View metadata, citation and similar papers at core.ac.uk brought to you by CORE R244 Dispatch provided by Elsevier - Publisher Connector
Zebrafish genetics: Harnessing horizontal gene transfer Eric S. Weinberg
The promiscuous spread of Tc1/mariner transposons involvement in the particular developmental process across species implies that host factors are relatively defective in the mutant. More systematic global unimportant for their transposition. Heterologous approaches are of course necessary. Positional cloning will elements can integrate on expression of the undoubtedly be effective, as demonstrated by the use of a corresponding transposases, an approach that should map-position-based approach for the recent identification greatly facilitate genetic analysis in the zebrafish. of one eyed pinhead as a gene encoding a ligand related to epidermal growth factor (EGF) [5]. But such approaches Address: Department of Biology, University of Pennsylvania, Philadelphia, Pennsylvania 19104, USA. are labor intensive and time consuming, even when a high- E-mail: [email protected] resolution map is available.
Current Biology 1998, 8:R244–R247 A promising alternative to positional cloning is the use of http://biomednet.com/elecref/09609822008R0244 a pseudotyped retroviral vector which can be transmitted © Current Biology Ltd ISSN 0960-9822 at high efficiency through the germline after integration into zebrafish DNA [6]. Such a vector has recently been The zebrafish, Danio rerio, has become a favored organism shown to be effective in causing mutations by insertion for attempts to identify genes that are important in devel- and allowing identification of the affected locus by isola- opmental and physiological processes in vertebrates. tion of the vector tag [7,8]. And a second insertional muta- Among the advantages of this organism [1,2] are the genesis method may be close at hand, thanks to the optical clarity of the embryos, the ease of breeding fish exploitation of a class of transposable elements that are and obtaining gametes, a generation time short enough to known to be promiscuous in their transmission from host allow practical genetic experiments, and amenability to to host during evolution [9,10]. the induction of mutations at high frequency. These fea- tures of the zebrafish have permitted large-scale mutant Gene transfer between species, a phenomenon known as hunts after treatment of fish with the chemical mutagen horizontal gene transmission, appears to have played an ethyl-N-nitrosourea (ENU) [3,4]. important role in the evolution of the Tc1/mariner super- family of transposons [11–14]. These DNA elements, Several thousand mutants were initially identified in two found in most, if not all, animal phyla, transpose by a cut- separate screens, and many of these mutants have already and-paste mechanism via a DNA intermediate, using an been grouped into several hundred genes by complemen- element-encoded transposase of the D,D(35)E superfam- tation analysis [3,4]. The observed phenotypes included ily [14,15]. They have short, inverted terminal repeats defects in development of a number of organs and tissues, and duplicate a TA target site upon insertion. Although disturbances in axonal pathfinding, defects in cell motil- the amino-acid identities within the transposase open ity, and abnormal gastrulation and cleavage patterns. The reading frame between the mariner and Tc1 subgroups are finding that most of the complementation groups had only only 10–20%, particular amino acids scattered throughout one allele indicates that each of the screens was far from the protein are conserved in virtually all family members saturation. A number of the mutants were due to new [14]. Horizontal transmission is inferred from the occur- alleles of genes previously identified by gamma-ray- rence of very similar transposon sequences in distantly induced or spontaneous mutations [1]. At this time, many related species [11–14]. other groups are employing variations of these screening methods to identify additional mutants. It is expected that The extremely broad range of these elements, an indica- that, within a year, hundreds of additional zebrafish tion of the lack of importance of species-specific host mutants will be identified with defects in developmental factors in the transposition process, has led to the sugges- processes such as gastrulation, pattern formation, organo- tion that Tc1/mariner elements could be used as general- genesis and cell differentiation. ized DNA vectors [12]. The first demonstration that a Tc1/mariner element could function in a foreign species The obvious next step is to identify the genes in which was the use of Mos1, an autonomous mariner-like element these mutations have occurred. A surprisingly large from Drosophila mauritiana, to direct integration of a non- number of the affected loci (perhaps a score at this point) autonomous target element into the genome of Drosophila have already been identified using a candidate gene melanogaster [16]. The Mos1 transposon was recently approach. However, such identifications depend on having shown to transpose in the genome of the trypanosomatid cloned genes with expression patterns suggestive of protozoan Leishmania major, resulting in the inactivation of Dispatch R245
Figure 1
(a) Constructs used for Sleeping Beauty- (a) pT/neo (b) pTc3GFP mediated genomic integration [9]. pT/neo is neo unc-22 unc-22 the substrate construct composed of inverted repeats of the salmonid Tc1-like element L48685 from Tanichtlys albonubes flanking a SV40 EF1α enhancer/ promoter promoter neo gene under the control of the SV40 SV40 poly(A) enhancer/promoter. pSB10 contains the GFP SV40 poly(A) pSB10 Tc3GFP resurrected Sleeping Beauty transposase SB transposase open reading frame under the control of the human CMV immediate early (IE) gene (b) CMV IE enhancer/promoter. Constructs used for enhancer/ promoter Tc3 element insertion in the zebrafish [10]. Tc3A mRNA pTc3GFP is the substrate construct Cap AAAA containing the inverted repeats from Current Biology transposon Tc3 flanking a GFP open reading frame expressed from a Xenopus EF1α Also shown is the Tc3GFP element after sequences; thick blue lines, flanking zebrafish promoter. As the Tc3 used to make this integration into zebrafish DNA. The elements sequences; red arrows, transposon inverted construct was derived from an integration in from both of these contexts were mobilized by repeats; green arrows, enhancer/promoter C. elegans, the inverted repeats are flanked injection of Tc3A transposase mRNA elements; other elements are labelled and by unc-22 gene sequence which was the prepared in vitro. Grey, transposon variously coloured. integration site of that particular insertion. sequences; thick black lines, flanking plasmid
at least one specific gene [17]. This was the first demon- between terminal inverted repeats from a salmonid-type stration that a Tc1/mariner transposase from one species element (Figure 1a). Integration of the donor element was can function in species of a different order. These results enhanced over background (a certain level of integration are consistent with the finding that recombinant purified was probably due to non-legitimate integrations typical in transposase is the only protein factor necessary for the transfected cells) in all three cell types, most notably in transposition of Tc1/mariner elements in vitro [18,19]. the HeLa cells which showed a 20-fold increase in G-418- resistant colonies. A promising system for genetic transformation and insertional mutagenesis that works in vertebrates has The enhancement of integration required a full-length recently been developed by Ivics et al. [9] using a Tc1-like transposase, as proteins that were incompletely changed to transposase encoded by a sequence reconstructed from a the Sleeping Beauty consensus, or that lacked the catalytic group of fish transposons of the salmonid Tc1-like sub- domain, did not promote integration. The presence of two group. The salmonid transposons, like other vertebrate inverted repeats in the donor substrate was essential for Tc1/mariner elements cloned and sequenced thus far, integration. Southern blots of DNA from the G-418 contain transposase pseudogenes and so cannot function selected cells showed that transgenes were integrated into autonomously. Aligning the sequence of 12 partial different locations of the human genome, and sequencing salmonid-type elements, Ivics et al. [9] restored an open of junction fragments revealed that they had the expected reading frame by removing premature translational stop duplicated flanking TA dinucleotides and intact inverted codons and frameshifts and systematically changing the repeat sequences. These findings indicated that the amino acids at 24 positions, creating a putative full-length Sleeping Beauty transposase can function faithfully in the transposase gene that matched the aligned 340 amino acid heterologous cultured cells. Thus, the transposase is a consensus sequence. This procedure resulted in the resur- potentially effective reagent for introduction of DNA for rection of an active transposase gene that was endearingly transgenesis and transposon tagging in vertebrates, includ- named Sleeping Beauty. ing the zebrafish.
After showing that the reconstructed transposase had an In a recent issue of Current Biology, Raz et al. [10] reported active nuclear localization signal domain and could specifi- a direct demonstration that a heterologous Tc1/mariner cally bind inverted repeat sequences of salmonid-type transposon system can work in the zebrafish. As a element, Ivics et al. [9] tested for integration activity in transposon source, they employed a plasmid construct, vertebrate cells. They co-transfected cultured carp, mouse pTc3GFP, containing the inverted repeats of the Tc1- and HeLa cells with a helper construct encoding a related element Tc3 from Caenorhabditis elegans, flanking cytomegalovirus (CMV) enhancer/promoter-driven Sleep- the gene for green fluorescent protein (GFP) under the ing Beauty transposase, and a donor element consisting of a control of the Xenopus elongation factor 1α (EF1α) promoter simian virus 40 (SV40) promoter-driven neo gene placed (Figure 1b). The Tc3 element used for this construct was R246 Current Biology, Vol 8 No 7
flanked by C. elegans unc-22 sequence, as this was the origi- the fish cells was enhanced far less than in HeLa cells. nal site of insertion into the C. elegans genome of this par- Moreover, Ivics et al. [9] reported only relative numbers of ticular Tc3 insertion. G-418 selectable transformants, and not absolute numbers of integration events. In a direct test of integration in One-cell-stage zebrafish embryos were co-injected with zebrafish embryos, Raz et al. [10] demonstrated only one pTc3GFP and in vitro transcribed mRNA encoding Tc3A legitimate integration in 40 embryos co-injected with transposase. Of 40 embryos raised from such injected transposase mRNA and substrate DNA. Although excision embryos, one fish was shown to have a legitimate transpo- of integrated transposons after exposure to injected trans- son insertion that could be transmitted through the posase mRNA was clearly demonstrated, there was no germline (line 3-2), and two fish carried non-legitimate direct evidence for reintegration of these same transposon integrations (containing flanking plasmid and unc-22 sequences. Furthermore, no data on the efficiency of exci- sequences) of the type that is normally seen after injection sion were presented. of any DNA into fish embryos. The line 3-2 founder fish transmitted the transposon to at least 7% of its progeny, One potential worry is that efficient mobilization may take which were able to express GFP. The integrated transpo- place only within a narrow range of transposase concentra- son had the expected repeat of an endogenous TA target tion. At least one mariner transposase has the unusual and intact inverted repeat sequences. property, termed overproduction inhibition, of being less active at higher transposase concentration [20]. This The question now was whether an integrated transposon feature may have to be considered in designing conditions could be correctly mobilized after another round of expo- that optimize the efficiency of mobilization. The large- sure to Tc3A transposase. For this test, Raz et al. [10] used scale zebrafish mutant screens [1,2] were successful in line 3-2 carrying the legitimate integration, as well as large part because of the high efficiency of chemical muta- another line of fish (line 3-7) carrying a non-legitimate genesis. The mutagenesis and mating protocols allowed integration of pTc3GFP derived from an injected embryo the screening of thousands of mutagenized haploid that did not receive transposase mRNA. One-cell-stage genomes from relatively few ENU-treated adult male fish embryos from both of these fish lines were injected with (49 in one screen, 240 in the other). Each mutagenized transposase mRNA, and after one day of embryogenesis, haploid-genome equivalent assayed in the screen carried excision events were assayed by PCR using oligonu- approximately one mutation that resulted in a visible phe- cleotides specific for sequences flanking the transposon. A notype when homozygous. high proportion of embryos (>80%) from both fish lines appeared to support specific transposon excision. More- Insertional mutagenesis screens will undoubtedly have to over, the sites of excision were PCR amplified and be carried out by raising thousands of founder fish from sequenced and shown to contain a small CAG or CTG injected embryos. The pseudotyped viral vector devel- footprint flanked by TA repeats. Although the footprint is oped by Hopkins and colleagues [6] is capable of generat- one base larger than the usual case for Tc3 excision, the ing 10 insertions per founder fish, and it is estimated that transposons have clearly been specifically and cleanly 100,000–200,000 insertions could be produced by injecting mobilized in the context of either flanking zebrafish (line 10,000–20,000 founders, a undertaking that might take 3-2) or flanking unc22 (line 3-7) sequences. four to six workers approximately three months [6]. Depending on the proportion of such insertions that result Lastly, Raz et al. [10] followed the ability of fish to express in mutant phenotypes, such an effort may yield numbers GFP from the integrated transposon. They observed GFP of mutants comparable to what has been achieved by expression after three generations of germline expression chemical mutagenesis. Whether the Tc1/mariner transpo- in zebrafish lines 3-2 and 3-7, with a more constant expres- son-based vector systems can also function at this level of sion in the legitimate insertion line 3-2. Expression was efficiency is still very much unknown. Tc1 and Tc3 both maintained in embryos at least to day 5 in line 3-2, but recognize consensus sequences for integration, and target faded more quickly in line 3-7. These experiments have choice sites are far from random [15]. Thus, although now set the stage for direct tests of insertional mutagene- many C. elegans genes have been cloned by Tc1 transposon sis, transposon tagging and enhancer/gene trap analysis in tagging, there are many genes that are refractive to trans- the zebrafish. poson mutagenesis even in this organism.
The work of Ivics et al. [9] and Raz et al. [10] shows Attention has been given to the specificity of mobilization promise, but a number of important questions remain. with heterologous transposases. The zebrafish genome First, there is some concern about the efficiency of inte- does in fact harbor Tc1-like elements [21–24], at least one gration and mobilization. Great variability in the efficiency of which can mobilize spontaneously [24]. Sleeping Beauty of integration of the Sleeping Beauty substrate was found in transposase binds to the inverted repeats of salmonid cultured carp, mouse and human cells, and integration in transposons, but apparently not to the analogous region of Dispatch R247
the zebrafish Tdr1 element [9]. As the zebrafish elements 11. Maruyama K, Hartl DL: Evidence for interspecific transfer of the transposable element mariner between Drosophila and found thus far appear to be distinct from the salmonid ele- Zaprionus. J Mol Evol 1991, 33:514-524. ments in their inverted repeat sequences, including differ- 12. Kidwell MG: Horizontal transfer. Curr Opin Genet Dev 1992, 2:868- ences in a portion of the sequence of the transposase 873. 13. Kidwell MG: Voyage of an ancient mariner. Nature 1993, 362:202. binding site [9], it is reasonable to expect that Sleeping 14. Robertson HM: The Tc1-mariner superfamily of transposons in Beauty transposase will not mobilize endogenous zebrafish animals. J Insect Physiol 1995, 41:99-105. 15. Plasterk RHA: The Tc1/mariner transposon family. Curr Top transposons. The Tc3 transposon also has inverted repeats Microbiol Immunol 1996, 204:125-143. that are distinct from those of the described zebrafish 16. Garza D, Medhora M, Koga A, Hartl DL: Introduction of the endogenous elements. It is thus unlikely that Tc3 trans- transposable element mariner into the germline of Drosophila melanogaster. Genetics 1991, 128:303-310. posase would mobilize these elements, although direct 17. Gueiros-Filho FJ, Beverley SM: Trans-kingdom transposition of the tests have not been performed. Drosophila element mariner within the protozoan Leishmania. Science 1997, 276:1716-1719. 18. Vos JC, De Barre I, Plasterk RHA: Transposase is the only The Sleeping Beauty and Tc3 transposon systems are most nematode protein required for in vitro transposition of Tc1. Genes encouraging with respect to expression of genes carried by Dev 1996, 10:755-761. 19. Lampe DJ, Churchill MEA, Robertson HM: A purified mariner the transposed substrates. Expression of neo in cultured transposase is sufficient to mediate transposition in vitro. EMBO cells, and of GFP in injected embryos and derived fish J 1996, 15:5470-5479. lines, suggests that the transposon systems may be 20. Lohe AR, Hartl DL: Autoregulation of mariner transposase activity by overproduction and dominant-negative complementation. Mol amenable to enhancer and gene trap approaches, as well as Biol Evol 1996, 13:549-555. serving as a good general system for transgenesis in fish. It 21. Radice AD, Bugaj B, Fitch DHA, Emmons SW: Widespread will now be of great interest to test for activity of zebrafish occurrence of the Tc1 transposon family: Tc1-like transposons from teleost fish. Mol Gen Genet 1994, 244:606-612. promoter elements and additional ubiquitously active 22. Izvák Z, Ivics Z, Hackett PB: Characterization of a Tc1-like transcriptional elements using the Tc3 vector. Another transposable element in zebrafish (Danio rerio). Mol Gen Genet 1995, 247:312-322. potential advantage is the large size of DNA — up to 23. Ivics Z, Izsvák Z, Minter A, Hackett PB: Identification of functional 40 kb according to unpublished results cited in Raz et al. domains and evolution of Tc1-like transposable elements. Proc [10] — that can be transferred by transposons, in contrast Natl Acad Sci USA 1996, 93:5008-5013. 24. Lam WL, Lee T-S, Gilbert W: Active transposition in zebrafish. Proc to more restricted insert sizes possible with viral vectors. Natl Acad Sci USA 1996, 93:10870-10875. The use in zebrafish of Tc1/mariner transposons for trans- genesis, insertional mutagenesis, gene and enhancer traps, and genomic manipulations is an exciting prospect and, if successful, will greatly enhance the potential of this verte- brate for genetic analysis.
References 1. Kimmel CB: Genetics and early development of zebrafish. Trends Genet 1989, 5:283-288. 2. Weinberg ES: Analysis of early development in the zebrafish embryo. Res Problems Cell Diff 1992, 18:91-150. 3. Hafter P, Grantao M, Brand M, Mullins MC, Hammerschmidt M, Kane DA, Odenthal J, van Eeden FJM, Jiang Y-J, Heisenberg C-P, et al.: The identification of genes with unique and essential functions in the development of the zebrafish, Danio rerio. Development 1996, 123:1-36. 4. Driever W, Solnica-Krezel L, Schier AF, Neuhauss SCF, Malicki J, Stemple DL, Stainier DYR, Zwartkruis F, Abdelilah S, Rangini Z, et al.: A genetic screen for mutations affecting embryogenesis in zebrafish. Development 1996, 123:37-46. 5. Zhang J, Talbot WS, Schier AF: Positional cloning identifies zebrafish one-eye pinhead as a permissive EGF-related ligand required during gastrulation. Cell 1998, 92:241-251. 6. Gaiano N, Allende M, Amsterdam A, Kawakami K, Hopkins N: Highly efficient germ-line transmission of proviral insertions in zebrafish. Proc Natl Acad Sci USA 1996, 93:7777-7782. 7. Gaiano N, Amsterdam A, Kawakami K, Allende M, Becker T, Hopkins N: Insertional mutagenesis and rapid cloning of essential genes in zebrafish. Nature 1996, 383:829-832. 8. Allende M, Amsterdam A, Becker T, Kawakami K, Gaiano N, Hopkins N: Insertional mutagenesis in zebrafish identifies two novel genes, pescadillo and dead eye, essential for embryonic development. Genes Dev 1996, 10:3141-3155. 9. Ivics Z, Hackett PB, Plasterk RH, Izsvák Z: Molecular reconstruction of Sleeping Beauty, a Tc1-like transposon from fish, and its transposition in human cells. Cell 1997, 91:501-510. 10. Raz E, van Luenen HGAM, Schaerringer B, Plasterk RHA, Driever W: Transposition of the nematode Caenorhabditis elegans Tc3 element in the zebrafish Danio rerio. Curr Biol 1998, 8:82-88.