Targeted gene addition to a predetermined site in the human genome using a ZFN-based nicking enzyme

Zinc-finger nucleases (ZFNs) are chimeric endonucleases created by fusing the cleavage domain from the type IIS restriction enzyme FokI to a designed zinc-finger protein (ZFP). In order to generate a DNA double strand break (DSB) at a predetermined genomic site, two engineered ZFNs must be designed to bind DNA with the appropriate orientation and spacing (i.e., on opposite sides of the DNA with 5 or 6 bp of sequence separating them) (Carroll 2008; Cathomen and Joung 2008). This architecture (analogous to that depicted in Fig. 1A) permits the two FokI cleavage domains to heterodimerize—a prerequisite for subsequent cleavage of the targeted DNA (Bitinaite et al. 1998). Figure 1. Mutation of the catalytic domain of a ZFN monomer generates a ZFN heterodimer with nicking activity. (A) Illustration of the ZFN heterodimer architecture for single-stranded break (SSB) generation. The FokI catalytic domain of ZFN (left, blue) is mutated ... Repair of the ZFN-induced site-specific DSB provides the molecular basis for gene disruption, gene correction, or gene addition—the different potential outcomes of ZFN-mediated genome editing (for review, see Urnov et al. 2010). Specifically, a given DSB may be repaired via either the nonhomologous end joining (NHEJ) or homology-directed repair (HDR) pathways (Kowalczykowski 2000; Sung and Klein 2006; Wyman and Kanaar 2006; Brugmans et al. 2007). HDR utilizes a homologous donor sequence as template for the conservative repair of the DNA break. Provision of a suitably designed donor DNA molecule can, therefore, specify gene addition or gene correction at the ZFN-induced DSB. NHEJ, on the other hand, simply catalyzes the rejoining of the two DNA ends—a process that can result in the deletion or insertion of nucleotides at the repair junction. DNA repair via NHEJ is, therefore, mutagenic. To date, ZFN-induced DNA repair via HDR or NHEJ has been utilized to target modifications to the genomes of numerous species (Le Provost et al. 2010) including Caenorhabditis elegans (Morton et al. 2006), Drosophila melanogaster (Bibikova et al. 2002), silkworms (Takasu et al. 2010), zebrafish (Doyon et al. 2008; Meng et al. 2008), sea urchins (Ochiai et al. 2010), Arabidopsis thaliana (Osakabe et al. 2010; Zhang et al. 2010), tobacco (Maeder et al. 2008; Cai et al. 2009; Townsend et al. 2009), corn (Shukla et al. 2009), mice (Carbery et al. 2010; Meyer et al. 2010), and rats (Geurts et al. 2009). ZFNs have also been used to efficiently engineer a diverse range of mammalian cell types (Urnov et al. 2010), including human stem cells, with subsequent retention of full potency and normal growth characteristics (Hockemeyer et al. 2009; Zou et al. 2009; Holt et al. 2010). Moreover, the therapeutic potential of ZFNs is currently under evaluation in clinical trials (clinicaltrials.gov identifiers NCT00842634, NCT01252641, NCT01044654). Given the markedly different genome editing outcomes mediated by NHEJ and HDR, it would be desirable to have control over the choice of pathway used and thus specify the outcome of repair. In practice, for gene disruption this can already be achieved, since NHEJ is the predominant pathway in mammalian cells (Lieber 2008) and codelivery of donor DNA is unnecessary. For gene correction and addition, however, NHEJ events represent competition to the desired homology-directed outcome at the target locus. One potential strategy for enforcing DNA repair pathway bias would be the targeted introduction of a DNA single strand break (SSB) or nick. Like the DSB, a DNA SSB/nick can theoretically stimulate homology-directed repair pathways (Holliday 1964; Meselson and Radding 1975; Radding 1982). For example, SSBs induced by the bacteriophage fd gene II protein (gIIp) stimulated HDR in yeast (Galli and Schiestl 1998), while nicks generated by specific mutants of the RAG proteins—critical components of the V(D)J recombination system—are repaired via initiation of homologous recombination (Lee et al. 2004). Moreover, experiments using a nick-inducing variant of the I-Anil LAGLIDADG homing endonuclease showed that the SSB/nicks induced by these engineered enzymes are also repaired via HDR (Smith et al. 2009). Importantly, and in contrast to a DSB, a DNA nick or SSB is not a bona fide substrate for repair via the NHEJ pathway. Thus, a targeted SSB has the potential to restrict repair to the homology-directed pathway. Here, we describe the engineering of a zinc finger nickase (ZFNickase) by mutation of a residue critical for FokI cleavage activity in one monomer of the ZFN heterodimer, thus allowing the ZFNs to heterodimerize on DNA but restricting cleavage to a single DNA strand. We show that targeting of these ZFNickases to the endogenous CCR5 locus stimulates gene correction and gene addition via HDR at frequencies of between ∼1%–8% in both transformed and primary human lines. Importantly, we show that the CCR5 ZFNickase-induced gene addition occurs without a marked increase in detectable mutagenesis by the error-prone NHEJ repair pathway. Specific outcomes of either gene addition/correction or gene disruption may, therefore, be modulated via the targeted introduction of a single- or double-strand break, respectively. Moreover, beyond the targeted event itself, elimination of DNA repair via NHEJ has the potential to eradicate mutagenesis at putative off-target cleavage sites, further improving the fidelity of genome editing.