Trinucleotide repeat expansions are responsible for more than two dozens severe neurological disorders in humans. A double-strand break between two short CAG/CTG trinucleotide repeats was formerly shown to induce a high frequency of repeat contractions in yeast. Here, using a dedicated TALEN, we show that induction of a double-strand break into a CAG/CTG trinucleotide repeat in heterozygous yeast diploid cells results in gene conversion of the repeat tract with near 100% efficacy, deleting the repeat tract. Induction of the same TALEN in homozygous yeast diploids leads to contractions of both repeats to a final length of 3–13 triplets, with 100% efficacy in cells that survived the double-strand breaks. Whole-genome sequencing of surviving yeast cells shows that the TALEN does not increase mutation rate. No other CAG/CTG repeat of the yeast genome showed any length alteration or mutation. No large genomic rearrangement such as aneuploidy, segmental duplication or translocation was detected. It is the first demonstration that induction of a TALEN in an eukaryotic cell leads to shortening of trinucleotide repeat tracts to lengths below pathological thresholds in humans, with 100% efficacy and very high specificity.
Citation: Richard G-F, Viterbo D, Khanna V, Mosbach V, Castelain L, Dujon B (2014) Highly Specific Contractions of a Single CAG/CTG Trinucleotide Repeat by TALEN in Yeast. PLoS ONE 9(4): e95611. doi:10.1371/journal.pone.0095611
Editor: Tetsuo Ashizawa, University of Florida, United States of America
Received: November 13, 2013; Accepted: March 28, 2014; Published: April 18, 2014
Copyright: © 2014 Richard et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the Association pour la Recherche contre le Cancer (ARC) to G.-F. Richard (grant number: SFI20111203641) and by ANR blanc 2011 (DYGEVO) to B. Dujon. V. Mosbach is the recipient of a doctoral fellowship from the Fondation pour la Recherche Médicale (FRM) (grant number: PLP20131028794). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Trinucleotide repeat expansions are involved in at least two dozens dramatic neurological and developmental disorders in human , , , , . A large amount of studies were devoted to understanding the mechanisms responsible for large CAG/CTG repeat expansions, using model systems as diverse as bacteria , , yeast , , , drosophila , mice , , ,  or human cell lines , , . Over the last 20 years or so, it was demonstrated that replication slippage, double-strand break repair, base excision repair, nucleotide excision repair, mismatch repair, basically any mechanism involving de novo DNA synthesis within CAG/CTG triplet repeats would favor repeat size changes (reviewed in: , , , . However, the precise mechanism by which hundreds or thousands of triplets are added in one single human generation is still obscure.
Given that trinucleotide repeat disorders are always associated to an expansion of the repeat array, shortening the expanded array to non-pathological length should suppress the pathology. Indeed, when a large trinucleotide repeat contraction occured during transmission from father to daughter, of an expanded myotonic dystrophy allele, complete clinical examination of the daughter showed no sign of myotonic dystrophy symptoms . It was previously reported that frequent expansions and contractions of a CAG/CTG repeat occured during double-strand break repair induced by a specific endonuclease such as I-Sce I  or HO , in Saccharomyces cerevisiae , , , . More specifically, when an I-Sce I recognition site was inserted between two short (CAG)5 repeats and a double-strand break (DSB) was induced, two-thirds of the repair events led to shortening of the repeat array by single-strand annealing, even though a homologous template was available to repair the break by gene conversion . This observation led us to the idea that inducing a specific DSB within a given trinucleotide repeat could lead to its shortening to non-pathological length.
Historically, the first nucleases used to induce a specific double-strand break into an eukaryotic chromosome were I-Sce I ,  and HO . Subsequently, homing endonucleases of the LAGLIDAGD family were engineered to recognize a large variety of restriction sites  and used for gene targeting in a wide variety of eukaryotes, including human cells , . However, the efficacy of such engineered nucleases is highly variable between the different genomic targets tested . Zinc-finger nucleases (ZFN) were developed for the same purpose. They were built by fusing modular zinc-finger DNA-binding domains to the catalytic domain of the Fok I endonuclease . They were used for targeted gene editing in eukaryotes , , but seem to induce high toxicity and a high frequency of off-target mutations, probably due to recognition and cutting of many degenerate sequences differing only slightly from the targeted sequence . More recently, a new family of specific endonucleases, called TALEN, was developed. TALENs relie on modular transcription factors, TAL effectors discovered in the genus Xanthomonas, a plant pathogen, that can be assembled to recognize any specific DNA sequence , . TAL effectors were subsequently fused to Fok I ,  or more recently to Tev I catalytic domains , to create modular proteins used in genome editing , , , , .
In the present work, a TALEN designed to recognize and cut a CAG/CTG trinucleotide repeat was assayed in a dedicated yeast experimental system. We show that, in a diploid strain containing a CAG/CTG trinucleotide repeat integrated in only one of the two homologues, the repeat-containing locus was replaced by its allelic copy by gene conversion, following TALEN induction. In a diploid strain containing CAG/CTG trinucleotide repeats integrated in both homologues, both repeats were shortened by TALEN induction. Deep-sequencing of yeast colonies in which the TALEN was expressed or not expressed showed that induction of the nuclease did not increase the mutation rate, nor did it induce formation of genomic rearrangements, segmental duplications or chromosomal translocations. Therefore, TALENs appear to be the safest and most straightforward way, at the present time, to shorten a trinucleotide repeat to non-pathological lengths.
Materials and Methods
Yeast Strains and Plasmids
Both TALEN and split-TALEN were designed and ordered at Cellectis (Paris, France). The target sequence was chosen at the centromeric-proximal junction of the trinucleotide repeat tract. Cellectis validated the design with their own appropriate bioinformatics tools, built the TALEN and validated its cutting efficiency with a dedicated single-strand annealing assay (cleavage rate 0.5 on a scale 0 to 1). Plasmid pCLS9996 (marked with KANMX), carrying the TALEN right arm was transformed in strain GFY40 (MATa ura3Δ851 leu2Δ1 his3Δ200 lys2Δ202 ade2-opal SUP4-opal), or GFY6162-3C (MATa ura3Δ851 leu2Δ1 his3Δ200 lys2Δ202 ade2-opal sup4::CAG), both previously described , . Strain GFY6162-3D (MATalpha ura3Δ851 leu2Δ1 his3Δ200 trp1Δ63 ade2-opal sup4::CAG) was transformed with pCLS16715 (TALEN left arm) or with pCLS9984 (split-TALEN left arm), both marked with LEU2. Haploid transformants were crossed on rich medium (YPGlucose), and diploids containing both TALEN arms were selected on SC-Leu supplemented with G418 sulfate (200 µg/ml). Repeat lengths were checked by Southern blot in several independent diploids before galactose induction (Figure S1).
Yeast cells were grown overnight in liquid SC -Leu medium supplemented with 200 µg/ml G418 sulfate. Cultures were washed twice with water, diluted to ca. 106 cells/ml and grown in YPLactate for five more hours (one generation). They were diluted to an appropriate concentration, then plated on SC -Leu plates supplemented with 200 µg/ml G418 sulfate, containing either 20 g/l glucose or galactose. Survival was determined as the ratio of CFU on galactose plates over CFU on glucose plates, after 3–5 days of growth at 30°C. Alternatively, after growth in YPLactate, cells were diluted to an appropriate concentration, then plated on SC -Leu -Ade plates supplemented with 200 µg/ml G418 sulfate, containing either 20 g/l glucose or galactose and a small amount of adenine (6.4 mg/l) in order to score colony color. In all induction experiments performed, only a subset of the plated colonies was analyzed by Southern blots and/or PCR.
Analysis of Trinucleotide Repeat Size
Red and white colonies were picked, total genomic DNA was extracted, digested with Eco RV or Ssp I, loaded on a 1% agarose gel and run overnight at 1 V/cm. The gel was vacuum transfered in alkaline conditions to a Hybond-XL nylon membrane (GE Healthcare) and hybridized with a randomly-labeled probe  specific of a unique region downstream of SUP4. After washing, the membrane was overnight exposed on a Fujifilm FLA-9000. When repeat tracts were short enough, it was possible to PCR amplify the SUP4 locus, using primers su3 (TTTCTCGTGTCCCCTCTTCCGT) and su9 (TTCTCTCTGGGTATGTAGGAAT). The PCR fragment was sequenced using a primer (su7: TTCAAGTATTTGTTCATTAATTT) located ca. 210 bp upstream of the repeat tract. Sanger sequencing was performed by GATC Biotech.
Library Preparation and Deep-sequencing of Yeast Colonies
Each colony, collected on a glucose or on a galactose plate, was grown in non-selective rich medium (YPGlu), whose DNA was extracted and sonicated to an average size of 500 bp (Bioruptor, maximum power (H), 30′′ ON/30′′ OFF cycles, 9 cycles). DNA ends were subsequently repaired with T4 DNA polymerase (15 units, NEBiolabs) and Klenow DNA polymerase (5 units, NEBiolabs) and phosphorylated with T4 DNA kinase (50 units, NEBiolabs). Repaired DNA was purified on two MinElute columns (Qiagen) and eluted in 16 µl (32 µl final for each library). Addition of a 3′ dATP was performed with Klenow DNA polymerase (exo-) (15 units, NEBiolabs). Home-made adapters containing a 4-bp unique tag used for multiplexing, were ligated with 2 µl T4 DNA ligase (NEBiolabs, high concentration, 2×106 units/µl). DNA was size fractionated on a Pippin Prep (Sage Science) and the fraction containing 400–600 bp DNA fragments was recovered in LoBind microtubes (Eppendorf). DNA was PCR amplified with Illumina primers PE1.0 and PE2.0 and Phusion DNA polymerase (1 unit, Thermo Scientific). Depending on PCR efficiency, 9, 12 or 15 PCR cycles were performed on each libary. Twenty-four PCR reactions were pooled, for each library, and purified on Qiagen purification columns (two columns were used for 24 PCR reactions). Elution was performed in 60 µl (twice 30 µl) and DNA was quantified on a spectrophotometer and on an agarose gel.
Analysis of Paired-end Illumina Reads
One library or two multiplexed libraries were loaded on each lane of a HiSeq 2000 (Illumina), and 110 bp paired-end reads were generated. Reads quality was evalued by FastQC v.0.10.1 (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/) and trimmed off using the paired-end mode of Trimmomatic v0.30 (http://www.usadellab.org/cms/index.php?page=trimmomatic). Trimmed reads were mapped along S288C chromosomes reference sequence (GenBank NC_001133 to NC_001148, PLN 06-DEC-2008), plus the two SUP4 alleles (SUP4-opal and sup4-(CAG)) using the paired-end mapping mode of BWA v0.6.2  with default parameters. The output SAM files were converted and sorted to BAM files using SAMtools v0.1.18 . The command IndelRealigner from GATK v2.2  was used to realigned the reads. Duplicated reads were removed using the option “MarkDuplicates” implemented in Picard v1.81 (http://picard.sourceforge.net/). Reads uniquely mapped to the reference sequence with a minimum mapping quality of 30 (Phred-scaled) were kept. Mpileup files were generated by SAMtools without BAQ adjustments. SNPs and INDELs were called by the options “mpileup2snp” and “mpileup2indel” of Varscan2 v2.3.5  with a minimum depth of 5 reads and a threshold of 0.45 for minimum variant allele frequency (strains are diploids). Mismatches were kept when they represented at least 20% of the reads supporting the variant on each strand. They were manually examined and compared between all sequenced libraries for interpretation.
TALEN Induction Induces Trinucleotide Repeat Deletions and Contractions
The genetic assay used in the present work relies on a modified suppressor tRNA gene (SUP4) in which the natural intron was replaced by either a short spacer sequence (18 bp, hereafter called SUP4-opal) or a CAG/CTG trinucleotide repeat (30–75 triplets, hereafter called sup4::CAG). The SUP4-opal allele is functional and suppresses an ade2-opal non-sense mutation that accumulates a red pigment into yeast cells, whereas the sup4::CAG is not functional , . Diploid yeast cells carrying homozygous ade2-opal mutations are red if only one copy of SUP4-opal is present, but they revert to white if two copies are present (Figure 1A). Haploid cells of opposite mating types containing either SUP4-opal or sup4::CAG, were transformed with one of the two TALEN arms. As a control, a TALEN arm modified to bind a recognition site split in two halves separated by 49 bp, was also transformed in one of the two haploid strains. The left arm of this split-TALEN should not be able to bind its cognate site and therefore no double-strand break should be induced (Figure 1B). TALEN arms are carried by multicopy plasmids (2 micron) and their expression is under the control of the inducible GAL1-10 promoter , . Cells were simultaneously plated on glucose and galactose media and colonies were scored after 3–5 days of growth. Yeast survival to the TALEN induction was 81.4%±7.2%, slightly less than survival to the split-TALEN induction (96.4%, Figure 2A). White colonies were scored and represent a majority of cells on both media, even though they are more frequent on galactose (82.5% of white colonies) as compared to glucose (66.7%). This suggests that even in repressing conditions (glucose), the GAL1-10 promoter shows some level of leakiness which is, associated to multicopy plasmids, apparently sufficient to induce TALEN expression. In support of this observation, we noticed that when crossing two haploids strains containing a stable trinucleotide repeat and one of the two TALEN arms, none of the diploids obtained contained a repeat longer than 30 triplets, strongly suggesting than even in repressing conditions, leaky expression of both TALEN arms occur to a level high enough to induce repeat contractions when both plasmids are in the same diploid cell (Figure S1). Quantification of steady-state TALEN transcripts in glucose and galactose media confirmed a low but detectable level of transcripts in glucose (Figure S2).
A: Plasmids pCLS9996 and pCLS16715, carrying the two TALEN arms, were transformed into MATa and MATalpha haploid strains, and strains were crossed to obtain diploids containing both TALEN arms. The TALEN is normally repressed on glucose medium. One copy of the active SUP4 tRNA being insufficient to suppress the ade2-opal mutation, yeast cells are red , ,  (top). In the presence of galactose, the TALEN is expressed, binds CAG/CTG trinucleotide repeats and induces a double-strand break into the repeat tract. If a second copy of an active SUP4 tRNA is generated during double-strand break repair, the ade2-opal mutation will be suppressed and yeast cells will now be white (bottom). B: Sequences recognized by both TALE DNA-binding domains and by the split-TALE. The length of the minimal spacer (18 bp) needed to induce a DSB was deduced from repeat tract lengths analyzed in surviving cells after TALEN induction (see text).
A: Survival after galactose induction. Survival was determined as the ratio of CFU on galactose plates over CFU on glucose plates. GFY620, 621, 622 and 623 are strain names used in these experiments. Error bars indicate 95% confidence intervals. B: Molecular analysis of heterozygous diploids (SUP4-opal/sup4::CAG), on glucose or galactose plates, in strains expressing either the TALEN (GFY621) or the split-TALEN (GFY620). C: PCR analysis of heterozygous diploids on glucose or galactose. PCR products were digested using restriction enzyme I-Sce I (I) or Pst I (P). For each clone, numbered 1 to 20, the two lanes show the result of restriction with one of the two enzymes. When both alleles are present, bands of slightly different sizes corresponding to uncut alleles are visible in both lanes (arrow labeled “Uncut”), along with restriction products of cut alleles (arrows labeled “Cut”). When only the SUP4-opal allele is present, no cut product is detected in the ‘I’ lane (clones 8 and 11 to 20). Note that these 20 survivors correspond to the same clones in figure 2B (strain GFY621). D: Molecular analysis of homozygous diploids (sup4::CAG/sup4::CAG), on glucose or galactose plates, in strains expressing either the TALEN (GFY622) or the split-TALEN (GFY623). Note that in all induction experiments, only a subset of all colonies growing on glucose or galactose was analyzed by Southern blot and/or PCR.
DNA originating from red and white colonies was subsequently analyzed by Southern blotting. Forty-nine out of 52 red colonies contain the two alleles, only three colonies showed the complete deletion of the sup4::CAG allele (Figure 2B). Conversely, 119 out of 120 white colonies only contain the SUP4-opal allele, whose signal intensity was twice the intensity detected in red colonies, suggesting that it corresponds to a near-complete deletion of the sup4::CAG allele. We took advantage of a restriction site polymorphism between SUP4-opal and sup4::CAG alleles, to discriminate between a perfect homozygotization and a large contraction of the sup4::CAG allele. DNA extracted from red or white diploid survivors was amplified and digested with enzymes recognizing one of the two alleles. In all ten white survivors analyzed, restrictions showed the presence of only the SUP4-opal allele (Figure 2C). Sequencing the same PCR products amplified from white diploid survivors confirmed that only one sequence was present, and not a mix of two different sequences, as would be expected for an heterozygous SUP4/sup4 locus. These experiments proved that gene conversion of the sup4::CAG allele by the SUP4-opal allele was more than 99% efficient following TALEN expression. Comparatively, there was no difference between glucose and galactose and no gene conversion was detected when inducing the split-TALEN (Figure 2B).
In a second set of experiments, we built a diploid strain containing two sup4::CAG alleles of different lengths. In such a strain, it is not possible to screen for white colonies, since both alleles are deficient in suppressing ade2-opal mutation. In the diploid strain containing the split-TALEN (GFY623), survival to galactose induction was 78.1%±13.7%, a slightly lower figure than survival of the SUP4-opal/sup4::CAG heterozygote (GFY620, 96.4%, Figure 2A). However, in the diploid strain containing the TALEN (GFY622), survival dropped to 37.1%±18%, a figure 2.2 fold lower than survival of the SUP4-opal/sup4::CAG heterozygote. This shows that cutting both chromosomes instead of one decreases viability by about a two-fold factor. Molecular analysis showed that ca. 5% of colonies on glucose (2 out of 37) showed a small expansion, whereas 59% (22 out of 37) of colonies exhibited a contracted or deleted allele (Figure 2D), suggesting again that some TALEN induction occurs in repressing conditions.
In galactose, 100% of the 153 colonies analyzed showed one single band corresponding in size to the near-complete contraction of both repeat tracts (Figure 2D: Double contraction). However, Southern blot resolution was not sufficient to determine if both alleles harbored repeats of the exact same length. DNA extracted from diploid survivors was therefore amplified and sequenced. In 23 out of 60 sequenced survivors (38%), only one sequence was present, as shown by good quality, evenly spaced peaks (Figure 3A). In 37 out of 60 survivors (62%), a mix of two DNA sequences was read after the repeat tract, indicating that the two alleles carry repeat tracts of different lengths. Using this approach, only the shortest of the two repeat tract lengths could be determined, and was found to range from three to 13 triplets (with one exception, one sequence of 20 triplets was found). Therefore, the minimal spacing between the two TALE DNA-binding domains necessary to obtain active dimerization of the Fok I nuclease and subsequent DSB formation was calculated as being 39 nt (13 triplets) minus the number of triplet repeat nucleotides bound to the left TALE (5 nt) and the number of triplet repeat nucleotides bound to the right TALE (16 nt, Figure 1B), for a total of 18 nt, a figure slightly higher than expected but compatible with former reports .
A: Sanger sequencing of survivors. PCR fragment amplified with su3/su9 (Figure 2C) was sequenced using a primer (su7) located ca. 210 bp upstream of the repeat tract. Left: when only one allele was present, one unique sequence was read (upper graph, homozygous tract length). When two alleles of different lengths were present, the sequence was blurry and unreadable after the shortest of the two repeat tracts (lower graph, heterozygous tract length). The freeware 4Peaks was used to visualize sequences. Right: length distribution of alleles in homozygous tract length (green bars) and heterozygous tract length (orange bars) survivors to TALEN induction. Homozygous tract lengths are shorter on the average (mean = 7 triplets) than heterozygous tract lengths (mean = 9 triplets), this difference being very significant (Wilcoxon test, p-value = 0.0021). Note that for heterozygous alleles only the length of the shortest repeat can be precisely known, hence the statistical difference observed between the two distributions is more significant than shown. B: Deep sequencing of yeast genomes from yeast colonies isolated on glucose or galactose plates. Each of the 15 yeast genomes was resequenced to 700 X coverage, on the average (see Table S1). For each colony, the number of unique SNPs, insertions/deletions or size changes in other CAG/CTG triplet repeats of the yeast genome, are indicated. C: Pulse-field gel electrophoresis of red and white colonies after galactose induction. Chromosomal DNA was prepared from yeast cells embedded in agarose plugs according to standard methods . Agarose plugs were loaded on 1% agarose gels (SeaKem GTG, TEBU) and electrophoresis was run on Rotaphor (Biometra) at 12°C in 0.25X TBE buffer at pH8.3, 140V, with decreasing pulse ramp 140 sec to 80 sec, and field angle 120°. Karyotypes are identical among all clones and do not show any large chromosomal rearrangement, neither on chromosome X (bearing SUP4) nor on any other chromosome. D: Two models proposing how heterozygous and homozygous tract lengths may be formed following TALEN induction (see text).
When survivors of the sup4::CAG/sup4::CAG strain containing the split-TALEN were analyzed, no colony containing two contracted alleles was detected (Figure 2D). However, six colonies out of ten on glucose and 17 colonies out of 20 on galactose showed a contraction of the largest of the two alleles (scored as “Contraction”, Figure 2D). This strongly suggests that presence of the right TALE DNA-binding domain is sufficient to increase the instability of a CAG trinucleotide repeat, probably by interfering with repeat tract replication. Indeed, when only the right TALE DNA-binding domain was expressed in a haploid strain containing a sup4::CAG allele, stability of the repeat tract was decreased (our own unpublished data).
TALEN Induction does not Increase Mutation Rates
In order to determine TALEN specificity, particularly if an increase in off-site mutations was associated with its expression, we completely resequenced eight colonies growing on glucose plates and seven colonies growing on galactose plates. Paired-end Illumina reads were generated and mapped to the S288C reference genome for each colony (Table S1). After removal of duplicates, coverage of unique sequences was homogeneous in all 15 clones sequenced, showing no aneuploidy nor segmental duplication. Among eight glucose colonies, seven unique heterozygous SNPs were detected, whereas among seven galactose colonies no heterozygous SNP was detected (Figure 3B). These numbers are not significantly different from each other and are in good agreement with predictions. Lynch and colleagues  determined that the average base substitution rate per nucleotide site was 3.3×10−10 per cell division, in S. cerevisiae. Given that glucose and galactose colonies underwent respectively 33 and 30 cell divisions before DNA was extracted and sequenced, following a Poisson distribution it was expected that ca. 11% of the sequenced colonies (one or two colonies out of 15) contained at least one mutation. Four colonies out of 15 (27%) actually contained at least one mutation, a number not statistically different from expected (Fisher exact test p-value: 0.43 for two mutant colonies, 0.21 for one mutant colony).
Mutant colonies contained between one (two colonies) and three SNPs (one colony). Altogether, three transitions for four transversions were found, a proportion not statistically different from expected (expected ratio: 0.61 ). Insertions and deletions (Indels) of one base pair in non-monotonous DNA are expected to be ten times less frequent than base substitutions , whereas indels within long poly-A/T stretches (12 bp) are more frequent . Indeed, we only found one deletion of a GC dinucleotide in an intergenic region (zero expected) and three colonies containing indels in monotonous poly-A/T stretches (four expected). More importantly no mutation was detected in any one of the naturally occuring 88 CAG/CTG trinucleotide repeats (at least five triplet long ) of the S288C genome . All indels and one out of seven SNPs fall within intergenic regions. Out of six remaining SNPs in coding regions, two are synonymous (third codon base) whereas four are non synonymous and encode point mutations in five different genes (Table S2). We concluded that expression of a TALEN targeted to a specific CAG/CTG trinucleotide repeat has no effect on other triplet repeats nor on the overall mutation rate of the yeast genome.
Since deep-sequencing cannot reveal reciprocal translocations that could be induced by the TALEN, as a last control experiment, a PFGE was run on the heterozygous SUP4-opal/sup4::CAG strain. DNA from two colonies grown on glucose and 20 colonies grown on galactose was prepared embedded in agarose plugs and loaded on a PFGE. All karyotypes were normal, showing no evidence for aneuploidies, large segmental duplications or translocations (Figure 3C).
In the present work, we show that a TALEN designed to recognize and cut a CAG/CTG trinucleotide repeat integrated in a yeast chromosome was 100% efficient in shortening the repeat tract, without inducing any other mutation in the yeast genome. In a former similar experiment, a zinc-finger nuclease was designed to recognize and cut a plasmid-born CAG/CTG trinucleotide repeat tract, in human cells. It was shown to increase triplet repeat instability by 15 fold, inducing repeat contractions, deletions of the complete repeat tract along with flanking DNA sequences, and insertions of plasmidic DNA within the repeat tract . In other experiments, in which a ZFN was directed at chromosome-borne CAG/CTG repeats in human cells , frequent contractions were observed, along with less frequent expansions of the repeat tract. These expansions were proposed to occur by homologous recombination between sister chromatids in S or G2 phase of the cell cycle. However, in these two previous experiments with ZFN, no estimate of the rate of genome wide off-target mutations induced by the nuclease was provided. Targeting efficiencies between ZFN and TALEN have been compared in nematode  and drosophila . Both ZFN and TALEN are mutagenic, making a double-strand break that will be repaired by unfaithful non-homologous end joining, generating mutations at the broken locus. However, frequencies of mutagenesis vary greatly between ZFN and TALEN, depending on the locus targeted. It is possible that TALEN efficacy, as compared to ZFN, at recognizing and cutting CAG/CTG secondary structures, is due to different DNA-protein structures. TAL Effectors wrap around DNA, each repeated motif consisting of two alpha helices connected by a short loop containing the Repeat Variable Diresidue (RVD). Each RVD contacts its cognate nucleotide within the major groove, and the protein is wrapped around DNA in a superhelical structure . Therefore, if some secondary structures are formed within triplet repeats, it is possible that they are disrupted by the binding of the TALEN left arm at the repeat junction (Figure 1B), allowing efficient binding of the TALEN right arm on the triplet repeat sequence. Very recently, a new family of highly specific endonucleases, called CRISPR, based on a guide RNA associated to a bacterial nuclease (Cas9), was engineered to modify eukaryotic genomes. The guide RNA is homologous to the target sequence and only 20–30 base pairs of homology are required for the RNA-nuclease complex to recognize and cut its cognate sequence . Given that CUG-containing RNAs are known to form stable secondary structures , it is probable that these structures will interfere either with binding of the RNA guide to the nuclease or recognition of the cognate sequence, making it unlikely that CRISPR will be very efficient at recognizing and cutting structured sequences. More importantly, the CRISPR-associated protein (Cas9) requires a protospacer adjacent motif (PAM) whose sequence is 5′-NGG-3′, in order to bind and cleave its target . Since there is no NGG triplet in a CAG/CTG trinucleotide repeat, at the present time the CRISPR technology cannot be used to induce a DSB into such repeats.
It will be interesting to test the efficacy and specificity of a similar TALEN in mammalian cells containing large CAG/CTG trinucleotide repeat expansions. The DM1 locus, containing such an expansion in myotonic dystrophy patients has been well studied by several authors. It was shown that the triplet repeat instability at this locus was dependent on the presence and activity of a nearby CTCF binding site . Given that CTCF is a regulatory factor involved in chromatin remodeling, and that it plays a direct role in regulating replication (and therefore DNA accessibility) at the DM1 locus , its presence could affect the efficacy of recognition and binding of the trinucleotide repeat by the TALEN. Additional experiments will be needed to properly adress these questions.
The mechanism by which CAG/CTG trinucleotide repeats are shortened by the TALEN can only be infered from former experiments with I-Sce I  and from known pathways of DSB-repair in yeast . DSBs made in heterozygous diploids are repaired almost exclusively by gene conversion, effectively removing the repeat tract (Figure 2B). However, six cases of repeat contractions (in red colonies) were detected on galactose plates. These contractions may correspond to intramolecular repair of the DSB by single-strand annealing, leading to small shortenings of the repeat tract. Alternatively, it may also correspond to natural instability of the trinucleotide repeat. In diploids homozygous for repeat tracts, it is likely that DSBs are repaired by single-strand annealing, although this must be confirmed by redoing similar experiments in dedicated yeast mutants. Homozygous survivors may result from iterative coordinated or uncoordinated breaks on both chromosomes, one (or two) allele(s) being cut and repaired by intra-molecular mechanism, while the other allele is repaired by gene conversion using the shortest one as a template (Figure 3D). Heterozygous survivors may result as before, from iterative coordinated or uncoordinated breaks, that will not be repaired by gene conversion and will therefore lead to repeat tracts of different lengths. This may be due to the presence of CAG repeats at DSB ends, which may impede one or more steps of homologous recombination, including correct processing of the break, subsequent formation of Rad51 nucleofilament, or strand invasion of the homologous template (which also contains CAG repeats). In support of this hypothesis, distribution of repeat tract lengths among heterozygous and homozygous survivors shows that homozygous tract lengths are shorter on the average (mean = 7 triplets) than heterozygous tract lengths (mean = 9 triplets), this difference being very significant (Wilcoxon test, p-value = 0.0021, Figure 3A). This suggests that gene conversion between repeat tracts may be hindered when tract lengths are too long, probably inhibiting an early step in the recombination process. In these cases, intramolecular repair is favored, giving rise to longer repeat tracts of unequal lengths. However, we cannot totally exclude that heterozygous survivors result from slippage occuring during DNA synthesis associated to gene conversion, as was previously demonstrated for CAG/CTG trinucleotide repeats , , . In our previous work, when an I-Sce I DSB was induced between two short (CAG)5 tracts, the break was repaired by annealing between the two repeats 67% of the time , although a homologous donor sequence was also available, a figure close to the proportion of heterozygous survivors obtained here. This suggests that when competition is possible between intra- and intermolecular repair mechanisms, intramolecular events are favored, even though gene conversion is highly efficient in yeast . Interestingly, it was very recently shown that induction of a TALEN, 129 bp downstream of a (TG)70 dinucleotide repeat in zebrafish induced frequent contractions of the repeat tract . In this experiment, three types of mutations were obtained: 56% of the sequenced zebrafish embryos showed a contraction of the TG tract but no modification of the TALEN recognition site, 15% of the embryos exhibited mutations of the recognition site but an unchanged TG tract, and 5% of the embryos showed both modifications of the recognition site and repeat contraction. Various modes of DSB-repair are proposed to account for mechanism(s) contracting tandem repeats 129 bp away from a DSB. It would be interesting to know whether mechanisms involved in zebrafish to repair TALEN-induced DSBs are similar to those happening in yeast.
TALEN expression leads to trinucleotide repeat contractions with a 100% efficacy in yeast cells, giving rise to survivors containing homozygous or heterozygous shorter alleles. Although precise molecular mechanisms by which contractions occur following TALEN induction may only be infered from our knowledge of DSB-repair following irradiation, drugs or meganuclease action, yeast will certainly prove to be helpful in dissecting mechanisms of trinucleotide repeat contractions induced by a TALEN.
Instability of trinucleotide repeats in diploid strains containing TALEN or split-TALEN, on glucose medium. A: Left: strains GFY6161-3C (MATa leu2Δ1 his3Δ200 lys2Δ202 ade2-opal sup4::(CAG)30) and GFY6162-3D (MATalpha ura3Δ851 leu2Δ1 his3Δ200 trp1Δ65 ade2-opal sup4::(CAG)75) were respectively transformed with pCLS9996 (KANMX marker) or pCLS16715 (LEU2 marker). Six transformants were analyzed by Southern blot, for each strain, to estimate repeat length variability after transformation, as well as the untransformed strain as a size control (labeled “C”). On each gel a ladder corresponding in size to different triplet repeat lengths, hybridizing with the probe, was loaded in the rightmost lane. Transformant #3 in strain GFY6162-3C shows extensive contractions of the repeat tract, but all other transformants exhibit stable trinucleotide repeats after transformation. Right: Transformants GFY6162-3C/1 and GFY6162-3D/1 were crossed, and diploids were selected on glucose SC-Leu plates supplemented with G418 sulfate (200 µg/ml). Twelve independent diploids were analyzed by Southern blot, as previously. None of the diploids contained the repeat band around 75 triplets, showing that it was contracted during or right after the cross, even though cells were crossed on glucose medium. In this particular cross, diploid #5 was selected for further induction experiments. B: Left: strain GFY6162-3D (MATalpha ura3Δ851 leu2Δ1 his3Δ200 trp1Δ65 ade2-opal sup4::(CAG)75) was transformed with pCLS9984 (split-TALEN left arm) and 12 independant transformants were analyzed by Southern blot. Transformant #3 shows an expansion and transformant #6 shows a contraction of the repeat tract, but all other transformants exhibit stable trinucleotide repeats after transformation. Clone GFY6162-3D/2 was crossed to GFY6162-3C/1, and diploids were selected on glucose SC-Leu plates supplemented with G418 sulfate (200 µg/ml). Right: Molecular analysis of four diploids shows that two of them (#3 and #4) exhibit a large contraction of the (CAG)75 repeat, that occured during or right after the cross, even though cells were crossed on glucose medium and the split-TALEN is not active. The two haploid parental strains are used as an additional size control (labeled “C”). In this particular cross, diploids #3 and #4 were selected for further induction experiments of the split-TALEN.
Steady-state level of TALEN transcript in repressing (glucose) and inducing (galactose) conditions. Strains GFY621 (SUP4-opal/sup4::CAG) and GFY622 (sup4::CAG/sup4::CAG) were grown overnight in liquid SC -Leu glucose medium supplemented with 200 µg/ml G418 sulfate. Cultures were washed twice with water, diluted to ca. 107 cells/ml and grown in 20 ml SC -Leu glucose or galactose medium supplemented with 200 µg/ml G418 sulfate, for four hours. Total RNAs were extracted and analyzed by Northern blot as previously described , . The full left TALE arm purified from pCLS16715 was labeled by random priming and used as probe. Blots were stripped in boiling 0.5% SDS and rehybridized with a randomly labeled actin probe, covering the ACT1 yeast gene. Membranes were exposed and signals were quantified on a Fujifilm FLA-9000. Relative amounts of TALEN as compared to actin transcripts are shown on the graph, in both growth conditions. There is a 10–32 fold increase of TALEN transcripts in galactose as compared to glucose, depending of the strain. In GFY622 the level of TALEN transcripts is lower than in GFY621, in both conditions. The reason for this difference was not further investigated. Note that the level of actin is lower in galactose as compared to glucose, reflecting that cells in galactose grew more slowly than in glucose medium, reducing the final number of cells, and therefore the final amount of RNAs extracted in glucose as compared to galactose.
Illumina sequencing data. Each library corresponds to one individual colony, collected on glucose or galactose plates (Origin). Total number of reads, initial read lengths, lengths after trimming and sequencing depths are indicated for each sequenced library.
Summary of mutations detected in the 15 sequenced colonies.
The authors thank Laurence Ma and Christiane Bouchier (Genomics platform, Institut Pasteur) for efficient generation of paired-end Illumina reads.
Conceived and designed the experiments: GFR. Performed the experiments: GFR DV LC VM. Analyzed the data: GFR VK VM. Contributed reagents/materials/analysis tools: GFR BD. Wrote the paper: GFR.
- 1. McMurray CT (2010) Mechanisms of trinucleotide repeat instability during human development. Nat Rev Genet 11: 786–799.
- 2. Mirkin SM (2006) DNA structures, repeat expansions and human hereditary disorders. Curr Opin Struct Biol 16: 351–358.
- 3. Orr HT, Zoghbi HY (2007) Trinucleotide repeat disorders. Annu Rev Neurosci 30: 575–621.
- 4. Pearson CE, Edamura KN, Cleary JD (2005) Repeat instability: mechanisms of dynamic mutations. Nature Reviews Genetics 6: 729–742.
- 5. Brown LY, Brown SA (2004) Alanine tracts: the expanding story of human illness and trinucleotide repeats. Trends in Genetics 20: 51–58.
- 6. Blackwood JK, Okely EA, Zahra R, Eykelenboom JK, Leach DR (2010) DNA tandem repeat instability in the Escherichia coli chromosome is stimulated by mismatch repair at an adjacent CAG.CTG trinucleotide repeat. Proceedings of the National Academy of Sciences of the United States of America 107: 22582–22586.
- 7. Kang S, Jaworski A, Ohshima K, Wells RD (1995) Expansion and deletion of CTG repeats from human disease genes are determined by the direction of replication in E. coli. Nature Genet 10: 213–217.
- 8. Bhattacharyya S, Lahue RS (2004) Saccharomyces cerevisiae Srs2 DNA helicase selectively blocks expansions of trinucleotide repeats. Mol Cell Biol 24: 7324–7330.
- 9. Freudenreich CH, Kantrow SM, Zakian VA (1998) Expansion and length-dependent fragility of CTG repeats in yeast. Science 279: 853–856.
- 10. Kerrest A, Anand R, Sundararajan R, Bermejo R, Liberi G, et al. (2009) SRS2 and SGS1 prevent chromosomal breaks and stabilize triplet repeats by restraining recombination. Nature Structural and Molecular Biology 16: 159–167.
- 11. Jung J, van Jaarsveld MT, Shieh SY, Xu K, Bonini NM (2011) Defining genetic factors that modulate intergenerational CAG repeat instability in Drosophila melanogaster. Genetics 187: 61–71.
- 12. Gomes-Pereira M, Monckton DG (2004) Chemically induced increases and decreases in the rate of expansion of a CAG*CTG triplet repeat. Nucleic Acids Res 32: 2865–2872.
- 13. Savouret C, Brisson E, Essers J, Kanaar R, Pastink A, et al. (2003) CTG repeat instability and size variation timing in DNA repair-deficient mice. Embo J 22: 2264–2273.
- 14. Kovtun IV, Liu Y, Bjoras M, Klungland A, Wilson SH, et al. (2007) OGG1 initiates age-dependent CAG trinucleotide expansion in somatic cells. Nature 447: 447–452.
- 15. Dion V, Lin Y, Hubert L Jr, Waterland RA, Wilson JH (2008) Dnmt1 deficiency promotes CAG repeat expansion in the mouse germline. Hum Mol Genet 17: 1306–1317.
- 16. Claassen DA, Lahue RS (2007) Expansions of CAG.CTG repeats in immortalized human astrocytes. Human Molecular Genetics 16: 3088–3096.
- 17. Gorbunova V, Seluanov A, Dion V, Sandor Z, Meservy JL, et al. (2003) Selectable system for monitoring the instability of CTG/CAG triplet repeats in mammalian cells. Mol Cell Biol 23: 4485–4493.
- 18. Lopez Castel A, Nakamori M, Tome S, Chitayat D, Gourdon G, et al. (2011) Expanded CTG repeat demarcates a boundary for abnormal CpG methylation in myotonic dystrophy patient tissues. Hum Mol Genet 20: 1–15.
- 19. Cleary JD, Pearson CE (2005) Replication fork dynamics and dynamic mutations: the fork-shift model of repeat instability. Trends in Genetics 21: 272–280.
- 20. Lenzmeier BA, Freudenreich CH (2003) Trinucleotide repeat instability: a hairpin curve at the crossroads of replication, recombination, and repair. Cytogenetic and Genome Research 100: 7–24.
- 21. Richard GF, Kerrest A, Dujon B (2008) Comparative genomics and molecular dynamics of DNA repeats in eukaryotes. Microbiol Mol Biol Rev 72: 686–727.
- 22. O’Hoy KL, Tsilfidis C, Mahadevan MS, Neville CE, Barcelo J, et al. (1993) Reduction in size of the myotonic dystrophy trinucleotide repeat mutation during transmission. Science 259: 809–812.
- 23. Colleaux L, d’Auriol L, Betermier M, Cottarel G, Jacquier A, et al. (1986) Universal code equivalent of a yeast mitochondrial intron reading frame is expressed into E. Coli as a specific double strand break endonuclease. Cell 44: 521–533.
- 24. Kostriken R, Strathern JN, Klar AJ, Hicks JB, Heffron F (1983) A site-specific endonuclease essential for mating-type switching in Saccharomyces cerevisiae. Cell 35: 167–174.
- 25. Richard G-F, Cyncynatus C, Dujon B (2003) Contractions and expansions of CAG/CTG trinucleotide repeats occur during ectopic gene conversion in yeast, by a MUS81-independent mechanism. Journal of Molecular Biology 326: 769–782.
- 26. Richard G-F, Dujon B, Haber JE (1999) Double-strand break repair can lead to high frequencies of deletions within short CAG/CTG trinucleotide repeats. Mol Gen Genet 261: 871–882.
- 27. Richard G-F, Goellner GM, McMurray CT, Haber JE (2000) Recombination-induced CAG trinucleotide repeat expansions in yeast involve the MRE11/RAD50/XRS2 complex. EMBO J 19: 2381–2390.
- 28. Richard G-F, Pâques F (2000) Mini- and microsatellite expansions: the recombination connection. EMBO Reports 1: 122–126.
- 29. Fairhead C, Dujon B (1993) Consequences of unique double-stranded breaks in yeast chromosomes: death or homozygosis. Mol Gen Genet 240: 170–180.
- 30. Choulika A, Perrin A, Dujon B, Nicolas JF (1995) Induction of homologous recombination in mammalian chromosomes by using the I-SceI system of Saccharomyces cerevisiae. Mol Cell Biol 15: 1968–1973.
- 31. Haber JE (1995) In vivo biochemistry: physical monitoring of recombination induced by site-specific endonucleases. BioEssays 17: 609–620.
- 32. Arnould S, Chames P, Perez C, Lacroix E, Duclert A, et al. (2006) Engineering of large numbers of highly specific homing endonucleases that induce recombination on novel DNA targets. J Mol Biol 355: 443–458.
- 33. Pâques F, Duchâteau P (2007) Meganucleases and DNA double-strand break-induced recombination: perspectives for gene therapy. Curr Gene Ther 7: 49–66.
- 34. Stoddard BL (2011) Homing endonucleases: from microbial genetic invaders to reagents for targeted DNA modification. Structure 19: 7–15.
- 35. Daboussi F, Zaslavskiy M, Poirot L, Loperfido M, Gouble A, et al. (2012) Chromosomal context and epigenetic mechanisms control the efficacy of genome editing by rare-cutting designer endonucleases. Nucleic Acids Res 40: 6367–6379.
- 36. Kim YG, Cha J, Chandrasegaran S (1996) Hybrid restriction enzymes: zinc finger fusions to Fok I cleavage domain. Proc Natl Acad Sci U S A 93: 1156–1160.
- 37. Chen F, Pruett-Miller SM, Huang Y, Gjoka M, Duda K, et al. (2011) High-frequency genome editing using ssDNA oligonucleotides with zinc-finger nucleases. Nat Methods 8: 753–755.
- 38. Doyon Y, McCammon JM, Miller JC, Faraji F, Ngo C, et al. (2008) Heritable targeted gene disruption in zebrafish using designed zinc-finger nucleases. Nat Biotechnol 26: 702–708.
- 39. Boch J, Scholze H, Schornack S, Landgraf A, Hahn S, et al. (2009) Breaking the code of DNA binding specificity of TAL-type III effectors. Science 326: 1509–1512.
- 40. Moscou MJ, Bogdanove AJ (2009) A simple cipher governs DNA recognition by TAL effectors. Science 326: 1501.
- 41. Christian M, Cermak T, Doyle EL, Schmidt C, Zhang F, et al. (2010) Targeting DNA double-strand breaks with TAL effector nucleases. Genetics 186: 757–761.
- 42. Li T, Huang S, Jiang WZ, Wright D, Spalding MH, et al. (2011) TAL nucleases (TALNs): hybrid proteins composed of TAL effectors and FokI DNA-cleavage domain. Nucleic Acids Res 39: 359–372.
- 43. Beurdeley M, Bietz F, Li J, Thomas S, Stoddard T, et al. (2013) Compact designer TALENs for efficient genome engineering. Nat Commun 4: 1762.
- 44. Cermak T, Doyle EL, Christian M, Wang L, Zhang Y, et al. (2011) Efficient design and assembly of custom TALEN and other TAL effector-based constructs for DNA targeting. Nucleic Acids Res 39: e82.
- 45. Bedell VM, Wang Y, Campbell JM, Poshusta TL, Starker CG, et al. (2012) In vivo genome editing using a high-efficiency TALEN system. Nature 491: 114–118.
- 46. Cade L, Reyon D, Hwang WY, Tsai SQ, Patel S, et al. (2012) Highly efficient generation of heritable zebrafish gene mutations using homo- and heterodimeric TALENs. Nucleic Acids Res 40: 8001–8010.
- 47. Chen S, Oikonomou G, Chiu CN, Niles BJ, Liu J, et al. (2013) A large-scale in vivo analysis reveals that TALENs are significantly more mutagenic than ZFNs generated using context-dependent assembly. Nucleic Acids Res 41: 2769–2778.
- 48. Qiu Z, Liu M, Chen Z, Shao Y, Pan H, et al.. (2013) High-efficiency and heritable gene targeting in mouse by transcription activator-like effector nucleases. Nucleic Acids Res.
- 49. Feinberg AP, Vogelstein B (1983) A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132: 6–13.
- 50. Li H, Durbin R (2009) Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25: 1754–1760.
- 51. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, et al. (2009) The Sequence Alignment/Map format and SAMtools. Bioinformatics 25: 2078–2079.
- 52. DePristo MA, Banks E, Poplin R, Garimella KV, Maguire JR, et al. (2011) A framework for variation discovery and genotyping using next-generation DNA sequencing data. Nat Genet 43: 491–498.
- 53. Koboldt DC, Zhang Q, Larson DE, Shen D, McLellan MD, et al. (2012) VarScan 2: somatic mutation and copy number alteration discovery in cancer by exome sequencing. Genome Res 22: 568–576.
- 54. Guarente L, Yocum RR, Gifford P (1982) A GAL10-CYC1 hybrid yeast promoter identifies the GAL4 regulatory region as an upstream site. Proc Natl Acad Sci U S A 79: 7410–7414.
- 55. Giniger E, Varnum SM, Ptashne M (1985) Specific DNA binding of GAL4, a positive regulatory protein of yeast. Cell 40: 767–774.
- 56. Bogdanove AJ, Voytas DF (2011) TAL effectors: customizable proteins for DNA targeting. Science 333: 1843–1846.
- 57. Lynch M, Sung W, Morris K, Coffey N, Landry CR, et al. (2008) A genome-wide view of the spectrum of spontaneous mutations in yeast. Proc Natl Acad Sci U S A 105: 9272–9277.
- 58. Lang GI, Murray AW (2008) Estimating the per-base-pair mutation rate in the yeast Saccharomyces cerevisiae. Genetics 178: 67–82.
- 59. Richard G-F, Dujon B (1997) Trinucleotide repeats in yeast. Res Microbiol 148: 731–744.
- 60. Mittelman D, Moye C, Morton J, Sykoudis K, Lin Y, et al. (2009) Zinc-finger directed double-strand breaks within CAG repeat tracts promote repeat instability in human cells. Proc Natl Acad Sci U S A 106: 9607–9612.
- 61. Liu G, Chen X, Bissler JJ, Sinden RR, Leffak M (2010) Replication-dependent instability at (CTG) x (CAG) repeat hairpins in human cells. Nat Chem Biol 6: 652–659.
- 62. Wood AJ, Lo TW, Zeitler B, Pickle CS, Ralston EJ, et al. (2011) Targeted genome editing across species using ZFNs and TALENs. Science 333: 307.
- 63. Beumer KJ, Trautman JK, Christian M, Dahlem TJ, Lake CM, et al. (2013) Comparing Zinc Finger Nucleases and Transcription Activator-Like Effector Nucleases for Gene Targeting in Drosophila. G3 (Bethesda) 3: 1717–1725.
- 64. Deng D, Yan C, Pan X, Mahfouz M, Wang J, et al. (2012) Structural basis for sequence-specific recognition of DNA by TAL effectors. Science 335: 720–723.
- 65. Pennisi E (2013) The CRISPR craze. Science 341: 833–836.
- 66. Pinheiro P, Scarlett G, Rodgers A, Rodger PM, Murray A, et al. (2002) Structures of CUG repeats in RNA. The Journal of Biological Chemistry 277: 35183–35190.
- 67. Sternberg SH, Redding S, Jinek M, Greene EC, Doudna JA (2014) DNA interrogation by the CRISPR RNA-guided endonuclease Cas9. Nature.
- 68. Libby RT, Hagerman KA, Pineda VV, Lau R, Cho DH, et al. (2008) CTCF cis-regulates trinucleotide repeat instability in an epigenetic manner: a novel basis for mutational hot spot determination. PLoS Genet 4: e1000257.
- 69. Cleary JD, Tome S, Lopez Castel A, Panigrahi GB, Foiry L, et al. (2010) Tissue- and age-specific DNA replication patterns at the CTG/CAG-expanded human myotonic dystrophy type 1 locus. Nat Struct Mol Biol 17: 1079–1087.
- 70. Pâques F, Haber JE (1999) Multiple Pathways of Recombination Induced by Double-Strand Breaks in Saccharomyces cerevisiae. Microbiol Mol Biol Rev 63: 349–404.
- 71. Huang W, Zheng J, He Y, Luo C (2013) Tandem Repeat Modification during Double-Strand Break Repair Induced by an Engineered TAL Effector Nuclease in Zebrafish Genome. PLoS One 8: e84176.
- 72. Schwartz DC, Cantor CR (1984) Separation of yeast chromosome-sized DNAs by pulsed field gradient gel electrophoresis. Cell 37: 67–75.
- 73. Richard G-F, Fairhead C, Dujon B (1997) Complete transcriptional map of yeast chromosome XI in different life conditions. J Mol Biol 268: 303–321.
- 74. Fairhead C, Dujon B (1994) Transcript map of two regions from chromosome XI of Saccharomyces cerevisiae for interpretation of systematic sequencing results. Yeast 10: 1403–1413.