SEARCH

SEARCH BY CITATION

Keywords:

  • double-strand break;
  • I-CreI;
  • mutation;
  • transgenic plant;
  • maize;
  • Zea mays L

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

The liguleless locus (liguleless1) was chosen for demonstration of targeted mutagenesis in maize using an engineered endonuclease derived from the I-CreI homing endonuclease. A single-chain endonuclease, comprising a pair of I-CreI monomers fused into a single polypeptide, was designed to recognize a target sequence adjacent to the LIGULELESS1 (LG1) gene promoter. The endonuclease gene was delivered to maize cells by Agrobacterium-mediated transformation of immature embryos, and transgenic T0 plants were screened for mutations introduced at the liguleless1 locus. We found mutations at the target locus in 3% of the T0 plants, each of which was regenerated from independently selected callus. Plants that were monoallelic, biallelic and chimeric for mutations at the liguleless1 locus were found. Relatively short deletions (shortest 2 bp, longest 220 bp) were most frequently identified at the expected cut site, although short insertions were also detected at this site. We show that rational re-design of an endonuclease can produce a functional enzyme capable of introducing double-strand breaks at selected chromosomal loci. In combination with DNA repair mechanisms, the system produces targeted mutations with sufficient frequency that dedicated selection for such mutations is not required. Re-designed homing endonucleases are a useful molecular tool for introducing targeted mutations in a living organism, specifically a maize plant.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Chromosomal DNA breaks are highly recombinogenic events that frequently lead to genetic lesions (Puchta, 2005; Paques and Duchateau, 2007). DNA deletions, insertions, and homology-driven replacements of DNA segments are produced at the endogenous chromosomal sites of double-strand breaks (DSBs) by non-homologous end joining and/or homologous recombination reactions in higher eukaryotic organisms such as Drosophila, Caenorhabditis elegans, Danio rerio, Nicotiana tabacum and Zea mays (Bibikova et al., 2003; Morton et al., 2006; Doyon et al., 2008; Maeder et al., 2008; Meng et al., 2008; Cai et al., 2009; Shukla et al., 2009; Townsend et al., 2009), or higher eukaryotic cells (Moehle et al., 2007; Maeder et al., 2008; Santiago et al., 2008). A key element of this approach is the introduction of DSBs at a chromosomal locus of interest. Currently, the predominant reagents for targeting such breaks are zinc-finger nucleases (ZFNs), which consist of an engineered zinc finger DNA-binding domain and the DNA cleavage domain from the FokI restriction enzyme (Kim et al., 1996; Porteus and Carroll, 2005). Natural rare-cutting endonucleases called homing endonucleases provide an alternative starting material for development of ‘molecular scissors’ (Stephens et al., 1997; Seligman et al., 2002). These proteins are bigger than zinc-finger proteins and make numerous contacts with the DNA recognition sequences (Stoddard, 2005; Moure et al., 2008). Unlike ZFNs, the DNA-binding domains of most homing endonucleases are not clearly separated from the catalytic domains. This makes protein engineering procedures more complex (Prieto et al., 2007; Szczepek et al., 2007), but may confer greater selectivity for the desired target site than is possible with ZFNs (Redondo et al., 2008). Therefore, there is great interest in re-engineering natural homing endonucleases to develop reagents for targeted genome modification (Arnould et al., 2006; Ashworth et al., 2006; Rosen et al., 2006).

ZFNs have successfully been explored for introducing targeted DNA modifications in eukaryotic cells for several years (Carroll, 2008), but the postulated application of re-designed homing endonucleases has only recently been tested in the natural chromosomal context of endogenous loci (Grizot et al., 2009). Great effort has been devoted to the production of comprehensively re-engineered homing endonucleases that are capable of recognizing DNA sequences found at loci of interest (Paques and Duchateau, 2007). A popular endonuclease scaffold for such designs is the I-CreI homing endonuclease that is found in chloroplasts of Chlamydomonas reinhardti (Thompson et al., 1992). This endonuclease is a homodimer that recognizes a pseudo-palindromic 22 bp DNA site in the 23S rRNA gene and creates a DSB that is used for the introduction of an intron in a process referred to as ‘homing’ (Durrenberger and Rochaix, 1993). Experimental selection and computational design methods have been developed to re-engineer the cleavage specificity of I-CreI and its homologs (Ashworth et al., 2006; Smith et al., 2006; Arnould et al., 2007). In addition, methods have been developed to force the heterodimerization of differentially engineered I-CreI monomers, thus yielding engineered endonucleases that recognize non-palindromic DNA sites (Fajardo-Sanchez et al., 2008). These efforts recently culminated in the production of re-designed endonucleases that create DSBs in the human XPC and RAG1 genes (Redondo et al., 2008; Grizot et al., 2009).

In plants, extensive work has been carried out on DSB repair mechanisms using the yeast mitochondrial I-SceI homing endonuclease in combination with artificial I-SceI recognition sequences (Puchta, 2005). It has been shown that plant DSBs are predominantly repaired by non-homologous end-joining pathways, although homologous recombination may significantly contribute to the repair process under appropriate conditions (Salomon and Puchta, 1998; Orel et al., 2003; D’Halluin et al., 2008). This has enabled precise genetic alterations to be made at the site of I-SceI-induced DNA breaks in Arabidopsis and tobacco, as well as agronomically important plants such as maize (D’Halluin et al., 2008; Yang et al., 2009). Likewise, engineered ZFNs have been used successfully to target genetic lesions to introduced sites in Arabidopsis plants and tobacco protoplasts (Lloyd et al., 2005; Wright et al., 2005). More recently, this work was extended to endogenous loci by research groups who demonstrated the induction of DSBs at the endogenous endochitinase (CHN50) and SuRA genes in tobacco and the IPK1 locus in maize using engineered ZFNs (Maeder et al., 2008, Cai et al., 2009;Shukla et al., 2009; Townsend et al., 2009).

The next milestone in the achievement of targeted mutagenesis stimulated by engineered homing endonucleases such as I-SceI or I-CreI is the induction of DSBs at natural chromosomal sites of choice. While this paper was under review, Grizot et al. (2009) published results documenting the successful use of a single-chain I-CreI-based V2/V3 homing endonuclease for targeted modification of an SCID gene in human cells. Here we demonstrate targeted mutagenesis in maize (Zea mays) using an engineered I-CreI endonuclease that was designed to produce DSBs at the liguleless1 chromosomal locus.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Target site selection and design of the endonuclease

We selected the Lig34-site in the vicinity of the LIGULELESS1 (LG1) gene on the short arm of chromosome 2 of Zea mays for DSB induction because this locus is close to an existing transgenic event with significant flanking sequence data available. Also, inadvertent disruption of the LG1 gene should not be detrimental for plant growth and development. The search was limited to the 8.1 kb flanking sequences around the LG1 coding sequences. Two potentially useful target sites were originally identified: Lig12 and Lig34. The LIG 1::2 endonuclease, which was designed to recognize the DNA sequence 5′-GTGCTGTACGTACGAGAATTTC-3′, was found to cleave its intended recognition site in vitro and in a homologous recombination assay in a human cell line, but did not generate mutations at the intended chromosomal site in initial maize transformations (data not shown). Therefore, LIG 1::2 was not investigated further. The Lig34-site was 1171 bp upstream of the translation start codon of the LG1 gene (Figure 1a). Among known genomic maize DNA, it is a unique sequence in ETX, an inbred maize genotype used for transformation. The Lig34-site sequence differs from the recognition sequence for wild-type I-CreI at 14 of 22 base pairs (Figure 1b). To produce an engineered endonuclease that recognizes the Lig34-site sequence, we used a structure-based protein design method to modify the DNA-binding characteristics of I-CreI. Briefly, based on visual inspection of the I-CreI–DNA co-crystal structure, we predicted a large number of amino acid substitutions that change I-CreI base preference at particular positions in its recognition site. Individual amino acid substitutions were evaluated experimentally, and those that conferred the desired change in base preference were added to a database of mutations that can be ‘mixed and matched’ to generate derivatives of I-CreI that recognize highly divergent DNA sites. In theory, the combinatorial diversity available using our current mutation database is sufficient to target an engineered endonuclease approximately every 1000 bp in a random DNA sequence. To target the Lig34-site sequence, we introduced five amino acid substitutions (I24K, Y33C, R68Y, I77E and E80Q) into the first I-CreI monomer (referred to as ‘LIG 3’) and ten amino acid substitutions (Q26S, N30R, S32C, Q38E, S40R, T42E, Q44T, R68Y, I77R and E80Q) into the second I-CreI monomer (referred to as ‘LIG 4’, Figure 1c). Most of these substitutions were introduced to modify base preference. In each case, however, one mutation (E80Q) was introduced to increase the activity of the engineered endonuclease by removing an unfavorable electrostatic interaction between I-CreI and the negatively charged DNA backbone. The LIG 3 and LIG 4 monomers were then fused into a single polypeptide using a 38 amino acid linker, which joined the C-terminus of the LIG 3 monomer to the N-terminus of the LIG 4 monomer (Figure 1c). The resulting ‘single-chain’ endonuclease was called LIG 3::4.

image

Figure 1.  The selected liguleless1 chromosomal target locus and engineered I-CreI monomers. (a) The LG1 exons are shown as shaded arrows. The distance between the ATG codon and the site of the double-strand break is 1171 bp. The length of the PCR diagnostic fragments is indicated before and after digestion with LIG 3::4. (b) The selected target recognition sequence for the LIG 3::4 endonuclease aligned with the native target site recognized by the wild-type I-CreI (mismatched nucleotides are shown in red). The LIG 3 monomer binds to the left-hand side of the sequence (Lig 3), and the LIG 4 monomer binds to the right-hand side (Lig 4). (c) DNA–protein contacts of the wild-type I-CreI monomer and the predicted DNA–protein contacts for the re-designed I-CreI monomers LIG 3 and LIG 4. The amino acid substitutions are marked in red.

Download figure to PowerPoint

Verification of the re-designed I-CreI protein activity

The LIG 3::4 protein was expressed in Escherichia coli and partially purified using a C-terminal six-histidine tag. The enzyme was then evaluated in vitro by incubating varying concentrations of LIG 3::4 protein with a linearized plasmid substrate carrying the Lig34-site sequence (Figure 2a). Under these experimental conditions, LIG 3::4 was found to cleave its intended recognition site with activity comparable to that of wild-type I-CreI (Figure 2b). In addition, there was no detectable cleavage by LIG 3::4 of the wild-type I-CreI recognition site, nor was I-CreI able to cut the Lig34-site sequence. The LIG 3::4 endonuclease was next evaluated for its ability to cleave its intended recognition site in an extra-chromosomal recombination assay in Chinese hamster ovary (CHO) cells. The cells were co-transfected with a mammalian expression plasmid encoding LIG 3::4 and a reporter plasmid harboring a disrupted GFP reporter cassette in which the Lig34-site sequence was positioned between direct repeats of the GFP gene (Figure 2b). In this system, cleavage of the Lig34-site sequence by the LIG 3::4 endonuclease was expected to stimulate recombination between the GFP repeats to yield a functional GFP gene, and the number of GFP-positive cells could then be quantified by flow cytometry. To partially evaluate the cleavage specificity of LIG 3::4, we also used this reporter assay to determine whether or not LIG 3::4 is able to cleave ten potential ‘off-target’ sites found in the maize genome. These ten additional sites, which deviate from the Lig34-site by three or four base pairs each, represent the sites in the maize B73 genome that are the most similar to the Lig34-site. Using this assay, LIG 3::4 was found to result in approximately 7.7% of cells being GFP-positive when paired with its intended recognition site (Figure 2d). This represents an approximately 16-fold increase in the number of GFP-positive cells over background (approximately 0.48%) determined using an empty endonuclease expression vector. Wild-type I-CreI was found to result in approximately 13.4% of cells being GFP-positive, an approximately 32-fold increase over background (0.42%). LIG 3::4 did not result in an increase in the number of GFP-positive cells when paired with any of the ten potential off-target sites, suggesting that the engineered endonuclease is unlikely to cleave any of these sites in the maize genome with appreciable frequency. These results indicated that the LIG 3::4 endonuclease was able to specifically cleave an artificial Lig34-site sequence within a eukaryotic cell.

image

Figure 2.  Preliminary analyses of LIG 3::4. (a) In vitro target site cleavage by wild-type I-CreI and LIG-3::4 over 1 h at 37°C. Digests contained 8 nm target plasmid carrying an I-CreI or LIG 3::4 recognition site, with protein concentrations of 0, 3.13, 6.25, 12.5, 25, 50, 100, 200 or 400 nm (lanes 1–9) or 0 nm (lane A) or 400 nm (lane B). Digests with wild-type I-CreI are shown in the top gel, and those with LIG 3::4 are shown in the bottom gel. In each case, lanes 1–9 pair the endonuclease with its cognate recognition site, whereas lanes A and B pair non-cognate endonuclease and recognition site. (b) Endonuclease activity assay in a mammalian cell line. CHO cells were co-transfected with a LIG 3::4 or I-CreI expression vector as well as a reporter vector carrying a GFP gene interrupted by a potential endonuclease recognition site. In vivo cleavage of the GFP reporter cassette stimulates recombination between direct repeats of the GFP gene to yield a functional gene. The number of GFP-positive cells can then be determined by flow cytometry. (c) Sequence alignment of the LIG 3::4 recognition site as well as ten potential off-target sites found in the maize genome. Bases in the potential off-target site that deviate from the Lig34-site are shaded. (d) Data from the assay described in (b) showing the percentage of GFP-positive cells obtained when using LIG 3::4 (columns 1–11) or wild-type I-CreI (column 12) paired with the indicated recognition site. Values shown are corrected for background (<0.5% in each case) and transfection efficiency (39%).

Download figure to PowerPoint

Induction of mutations at the maize liguleless1 locus

The LIG 3::4 coding sequence was modified to include maize-preferred amino acid codons and put under the control of a maize ubiquitin promoter. The potato proteinase inhibitor II gene terminator was added to facilitate mRNA processing, and a ten amino acid SV40 nuclear localization signal (NLS) was fused to the N-terminus of the protein (MAPKKKRKVI). We delivered the LIG 3::4 expression cassette to maize cells by the Agrobacterium-mediated transformation procedure. A bar selectable marker cassette was incorporated into the T-DNA allowing selection of stably transformed cells on medium containing 1.5 mg L−1 Bialaphos. No selection was applied for mutations at the Lig34-site sequence.

A summary of the transformation results from two experiments is shown in Table 1. The first small-scale experiment provided preliminary observations on mutations induced by LIG 3::4. Two identified mutant T0 plants were designated 7663 and 7698. The second large-scale experiment produced 718 T0 plants. All together, 781 T0 transgenic plants were evaluated by PCR, and 23 T0 plants were identified that contained mutations at the liguleless1 locus based on visual screening of the Lig34-site PCR amplification products digested with the LIG 3::4 protein (Figure 3a). The PCR reactions were run twice for each T0 plant (independent sampling) in order to reduce the number of false positives (Figure 3b, lanes 1H04 and 6F01). Incomplete digestion of the PCR products by the LIG 3::4 protein could produce false positives, while those plants that were highly chimeric for Lig34-site mutations could escape detection. Thus, all selected events were subjected to an additional sequence analysis, and the number of recovered chimeric events should be considered as an approximation.

Table 1.   Transformation summary and initial screening results for induction of mutations at the liguleless1 locus
Vector PHP 34090Number of embryosNumber of T0 plantsNumber of plants analyzedMutations
TotalMonoallelicBiallelicChimeric
  1. *Only 100 T0 plants were analyzed in the first preliminary experiment. This number of T0 plants corresponds to 390 co-cultivated embryos.

Experiment 1 390*100100 4202
Experiment 23238718681196211
Total  781238213
image

Figure 3.  Illustration of the screening procedure for identification of T0 plants containing mutations at the lig34-site locus, and the lig34-site mutagenized sequences. (a) The PCR products were digested with the LIG 3::4 enzyme, producing two fragments if the wild-type Lig34-site sequences were not mutated. The screening method relied on visual identification of undigested PCR products (lanes B10, C07 and H10). It is likely that highly chimeric events escaped our selection procedure. (b) A set of selected T0 plants was re-evaluated for mutations at the lig34-site locus by another round of PCR reactions The ETX lanes represent digestion of the PCR products from non-transgenic maize plants. (c) The upper sequence in bold represents the original, wild-type Lig34-site. The 3′ overhang presumably produced by the LIG 3::4 protein is indicated in red. Insertions are marked in blue. Empty spaces indicate deletions, and underlined nucleotides indicate micro-homologies.

Download figure to PowerPoint

PCR products from the selected T0 plants were cloned and sequenced. Only wild-type Lig34-site sequences were found for two events, and those events were discarded (Table 2). We identified two events (3H10 and 4D05) that did not produce any wild-type PCR clones. These two events were pre-classified as biallelic. The other events, producing just one type of mutation, were considered as potential monoallelic events. Assuming that the cloning procedure was not biased towards particular mutations, a population of cloned PCR products should contain the same proportion of mutant and wild-type PCR sequences in the monoallelic events (= 0.50). We tested this hypothesis by calculating the z-score, which yields the value of the test statistic for the sample proportion (an estimate of the population proportion). We pre-classified events as monoallelic at significance level 0.05 (the z-score within ±1.645). The other events, including those yielding more than two mutations (7A06 and 8C05), were considered chimeric at Lig34-site.

Table 2.   The Lig34-site mutations in the sequenced PCR fragments
Sample nameSample sizeT0 mutationsSample proportionz-scoreStatus
ABCWT
  1. Random samples of the PCR amplification products were sequenced to determine the frequency of particular mutations. Sample proportion was arbitrary calculated using the number of type A mutations divided by the total number of sequenced PCR clones. Plants that contained more than two types of mutation were classified as chimeric. Sample proportions that were more than 1.645 standard deviations above 0.5 (i.e. the z-score exceeded 1.645) were also classified as chimeric.

01G0645110 340.24−3.4Chimeric
01H044600 460.00−6.8WT
02B014410 430.02−6.3Chimeric
02H044710 460.02−6.6Chimeric
03B104280 340.19−4.0Chimeric
03C0744190 250.43−0.9Monoallelic
03H10401822 01.006.3Bialleleic
04B032210 210.05−4.3Chimeric
04B0747121 340.28−3.1Chimeric
04D05672542 01.008.2Bialleleic
04E0148160 320.33−2.3Chimeric
05B1036150 210.42−1.0Monoallelic
05C0546190 270.41−1.2Monoallelic
06B074150 360.12−4.8Chimeric
06F014600 460.00−6.8WT
07A0643121390.09−5.3Chimeric
07D0523140 90.611.0Monoallelic
07G091870 110.39−0.9Monoallelic
07G112290 130.41−0.9Monoallelic
08B092419 140.42−0.8Monoallelic
08C0534111310.09−4.8Chimeric

Mutations at the maize liguleless1 locus

The structural features of the induced mutations are shown in Figure 3(c). Most mutations were centered on the Lig34-site sequence. Six of 27 sequences contained a 2 bp deletion of the presumptive 3′ overhang produced by cutting the Lig34-site sequence. Another seven mutations were produced by deletion of 7 bp from the upstream side of the expected cut site, with additional deletions of variable length on the other side of the break. The longest reported deletion in the second experiment (71 bp in the 7D05 event) also produced a PCR band that was clearly shorter than the undigested wild-type allele (four PCR bands in Figure 3b). Five sequenced PCR clones contained short insertions of undefined origin. Micro-homologies of various lengths were common at the ends of deletions (underlined nucleotides). The wide range of deletions and insertions encompassing the Lig34-site indicated that the DSBs were produced by the LIG 3::4 endonuclease and subsequently repaired by non-homologous end-joining.

LIG 3::4-induced mutations recovered in the T1 generation

Eight T0 plants were grown to maturity and back-crossed to non-transgenic plants. A sample of T1 progeny plants was tested for the presence of mutated lig34-site sequences (Table 3 and Figure 4). Representative T1 segregation data are shown for the 7698 event, which carries a 220 bp deletion at the Lig34-site (Figure 4a). As expected, the mutant lig34-site alleles segregated in the T1 population according to the Mendelian ratio (1:1) for a hemizygous single-locus gene (Table 3). The PCR products from the three T1-7698 back-crossed plants were cloned and sequenced. We found mutant lig34-site:wild-type Lig34-site ratios of 13:11, 10:13 and 9:13 in a random sample of sequenced clones. All mutations were 220 bp deletions as observed in the T0-7698 plant. The T1-7698 heterozygous (Lig34-site/lig34-site) plants were selfed in order to produce homozygous T2 plants with the 220 bp deletion at the lig34-site locus. Those plants did not show a liguleless phenotype, indicating that the introduced mutation did not knock out expression of the LG1 gene. Two biallelic lig34-site/lig34-site T0 plants (T0-3H10 and T0-4D05) also did not produce a liguleless phenotype.

Table 3.   Segregation of the lig34-site mutant alleles in the T1 back-crosses. The sample proportion value was calculated by dividing the number of T1 plants containing a mutant allele by the sample size (0.5 for monoallelic events; 1 for biallelic events)
Vector/eventNumber of plantsWTMutationsSample proportion T0 status
7663199102970.49Monoallelic
769817077930.55Monoallelic
7A06575700Chimeric
1G065933260.44Monoalleic
5B106033270.45Monoallelic
5C055927320.54Monoallelic
3H10580581.00Biallelic
4D05540541.00Biallelic
image

Figure 4.  Segregation of mutated alleles of the lig34-site locus in the T1 progeny. (a) PCR primers amplified two alleles of the lig34-site locus. One Lig34-site allele contained a 220 bp deletion in the T0-7698 plant, producing a shorter PCR amplification product (316 bp, compared to the wild-type 536 bp). The shorter alleles were found in approximately 50% of analyzed samples. (b) PCR-based screening for segregation of the mutant lig34-site sequences in the T1 back-crosses of T0-3H10, T0-5B10 and T0-7A06. The PCR amplification products were incubated with the LIG 3::4 protein and separated on an agarose gel. The biallelic 3H10 event produced undigested PCR fragments in all analyzed T1 samples. The mutant lig34-site allele segregated in the T1 back-cross progeny in the monoallelic 5B10 event, while the 7A06 event produced only restricted PCR fragments, indicating that this plant did not transmit its mutations to the next generation.

Download figure to PowerPoint

Mendelian segregation ratios indicative of heterozygous T0 plants were found in the T1 back-crossed plants from the T0-1G06, T0-5B10 and T0-5C05 events (Table 3 and Figure 4b). In each case, sequence analysis revealed only one type of mutated allele. Two events (T0-3H10 and T0-4D05) produced a population of T1 plants containing mutant lig34-site alleles in all tested T1 progeny plants (Table 3 and Figure 4b for the 3H10 event). Cloned and sequenced PCR fragments revealed 45 wild-type Lig34-site and 40 lig34-site mutant alleles in the T1-3H10 samples and 41 wild-type Lig34-site and 37 lig34-site mutant alleles in the T1-4D05 samples, indicating that the sampled T1 plants were heterozygous at the lig34-site locus. In each case, the sequences of the mutant alleles were identical to those observed in T0 plants. These results showed that the LIG 3::4 endonuclease was able to produce heritable, biallelic mutations at its intended target site. The T0-7A06 plant, originally classified as a chimeric event, did not transmit the lig34-site alleles to the next generation (Table 3 and Figure 4b). In summary, the segregation analysis of the T1 back-crossed populations validated our classification of events based on PCR and sequencing analysis of the T0 plants as described in Table 2.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We have demonstrated targeted mutagenesis at a native chromosomal site in a higher eukaryotic organism using an engineered homing endonuclease. Although numerous studies have addressed the issue of re-designing homing endonuclease DNA-binding and cleavage specificities (Seligman et al., 2002; Arnould et al., 2006; Ashworth et al., 2006; Rosen et al., 2006; Smith et al., 2006), this research has not been extended to targeted mutagenesis of chromosomal endogenous sites. A basic assumption of this study was that the re-designed I-CreI endonuclease would recognize the selected target site and introduce DSBs, resulting in erroneous repair (mutations). Although DSBs can be conservatively repaired by either re-ligation of the broken ends or homologous recombination, the frequency of such events is not high enough to mask the occurrence of DSBs at a particular chromosomal locus (Honma et al., 2007; Mansour et al., 2008). Therefore, the rate of mutagenesis can be used to assess the functionality of a re-designed endonuclease and to evaluate the feasibility of targeted mutagenesis at the selected chromosomal locus.

Members of the LAGLIDADG family of homing endonucleases, such as yeast mitochondrial I-SceI, Chlamydomonas reinhardti chloroplast I-CreI, Monomastix sp. I-MsoI and archaeal I-DmoI, have been the subjects of intensive structural studies and re-engineering efforts, leading to the production of protein variants with altered DNA-binding specificities (Guhan and Muniyappa, 2003; Silva and Belfort, 2004; Ashworth et al., 2006; Moure et al., 2008; Prieto et al., 2008). We produced an endonuclease to cleave the Lig34-site using a method that is free of genetic selection or screening. This method comprised three steps. First, the cleavage-site specificity of a pair of I-CreI monomers was modified by introducing amino acid substitutions at the protein–DNA interface. Two of these amino acid substitutions (Y33C in LIG 3 and N30R in LIG 4) have been described previously (Seligman et al., 2002; Rosen et al., 2006). The remaining specificity-altering amino acid substitutions (I24K, Q26S, S32C, Q38E, S40R, T42E, Q44T, R68Y, I77E and I77R) were predicted based on visual inspection of the I-CreI–DNA co-crystal structure (Chevalier et al., 2003). The second step in the endonuclease re-design process was the introduction of the E80Q mutation into both monomers. This mutation has been shown to significantly increase the DNA cleavage activity of engineered I-CreI endonucleases, and can compensate for cleavage activity that is lost as a consequence of re-designing the protein–DNA interface (unpublished results). The third and final step of our re-design process was the joining of the LIG 3 and LIG 4 monomers into a single-chain endonuclease using a 38 amino acid linker. This step served two important functions: first, it forced the two engineered LIG subunits to ‘dimerize’ with one another rather than forming homodimers that may cut at unintended sites; second, it greatly simplified transformation experiments by enabling the functional ‘heterodimeric’ endonuclease to be expressed from a single gene (unpublished results). A similar design (32 amino acid linker) was successfully used for producing genetic modifications at a SCID gene in 293H human cells (Grizot et al., 2009).

It is important to note that we did not observe obvious or severe toxicity due to LIG 3::4 expression in transgenic maize plants. Toxicity is frequently cited as a limitation of ZFNs (Cornu et al., 2008; Meng et al., 2008). Wild-type I-CreI has also been shown to be toxic to a wide range of organisms (Seligman et al., 2002; Maggert and Golic, 2005). The reported 22% transformation efficiency with the LIG 3::4 expression cassettes (number of T0 plants selected or regenerated/number of embryos treated) is standard for many other transgenes used under similar transformation conditions, including I-SceI (unpublished results). However, there is a possibility that selection against transformants that express a high level of endonuclease would reduce the efficiency of mutagenesis if the specificity of the LIG 3::4 was compromised. Also, selection against stable LIG 3::4 expression might have contributed to the large proportion of events identified that were chimeric for the Lig34-site mutations. The extent to which a potential change in the specificity and catalytic activity of LIG 3::4 might contribute to the observed efficiency of targeted mutagenesis in maize remains to be investigated.

We found mutations at Lig34-site in 19 out of 681 transgenic T0 maize plants, two of which contained two independently mutagenized alleles. The repair of many DSBs resulted in relatively short deletions; only five insertions were identified, four of which were accompanied by deletions. Similar outcomes of DSB repair have been reported for other organisms and other endonucleases including the ZFNs (Bibikova et al., 2002; Lloyd et al., 2005; Morton et al., 2006; Beumer et al., 2008; Mansour et al., 2008; Santiago et al., 2008). For example, 22–52 bp deletions were predominant in the products of DSB repair in Arabidopsis (Lloyd et al., 2005). It appears that the DSB repair pathways in maize produce mutations by non-homologous end-joining, a DNA repair pathway that is ubiquitous in many higher eukaryotic organisms (Honma et al., 2007). Micro-homologies found at the ends of deletions are another common feature of DNA repair by non-homologous end-joining (Rebuzzini et al., 2005). Similar products of DSB repair have been found in maize cells resulting from DSB induction by I-SceI or ZFNs (Yang et al., 2009; Shukla et al., 2009).

The frequency of mutagenesis identified here (approximately 3%) using an engineered homing endonuclease is comparable to, but slightly lower than, that found for similar experiments performed with ZFNs. Maeder et al. (2008) found a 5% mutation frequency at the SuRA locus induced by the ZFN SR2163. Heat-inducible expression of a ZFN in Arabidopsis produced approximately 7% mutated progeny seedlings with an artificial target site (Lloyd et al., 2005). These results are comparable with those obtained by direct Drosophila embryo injections with ZFN mRNA, which resulted in approximately 5% mutants at the coli target site (Beumer et al., 2008), or the targeted DHFR (dihyrofolate reductase) gene knockout at a frequency of 7% in Chinese hamster ovary cells (Santiago et al., 2008). The latter experiments also produced biallelic gene disruption at a frequency of approximately 1%.

Approximately 50% (10/23) of the mutant T0 plants contained monoallelic and biallelic mutations at the lig34-site locus, indicating that the mutation occurred very early in the development of the transformed plant. This could be an advantage for targeted transgene insertion if the endonuclease is actively producing DSBs at the target locus concurrently with transgene delivery. Also, the identified biallelic events proved that two LIG34-site alleles were available for restriction (the target site has the same sequence in both alleles), but the wild-type alleles were still found to be transmitted to the T1 generation in monoallelic events. The fact that we found no new mutations in the analyzed T1 plants indicate that de novo LIG 3::4-induced mutagenesis in planta may be an exception rather than a rule. It appears that a burst of transient endonuclease activity played a significant role in inducing mutations at the LIG 3::4 target sites. This conclusion is consistent with an increasingly popular delivery of ZFNs by micro-injection of mRNA (Beumer et al., 2008; Doyon et al., 2008; Meng et al., 2008; Foley et al., 2009). The pattern of DSB-inducing activity of endonucleases during development of transgenic organisms requires separate research that is beyond the scope of this study.

In summary, while more studies on targeted modifications of chromosomal loci by re-designed homing endonucleases are anticipated, the results presented here suggest that this approach holds great promise. Endonuclease-targeted genome modification in crop species, or any other species for that matter, could greatly accelerate production of transgenic organisms, as well as enabling genetic modifications that are not readily achievable using conventional breeding techniques.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Synthesis of the LIG 3::4 endonuclease

The LIG 3::4 endonuclease was produced by independently mutagenizing a pair of I-CreI monomers by PCR to produce LIG 3 and LIG 4. To produce LIG 3, we introduced five amino acid changes: I24K, Y33C, R68Y, I77E and E80Q. To produce LIG 4, we introduced ten amino acid changes: Q26S, N30R, S32C, Q38E, S40R, T42E, Q44T, R68Y, I77R and E80Q. The LIG 3 and LIG 4 monomers were then joined into a single polypeptide using a 38 amino acid linker with the sequence PGSVGGLSPSQASSAASSASSSPGSGISEALRAGATKS, which joined the N-terminal LIG 3 monomer truncated at L155 to the C-terminal LIG 4 monomer starting at K7. For bacterial expression, a six-histidine tag was added to the C-terminus of the fusion protein. For all experiments in human cell lines and plants, the histidine tag was removed and an SV40 nuclear localization signal (sequence MAPKKKRKVI) was added to the N-terminus of the fusion protein.

Protein purification and in vitro endonuclease assay

The coding sequences for LIG 3::4 and wild-type I-CreI were sub-cloned into a bacterial expression vector (pET-21a, Novagen, http://www.merck-chemicals.com/life-sciences-research). Both genes carried a C-terminal six-histidine tag to facilitate purification. BL21(DE3) cells were transformed with each plasmid and cultured on standard 2× YT medium containing 200 μg ml−1 ampicillin. Protein expression was induced by addition of 1 mm IPTG after reducing the growth temperature from 37 to 22°C. Three hours after induction, the cells were pelleted by centrifugation for 10 min at 6000 g, and the pellets were resuspended in 1 ml binding buffer (20 mm Tris/HCl, pH 8.0, 500 mm NaCl, 10 mm imidazole) by vortexing. The cells were disrupted using 12 pulses of sonication (50% power), and the cell debris was pelleted by centrifugation for 15 min at 14 000 g. The cell supernatant was diluted in 4 ml binding buffer and loaded onto a 200 μl nickel-charged metal-chelating Sepharose column. The column was washed with 4 ml wash buffer (20 mm Tris/HCl, pH 8.0, 500 mm NaCl, 60 mm imidazole) and then 0.2 ml elution buffer (20 mm Tris/HCl, pH 8.0, 500 mm NaCl, 400 mm imidazole). The enzymes were eluted in 0.6 ml elution buffer and concentrated to 50–130 μl using Vivaspin disposable concentrators (ISC, http://www.bioexpress.com). The enzymes were exchanged into SA buffer (25 mm Tris/HCl, pH 8.0, 100 mm NaCl, 5 mm MgCl2, 5 mm EDTA) for assays and storage using Zeba spin desalting columns (Thermo Scientific, http://www.thermo.com). The purity and molecular weight of the enzymes were then confirmed by MALDI-TOF mass spectrometry.

For in vitro cleavage assays, 25 pmol of a pUC19 plasmid harboring the meganuclease recognition sequence was linearized using XmnI and then incubated with the indicated concentration of purified meganuclease for 1 h at 37°C in 10 mm Tris, pH 8.0, 50 mm NaCl, 10 mm MgCl2. Reactions were stopped by addition of 0.5% SDS, 25 mm EDTA and 10 μg Proteinase K (New England Biolabs, http://www.neb.com). After an additional 1 h incubation at 37°C, plasmid digestions were separated by gel electrophoresis, and the cut and uncut DNA bands were quantified using the ImageJ program (http://rsbweb.nih.gov/ij).

DNA break/repair assay in a human cell line

A defective GFP reporter cassette was generated by first cloning a 5′ 480 bp fragment of the GFP gene into NheI/HindIII-digested pcDNA5/FRT (Invitrogen, http://www.invitrogen.com/) resulting in the plasmid pGF. Next, a 3′ 480 bp fragment of the GFP gene (including a 240 bp sequence duplicated in the 5′ 480 bp fragment) was cloned into BamHI/XhoI-digested pGF. The resulting plasmid, pGFFP, consists of the 5′ two-thirds of the GFP gene followed by the 3′ two-thirds of the GFP gene, interrupted by 24 bp of the pcDNA5/FRT polylinker. To insert the LIG 3::4 binding site LIG34-site, the I-CreI site or the ten potential off-target sites into pGFFP, complementary oligonucleotides encoding the binding site were annealed and ligated into HindIII/BamHI-digested pGFFP.

The coding sequences of LIG 3::4 and wild-type I-CreI were inserted into to the mammalian expression vector pCI (Promega, http://www.promega.com/) under the control of a constitutive (CMV) promoter. Chinese hamster ovary (CHO) cells at approximately 90% confluence were transfected in 96-well plates with 150 ng pGFFP reporter plasmid and 50 ng pCI LIG 3::4, 50 ng pCI I-CreI, or, to determine background, 50 ng of empty pCI, using Lipofectamine 2000 according to the manufacturer’s instructions (Invitrogen). To determine transfection efficiency, CHO cells were transfected with 200 ng pCI GFP. Cells were washed in PBS 24 h post-transfection, trypsinized and resuspended in PBS supplemented with 3% fetal bovine serum. Cells were assayed for GFP activity using a Cell Lab Quanta SC MPL flow cytometer and the accompanying Cell Lab Quanta analysis software (Beckman Coulter, http://www.beckmancoulter.com). The values presented in Figure 2(d) were calculated by subtracting the number of GFP-positive cells produced in background transfections using the empty pCI vector from the number of GFP-positive cells produced in transfections with pCI LIG3::4 or pCI I-CreI. This value was then divided by the transfection efficiency (0.39). All transfections were performed in triplicate and standard deviations are shown.

Plant LIG 3::4 expression vector

The coding sequence of the LIG 3::4 gene was excised from pUC19 and ligated into a vector containing the maize ubiquitin-1 promoter and the first exon and intron (Christensen and Quail, 1996). The potato pinII terminator (An et al., 1989) was ligated to the end of the coding sequence. The maize expression cassette was then transferred into the pSB11 plasmid backbone as a HindIII/NotI fragment (Komari et al., 1996). In this plasmid, which contains T-DNA borders, the BAR gene driven by the CaMV 35S promoter was used as a transformation selection marker. The resulting vector was then co-integrated into the super-binary vector pSB1 inAgrobacterium tumefaciens strain LBA4404 (Komari et al., 1996) by electroporation using a Bio-Rad Gene Pulser II set at 2.0 kV and 25 μF, with a resistance between 200 and 600 Ω.

Production of transgenic plants

Zea mays ETX immature embryos were transformed by a modified Agrobacterium-mediated transformation procedure as described by Djukanovic et al. (2006). The day before transformation, Agrobacterium was transferred onto medium containing 5.0 g L−1 yeast extract, 10.0 g L−1 peptone, 5.0 g L−1 sodium chloride, 15.0 g L−1 Bacto-Agar (VWR, http://www.vwrsp.com) and 50.0 mg L−1 spectinomycin, and grown overnight at 30°C. On the day of transformation, 10–11-day-old immature embryos (1.3–1.8 mm) were dissected from sterilized kernels and placed into 2 ml of liquid medium [4.0 g L−1 N6 basal salts (C-1416; Sigma-Aldrich Co., http://www.sigmaaldrich.com/), 1.0 ml L−1 Eriksson’s vitamin mix (E-1511; Sigma-Aldrich Co.), 1.0 mg L−1 thiamine HCl, 1.5 mg L−1 2,4-dichlorophenoxyacetic acid (2,4-D), 0.690 g L−1l-proline, 68.5 g L−1 sucrose, 36.0 g L−1 glucose, pH 5.2]. After embryo isolation, the infection medium was prepared by scraping up a sterile loop-full of the Agrobacterium streaked out the previous night and resuspending it in fresh liquid medium. The optical density was checked using a spectrophotometer set at 550 nm, and the suspension was diluted to an optical density of 0.175. The embryo-containing medium was replaced with 1 ml of the Agrobacterium suspension, and the embryos were incubated for 5 min at room temperature. After incubating, the Agrobacterium suspension was removed and the embryos (40 embryos/plate) were transferred, embryo axis down, onto a plate containing 4.0 g L−1 N6 basal salts, 1.0 ml L−1 Eriksson’s vitamin mix, 1.0 mg L−1 thiamine HCl, 1.5 mg L−1 2,4-D, 0.690 g L−1l-proline, 30.0 g L−1 sucrose, 0.85 mg L−1 silver nitrate, 0.1 nm acetosyringone and 3.0 g L−1 Gelrite (Sigma-Aldrich Co.), pH 5.8. Embryos were incubatefd in the dark for 3–4 days at 21°C, and then transferred to medium containing 4.0 g L−1 N6 basal salts, 1.0 ml L−1 Eriksson’s vitamin mix, 0.5 mg L−1 thiamine HCl, 1.5 mg L−1 2,4-D, 0.690 g L−1l-proline, 30.0 g L−1 sucrose, 0.5 g L−1 MES buffer, 0.85 mg L−1 silver nitrate, 100 mg L−1 carbenicillin and 8 g L−1 agar (Sigma) for an additional 4 days of incubation in the dark at 28°C. The embryos were then transferred (19 embryos/plate) onto new plates containing 4.0 g L−1 N6 basal salts, 1.0 ml L−1 Eriksson’s vitamin mix, 0.5 mg L−1 thiamine HCl, 1.5 mg L−1 2,4-D, 0.69 g L−1l-proline, 30.0 g L−1 sucrose, 0.5 g L−1 MES buffer, 0.85 mg L−1 silver nitrate, 1.5 mg L−1 Bialaphos, 100 mg L−1 carbenicillin, 8.0 g L−1 agar, pH 5.8, and placed in the dark at 28°C. After 3 weeks, the responding calli (7 calli/plate) was sub-cultured onto medium containing 4.0 g L−1 N6 basal salts, 1.0 ml L−1 Eriksson’s vitamin mix, 0.5 mg L−1 thiamine HCl, 1.5 mg L−1 2,4-D, 0.69 g L−1l-proline, 30.0 g L−1 sucrose, 0.5 g L−1 MES buffer, 0.85 mg L−1 silver nitrate, 3.0 mg L−1 Bialaphos, 100 mg L−1 carbenicillin, 8.0 g L−1 agar, pH 5.8. After 5 weeks in the dark at 28°C, somatic embryogenesis was induced by transferring a small amount of selected tissue onto regeneration medium (one transgenic event/plate) containing 4.3 g L−1 Murashige & Skoog (MS) salts (M 524, Phytotechnology Labs), 5.0 ml L−1 MS vitamins stock solution (M3900; Sigma-Aldrich Co.), 100 mg L−1myo-inositol, 0.1 μm abscisic acid, 1 mg L−1 indoleacetic acid, 0.5 mg L−1 zeatin, 60.0 g L−1 sucrose, 3.0 mg L−1 Bialaphos (purified from Herbiace herbicide), 100 mg L−1 carbenicillin, 6.0 g L−1 Ultrapure Agar (VWR), pH 5.6. The plates were incubated in the dark for 2–3 weeks at 28°C. All material with visible shoots and roots was transferred (5–10 shoots per event/plate) onto medium containing 4.3 g L−1 MS salts (Gibco 11117), 5.0 ml L−1 MS vitamins stock solution, 100 mg L−1myo-inositol, 40.0 g L−1 sucrose, 3 mg L−1 Bialophos, 100 mg L−1 Benomyl (Sigma-Aldrich Co.), 6.0 g L−1 Bacto-Agar, pH 5.6, and incubated under artificial light at 28°C. One week later, plantlets were moved into glass tubes (one plantlet per tube) containing the same medium minus the Bialophos, and grown under artificial light until sampling and/or transplantation into soil.

Polymerase chain reactions

DNA was extracted by placing two leaf punches, one stainless steel bead and 450 μl of extraction buffer (250 mm NaCl, 200 mm Tris/HCl pH 7.4, 25 mm EDTA, 4.2 m guanidine HCl) into each tube of a Mega titer rack (catalog number 40002-012; VWR). The samples were homogenized using a Genogrinder (SPEX SamplePrep 2000, Geno/Grinder, http://www.btc-bti.com) at 1650 rpm for 1 min, followed by centrifugation for 20 min at 2750 g. Supernatant (300 μl) was pipetted into wells of a GF/F filter plate (catalog number 7700-3310; Whatman Inc., http://www.whatman.com), and washed with 100 μl of extraction buffer followed by 200 μl of wash buffer applied twice (50 mm Tris/HCl, pH 7.4, 200 mm NaCl, 70% EtOH according to manufacturer's instructions). The plate was centrifuged for 10 min at 2750 g, and DNA was eluted with 100 μl of water by centrifugation at 2000 g for 1 min.

PCR reactions contained 100 ng DNA template, 5 pmol of each primer (forward: TAATTAGGGAGAGAAAAATAGAGCACC; reverse: ATGTGCATTGCATCGCTCTTCTCT), 10 μl of 2× RedExtract-N-AmpPCR mix (Sigma-Aldrich Co.) in a total volume of 20 μl. The initial incubation was at 94°C for 3 min, followed by 35 cycles at 95°C for 30 sec, 63°C for 30 sec and 72°C for 30 sec. Aliquots (5 μl) of the PCR reaction mix were digested with the LIG 3::4 endonuclease (25 mm Tris/HCl, pH 8.0, 100 mm NaCl, 5 mm MgCl2, 5 mm EDTA, 7 μm LIG 3::4 protein) at 37°C for 8 h. SDS at 0.1% concentration was added to the enzyme digestion mixture before loading samples on a 2% agarose gel stained with ethidium bromide. The uncut PCR fragments were purified using a Qiagen gel extraction kit (catalog number 28704, Qiagen, http://www.qiagen.com/), and cloned into the TOPO TA vector (Invitrogen) for sequencing.

Sequencing

DNA sequencing was performed by BigDye Terminator chemistry on ABI 3700 capillary sequencing machines (Applied Biosystems, http://www.appliedbiosystems.com/). Each sample contained 0.4–0.5 μg plasmid DNA or approximately 100 ng of the PCR products and 6.4 pmol primer.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

The authors acknowledge Carl Simmons’ (Pioneer Hi-Bred International, Department of Trait Charaterization and Development) contribution to this work in designing the maize codon-optimized I-CreI. We thank Jessica Stagg for her excellent technical assistance. Novel materials described in this publication may be available for non-commercial research purposes upon acceptance and signing of a material transfer agreement. In some cases, such materials may contain or be derived from materials obtained from a third party. In such cases, distribution of material will be subject to obtaining the requisite permission from any third-party owners, licensors or controllers of all or parts of the material. Plant germplasm and transgenic material will not be made available except at the discretion of the owner and then only in accordance with all applicable governmental regulations. Obtaining any permissions will be the sole responsibility of the person requesting the material.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • An, G., Mitra, A., Choi, H.K., Costa, M.A., An, K., Thornburg, R.W. and Ryan, C.A. (1989) Functional analysis of the 3′ control region of the potato wound-inducible proteinase inhibitor II gene. Plant Cell, 1, 115122.
  • Arnould, S., Chames, P., Perez, C. et al. (2006) Engineering of large numbers of highly specific homing endonucleases that induce recombination on novel DNA targets. J. Mol. Biol. 355, 443458.
  • Arnould, S., Perez, C., Cabaniols, J.P., Smith, J., Gouble, A., Grizot, S., Epinat, J.C., Duclert, A., Duchateau, P. and Paques, F. (2007) Engineered I–CreI derivatives cleaving sequences from the human XPC gene can induce highly efficient gene correction in mammalian cells. J. Mol. Biol. 371, 4965.
  • Ashworth, J., Havranek, J.J., Duarte, C.M., Sussman, D., Monnat, R.J. Jr, Stoddard, B.L. and Baker, D. (2006) Computational redesign of endonuclease DNA binding and cleavage specificity. Nature 441, 656659.
  • Beumer, K.J., Trautman, J.K., Bozas, A., Liu, J.L., Rutter, J., Gall, J.G. and Carroll, D. (2008) Efficient gene targeting in Drosophila by direct embryo injection with zinc-finger nucleases. Proc. Natl Acad. Sci. USA, 105, 1982119826.
  • Bibikova, M., Golic, M., Golic, K.G. and Carroll, D. (2002) Targeted chromosomal cleavage and mutagenesis in Drosophila using zinc-finger nucleases. Genetics 161, 11691175.
  • Bibikova, M., Beumer, K., Trautman, J.K. and Carroll, D. (2003) Enhancing gene targeting with designed zinc finger nucleases. Science 300, 764.
  • Cai, C.Q., Doyon, Y., Ainley, W.M. et al. (2009) Targeted transgene integration in plant cells using designed zinc finger nucleases. Plant Mol. Biol. 69, 699709.
  • Carroll, D. (2008) Progress and prospects: zinc-finger nucleases as gene therapy agents. Gene Ther. 15, 14631468.
  • Chevalier, B., Turmel, M., Lemieux, C., Monnat, R.J. Jr, and Stoddard, B.L. (2003) Flexible DNA target site recognition by divergent homing endonuclease isoschizomers I CreI and I MsoI. J. Mol. Biol. 329, 253269.
  • Christensen, A.H. and Quail, P.H. (1996) Ubiquitin promoter-based vectors for high-level expression of selectable and/or screenable marker genes in monocotyledonous plants. Transgenic Res. 5, 213218.
  • Cornu, T.I., Thibodeau-Beganny, S., Guhl, E., Alwin, S., Eichtinger, M., Joung, J.K. and Cathomen, T. (2008) DNA-binding specificity is a major determinant of the activity and toxicity of zinc-finger nucleases. Mol. Ther. 16, 352358.
  • D’Halluin, K., Vanderstraeten, C., Stals, E., Cornelissen, M. and Ruiter, R. (2008) Homologous recombination: a basis for targeted genome optimization in crop species such as maize. Plant Biotechnol. J. 6, 93102.
  • Djukanovic, V., Orczyk, W., Gao, H., Sun, X., Garrett, N., Zhen, S., Gordon-Kamm, W., Barton, J. and Lyznik, L.A. (2006) Gene conversion in transgenic maize plants expressing FLP/FRT and Cre/loxP site-specific recombination systems. Plant Biotechnol. J. 4, 345357.
  • Doyon, Y., McCammon, J.M., Miller, J.C. et al. (2008) Heritable targeted gene disruption in zebrafish using designed zinc-finger nucleases. Nat. Biotechnol. 26, 702708.
  • Durrenberger, F. and Rochaix, J.D. (1993) Characterization of the cleavage site and the recognition sequence of the I CreI DNA endonuclease encoded by the chloroplast ribosomal intron of Chlamydomonas reinhardtii. Mol. Gen. Genet. 236, 409414.
  • Fajardo-Sanchez, E., Stricher, F., Paques, F., Isalan, M. and Serrano, L. (2008) Computer design of obligate heterodimer meganucleases allows efficient cutting of custom DNA sequences. Nucleic Acids Res. 36, 21632173.
  • Foley, J.E., Yeh, J.R., Maeder, M.L., Reyon, D., Sander, J.D., Peterson, R.T. and Joung, J.K. (2009) Rapid mutation of endogenous zebrafish genes using zinc finger nucleases made by Oligomerized Pool ENgineering (OPEN). PLoS ONE, 4, e4348.
  • Grizot, S., Smith, J., Daboussi, F. et al. (2009) Efficient targeting of a SCID gene by an engineered single-chain homing endonuclease. Nucleic Acids Res. 37, 54055419.
  • Guhan, N. and Muniyappa, K. (2003) Structural and functional characteristics of homing endonucleases. Crit. Rev. Biochem. Mol. Biol. 38, 199248.
  • Honma, M., Sakuraba, M., Koizumi, T., Takashima, Y., Sakamoto, H. and Hayashi, M. (2007) Non-homologous end-joining for repairing I SceI-induced DNA double strand breaks in human cells. DNA Repair (Amst.), 6, 781788.
  • Kim, Y.G., Cha, J. and Chandrasegaran, S. (1996) Hybrid restriction enzymes: zinc finger fusions to Fok I cleavage domain. Proc. Natl. Acad.Sci. USA, 93, 11561160.
  • Komari, T., Hiei, Y., Saito, Y., Murai, N. and Kumashiro, T. (1996) Vectors carrying two separate T DNAs for co-transformation of higher plants mediated by Agrobacterium tumefaciens and segregation of transformants free from selection markers. Plant J. 10, 165174.
  • Lloyd, A., Plaisier, C.L., Carroll, D. and Drews, G.N. (2005) Targeted mutagenesis using zinc-finger nucleases in Arabidopsis. Proc. Natl Acad. Sci. USA, 102, 22322237.
  • Maeder, M.L., Thibodeau-Beganny, S., Osiak, A. et al. (2008) Rapid ‘open-source’ engineering of customized zinc-finger nucleases for highly efficient gene modification. Mol. Cell, 31, 294301.
  • Maggert, K.A. and Golic, K.G. (2005) Highly efficient sex chromosome interchanges produced by I CreI expression in Drosophila. Genetics 171, 11031114.
  • Mansour, W.Y., Schumacher, S., Rosskopf, R., Rhein, T., Schmidt-Petersen, F., Gatzemeier, F., Haag, F., Borgmann, K., Willers, H. and Dahm-Daphi, J. (2008) Hierarchy of nonhomologous end-joining, single-strand annealing and gene conversion at site-directed DNA double-strand breaks. Nucleic Acids Res. 36, 40884098.
  • Meng, X., Noyes, M.B., Zhu, L.J., Lawson, N.D. and Wolfe, S.A. (2008) Targeted gene inactivation in zebrafish using engineered zinc-finger nucleases. Nat. Biotechnol. 26, 695701.
  • Moehle, E.A., Rock, J.M., Lee, Y.L., Jouvenot, Y., DeKelver, R.C., Gregory, P.D., Urnov, F.D. and Holmes, M.C. (2007) Targeted gene addition into a specified location in the human genome using designed zinc finger nucleases. Proc. Natl Acad. Sci. USA, 104, 30553060.
  • Morton, J., Davis, M.W., Jorgensen, E.M. and Carroll, D. (2006) Induction and repair of zinc-finger nuclease-targeted double-strand breaks in Caenorhabditis elegans somatic cells. Proc. Natl Acad. Sci. USA, 103, 1637016375.
  • Moure, C.M., Gimble, F.S. and Quiocho, F.A. (2008) Crystal structures of I SceI complexed to nicked DNA substrates: snapshots of intermediates along the DNA cleavage reaction pathway. Nucleic Acids Res. 36, 32873296.
  • Orel, N., Kyryk, A. and Puchta, H. (2003) Different pathways of homologous recombination are used for the repair of double-strand breaks within tandemly arranged sequences in the plant genome. Plant J. 35, 604612.
  • Paques, F. and Duchateau, P. (2007) Meganucleases and DNA double-strand break-induced recombination: perspectives for gene therapy. Curr. Gene Ther. 7, 4966.
  • Porteus, M.H. and Carroll, D. (2005) Gene targeting using zinc finger nucleases. Nat. Biotechnol. 23, 967973.
  • Prieto, J., Redondo, P., Padro, D., Arnould, S., Epinat, J.C., Paques, F., Blanco, F.J. and Montoya, G. (2007) The C terminal loop of the homing endonuclease I CreI is essential for site recognition, DNA binding and cleavage. Nucleic Acids Res. 35, 32623271.
  • Prieto, J., Epinat, J.C., Redondo, P., Ramos, E., Padro, D., Cedrone, F., Montoya, G., Paques, F. and Blanco, F.J. (2008) Generation and analysis of mesophilic variants of the thermostable archaeal I DmoI homing endonuclease. J. Biol. Chem. 283, 43644374.
  • Puchta, H. (2005) The repair of double-strand breaks in plants: mechanisms and consequences for genome evolution. J. Exp. Bot. 56, 114.
  • Rebuzzini, P., Khoriauli, L., Azzalin, C.M., Magnani, E., Mondello, C. and Giulotto, E. (2005) New mammalian cellular systems to study mutations introduced at the break site by non-homologous end-joining. DNA Repair (Amst.), 4, 546555.
  • Redondo, P., Prieto, J., Munoz, I.G. et al. (2008) Molecular basis of xeroderma pigmentosum group C DNA recognition by engineered meganucleases. Nature 456, 107111.
  • Rosen, L.E., Morrison, H.A., Masri, S., Brown, M.J., Springstubb, B., Sussman, D., Stoddard, B.L. and Seligman, L.M. (2006) Homing endonuclease I CreI derivatives with novel DNA target specificities. Nucleic Acids Res. 34, 47914800.
  • Salomon, S. and Puchta, H. (1998) Capture of genomic and T DNA sequences during double-strand break repair in somatic plant cells. EMBO J. 17, 60866095.
  • Santiago, Y., Chan, E., Liu, P.Q. et al. (2008) Targeted gene knockout in mammalian cells by using engineered zinc-finger nucleases. Proc. Natl Acad. Sci. USA, 105, 58095814.
  • Seligman, L.M., Chisholm, K.M., Chevalier, B.S., Chadsey, M.S., Edwards, S.T., Savage, J.H. and Veillet, A.L. (2002) Mutations altering the cleavage specificity of a homing endonuclease. Nucleic Acids Res. 30, 38703879.
  • Shukla, V.K., Doyon, Y., Miller, J.C. et al. (2009) Precise genome modification in the crop species Zea mays using zinc-finger nucleases. Nature 459, 437441.
  • Silva, G.H. and Belfort, M. (2004) Analysis of the LAGLIDADG interface of the monomeric homing endonuclease I DmoI. Nucleic Acids Res. 32, 31563168.
  • Smith, J., Grizot, S., Arnould, S. et al. (2006) A combinatorial approach to create artificial homing endonucleases cleaving chosen sequences. Nucleic Acids Res. 34, e149.
  • Stephens, K.M., Monnat, R.J. Jr, Heath, P.J. and Stoddard, B.L. (1997) Crystallization and preliminary X ray studies of I CreI: a group I intron-encoded endonuclease from C. reinhardtii. Proteins, 28, 137139.
  • Stoddard, B.L. (2005) Homing endonuclease structure and function. Q. Rev. Biophys. 38, 4995.
  • Szczepek, M., Brondani, V., Buchel, J., Serrano, L., Segal, D.J. and Cathomen, T. (2007) Structure-based redesign of the dimerization interface reduces the toxicity of zinc-finger nucleases. Nat. Biotechnol. 25, 786793.
  • Thompson, A.J., Yuan, X., Kudlicki, W. and Herrin, D.L. (1992) Cleavage and recognition pattern of a double-strand-specific endonuclease (I creI) encoded by the chloroplast 23S rRNA intron of Chlamydomonas reinhardtii. Gene 119, 247251.
  • Townsend, J.A., Wright, D.A., Winfrey, R.J., Fu, F., Maeder, M.L., Joung, J.K. and Voytas, D.F. (2009) High-frequency modification of plant genes using engineered zinc-finger nucleases. Nature 459, 442445.
  • Wright, D.A., Townsend, J.A., Winfrey, R.J. Jr, Irwin, P.A., Rajagopal, J., Lonosky, P.M., Hall, B.D., Jondle, M.D. and Voytas, D.F. (2005) High-frequency homologous recombination in plants mediated by zinc-finger nucleases. Plant J. 44, 693705.
  • Yang, M., Djukanovic, V., Stagg, J., Lenderts, B., Bidney, D., Falco, S.C. and Lyznik, L.A. (2009) Targeted mutagenesis in the progeny of maize transgenic plants. Plant Mol. Biol. 70, 669779.