Abstract
We report a simple yet extremely efficient platform for systematic gene targeting by the RNA-guided endonuclease Cas9 in Drosophila. The system comprises two transgenic strains: one expressing Cas9 protein from the germline-specific nanos promoter and the other ubiquitously expressing a custom guide RNA (gRNA) that targets a unique site in the genome. The two strains are crossed to form an active Cas9–gRNA complex specifically in germ cells, which cleaves and mutates the target site. We demonstrate rapid generation of mutants in seven neuropeptide and two microRNA genes in which no mutants have been described. Founder animals stably expressing Cas9–gRNA transmitted germline mutations to an average of 60% of their progeny, a dramatic improvement in efficiency over the previous methods based on transient Cas9 expression. Simultaneous cleavage of two sites by co-expression of two gRNAs efficiently induced internal deletion with frequencies of 4.3–23%. Our method is readily scalable to high-throughput gene targeting, thereby accelerating comprehensive functional annotation of the Drosophila genome.
PHENOTYPES of mutant animals have provided deeper insight into gene function than any other means. In Drosophila melanogaster, hundreds of genetic screens have been conducted over the last century aiming to isolate new mutants. Even today, however, loss-of-function mutants in the vast majority of genes are yet to be described, calling for a breakthrough technology that allows systematic gene targeting to complete our understanding of the Drosophila genome. Targeted gene disruption in Drosophila has been conventionally achieved by deletion of flanking sequences by imprecise excision of transposons or, more recently, by gene replacement using homologous recombination (Rong and Golic 2000; Ryder and Russell 2003). Each of these techniques, however, has its own limitations: Although imprecise excision of transposons is moderately efficient, it is limited by its requirement of an existing transposon near the target locus, which is missing for quite a few loci (Ryder and Russell 2003). Homologous recombination suffers from extremely low efficiency, often requiring the screening of more than 1 × 106 flies (Huang et al. 2009).
Designer nucleases are a promising new technology that offers an alternative route to gene targeting. They represent genetically encoded nucleases that can be programmed to target an arbitrary sequence. Heterologously expressed designer nucleases cause a double-strand break at a specified target sequence in the genome, which gives rise to an insertion-deletion (indel) mutation through inaccurate DNA repair involving nonhomologous end joining. Zinc-finger nucleases (ZFNs) and TALE nucleases (TALENs) are heterodimeric nucleases with a programmable DNA-binding domain assembled from DNA-binding peptides (Kim et al. 1996; Miller et al. 2011). They have been widely used to mutate genes in a broad range of animal species. Despite their demonstrated high efficiency (Bibikova et al. 2002; Liu et al. 2012), ZFNs and TALENs have not received as much enthusiasm in Drosophila research as they have in other model organisms, most likely due to the difficulty of assembling DNA-binding domains and microinjecting nuclease messenger RNA into embryos.
CRISPR/Cas9 is the latest addition to the designer nuclease family. Originally identified as a component of the CRISPR bacterial innate immunity system, Cas9 was found to encode a novel class of sequence-specific endonuclease whose target specificity is determined by its guide-RNA (gRNA) component (Jinek et al. 2012; Cong et al. 2013; Mali et al. 2013). Cas9 can be easily reprogrammed to target a new sequence simply by replacing the specificity-determining 20-bp sequence of the gRNA. Constraints to selecting suitable 20-bp targets are minimal, the only requirements being that they must be followed by an “NGG” sequence known as the protospacer adjacent motif (PAM) (Jinek et al. 2012). The ease and flexibility of the design make Cas9 an ideal tool for genome-scale applications. Cas9 has been reported to be extremely effective when injected into zygotes of zebrafish and mice, with somatic mutation frequencies exceeding 50% (Hwang et al. 2013; Wang et al. 2013). Attempts to adapt Cas9 to genome engineering in Drosophila were recently reported, in which plasmid vectors or in vitro-transcribed RNAs encoding Cas9 and gRNA were injected into fertilized eggs to transiently express Cas9–gRNA (Bassett et al. 2013; Gratz et al. 2013; Yu et al. 2013). The reported procedures, however, resulted in somewhat variable germline mutation frequencies, calling for a more robust system in which any gene can be mutated with high efficiency. In the present study, we describe a method of deriving mutant progeny from transgenic flies stably expressing Cas9–gRNA in germ cells, with a dramatic improvement in consistency and efficiency.
Materials and Methods
Plasmid construction
Standard molecular biology techniques were used to construct plasmid vectors. We first constructed a general transformation vector pBFv, which has an attB sequence for site-specific integration by phiC31 integrase and a vermilion (v) gene as a visible marker. A 1264-bp promoter sequence and a 965-bp, 3′-UTR-containing sequence of the nanos gene, which have been shown to drive highly specific germline expression (Van Doren et al. 1998), were cloned into pBFv to create the germline-expression vector pBFv-nosP. A Cas9 cDNA fragment was cut from hCas9 (Mali et al. 2013) by XbaI/AgeI and cloned into pBFv-nosP to create pBFv-nosP-Cas9. The general gRNA expression vector pBFv-U6.2 was constructed by cloning a 399-bp promoter sequence of the Drosophila snRNA:U6:96Ab gene into pBFv. To construct a gRNA expression vector for each target gene, two complementary 24-bp oligonucleotides with a 20-bp target sequence were annealed to generate a double-strand DNA with 4-bp overhangs on both ends and cloned into BbsI-digested pBFv-U6.2 (Figure S2). Six gRNA vectors targeting the white gene and seven gRNA vectors targeting different neuropeptide genes were generated. They were named pBFv-U6.2-“gRNA ID” (pBFv-U6.2-w-ex3-1, pBFv-U6.2-w-ex6-1, pBFv-U6.2-Ast-1, etc.).
To construct a double-gRNA vector in which two different gRNAs are separately expressed from their own U6 promoters, a first gRNA was cloned into pBFv-U6.2 and a second gRNA into pBFv-U6.2B, a variant of pBFv-U6.2 that has a dummy sequence flanked by EcoRI and NotI sites upstream of the U6 promoter (Figure S1). A fragment containing the U6 promoter and the first gRNA was cut from pBFv-U6.2-gRNA#1 by EcoRI and NotI and ligated with pBFv-U6.2B-gRNA#2 linearized with EcoRI and NotI. Double-gRNA vectors targeting white, mir-219, and mir-315 were constructed. They were named pBFv-U6.2x2-w, pBFv-U6.2x2-mir-219, and pBFv-U6.2x2-mir-315.
The sequences of the oligonucleotides used to construct each gRNA vectors are shown in Table S1.
Fly transformation
All vectors were integrated into the attP40 landing site on the second chromosome by phiC31 integrase (Markstein et al. 2008). Plasmid DNA (100 ng/µl) was injected into the y1 v1 nos-phiC31; attP40 host (Bischof et al. 2007). Surviving G0 males were individually crossed to y2 cho2 v1 virgins. In a cho2 v1 background, a v+ transgene turns the eye color from light orange to dark brown, making it easier to identify transformants than in a v1 background. A single male transformant from each cross was mated to y2 cho2 v1; Sp/CyO virgins. Offspring in which the transgene was balanced were collected to establish a stock.
Two vectors, pBFv-nos-Cas9 and pBFv-U6.2-w-ex3-1, were individually injected. The other vectors were injected in three pools of multiple vectors: The first pool contained the remaining five gRNA vectors targeting the white gene. The second pool contained the seven vectors targeting neuropeptide genes. The third pool contained the three double-gRNA vectors. After establishing lines from individual F1 transformants, PCR analysis was performed to identify which vector was integrated in each line. For all three pools, screening of 16 lines was sufficient to recover at least one transgenic line for each of the pooled vectors.
Fly genetics
All the U6-gRNA lines used in this study have the y2 cho2 v1; attP40{U6-gRNA}/CyO genotype. The nos-Cas9 line has the y2 cho2 v1; attP40{nos-Cas9}/CyO genotype.
Seven U6-gRNA lines—U6.2-w-ex3-1, U6.2-w-ex3-2, U6.2-w-ex3-3, U6.2-w-ex3-4, U6.2-w-ex3-5, U6.2-w-ex6-1, and U6.2x2-w—were used to induce mutations at the white locus. Females carrying a U6-gRNA transgene were crossed to nos-Cas9 males to obtain founder animals that have both the U6-gRNA and the nos-Cas9 transgenes. Each male founder was crossed to four virgin females carrying the compound-X chromosome. We used the compound-X chromosome strain Bx3/C(1)DX y1 w1 f1. Female founders were individually put in a vial after being allowed to mate with the siblings. Sixteen to 18 crosses were set up for each experimental condition. The number of white mutants in the total progeny from each cross was counted to estimate mutation frequency. Crosses that produced <10 offspring were excluded from analysis. The female germline was not examined for U6.2x2-w.
To induce mutations at the neuropeptide and microRNA (miRNA) genes, the following U6-gRNA lines were used: U6.2-Ast-1, U6.2-capa-1, U6.2-Ccap-1, U6.2-Crz-1, U6.2-Eh-1, U6.2-Mip-1, U6.2-npf-1, U6.2x2-mir-219, and U6.2x2-mir-315. Founder animals were obtained by crossing U6-gRNA females to nos-Cas9 males. Fifteen to 20 male founders were crossed en masse to 15–20 w; Dr/TM6B virgins. Genomic DNA was extracted from each of the resultant offspring and used for molecular characterization.
PCR amplification of target loci
To molecularly characterize induced mutations, target loci were first amplified by PCR using genomic DNA extracted from individual flies. Each fly was crushed in 50 µl of DNA extraction buffer (50 mM NaOH, 0.5% Triton-X). The lysate was heated at 95° for 5 min, cooled down to 4°, and immediately neutralized with 50 µl of 10 mM Tris (pH 8.0). Using 0.5 µl of the crude DNA extract as a template, a DNA sequence surrounding the target site was amplified by PCR for 35–40 cycles in a 10-µl reaction with 0.2 U of KOD FX Neo (Toyobo). The PCR product was analyzed by agarose gel electrophoresis, the T7 endonuclease I (T7EI) assay, and DNA sequencing. Primers used for PCR and DNA sequencing are listed in Table S1.
T7EI assay
To identify animals with a heterozygous mutation, we used T7EI, a structure-specific nuclease that cleaves DNA heteroduplexes at mismatch sites. First, genomic DNA of each fly was amplified by PCR for 40 cycles as described. If a fly carries a heterozygous mutation at the target site, the PCR product forms heteroduplex DNA with a mismatch in later cycles of PCR after DNA amplification levels off. Two microliters of the PCR product was directly treated with 1 U of T7 endonuclease I (New England Biolabs) in a total volume of 10 µl. After incubation for 15 min at 37°, the sample was immediately put on ice to terminate enzyme reaction. Timely termination is critical, as prolonged incubation results in degradation of cleaved DNA fragments. The sample was directly analyzed by electrophoresis on a 2–3% agarose gel in Tris-acetate-EDTA buffer. Of 54 samples analyzed by both the T7EI assay and DNA sequencing, 29 had an indel mutation detected by both assays. Three mutations identified by DNA sequencing were not detectable in the T7EI assay.
Estimation of off-target mutation frequency
Male animals of the genotype y cho v/Y; nos-Cas9/U6-gRNA (G0) were crossed to */FM7 virgins. Female offspring (F1) of the genotype y cho v/FM7 were individually mated to FM7/Y males in separate vials. Absence of y cho v/Y males in their progeny (F2) indicates that a lethal mutation was induced on the X chromosome in the G0 germline. Approximately 200 F1 females were scored to estimate the frequency of de novo lethal mutations on the X chromosome. Four gRNAs targeting autosomal genes (capa, Ccap, Mip, npf) were tested. The results are shown in Table S2.
Results
To express Cas9–gRNA in germ cells of Drosophila, we chose to establish transgenic flies carrying genomic sources of the enzyme components. While transgenes could be transiently expressed by microinjection of plasmid DNA or in vitro-transcribed RNA, previous observations indicate that genomic sources of various DNA-modifying enzymes produce significantly higher activity than injected DNA or RNA in Drosophila (Robertson et al. 1988; Keeler et al. 1996; Bischof et al. 2007; Bateman and Wu 2008; Holtzman et al. 2010). We constructed two transformation vectors that expressed Cas9 protein or a custom gRNA (Figure 1 and Supporting Information, Figure S1). Cas9 is expressed specifically in germ cells by the promoter of the nanos (nos) gene (Van Doren et al. 1998), while gRNA is expressed from a ubiquitous pol III promoter derived from the upstream sequence of a Drosophila U6 small nuclear RNA gene (Wakiyama et al. 2005). The gRNA vector has a cloning site that allows seamless cloning of a 20-bp specificity-determining sequence by accepting annealed oligonucleotides (Figure S2). We separately integrated the Cas9 and gRNA vectors into a fixed landing site in the Drosophila genome by phiC31 integrase (Bischof et al. 2007; Markstein et al. 2008). Transgenic animals carrying nos-Cas9 or U6-gRNA were fully viable and fertile, indicating that neither Cas9 nor gRNA alone have a deleterious effect on cells. The nos-Cas9 and U6-gRNA lines are crossed to obtain founder animals that expressed an active Cas9–gRNA complex specifically in the germline, thereby inducing mutations in target genes. Once a mutant is obtained, the integrated nos-Cas9 and U6-gRNA transgenes can be conveniently removed by crossing the mutant into a yellow or vermilion mutant background, as they are marked with yellow and vermilion transgenes.
Schematic overview of the transgenic Cas9–gRNA system. For each gene of interest, a transgenic strain that ubiquitously expresses a gRNA targeting the gene is established. The gRNA strain is crossed to the nos-Cas9 strain, which expresses Cas9 protein specifically in germ cells. The founder animals obtained from the cross express active Cas9–gRNA nuclease complex specifically in germ cells. Mutation is induced at a certain frequency in the founder germline and transmitted to the next generation.
To test the efficacy of our system, we first attempted to mutate the sex-linked white locus, in which mutants have white eyes due to complete lack of pigment. We established transgenic lines each expressing one of six gRNAs designed to target different locations of the white gene (Figure 1). They were crossed to nos-Cas9 to obtain founder animals carrying both the nos-Cas9 and U6-gRNA transgenes. We estimated mutation frequency in the germline of these founder animals. The male and female germlines were individually examined because of potential sex-dependent differences in choice of DNA repair pathways because the X chromosome is hemizygous in Drosophila. We crossed individual male founders to females with a compound-X chromosome, such that their male progeny inherit the X chromosome from the father. Female founders were individually crossed to male siblings. The male progeny from each cross were scored for loss of eye color.
Each of the six gRNAs was highly effective in inducing white mutants. All founder males expressing Cas9–gRNA produced at least one white mutant offspring. Germline mutation frequencies, calculated as a percentage of white mutants in total progeny, exceeded 85% in four of the gRNAs. The other two were less effective with mutation frequencies of 12 and 57%. In the female germline, mutation frequencies were consistently lower than in the male germline, ranging from 3.4 to 93% (Figure 2B). In the progeny of female founders, we often obtained mosaic mutant animals of both sexes in which one of the eyes or part of an eye is mutant. In these animals, mutations presumably arose in somatic cells early during development by maternally deposited Cas9 and gRNA. Generation of white mutations was dependent on gRNAs targeting the white gene: a gRNA targeting an unrelated gene Ccap did not produce any white mutants (Figure 2B). To further confirm the target specificity of the Cas9–gRNA at the molecular level, we sequenced the white locus in obtained mutants. In all mutant animals examined, we identified small indel mutations encompassing the target site (Figure 2B and Figure S3).
The Cas9-induced deletions were typically small, ranging in size from 1 to 20 bp (Figure 2C and Figure S3). We asked if larger deletions could be induced by simultaneous cleavage of two sites by Cas9–gRNA. We constructed a transformation vector that contains two U6-gRNA transgenes (Figure 2D; see Materials and Methods for details). Each of the two gRNAs, w-ex3-1 and w-ex6-1, targeted one of two sites separated by 1.6 kb (Figure 2, A and E). From the cross between male flies carrying double-gRNA and nos-Cas9 and compound-X females, white mutants were obtained in the progeny at a frequency of 91% (Figure 2D). Of the 94 white mutant animals examined by PCR, 13 (14%) had deletions of various sizes (Figure 2D). Breakpoint sequences of the deletions found that five of them completely uncovered the region between the two target sites (Figure 2E and Figure S4), representing ∼5% of total progeny.
Mutagenesis of the white locus by transgenic Cas9–gRNA. (A) Schematic of the white locus. Exons are shown as boxes. Coding regions are depicted in black and the 5′- and 3′-UTRs in gray. Locations and sequences of the gRNA targets are indicated with the PAM in green. (B) Frequencies of mutations induced by Cas9–gRNA. For each gRNA, the mutation frequency is shown as a percentage of white mutants in total progeny. The percentage of founders producing at least one mutant offspring (yielders) is also shown. Male and female germlines were separately scored. (C) Sequences of mutations induced by the gRNA w-ex3-1. The wild-type sequence is shown at the top with the gRNA target sequence highlighted by a top line. The cleavage site is indicated by an arrowhead. Deleted residues are shown as dashes. Inserted residues are shown in blue lowercase letters. The indel size and the number of occurrences are shown on the right of each mutant sequence. (D) PCR analysis of internal deletions induced by simultaneous expression of two gRNAs. Locations of primers and cleavage sites are shown in E. The wild type produces a band of 2.3 kb, while complete deletion between the two target sites produces a band of 0.7 kb. (E) The extent of deletions in D was determined by sequencing the breakpoints. Primer locations and cleavage sites are indicated by arrowheads and arrows, respectively.
To investigate if other genes could be mutated with similarly high efficiency, we extended our analysis to autosomal genes. We chose to target neuropeptide and miRNA genes because most of these genes have no reported mutant alleles despite their apparent importance in development, physiology, and behavior. We generated U6-gRNA transgenic lines that target each of seven neuropeptide genes: Allatostatin (Ast), Eclosion hormone (Eh), Cardioacceleratory peptide (Ccap), capability (capa), Corazonin (Crz), Myoinhibiting peptide (Mip), and neuropeptide F (npf). Males carrying nos-Cas9 and a gRNA transgene were crossed to wild-type flies by mass mating, and their progeny were screened for the presence of a heterozygous indel mutation by the T7EI assay, which detects mismatches in heteroduplex DNA (see Materials and Methods for details). We obtained T7EI-positive progeny at frequencies of 8.7–98% for each target gene (Figure 3, A and B). The results of the T7EI assay were further confirmed by DNA sequencing (Figure 3C and Figure S5). To disrupt two miRNA genes, mir-219 and mir-315, we took the double-gRNA approach, thereby generating deletions that uncover their entire sequences because single gRNAs could not be designed inside these genes. For each of the two miRNA genes, two gRNAs were designed to delete 129- and 369-bp sequences, respectively. Complete gene deletion was observed in 7 of 39 and 4 of 94 offspring, respectively (Figure 3, D and E; Figure S6).
Mutagenesis of neuropeptide and miRNA genes. (A) Identification of heterozygous mutants by the T7EI assay. Each panel shows PCR products amplified from a wild-type animal and a representative heterozygous mutant treated or not treated with T7EI. Sizes of the untreated PCR product and the cleavage products are shown on the left side of each panel in A. (B) Frequencies of Cas9-induced mutations in neuropeptide genes. The percentage of samples that were positive in the T7EI assay is shown for each locus. (C) Sequencing chromatograms of a wild-type allele and a heterozygous mutant. The presence of induced mutations in T7EI-positive animals was confirmed by direct sequencing of PCR products. The sequence of the mutant allele was inferred by subtracting a wild-type sequence from the mixed sequence. The deleted sequence is highlighted in yellow. (D) Schematic of the miRNA loci targeted using the double-gRNA method. Arrows and arrowheads indicate primer locations and cleavage sites, respectively. (E) Identification of heterozygous deletion mutants in mir-219 and mir-315 genes by PCR. Examples of wild-type and mutant PCR products are shown. Heterozygous deletion results in appearance of an additional band of a smaller size. The mutation frequency is shown at the top of each gel image.
A major concern in mutagenesis by designer nucleases is off-target effect, whereby unintended sequences are haphazardly cleaved and mutated. It has been shown that Cas9 is capable of cleaving sequences that have multiple mismatches with the gRNA sequence with moderate efficiency (Fu et al. 2013; Hsu et al. 2013; Pattanayak et al. 2013). To estimate the extent of off-target effect in our system, we chose four gRNAs targeting different autosomal genes and tested if they induced unintended sex-linked lethal mutations. For each gRNA, we scored X chromosomes in female offspring of founder animals carrying nos-Cas9 and U6-gRNA. We found no sex-linked lethal mutations in a total of 766 X chromosomes examined (Table S2), suggesting that off-target mutations that disrupt gene function are rare.
Discussion
Here we have demonstrated that transgenic germline expression of Cas9–gRNA has the capacity to induce targeted mutations with extremely high efficiencies. In addition to the X-linked white gene in which mutation leads to a visible phenotype, we demonstrated that our method could indeed be applied to mutating autosomal genes whose mutant phenotypes were unknown. The overall mutation frequencies obtained using our system were much higher than the previously reported procedures based on transient expression of Cas9 and gRNA from injected DNA or RNA: The average mutation frequency of all gRNAs in our study was 57%, while those in the previous studies were 19% (Bassett et al. 2013), 29% (Yu et al. 2013), and 1% (Gratz et al. 2013). It should be noted, however, that direct comparison between the previous studies and ours is difficult as gRNAs with different sequences were used.
We further showed that simultaneous expression of two gRNAs efficiently induced internal deletion between the two target sites. The technique is useful not only for making large deletions but also for targeting very small genes, such as miRNAs, in which no gRNAs can be designed due to sequence constraints. Interestingly, the effect of the simultaneous expression of two gRNAs on mutation frequency was more than additive. While the gRNAs w-ex3-1 and w-ex6-1 induced white mutants at frequencies of 12 and 57%, respectively, in the male germline (Figure 2B), simultaneous expression of the two resulted in a mutation frequency of 91% (Figure 2D). Thus the double-gRNA approach will also be useful in cases where utmost mutation rates are required.
Differences in the male and female germlines should also be noted from a practical standpoint. Whatever their underlying molecular mechanisms, mutation frequencies in the female germline were often much lower than in the male germline (Figure 2B). In addition, somatic mosaic mutants were frequently produced by maternally deposited Cas9–gRNA, which would complicate identification of true mutants derived from parental germline mutations. In actual gene-targeting experiments using our system, it is therefore desirable to induce mutations in the male germline.
Recent studies have shown that Cas9 accommodates multiple mismatches in the 20-bp target sequence, raising concerns about off-target effect (Fu et al. 2013; Hsu et al. 2013; Pattanayak et al. 2013). We showed that none of the four gRNAs targeting autosomal genes were able to induce any sex-linked lethal mutations in ∼800 gametes examined. Given that the spontaneous mutation rate for sex-linked lethals is 0.2% (Wallace 1970), nonspecific deleterious mutations induced by Cas9 are extremely rare. Nonetheless, it should be borne in mind that possible existence of off-target mutations cannot be entirely excluded. Fortunately, Drosophila genetics offers various tools to confirm mutant phenotypes, such as chromosomal deficiencies and transgenic rescue.
Another major application of the designer nuclease technologies is precise modification of the genome using donor templates by homology-directed repair. Gap repair of transposon-induced double-strand breaks has been successfully used for targeted gene replacement (Gloor et al. 1991; Keeler et al. 1996; Lee et al. 2006). More recently, it has been shown that simple injection of a template DNA together with zinc-finger nucleases or Cas9–gRNA induces incorporation of the template sequence into the genome, allowing introduction of single-base-pair changes, deletions with defined breakpoints, and insertion of foreign DNA sequences (Beumer et al. 2008; Gratz et al. 2013). It would be possible to induce homology-directed repair by injecting a donor template into fertilized eggs that carry the nos-Cas9 and gRNA transgenes.
Our method requires generation of transgenic lines for each target gene. Although it takes an additional time of at least 1 month compared to direct injection of DNA or RNA encoding Cas9 and gRNA genes, we presume that the high mutation frequency of our method more than compensates for the extra time. Large-scale applications will further benefit from the transgenic approach as the time-consuming injection process can be drastically reduced by pooling multiple plasmid vectors (Bischof et al. 2013). Our approach can even obviate the need for microinjection experiments altogether by outsourcing transgenic production, which is currently the standard method of choice in Drosophila research.
The extremely high mutagenesis efficiency and the ease with which U6-gRNA vectors and transgenic lines are constructed make our methodology an ideal platform for generating genome-scale mutant resources in Drosophila. We and others have successfully generated genome-wide collections of transgenic RNA interference lines and distributed them to the community (Leulier et al. 2002; Dietzl et al. 2007; Ni et al. 2011). Likewise, it is highly feasible to generate transgenic U6-gRNA lines targeting all genes and derive mutants from them in a reasonable time frame.
The transgenic strains and vectors used in the present study will be made available through the NIG-Fly Stock Center (http://www.shigen.nig.ac.jp/fly/nigfly). We have set up a dedicated website for our Cas9 method (http://www.shigen.nig.ac.jp/fly/nigfly/cas9).
Acknowledgments
We thank all members of the Ueda lab for generating transgenic flies and for their help with fly maintenance and research experiments and the Bloomington Stock Center for providing fly stocks.
Footnotes
Communicating editor: C.-ting Wu
Literature Cited
Author notes
Supporting information is available online at http://www.genetics.org/lookup/suppl/doi:10.1534/genetics.113.156737/-/DC1.
