofection with magnetic nanoparticles: a novel platform for nonviral tissue engineering. Artif Organs, 2008. 32(3): p. 195-204. 126. Gupta, A. K. and Gupta, M., Synthesis and surface engineering of iron oxide nanoparticles for biomedical applications. Biomaterials, 2005. 26(18): p. 3995-4021. 127. Wagner, D. E. and Bhaduri, S. B., Progress and outlook of inorganic nanoparticles for delivery of nucleic Acid sequences related to orthopedic pathologies: a review. Tissue Eng Part B Rev, 2012. 18(1): p. 1-14. 128. Luo, D. and Saltzman, W. M., Enhancement of transfection by physical concentration of DNA at the cell surface. Nat Biotechnol, 2000. 18(8): p. 893-5. 129. Luo, D., Han, E., Belcheva, N. and Saltzman, W. M., A self-assembled, modular DNA delivery system mediated by silica nanoparticles. J Control Release, 2004. 95(2): p. 333-41. 130. Chen, Y., Xue, Z., Zheng, D., Xia, K., Zhao, Y., Liu, T., Long, Z. and Xia, J., Sodium chloride modified silica nanoparticles as a non-viral vector with a high efficiency of DNA transfer into cells. Curr Gene Ther, 2003. 3(3): p. 273-9. 131. Moudgil, S. and Ying, J. Y., Calcium-doped organosilicate nanoparticles for gene delivery vehicles for bone cells. Advanced Materials, 2007. 19(20): p. 3130-+. 132. Muangmoonchai, R., Wong, S. C., Smirlis, D., Phillips, I. R. and Shephardl, E. A., Transfection of liver in vivo by biolistic particle delivery: its use in the investigation of cytochrome P450 gene regulation. Mol Biotechnol, 2002. 20(2): p. 145-51. 133. Ghosh, P., Han, G., De, M., Kim, C. K. and Rotello, V. M., Gold nanoparticles in delivery applications. Adv Drug Deliv Rev, 2008. 60(11): p. 1307-15. 134. Kuroda, S., Kondo, H., Ohya, K. and Kasugai, S., A new technique with calcium phosphate precipitate enhances efficiency of in vivo plasmid DNA gene transfer. J Pharmacol Sci, 2005. 97(2): p. 227-33. References 63

135. Batard, P., Jordan, M. and Wurm, F., Transfer of high copy number plasmid into mammalian cells by calcium phosphate transfection. Gene, 2001. 270(1-2): p. 61-8. 136. Bhakta, G., Mitra, S. and Maitra, A., DNA encapsulated magnesium and manganous phosphate nanoparticles: potential non-viral vectors for gene delivery. Biomaterials, 2005. 26(14): p. 2157-63. 137. Pfeifer, A. and Verma, I. M., Gene therapy: promises and problems. Annu Rev Genomics Hum Genet, 2001. 2: p. 177-211. 138. Coffin, J. M., Hughes, S. H. and Varmus, H. E., The Interactions of Retroviruses and their Hosts. 1997. 139. Osten, P., Grinevich, V. and Cetin, A., Viral vectors: a wide range of choices and high levels of service. Handb Exp Pharmacol, 2007(178): p. 177-202. 140. Zhang, X. and Godbey, W. T., Viral vectors for gene delivery in tissue engineering. Adv Drug Deliv Rev, 2006. 58(4): p. 515-34. 141. Telesnitsky, A. and Goff, S. P., Reverse Transcriptase and the Generation of Retroviral DNA. 1997. 142. Brown, P. O., Integration. In: Coffin JM, Hughes SH, Varmus HE, eds. Retroviruses., 1997: p. 161-203. 143. Buchschacher Jr, G. L., Safety Considerations Associated with Development and Clinical Application of Lentiviral Vector Systems for Gene Transfer. Current Genomics, 2004. 5(1): p. 19-35. 144. Verma, I. M. and Weitzman, M. D., Gene therapy: twenty-first century medicine. Annu Rev Biochem, 2005. 74: p. 711-38. 145. Blaese, R. M., Culver, K. W., Miller, A. D., Carter, C. S., Fleisher, T., Clerici, M., Shearer, G., Chang, L., Chiang, Y., Tolstoshev, P., Greenblatt, J. J., Rosenberg, S. A., Klein, H., Berger, M., Mullen, C. A., Ramsey, W. J., Muul, L., Morgan, R. A. and Anderson, W. F., T lymphocyte-directed gene therapy for ADA- SCID: initial trial results after 4 years. Science, 1995. 270(5235): p. 475-80. 146. Coffin, J.M., Retroviridae: the viruses and their replication. in: Field, B.N., Knipe, D.M., Howley, P.M., Chanock, R.M., Melnick, J.L., Monath, T.P., Roizman, B., and Straus, S. E. (eds.), Fields Virology, 1996: p. 1767-1847. 147. Miller, A. D., Development and Applications of Retroviral Vectors. In: Coffin JM, Hughes SH, Varmus HE, (eds) Retroviruses, 1997. 148. Walther, W. and Stein, U., Viral vectors for gene transfer: a review of their use in the treatment of human diseases. Drugs, 2000. 60(2): p. 249-71. 149. Hirsch, M.S., Curran, J., Human immunodeficiency viruses. in: Field, B.N., Knipe, D.M., Howley, P.M., Chanock, R.M., Melnick, J.L., Monath, T.P., Roizman, B., and Straus, S. E. (eds.), Fields Virology, 1996: p. 1953-75. 150. Joag, S.V., Stephens, E.B., Narayan, O., Lentiviruses. in: Field, B.N., Knipe, D.M., Howley, P.M., Chanock, R.M., Melnick, J.L., Monath, T.P., Roizman, B., and Straus, S. E. (eds.), Fields Virology, 1996: p. 1977-96. 151. Malim, M. H., Hauber, J., Le, S. Y., Maizel, J. V. and Cullen, B. R., The HIV-1 rev trans-activator acts through a structured target sequence to activate nuclear export of unspliced viral mRNA. Nature, 1989. 338(6212): p. 254-7. 152. Heinzinger, N. K., Bukinsky, M. I., Haggerty, S. A., Ragland, A. M., Kewalramani, V., Lee, M. A., Gendelman, H. E., Ratner, L., Stevenson, M. and Emerman, M., The Vpr protein of human immunodeficiency virus type 1 influences nuclear localization of viral nucleic acids in nondividing host cells. Proc Natl Acad Sci U S A, 1994. 91(15): p. 7311-5. 153. Poeschla, E., Gilbert, J., Li, X., Huang, S., Ho, A. and Wong-Staal, F., Identification of a human immunodeficiency virus type 2 (HIV-2) encapsidation determinant and transduction of nondividing human cells by HIV-2-based lentivirus vectors. J Virol, 1998. 72(8): p. 6527-36. 154. Poeschla, E. M., Wong-Staal, F. and Looney, D. J., Efficient transduction of nondividing human cells by feline immunodeficiency virus lentiviral vectors. Nat Med, 1998. 4(3): p. 354-7. References 64

155. Olsen, J. C., Gene transfer vectors derived from equine infectious anemia virus. Gene Ther, 1998. 5(11): p. 1481-7. 156. Mangeot, P. E., Negre, D., Dubois, B., Winter, A. J., Leissner, P., Mehtali, M., Kaiserlian, D., Cosset, F. L. and Darlix, J. L., Development of minimal lentivirus vectors derived from simian immunodeficiency virus (SIVmac251) and their use for gene transfer into human dendritic cells. J Virol, 2000. 74(18): p. 8307-15. 157. Berkowitz, R. D., Ilves, H., Plavec, I. and Veres, G., Gene transfer systems derived from Visna virus: analysis of virus production and infectivity. Virology, 2001. 279(1): p. 116-29. 158. Naldini, L., Blomer, U., Gallay, P., Ory, D., Mulligan, R., Gage, F. H., Verma, I. M. and Trono, D., In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science, 1996. 272(5259): p. 263-7. 159. Dull, T., Zufferey, R., Kelly, M., Mandel, R. J., Nguyen, M., Trono, D. and Naldini, L., A third-generation lentivirus vector with a conditional packaging system. J Virol, 1998. 72(11): p. 8463-71. 160. Zufferey, R., Dull, T., Mandel, R. J., Bukovsky, A., Quiroz, D., Naldini, L. and Trono, D., Self-inactivating lentivirus vector for safe and efficient in vivo gene delivery. J Virol, 1998. 72(12): p. 9873-80. 161. Wiznerowicz, M. and Trono, D., Harnessing HIV for therapy, basic research and biotechnology. Trends in Biotechnology, 2005. 23(1): p. 42-47. 162. Naldini, L., Blomer, U., Gage, F. H., Trono, D. and Verma, I. M., Efficient transfer, integration, and sustained long-term expression of the transgene in adult rat brains injected with a lentiviral vector. Proc Natl Acad Sci U S A, 1996. 93(21): p. 11382-8. 163. Reiser, J., Harmison, G., Kluepfel-Stahl, S., Brady, R. O., Karlsson, S. and Schubert, M., Transduction of nondividing cells using pseudotyped defective high- titer HIV type 1 particles. Proc Natl Acad Sci U S A, 1996. 93(26): p. 15266-71. 164. Lois, C., Hong, E. J., Pease, S., Brown, E. J. and Baltimore, D., Germline transmission and tissue-specific expression of transgenes delivered by lentiviral vectors. Science, 2002. 295(5556): p. 868-72. 165. Warnock, J. N., Merten, O. W. and Al-Rubeai, M., Cell culture processes for the production of viral vectors for gene therapy purposes. Cytotechnology, 2006. 50(1- 3): p. 141-62. 166. Buchschacher, G. L., Jr. and Wong-Staal, F., Development of lentiviral vectors for gene therapy for human diseases. Blood, 2000. 95(8): p. 2499-504. 167. Follenzi, A., Santambrogio, L. and Annoni, A., Immune responses to lentiviral vectors. Curr Gene Ther, 2007. 7(5): p. 306-15. 168. Themis, M., Waddington, S. N., Schmidt, M., von Kalle, C., Wang, Y., Al-Allaf, F., Gregory, L. G., Nivsarkar, M., Holder, M. V., Buckley, S. M., Dighe, N., Ruthe, A. T., Mistry, A., Bigger, B., Rahim, A., Nguyen, T. H., Trono, D., Thrasher, A. J. and Coutelle, C., Oncogenesis following delivery of a nonprimate lentiviral gene therapy vector to fetal and neonatal mice. Mol Ther, 2005. 12(4): p. 763-71. 169. Williams, D. A., ESCGT 2008: progress in clinical gene therapy. Mol Ther, 2009. 17(1): p. 1-2. 170. Bank, A., Dorazio, R. and Leboulch, P., A phase I/II clinical trial of beta-globin gene therapy for beta-thalassemia. Ann N Y Acad Sci, 2005. 1054: p. 308-16. 171. Li, M. J., Kim, J., Li, S., Zaia, J., Yee, J. K., Anderson, J., Akkina, R. and Rossi, J. J., Long-term inhibition of HIV-1 infection in primary hematopoietic cells by lentiviral vector delivery of a triple combination of anti-HIV shRNA, anti-CCR5 ribozyme, and a nucleolar-localizing TAR decoy. Mol Ther, 2005. 12(5): p. 900-9. 172. Chhabra, A., Yang, L., Wang, P., Comin-Anduix, B., Das, R., Chakraborty, N. G., Ray, S., Mehrotra, S., Yang, H., Hardee, C. L., Hollis, R., Dorsky, D. I., Koya, R., Kohn, D. B., Ribas, A., Economou, J. S., Baltimore, D. and Mukherji, B., CD4+CD25- T cells transduced to express MHC class I-restricted epitope-specific TCR synthesize Th1 cytokines and exhibit MHC class I-restricted cytolytic effector function in a human melanoma model. J Immunol, 2008. 181(2): p. 1063-70. References 65

173. Rowe, W. P., Huebner, R. J., Gilmore, L. K., Parrott, R. H. and Ward, T. G., Isolation of a cytopathogenic agent from human adenoids undergoing spontaneous degeneration in tissue culture. Proc Soc Exp Biol Med, 1953. 84(3): p. 570-3. 174. Shenk, T., Adenoviruses: the viruses and their replication. in: Field, B.N., Knipe, D.M., Howley, P.M., Chanock, R.M., Melnick, J.L., Monath, T.P., Roizman, B., and Straus, S. E. (eds.), Fields Virology, 1996: p. 2111-48. 175. Shenk, T., Adenoviruses: the viruses and their replication. in: Field, B.N., Knipe, D.M., Howley, P.M., Chanock, R.M., Melnick, J.L., Monath, T.P., Roizman, B., and Straus, S. E. (eds.), Fields Virology, 2001: p. 2265-326. 176. Horwitz, M.S., Adenoviruses. in: Field, B.N., Knipe, D.M., Howley, P.M., Chanock, R.M., Melnick, J.L., Monath, T.P., Roizman, B., and Straus, S. E. (eds.), Fields Virology, 1996: p. 2149-71. 177. Douglas, J. T., Adenoviral vectors for gene therapy. Mol Biotechnol, 2007. 36(1): p. 71-80. 178. Rux, J.J., Burnett, R. , Adenovirus Capsid Proteins. In: Seth, P. (eds). Adenoviruses: Basic Biology to Gene Therapy, 1999: p. 5-15. 179. Boyer, J., Ketner, G., Adenovirus Late Gene Expression. In: Seth, P. (eds). Adenoviruses: Basic Biology to Gene Therapy, 1999: p. 69-77. 180. Bergelson, J. M., Cunningham, J. A., Droguett, G., Kurt-Jones, E. A., Krithivas, A., Hong, J. S., Horwitz, M. S., Crowell, R. L. and Finberg, R. W., Isolation of a common receptor for Coxsackie B viruses and adenoviruses 2 and 5. Science, 1997. 275(5304): p. 1320-3. 181. Wickham, T. J., Mathias, P., Cheresh, D. A. and Nemerow, G. R., Integrins alpha v beta 3 and alpha v beta 5 promote adenovirus internalization but not virus attachment. Cell, 1993. 73(2): p. 309-19. 182. Kovesdi, I., Brough, D. E., Bruder, J. T. and Wickham, T. J., Adenoviral vectors for gene transfer. Curr Opin Biotechnol, 1997. 8(5): p. 583-9. 183. Berkner, K. L., Development of adenovirus vectors for the expression of heterologous genes. Biotechniques, 1988. 6(7): p. 616-29. 184. Kochanek, S., Schiedner, G. and Volpers, C., High-capacity 'gutless' adenoviral vectors. Curr Opin Mol Ther, 2001. 3(5): p. 454-63. 185. Leiden, J. M., Human gene therapy: the good, the bad, and the ugly. Circ Res, 2000. 86(9): p. 923-5. 186. Mah, C., Byrne, B. J. and Flotte, T. R., Virus-based gene delivery systems. Clin Pharmacokinet, 2002. 41(12): p. 901-11. 187. Yei, S., Mittereder, N., Tang, K., O'Sullivan, C. and Trapnell, B. C., Adenovirus- mediated gene transfer for cystic fibrosis: quantitative evaluation of repeated in vivo vector administration to the lung. Gene Ther, 1994. 1(3): p. 192-200. 188. Roth, J. A., Modification of tumor suppressor gene expression and induction of apoptosis in non-small cell lung cancer (NSCLC) with an adenovirus vector expressing wildtype p53 and cisplatin. Hum Gene Ther, 1996. 7(8): p. 1013-30. 189. Batshaw, M. L., Wilson, J. M., Raper, S., Yudkoff, M. and Robinson, M. B., Recombinant adenovirus gene transfer in adults with partial ornithine transcarbamylase deficiency (OTCD). Hum Gene Ther, 1999. 10(14): p. 2419-37. 190. Suckau, L., Fechner, H., Chemaly, E., Krohn, S., Hadri, L., Kockskamper, J., Westermann, D., Bisping, E., Ly, H., Wang, X., Kawase, Y., Chen, J., Liang, L., Sipo, I., Vetter, R., Weger, S., Kurreck, J., Erdmann, V., Tschope, C., Pieske, B., Lebeche, D., Schultheiss, H. P., Hajjar, R. J. and Poller, W. C., Long-term cardiac- targeted RNA interference for the treatment of heart failure restores cardiac function and reduces pathological hypertrophy. Circulation, 2009. 119(9): p. 1241- 52. 191. Happel, K. I., Lockhart, E. A., Mason, C. M., Porretta, E., Keoshkerian, E., Odden, A. R., Nelson, S. and Ramsay, A. J., Pulmonary interleukin-23 gene delivery increases local T-cell immunity and controls growth of Mycobacterium tuberculosis in the lungs. Infect Immun, 2005. 73(9): p. 5782-8. References 66

192. Shiver, J. W., Fu, T. M., Chen, L., Casimiro, D. R., Davies, M. E., Evans, R. K., Zhang, Z. Q., Simon, A. J., Trigona, W. L., Dubey, S. A., Huang, L., Harris, V. A., Long, R. S., Liang, X., Handt, L., Schleif, W. A., Zhu, L., Freed, D. C., Persaud, N. V., Guan, L., Punt, K. S., Tang, A., Chen, M., Wilson, K. A., Collins, K. B., Heidecker, G. J., Fernandez, V. R., Perry, H. C., Joyce, J. G., Grimm, K. M., Cook, J. C., Keller, P. M., Kresock, D. S., Mach, H., Troutman, R. D., Isopi, L. A., Williams, D. M., Xu, Z., Bohannon, K. E., Volkin, D. B., Montefiori, D. C., Miura, A., Krivulka, G. R., Lifton, M. A., Kuroda, M. J., Schmitz, J. E., Letvin, N. L., Caulfield, M. J., Bett, A. J., Youil, R., Kaslow, D. C. and Emini, E. A., Replication-incompetent adenoviral vaccine vector elicits effective anti-immunodeficiency-virus immunity. Nature, 2002. 415(6869): p. 331-5. 193. Ferber, S., Halkin, A., Cohen, H., Ber, I., Einav, Y., Goldberg, I., Barshack, I., Seijffers, R., Kopolovic, J., Kaiser, N. and Karasik, A., Pancreatic and duodenal homeobox gene 1 induces expression of insulin genes in liver and ameliorates streptozotocin-induced hyperglycemia. Nat Med, 2000. 6(5): p. 568-72. 194. Okuda, T., Tagawa, K., Qi, M. L., Hoshio, M., Ueda, H., Kawano, H., Kanazawa, I., Muramatsu, M. and Okazawa, H., Oct-3/4 repression accelerates differentiation of neural progenitor cells in vitro and in vivo. Brain Res Mol Brain Res, 2004. 132(1): p. 18-30. 195. Goncalves, M. A., Adeno-associated virus: from defective virus to effective vector. Virol J, 2005. 2: p. 43. 196. Yalkinoglu, A. O., Heilbronn, R., Burkle, A., Schlehofer, J. R. and zur Hausen, H., DNA amplification of adeno-associated virus as a response to cellular genotoxic stress. Cancer Res, 1988. 48(11): p. 3123-9. 197. Yakobson, B., Koch, T. and Winocour, E., Replication of adeno-associated virus in synchronized cells without the addition of a helper virus. J Virol, 1987. 61(4): p. 972-81. 198. Mori, S., Wang, L., Takeuchi, T. and Kanda, T., Two novel adeno-associated viruses from cynomolgus monkey: pseudotyping characterization of capsid protein. Virology, 2004. 330(2): p. 375-83. 199. Schmidt, M., Grot, E., Cervenka, P., Wainer, S., Buck, C. and Chiorini, J. A., Identification and characterization of novel adeno-associated virus isolates in ATCC virus stocks. J Virol, 2006. 80(10): p. 5082-5. 200. Wu, Z., Asokan, A. and Samulski, R. J., Adeno-associated virus serotypes: vector toolkit for human gene therapy. Mol Ther, 2006. 14(3): p. 316-27. 201. Atchison, R. W., Casto, B. C. and Hammon, W. M., Adenovirus-Associated Defective Virus Particles. Science, 1965. 149(3685): p. 754-6. 202. Hoggan, M. D., Blacklow, N. R. and Rowe, W. P., Studies of small DNA viruses found in various adenovirus preparations: physical, biological, and immunological characteristics. Proc Natl Acad Sci U S A, 1966. 55(6): p. 1467-74. 203. Parks, W. P., Green, M., Pina, M. and Melnick, J. L., Physicochemical characterization of adeno-associated satellite virus type 4 and its nucleic acid. J Virol, 1967. 1(5): p. 980-7. 204. Bantel-Schaal, U. and zur Hausen, H., Characterization of the DNA of a defective human parvovirus isolated from a genital site. Virology, 1984. 134(1): p. 52-63. 205. Gao, G., Vandenberghe, L. H., Alvira, M. R., Lu, Y., Calcedo, R., Zhou, X. and Wilson, J. M., Clades of Adeno-associated viruses are widely disseminated in human tissues. J Virol, 2004. 78(12): p. 6381-8. 206. Jacobson, E. R., Kopit, W., Kennedy, F. A. and Funk, R. S., Coinfection of a bearded dragon, Pogona vitticeps, with adenovirus- and dependovirus-like viruses. Vet Pathol, 1996. 33(3): p. 343-6. 207. Farkas, S. L., Zadori, Z., Benko, M., Essbauer, S., Harrach, B. and Tijssen, P., A parvovirus isolated from royal python (Python regius) is a member of the genus Dependovirus. J Gen Virol, 2004. 85(Pt 3): p. 555-61. 208. Bossis, I. and Chiorini, J. A., Cloning of an avian adeno-associated virus (AAAV) and generation of recombinant AAAV particles. J Virol, 2003. 77(12): p. 6799-810. References 67

209. Olson, E. J., Haskell, S. R., Frank, R. K., Lehmkuhl, H. D., Hobbs, L. A., Warg, J. V., Landgraf, J. G. and Wunschmann, A., Isolation of an adenovirus and an adeno- associated virus from goat kids with enteritis. J Vet Diagn Invest, 2004. 16(5): p. 461-4. 210. Arbetman, A. E., Lochrie, M., Zhou, S., Wellman, J., Scallan, C., Doroudchi, M. M., Randlev, B., Patarroyo-White, S., Liu, T., Smith, P., Lehmkuhl, H., Hobbs, L. A., Pierce, G. F. and Colosi, P., Novel caprine adeno-associated virus (AAV) capsid (AAV-Go.1) is closely related to the primate AAV-5 and has unique tropism and neutralization properties. J Virol, 2005. 79(24): p. 15238-45. 211. Schmidt, M., Katano, H., Bossis, I. and Chiorini, J. A., Cloning and characterization of a bovine adeno-associated virus. J Virol, 2004. 78(12): p. 6509-16. 212. Dutta, S. K., Isolation and characterization of an adenovirus and isolation of its adenovirus-associated virus in cell culture from foals with respiratory tract disease. Am J Vet Res, 1975. 36(3): p. 247-50. 213. Myers, M. W. and Carter, B. J., Assembly of adeno-associated virus. Virology, 1980. 102(1): p. 71-82. 214. de la Maza, L. M. and Carter, B. J., Molecular structure of adeno-associated virus variant DNA. J Biol Chem, 1980. 255(7): p. 3194-203. 215. King, J. A., Dubielzig, R., Grimm, D. and Kleinschmidt, J. A., DNA helicase- mediated packaging of adeno-associated virus type 2 genomes into preformed capsids. EMBO J, 2001. 20(12): p. 3282-91. 216. Steinbach, S., Wistuba, A., Bock, T. and Kleinschmidt, J. A., Assembly of adeno- associated virus type 2 capsids in vitro. J Gen Virol, 1997. 78 ( Pt 6): p. 1453-62. 217. Hermonat, P. L., Labow, M. A., Wright, R., Berns, K. I. and Muzyczka, N., Genetics of adeno-associated virus: isolation and preliminary characterization of adeno- associated virus type 2 mutants. J Virol, 1984. 51(2): p. 329-39. 218. Warrington, K. H., Jr., Gorbatyuk, O. S., Harrison, J. K., Opie, S. R., Zolotukhin, S. and Muzyczka, N., Adeno-associated virus type 2 VP2 capsid protein is nonessential and can tolerate large peptide insertions at its N terminus. J Virol, 2004. 78(12): p. 6595-609. 219. Summerford, C., Bartlett, J. S. and Samulski, R. J., AlphaVbeta5 integrin: a co- receptor for adeno-associated virus type 2 infection. Nat Med, 1999. 5(1): p. 78-82. 220. Kashiwakura, Y., Tamayose, K., Iwabuchi, K., Hirai, Y., Shimada, T., Matsumoto, K., Nakamura, T., Watanabe, M., Oshimi, K. and Daida, H., Hepatocyte growth factor receptor is a coreceptor for adeno-associated virus type 2 infection. J Virol, 2005. 79(1): p. 609-14. 221. Blackburn, S. D., Steadman, R. A. and Johnson, F. B., Attachment of adeno- associated virus type 3H to fibroblast growth factor receptor 1. Arch Virol, 2006. 151(3): p. 617-23. 222. Di Pasquale, G., Davidson, B. L., Stein, C. S., Martins, I., Scudiero, D., Monks, A. and Chiorini, J. A., Identification of PDGFR as a receptor for AAV-5 transduction. Nat Med, 2003. 9(10): p. 1306-12. 223. Walters, R. W., Yi, S. M., Keshavjee, S., Brown, K. E., Welsh, M. J., Chiorini, J. A. and Zabner, J., Binding of adeno-associated virus type 5 to 2,3-linked sialic acid is required for gene transfer. J Biol Chem, 2001. 276(23): p. 20610-6. 224. Kaludov, N., Brown, K. E., Walters, R. W., Zabner, J. and Chiorini, J. A., Adeno- associated virus serotype 4 (AAV4) and AAV5 both require sialic acid binding for hemagglutination and efficient transduction but differ in sialic acid linkage specificity. J Virol, 2001. 75(15): p. 6884-93. 225. Wu, Z., Miller, E., Agbandje-McKenna, M. and Samulski, R. J., Alpha2,3 and alpha2,6 N-linked sialic acids facilitate efficient binding and transduction by adeno- associated virus types 1 and 6. J Virol, 2006. 80(18): p. 9093-103. 226. Akache, B., Grimm, D., Pandey, K., Yant, S. R., Xu, H. and Kay, M. A., The 37/67- kilodalton laminin receptor is a receptor for adeno-associated virus serotypes 8, 2, 3, and 9. J Virol, 2006. 80(19): p. 9831-6. References 68

227. McCarty, D. M., Fu, H., Monahan, P. E., Toulson, C. E., Naik, P. and Samulski, R. J., Adeno-associated virus terminal repeat (TR) mutant generates self- complementary vectors to overcome the rate-limiting step to transduction in vivo. Gene Ther, 2003. 10(26): p. 2112-8. 228. Kyostio, S. R., Owens, R. A., Weitzman, M. D., Antoni, B. A., Chejanovsky, N. and Carter, B. J., Analysis of adeno-associated virus (AAV) wild-type and mutant Rep proteins for their abilities to negatively regulate AAV p5 and p19 mRNA levels. J Virol, 1994. 68(5): p. 2947-57. 229. Jay, F. T., Laughlin, C. A. and Carter, B. J., Eukaryotic translational control: adeno- associated virus protein synthesis is affected by a mutation in the adenovirus DNA- binding protein. Proc Natl Acad Sci U S A, 1981. 78(5): p. 2927-31. 230. Weitzman, M. D., Kyostio, S. R., Kotin, R. M. and Owens, R. A., Adeno-associated virus (AAV) Rep proteins mediate complex formation between AAV DNA and its integration site in human DNA. Proc Natl Acad Sci U S A, 1994. 91(13): p. 5808-12. 231. Cathomen, T., Stracker, T. H., Gilbert, L. B. and Weitzman, M. D., A genetic screen identifies a cellular regulator of adeno-associated virus. Proc Natl Acad Sci U S A, 2001. 98(26): p. 14991-6. 232. McCarty, D. M., Ryan, J. H., Zolotukhin, S., Zhou, X. and Muzyczka, N., Interaction of the adeno-associated virus Rep protein with a sequence within the A palindrome of the viral terminal repeat. J Virol, 1994. 68(8): p. 4998-5006. 233. Ryan, J. H., Zolotukhin, S. and Muzyczka, N., Sequence requirements for binding of Rep68 to the adeno-associated virus terminal repeats. J Virol, 1996. 70(3): p. 1542-53. 234. Nony, P., Tessier, J., Chadeuf, G., Ward, P., Giraud, A., Dugast, M., Linden, R. M., Moullier, P. and Salvetti, A., Novel cis-acting replication element in the adeno- associated virus type 2 genome is involved in amplification of integrated rep-cap sequences. J Virol, 2001. 75(20): p. 9991-4. 235. Weitzman, M. D., Young, S. M., Jr., Cathomen, T. and Samulski, R. J., Targeted integration by adeno-associated virus. Methods Mol Med, 2003. 76: p. 201-19. 236. Ding, W., Zhang, L., Yan, Z. and Engelhardt, J. F., Intracellular trafficking of adeno- associated viral vectors. Gene Ther, 2005. 12(11): p. 873-80. 237. Yan, Z., Duan, D., Engelhardt, J.F., Mechanism of recombinant adeno-associated virus transduction. In: Kerr, J.R., Cotmore, S.F., Bloom, M.E., Linden, R.M., Parrisch, C.R. (eds) Parvoviruses, 2006: p. 511-524. 238. Seisenberger, G., Ried, M. U., Endress, T., Buning, H., Hallek, M. and Brauchle, C., Real-time single-molecule imaging of the infection pathway of an adeno-associated virus. Science, 2001. 294(5548): p. 1929-32. 239. Cheung, A. K., Hoggan, M. D., Hauswirth, W. W. and Berns, K. I., Integration of the adeno-associated virus genome into cellular DNA in latently infected human Detroit 6 cells. J Virol, 1980. 33(2): p. 739-48. 240. McCarty, D. M., Young, S. M., Jr. and Samulski, R. J., Integration of adeno- associated virus (AAV) and recombinant AAV vectors. Annu Rev Genet, 2004. 38: p. 819-45. 241. Cathomen, T., rAAV vectors for gene targeting. In: Kerr, J.R., Cotmore, S.F., Bloom, M.E., Linden, R.M., Parrisch, C.R. (eds) Parvoviruses, 2006: p. 543-49. 242. Samulski, R. J., Chang, L. S. and Shenk, T., Helper-free stocks of recombinant adeno-associated viruses: normal integration does not require viral gene expression. J Virol, 1989. 63(9): p. 3822-8. 243. Xiao, X., Li, J. and Samulski, R. J., Production of high-titer recombinant adeno- associated virus vectors in the absence of helper adenovirus. J Virol, 1998. 72(3): p. 2224-32. 244. Grimm, D., Kay, M. A. and Kleinschmidt, J. A., Helper virus-free, optically controllable, and two-plasmid-based production of adeno-associated virus vectors of serotypes 1 to 6. Mol Ther, 2003. 7(6): p. 839-50. References 69

245. Xiao, X., Li, J. and Samulski, R. J., Efficient long-term gene transfer into muscle tissue of immunocompetent mice by adeno-associated virus vector. J Virol, 1996. 70(11): p. 8098-108. 246. Zolotukhin, S., Byrne, B. J., Mason, E., Zolotukhin, I., Potter, M., Chesnut, K., Summerford, C., Samulski, R. J. and Muzyczka, N., Recombinant adeno- associated virus purification using novel methods improves infectious titer and yield. Gene Ther, 1999. 6(6): p. 973-85. 247. Summerford, C. and Samulski, R. J., Membrane-associated heparan sulfate proteoglycan is a receptor for adeno-associated virus type 2 virions. J Virol, 1998. 72(2): p. 1438-45. 248. Clark, K. R., Liu, X., McGrath, J. P. and Johnson, P. R., Highly purified recombinant adeno-associated virus vectors are biologically active and free of detectable helper and wild-type viruses. Hum Gene Ther, 1999. 10(6): p. 1031-9. 249. Paterna, J. C. and Bueler, H., Recombinant adeno-associated virus vector design and gene expression in the mammalian brain. Methods, 2002. 28(2): p. 208-18. 250. Coura Rdos, S. and Nardi, N. B., The state of the art of adeno-associated virus- based vectors in gene therapy. Virol J, 2007. 4: p. 99. 251. Hoare, J., Waddington, S., Thomas, H. C., Coutelle, C. and McGarvey, M. J., Complement inhibition rescued mice allowing observation of transgene expression following intraportal delivery of baculovirus in mice. J Gene Med, 2005. 7(3): p. 325- 33. 252. Casper, D., Engstrom, S. J., Mirchandani, G. R., Pidel, A., Palencia, D., Cho, P. H., Brownlee, M., Edelstein, D., Federoff, H. J. and Sonstein, W. J., Enhanced vascularization and survival of neural transplants with ex vivo angiogenic gene transfer. Cell Transplant, 2002. 11(4): p. 331-49. 253. Gomez, C. E., Najera, J. L., Krupa, M. and Esteban, M., The poxvirus vectors MVA and NYVAC as gene delivery systems for vaccination against infectious diseases and cancer. Curr Gene Ther, 2008. 8(2): p. 97-120. 254. Johnston, K. M., Jacoby, D., Pechan, P. A., Fraefel, C., Borghesani, P., Schuback, D., Dunn, R. J., Smith, F. I. and Breakefield, X. O., HSV/AAV hybrid amplicon vectors extend transgene expression in human glioma cells. Hum Gene Ther, 1997. 8(3): p. 359-70. 255. Wang, Y., Camp, S. M., Niwano, M., Shen, X., Bakowska, J. C., Breakefield, X. O. and Allen, P. D., Herpes simplex virus type 1/adeno-associated virus rep(+) hybrid amplicon vector improves the stability of transgene expression in human cells by site-specific integration. J Virol, 2002. 76(14): p. 7150-62. 256. Recchia, A., Parks, R. J., Lamartina, S., Toniatti, C., Pieroni, L., Palombo, F., Ciliberto, G., Graham, F. L., Cortese, R., La Monica, N. and Colloca, S., Site- specific integration mediated by a hybrid adenovirus/adeno-associated virus vector. Proc Natl Acad Sci U S A, 1999. 96(6): p. 2615-20. 257. Curiel, D. T., Agarwal, S., Wagner, E. and Cotten, M., Adenovirus enhancement of transferrin-polylysine-mediated gene delivery. Proc Natl Acad Sci U S A, 1991. 88(19): p. 8850-4. 258. Schuster, M. J., Wu, G. Y., Walton, C. M. and Wu, C. H., Multicomponent DNA carrier with a vesicular stomatitis virus G-peptide greatly enhances liver-targeted gene expression in mice. Bioconjug Chem, 1999. 10(6): p. 1075-83. 259. Klug, A. and Schwabe, J. W., Protein motifs 5. Zinc fingers. FASEB J, 1995. 9(8): p. 597-604. 260. Krishna, S. S., Majumdar, I. and Grishin, N. V., Structural classification of zinc fingers: survey and summary. Nucleic Acids Res, 2003. 31(2): p. 532-50. 261. Pavletich, N. P. and Pabo, C. O., Zinc finger-DNA recognition: crystal structure of a Zif268-DNA complex at 2.1 A. Science, 1991. 252(5007): p. 809-17. 262. Segal, D. J., Beerli, R. R., Blancafort, P., Dreier, B., Effertz, K., Huber, A., Koksch, B., Lund, C. V., Magnenat, L., Valente, D. and Barbas, C. F., 3rd, Evaluation of a modular strategy for the construction of novel polydactyl zinc finger DNA-binding proteins. Biochemistry, 2003. 42(7): p. 2137-48. References 70

263. Maeder, M. L., Thibodeau-Beganny, S., Sander, J. D., Voytas, D. F. and Joung, J. K., Oligomerized pool engineering (OPEN): an 'open-source' protocol for making customized zinc-finger arrays. Nat Protoc, 2009. 4(10): p. 1471-501. 264. Sander, J. D., Dahlborg, E. J., Goodwin, M. J., Cade, L., Zhang, F., Cifuentes, D., Curtin, S. J., Blackburn, J. S., Thibodeau-Beganny, S., Qi, Y., Pierick, C. J., Hoffman, E., Maeder, M. L., Khayter, C., Reyon, D., Dobbs, D., Langenau, D. M., Stupar, R. M., Giraldez, A. J., Voytas, D. F., Peterson, R. T., Yeh, J. R. and Joung, J. K., Selection-free zinc-finger-nuclease engineering by context-dependent assembly (CoDA). Nat Methods, 2011. 8(1): p. 67-9. 265. Li, L., Wu, L. P. and Chandrasegaran, S., Functional domains in Fok I restriction endonuclease. Proc Natl Acad Sci U S A, 1992. 89(10): p. 4275-9. 266. Wah, D. A., Bitinaite, J., Schildkraut, I. and Aggarwal, A. K., Structure of FokI has implications for DNA cleavage. Proc Natl Acad Sci U S A, 1998. 95(18): p. 10564-9. 267. Kim, Y. G. and Chandrasegaran, S., Chimeric restriction endonuclease. Proc Natl Acad Sci U S A, 1994. 91(3): p. 883-7. 268. Bitinaite, J., Wah, D. A., Aggarwal, A. K. and Schildkraut, I., FokI dimerization is required for DNA cleavage. Proc Natl Acad Sci U S A, 1998. 95(18): p. 10570-5. 269. Smith, J., Bibikova, M., Whitby, F. G., Reddy, A. R., Chandrasegaran, S. and Carroll, D., Requirements for double-strand cleavage by chimeric restriction enzymes with zinc finger DNA-recognition domains. Nucleic Acids Res, 2000. 28(17): p. 3361-9. 270. Bibikova, M., Carroll, D., Segal, D. J., Trautman, J. K., Smith, J., Kim, Y. G. and Chandrasegaran, S., Stimulation of homologous recombination through targeted cleavage by chimeric nucleases. Mol Cell Biol, 2001. 21(1): p. 289-97. 271. Shukla, V. K., Doyon, Y., Miller, J. C., DeKelver, R. C., Moehle, E. A., Worden, S. E., Mitchell, J. C., Arnold, N. L., Gopalan, S., Meng, X., Choi, V. M., Rock, J. M., Wu, Y. Y., Katibah, G. E., Zhifang, G., McCaskill, D., Simpson, M. A., Blakeslee, B., Greenwalt, S. A., Butler, H. J., Hinkley, S. J., Zhang, L., Rebar, E. J., Gregory, P. D. and Urnov, F. D., Precise genome modification in the crop species Zea mays using zinc-finger nucleases. Nature, 2009. 459(7245): p. 437-41. 272. Marton, I., Zuker, A., Shklarman, E., Zeevi, V., Tovkach, A., Roffe, S., Ovadis, M., Tzfira, T. and Vainstein, A., Nontransgenic genome modification in plant cells. Plant Physiol, 2010. 154(3): p. 1079-87. 273. Cai, C. Q., Doyon, Y., Ainley, W. M., Miller, J. C., Dekelver, R. C., Moehle, E. A., Rock, J. M., Lee, Y. L., Garrison, R., Schulenberg, L., Blue, R., Worden, A., Baker, L., Faraji, F., Zhang, L., Holmes, M. C., Rebar, E. J., Collingwood, T. N., Rubin- Wilson, B., Gregory, P. D., Urnov, F. D. and Petolino, J. F., Targeted transgene integration in plant cells using designed zinc finger nucleases. Plant Mol Biol, 2009. 69(6): p. 699-709. 274. Beumer, K., Bhattacharyya, G., Bibikova, M., Trautman, J. K. and Carroll, D., Efficient gene targeting in Drosophila with zinc-finger nucleases. Genetics, 2006. 172(4): p. 2391-403. 275. Morton, J., Davis, M. W., Jorgensen, E. M. and Carroll, D., Induction and repair of zinc-finger nuclease-targeted double-strand breaks in Caenorhabditis elegans somatic cells. Proc Natl Acad Sci U S A, 2006. 103(44): p. 16370-5. 276. Takasu, Y., Kobayashi, I., Beumer, K., Uchino, K., Sezutsu, H., Sajwan, S., Carroll, D., Tamura, T. and Zurovec, M., Targeted mutagenesis in the silkworm Bombyx mori using zinc finger nuclease mRNA injection. Insect Biochem Mol Biol, 2010. 40(10): p. 759-65. 277. Foley, J. E., Yeh, J. R., Maeder, M. L., Reyon, D., Sander, J. D., Peterson, R. T. and Joung, J. K., Rapid mutation of endogenous zebrafish genes using zinc finger nucleases made by Oligomerized Pool ENgineering (OPEN). PLoS One, 2009. 4(2): p. e4348. 278. Ochiai, H., Fujita, K., Suzuki, K., Nishikawa, M., Shibata, T., Sakamoto, N. and Yamamoto, T., Targeted mutagenesis in the sea urchin embryo using zinc-finger nucleases. Genes Cells, 2010. 15(8): p. 875-85. References 71

279. Young, J. J., Cherone, J. M., Doyon, Y., Ankoudinova, I., Faraji, F. M., Lee, A. H., Ngo, C., Guschin, D. Y., Paschon, D. E., Miller, J. C., Zhang, L., Rebar, E. J., Gregory, P. D., Urnov, F. D., Harland, R. M. and Zeitler, B., Efficient targeted gene disruption in the soma and germ line of the frog Xenopus tropicalis using engineered zinc-finger nucleases. Proc Natl Acad Sci U S A, 2011. 108(17): p. 7052-7. 280. Mashimo, T., Takizawa, A., Voigt, B., Yoshimi, K., Hiai, H., Kuramoto, T. and Serikawa, T., Generation of knockout rats with X-linked severe combined immunodeficiency (X-SCID) using zinc-finger nucleases. PLoS One, 2010. 5(1): p. e8870. 281. Cui, X., Ji, D., Fisher, D. A., Wu, Y., Briner, D. M. and Weinstein, E. J., Targeted integration in rat and mouse embryos with zinc-finger nucleases. Nat Biotechnol, 2011. 29(1): p. 64-7. 282. Liu, P. Q., Chan, E. M., Cost, G. J., Zhang, L., Wang, J., Miller, J. C., Guschin, D. Y., Reik, A., Holmes, M. C., Mott, J. E., Collingwood, T. N. and Gregory, P. D., Generation of a triple-gene knockout mammalian cell line using engineered zinc- finger nucleases. Biotechnol Bioeng, 2010. 106(1): p. 97-105. 283. Watanabe, M., Umeyama, K., Matsunari, H., Takayanagi, S., Haruyama, E., Nakano, K., Fujiwara, T., Ikezawa, Y., Nakauchi, H. and Nagashima, H., Knockout of exogenous EGFP gene in porcine somatic cells using zinc-finger nucleases. Biochem Biophys Res Commun, 2010. 402(1): p. 14-8. 284. Yuan, J., Wang, J., Crain, K., Fearns, C., Kim, K. A., Hua, K. L., Gregory, P. D., Holmes, M. C. and Torbett, B. E., Zinc-finger Nuclease Editing of Human cxcr4 Promotes HIV-1 CD4(+) T Cell Resistance and Enrichment. Mol Ther, 2012. 285. Soldner, F., Laganiere, J., Cheng, A. W., Hockemeyer, D., Gao, Q., Alagappan, R., Khurana, V., Golbe, L. I., Myers, R. H., Lindquist, S., Zhang, L., Guschin, D., Fong, L. K., Vu, B. J., Meng, X., Urnov, F. D., Rebar, E. J., Gregory, P. D., Zhang, H. S. and Jaenisch, R., Generation of isogenic pluripotent stem cells differing exclusively at two early onset Parkinson point mutations. Cell, 2011. 146(2): p. 318-31. 286. Perez, E. E., Wang, J., Miller, J. C., Jouvenot, Y., Kim, K. A., Liu, O., Wang, N., Lee, G., Bartsevich, V. V., Lee, Y. L., Guschin, D. Y., Rupniewski, I., Waite, A. J., Carpenito, C., Carroll, R. G., Orange, J. S., Urnov, F. D., Rebar, E. J., Ando, D., Gregory, P. D., Riley, J. L., Holmes, M. C. and June, C. H., Establishment of HIV-1 resistance in CD4+ T cells by genome editing using zinc-finger nucleases. Nat Biotechnol, 2008. 26(7): p. 808-16. 287. Cornu, T. I., Thibodeau-Beganny, S., Guhl, E., Alwin, S., Eichtinger, M., Joung, J. K. and Cathomen, T., DNA-binding specificity is a major determinant of the activity and toxicity of zinc-finger nucleases. Mol Ther, 2008. 16(2): p. 352-8. 288. Xu, Q., Chou, B., Fitzsimmons, B., Miyanohara, A., Shubayev, V., Santucci, C., Hefferan, M., Marsala, M. and Hua, X. Y., In vivo gene knockdown in rat dorsal root Ganglia mediated by self-complementary adeno-associated virus serotype 5 following intrathecal delivery. PLoS One, 2012. 7(3): p. e32581. 289. Kacherovsky, N., Harkey, M. A., Blau, C. A., Giachelli, C. M. and Pun, S. H., Combination of Sleeping Beauty transposition and chemically induced dimerization selection for robust production of engineered cells. Nucleic Acids Res, 2012. 290. Kuan, J. Y. and Glazer, P. M., Targeted gene modification using triplex-forming oligonucleotides. Methods Mol Biol, 2004. 262: p. 173-94. 291. Komiyama, M., Aiba, Y., Yamamoto, Y. and Sumaoka, J., Artificial restriction DNA cutter for site-selective scission of double-stranded DNA with tunable scission site and specificity. Nat Protoc, 2008. 3(4): p. 655-62. 292. Wurtz, N. R. and Dervan, P. B., Sequence specific alkylation of DNA by hairpin pyrrole-imidazole polyamide conjugates. Chem Biol, 2000. 7(3): p. 153-61. 293. Dervan, P. B. and Edelson, B. S., Recognition of the DNA minor groove by pyrrole- imidazole polyamides. Curr Opin Struct Biol, 2003. 13(3): p. 284-99. 294. Kaihatsu, K., Janowski, B. A. and Corey, D. R., Recognition of chromosomal DNA by PNAs. Chem Biol, 2004. 11(6): p. 749-58. References 72

295. Thermes, V., Grabher, C., Ristoratore, F., Bourrat, F., Choulika, A., Wittbrodt, J. and Joly, J. S., I-SceI meganuclease mediates highly efficient transgenesis in fish. Mech Dev, 2002. 118(1-2): p. 91-8. 296. Porteus, M. H. and Carroll, D., Gene targeting using zinc finger nucleases. Nat Biotechnol, 2005. 23(8): p. 967-73. 297. Schambach, A., Cantz, T., Baum, C. and Cathomen, T., Generation and genetic modification of induced pluripotent stem cells. Expert Opin Biol Ther, 2010. 10(7): p. 1089-103. 298. Hockemeyer, D., Wang, H., Kiani, S., Lai, C. S., Gao, Q., Cassady, J. P., Cost, G. J., Zhang, L., Santiago, Y., Miller, J. C., Zeitler, B., Cherone, J. M., Meng, X., Hinkley, S. J., Rebar, E. J., Gregory, P. D., Urnov, F. D. and Jaenisch, R., Genetic engineering of human pluripotent cells using TALE nucleases. Nat Biotechnol, 2011. 29(8): p. 731-4. 299. Szczepek, M., Brondani, V., Buchel, J., Serrano, L., Segal, D. J. and Cathomen, T., Structure-based redesign of the dimerization interface reduces the toxicity of zinc- finger nucleases. Nat Biotechnol, 2007. 25(7): p. 786-93. 300. Alwin, S., Gere, M. B., Guhl, E., Effertz, K., Barbas, C. F., 3rd, Segal, D. J., Weitzman, M. D. and Cathomen, T., Custom zinc-finger nucleases for use in human cells. Mol Ther, 2005. 12(4): p. 610-7. 301. Cathomen, T. and Joung, J. K., Zinc-finger nucleases: the next generation emerges. Mol Ther, 2008. 16(7): p. 1200-7. 302. Gabriel, R., Lombardo, A., Arens, A., Miller, J. C., Genovese, P., Kaeppel, C., Nowrouzi, A., Bartholomae, C. C., Wang, J., Friedman, G., Holmes, M. C., Gregory, P. D., Glimm, H., Schmidt, M., Naldini, L. and von Kalle, C., An unbiased genome- wide analysis of zinc-finger nuclease specificity. Nat Biotechnol, 2011. 29(9): p. 816-23. 303. Benabdallah, B. F., Allard, E., Yao, S., Friedman, G., Gregory, P. D., Eliopoulos, N., Fradette, J., Spees, J. L., Haddad, E., Holmes, M. C. and Beausejour, C. M., Targeted gene addition to human mesenchymal stromal cells as a cell-based plasma-soluble protein delivery platform. Cytotherapy, 2010. 12(3): p. 394-9. 304. Morgan, J. R., Barrandon, Y., Green, H. and Mulligan, R. C., Expression of an exogenous growth hormone gene by transplantable human epidermal cells. Science, 1987. 237(4821): p. 1476-9. 305. Handel, E. M. and Cathomen, T., Zinc-finger nuclease based genome surgery: it's all about specificity. Curr Gene Ther, 2011. 11(1): p. 28-37. 306. Meng, X., Noyes, M. B., Zhu, L. J., Lawson, N. D. and Wolfe, S. A., Targeted gene inactivation in zebrafish using engineered zinc-finger nucleases. Nat Biotechnol, 2008. 26(6): p. 695-701. 307. Doyon, Y., McCammon, J. M., Miller, J. C., Faraji, F., Ngo, C., Katibah, G. E., Amora, R., Hocking, T. D., Zhang, L., Rebar, E. J., Gregory, P. D., Urnov, F. D. and Amacher, S. L., Heritable targeted gene disruption in zebrafish using designed zinc- finger nucleases. Nat Biotechnol, 2008. 26(6): p. 702-8. 308. Pruett-Miller, S. M., Reading, D. W., Porter, S. N. and Porteus, M. H., Attenuation of zinc finger nuclease toxicity by small-molecule regulation of protein levels. PLoS Genet, 2009. 5(2): p. e1000376. 309. Selvaraj, V., Plane, J. M., Williams, A. J. and Deng, W., Switching cell fate: the remarkable rise of induced pluripotent stem cells and lineage reprogramming technologies. Trends Biotechnol, 2010. 28(4): p. 214-23. 310. Marion, R. M., Strati, K., Li, H., Murga, M., Blanco, R., Ortega, S., Fernandez- Capetillo, O., Serrano, M. and Blasco, M. A., A p53-mediated DNA damage response limits reprogramming to ensure iPS cell genomic integrity. Nature, 2009. 460(7259): p. 1149-53. 311. Mitsui, K., Suzuki, K., Aizawa, E., Kawase, E., Suemori, H., Nakatsuji, N. and Mitani, K., Gene targeting in human pluripotent stem cells with adeno-associated virus vectors. Biochem Biophys Res Commun, 2009. 388(4): p. 711-7. References 73

312. Guidance for Industry: Gene Therapy Clinical Trials - Observing Subjects for Delayed Adverse Events Availble at:. Availble at: http://www.fda.gov/BiologicsBloodVaccines/GuidanceComplianceRegulatoryInform ation/Guidances/CellularandGeneTherapy/ucm072957.htm, Accessed on 24/02/2012. 313. Miller, D. G., Petek, L. M. and Russell, D. W., Adeno-associated virus vectors integrate at chromosome breakage sites. Nat Genet, 2004. 36(7): p. 767-73. 314. Cornu, T. I. and Cathomen, T., Targeted genome modifications using integrase- deficient lentiviral vectors. Mol Ther, 2007. 15(12): p. 2107-13. 315. Gellhaus, K., Cornu, T. I., Heilbronn, R. and Cathomen, T., Fate of recombinant adeno-associated viral vector genomes during DNA double-strand break-induced gene targeting in human cells. Hum Gene Ther, 2010. 21(5): p. 543-53. 316. Petek, L. M., Russell, D. W. and Miller, D. G., Frequent endonuclease cleavage at off-target locations in vivo. Mol Ther, 2010. 18(5): p. 983-6. 317. Khan, I. F., Hirata, R. K., Wang, P. R., Li, Y., Kho, J., Nelson, A., Huo, Y., Zavaljevski, M., Ware, C. and Russell, D. W., Engineering of human pluripotent stem cells by AAV-mediated gene targeting. Mol Ther, 2010. 18(6): p. 1192-9. 318. Khan, I. F., Hirata, R. K. and Russell, D. W., AAV-mediated gene targeting methods for human cells. Nat Protoc, 2011. 6(4): p. 482-501. 319. Miller, D. G., Wang, P. R., Petek, L. M., Hirata, R. K., Sands, M. S. and Russell, D. W., Gene targeting in vivo by adeno-associated virus vectors. Nat Biotechnol, 2006. 24(8): p. 1022-6. 320. Donsante, A., Miller, D. G., Li, Y., Vogler, C., Brunt, E. M., Russell, D. W. and Sands, M. S., AAV vector integration sites in mouse hepatocellular carcinoma. Science, 2007. 317(5837): p. 477. 321. Hirata, R. K. and Russell, D. W., Design and packaging of adeno-associated virus gene targeting vectors. J Virol, 2000. 74(10): p. 4612-20. 322. Moehle, E. A., Rock, J. M., Lee, Y. L., Jouvenot, Y., DeKelver, R. C., Gregory, P. D., Urnov, F. D. and Holmes, M. C., Targeted gene addition into a specified location in the human genome using designed zinc finger nucleases. Proc Natl Acad Sci U S A, 2007. 104(9): p. 3055-60. 323. Chamberlain, J. R., Schwarze, U., Wang, P. R., Hirata, R. K., Hankenson, K. D., Pace, J. M., Underwood, R. A., Song, K. M., Sussman, M., Byers, P. H. and Russell, D. W., Gene targeting in stem cells from individuals with osteogenesis imperfecta. Science, 2004. 303(5661): p. 1198-201. 324. Hirsch, M. L., Green, L., Porteus, M. H. and Samulski, R. J., Self-complementary AAV mediates gene targeting and enhances endonuclease delivery for double- strand break repair. Gene Ther, 2010. 17(9): p. 1175-80. 325. Miller, D. G., Petek, L. M. and Russell, D. W., Human gene targeting by adeno- associated virus vectors is enhanced by DNA double-strand breaks. Mol Cell Biol, 2003. 23(10): p. 3550-7. 326. Porteus, M. H., Cathomen, T., Weitzman, M. D. and Baltimore, D., Efficient gene targeting mediated by adeno-associated virus and DNA double-strand breaks. Mol Cell Biol, 2003. 23(10): p. 3558-65. 327. Lombardo, A., Genovese, P., Beausejour, C. M., Colleoni, S., Lee, Y. L., Kim, K. A., Ando, D., Urnov, F. D., Galli, C., Gregory, P. D., Holmes, M. C. and Naldini, L., Gene editing in human stem cells using zinc finger nucleases and integrase- defective lentiviral vector delivery. Nat Biotechnol, 2007. 25(11): p. 1298-306. 328. Wang, Z., Ma, H. I., Li, J., Sun, L., Zhang, J. and Xiao, X., Rapid and highly efficient transduction by double-stranded adeno-associated virus vectors in vitro and in vivo. Gene Ther, 2003. 10(26): p. 2105-11. 329. Bohne, J. and Cathomen, T., Genotoxicity in gene therapy: an account of vector integration and designer nucleases. Curr Opin Mol Ther, 2008. 10(3): p. 214-23. 330. Porteus, M. H. and Baltimore, D., Chimeric nucleases stimulate gene targeting in human cells. Science, 2003. 300(5620): p. 763. References 74

331. Bibikova, M., Golic, M., Golic, K. G. and Carroll, D., Targeted chromosomal cleavage and mutagenesis in Drosophila using zinc-finger nucleases. Genetics, 2002. 161(3): p. 1169-75. 332. Miller, J. C., Holmes, M. C., Wang, J., Guschin, D. Y., Lee, Y. L., Rupniewski, I., Beausejour, C. M., Waite, A. J., Wang, N. S., Kim, K. A., Gregory, P. D., Pabo, C. O. and Rebar, E. J., An improved zinc-finger nuclease architecture for highly specific genome editing. Nat Biotechnol, 2007. 25(7): p. 778-85. 333. Bunz, F., Dutriaux, A., Lengauer, C., Waldman, T., Zhou, S., Brown, J. P., Sedivy, J. M., Kinzler, K. W. and Vogelstein, B., Requirement for p53 and p21 to sustain G2 arrest after DNA damage. Science, 1998. 282(5393): p. 1497-501. 334. Chan, T. A., Hermeking, H., Lengauer, C., Kinzler, K. W. and Vogelstein, B., 14-3- 3Sigma is required to prevent mitotic catastrophe after DNA damage. Nature, 1999. 401(6753): p. 616-20. 335. Raj, K., Ogston, P. and Beard, P., Virus-mediated killing of cells that lack p53 activity. Nature, 2001. 412(6850): p. 914-7. 336. Saudan, P., Vlach, J. and Beard, P., Inhibition of S-phase progression by adeno- associated virus Rep78 protein is mediated by hypophosphorylated pRb. EMBO J, 2000. 19(16): p. 4351-61. 337. Li, H., Haurigot, V., Doyon, Y., Li, T., Wong, S. Y., Bhagwat, A. S., Malani, N., Anguela, X. M., Sharma, R., Ivanciu, L., Murphy, S. L., Finn, J. D., Khazi, F. R., Zhou, S., Paschon, D. E., Rebar, E. J., Bushman, F. D., Gregory, P. D., Holmes, M. C. and High, K. A., In vivo genome editing restores haemostasis in a mouse model of haemophilia. Nature, 2011. 475(7355): p. 217-21. 338. Okabe, M., Ikawa, M., Kominami, K., Nakanishi, T. and Nishimune, Y., 'Green mice' as a source of ubiquitous green cells. FEBS Lett, 1997. 407(3): p. 313-9. 339. Franz, W. M., Breves, D., Klingel, K., Brem, G., Hofschneider, P. H. and Kandolf, R., Heart-specific targeting of firefly luciferase by the myosin light chain-2 promoter and developmental regulation in transgenic mice. Circ Res, 1993. 73(4): p. 629-38. 340. Zincarelli, C., Soltys, S., Rengo, G. and Rabinowitz, J. E., Analysis of AAV serotypes 1-9 mediated gene expression and tropism in mice after systemic injection. Mol Ther, 2008. 16(6): p. 1073-80. 341. Pacak, C. A., Mah, C. S., Thattaliyath, B. D., Conlon, T. J., Lewis, M. A., Cloutier, D. E., Zolotukhin, I., Tarantal, A. F. and Byrne, B. J., Recombinant adeno- associated virus serotype 9 leads to preferential cardiac transduction in vivo. Circ Res, 2006. 99(4): p. e3-9. 342. Yang, L., Jiang, J., Drouin, L. M., Agbandje-McKenna, M., Chen, C., Qiao, C., Pu, D., Hu, X., Wang, D. Z., Li, J. and Xiao, X., A myocardium tropic adeno-associated virus (AAV) evolved by DNA shuffling and in vivo selection. Proc Natl Acad Sci U S A, 2009. 106(10): p. 3946-51. 343. Inagaki, K., Fuess, S., Storm, T. A., Gibson, G. A., McTiernan, C. F., Kay, M. A. and Nakai, H., Robust systemic transduction with AAV9 vectors in mice: efficient global cardiac gene transfer superior to that of AAV8. Mol Ther, 2006. 14(1): p. 45- 53. 344. Chu, D., Sullivan, C. C., Weitzman, M. D., Du, L., Wolf, P. L., Jamieson, S. W. and Thistlethwaite, P. A., Direct comparison of efficiency and stability of gene transfer into the mammalian heart using adeno-associated virus versus adenovirus vectors. J Thorac Cardiovasc Surg, 2003. 126(3): p. 671-9.

Appendix 75

9. Appendix

9.1 List of own publication

1. Hoher, T., Wallace, L., Khan, K., Cathomen, T. and Reichelt, J., Highly Efficient Zinc-Finger Nuclease-Mediated Disruption of an eGFP Transgene in Keratinocyte Stem Cells without Impairment of Stem Cell Properties. Stem Cell Rev, 2011.

2. Handel, E. M., Gellhaus, K., Khan, K., Bednarski, C., Cornu, T. I., Muller-Lerch, F., Kotin, R. M., Heilbronn, R. and Cathomen, T., Versatile and Efficient Genome Editing in Human Cells by Combining Zinc-Finger Nucleases With Adeno- Associated Viral Vectors. Hum Gene Ther, 2011.

3. Wali, A., Chishti, M. S., Ayub, M., Yasinzai, M., Kafaitullah, Ali, G., John, P. and Ahmad, W., Localization of a novel autosomal recessive hypotrichosis locus (LAH3) to chromosome 13q14.11-q21.32. Clin Genet, 2007. 72(1): p. 23-9.

4. Azeem, Z., Wasif, N., Basit, S., Razak, S., Waheed, R. A., Islam, A., Ayub, M., Kafaitullah, Kamran-ul-hassan Naqvi, S., Ali, G. and Ahmad, W., Congenital atrichia with papular lesions resulting from novel mutations in human hairless gene in four consanguineous families. J Dermatol, 2011. 38(8): p. 755-60. Appendix 77

9.3 Acknowledgements

All the praise belongs to ALLAH the most gracious and the most merciful. Who bestowed the man with the wisdom and knowledge to think and search in the universe. All regards and respects are for the Holy prophet Hazrat Muhammad (sallAllahu alaihi wa sallam).

Afterward, I express my indebtedness to my supervisor Prof. Dr. Toni Cathomen for his devoted guidance, unflinching support, patience, funding and especially open door policy to his students, which made it possible for me to complete this study.

I would also like to thank my co-supervisor PD Dr. Albert Heim for his constructive criticism and support.

Special thanks are extended to Prof. Dr. Christopher Baum for his encouragement, support, and guidance during my stay here in the Institute of Experimental Hematology as well as his kind and cordial treatment of colleagues.

I am obliged to Prof. Dr. Denise Hilfiker-Kleiner for her sincere help, especially for the completion of in vivo experiments.

I would like to thank my examiner and reviewer Prof. Dr. Jürgen Alves.

I would like to thank all of my colleagues from the Institute of Experimental Hematology especially our own group including Cem Söllü, Christien Bednarski, Nico Jäschke, Jamal Al-Zu'bi, Anne-Kathrin Dreyer, Sylwia Bobis-Wozowicz, Deepika Pothiraju, Shamim Rahman, Claudio Mussolino, Nadine Dannemann, Jessica Wenzl, Fabienne Lütge, and Eva-Maria Händel for their help, support and time.

I owe my gratitude to my parents and grandmother for their unconditional love, support, encouragement, care, teaching and prayer for me. I have no words to express my feeling for them. I also wish to thank my brothers Haji Samiullah Khan, Hafiz Ghulamullah Khan and sister as well as uncles Niamatullah khan, Mateeullah khan, Muhammad Karim, Muhammad Akram, Hassan Amin and all aunts for their support, help, love and care. I also extended my thanks to my parent-in-law. My special thanks are for my wife for her support, encouragement and patience during my stay here in Germany.

Last but not least I like to thanks Deutscher Akademischer Austauschdienst (DAAD), Higher Education Commission (HEC) of Pakistan, University of Balochistan and Hannover Medical School for financial and administrative support.

Appendix 78

9.4 List of abbreviations

AAV Adeno-associated virus AAVS1 Adeno-associated virus integration site 1 A1AT Alpha 1-anitrypsin A1ATD Alpha1-antitrypsin deficiency Ad Adenovirus AIDS Acquired immune deficiency syndrome ALD Adrenoleukodystrophy bp Base pair CAR Coxsackievirus and adenovirus receptor CARE Cis-acting replication element CMV Cytomegalovirus CoDA Context-dependent assembly DSB Double strand break dsDNA Double-stranded DNA EIA Equine infectious anaemia virus FGFR-1 Fibroblast growth factors receptor-1 FIV Feline immunodeficiency virus HIV1 Human immunodeficiency virus 1 HIV2 Human immunodeficiency virus 2 HR Homologous recombination HSPG Heparin sulfate proteoglycan HSV Herpes simplex virus iPSC Induced Pluripotent Stem Cells ITR Inverted terminal repeat kDa Kilodaltons KSCs Keratinocyte stem cells LamR Laminin receptor LTR Long terminal repeat MLC Myosin light chain mRNA Messenger ribonucleic acid NHEJ Non-homologous end joining NMD Nonsense-mediated decay OPEN Oligomerized pool engineering OTC Ornithine transcarbamylase PDGFR Platelet-derived growth factor receptor PEI Polyethyleneimine PLL Polylysine Appendix 79

psi Packaging signal rAAV Recombinant adeno-associated virus RBE Rep binding element RNA Ribonucleic acid SERPINA1 Serpin peptidase inhibitor, clade A (alpha-1 antiproteinase, antitrypsin), member 1 siRNA Small interfering RNA SV15 Simian adenovirus type 15 stock SIV Simian immunodeficiency virus TALENs Transcription Activator-Like Effector Nucleases TFOs Triplex-forming oligonucleotides T7E1 T7 endonuclease I VP Viral protein VSV-G Vesicular stomatitis virus glycoprotein WAS Wiskott-Aldrich Syndrome ZF Zinc finger ZFN Zinc finger nuclease

Appendix 80

9.5 Erklärung

Hiermit erkläre ich, dass ich die Dissertation “Establishing a zinc finger nuclease platform for targeted genome editing in human cell lines, primary keratinocyte stem cells and cardiomyocytes in vivo” selbstständig verfasst habe.

Bei der Anfertigung wurden folgende Hilfen Dritter in Anspruch genommen (namentliche Nennung weiterer an der Dissertation beteiligter Personen und ihre Funktion bei der Erstellung der Dissertation):

Prof. Dr. Denise Hilfiker-Kleine’s research group performed intramyocardial injection and immunohistology for the unpublished data.

Ich habe keine entgeltliche Hilfe von Vermittlungs- bzw. Beratungsdiensten (Promotionsberater oder anderer Personen) in Anspruch genommen. Niemand hat von mir unmittelbar oder mittelbar entgeltliche Leistungen für Arbeiten erhalten, die im Zusammenhang mit dem Inhalt der vorgelegten Dissertation stehen.

Ich habe die Dissertation an folgenden Institutionen angefertigt:

Institute of Virology, Charité - Universitätsmedizin Berlin and Institute of Experimental Hematology, Hannover Medical School, Hannover

Die Dissertation wurde bisher nicht für eine Prüfung oder Promotion oder für einen ähnlichen Zweck zur Beurteilung eingereicht. (Ist die Dissertation in einer auswärtigen Institution angefertigt worden, so ist zugleich eine Erklärung der betr. Leiterin oder des Leiters beizufügen, dass sie oder er mit der Einreichung der Arbeit als Dissertation an der Medizinischen Hochschule einverstanden ist.) Ich versichere, dass ich die vorstehenden Angaben nach bestem Wissen vollständig und der Wahrheit entsprechend gemacht habe.

14.03.2012

Datum und Unterschrift