Mar Biotechnol (2014) 16:125–134 DOI 10.1007/s10126-013-9556-6
ORIGINAL ARTICLE
Parameters and Efficiency of Direct Gene Disruption by Zinc Finger Nucleases in Medaka Embryos Xi Zhang & Guijun Guan & Jianbin Chen & Kiyoshi Naruse & Yunhan Hong
Received: 18 April 2013 / Accepted: 15 July 2013 / Published online: 23 October 2013 # Springer Science+Business Media New York 2013
Abstract Zinc finger nucleases (ZFNs) can generate targeted gene disruption (GD) directly in developing embryos of zebrafish, mouse and human. In the fish medaka, ZFNs have been attempted on a transgene. Here, we developed procedures and parameters for ZFN-mediated direct GD on the gonad-specifically expressed gsdf locus in medaka. A pair of ZFNs was designed to target the first exon of gsdf and their synthetic mRNAs were microinjected into 1-cell stage embryos. We reveal dose-dependent survival rate and GD efficiency. In fry, ZFN mRNA injection at 10 ng/μl led to a GD efficiency of 30 %. This value increased up to nearly 100 % when the dose was enhanced to 40 ng/μl. In a typical series of experiments of ZFN mRNA injection at 10 ng/μl, 420 injected embryos developed into 94 adults, 4 of which had altered gsdf alleles. This leads to a GD efficacy of ∼4 % in the adulthood. Sequencing revealed a wide variety of subtle allelic alterations including additions and deletions of 1∼18 bp in length in ZFN-injected samples. Most importantly, one of the 4 adults examined was capable of germline transmission to 15.2 % of its F1 progeny. Interestingly, ontogenic analyses of the allelic profile revealed that GD commenced early in development, continued during subsequent stages of development and in primordia for different adult organs of the three germ layers. Xi Zhang and Guijun Guan contributed equally to this work. Electronic supplementary material The online version of this article (doi:10.1007/s10126-013-9556-6) contains supplementary material, which is available to authorized users. X. Zhang : G. Guan : J. Chen : Y. Hong (*) Department of Biological Sciences, National University of Singapore, 14 Science Drive 4, Singapore 117543, Singapore e-mail: [email protected] G. Guan : K. Naruse (*) Department of Bioresources, National Institute for Basic Biology, Okazaki, Aichi 444-8585, Japan e-mail: [email protected]
These results demonstrate the feasibility and—for the first time to our knowledge—the efficacy of ZFN-mediated direct GD on a chromosomal gene in medaka embryos. Keywords Gene disruption . Gene targeting . Genome editing . gsdf . Zinc finger nuclease . ZFN Abbreviations GD Gene disruption gsdf Gonadal soma derived factor Hm Homoduplex Ht Heteroduplex WT Wild-type ZFN Zinc finger nuclease
Introduction Classical gene targeting (GT) involves the replacement of an endogenous DNA segment with an exogenously introduced DNA fragment via homologous recombination (HR) (Capecchi 2005; Wang et al. 2008). GT in embryonic stem (ES) cells followed by the formation of germline chimeras has become a routine in mice and produced thousands of transgenic animals to characterize specific gene functions or establish human disease models (Xu et al. 1999). An inherently low HR frequency requires a large number of cultured cells to obtain genomes containing the rare event and the ability of these cells to produce whole animals, which are pluripotent ES cells. The availability of ES cells has thus limited GT to mouse (Collins et al. 2007), rat (Tong et al. 2010) and most recently to medaka (Yan et al. 2013). Recently, engineered zinc finger nucleases (ZFNs) have demonstrated their high efficiency to mediate gene disruption (GD) by creating a double-stranded DNA break (DSB) at a
126
specific chromosomal site (Carroll 2011). The break can be repaired by either non-homologous end joining (NHEJ) with imperfect fidelity or HR (Urnov et al. 2010; Baker 2012). NHEJ will lead to random heterogeneous genetic insertions or deletions at the targeted locus. HR with a donor plasmid will result in gene replacement. ZFNs have been successfully employed for direct GD in diverse organisms including four invertebrate species, Drosophila, nematode, sea urchin and silkworm, 4 vertebrate species, zebrafish, frog, mouse and rat, four plant species, cress, tobacco, maize and petunia as well as cell cultures from 4 mammalian species, human, mouse, hamster and pig (Carroll 2011). Recently, ZFN was applied in the functional study of an immune-related gene sdY in the sex determination of rainbow trout (Yano et al. 2012). The laboratory fish medaka (Oryzias latipes) is an excellent model for genetic analysis and manipulation. This organism has stable stem cell lines required for classical GT experiments. These include diploid ES cells (Hong et al. 1996; Hong et al. 1998), haploid ES cells (Yi et al. 2009), and male germ stem cells (Hong et al. 2004). Its embryo is transparent and robust for microinjection (Li et al. 2009), nuclear transfer (Yi et al. 2009; Liu et al. 2011) and cell transplantation (Hong et al. 2010). In this organism and other fish species, however, production of whole animals with a GT event has not been available, and reverse genetic approaches have so far relied on the use of antisense morpholino oligos (Nasevicius and Ekker 2000; Li et al. 2009). Recently, ZFNs have been attempted for GD on an introduced gfp transgene (Ansai et al. 2012). The fish-specific gene gsdf encodes gonadal soma derived factor, which was first identified in the rainbow trout (Sawatari et al. 2007). As a fish-specific gene, gsdf is conserved in several fish species but absent in the tetrapod lineage (Gautier et al. 2011). This gene shows gonad-specific expression and its embryonic expression correlates with the initiation of testicular differentiation (Shibata et al. 2010). Accumulated data suggest that gsdf may play a role in sex and germ cell development (Myosho et al. 2012). This gene serves as a good candidate for establishing the ZFN technology for direct GD in developing embryos of medaka as a lower vertebrate model, as its loss would have little adverse effect on somatic development and viability. Previously, we have shown that ZFN-mediated GD is possible in medaka embryos (Chen et al. 2012). This study was aimed at the establishment of procedures and parameters for direct GD at the chromosomal gsdf locus in medaka. We show that ZFN mRNA injection allows for direct GD in medaka embryos. A high dose (40 ng/μl) of ZFN mRNAs leads to a proficient GD efficiency of up to 100 % in randomly sampled fry. Independent GD events result in a wide variety of subtle allelic alterations including minor additions and deletions capable of germline transmission. Our data demonstrate the feasibility and efficacy of ZFN-mediated direct GD in medaka embryos.
Mar Biotechnol (2014) 16:125–134
Materials and Methods Fish and Reagents Work with fish was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Advisory Committee for Laboratory Animal Research in Singapore and approved by this committee (Permit Number: 27/09). Medaka strains af and Hd-rR were maintained under an artificial photoperiod of 14-h light to 10-h darkness at 26∼28 °C as described (Hong et al. 2010; Zhao et al. 2011). Embryos were maintained at 26∼28 °C and staged as described (Iwamatsu 2004). Unless otherwise indicated, chemicals were purchased from Sigma, enzymes and PCR reagents were from Promega and TaKaRa, respectively. Plasmids A pair of ZFN expression vectors, pZN1gsdf and pZN2gsdf was commercially synthesized (ToolGen, South Korea) to target the medaka gsdf gene within its first exon: pZN1gsdf contains an array of 4 zinc fingers that recognize 12-bp sequence (CCTTTGTCCTGC); pZN2gsdf contains an array of 3 zinc fingers that recognize 9-bp sequence (CAAGGGAAG). The zinc finger array is fused in frame with FokI endonucleases, forming a ZFN. The target sites of the two ZFNs are separated by a 6-bp spacer (ATCCGT) to allow for dimer formation. RNA Synthesis RNA synthesis was done as described (Li et al. 2009). Briefly, pZN1gsdf and pZN2gsdf were linearized with ApaI and used for mRNA synthesis by using the mMessage mMachine T7 Ultra Kit (Ambion). RNA was then diluted and stored at −80 °C until use. Microinjection Microinjection into embryos at the 1-cell stage was done as described (Li et al. 2009). DNA Extraction Genomic DNA was isolated from embryos and adult organs as described (Hong et al. 1998; Chen et al. 2012). Briefly, embryos were collected in 1.5-ml Eppendorf tubes and stored at −80 °C. Frozen embryos were treated with 200 μl of TNES6U buffer (10 mM Tris–HCl, pH 7.5; 125 mM NaCl, 10 mM EDTA, 1 % SDS, 6 M Urea) and homogenized with a plastic pestle. After the addition of 5 μl of proteinase K solution (20 mg/ml) the tube was incubated at 37 °C overnight.
Mar Biotechnol (2014) 16:125–134
Following extraction with 300 μl phenol/chloroform mixture, DNA was precipitated by adding 2.5 volume of 100 % ethanol and dissolved in 50 μl of TE (10 mM Tris–HCl, 1 mM EDTA, pH 8.0) (Hong et al. 1998). Preparation of caudal fin clips and DNA isolation from them were performed as described (Chen et al. 2012). PCR PCR was run in a 25-μl volume with 50 ng of genomic DNA as the template for 35 cycles (94 °C for 15 s, 60 °C for 15 s and 72 °C for 30 s). In the gel recovery and subsequent PCR (grsPCR) procedure to enrich for rare alleles, PCR products were gel-recovered and ground in 10∼20 μl of TE. After incubation overnight at room temperature, the supernatant containing DNA was used as templates to run 30 cycles of subsequent PCR under the same conditions. Gel Electrophoresis Agarose gel electrophoresis (AGE), T7 endonuclease I digestion and agarose electrophoresis (TAGE) and polyacrylamide gel electrophoresis (PAGE) were done as described (Chen et al. 2012). Briefly, PCR products or plasmid DNA were separated on 1 % agarose gels or 8 % natural polyacrylamide gels and documented on a bio-imaging system (Vilber Lourmat). Band intensity and abundance on PAGE gel is quantified with Gel-Pro analyzer software. Cloning and Sequencing PCR products were TA-cloned into pGEM-T Easy vectors (Promega). Recombinant plasmids were test-digested and sequenced on the 3130xl Sequencer (Applied Biosystems). Sequence analyses were run on Vector NTI and DNAman software.
127
org/Oryzias_latipes/Info/Index) revealed that this target sequence exists only once and is unique to the gsdf. Upon targeted GD via subtle addition or deletion at or around the target site, the gsdf would be structurally disrupted, resulting in loss-of-function alleles due to frame shift mutations or premature truncation of translation. A pair of primers flanking the target site was used for PCR genotyping (Fig. 1) (Chen et al. 2012). Dose-Dependent Survival Previously, we have shown the feasibility of gsdf GD by microinjection of ZFN mRNAs (Chen et al. 2012). To optimize the procedure and efficiency, we first examined the effect of various dosages of ZFN mRNAs on the survival rate of injected embryos. To this end, medaka embryos at the 1-cell stage were injected with ZFN mRNAs at different doses and the survival rate was monitored at the gastrulation, hatching and fry (3 days post hatching) stages. Control embryos showed a 95 % survival rate of fry development, whereas this value decreased to 73, 34, 27, and 13 % when the ZFN mRNAs were injected at 10, 20, 40, and 60 ng/ μl, respectively (Table 1). Further higher doses of injection abolished survival until gastrulation and thus prevented fry production (Table 1). These results indicate that an excess of ZFN mRNAs has adverse effect on embryogenesis and survival, and that injection of ZFN mRNAs at 10∼40 ng/μl ensures a satisfactory survival rate of fry production. Worth mentioning was that all malformed fry died within 1 day post hatching. Therefore, the survived fry we recorded are all in regular form. The malformation rate recorded right after hatching is given in Table S1.
Results Experimental Design The medaka gsdf mRNA is 2297 bp in length and encode 216 amino acids (aa) of a transforming growth factor-β superfamily member, with the first 19 aa serves as a putative signal peptide (Shibata et al. 2010). The medaka gsdf gene consists of 5 exons, and the first exon encodes the first 40 aa. The target site for a ZFN pair was designed to contain two recognition sites and a 6-bp spacer spanning codons 19∼28 (Fig. 1), which is 27 bp long and reads CCTTTGTCC TGCATCCGT CAAGGGAAG (underlined: spacer). The designed ZFNs could specifically target the gsdf gene, as a blast search against the medaka genome (http://www.ensembl.
Fig. 1 Gene structure and ZFN target site. The ZFN target site consists of two recognition sequences and a spacer. One ZFN has four zinc fingers (Z1∼4) and the other three (Z1∼Z3). The site resides within the coding sequence of exon 1 of the wild-type (WT) gsdf locus. Amino acids are shown with their positions above the coding triplets. ZFNs generate a double-stranded DNA break at the target site (dot) to stimulate targeted allelic alterations, which create mutant alleles (MT) with subtle additions and deletions within or near the target site (asterisk). Arrowheads define the positions and extension directions of PCR primers for genotyping
128
Mar Biotechnol (2014) 16:125–134
Table 1 GD survival rate. Table showing the relationship between survival rate and ZFN mRNA with various concentrations. A series of concentrations of ZFN mRNA are applied: 10, 20, 40, 60, 80 ng/μl with DEPC-water (0 ng/μl) as negative control. Survival rates regarding different stages, gastrula, before hatching, fry are recorded. With the increase of dosage, survival rate decreases
We then examined the effect of various dosages of ZFN mRNAs on the GD efficiency in injected embryos. To this end, groups of fry from embryos injected with different doses of ZFN mRNAs were sampled in group of 5 and used for PCR amplification of both wild-type (WT) and mutant (MT) alleles. PCR products were detected, quantified and enriched for by polyacrylamide gel electrophoresis (PAGE) (Chen et al. 2012). DNA from control fry produced merely two homoduplexes (Hm), whereas DNA from ZFN mRNA-injected specimen generated heteroduplexes (Ht) besides Hm forms (Fig. 2a). The intensity and number of Ht forms correlated with the doses of ZFN mRNAs injected, which became more evident after grsPCR (Fig. 2b). For instance, the relative intensity of Ht
forms was 6.1 % at 10 ng/μl and increased to 28.3 and 57.7 % at 20 and 40 ng/μl (Fig. 3 and Figure S1). Concurrently, the number of Ht forms was 4, 9, and 11 for doses at 10, 20 and 40 ng/μl, respectively (Fig. 3 and Figure S1). Notably, the abundance of Ht forms did not increase but decreased at 60 ng/μl compared to 40 ng/μl (Fig. 3 and Figure S1), possibly due to the loss by death of embryos that received this extremely high dose of ZFN mRNAs. Therefore, the GD efficiency positively correlates with the dose of ZFN mRNA injection within the appropriate range, which is 10∼40 ng/μl in the case of gsdf ZFN mRNAs. Taken together, doses at 10∼40 ng/μl ensure both a high survival rate and high efficiency of gsdf GD. The above experiments described GD detection in groups of specimens. We wanted to determine the GD efficiency in individual fry. To facilitate detection, embryos were injected with a high dose (40 ng/μl) of ZFN mRNAs and resultant fry were individually sampled for genotyping. On an agarose gel, PCR products of 346 bp were expectedly apparent in control and ZFN-injected fry (Fig. 4a), whereas Ht forms diagnostic of GD events were absent in control but present in all of ZFN-injected fry examined (n =25). Seven representatives are shown in Fig. 4a. Hence, injection of ZFN mRNAs at 40 ng/μl is capable of nearly 100 % gsdf GD in medaka embryos. We extended our examination into adulthood. To ensure a high survival rate of embryonic development and perhaps also post-hatching growth, a total of 420 embryos were injected with a low dose (10 ng/μl) of ZFN mRNAs. This resulted in 300 fry that were grown into 94 adults. DNA was extracted from fin clips and PCR products were surveyed for genotype by the TAGE. The T7 endonuclease I specifically cleaves a Ht DNA fragment at the mismatch site and produces two smaller
Fig. 2 Dose-dependent GD efficiency in medaka embryos. Embryos at the 1-cell stage were injected with indicated doses (above the lanes) of ZFN mRNA and groups of 5 fry each derived from the injected embryos were analyzed by PAGE. a PAGE showing the PCR products. Multiple heteroduplex bands indicative of GD are marginally detectable (framed).
Sizes in base pairs are given to the left. b PAGE showing the PCR products after the grsPCR procedure. Heteroduplex products (framed) become much more obvious after this round of enrichment PCR. Hm homoduplex, Ht heteroduplex, MT mutant alleles, WT wild-type allele, neg negative PCR control without template DNA
ZFN mRNA (ng/μl)
Embryos injected n
Survivor n (%) Gastrula
Hatching
Fry
0 10
95 85
92 (96) 78 (92)
90 (95) 65 (77)
90 (95) 62 (73)
20 40 60 80
167 180 240 271
116 (70) 130 (72) 84 (35) 1 (0.4)
60 (36) 60 (33) 52 (22) 0 (0)
56 (34) 48 (27) 30 (13) 0 (0)
Dose-Dependent GD Efficiency
Mar Biotechnol (2014) 16:125–134
129
5 fish, namely #23, #25, #38 and #52 (Fig. 4b). This gives rise to ∼4 % efficiency for ZFN-mediated gsdf GD in adult medaka. Interestingly, the 4 fish exhibited differences in the number and intensity of Ht bands, and more specifically, one of them, namely #38, had 15 Ht bands (Fig. 4b), suggesting multiple and independent GD events in this fish. Fish #38 was male (Fig. 4c), and the 3 others including #52 were female (Fig. 4d). Taken together, ZFNs are effective to induce gsdf GD in medaka fry capable of development into male and female adults. Fig. 3 Dose-dependent GD efficiency in medaka embryos. Curves of dose-dependent GD efficiency. Dashed curve is the curve of dose-dependent intensity (%) of heteroduplex bands; solid curve is the curve of dosedependent number of heteroduplex bands
fragments, which allows for distinction on an agarose gel between the WT allele and MT alleles in a Ht (Chen et al. 2012). Five adults were found to show faint bands of Ht forms. Part of the TAGE profile has already been published: fish #4 and #6 (numbered as #23 and #25, respectively here) exhibited a smear of faint band indicative of Ht products (Chen et al. 2012). A total of eight more representatives are shown in Figure S2, where #52 exhibited smaller bands (asterisk), which are faint on Figure S2 but more detectable on gel. PAGE revealed easily detectable Ht forms in 4 of these
Fig. 4 Genotyping of fry and adults. a Genotyping of fry. A major band of 346 bp for the wild-type locus is clearly visible by agarose gel electrophoresis (AGE) and polyacrylamide gel electrophoresis (PAGE). Multiple heteroduplex bands indicative of GD are clearly seen on PAGE (framed). 1∼7, representative fry from embryos injected with 40 ng/μl of ZFN mRNA; ctrl control fry from a non-injected embryo; neg negative PCR control without template DNA. Sizes in base pairs are given to the
Diversity of GD Alleles Three of the adult fish described above, namely #25, #38 and #52, were used to identify WT and MT gsdf alleles. To this end, DNA from fin clips was subjected to grsPCR and the resultant PCR products were cloned for sequencing. A total of 140 recombinant clones were successfully sequenced, which led to the identification of 7 different GD alleles, which fell into 3 categories (Fig. 5, Table 2). The first is the simple deletion (D) of a few bps within the target site, such as D1, D6 and D7 that have a 1-, 6- and 7-bp deletion. The second is the compound deletion, such as D7A4 containing a 7-bp deletion and a 4-bp addition (A), with the net consequence being a 3-bp deletion. The third is compound addition, where a larger addition accompanies a small deletion, with the net
left. b Genotyping of adults by PAGE, highlighting the presence of multiple heteroduplex bands (framed) diagnostic of multiple GD events in four representative animals from embryos injected with 10 ng/μl of ZFN mRNAs. marker, kb size markers with three bands being shown in base pairs to the left. (c and d) Adult male (c) and female fish (d) from ZFN-injected embryos. Df dorsal fin; cf caudal fin. Scale bars 0.5 cm
130
Mar Biotechnol (2014) 16:125–134
Fig. 5 Sequences of GD alleles. DNA sequences encompassing the ZFN-targeted site of three founder fish. D deletion; A addition. Spacer region is highlighted in bold
consequence being an addition by one to several bps, such as A16D3, A21D4 and A22D4 that are 13, 17 and 18 bp longer than the WT allele. These results demonstrate the diversity of ZFN-induced GD alleles. Germline Transmission The four adults described above were examined for germline transmission of GD alleles by progeny test. After crossing with control partners of strain Hd-rR, one of them showed germline transmission, which was male fish #38. The other female fish were not informative because the number of their progeny (11∼24) was not big enough. A total of 46 progeny fry from #38 were genotyped by PAGE, 7 of which had a GD event as they formed distinct Ht bands (Fig. 6a), producing a germline transmission rate of 15.2 %, which is within the range, 8.1 to 100 %, reported in ZFN-mediated exogenous GFP disruption in medaka (Ansai et al. 2012). As illustrated in Fig. 6 for 10 representative progeny fry, 7 fry were positive for GD. Since B =N 2 −N (where B is the number of Ht bands and N is the number of different alleles) as described previously
(Chen et al. 2012), a single GD allele could form two Ht bands (Fig. 6b). A closer inspection revealed that all the 7 fry exhibited two Ht bands of a similar size, suggesting the germline transmission of one and the same GD allele. Cloning and sequencing validated this observation (Fig. 6c). This GD allele had a 7-bp deletion covering the spacer. Interestingly, this GD allele in #38 fish was present in the somatic fin clip, in the testis, and further inherited to the next generation, strongly suggesting that this GD event could occur early in development before the separation between the soma and germline. GD Ontogeny The observation that injection of ZFN mRNAs could induce multiple GD events evoked us to examine GD ontogeny. To this end, embryos were injected again with a low dose (10 ng/μl) of ZFN mRNAs and examined in groups of 5 specimens each at different stages of development. Ht forms containing GD alleles were detectable from 1 day post fertilization (dpf) onwards (Fig. 7a). More specifically, a closer inspection revealed that the intensity and number of Ht bands increased
Table 2 Categories of GD alleles. Table showing the frequencies of each mutation type from the three founder fish Fish
Plasmid clones
Alleles, n (%) (second PCR products) WT
MT Deletion (D)
#25 #38 #52
56 58 26
D deletion; A addition
37 (66.0) 29 (50.0) 17 (65.4)
Irregular cases
D1
D6
D7
A21D4
1 (1.8)
4 (7.2) 1 (1.7) 8 (30.8)
16 (27.6)
12 (20.7)
D7A4
1 (3.8)
A22D4
A16D3
11 (19.6)
3 (5.4)
Mar Biotechnol (2014) 16:125–134
131
Fig. 6 Germline transmission of ZFN-mediated GD. a PAGE profile of F1 progeny of founder #38. 1∼10, representative F1 fry from founder #38; ctrl control fry from a non-injected embryo; neg negative PCR control without template DNA. Arrows depict heteroduplexes between WT and MT alleles. b Schematic view of heteroduplexes. A single GD
allele (Aa) with a WT allele (Bb) could form two Ht bands (Ab and Ba). c Allelic sequences. WT allele is shown at the top. The GD allele has a 7-bp deletion (D7), which is seen in the tail, testis of #38 and its F1 progeny. Spacer region is highlighted in bold
when embryogenesis proceeded (Fig. 7a). We then extended our observation into adulthood. Fish fins, in particular the caudal fin, have a high regeneration activity (Katogi et al. 2004). We utilized this property to examine the gsdf allelic
profile in fish #38 at adulthood at two different time point, 3 and 6 months post hatching. After a cut, the caudal fin regenerated, then re-cut is performed for DNA preparation and analysis. As illustrated in Fig. 7b, the band pattern altered
Fig. 7 GD ontogeny. a GD alleles at different stages of embryonic development. Groups of five embryos injected with 40 ng/μl were analyzed at each stage before (left panel) and after grsPCR (right panel). b GD alleles at different stages of posthatching development of fish #38. c GD alleles in adult organs. Fish #38 (top panel) and #52 (bottom panel) were used for DNA isolation from organs indicated above lanes and subjected to PAGE analyses. Hm homoduplex; Ht heteroduplex; mph months post hatching
132
with time. Finally, different organs representative the three germ layers were dissected from fish #38 and #52 after progeny test for germline transmission and analyzed for WT and MT alleles. This revealed that all organs of both fish had GD alleles by the presence of Ht forms and more importantly, the majority of organs examined—including those that share the same developmental origin such as the gut and liver—exhibited heterogeneous band patterns (Fig. 7c). On the other hand, organs of different germ layers shared a similar band pattern, as is the case between the fin and gonad or between the gill and gut in fish #52. Cloning and sequencing validated the presence of different GD alleles in the seven tissues or organs of fish #38 (Figure S3). The heterogeneity of GD alleles suggests multiple and independent GD events in medaka ontogeny. Since germ layers are formed during gastrulation and organogenesis takes places late (2∼3 dpf) in development, ZFN-induced GD must commence early in development and occurs repeatedly throughout embryogenesis. In summary, we have established procedures and parameters for successful gsdf GD by direct ZFN mRNA injection in medaka embryos. We show that such GD events occur efficiently during embryogenesis and in primordia that ultimately give rise to somatic organs and gonads of both sexes. Most importantly, we demonstrate that GD-bearing embryos can develop into normal adults capable of germline transmission of GD alleles into the next generation.
Discussion In this study, we have demonstrated the feasibility, and—for the first time to our knowledge—the efficacy of ZFNmediated direct GD at a chromosomal locus in medaka embryos. We have also established procedures and parameters that allow for maximal survival or maximal GD efficiency. Specifically, the dosage of ZFN mRNA injection correlates negatively with survival but positively with GD efficiency. RNA injection at a high dose of 40 ng/μl gives rise to nearly 100 % GD efficiency in operated fry, whereas RNA injection at a low dose of 10 ng/μl ensures a high survival rate of fry development. A high efficiency of direct GD at a chromosomal locus and parameters for a satisfactory balance between survival and GD efficiency will provide valuable information for setting up new ZFN-mediated GD experiments in medaka in particular and other fish species in general. The ZFN approach has proven its power for direct GD in several organisms (Carroll 2011). An important requirement for the introduction of ZFNs into a new model is an appropriate procedure, as well as detailed practical parameters to predict and implement effective GD. In our case, the streamlined protocol mainly consists of: design and construct ZFNs to target the desired gene; synthesize ZFN mRNA; efficiency test for designed ZFNs and optimize the parameters in
Mar Biotechnol (2014) 16:125–134
embryos and fry; GD detection assay for somatic and germline mutagenesis. The experience that we have in designing ZFNs is that we chose the target AA following closely downstream the signal peptide of the target gene. As the target locus is located at 5′ end, truncated gsdf protein, which may contribute to partial function of the intact gsdf, is avoided to the maximum extent. We chose mRNA instead of DNA as the ZFN delivery form since mRNA eliminates the risk of random genome integration, and lower cytotoxicity and off-target events by exposing cells to ZFNs for a shorter time. Additionally, we found that medaka cells do not tolerate input ZFN plasmid since all embryos are dead upon microinjection of a pair of ZFN plasmid (20 ng/μl). Although embryo injection of mRNAs for ZFN expression has proven to be practical in several organisms, zebrafish, rat, mouse, sea urchin, frog, and so on (Carroll 2011), parameters for efficient GD have, however, remained hardly elucidated. One of the critical issues for setting up a ZFN experiment is the dosage of mRNA injection. Fry test reveals that the dose of ZFN mRNA injection correlates negatively with the survival rate but positively with the GD efficiency, which ranges from 30 to 100 % when the ZFN mRNAs are injected at 10 to 40 ng/μl. Interestingly, when the dosage is raised to 60 ng/μl, the efficiency is lower than that of 40 ng/μl. One explanation is that high concentration of ZFN may cause toxicity to embryos or generate off-target mutations, leaving many GD embryos dead. In our example, GD occurs at a ∼4 % efficiency when the survival rate until the fry stage is ∼70 %. More intriguingly, with a relatively low dosage of 10 ng/μl, the percentage of individuals with successful GD falls from 30 % (Figure S4) to ∼4 % from fry to adults, indicating that the embryo containing considerable GD events due to a high dosage of ZFN RNA injection is less viable to its normal counterpart. In ZFN-injected medaka embryos, the number of different alleles increases when the development proceeds from 1 to 3 dpf, suggesting that introduced ZFNs persist and retain their specific cleavage activity throughout this period, thus continue to generate independent GD events. The persistence of ZFN activity could produce highly mosaic organs not only in the relative proportion of GD events but also in the number of various GD alleles. ZFN-induced mosaicism has been reported in the tail of rodents (Cui et al. 2011). In medaka, we have expanded such examination into a total of eight organs or tissues representative of the three germ layers that are formed during gastrulation in two F0 fish. This leads to an observation that the band pattern of GD alleles is different among all these organs, even between the gut and liver, which are both originated from endoderm during organogenesis and become discernable at 4 dpf in medaka (Watanabe et al. 2004). A difference between these two organs in GD band pattern and alleles strongly suggests that ZFNs continue to operate at least until the organogenesis stage.
Mar Biotechnol (2014) 16:125–134
One of the ultimate goals of genome editing is the germline transmission to produce whole animals containing a defined genetic alteration. In medaka, we notice that the efficiency of gsdf GD is higher in the soma than the germline. For example, only one of four adults positive for GD in the caudal fin is capable of germline transmission. Non-transmitting fish may contain an easily detectable proportion of GD alleles in many somatic organs and gonad. Non-transmitting fish #52, for example, is comparable to the transmitter #38 in the number and abundance of GD alleles in various organs. The presence of a discrepancy between a high efficiency of GD in the soma and a lack of GD germline transmission is not surprising, because the number of cell types and mass in the soma is substantially larger than the germline, the latter comprises only a single cell lineage, namely germ cells responsible for germline transmission. It deserves to note that the adult gonad, the ovary, and testis, contains also several types of somatic cells besides germ cells. For example, there are three larger Ht bands shown in the sample from #38 testis (Fig. 7c), in the F1 progeny of this founder, only two higher bands were observed (Fig. 6a), indicating there is another kind of GD allele existed in the testis of #38, present either in the gonadal somatic cells and/or in the germ cells that have not been transmitted to the next generation (Figure S3). However, compared to our result, the previous ZFN-mediated GFP knockout case in medaka reported higher germline transmitter rates ranging from 6.25 to 50 % under nearly the same injection dosages (Ansai et al. 2012). Besides the variations in customized ZFN activity, the targeted locus configuration and the efficacy of mutation detection, another explanation for the difference in germline transmission efficiency is that gsdf GD might affect germ cell development for gamete production. This possibility is supported by the report that in rainbow trout, gsdf knockdown suppresses primordial germ cell proliferation, and the addition of a recombinant GSDF to testicular cell culture enhances the proliferation of male germ stem cells spermatogonia (Sawatari et al. 2007). Targeted GD results in a wide variety of subtle allelic alterations including minor additions and deletions in fry and adults. A total of eight allelic alterations were defined by sequencing, including the seven mutation types listed in Fig. 4. All deletions are the consequence of loss by 1∼10 bp, whereas in the case of additions, besides a simple addition of several base pairs, additions can also result from a combination of additions and deletions. These results conform to previous studies in zebrafish (Doyon et al. 2008; Meng et al. 2008). Our data demonstrate that ZFNs can produce a wide variety of allelic alterations in medaka, which may lead to loss (frame shift mutations) and modification (in frame additions/deletions) of function of the target gene. The diversity of GD alleles will allow for detailed analyses of various subtle alterations on the activity and function of the target gene.
133
Engineered endonucleases, like ZFNs, have opened a new era of genome editing. Most recently, targeted mutagenesis in medaka embryos using custom-designed TALENs has been reported (Ansai et al. 2013). In this study, we have established procedures and parameters for a high efficiency of ZFNmediated GD in the soma and germline directly in medaka embryos. This procedures and parameters should be informative and applicable for ZFN-mediated direct GT in other fish species including those of aquaculture importance. Acknowledgments We thank J. Deng for fish breeding. This work was supported by the National Research Foundation Singapore (NRF-CRP72010-03) and the Japan Core Research for Evolutional Science and Technology (CREST). We acknowledge the NUS for scholarship to X. Zhang and J. B. Chen. Conflict of Interest The authors declare that they have no conflict of interest. Supporting Information Additional information noted in text includes supplementary figures.
References Ansai S, Ochiai H, Kanie Y, Kamei Y, Gou Y, Kitano T, Yamamoto T, Kinoshita M (2012) Targeted disruption of exogenous EGFP gene in medaka using zinc-finger nucleases. Develop Growth Differ 54: 546–556 Ansai S, Sakuma T, Yamamoto T, Ariga H, Uemura N, Takahashi R, Kinoshita M (2013) Efficient targeted mutagenesis in medaka using custom-designed transcription activator-like effector nucleases. Genetics 193:739–749 Baker M (2012) Gene-editing nucleases. Nat Meth 9:23–26 Capecchi MR (2005) Gene targeting in mice: functional analysis of the mammalian genome for the twenty-first century. Nat Rev Genet 6: 507–512 Carroll D (2011) Genome engineering with zinc-finger nucleases. Genetics 188:773–782 Chen J, Zhang X, Wang T, Li Z, Guan G, Hong Y (2012) Efficient detection, quantification and enrichment of subtle allelic alterations. DNA Res 19:423–433 Collins FS, Rossant J, Wurst W (2007) A mouse for all reasons. Cell 128: 9–13 Cui X, Ji D, Fisher DA, Wu Y, Briner DM, Weinstein EJ (2011) Targeted integration in rat and mouse embryos with zinc-finger nucleases. Nat Biotechnol 29:64–67 Doyon Y, Mccammon JM, Miller JC, Faraji F, Ngo C, Katibah GE, Amora R, Hocking TD, Zhang L, Rebar EJ, Gregory PD, Urnov FD, Amacher SL (2008) Heritable targeted gene disruption in zebrafish using designed zinc-finger nucleases. Nat Biotech 26: 702–708 Gautier A, Le Gac F, Lareyre JJ (2011) The gsdf gene locus harbors evolutionary conserved and clustered genes preferentially expressed in fish previtellogenic oocytes. Gene 472:7–17 Hong Y, Winkler C, Schartl M (1996) Pluripotency and differentiation of embryonic stem cell lines from the medakafish (Oryzias latipes). Mech Dev 60:33–44 Hong Y, Winkler C, Schartl M (1998) Production of medakafish chimeras from a stable embryonic stem cell line. Proc Natl Acad Sci U S A 95: 3679–3684
134 Hong Y, Liu T, Zhao H, Xu H, Wang W, Liu R, Chen T, Deng J, Gui J (2004) Establishment of a normal medakafish spermatogonial cell line capable of sperm production in vitro. Proc Natl Acad Sci U S A 101:8011–8016 Hong N, Li M, Zeng Z, Yi M, Deng J, Gui J, Winkler C, Schartl M, Hong Y (2010) Accessibility of host cell lineages to medaka stem cells depends on genetic background and irradiation of recipient embryos. Cell Mol Life Sci 67:1189–1202 Iwamatsu T (2004) Stages of normal development in the medaka Oryzias latipes. Mech Dev 121:605–618 Katogi R, Nakatani Y, Shin-I T, Kohara Y, Inohaya K, Kudo A (2004) Largescale analysis of the genes involved in fin regeneration and blastema formation in the medaka, Oryzias latipes. Mech Dev 121:861–872 Li M, Hong N, Xu H, Yi M, Li C, Gui J, Hong Y (2009) Medaka vasa is required for migration but not survival of primordial germ cells. Mech Dev 126:366–381 Liu T, Liu L, Wei Q, Hong Y (2011) Sperm nuclear transfer and transgenic production in the fish medaka. Int J Biol Sci 7:469–475 Meng X, Noyes MB, Zhu LJ, Lawson ND, Wolfe SA (2008) Targeted gene inactivation in zebrafish using engineered zinc-finger nucleases. Nat Biotechnol 26:695–701 Myosho T, Otake H, Masuyama H, Matsuda M, Kuroki Y, Fujiyama A, Naruse K, Hamaguchi S, Sakaizumi M (2012) Tracing the emergence of a novel sex-determining gene in medaka, Oryzias luzonensis. Genetics 191:163–170 Nasevicius A, Ekker SC (2000) Effective targeted gene ‘knockdown’ in zebrafish. Nat Genet 26:216–220 Sawatari E, Shikina S, Takeuchi T, Yoshizaki G (2007) A novel transforming growth factor-beta superfamily member expressed in gonadal somatic cells enhances primordial germ cell and spermatogonial proliferation in rainbow trout (Oncorhynchus mykiss). Dev Biol 301:266–275 Shibata Y, Paul-Prasanth B, Suzuki A, Usami T, Nakamoto M, Matsuda M, Nagahama Y (2010) Expression of gonadal soma derived factor (GSDF) is spatially and temporally correlated with early testicular differentiation in medaka. Gene Expr Patterns 10:283–289
Mar Biotechnol (2014) 16:125–134 Tong C, Li P, Wu NL, Yan Y, Ying Q-L (2010) Production of p53 gene knockout rats by homologous recombination in embryonic stem cells. Nature 467:211–213 Urnov FD, Rebar EJ, Holmes MC, Zhang HS, Gregory PD (2010) Genome editing with engineered zinc finger nucleases. Nat Rev Genet 11:636–646 Wang R-H, Sengupta K, Li C, Kim H-S, Cao L, Xiao C, Kim S, Xu X, Zheng Y, Chilton B, Jia R, Zheng Z-M, Appella E, Wang XW, Ried T, Deng C-X (2008) Impaired DNA damage response, genome instability, and tumorigenesis in SIRT1 mutant mice. Cancer Cell 14:312–323 Watanabe T, Asaka S, Kitagawa D, Saito K, Kurashige R, Sasado T, Morinaga C, Suwa H, Niwa K, Henrich T, Hirose Y, Yasuoka A, Yoda H, Deguchi T, Iwanami N, Kunimatsu S, Osakada M, Loosli F, Quiring R, Carl M, Grabher C, Winkler S, Del Bene F, Wittbrodt J, Abe K, Takahama Y, Takahashi K, Katada T, Nishina H, Kondoh H, Furutani-Seiki M (2004) Mutations affecting liver development and function in Medaka, Oryzias latipes, screened by multiple criteria. Mech Dev 121:791–802 Xu X, Wagner KU, Larson D, Weaver Z, Li C, Ried T, Hennighausen L, Wynshaw-Boris A, Deng CX (1999) Conditional mutation of Brca1 in mammary epithelial cells results in blunted ductal morphogenesis and tumour formation. Nat Genet 22:37–43 Yan Y, Hong N, Chen T, Li M, Wang T, Guan G, Qiao Y, Chen S, Schartl M, Li CM, Hong Y (2013) p53 gene targeting by homologous recombination in fish ES cells. PLoS One 8:e59400 Yano A, Guyomard R, Nicol B, Jouanno E, Quillet E, Klopp C, Cabau C, Bouchez O, Fostier A, Guiguen Y (2012) An immune-related gene evolved into the master sex-determining gene in rainbow trout, Oncorhynchus mykiss. Curr Biol 22:1423–1428 Yi M, Hong N, Hong Y (2009) Generation of medaka fish haploid embryonic stem cells. Science 326:430–433 Zhao H, Hong N, Lu W, Zeng H, Song J, Hong Y (2011) Fusion gene vectors allowing for simultaneous drug selection, cell labeling, and reporter assay in vitro and in vivo. Anal Chem 84:987–993
ORIGINAL ARTICLE
Parameters and Efficiency of Direct Gene Disruption by Zinc Finger Nucleases in Medaka Embryos Xi Zhang & Guijun Guan & Jianbin Chen & Kiyoshi Naruse & Yunhan Hong
Received: 18 April 2013 / Accepted: 15 July 2013 / Published online: 23 October 2013 # Springer Science+Business Media New York 2013
Abstract Zinc finger nucleases (ZFNs) can generate targeted gene disruption (GD) directly in developing embryos of zebrafish, mouse and human. In the fish medaka, ZFNs have been attempted on a transgene. Here, we developed procedures and parameters for ZFN-mediated direct GD on the gonad-specifically expressed gsdf locus in medaka. A pair of ZFNs was designed to target the first exon of gsdf and their synthetic mRNAs were microinjected into 1-cell stage embryos. We reveal dose-dependent survival rate and GD efficiency. In fry, ZFN mRNA injection at 10 ng/μl led to a GD efficiency of 30 %. This value increased up to nearly 100 % when the dose was enhanced to 40 ng/μl. In a typical series of experiments of ZFN mRNA injection at 10 ng/μl, 420 injected embryos developed into 94 adults, 4 of which had altered gsdf alleles. This leads to a GD efficacy of ∼4 % in the adulthood. Sequencing revealed a wide variety of subtle allelic alterations including additions and deletions of 1∼18 bp in length in ZFN-injected samples. Most importantly, one of the 4 adults examined was capable of germline transmission to 15.2 % of its F1 progeny. Interestingly, ontogenic analyses of the allelic profile revealed that GD commenced early in development, continued during subsequent stages of development and in primordia for different adult organs of the three germ layers. Xi Zhang and Guijun Guan contributed equally to this work. Electronic supplementary material The online version of this article (doi:10.1007/s10126-013-9556-6) contains supplementary material, which is available to authorized users. X. Zhang : G. Guan : J. Chen : Y. Hong (*) Department of Biological Sciences, National University of Singapore, 14 Science Drive 4, Singapore 117543, Singapore e-mail: [email protected] G. Guan : K. Naruse (*) Department of Bioresources, National Institute for Basic Biology, Okazaki, Aichi 444-8585, Japan e-mail: [email protected]
These results demonstrate the feasibility and—for the first time to our knowledge—the efficacy of ZFN-mediated direct GD on a chromosomal gene in medaka embryos. Keywords Gene disruption . Gene targeting . Genome editing . gsdf . Zinc finger nuclease . ZFN Abbreviations GD Gene disruption gsdf Gonadal soma derived factor Hm Homoduplex Ht Heteroduplex WT Wild-type ZFN Zinc finger nuclease
Introduction Classical gene targeting (GT) involves the replacement of an endogenous DNA segment with an exogenously introduced DNA fragment via homologous recombination (HR) (Capecchi 2005; Wang et al. 2008). GT in embryonic stem (ES) cells followed by the formation of germline chimeras has become a routine in mice and produced thousands of transgenic animals to characterize specific gene functions or establish human disease models (Xu et al. 1999). An inherently low HR frequency requires a large number of cultured cells to obtain genomes containing the rare event and the ability of these cells to produce whole animals, which are pluripotent ES cells. The availability of ES cells has thus limited GT to mouse (Collins et al. 2007), rat (Tong et al. 2010) and most recently to medaka (Yan et al. 2013). Recently, engineered zinc finger nucleases (ZFNs) have demonstrated their high efficiency to mediate gene disruption (GD) by creating a double-stranded DNA break (DSB) at a
126
specific chromosomal site (Carroll 2011). The break can be repaired by either non-homologous end joining (NHEJ) with imperfect fidelity or HR (Urnov et al. 2010; Baker 2012). NHEJ will lead to random heterogeneous genetic insertions or deletions at the targeted locus. HR with a donor plasmid will result in gene replacement. ZFNs have been successfully employed for direct GD in diverse organisms including four invertebrate species, Drosophila, nematode, sea urchin and silkworm, 4 vertebrate species, zebrafish, frog, mouse and rat, four plant species, cress, tobacco, maize and petunia as well as cell cultures from 4 mammalian species, human, mouse, hamster and pig (Carroll 2011). Recently, ZFN was applied in the functional study of an immune-related gene sdY in the sex determination of rainbow trout (Yano et al. 2012). The laboratory fish medaka (Oryzias latipes) is an excellent model for genetic analysis and manipulation. This organism has stable stem cell lines required for classical GT experiments. These include diploid ES cells (Hong et al. 1996; Hong et al. 1998), haploid ES cells (Yi et al. 2009), and male germ stem cells (Hong et al. 2004). Its embryo is transparent and robust for microinjection (Li et al. 2009), nuclear transfer (Yi et al. 2009; Liu et al. 2011) and cell transplantation (Hong et al. 2010). In this organism and other fish species, however, production of whole animals with a GT event has not been available, and reverse genetic approaches have so far relied on the use of antisense morpholino oligos (Nasevicius and Ekker 2000; Li et al. 2009). Recently, ZFNs have been attempted for GD on an introduced gfp transgene (Ansai et al. 2012). The fish-specific gene gsdf encodes gonadal soma derived factor, which was first identified in the rainbow trout (Sawatari et al. 2007). As a fish-specific gene, gsdf is conserved in several fish species but absent in the tetrapod lineage (Gautier et al. 2011). This gene shows gonad-specific expression and its embryonic expression correlates with the initiation of testicular differentiation (Shibata et al. 2010). Accumulated data suggest that gsdf may play a role in sex and germ cell development (Myosho et al. 2012). This gene serves as a good candidate for establishing the ZFN technology for direct GD in developing embryos of medaka as a lower vertebrate model, as its loss would have little adverse effect on somatic development and viability. Previously, we have shown that ZFN-mediated GD is possible in medaka embryos (Chen et al. 2012). This study was aimed at the establishment of procedures and parameters for direct GD at the chromosomal gsdf locus in medaka. We show that ZFN mRNA injection allows for direct GD in medaka embryos. A high dose (40 ng/μl) of ZFN mRNAs leads to a proficient GD efficiency of up to 100 % in randomly sampled fry. Independent GD events result in a wide variety of subtle allelic alterations including minor additions and deletions capable of germline transmission. Our data demonstrate the feasibility and efficacy of ZFN-mediated direct GD in medaka embryos.
Mar Biotechnol (2014) 16:125–134
Materials and Methods Fish and Reagents Work with fish was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Advisory Committee for Laboratory Animal Research in Singapore and approved by this committee (Permit Number: 27/09). Medaka strains af and Hd-rR were maintained under an artificial photoperiod of 14-h light to 10-h darkness at 26∼28 °C as described (Hong et al. 2010; Zhao et al. 2011). Embryos were maintained at 26∼28 °C and staged as described (Iwamatsu 2004). Unless otherwise indicated, chemicals were purchased from Sigma, enzymes and PCR reagents were from Promega and TaKaRa, respectively. Plasmids A pair of ZFN expression vectors, pZN1gsdf and pZN2gsdf was commercially synthesized (ToolGen, South Korea) to target the medaka gsdf gene within its first exon: pZN1gsdf contains an array of 4 zinc fingers that recognize 12-bp sequence (CCTTTGTCCTGC); pZN2gsdf contains an array of 3 zinc fingers that recognize 9-bp sequence (CAAGGGAAG). The zinc finger array is fused in frame with FokI endonucleases, forming a ZFN. The target sites of the two ZFNs are separated by a 6-bp spacer (ATCCGT) to allow for dimer formation. RNA Synthesis RNA synthesis was done as described (Li et al. 2009). Briefly, pZN1gsdf and pZN2gsdf were linearized with ApaI and used for mRNA synthesis by using the mMessage mMachine T7 Ultra Kit (Ambion). RNA was then diluted and stored at −80 °C until use. Microinjection Microinjection into embryos at the 1-cell stage was done as described (Li et al. 2009). DNA Extraction Genomic DNA was isolated from embryos and adult organs as described (Hong et al. 1998; Chen et al. 2012). Briefly, embryos were collected in 1.5-ml Eppendorf tubes and stored at −80 °C. Frozen embryos were treated with 200 μl of TNES6U buffer (10 mM Tris–HCl, pH 7.5; 125 mM NaCl, 10 mM EDTA, 1 % SDS, 6 M Urea) and homogenized with a plastic pestle. After the addition of 5 μl of proteinase K solution (20 mg/ml) the tube was incubated at 37 °C overnight.
Mar Biotechnol (2014) 16:125–134
Following extraction with 300 μl phenol/chloroform mixture, DNA was precipitated by adding 2.5 volume of 100 % ethanol and dissolved in 50 μl of TE (10 mM Tris–HCl, 1 mM EDTA, pH 8.0) (Hong et al. 1998). Preparation of caudal fin clips and DNA isolation from them were performed as described (Chen et al. 2012). PCR PCR was run in a 25-μl volume with 50 ng of genomic DNA as the template for 35 cycles (94 °C for 15 s, 60 °C for 15 s and 72 °C for 30 s). In the gel recovery and subsequent PCR (grsPCR) procedure to enrich for rare alleles, PCR products were gel-recovered and ground in 10∼20 μl of TE. After incubation overnight at room temperature, the supernatant containing DNA was used as templates to run 30 cycles of subsequent PCR under the same conditions. Gel Electrophoresis Agarose gel electrophoresis (AGE), T7 endonuclease I digestion and agarose electrophoresis (TAGE) and polyacrylamide gel electrophoresis (PAGE) were done as described (Chen et al. 2012). Briefly, PCR products or plasmid DNA were separated on 1 % agarose gels or 8 % natural polyacrylamide gels and documented on a bio-imaging system (Vilber Lourmat). Band intensity and abundance on PAGE gel is quantified with Gel-Pro analyzer software. Cloning and Sequencing PCR products were TA-cloned into pGEM-T Easy vectors (Promega). Recombinant plasmids were test-digested and sequenced on the 3130xl Sequencer (Applied Biosystems). Sequence analyses were run on Vector NTI and DNAman software.
127
org/Oryzias_latipes/Info/Index) revealed that this target sequence exists only once and is unique to the gsdf. Upon targeted GD via subtle addition or deletion at or around the target site, the gsdf would be structurally disrupted, resulting in loss-of-function alleles due to frame shift mutations or premature truncation of translation. A pair of primers flanking the target site was used for PCR genotyping (Fig. 1) (Chen et al. 2012). Dose-Dependent Survival Previously, we have shown the feasibility of gsdf GD by microinjection of ZFN mRNAs (Chen et al. 2012). To optimize the procedure and efficiency, we first examined the effect of various dosages of ZFN mRNAs on the survival rate of injected embryos. To this end, medaka embryos at the 1-cell stage were injected with ZFN mRNAs at different doses and the survival rate was monitored at the gastrulation, hatching and fry (3 days post hatching) stages. Control embryos showed a 95 % survival rate of fry development, whereas this value decreased to 73, 34, 27, and 13 % when the ZFN mRNAs were injected at 10, 20, 40, and 60 ng/ μl, respectively (Table 1). Further higher doses of injection abolished survival until gastrulation and thus prevented fry production (Table 1). These results indicate that an excess of ZFN mRNAs has adverse effect on embryogenesis and survival, and that injection of ZFN mRNAs at 10∼40 ng/μl ensures a satisfactory survival rate of fry production. Worth mentioning was that all malformed fry died within 1 day post hatching. Therefore, the survived fry we recorded are all in regular form. The malformation rate recorded right after hatching is given in Table S1.
Results Experimental Design The medaka gsdf mRNA is 2297 bp in length and encode 216 amino acids (aa) of a transforming growth factor-β superfamily member, with the first 19 aa serves as a putative signal peptide (Shibata et al. 2010). The medaka gsdf gene consists of 5 exons, and the first exon encodes the first 40 aa. The target site for a ZFN pair was designed to contain two recognition sites and a 6-bp spacer spanning codons 19∼28 (Fig. 1), which is 27 bp long and reads CCTTTGTCC TGCATCCGT CAAGGGAAG (underlined: spacer). The designed ZFNs could specifically target the gsdf gene, as a blast search against the medaka genome (http://www.ensembl.
Fig. 1 Gene structure and ZFN target site. The ZFN target site consists of two recognition sequences and a spacer. One ZFN has four zinc fingers (Z1∼4) and the other three (Z1∼Z3). The site resides within the coding sequence of exon 1 of the wild-type (WT) gsdf locus. Amino acids are shown with their positions above the coding triplets. ZFNs generate a double-stranded DNA break at the target site (dot) to stimulate targeted allelic alterations, which create mutant alleles (MT) with subtle additions and deletions within or near the target site (asterisk). Arrowheads define the positions and extension directions of PCR primers for genotyping
128
Mar Biotechnol (2014) 16:125–134
Table 1 GD survival rate. Table showing the relationship between survival rate and ZFN mRNA with various concentrations. A series of concentrations of ZFN mRNA are applied: 10, 20, 40, 60, 80 ng/μl with DEPC-water (0 ng/μl) as negative control. Survival rates regarding different stages, gastrula, before hatching, fry are recorded. With the increase of dosage, survival rate decreases
We then examined the effect of various dosages of ZFN mRNAs on the GD efficiency in injected embryos. To this end, groups of fry from embryos injected with different doses of ZFN mRNAs were sampled in group of 5 and used for PCR amplification of both wild-type (WT) and mutant (MT) alleles. PCR products were detected, quantified and enriched for by polyacrylamide gel electrophoresis (PAGE) (Chen et al. 2012). DNA from control fry produced merely two homoduplexes (Hm), whereas DNA from ZFN mRNA-injected specimen generated heteroduplexes (Ht) besides Hm forms (Fig. 2a). The intensity and number of Ht forms correlated with the doses of ZFN mRNAs injected, which became more evident after grsPCR (Fig. 2b). For instance, the relative intensity of Ht
forms was 6.1 % at 10 ng/μl and increased to 28.3 and 57.7 % at 20 and 40 ng/μl (Fig. 3 and Figure S1). Concurrently, the number of Ht forms was 4, 9, and 11 for doses at 10, 20 and 40 ng/μl, respectively (Fig. 3 and Figure S1). Notably, the abundance of Ht forms did not increase but decreased at 60 ng/μl compared to 40 ng/μl (Fig. 3 and Figure S1), possibly due to the loss by death of embryos that received this extremely high dose of ZFN mRNAs. Therefore, the GD efficiency positively correlates with the dose of ZFN mRNA injection within the appropriate range, which is 10∼40 ng/μl in the case of gsdf ZFN mRNAs. Taken together, doses at 10∼40 ng/μl ensure both a high survival rate and high efficiency of gsdf GD. The above experiments described GD detection in groups of specimens. We wanted to determine the GD efficiency in individual fry. To facilitate detection, embryos were injected with a high dose (40 ng/μl) of ZFN mRNAs and resultant fry were individually sampled for genotyping. On an agarose gel, PCR products of 346 bp were expectedly apparent in control and ZFN-injected fry (Fig. 4a), whereas Ht forms diagnostic of GD events were absent in control but present in all of ZFN-injected fry examined (n =25). Seven representatives are shown in Fig. 4a. Hence, injection of ZFN mRNAs at 40 ng/μl is capable of nearly 100 % gsdf GD in medaka embryos. We extended our examination into adulthood. To ensure a high survival rate of embryonic development and perhaps also post-hatching growth, a total of 420 embryos were injected with a low dose (10 ng/μl) of ZFN mRNAs. This resulted in 300 fry that were grown into 94 adults. DNA was extracted from fin clips and PCR products were surveyed for genotype by the TAGE. The T7 endonuclease I specifically cleaves a Ht DNA fragment at the mismatch site and produces two smaller
Fig. 2 Dose-dependent GD efficiency in medaka embryos. Embryos at the 1-cell stage were injected with indicated doses (above the lanes) of ZFN mRNA and groups of 5 fry each derived from the injected embryos were analyzed by PAGE. a PAGE showing the PCR products. Multiple heteroduplex bands indicative of GD are marginally detectable (framed).
Sizes in base pairs are given to the left. b PAGE showing the PCR products after the grsPCR procedure. Heteroduplex products (framed) become much more obvious after this round of enrichment PCR. Hm homoduplex, Ht heteroduplex, MT mutant alleles, WT wild-type allele, neg negative PCR control without template DNA
ZFN mRNA (ng/μl)
Embryos injected n
Survivor n (%) Gastrula
Hatching
Fry
0 10
95 85
92 (96) 78 (92)
90 (95) 65 (77)
90 (95) 62 (73)
20 40 60 80
167 180 240 271
116 (70) 130 (72) 84 (35) 1 (0.4)
60 (36) 60 (33) 52 (22) 0 (0)
56 (34) 48 (27) 30 (13) 0 (0)
Dose-Dependent GD Efficiency
Mar Biotechnol (2014) 16:125–134
129
5 fish, namely #23, #25, #38 and #52 (Fig. 4b). This gives rise to ∼4 % efficiency for ZFN-mediated gsdf GD in adult medaka. Interestingly, the 4 fish exhibited differences in the number and intensity of Ht bands, and more specifically, one of them, namely #38, had 15 Ht bands (Fig. 4b), suggesting multiple and independent GD events in this fish. Fish #38 was male (Fig. 4c), and the 3 others including #52 were female (Fig. 4d). Taken together, ZFNs are effective to induce gsdf GD in medaka fry capable of development into male and female adults. Fig. 3 Dose-dependent GD efficiency in medaka embryos. Curves of dose-dependent GD efficiency. Dashed curve is the curve of dose-dependent intensity (%) of heteroduplex bands; solid curve is the curve of dosedependent number of heteroduplex bands
fragments, which allows for distinction on an agarose gel between the WT allele and MT alleles in a Ht (Chen et al. 2012). Five adults were found to show faint bands of Ht forms. Part of the TAGE profile has already been published: fish #4 and #6 (numbered as #23 and #25, respectively here) exhibited a smear of faint band indicative of Ht products (Chen et al. 2012). A total of eight more representatives are shown in Figure S2, where #52 exhibited smaller bands (asterisk), which are faint on Figure S2 but more detectable on gel. PAGE revealed easily detectable Ht forms in 4 of these
Fig. 4 Genotyping of fry and adults. a Genotyping of fry. A major band of 346 bp for the wild-type locus is clearly visible by agarose gel electrophoresis (AGE) and polyacrylamide gel electrophoresis (PAGE). Multiple heteroduplex bands indicative of GD are clearly seen on PAGE (framed). 1∼7, representative fry from embryos injected with 40 ng/μl of ZFN mRNA; ctrl control fry from a non-injected embryo; neg negative PCR control without template DNA. Sizes in base pairs are given to the
Diversity of GD Alleles Three of the adult fish described above, namely #25, #38 and #52, were used to identify WT and MT gsdf alleles. To this end, DNA from fin clips was subjected to grsPCR and the resultant PCR products were cloned for sequencing. A total of 140 recombinant clones were successfully sequenced, which led to the identification of 7 different GD alleles, which fell into 3 categories (Fig. 5, Table 2). The first is the simple deletion (D) of a few bps within the target site, such as D1, D6 and D7 that have a 1-, 6- and 7-bp deletion. The second is the compound deletion, such as D7A4 containing a 7-bp deletion and a 4-bp addition (A), with the net consequence being a 3-bp deletion. The third is compound addition, where a larger addition accompanies a small deletion, with the net
left. b Genotyping of adults by PAGE, highlighting the presence of multiple heteroduplex bands (framed) diagnostic of multiple GD events in four representative animals from embryos injected with 10 ng/μl of ZFN mRNAs. marker, kb size markers with three bands being shown in base pairs to the left. (c and d) Adult male (c) and female fish (d) from ZFN-injected embryos. Df dorsal fin; cf caudal fin. Scale bars 0.5 cm
130
Mar Biotechnol (2014) 16:125–134
Fig. 5 Sequences of GD alleles. DNA sequences encompassing the ZFN-targeted site of three founder fish. D deletion; A addition. Spacer region is highlighted in bold
consequence being an addition by one to several bps, such as A16D3, A21D4 and A22D4 that are 13, 17 and 18 bp longer than the WT allele. These results demonstrate the diversity of ZFN-induced GD alleles. Germline Transmission The four adults described above were examined for germline transmission of GD alleles by progeny test. After crossing with control partners of strain Hd-rR, one of them showed germline transmission, which was male fish #38. The other female fish were not informative because the number of their progeny (11∼24) was not big enough. A total of 46 progeny fry from #38 were genotyped by PAGE, 7 of which had a GD event as they formed distinct Ht bands (Fig. 6a), producing a germline transmission rate of 15.2 %, which is within the range, 8.1 to 100 %, reported in ZFN-mediated exogenous GFP disruption in medaka (Ansai et al. 2012). As illustrated in Fig. 6 for 10 representative progeny fry, 7 fry were positive for GD. Since B =N 2 −N (where B is the number of Ht bands and N is the number of different alleles) as described previously
(Chen et al. 2012), a single GD allele could form two Ht bands (Fig. 6b). A closer inspection revealed that all the 7 fry exhibited two Ht bands of a similar size, suggesting the germline transmission of one and the same GD allele. Cloning and sequencing validated this observation (Fig. 6c). This GD allele had a 7-bp deletion covering the spacer. Interestingly, this GD allele in #38 fish was present in the somatic fin clip, in the testis, and further inherited to the next generation, strongly suggesting that this GD event could occur early in development before the separation between the soma and germline. GD Ontogeny The observation that injection of ZFN mRNAs could induce multiple GD events evoked us to examine GD ontogeny. To this end, embryos were injected again with a low dose (10 ng/μl) of ZFN mRNAs and examined in groups of 5 specimens each at different stages of development. Ht forms containing GD alleles were detectable from 1 day post fertilization (dpf) onwards (Fig. 7a). More specifically, a closer inspection revealed that the intensity and number of Ht bands increased
Table 2 Categories of GD alleles. Table showing the frequencies of each mutation type from the three founder fish Fish
Plasmid clones
Alleles, n (%) (second PCR products) WT
MT Deletion (D)
#25 #38 #52
56 58 26
D deletion; A addition
37 (66.0) 29 (50.0) 17 (65.4)
Irregular cases
D1
D6
D7
A21D4
1 (1.8)
4 (7.2) 1 (1.7) 8 (30.8)
16 (27.6)
12 (20.7)
D7A4
1 (3.8)
A22D4
A16D3
11 (19.6)
3 (5.4)
Mar Biotechnol (2014) 16:125–134
131
Fig. 6 Germline transmission of ZFN-mediated GD. a PAGE profile of F1 progeny of founder #38. 1∼10, representative F1 fry from founder #38; ctrl control fry from a non-injected embryo; neg negative PCR control without template DNA. Arrows depict heteroduplexes between WT and MT alleles. b Schematic view of heteroduplexes. A single GD
allele (Aa) with a WT allele (Bb) could form two Ht bands (Ab and Ba). c Allelic sequences. WT allele is shown at the top. The GD allele has a 7-bp deletion (D7), which is seen in the tail, testis of #38 and its F1 progeny. Spacer region is highlighted in bold
when embryogenesis proceeded (Fig. 7a). We then extended our observation into adulthood. Fish fins, in particular the caudal fin, have a high regeneration activity (Katogi et al. 2004). We utilized this property to examine the gsdf allelic
profile in fish #38 at adulthood at two different time point, 3 and 6 months post hatching. After a cut, the caudal fin regenerated, then re-cut is performed for DNA preparation and analysis. As illustrated in Fig. 7b, the band pattern altered
Fig. 7 GD ontogeny. a GD alleles at different stages of embryonic development. Groups of five embryos injected with 40 ng/μl were analyzed at each stage before (left panel) and after grsPCR (right panel). b GD alleles at different stages of posthatching development of fish #38. c GD alleles in adult organs. Fish #38 (top panel) and #52 (bottom panel) were used for DNA isolation from organs indicated above lanes and subjected to PAGE analyses. Hm homoduplex; Ht heteroduplex; mph months post hatching
132
with time. Finally, different organs representative the three germ layers were dissected from fish #38 and #52 after progeny test for germline transmission and analyzed for WT and MT alleles. This revealed that all organs of both fish had GD alleles by the presence of Ht forms and more importantly, the majority of organs examined—including those that share the same developmental origin such as the gut and liver—exhibited heterogeneous band patterns (Fig. 7c). On the other hand, organs of different germ layers shared a similar band pattern, as is the case between the fin and gonad or between the gill and gut in fish #52. Cloning and sequencing validated the presence of different GD alleles in the seven tissues or organs of fish #38 (Figure S3). The heterogeneity of GD alleles suggests multiple and independent GD events in medaka ontogeny. Since germ layers are formed during gastrulation and organogenesis takes places late (2∼3 dpf) in development, ZFN-induced GD must commence early in development and occurs repeatedly throughout embryogenesis. In summary, we have established procedures and parameters for successful gsdf GD by direct ZFN mRNA injection in medaka embryos. We show that such GD events occur efficiently during embryogenesis and in primordia that ultimately give rise to somatic organs and gonads of both sexes. Most importantly, we demonstrate that GD-bearing embryos can develop into normal adults capable of germline transmission of GD alleles into the next generation.
Discussion In this study, we have demonstrated the feasibility, and—for the first time to our knowledge—the efficacy of ZFNmediated direct GD at a chromosomal locus in medaka embryos. We have also established procedures and parameters that allow for maximal survival or maximal GD efficiency. Specifically, the dosage of ZFN mRNA injection correlates negatively with survival but positively with GD efficiency. RNA injection at a high dose of 40 ng/μl gives rise to nearly 100 % GD efficiency in operated fry, whereas RNA injection at a low dose of 10 ng/μl ensures a high survival rate of fry development. A high efficiency of direct GD at a chromosomal locus and parameters for a satisfactory balance between survival and GD efficiency will provide valuable information for setting up new ZFN-mediated GD experiments in medaka in particular and other fish species in general. The ZFN approach has proven its power for direct GD in several organisms (Carroll 2011). An important requirement for the introduction of ZFNs into a new model is an appropriate procedure, as well as detailed practical parameters to predict and implement effective GD. In our case, the streamlined protocol mainly consists of: design and construct ZFNs to target the desired gene; synthesize ZFN mRNA; efficiency test for designed ZFNs and optimize the parameters in
Mar Biotechnol (2014) 16:125–134
embryos and fry; GD detection assay for somatic and germline mutagenesis. The experience that we have in designing ZFNs is that we chose the target AA following closely downstream the signal peptide of the target gene. As the target locus is located at 5′ end, truncated gsdf protein, which may contribute to partial function of the intact gsdf, is avoided to the maximum extent. We chose mRNA instead of DNA as the ZFN delivery form since mRNA eliminates the risk of random genome integration, and lower cytotoxicity and off-target events by exposing cells to ZFNs for a shorter time. Additionally, we found that medaka cells do not tolerate input ZFN plasmid since all embryos are dead upon microinjection of a pair of ZFN plasmid (20 ng/μl). Although embryo injection of mRNAs for ZFN expression has proven to be practical in several organisms, zebrafish, rat, mouse, sea urchin, frog, and so on (Carroll 2011), parameters for efficient GD have, however, remained hardly elucidated. One of the critical issues for setting up a ZFN experiment is the dosage of mRNA injection. Fry test reveals that the dose of ZFN mRNA injection correlates negatively with the survival rate but positively with the GD efficiency, which ranges from 30 to 100 % when the ZFN mRNAs are injected at 10 to 40 ng/μl. Interestingly, when the dosage is raised to 60 ng/μl, the efficiency is lower than that of 40 ng/μl. One explanation is that high concentration of ZFN may cause toxicity to embryos or generate off-target mutations, leaving many GD embryos dead. In our example, GD occurs at a ∼4 % efficiency when the survival rate until the fry stage is ∼70 %. More intriguingly, with a relatively low dosage of 10 ng/μl, the percentage of individuals with successful GD falls from 30 % (Figure S4) to ∼4 % from fry to adults, indicating that the embryo containing considerable GD events due to a high dosage of ZFN RNA injection is less viable to its normal counterpart. In ZFN-injected medaka embryos, the number of different alleles increases when the development proceeds from 1 to 3 dpf, suggesting that introduced ZFNs persist and retain their specific cleavage activity throughout this period, thus continue to generate independent GD events. The persistence of ZFN activity could produce highly mosaic organs not only in the relative proportion of GD events but also in the number of various GD alleles. ZFN-induced mosaicism has been reported in the tail of rodents (Cui et al. 2011). In medaka, we have expanded such examination into a total of eight organs or tissues representative of the three germ layers that are formed during gastrulation in two F0 fish. This leads to an observation that the band pattern of GD alleles is different among all these organs, even between the gut and liver, which are both originated from endoderm during organogenesis and become discernable at 4 dpf in medaka (Watanabe et al. 2004). A difference between these two organs in GD band pattern and alleles strongly suggests that ZFNs continue to operate at least until the organogenesis stage.
Mar Biotechnol (2014) 16:125–134
One of the ultimate goals of genome editing is the germline transmission to produce whole animals containing a defined genetic alteration. In medaka, we notice that the efficiency of gsdf GD is higher in the soma than the germline. For example, only one of four adults positive for GD in the caudal fin is capable of germline transmission. Non-transmitting fish may contain an easily detectable proportion of GD alleles in many somatic organs and gonad. Non-transmitting fish #52, for example, is comparable to the transmitter #38 in the number and abundance of GD alleles in various organs. The presence of a discrepancy between a high efficiency of GD in the soma and a lack of GD germline transmission is not surprising, because the number of cell types and mass in the soma is substantially larger than the germline, the latter comprises only a single cell lineage, namely germ cells responsible for germline transmission. It deserves to note that the adult gonad, the ovary, and testis, contains also several types of somatic cells besides germ cells. For example, there are three larger Ht bands shown in the sample from #38 testis (Fig. 7c), in the F1 progeny of this founder, only two higher bands were observed (Fig. 6a), indicating there is another kind of GD allele existed in the testis of #38, present either in the gonadal somatic cells and/or in the germ cells that have not been transmitted to the next generation (Figure S3). However, compared to our result, the previous ZFN-mediated GFP knockout case in medaka reported higher germline transmitter rates ranging from 6.25 to 50 % under nearly the same injection dosages (Ansai et al. 2012). Besides the variations in customized ZFN activity, the targeted locus configuration and the efficacy of mutation detection, another explanation for the difference in germline transmission efficiency is that gsdf GD might affect germ cell development for gamete production. This possibility is supported by the report that in rainbow trout, gsdf knockdown suppresses primordial germ cell proliferation, and the addition of a recombinant GSDF to testicular cell culture enhances the proliferation of male germ stem cells spermatogonia (Sawatari et al. 2007). Targeted GD results in a wide variety of subtle allelic alterations including minor additions and deletions in fry and adults. A total of eight allelic alterations were defined by sequencing, including the seven mutation types listed in Fig. 4. All deletions are the consequence of loss by 1∼10 bp, whereas in the case of additions, besides a simple addition of several base pairs, additions can also result from a combination of additions and deletions. These results conform to previous studies in zebrafish (Doyon et al. 2008; Meng et al. 2008). Our data demonstrate that ZFNs can produce a wide variety of allelic alterations in medaka, which may lead to loss (frame shift mutations) and modification (in frame additions/deletions) of function of the target gene. The diversity of GD alleles will allow for detailed analyses of various subtle alterations on the activity and function of the target gene.
133
Engineered endonucleases, like ZFNs, have opened a new era of genome editing. Most recently, targeted mutagenesis in medaka embryos using custom-designed TALENs has been reported (Ansai et al. 2013). In this study, we have established procedures and parameters for a high efficiency of ZFNmediated GD in the soma and germline directly in medaka embryos. This procedures and parameters should be informative and applicable for ZFN-mediated direct GT in other fish species including those of aquaculture importance. Acknowledgments We thank J. Deng for fish breeding. This work was supported by the National Research Foundation Singapore (NRF-CRP72010-03) and the Japan Core Research for Evolutional Science and Technology (CREST). We acknowledge the NUS for scholarship to X. Zhang and J. B. Chen. Conflict of Interest The authors declare that they have no conflict of interest. Supporting Information Additional information noted in text includes supplementary figures.
References Ansai S, Ochiai H, Kanie Y, Kamei Y, Gou Y, Kitano T, Yamamoto T, Kinoshita M (2012) Targeted disruption of exogenous EGFP gene in medaka using zinc-finger nucleases. Develop Growth Differ 54: 546–556 Ansai S, Sakuma T, Yamamoto T, Ariga H, Uemura N, Takahashi R, Kinoshita M (2013) Efficient targeted mutagenesis in medaka using custom-designed transcription activator-like effector nucleases. Genetics 193:739–749 Baker M (2012) Gene-editing nucleases. Nat Meth 9:23–26 Capecchi MR (2005) Gene targeting in mice: functional analysis of the mammalian genome for the twenty-first century. Nat Rev Genet 6: 507–512 Carroll D (2011) Genome engineering with zinc-finger nucleases. Genetics 188:773–782 Chen J, Zhang X, Wang T, Li Z, Guan G, Hong Y (2012) Efficient detection, quantification and enrichment of subtle allelic alterations. DNA Res 19:423–433 Collins FS, Rossant J, Wurst W (2007) A mouse for all reasons. Cell 128: 9–13 Cui X, Ji D, Fisher DA, Wu Y, Briner DM, Weinstein EJ (2011) Targeted integration in rat and mouse embryos with zinc-finger nucleases. Nat Biotechnol 29:64–67 Doyon Y, Mccammon JM, Miller JC, Faraji F, Ngo C, Katibah GE, Amora R, Hocking TD, Zhang L, Rebar EJ, Gregory PD, Urnov FD, Amacher SL (2008) Heritable targeted gene disruption in zebrafish using designed zinc-finger nucleases. Nat Biotech 26: 702–708 Gautier A, Le Gac F, Lareyre JJ (2011) The gsdf gene locus harbors evolutionary conserved and clustered genes preferentially expressed in fish previtellogenic oocytes. Gene 472:7–17 Hong Y, Winkler C, Schartl M (1996) Pluripotency and differentiation of embryonic stem cell lines from the medakafish (Oryzias latipes). Mech Dev 60:33–44 Hong Y, Winkler C, Schartl M (1998) Production of medakafish chimeras from a stable embryonic stem cell line. Proc Natl Acad Sci U S A 95: 3679–3684
134 Hong Y, Liu T, Zhao H, Xu H, Wang W, Liu R, Chen T, Deng J, Gui J (2004) Establishment of a normal medakafish spermatogonial cell line capable of sperm production in vitro. Proc Natl Acad Sci U S A 101:8011–8016 Hong N, Li M, Zeng Z, Yi M, Deng J, Gui J, Winkler C, Schartl M, Hong Y (2010) Accessibility of host cell lineages to medaka stem cells depends on genetic background and irradiation of recipient embryos. Cell Mol Life Sci 67:1189–1202 Iwamatsu T (2004) Stages of normal development in the medaka Oryzias latipes. Mech Dev 121:605–618 Katogi R, Nakatani Y, Shin-I T, Kohara Y, Inohaya K, Kudo A (2004) Largescale analysis of the genes involved in fin regeneration and blastema formation in the medaka, Oryzias latipes. Mech Dev 121:861–872 Li M, Hong N, Xu H, Yi M, Li C, Gui J, Hong Y (2009) Medaka vasa is required for migration but not survival of primordial germ cells. Mech Dev 126:366–381 Liu T, Liu L, Wei Q, Hong Y (2011) Sperm nuclear transfer and transgenic production in the fish medaka. Int J Biol Sci 7:469–475 Meng X, Noyes MB, Zhu LJ, Lawson ND, Wolfe SA (2008) Targeted gene inactivation in zebrafish using engineered zinc-finger nucleases. Nat Biotechnol 26:695–701 Myosho T, Otake H, Masuyama H, Matsuda M, Kuroki Y, Fujiyama A, Naruse K, Hamaguchi S, Sakaizumi M (2012) Tracing the emergence of a novel sex-determining gene in medaka, Oryzias luzonensis. Genetics 191:163–170 Nasevicius A, Ekker SC (2000) Effective targeted gene ‘knockdown’ in zebrafish. Nat Genet 26:216–220 Sawatari E, Shikina S, Takeuchi T, Yoshizaki G (2007) A novel transforming growth factor-beta superfamily member expressed in gonadal somatic cells enhances primordial germ cell and spermatogonial proliferation in rainbow trout (Oncorhynchus mykiss). Dev Biol 301:266–275 Shibata Y, Paul-Prasanth B, Suzuki A, Usami T, Nakamoto M, Matsuda M, Nagahama Y (2010) Expression of gonadal soma derived factor (GSDF) is spatially and temporally correlated with early testicular differentiation in medaka. Gene Expr Patterns 10:283–289
Mar Biotechnol (2014) 16:125–134 Tong C, Li P, Wu NL, Yan Y, Ying Q-L (2010) Production of p53 gene knockout rats by homologous recombination in embryonic stem cells. Nature 467:211–213 Urnov FD, Rebar EJ, Holmes MC, Zhang HS, Gregory PD (2010) Genome editing with engineered zinc finger nucleases. Nat Rev Genet 11:636–646 Wang R-H, Sengupta K, Li C, Kim H-S, Cao L, Xiao C, Kim S, Xu X, Zheng Y, Chilton B, Jia R, Zheng Z-M, Appella E, Wang XW, Ried T, Deng C-X (2008) Impaired DNA damage response, genome instability, and tumorigenesis in SIRT1 mutant mice. Cancer Cell 14:312–323 Watanabe T, Asaka S, Kitagawa D, Saito K, Kurashige R, Sasado T, Morinaga C, Suwa H, Niwa K, Henrich T, Hirose Y, Yasuoka A, Yoda H, Deguchi T, Iwanami N, Kunimatsu S, Osakada M, Loosli F, Quiring R, Carl M, Grabher C, Winkler S, Del Bene F, Wittbrodt J, Abe K, Takahama Y, Takahashi K, Katada T, Nishina H, Kondoh H, Furutani-Seiki M (2004) Mutations affecting liver development and function in Medaka, Oryzias latipes, screened by multiple criteria. Mech Dev 121:791–802 Xu X, Wagner KU, Larson D, Weaver Z, Li C, Ried T, Hennighausen L, Wynshaw-Boris A, Deng CX (1999) Conditional mutation of Brca1 in mammary epithelial cells results in blunted ductal morphogenesis and tumour formation. Nat Genet 22:37–43 Yan Y, Hong N, Chen T, Li M, Wang T, Guan G, Qiao Y, Chen S, Schartl M, Li CM, Hong Y (2013) p53 gene targeting by homologous recombination in fish ES cells. PLoS One 8:e59400 Yano A, Guyomard R, Nicol B, Jouanno E, Quillet E, Klopp C, Cabau C, Bouchez O, Fostier A, Guiguen Y (2012) An immune-related gene evolved into the master sex-determining gene in rainbow trout, Oncorhynchus mykiss. Curr Biol 22:1423–1428 Yi M, Hong N, Hong Y (2009) Generation of medaka fish haploid embryonic stem cells. Science 326:430–433 Zhao H, Hong N, Lu W, Zeng H, Song J, Hong Y (2011) Fusion gene vectors allowing for simultaneous drug selection, cell labeling, and reporter assay in vitro and in vivo. Anal Chem 84:987–993
