Methods xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Methods journal homepage: www.elsevier.com/locate/ymeth

TALEN-mediated Drosophila genome editing: Protocols and applications Jiyong Liu a,b, Yixu Chen a, Renjie Jiao a,b,⇑ a b

State Key Laboratory of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences, China Sino-French Hoffmann Institute, Guangzhou Medical University, Dongfengxi Road 195, Guangzhou 510182, China

a r t i c l e

i n f o

Article history: Received 31 October 2013 Revised 8 April 2014 Accepted 10 April 2014 Available online xxxx Keywords: Drosophila TALEN TALE Genome editing

a b s t r a c t TALEs (transcription activator-like effectors) are a family of natural transcriptional activators originally isolated from the plant pathogen of Xanthomonas spp. The DNA binding motif of TALEs can be re-designed in such way that they bind specific DNA sequences other than their original targets. Fusion of customized TALEs with an endonuclease, Fok I, generates artificial enzymes that are targeted to specific DNA sites for cutting, allowing gene specific modification of both animal and plant genomes. Previously, we reported the use of TALEN (transcription activator-like effector nuclease) for the highly specific and efficient modification of the two Drosophila loci yellow and CG9797. Here, we describe a detailed protocol for TALEN-mediated genomic modification in Drosophila, with the aim of providing a practical bench guide for the Drosophila research community. Ó 2014 Elsevier Inc. All rights reserved.

1. Introduction The functional dissection of a genome greatly depends on the precision of tools available for modification of DNA at specific sites of interest, at the required size and the type of genomic modification, including nucleotide replacement, deletions, insertions, inversions or even translocations. Ever since the establishment of modern genetics over 100 years ago, finding such precision tools that allow for targeted genome modification has been a longstanding goal for generations of scientists. Over the course of the last century, a wide range of genome-modifying methods has been developed for Drosophila, including forward and reverse genetic techniques. However, one major obstacle for advancing novel insights into gene function has been the fact that EMS (ethylmethane sulphonate)- and transposon-mediated methods are not ideal methods for sequence specific modifications [1]. And whilst HR (homologous recombination)-based gene targeting (ends-out or ends-in) methods can be employed for gene specific mutations [2,3], their low efficiency of successful targeting of only about 1/500 to 1/800 gametes [4], together with the fact that they are time-consuming procedures, limits their prevalence in the fly community, especially for large-scale and high-throughput studies. Lastly, even though a customizable method based on ZFN (zincfinger nuclease) has been developed specifically for Drosophila mutagenesis, an approach distinguished by high sequence specific⇑ Corresponding author at: Institute of Biophysics, CAS, Datun Road 15, Beijing 100101, China. E-mail address: [email protected] (R. Jiao).

ity and considerable efficiency [5–7], the complexity of constructing DNA-binding zinc-finger arrays, as well as toxicity-related issues present a major drawback for wider application [8,9]. TALEs (transcription activator-like effectors) are a group of transcriptional activators specifically found in Xanthomonas spp. [10]. The plant pathogen translocates TALEs into host plant cells through a type III secretion system, resulting in the activation of target gene expression. Characteristically, a naturally occurring TALE is composed of three parts: a type III translocation signal-containing N-terminal, a C-terminal containing both a nuclear localization signal (NLS) and a transcriptional activation domain (AD), and a central DNA-binding domain [10] (Fig. 1A). The DNA binding domain consists of a variable number of highly repeated units (between 1.5 and 33.5), with each unit specifically recognizing one nucleotide. Each unit is composed of 33–35, but typically 34 amino acids, except for the last unit, which is composed of 20 amino acids. The amino acid sequences of all repeats are nearly identical, except for the hyper-variable residues in position 12 and 13, which are therefore called repeat-variable di-residues (RVDs) [11]. The DNA-binding specificity of a TALE is determined by its RVDs, with a simple cipher of HD to C, NG to T, NI to A, NN to A or G, and NS to A, C, G or T [12,13] (Fig. 1A and B). By fusing an artificially designed TALE, containing specific RVDs, with the endonuclease Fok I, it is possible for scientists to generate a customized fusion protein that can target this enzyme activity to any site within the genome. The resulting fusion protein, composed of a TALE and the C-terminal Fok I cleavage domain, is termed TALEN (transcription activator-like effector nuclease), which combines the features of both specific DNA binding and DNA cleavage activity. TALENs are

http://dx.doi.org/10.1016/j.ymeth.2014.04.010 1046-2023/Ó 2014 Elsevier Inc. All rights reserved.

Please cite this article in press as: J. Liu et al., Methods (2014), http://dx.doi.org/10.1016/j.ymeth.2014.04.010

2

J. Liu et al. / Methods xxx (2014) xxx–xxx

Fig. 1. Schematic view and principle of TALEN-mediated genome editing. (A) The structure of Xanthomonas TALE protein, with the amino acids sequence of a typical repeat depicted underneath. (B) Structure of a TALEN, with its recognized DNA sequence presented underneath. (C) TALEN-mediated genome editing. Left and right TALENs bind to the target sequence before the Fok I dimer induces a DSB in the spacer region. NHEJ repair produces small insertions and/or deletions (indels), and HR-mediated repair can be employed for gene tagging and gene correction. Double DSBs induced in the same chromosome can promote the generation of a large fragment deletion or an inversion. Double DSBs induced in different chromosomes may induce a chromosomal translocation. NLS, nuclear localization signal. AD, activation domain. TLS, translocation signal. Modified from Liu and colleagues [16].

designed in such way that they work in pairs: a left and a right TALEN bind to their respective target sequences, which are separated by a short DNA spacer region. In turn, the two Fok I form a dimer, which is required to cleave efficiently the spacer DNA, generating a DSB (double-strand break) as a result. The DSB is then repaired by one of two pathways: non-homologous end joining (NHEJ) or homologous recombination (HR). NHEJ is an error prone pathway that results in small insertions and/or deletions (indels) around the DSB. HR uses a homologous template to repair the broken DNA with high precision. When an exogenous homologous template containing modifications such as GFP tag is available, the desired sequence can then be incorporated into the genome, achieving a precise genome edition [14,15] (Fig. 1C). Recently, we have used this novel TALEN technique to target successfully two Drosophila genes, yellow and CG9797, with efficiencies ranging from 17.2% to 66.7% [16]. Thus far, TALEN has been employed in various species, including human cells [17], mouse [18], zebrafish [19], Xenopus [20], Caenorhabditis elegans [21], plants [22], bovine [23], silkworm [24], cricket [25] and mosquito [26]. The option of customizing this method for unique target regions makes TALEN an excellent tool to manipulate almost any genomic sequence. However, the high degree of similarity of the TALE repeats challenges the construction of customized TALENs. To overcome this constraint, several methods of constructing TALE repeats have been developed, which can be classified into three main types [14]: (1) The traditional, enzymatic digestion–ligation-based methods, including Unit Assembly [19] and REAL (restriction enzyme and ligation) [27]. Unit Assembly makes use of a collection of RVD mono-/di-/tri- and tetra-units. Taking advantage of the isocaudameric nature of enzymes such as Nhe I and Spe I, this method adds, step by step, Spe I/Hind III-digested fragments to the Nhe I/Hind III linearized vector, resulting in a complete set of TALE repeats. The construction of a TALE of no more than 17 RVD repeats can be achieved through 2 rounds of digestion–ligations (for experimental details, see Section 3.3). This method is suitable for small-scale construction of TALEs in most labs [16,28,29]. (2) Golden Gate-based methods employ an overhang generated by a type IIS endonuclease, from which the orderly assembly of up to 10 DNA fragments is directed in a single step [30]. Usually, a

full-length TALEN construct can be generated within a timeframe of approximately 1 week [22,31,32]. (3) Other methods include Solid-phase-supported methods such as FLASH (Fast Ligationbased Automatable Solid-phase High-throughput) system [33], ICA (Iterative Capped Assembly) [34], etc. [35]. Here, ligation is carried out on solid-phase magnet beads, avoiding manual PCRs and the requirement for cumbersome gel isolation steps. FLASH is a high-throughput method, which allows the automatic production of 96 different TALE arrays in less than 1 day [33]. Another highthroughput strategy is called Ligation-Independent Cloning (LIC). This method can handle samples in a high-throughput manner, which can process more than 600 different TALE arrays in 1 day [36]. Customized TALEs can now also be obtained commercially, greatly reducing time and man power at the bench. In addition to genomic modifications, TALE/TALEN has also been used in other research areas, including studies of gene expression regulation in Drosophila [37], DNA recombination in bacterial and human cells [38], as well as live visualization of chromatin dynamics in mouse cells [39] and others. In this article, we describe in detail a step-by-step protocol for gene mutagenesis in Drosophila using TALEN, from selection of target sites to molecular identification of designed mutations. In addition, we will discuss its use for other applications, including HR-based genomic modification.

2. Materials 2.1. Husbandry and handle of Drosophila The flies are fed with common corn yeast food as described in Stocker and Gallant [40], raised at 25 °C in an incubator or a room with the humidity of 60% and a 12 h each of light and dark cycle. Other materials for handling flies include: (1) (2) (3) (4)

Stereomicroscopes (Leica SE4); Forceps (Dumont, #55); Incubators (Percival, CAT# I-36VL); Fly-sleeping facility equipped with CO2.

Please cite this article in press as: J. Liu et al., Methods (2014), http://dx.doi.org/10.1016/j.ymeth.2014.04.010

J. Liu et al. / Methods xxx (2014) xxx–xxx

2.2. Construction of TALENs (1) Mono-/di-/tri- and tetra-units used in Unit Assembly (courtesy of Dr. Bo Zhang at Peking University, China): Plasmids containing the coding sequences of mono-/di-/tri- and tetra-TALE repeats. All coding sequences were cloned into the pMD18T vector with Spe I and Nhe I [19] (Fig. 3A); (2) pCS2-PEAS and pCS2-PERR (courtesy of Dr. Bo Zhang, Peking University, China): SP6 driven expression vectors containing the coding sequences of Fok I and the N-/C-termini of TALEN [19]. 2.3. Other molecular manipulations (1) Solution I (TaKaRa, CAT# D6022A): for ligation reactions; (2) AxyPrep DNA Gel Extraction Kit (Axygen, CAT# AP-GX-250): for gel purification of DNA; (3) AxyPrep Plasmid Miniprep Kit (Axygen, CAT# AP-MN-P250): for standard plasmid isolation; (4) QIAGEN Plasmid Midi Kit (QIAGEN, CAT# 12143): for high quality DNA purification, such as the donor DNAs that have to be RNase-free; (5) Alkaline Phosphatase (CIP) (NEB, CAT# M0290): for removal of 50 phosphate groups from DNA prior to ligation; (6) Proteinase K (Calbiochem, CAT# 539480). 2.4. In vitro transcription of TALENs (1) SP6 mMESSAGE mMACHINE Kit (Ambion, CAT# AM1340): for in vitro transcription of the left and right TALENs; (2) RNase-free water (TaKaRa, CA# D602): for dissolving in vitro transcribed mRNAs; (3) Eppendorf BioPhotometer: for RNA quantification. 2.5. Germ-line transformation of Drosophila (1) Leica microscope (DMIL): an inverted microscope used as microinjection platform; (2) Micromanipulator (Narishige MN-151): to manipulate needles for microinjection; (3) Microinjectors (Eppendorf, CellTram(Oil)): for microinjection; (4) Glass capillary tubes (Narishige, GD-1): for microinjection needles; (5) Magnetic glass microelectrode puller (Narishige, PN-30): to prepare the tapering needles for injection; (6) Eppendorf Microloader Tips (CAT# 5242 956.003): for filling of injection needles with injection solution; (7) Halocarbon oil 700 (Sigma, H8898): for covering microinjected embryos to maintain humidity; (8) Sartorius cellulose nitrate (CN) membrane: for embryo collection. 3. Methods 3.1. Selection of TALEN-targeting sites Based on the DNA-binding characteristics of natural TALE proteins, the selection of TALEN-targeting sites should follow the rules described below: 1. The binding sequence for a designed TALEN is recommended to be 12–17 bp. Although natural TALEs consist of a variable number of repeats, varying from 1.5 to 33.5 repeats and with an average number of 17.5 [10], usually more than 10.5 repeats are required

3

for efficient target activation [12]. Whilst long binding sequences (>17 bp) theoretically improve binding specificity, the construction of such TALEs is more time consuming and more laborious compared to short ones. When the Unit Assembly strategy is used for the construction of TALEN repeats, we recommend the TALENbinding target sequence to be no more than 17 bp, to ensure that the construction of TALE repeats can be finished in two rounds of molecular digestion–ligations. In our hands, the shortest TALEN binding sequence successfully devised was 13 bp. 2. The binding sequence of TALEN usually starts after a T or C. Although most natural TALEs bind to sequences that are preceded by a T [13], target sequences starting after a C can also be found [41]. It is reported that T-preceded sequences usually show higher activity, and thus it is the preferred option for TALEN-binding sequences [42]. 3. The length of spacer DNA between the left and right TALENs is usually 14–18 bp. In the process of TALEN-mediated DNA cleavage, Fok I is active only in the form of dimers, a manner that requires appropriate space on the DNA in between the two TALEN-binding sites, thus both the structure of TALENs and the length of spacer DNA affect the activity of a customized pair of TALENs [43,44]. For example, Christian and colleagues reported that TALENs with either 63 or 231 amino acids at the C-terminus both can cleave spacers comprising of 14–33 bp. But, when the C-terminus is truncated to the length of 18 amino acids, such TALENs can then only function efficiently on spacers with 13–16 bp in length [45]. In our hands, a spacer length of 14–18 bp was the best for getting optimal gene targeting results. In addition, for the convenience of molecular identification of mutant F0 (mosaic) and F1 flies, it is advisable to choose a spacer DNA that contains a restriction enzyme cutting site that appears nowhere else in the adjacent DNA (500 bp). Commonly used enzymes with high efficiency are recommended. 4. The presence of strong as well as evenly distributed RVDs is important for TALE activity. Streubel and colleagues reported that RVD composition, meaning the ratio of strong RVDs to weak RVDs, can greatly affect the activity of TALE effectors [46]. At least 3–4 strong RVDs (HD or NN) should be included in each TALE, which ideally should be evenly distributed to avoid multiple weak RVDs (NI, NG or NK) to appear in a tandem array. A consecutive sequence of weak RVDs will weaken overall TALE activity [46]. 5. For increased specificity, use NH or NK to specify G; for stronger RVDs to be incorporated, use NN to recognize G. Being the prevalent RVD to target G, NN is highly efficient, but it also recognizes A [12]. NH and NK are highly specific to G, but both exhibit compromised targeting efficiency [45–47]. Therefore, requirements for either high efficiency or specificity have to be balanced when deciding which of the RVDs to chose for targeting G. 6. For the purpose of disrupting gene function through mutagenesis, it is best to choose the targeting site from within the coding region. It is advisable to disrupt the sequence either near the ATG, in other essential domains, or at intron–exon boundaries. 7. TAL Effector-Nucleotide Targeter (TALE-NT) 2.0 (https://boglab.plp.iastate.edu/) is a powerful program to help identify potential TALEN sites from a given DNA sequence [42]. One can chose from various key parameters to specify and determine a target sequence, for instance repeat length, spacer length and others (Fig. 2A). In addition to TALE-NT, many other TALEN design tools have been developed, and are listed in Section 5 below. Based on these guidelines, a sample sequence was selected, using the TALE-NT program, for a pair of TALENs to target Drosophila yellow gene (Fig. 2B), 50 -GCCCCTATGCGGTAA-30 as the spacer, which includes a Bsl I recognition site. The left TALEN sequence was designed as 50 -ACCACCACTAATCCGT-30 , and the right sequence as 50 -GGTCAAGTCAAAGACAT-30 [16].

Please cite this article in press as: J. Liu et al., Methods (2014), http://dx.doi.org/10.1016/j.ymeth.2014.04.010

4

J. Liu et al. / Methods xxx (2014) xxx–xxx

Fig. 2. Selection of TALEN target sites. (A) The snapshot of the front page of a TALEN sites designing program, TALE-NT (TAL Effector-Nucleotide Targeter) 2.0. (B) The selected TALEN target sequence for Drosophila yellow mutagenesis. The sequences in the boxes are the left and right TALEN binding sites. The small-lettered sequence is the spacer DNA, in which the Bsl I site resides, and will be used for the molecular identification of mutations. Adapted from Liu and colleagues [16].

3.2. Confirmation of the selected TALEN sites

3.3. Construction of TALENs using ‘‘Unit Assembly’’

To efficiently and specifically edit the genomic sequence of interest, it is recommended to confirm the chosen target site using the methods described below:

Different strategies have been established for the construction of TALENs (see Section 1 for details). We routinely employ the Unit Assembly [19] to construct TALENs, for the following reasons: (1) it is suitable for small scale construction of TALENs; (2) the required reagents (mono-/di-/tri- and tetra-units for TALE repeats construction and pCS2-PEAS/pCS2-PERR for TALEN construction) are freely available (see Section 2.2); (3) traditional digestion–ligation reactions can be carried out in any lab without need for specialized equipments; (4) it is a simple procedure that works with high efficiency. Taking the left TALEN of yellow gene, pCS2-TALEN-yL (corresponding DNA target is 50 -ACCACCACTAATCCGT-30 ), as an example (Fig. 3), the assembly of a TALEN with the existing units is described below. For DNA cloning procedures, please refer to standard protocols published previously [49].

1. BLAST search to ensure that the selected target sequence is unique in the entire Drosophila genome. 2. Extract the gDNA of flies you intend to use for microinjection; design at least one pair of primers to amplify the sequence (about 500 bp) flanking the selected target site. Verify that the PCR product is clean and abundant. 3. (Important) Sequence the amplified PCR products, verifying that the sequence of this fly genome is identical to the selected targets. Any mismatch of a single nucleotide in the region of TALEN binding sequence may compromise the TALEN activity [48]. If the sequencing results indicate any possible SNPs (single nucleotide polymorphisms), it is recommended to re-select an alternative target sequence until the sequences are found to be identical. 4. (Optional) Digest the PCR product with the same enzyme that is designed to cut the spacer site, verifying that: (1) the PCR product can be completely digested; (2) the cut DNA can be easily distinguished by gel electrophoresis from the PCR product.

3.3.1. Construction of TALE repeats 1. pACCA, about 0.5–1 lg, is double-digested with Nhe I/Hind III to be a linearized vector. 3–4 lg of pCCAC (3.2 kb), containing the coding sequence of CCAC (about 0.4 kb), is digested with Spe I/Hind III. The digestion reaction is incubated at 37 °C for about 2–4 h. Subsequently, the vector DNA and the CCAC coding fragment are

Please cite this article in press as: J. Liu et al., Methods (2014), http://dx.doi.org/10.1016/j.ymeth.2014.04.010

J. Liu et al. / Methods xxx (2014) xxx–xxx

5

Fig. 3. The Unit Assembly strategy for TALEN construction. (A and B) Show the construction of TALE repeats. Spe I/Hind III double digested C (the coding sequence of the TALE unit recognizing C) is ligated to the Nhe I/Hind III digested vector (containing A) to get the double unit of AC (containing the sequence of TALE unit recognizing AC). The TALE repeats of the left TALEN for yellow is constructed in two rounds of digestion–ligations (B). (C) The constructed TALE repeats are subcloned into the Nhe I digested Fok I expression vector to get the final TALEN plasmid. S, Spe I. N, Nhe I. H, Hind III.

extracted from the gel and purified (Axygen, CAT# AP-GX-250), before ligated in Solution I (TaKaRa, CAT#: D6022A) for 2 h at 16 °C (with a ligation system of 6 lL that contains 0.1 lg DNA for ACCA and 0.3 lg DNA for CCAC) to obtain pACCACCAC. Similarly, the CCG coding sequence is cut from pCCG and cloned into the Nhe I/Hind III double-digested pTAAT to obtain pTAATCCG, in parallel with the construction of pACCACCAC (Fig. 3B). The last unit (recognizing T) is already embedded in the Fok I expression vectors (pCS2-PEAS and pCS2-PERR). 2. To identify positive colonies, we screen for positive clones by colony PCR using the pair of primers M13-47 (50 -CGCCAG GGTTTTCCCAGTCACGAC-30 ) and RV-M (50 -AGCGGATAACAATTTCACACAGGA-30 ). Usually, 60–100% of the picked clones are positive. For colony PCR, the picked colonies are incubated with a typical PCR lysis/reaction buffer containing 2.5 lL of 10  Taq buffer, 2 lL of dNTP Mix (2.5 mM for each dNTP), 0.5 lM of M13-47, 0.5 lM of RV-M, 1 U of Taq and distilled water for a total reaction volume of 25 lL. Please note that each unit is 102 bp long, and thus the extension time for PCR varies depending on the number of TALE units that the plasmid contains. In this case, the fragment to be amplified is 972 bp (156 bp (the backbone sequence) + 102 bp  8). 3. pACCACCAC and pTAATCCG from the positive colonies are then digested and the repeats ligated to get the full length pMD-TALE-yL (pACCACCACTAATCCG) (Fig. 3B) [16]. Sequencing

with the primer pair M13-47 and RV-M is recommended at this point, to ensure no mutations were introduced during the cloning procedure. 3.3.2. Construction of TALENs Constructed TALE repeats should be subcloned into the Fok I-carrying vector to obtain the final fused TALEN construct. Briefly, it goes as follows (Fig. 3C): 1. Use Nhe I to digest about 1 lg of each vector pCS2-PEAS (5.4 kb) and pCS2-PERR (5.5 kb), respectively, at 37 °C for about 2–4 h. Remove the 50 phospho group by CIP (NEB, CAT# M0290) treatment for 30 min, run on an agarose gel before extract and purify the vector DNAs. 2. Cut the left and right TALE repeats from pMD-TALE-yL (about 2 lg) and pMD-TALE-yR (about 2 lg), respectively, with Spe I and Nhe I, ligate the repeats (about 0.3 lg each) with the dephosphorylated pCS2-PEAS and pCS2-PERR (about 0.2 lg each), respectively, before transforming the ligated plasmids into the bacteria to get the TALEN expression plasmids of pCS2-TALEN-yL and pCS2-TALEN-yR [16]. (Note: in vivo, Fok I only functions in form of a heterodimer, thus the two subunits are encoded individually in the pCS2-PEAS and pCS2-PERR constructs). 3. Select positive colonies with the correct insertions. Because the vector plasmids pCS2-PEAS and pCS2-PERR are digested by a single enzyme, Nhe I, the inserted fragments could be in either

Please cite this article in press as: J. Liu et al., Methods (2014), http://dx.doi.org/10.1016/j.ymeth.2014.04.010

6

J. Liu et al. / Methods xxx (2014) xxx–xxx

direction, one of which is of use only. The insertion direction in our case is identified by Nhe I/Hind III double digestion, and only the clone which yields a fragment of 0.48 k contains the correct insertion. Finally, sequencing with SP6 primer is recommended to verify the direction of the insert. 3.4. In vitro transcription of TALENs and preparation of the microinjection solution In principle, there are at least three options to choose with the TALEN pairs as prepared above, in order to mutate a gene of interest in Drosophila: (1) to obtain transgenic flies which carry either the left or the right TALEN, before crossing them to generate flies with both transgenes. Ubiquitous or germ-line specific induction of the TALEN expression initiates a TALEN-mediated mutagenesis in germ cells [6,7,50]; (2) to directly inject the DNA constructs of the TALEN pairs into Drosophila embryos. Mutagenesis is expected to occur upon the expression of injected TALENs in the embryos; (3) to directly inject the in vitro transcribed capped mRNAs of the TALEN pairs into the embryos. This method does not require the step of embryonic transcription and mRNA maturation, and has proven to be highly efficient and time-saving [5]. In our lab, an mRNA injection strategy is used. We prepare mature mRNAs in vitro prior to the injection solution. The procedure is as follows: 1. Linearize the circular left and right TALEN plasmids (in the case of yellow described here, they are pCS2-TALEN-yL and pCS2TALEN-yR). Digest the TALEN DNAs with Not I for 2 h at 37 °C, purify the linear DNA using a DNA gel extraction Kit (Axygen, CAT# AP-GX-250). (Please note that the following steps should be absolutely RNase-free!) 2. Dissolve the DNA pellet in RNase-free water (TaKaRa CAT# D602) to get a final concentration of no less than 150 ng/lL. 3. Capped mRNAs are transcribed using the SP6 mMESSAGE mMACHINE Kit (Ambion, CAT# AM1340) with the linearized TALEN DNAs as templates at 37 °C for 2 h. The reaction is set up as: 5 lL of 2  NTP/CAP (mixture of ribonucleotides and cap analogs), 1 lL of 10  reaction buffer, about 500 ng of linear template DNA, 1 lL of SP6 enzyme mix, and add nuclease-free water to make the reaction to a final total volume of 10 lL. Note: reaction buffer to add after NTP/CAP and water, according to the manufacturer’s instructions. 4. The transcribed mRNAs are recovered by lithium chloride precipitation following the Kit’s instructions. Briefly, add 15 lL RNase-free water and 15 lL LiCl to precipitate the RNAs, vortex to mix thoroughly before keeping the sample at 20 °C for at least 30 min. Centrifuge at maximum speed for 15 min at 4 °C and remove the supernatant. Add 1 mL of 70% ethanol to wash the RNA pellets and remove the supernatant after another maximum-speed spin for 10 min at 4 °C. Resolve RNAs in 20–30 lL RNase-free water. 5. Take 1 lL of the RNA to run on the gel for quality control, and use 1 lL for quantification using a photometer (Eppendorf BioPhotometer). Alternatively, RNA quantity and quality can be tested in a one-step procedure by Bioanalyzer (Agilent, US), as per manufacturer’s instructions. 6. Mix the left and right TALEN mRNAs at a concentration of 250–500 ng/lL each. Usually, a total amount of 15 lL preparation is sufficient for injection of each TALEN pairs. Mix the solution thoroughly and centrifuge. Store the mRNAs at 70 °C. Note: the mRNA preparation can be stored in 3 aliquots, 5 lL each.

modification experiments. Flies should be healthy to allow sufficient numbers of eggs to be collected for injection. Generally speaking, yw or w1118 flies are suitable for most gene disruptions. In our experiment, w1118 flies were used for yellow mutagenesis, in order not to miss any yellow phenotypes upon disruption of the yellow gene. Flies of the yw strain were used specifically for CG9797 disruption. To disrupt genes that are close to the centromere, a FRT fly stock should be used for injection of TALEN mRNA, otherwise the mutations of such genes will be difficult to be recombined with the FRT chromosomes. If the purpose is to tag an endogenous gene, or generate a gene correction, in which the modification of the genome is based on homologous recombination, it is strongly advisable to use the lig4 mutant fly strain for injection (see Section 4.1). The reason for that is that lig4 is essential for the DSB repair pathway of NHEJ, and loss of DNA ligase IV blocks NHEJ and promotes HR, thus increasing the efficiency of precise gene replacement [5]. 3.5.2. Microinjection The standard procedure of microinjection of Drosophila embryos has been described in detail by Bachmann and Knust [51]. We performed our injections with minor modifications (it is highly recommended to wear clean lab coat, mask and gloves, in order to minimize the possibility of mRNA degradation in this procedure): (i) Prepare microinjection needles. We used glass capillary tubes (Narishige, GD-1) to produce microinjection needles. One capillary tube is heated and pulled by a magnetic microelectrode puller (Narishige, PN-30) into two tapering needles, according to the manufacturer’s instructions. (ii) Fill the sharpened injection needle with injection solution. Thaw an aliquot of the injection solution stored at 70 °C, vortex 2–3 s, centrifuge at maximum speed at 4 °C for 5 min, and keep tubes on ice. Immediately, use the Eppendorf Microloader Tips (CAT# 5242 956.003) to load 1.5–2 lL solution into the tapering tip of the prepared injection needle. Keep in mind to return the thawed solution to the 70 °C refrigerator instantly. (iii) Prepare the embryos. Collect the eggs laid within 30 min, dechorionate with sodium hypochlorite solution (8%) for about 40 s, align the dechorionated embryos before transferring onto a slide, dry for about 30 min (depending on the environmental temperature and humidity) before covering of the embryos with the halocarbon oil (Sigma, CAT# H8898). (iv) Adjust the pressure of the microinjector (Eppendorf, CellTram(Oil)) with care prior to injecting the mRNAs into the posterior of the embryos. One may find the liquid diffusing slowly from the tip of the needle. Do not inject too large a volume so to avoid breaking the embryos. (Suggestion: kill all embryos that are older than 2 h). (v) Keep the injected embryos (covered with oil) on the slides inside a moist chamber (to avoid drying out), for development at 18 °C. (vi) About 40 h later, embryos that survived the injection procedure will hatch in short succession. Gently pick up larvae and transfer into a food-containing vial. Place larvae at 18 °C for 1–2 days before transferring to 25 °C. (Note: The whole process of injection should be carried out at 18 °C, including the collection and treatment of embryos).

3.5. Microinjection of TALEN mRNA 3.6. Genetic and molecular identifications of the mutants 3.5.1. Flies Choose appropriate genotype of flies for microinjection according to the aims of the project, or suitable for the desired gene

1. Collect eclosed virgin females and males of the F0 flies (injected). Singly cross each F0 fly with 3–4 balancer flies of the

Please cite this article in press as: J. Liu et al., Methods (2014), http://dx.doi.org/10.1016/j.ymeth.2014.04.010

J. Liu et al. / Methods xxx (2014) xxx–xxx

opposite sex. It is recommended to use double balancers to save time and not to lose any mutant chromosomes. Taking the example of CG9797 located on chromosome 3, we cross F0 flies singly to TM6B, Tb/TM2, y+ flies to balance any chromosomes that might be mutated. For genes located on the 2nd chromosome, balancer flies of BCG/CyO can be used. 2. About 5 days later, when enough F1 animals have been produced, sacrifice the F0 flies for identification of successful mutation. TALEN-mediated mutagenesis can take place in both somatic and germ-line cells. Germ-line mutations produce heritable offsprings, whereas somatic mutations lead to visible (such as change of eye color or cuticle pigmentation or other developmental defects) or molecular phenotypes (such as disruption of restriction enzyme site) in F0, both of which can be viewed as indications of mutation events. In our example of disruption of the yellow gene, 43.8% of F0 male flies exhibited variable-sized yellow patches on their bodies, indicating yellow mutant cell clones (Fig. 4B and C) [16]. In the other example of CG9797 mutagenesis, 58.3% of the F0 male flies carried detectable somatic mutations (Fig. 4A and Table 1). For molecular identification of mutations in F0 flies, we take the CG9797 modification as an example: (i) Extract single fly genomes of the F0 generation. First, prepare extraction mixtures: 0.4 lL proteinase K (20 mg/mL, Calbiochem, CAT# 539480) plus 40 lL Squishing Buffer (SB) (10 mM Tris–HCl, pH 8.2, 1 mM EDTA, 25 mM NaCl and 0.2% Triton X-100); Second, homogenize a single F0 fly in a 0.2 mL tube containing 5 lL of extraction solution by using a pipette tip, before adding the rest of 35 lL extraction mixture. Third, incubate the lysate for 30 min at 37 °C, before heating it for 5 min at 95 °C. (ii) Centrifuge single fly gDNA extract at maximum speed for 2 min before taking 5 lL of each supernatant as template for a 50 lL PCR reaction, using the primer pair that are designed previously (Section 3.2 step 2). The single fly PCR system contains: 5 lL of 10  Taq buffer, 4 lL of dNTP Mix (2.5 mM for each dNTP), 0.4 lM of forward and reverse primers, 5 lL of DNA template, 1.25 U of Taq and distilled water for a total reaction volume of 50 lL. (iii) Purify the ethanol-precipitated PCR products and dissolve them in double distilled water of about 10–20 lL. The PCR products can be directly used for enzyme digestions if the enzyme used is highly efficient, in which case this step can be skipped. (iv) Use approx. 500 ng DNA for Bsp 1286I digestion (cutting the spacer DNA) [16]. To make sure the digestion is complete, it is recommended to incubate overnight at 37 °C. (v) Identify possible F0 flies containing intended mutations by gel electrophoresis. Incomplete digestion indicates that mutagenesis may have successfully occurred in some of the cells of the particular line (Fig. 4A). 3. Collect the virgin females and males of the F1 generation, singly cross them to the opposite sex of balancer flies (the same method as used for F0. For CG9797, it is TM6B, Tb/TM2, y+ flies) to establish stocks. Usually, we randomly pick five males and five female virgins from each F0 for the establishment of stocks. If you have identified mosaic F0, sufficient attention should be paid to their offspring. In the case of yellow, 65.4% of the mosaic male F0 produced mutant offspring, and in the case of CG9797, it was 85.7%. 4. Extract single fly gDNA from F1 flies (or wait for the homozygous F3 flies, which is more time-consuming), amplify by PCR, and digest gDNA just as was done with F0 gDNA to identify/confirm mutations. In this step, the mutations are easier to detect, because

7

if a F1 fly contains the mutation, it is present in 50% of the whole genome, which should be high enough for easy detection. 5. Sequence undigested DNA to confirm the mutations. 3.7. Additional notes for microinjection, genetic and molecular identification of mutations In general, by using TALEN-mediated genomic modifications, one can expect to identify somatic mutations in F0 and inheritable mutations in F1 flies, and stocks can be established in the third generation (F2) (Fig. 4A). Compared with previous HR-based gene targeting [2,3], this is a time-saving strategy, distinguished by high efficiency for mutation generation. A number of practical questions should be addressed here: How many embryos should one inject for mutagenesis, and how many single crosses should be prepared for the screening of mutants. In our hands, 30–50% of the larvae that hatched after injection eclose, and about 60% of the eclosed adults are fertile (Table 1). Among the fertile F0 flies, 17.2–66.7% yielded mutant offsprings [16]. Thus, it is recommended to make no less than 50 F0 single crosses for one mutagenesis procedure. We advise that 8–10 F1 flies from each F0 single cross should be used for stock establishment and molecular identification. 4. Other applications of TALEN-mediated genomic modifications 4.1. TALEN-induced specific local modifications TALEN has been mostly used for gene disruptions. This strategy depends on NHEJ repair of the DSBs generated by a pair of TALENs. In the absence of a homologous sequence, NHEJ repairs the DSB by a simple and imprecise process of trimming and ligation of the broken ends, in the process of which so-called indels are produced. However, when a donor DNA with homology to the DSB-flanking sequence is provided by co-microinjection with TALEN mRNAs, cells may repair the broken DNA via HR using the homologous donor. Based on this principle, one can manipulate a precise substitution of the genome sequence with an artificial exogenous sequence as required. Precise substitution allows us to correct a gene with mutations or to integrate an exogenous DNA into a specific site, etc. (Fig. 1C). TALEN-based gene correction for therapeutic purpose [52] and exogenous GFP knock-in in Drosophila [53] have been reported. For TALEN-mediated HR, several forms of donor DNA can be chosen from: single-strand oligonucleotides [54], and linear or circular double-strand DNA [5]. Oligonucleotide donors are convenient, though they are only suitable for small modifications of the genome, such as corrections of point mutations. In contrast, linear and circular double-strand DNA is more suitable for manipulation of large fragments. In our experiments in Drosophila, we deleted a genomic fragment of precisely 308 bp, including the entire miR-281 locus, by means of TALEN-mediated HR. Briefly, the 1.3 kb upstream and 1.9 kb downstream genomic sequences (excluding the sequences of designed deletion) of miR-281 were amplified and sub-cloned into the pBluescript KS to generate a circular donor. The donor DNA was purified using a plasmid Midi kit (QIAGEN, CAT# 12143), before being co-injected into the lig4 embryos at concentrations of 500 ng/lL for the donor DNA, and 250–500 ng/lL for each TALEN. Molecular confirmation identified 2 precise mutants out of 520 F1 flies [55]. In another experiment in Drosophila, we successfully used a circular DNA as donor, containing 2–3 kb long homologous arms that flanked each side of the TALEN site, to knock in a 5 kb exogenous DNA that included the white coding sequence [55].

Please cite this article in press as: J. Liu et al., Methods (2014), http://dx.doi.org/10.1016/j.ymeth.2014.04.010

8

J. Liu et al. / Methods xxx (2014) xxx–xxx

Fig. 4. Genetic and molecular identification of TALEN-induced genomic mutations. (A) The crossing scheme for the genetic screen for mutations on the 3rd chromosome. Mosaics can be found in F0 flies, and the heritable mutations can be identified from F1 flies. The gel photograph in (A) shows the undigested DNA (indicated by white arrows), indicating the occurrence of mutations in the Drosophila CG9797 locus. (B) The abdomen of wild type (w1118) male flies. (C) The mosaic yellow patches visible at the abdomen of a male F0 fly that was injected at its early embryonic stage with the yellow TALEN mRNAs. P, positive control. N, negative control. Asterisk ( ) indicates the chromosome with a potential mutation. Adapted from Liu and colleagues [16].

Table 1 Statistics of TALEN-induced mutations in the Drosophila CG9797 locus. Gene CG9797 a b

Sex of F0 Male Female

Eclosed flies/hatched larvae (%) 104/288(36.1)

Fertile flies/eclosed flies (%) 67/104(64.4)

Indels-yielding F0/total fertile F0 (%) a

10/15 (66.7) 10/29 (34.5)a

Mosaic F0/total F0 examined (%) 21/36 (58.3)b –

‘‘Indels-yielding’’ F0 represents the flies that produces ‘non-viable’ offsprings. Mosaic F0 describes flies whose DNA cannot be fully digested by the enzyme. Modified from Liu and colleagues [16].

4.2. TALEN induced large-fragment rearrangements In addition to the above-mentioned applications of one pair of TALENs, custom TALENs can also be used in a two-pair manner. As shown in Fig. 1C, simultaneous function of two pairs of TALENs induces two DSBs in the genome. If the two DSBs are located on the same chromosome, large fragment deletion or inversion can be generated. If the DSBs occur on different chromosomes, large fragment translocation can be induced. Recently, Beumer and colleagues have reported a 2.5 kb deletion of the Drosophila ry gene by this strategy [54]. Furthermore, using two pairs of TALENs in a zebrafish model, Gupta and colleagues reported a segmental

deletion of up to 5.5 Mb [56]. Similar work has also been accomplished by Zhang and colleagues [57].

5. Currently available TALE/TALEN resources Since the first reports of customized and innovative TALENs in genomic modifications have been published [17,19,22], the TALE/ TALEN technology has attracted a wide circle of scientists, who have now formed a global network, to provide each other with technical support and a forum for discussion. Many web-based resources, including open access sites for TALE/TALEN target

Please cite this article in press as: J. Liu et al., Methods (2014), http://dx.doi.org/10.1016/j.ymeth.2014.04.010

J. Liu et al. / Methods xxx (2014) xxx–xxx

design, have been established. Some of these useful links are listed here: (i) Target site design: E-TALEN (http://www.e-talen.org); idTALE (http://idtale.kaust.edu.sa/index.html); Mojo Hand (http://www.talendesign.org); TAL Effector-Nucleotide Targeter (https:// boglab.plp.iastate.edu/); TALEN Hit (http://talen-hit.cellectis-bioresearch.com/ search); ZiFit Targeter (http://zifit.partners.org/ZiFiT/); (ii) Prediction of TALEN off-target sites: TALENoffer (http://www.jstacs.de/index.php/TALENoffer) (iii) Reagents: Addgene (http://www.addgene.org/) (iv) Forums: https://groups.google.com/forum/#!forum/taleffectors https://groups.google.com/group/talengineering (v) Other: http://www.talengineering.org/ http://eendb. zfgenetics.org/ 6. Conclusions In this article, we describe the technology of TALEN-mediated genomic modification in Drosophila, and provide a detailed protocol for fly labs, who intend to utilize this approach for genetic manipulation. Because of its simplicity, high efficiency and specificity, versatile applications, TALEN has been widely appreciated by not only Drosophilists, but also biologists from other fields investigating a diverse range of organisms, from plants to humans. It is important to mention here that another gene targeting technique based on the CRISPR/Cas9 system has been developed in Drosophila [58–60], in which a gRNA guides the Cas9 nuclease to a specific site in the genome. It seems possible that the CRISPR/Cas9 system functions with even a higher efficiency than TALEN, though also with higher off-target frequencies [61,62]. Compared with the construction of TALE repeats, gRNA can be easily synthesized, which is certainly an advantage. However, TALE can be fused with different kinds of effectors such as transcription activators, repressors, methyltransferases, histone deacetylases, etc. to be applied in the study of processes such as transcription regulation, epigenetic organization/manipulation of the chromatin. The availability of many choices for construction of TALE repeats, and the easy access of commercially available, customized TALEs should guarantee an ever-increasing field for application of TALE/ TALENs. Therefore, a combination of the powerful genetics of Drosophila with the highly specific and efficient TALE/TALEN system is expected to greatly promote the advancement of functional genomic studies in Drosophila. Acknowledgements We thank Dr. Bo Zhang for providing reagents. We thank Ms. Xuehong Liang for technical assistance, Dr. Torsten Juelich for critical reading of the manuscript, and all members of the Jiao lab for stimulating discussions. This work was supported by the grants from NSFC (Nos. 31201007, 31271573, 31228015 and 31100889) and the 973 Program (2012CB825504). References [1] [2] [3] [4]

K.J. Venken, H.J. Bellen, Nat. Rev. Genet. 6 (2005) 167–178. Y.S. Rong, K.G. Golic, Science 288 (2000) 2013–2018. Y.S. Rong, K.G. Golic, Genetics 157 (2001) 1307–1312. W.J. Gong, K.G. Golic, Proc. Natl. Acad. Sci. U.S.A. 100 (2003) 2556–2561.

9

[5] K.J. Beumer, J.K. Trautman, A. Bozas, J.L. Liu, J. Rutter, J.G. Gall, D. Carroll, Proc. Natl. Acad. Sci. U.S.A. 105 (2008) 19821–19826. [6] K. Beumer, G. Bhattacharyya, M. Bibikova, J.K. Trautman, D. Carroll, Genetics 172 (2006) 2391–2403. [7] M. Bibikova, M. Golic, K.G. Golic, D. Carroll, Genetics 161 (2002) 1169–1175. [8] T.I. Cornu, S. Thibodeau-Beganny, E. Guhl, S. Alwin, M. Eichtinger, J. Joung, T. Cathomen, Mol. Ther. 16 (2008) 352–358. [9] M.L. Maeder, S. Thibodeau-Beganny, J.D. Sander, D.F. Voytas, J.K. Joung, Nat. protoc. 4 (2009) 1471–1501. [10] J. Boch, U. Bonas, Annu. Rev. Phytopathol. 48 (2010) 419–436. [11] A.J. Bogdanove, D.F. Voytas, Science 333 (2011) 1843–1846. [12] J. Boch, H. Scholze, S. Schornack, A. Landgraf, S. Hahn, S. Kay, T. Lahaye, A. Nickstadt, U. Bonas, Science 326 (2009) 1509–1512. [13] M.J. Moscou, A.J. Bogdanove, Science 326 (2009) 1501. [14] J.K. Joung, J.D. Sander, Nat. Rev. Mol. Cell Biol. 14 (2013) 49–55. [15] Y. Zu, X.J. Tong, Z.X. Wang, D. Liu, R.C. Pan, Z. Li, Y.Y. Hu, Z. Luo, P. Huang, Q. Wu, Z.Y. Zhu, B. Zhang, S. Lin, Nat. Methods 10 (2013) 329–331. [16] J. Liu, C. Li, Z. Yu, P. Huang, H. Wu, C. Wei, N. Zhu, Y. Shen, Y. Chen, B. Zhang, W.M. Deng, R. Jiao, J. Genet. Genomics 39 (2012) 209–215. [17] D. Hockemeyer, H. Wang, S. Kiani, C.S. Lai, Q. Gao, J.P. Cassady, G.J. Cost, L. Zhang, Y. Santiago, J.C. Miller, B. Zeitler, J.M. Cherone, X. Meng, S.J. Hinkley, E.J. Rebar, P.D. Gregory, F.D. Urnov, R. Jaenisch, Nat. Biotechnol. 29 (2011) 731– 734. [18] Y.H. Sung, I.J. Baek, D.H. Kim, J. Jeon, J. Lee, K. Lee, D. Jeong, J.S. Kim, H.W. Lee, Nat. Biotechnol. 31 (2013) 23–24. [19] P. Huang, A. Xiao, M.G. Zhou, Z.Y. Zhu, S. Lin, B. Zhang, Nat. Biotechnol. 29 (2011) 699–700. [20] Y. Lei, X. Guo, Y. Liu, Y. Cao, Y. Deng, X. Chen, C.H. Cheng, I.B. Dawid, Y. Chen, H. Zhao, Proc. Natl. Acad. Sci. U.S.A. 109 (2012) 17484–17489. [21] A.J. Wood, T.W. Lo, B. Zeitler, C.S. Pickle, E.J. Ralston, A.H. Lee, R. Amora, J.C. Miller, E. Leung, X.D. Meng, L. Zhang, E.J. Rebar, P.D. Gregory, F.D. Urnov, B.J. Meyer, Science 333 (2011) 307. [22] T. Cermak, E.L. Doyle, M. Christian, L. Wang, Y. Zhang, C. Schmidt, J.A. Baller, N.V. Somia, A.J. Bogdanove, D.F. Voytas, Nucleic Acids Res. 39 (2011) e82. [23] D.F. Carlson, W. Tan, S.G. Lillico, D. Stverakova, C. Proudfoot, M. Christian, D.F. Voytas, C.R. Long, C.B. Whitelaw, S.C. Fahrenkrug, Proc. Natl. Acad. Sci. U.S.A. 109 (2012) 17382–17387. [24] S. Ma, S. Zhang, F. Wang, Y. Liu, H. Xu, C. Liu, Y. Lin, P. Zhao, Q. Xia, PLoS One 7 (2012) e45035. [25] T. Watanabe, H. Ochiai, T. Sakuma, H.W. Horch, N. Hamaguchi, T. Nakamura, T. Bando, H. Ohuchi, T. Yamamoto, S. Noji, T. Mito, Nat. Commun. 3 (2012) 1017. [26] A.L. Smidler, O. Terenzi, J. Soichot, E.A. Levashina, E. Marois, PLoS One 8 (2013) e74511. [27] D. Reyon, C. Khayter, M.R. Regan, J.K. Joung, J.D. Sander, in: M. Frederick et al. (Eds.), Current Protocols in Molecular Biology, 2012, pp. 12.15.1–12.15.14 (Chapter 12, Unit 12.15). [28] Y. Zu, X. Tong, Z. Wang, D. Liu, R. Pan, Z. Li, Y. Hu, Z. Luo, P. Huang, Q. Wu, Z. Zhu, B. Zhang, S. Lin, Nat. Methods 10 (2013) 329–331. [29] Z. Cheng, P. Yi, X. Wang, Y. Chai, G. Feng, Y. Yang, X. Liang, Z. Zhu, W. Li, G. Ou, Nat. Biotechnol. 31 (2013) 934–937. [30] C. Engler, R. Gruetzner, R. Kandzia, S. Marillonnet, PLoS One 4 (2009) e5553. [31] T. Sakuma, S. Hosoi, K. Woltjen, K. Suzuki, K. Kashiwagi, H. Wada, H. Ochiai, T. Miyamoto, N. Kawai, Y. Sasakura, S. Matsuura, Y. Okada, A. Kawahara, S. Hayashi, T. Yamamoto, Genes Cells 18 (2013) 315–326. [32] N.E. Sanjana, L. Cong, Y. Zhou, M.M. Cunniff, G. Feng, F. Zhang, Nat. protoc. 7 (2012) 171–192. [33] D. Reyon, S.Q. Tsai, C. Khayter, J.A. Foden, J.D. Sander, J.K. Joung, Nat. Biotechnol. 30 (2012) 460–465. [34] A.W. Briggs, X. Rios, R. Chari, L. Yang, F. Zhang, P. Mali, G.M. Church, Nucleic Acids Res. 40 (2012) e117. [35] Z. Wang, J. Li, H. Huang, G.C. Wang, M.J. Jiang, S.Y. Yin, C.H. Sun, H.S. Zhang, F.F. Zhuang, J.J. Xi, Angew. Chem. Int. Ed. Engl. 51 (2012) 8505–8508. [36] J.L. Schmid-Burgk, T. Schmidt, V. Kaiser, K. Honing, V. Hornung, Nat. Biotechnol. 31 (2013) 76–81. [37] J. Crocker, D.L. Stern, Nat. Methods 10 (2013) 762–767. [38] A.C. Mercer, T. Gaj, R.P. Fuller, C.F. Barbas, Nucleic Acids Res. 40 (2012) 11163– 11172. [39] Y. Miyanari, C. Ziegler-Birling, M.E. Torres-Padilla, Nat. Struct. Mol. Biol. 20 (2013) 1321–1324. [40] H. Stocker, P. Gallant, Methods Mol. Biol. 420 (2008) 27–44. [41] Y. Yu, J. Streubel, S. Balzergue, A. Champion, J. Boch, R. Koebnik, J. Feng, V. Verdier, B. Szurek, Mol. Plant Microbe Interact. 24 (2011) 1102–1113. [42] E.L. Doyle, N.J. Booher, D.S. Standage, D.F. Voytas, V.P. Brendel, J.K. Vandyk, A.J. Bogdanove, Nucleic Acids Res. 40 (2012) W117–W122. [43] J.C. Miller, S. Tan, G. Qiao, K.A. Barlow, J. Wang, D.F. Xia, X. Meng, D.E. Paschon, E. Leung, S.J. Hinkley, G.P. Dulay, K.L. Hua, I. Ankoudinova, G.J. Cost, F.D. Urnov, H.S. Zhang, M.C. Holmes, L. Zhang, P.D. Gregory, E.J. Rebar, Nat. Biotechnol. 29 (2011) 143–148. [44] C. Mussolino, R. Morbitzer, F. Lutge, N. Dannemann, T. Lahaye, T. Cathomen, Nucleic Acids Res. 39 (2011) 9283–9293. [45] M.L. Christian, Z.L. Demorest, C.G. Starker, M.J. Osborn, M.D. Nyquist, Y. Zhang, D.F. Carlson, P. Bradley, A.J. Bogdanove, D.F. Voytas, PLoS One 7 (2012) e45383. [46] J. Streubel, C. Blucher, A. Landgraf, J. Boch, Nat. Biotechnol. 30 (2012) 593–595. [47] L. Cong, R. Zhou, Y.C. Kuo, M. Cunniff, F. Zhang, Nat. Commun. 3 (2012) 968. [48] R. Morbitzer, P. Romer, J. Boch, T. Lahaye, Proc. Natl. Acad. Sci. U.S.A. 107 (2010) 21617–21622.

Please cite this article in press as: J. Liu et al., Methods (2014), http://dx.doi.org/10.1016/j.ymeth.2014.04.010

10

J. Liu et al. / Methods xxx (2014) xxx–xxx

[49] M.R. Green, J. Sambrook, Molecular Cloning: A Laboratory Manual, fourth ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 2012. [50] M. Bibikova, K. Beumer, J.K. Trautman, D. Carroll, Science 300 (2003) 764. [51] A. Bachmann, E. Knust, Methods Mol. Biol. 420 (2008) 61–77. [52] M.J. Osborn, C.G. Starker, A.N. McElroy, B.R. Webber, M.J. Riddle, L. Xia, A.P. DeFeo, R. Gabriel, M. Schmidt, C. von Kalle, D.F. Carlson, M.L. Maeder, J.K. Joung, J.E. Wagner, D.F. Voytas, B.R. Blazar, J. Tolar, Mol. Ther. 21 (2013) 1151– 1159. [53] T. Katsuyama, A. Akmammedov, M. Seimiya, S.C. Hess, C. Sievers, R. Paro, Nucleic Acids Res. 41 (2013) e163. [54] K.J. Beumer, J.K. Trautman, M. Christian, T.J. Dahlem, C.M. Lake, R.S. Hawley, D.J. Grunwald, D.F. Voytas, D. Carroll, G3 3 (2013) 1717–1725. [55] Z. Yu, H. Chen, J. Liu, H. Zhang, Y. Yan, N. Zhu, Y. Guo, B. Yang, Y. Chang, F. Dai, X. Liang, Y. Chen, Y. Shen, W.M. Deng, J. Chen, B. Zhang, C. Li, R. Jiao, Biol. Open 3 (2014) 271–280.

[56] A. Gupta, V.L. Hall, F.O. Kok, M. Shin, J.C. McNulty, N.D. Lawson, S.A. Wolfe, Genome Res. 23 (2013) 1008–1017. [57] A. Xiao, Z. Wang, Y. Hu, Y. Wu, Z. Luo, Z. Yang, Y. Zu, W. Li, P. Huang, X. Tong, Z. Zhu, S. Lin, B. Zhang, Nucleic Acids Res. 41 (2013) e141. [58] Z. Yu, M. Ren, Z. Wang, B. Zhang, Y.S. Rong, R. Jiao, G. Gao, Genetics 195 (2013) 289–291. [59] A.R. Bassett, C. Tibbit, C.P. Ponting, J.L. Liu, Cell Rep. 4 (2013) 220–228. [60] S.J. Gratz, A.M. Cummings, J.N. Nguyen, D.C. Hamm, L.K. Donohue, M.M. Harrison, J. Wildonger, K.M. O’Connor-Giles, Genetics 194 (2013) 1029–1035. [61] Y.F. Fu, J.A. Foden, C. Khayter, M.L. Maeder, D. Reyon, J.K. Joung, J.D. Sander, Nat. Biotechnol. 31 (2013) 822–826. [62] P.D. Hsu, D.A. Scott, J.A. Weinstein, F.A. Ran, S. Konermann, V. Agarwala, Y.Q. Li, E.J. Fine, X.B. Wu, O. Shalem, T.J. Cradick, L.A. Marraffini, G. Bao, F. Zhang, Nat. Biotechnol. 31 (2013) 827–832.

Please cite this article in press as: J. Liu et al., Methods (2014), http://dx.doi.org/10.1016/j.ymeth.2014.04.010