Introduction

Zinc finger nucleases (ZFNs) are a class of engineered DNA-binding proteins that facilitate targeted editing of the genome by creating double-strand breaks in DNA at user-specified locations. Double-strand breaks are important for site-specific mutagenesis in that they stimulate the cell's natural DNA-repair processes, namely homologous recombination and Non-Homologous End Joining (NHEJ). By implementing established, field proven methods, these processes are harnessed to generate precisely targeted genomic edits, resulting in cell lines with targeted gene deletions, integrations, or modifications.

Zinc-finger nucleases (ZFNs) are the gene-targeting tools which are also known as targetable DNA cleavage reagents. ZFN induced double-strand breaks are subject to cellular DNA repair processes which thereby lead to two sub processes: targeted mutagenesis and targeted gene replacement. Considerable progress has been made in methods for the improvement in this technique and the approaches to design and selection of ZFNs are still being perfected.

The ability to modify with high fidelity a complex genome has revolutionized biomedical research in the late 1980s and early 1990s. The underlying technology is known as gene targeting and is based on the cellular homologous recombination (HR) pathway, which has evolved to promote genetic recombination during meiosis and the repair of DNA double strand breaks (DSBs) before mitosis. A recent technological breakthrough, however, has changed this dreary perspective.

By fusing engineered zinc-finger (ZF) DNA-binding domains to a non-specific nuclease domain, so called zinc-finger nucleases (ZFNs) were generated. These ZFNs can be designed to introduce a Double Stranded Break into a desired target locus and, as a consequence, stimulate gene targeting 100 to 10,000-fold by activating cellular DNA repair pathways. This technology might be used to correct inborn mutations in adult stem cells derived from patients with genetic disorders. These corrected cells could subsequently be used to repopulate an affected organ and thereby reverse the disorder.

The expression repertoire of all the genes in any given cell is governed by transcriptional activators and repressors that bind to specific sites in the genome. There are several protein folds that can elicit sequence specific DNA binding, including helix-turn-helix, leucine zipper and zinc finger domains. The C2H2 zinc finger motif, which comprises 20 to 30 amino acids containing two Cysteine and two Histidine residues coordinated by a zinc atom, has proven to be particularly versatile for protein engineering applications. An archetypal member of this family is Zif268, also known as EGR1, which is a transcriptional regulator initially found in mice. The crystal structure of the mouse Zif268 three zinc finger peptide bound to its target DNA sequence showed that the individual zinc fingers fold into two antiparallel β sheets and an α helix, with the α helix making sequence specific DNA contacts in the major grove of the DNA.

Each zinc finger binds three nucleotides, with the entire Zif268 polypeptide binding a nine base pair (bp) GCG-TGG-GCG DNA motif. Such a modular design immediately suggested the possibility for combining zinc fingers with distinct triplet recognition motifs, to create proteins that could potentially recognize any DNA sequence. This could include pre-existing zinc fingers with known triplet binding sequences, as well as entirely novel designer zinc fingers generated against new DNA sequences.

Indeed, even before the crystal structure and mode of binding was known, systematic mutations in α helix of the second zinc finger of the Sp1 transcription factor had been shown to shift DNA binding specificity. Moreover, mix and match shuffling of individual zinc fingers was shown to shift DNA binding specificity towards contiguous triplet DNA sequences accordingly.

Zinc finger nucleases (ZFNs) induce double-strand DNA breaks at specific recognition sites. ZFNs can dramatically increase the efficiency of incorporating desired insertions, deletions, or substitutions in living cells. These tools have revolutionized the field of genome engineering in several model organisms and cell types including zebrafish, rats, and human pluripotent stem cells.

Genetics is driven by the ability to connect genotype with phenotype. The classical approach is to identify a novel phenotype, whether occurring spontaneously or derived by mutagenesis, to identify the responsible gene(s) and to discover why mutations at that locus have the observed effect. A more modern approach, sometimes called reverse genetics, is to identify a gene from a genomic sequence to make mutations specifically in that gene and to characterize the resulting phenotype. Two types of gene-specific manipulations can be envisioned (Figure 1). In one, which we can call "targeted gene replacement," the goal is to make localized sequence changes, often ones that will create a null mutation. In targeted gene replacement, the goal is to replace an existing sequence with one designed in the laboratory. The latter allows the introduction of both more subtle and more extensive alterations.

Making directed genetic changes is often called "gene targeting." It sounds simple enough, but targeting a single gene within a large genome presents a substantial challenge. Procedures for gene replacement in baker's yeast, Saccharomyces cerevisiae, have been available for several decades (Scherer and Davis 1979; Rothstein 1983). Success in this case depends on several features: the ability to manipulate segments of yeast DNA in the laboratory, the ability to introduce DNA into yeast cells, interaction between donor and target DNA by homologous recombination, the near absence of competing reactions that would integrate the donor into alternative sites in the genome, and the ability to apply strong selection for the desired product. These properties are shared by some other fungi and many bacteria, but not by the majority of eukaryotic organisms.

Making targeted gene replacements has also become standard practice in mice, thanks to the availability of embryonic stem (ES) cells that can be manipulated in culture and the development of powerful selection procedures (Capecchi 2005). Like targeting in yeast, the process in mice depends on homologous recombination between the donor and the target. In addition, selection must be applied against the more common products of random integration. This is accomplished by placing a positive selectable marker inside the donor homology and a negative selectable marker outside the homology (Mansour et al. 1988). Double selection yields the desired replacements, and the pluripotency of the ES cells allows them to populate all cell lineages after injection into early embryos.

In both yeast and mouse cells, the absolute frequency of homologous recombination between donor and target sequences is quite low--on the order of one in every 104 to 107 cells. Selection in culture allows the recovery of the rare cells that have enjoyed the desired event. With other experimental organisms, ES cells are not available, screening or selection procedures are not adequate, and development of useful gene-targeting approaches is impeded by the low frequency of recombination.

Zinc finger nucleases to be useful for genome engineering, an endonuclease must exhibit an extraordinary combination of qualities: specific recognition of long target sequences (ideally, long enough for unique occurrence in a eukaryotic genome) coupled with sufficient adaptability for retargeting to user-defined sequences. The ZFN architecture meets these specifications by linking the DNA-binding domain of a versatile class of eukaryotic transcription factors. Zinc finger proteins (ZFPs) -- with the nuclease domain of the FokI restriction enzyme.

By virtue of their structure, ZFNs combine the favorable qualities of both components -- the DNA binding specificity and flexibility of ZFPs and a cleavage activity that is robust but restrained in the absence of a specific binding event -- while retaining functional modularity. As a consequence, both the DNA-binding and catalytic domains can be optimized in isolation, which simplifies retargeting and platform improvement efforts.

The zinc finger
A platform for the design of novel DNA binding domains

The ZFP region provides a ZFN with the ability to bind a discrete base sequence. This region contains a tandem array of Cys2-His2 fingers each recognizing approximately 3 bp of DNA. In early studies individual ZFNs used three fingers to bind a 9-bp target, which enabled ZFN dimers (the active species) to specify 18 bp of DNA per cleavage site. More recent studies have added more fingers (up to six per ZFN) to specify longer and rarer cleavage targets. A variety of strategies have been described for making ZFPs with new, user-chosen binding specificities.

The first emerged from observations of the initial ZFP- DNA co-crystal structure, which suggested a substantial degree of functional autonomy in the interaction of individual fingers with DNA8. The approach, which has been termed 'modular assembly', generates candidate ZFPs for a given target sequence by identifying fingers for each component triplet and linking them into a multifinger peptide targeted to the corresponding composite sequence. Fingers used for modular assembly have been developed for most triplet sequences 10-16. The method has been used to develop the zinc finger component of active ZFNs for a number of endogenous targets in higher eukaryotic cells. Besides modular assembly, several alternative strategies for making ZFPs have been developed. These newer methods were designed to accommodate the deviations from strict functional modularity observed for many zinc fingers.

Whatever the design method, the production of a DNA binding module evaluated in vitro for affinity and specificity towards its intended target provides only the first step towards use in vivo. Indeed ZFNs assembled from in vitro 'validated' ZFPs often fail to drive genome editing at the endogenous locus when tested in living cells.

One factor is specificity: complex genomes often contain naturally occurring multiple copies of a sequence that is identical or highly related to the intended target (for example, paralogues or pseudogenes), and these copies can act as additional targets for ZFNs. Researchers have addressed this issue by building up a detailed understanding of the rules governing protein-DNA interactions and by exploiting minor sequence divergences between related genomic regions. An additional problem is the chromatin structure at target sites, which may not be amenable to cleavage.

Targeted mutations induced by ZFN cleavage

The consequences of a double strand break induced by targeted ZFN expression vary upon several different factors, but in general cells repair these breaks by either non-homologous end joining (NHEJ) or homologous recombination (HR). The FokI-mediated DNA cleavage leaves overhanging ends, which can either be filled in, completely or incompletely, or chewed back by limited exonuclease activity, followed by ligation by endogenous DNA ligase activity.

In some cases small regions of single stranded micro homology are used to initiate strand annealing and ligation. The net result of this process is that small insertions or deletions are usually generated in the region flanking the ZFN mediated processing. When the ZFPs are targeted to coding regions, the outcome is often a shift in the reading frame, usually leading to a null allele of the targeted gene.

Gene replacement induced by ZFN

The second method used by cells to repair damaged DNA is homologous recombination (HR). This can be favourably exploited in the context of ZFN-mediated cleavage by introducing exogenously added donor plasmids containing homologous DNA stretches. Initial proof-of-concept experiments in human cells showed the feasibility of this approach by demonstrating the ability to repair a mutated form of GFP stably integrated into the genome. The first example of ZFN-mediated repair of an endogenous mutated human gene was presented by Urnov et al. In this study, the authors designed ZFNs targeting IL2Rγ, the gene mutated in X-linked severe combined immune deficiency (SCID). They used ZFNs targeted to exon 5 of IL2Rγ, and a plasmid carrying this exon with a restriction site to monitor incorporation, flanked by approximately 1 kb homology either side of the mutation.

Co-transfection of ZFNs and this plasmid into K562 cells showed a 10 to 20% modification rate of the IL2Rγ exon 5 across a polyclonal cell population, with similar frequencies documented by limiting dilution cloning (13/96 (14%) heterozygous clones, of which 6/9 (67%) were homozygous for the introduced mutation). The authors then introduced a frameshift mutation into one or both alleles of IL2Rγ using these reagents, thereby mimicking the defect seen in the human disease, and finally fixed the mutation using a donor plasmid containing the wild type exon. This latter experiment shows the path for use of this approach in gene therapy. Primary patient T cells could be extracted and modified ex vivo, before subsequent re-implantation. A huge benefit of this approach compared to viral based gene therapy methods, is the elimination of potential detrimental gene reactivation events due to insertional mutagenesis [56]. Mouse cells have also been shown to be responsive to ZFN-mediated genome editing through homologous recombination. Melanocytes derived from albino mice contain a point mutation in the first exon of the tyrosinase gene. Cotransfection of these melanocytes with ZFNs targeting a sequence 80 bps from the Cys85Ser mutation, together with the WT donor sequence resulted in pigmented cells as early as four days post-transfection.

Finally, successful genetic modification of mouse ES cells has also been enabled by ZFNs, which is discussed in more detail in a later section.

Benefits

1. Rapid disruption of, or integration into, any genomic loci
2. Mutations made are permanent and heritable
3. Works in a variety of mammalian somatic cell types
4. Edits induced through a single transfection experiment
5. Knockout or knock-in cell lines in as little as two months
6. Single or biallelic edits occur in 1-20% of clone population
7. No antibiotic selection required for screening

ZFN-associated toxicity

ZFN-induced cytotoxicity is a major potential issue, which has been reported in several studies. Cell death and apoptosis associated with ZFN expression is most likely the result of excessive cleavage at off-target sites, which in turn suggests imperfect target site recognition by the ZF DNA-binding domains. Since therapeutic gene targeting strongly depends on creation of a DSB at a specific target site, the implementation of quantitative assays to assess immediate and long-term genotoxicity of artificial nucleases is paramount. In some studies the extent of cytotoxicity upon ZFN expression was quantified by measuring cell survival or apoptosis. Although these quantitative assays can be used to characterize the specificity and immediate genotoxicity of any artificial nuclease of interest, they are not very sensitive and they do not provide information about the sites at which off-target DSBs occur.

The long-term effects of ZFN-induced mutagenesis to induce unpredictable oncogenicity can be assessed by soft agar transformation studies or in vitro transformation assays using purified lineage-negative cells from murine bone marrow. Moreover, cytogenetic analyses, like spectral karyotyping, can provide information about whether ZFN activity induces chromosomal abnormalities and/or translocations. However this is now clear that the long-term consequences of ZFN-induced DSBs can only be studied in vivo. Assays previously developed to evaluate the genotoxicity of retroviral vectors in gene therapy protocols should prove useful to study the malignant potential of cells upon overexpression of ZFNs.


ZFNs in therapeutics

It has long been envisaged that by applying the ZFN technology to stem cells, inherited mutations could be repaired ex vivo and functionally corrected stem cells transplanted back into patients to repopulate the affected tissues and cure the disease. Importantly, gene correction would restore the functionality of the affected gene product and, at the same time, retain its normal endogenous expression pattern, thereby overcoming a major limitation of conventional gene therapy approaches. Gene correction might work even more efficiently if the repaired gene provides the modified stem cell with a growth advantage. For example, in the case of X-SCID, which is caused by mutations in the IL2Rγ locus, correction of only a small number of genetically corrected HSCs will be sufficient to restore proper function of the immune system. Several obstacles continue to limit the exploitation of ZFNs in a therapeutic setting. The following criteria should be met in order for ZFNs to be successfully applied in a clinical setting:

→ ZFN architecture: As outlined above, ZFNs must consist of a DNA-binding domain with high specificity to the target site and a "regulated" nuclease cleavage domain.

→ Choice of delivery system: To date four different systems have been reported to be suitable to mediate DSB-stimulated targeted genome editing in human cells: plasmid-DNA introduced by transfection AAV vectors; integrase-deficient lentiviral vectors (IDLVs) and modified adenoviral vectors. Plasmid-DNA has been the most commonly used ZFN expression vector thus far.

Owing to their superior transduction record, AAV vectors, IDLVs and adenovirus type 5 vectors substituted with a type 35 fiber structure (Ad5/35) are promising tools to deliver high numbers of ZFN expression cassettes into stem cells, such as hematopoietic and mesenchymal human stem cells. However, independent of the nature of the expression vector, the delivery of DNA expression cassettes containing strong promoters - and this can include donor DNAs for targeted gene addition - is associated with the potential risk of insertional mutagenesis, as reported for both integrating and "episomal" vectors. Moreover, transient expression of ZFNs is strongly preferred over permanent expression of the nucleases. The efficient ex vivo delivery of ZFNs in the form of mRNA has therefore been an important breakthrough to achieve this goal and to reduce the risk of vector integration.

→ Genotoxic side effects: As mentioned above, a comprehensive evaluation of treated cells for potential ZFN-induced side-effects both short-term and long-term is paramount.

→ Immune response: Assessment of the potential immune reactivity against ZFNs, especially against the bacterial FokI domain should be included, especially when applying ZFNs in vivo.


Conclusions

The main advantage of using ZFN-stimulated gene targeting over conventional gene-addition type gene therapy is the potential to preserve temporal and tissue-specific gene expression. Gene knockout via NHEJ-mediated repair of ZFN-induced DSBs is another promising application of this technology. While the overall efficiency of ZFN-induced genome editing depends on the activity and specificity of custom-made ZFNs, additional parameters, such as the apoptotic threshold of a cell and the proficiency to activate the appropriate DNA repair pathways, both of which may vary significantly among different cell types, will be important too.

Despite tremendous recent progress, the development of methods to better understand and quantify ZFN-associated genotoxicity remains a major priority and challenge for future research.
Cleavage at unintended, off-target DNA sites is likely to be the decisive factor for ZFN-associated toxicity. Although the development of a malignant phenotype is hypothesized to require multiple genetic insults, every genetic manipulation poses a risk, especially in stem and progenitor cells with their high proliferative potential. In view of recent adverse events in gene therapy trials to treat XSCID, development of both "safer" ZFNs and "safer" delivery methods to minimize genotoxic side effects will be important in order to bring the next generation of gene therapy tools into the clinic.

Manipulating endogenous genes at will has long been the goal for researchers wishing to understand and treat human diseases. The use of engineered zinc fingers to modify specific genomic loci is a relatively recent addition to this area, but is rapidly showing enormous promise at becoming a reliable research and therapeutic tool. In fact, designer ZFPs are already in clinical trials, and a multitude of others are in development. However, there are a few cautions that prevent unbridled enthusiasm at this time and point to the need for further technical development. For example, it is clear that some regions of the genome are more targetable than others by ZFPs; the reason for this is not completely clear, and until this is resolved, progress will remain somewhat impeded. In addition, the potential toxicities as a result of off-target binding and cutting (in the case of ZFNs) are still not completely understood, and the methods currently available to even monitor these events are laborious. While ZFNs have been successfully employed in numerous model organisms to generate gene knock-outs, the ability of ZFNs to enhance the generation of knock-in animals remains largely untested. Finally, the expansion of therapeutic ZFPs will also be limited by the ability to efficiently deliver the ZFP into the disease relevant cell. Nevertheless, given the explosive progress made in just the last 10 to 15 years, one cannot help but be excited about what will happen in the next equivalent time period.

About Author / Additional Info:
1. Targeted Modifications of the Human Genome using Zinc-Finger Nucleases
Toni Cathomen
Charite Medical School, Institute of Virology (CBF)-12203 Berlin, Germany
2. Zinc Finger Nucleases as tools to understand and treat human diseases
David Davis and David Stokoe
3. Genome editing with engineered zinc finger nucleases
Fyodor D. Urnov, Edward J. Rebar, Michael C. Holmes, H. Steve Zhang and Philip D. Gregory
4. Genome Engineering With Zinc-Finger Nucleases
Dana Carroll
Department of Biochemistry, University of Utah School of Medicine, Salt Lake City, Utah 84112-5650
5. http://www.sigmaaldrich.com/life-science/zinc-finger-nuclease-technology/learning-center/what-is-zfn.html
6. Image source: wikipedia.org/wiki/Zinc_finger_nuclease - By Dana Carroll