WO2004044150A2

WO2004044150A2 - Systems for gene targeting and producing stable genomic transgen e insertions

Info

Publication number: WO2004044150A2
Application number: PCT/US2003/035587
Authority: WO
Inventors: Alfred M. Handler; Carsten Horn
Original assignee: US Department of Agriculture USDA
Current assignee: US Department of Agriculture USDA
Priority date: 2002-11-07
Filing date: 2003-11-07
Publication date: 2004-05-27
Anticipated expiration: 2005-05-07
Also published as: AU2003290660A8; WO2004044150A3; AU2003290660A1; DE10251918A1

Abstract

The novel germ-line transformation systems disclosed in this patent application allow the physical deletion of transposon DNA following the transformation process, and the targeting of transgene integrations into predefined target sites. In this way, transposasemediated mobilization of genes-of-interest are excluded mechanistically and random genomic integrations eliminated. In contrast to conventional germ-line transformation technology, our systems provide enhanced stability to the transgene insertion. Furthermore, DNA sequences required for the transgene modification (e.g.transformation marker genes, transposase or recombinase target sites), are largely removed from the genome after the final transgene insertion, thereby eliminating the possibility for instability generated by these processes. The RMCE technology, which is disclosed in this patent application for invertebrate organisms (exemplified in Drosophila melanogaster) represents an extremely versatile tool with application potential far beyond the goal of transgene immobilization. RMCE makes possible the targeted integration of DNA cassettes into a specific genomic loci that are pre-defined by the integration of the RMCE acceptor plasmid. The loci can be characterized prior to a targeting experiment allowing optimal integration sites to be pre-selected for specific applications, and allowing selection of host strains with optimal fitness. In addition, multiple cassette exchange reactions can be performed in a repetitive way where an acceptor cassette can be repetitively exchanged by multiple donor cassettes. In this way several different transgenes can be placed precisely at the same genomic locus, allowing, for the first time, the ability to eliminate genomic positional effects and to comparatively study the biological effects of different transgenes.

Description

Systems for gene targeting and producing stable genomic transgene insertions

Field of the Invention

The invention relates to novel methods and techniques to produce transgenic, or genetically modified, organisms (transgenesis). The focus of the innovation is on manipulation techniques that allow for the targeting and the stable anchoring of homologous or heterologous DNA- sequences (in the following description referred to as: "transgene" or "gene-of-interest") into the genome of a target species. To achieve this goal, we have developed three different systems of transformation vectors that are capable of integrating a transgene into invertebrate and vertebrate organisms via transposon- or recombinase-mediated transformation events. In addition, following the germline transformation procedure, both systems make possible the physical deletion of mobile DNA-sequences, brought in with the vector, from the target genome and therefore to stabilize the gene-of-interest. Stable (genomic) transgene insertions are regarded to be an essential pre-requisite for the safe production of genetically modified organisms at a large industrial scale.

Description of the Related Art

Current state-of-the-art technology to produce genetically modified insect organisms relies on transposon-mediated germ-line transformation. This transformational technique is based on mobilizable DNA, i.e. transformation vectors derived from Class II transposable elements having terminal inverted sequences, which transpose via a DNA-mediated process (see Finnegan, D.J., 1989. Eucaryotic transposable elements and genome evolution. Trends Genet. 5, 103-107, and Atkinson, P.W., Pinkerton, A.C., O'Brochta, D.A., 2001. Genetic transformation systems in insects. Annu. Rev. Entomol. 46, 317-346, the contents of which are incorporated herein by reference). The two ends of such a transposable element carrying within all functional parts necessary and sufficient for in vivo mobilization are termed TransposonL (5' end) and TransposonR (3' end). Several different germ-line transformation systems have in common that a gene-of-interest/transgene originally located within a transgene construct is transferred into genomic DNA of germ-line cells of the target species. The transformation process is catalyzed by the transposase enzyme provided by a helper plasmid. This enzyme recognizes DNA target sites flanking the gene-of-interest/transgene and mobilizes the transgene into the genome of germ-line cells of the insect species. In addition, transformed DNA contains a marker gene that allows detection of successful germ-line transformation events (by producing a dominantly visible phenotype).

Transposon-mediated germ-line transformation systems are currently available for a diverse spectrum of insect species. Systems based on the R-element revolutionized the genetics of the vinegar fly Drosophila melanogaster (see Engels, W.R. (1996). P elements in Drosophila. Curr. Top. Microbiol. Immunol. 204, 103-123, the contents of which are incorporated herein by reference), but they were not applicable to non-drosophilid insect species because of the dependence of P-elements on røsop α-endogenous host factors (see Rio, D.C. & Rubin, G.M. (1988). Identification and purification of a Drosophila protein that binds to the terminal 31 -base-pair inverted repeats of the P transposable element. Proc. Natl. Acad. Sci. USA 85, 8929-8933, the contents of which are incorporated herein by reference). Therefore, insect species of medical or economic importance have been transformed using host factor- independent "broad host range" transposable elements (see Atkinson, P.W. & James, A.A. (2002). Germline transformants spreading out to many insect species. Adv. Genet. 47, 49-86, the contents of which are incorporated herein by reference). Germline transformation systems based on the transposable elements piggyBac (see United States Patent No. US 6,218,185; WO 01/14537; and Handler, A.M., McCombs, S.D., Fraser, M.J., Saul, S.H. (1998). The lepidopteran transposon vector, piggyBac, mediates germline transformation in the Mediterranean fruitfly. Proc. Natl. Acad. Sci. USA 95, 7520-7525, the contents of which are incorporated by reference herein), Hermes (see United States Patent No. US 5,614,398, the contents of which are incorporated herein by reference), Minos (see European Patent No. EP 0 955 364 A36, the contents of which are incorporated herein by reference) and mariner (see WO 99/09817, the contents of which are incorporated herein by reference) are currently state- of-the-art technology to genetically modify important pest or useful insect species including, for example, malaria transmitting anopheline or culicine mosquitoes (Anopheles gambiae, Anopheles stephensi, Anopheles albimanus, Culex quinquefasciatus, Aedes aegypti; see Catteruccia, F., Nolan, T., Loukeris, T.G., Blass, C, Savakis, C, Kafatos, F.C. & Crisanti, A. (2000). Stable germline transformation of the malaria mosquito Anopheles stephensi. Nature 405, 959-962, and Allen, M.L., O'Brochta, D.A., Atkinson, P.W. & Levesque, C.S. (2001). Stable, germ-line transformation of Culex quinquefasciatus (Diptera : Culicidae). J. Med. Entomol. 38, 701-710, and Coates J.C., Jasinskiene, N., Miyashiro, L. & James, A.A. (1998). Mariner transposition and transformation of the yellow fever mosquito, Aedes aegypti. Proc. Natl. Acad. Sci. USA 95, 3748-3751, and Jasinskiene, N., Coates, C.J., Benedict, M.Q., Cornel, A.J., Rafferty, C.S., James, A.A. & Collins, F.H. (1998). Stable transformation of the yellow fever mosquito, Aedes aegypti, with the Hermes element from the housefly. Proc. Natl. Acad. Sci. USA 95, 3743-3747, and Perera, O.P., Harrell, R.A., Handler, A.M. (2002) Germline transformation of the South American malaria vector, Anopheles albimanus, with a piggyBac/EGFP transposon vector is routine and highly efficient. Insect Mol. Biol., 11, 291- 297, the contents of which are incorporated herein by reference), the Mediterranean fruit fly, Ceratitis capitata (see Handler, A.M., McCombs, S.D., Fraser, M.J., Saul, S.H. (1998). The lepidopteran transposon vector, piggyBac, mediates germline transformation in the Mediterranean fruitfly. Proc. Natl. Acad. Sci. USA 95, 7520-7525 and Loukeris, G.T., Livadaras, I., Area, B, Zabalou, S. & Savakis, C. (1995). Gene transfer into the Medfly, Ceratitis capitata, with a Drosophila hydei transposable element. Science 270, 2002-2005, the contents of which are incorporated herein by reference) and the silkworm, Bombyx mori (see Tamura, T. et al. (2000). Germline transformation of the silkworm Bombyx mori L. using a piggyBac transposon-derived vector. Nat. Biotechnol. 18, 81-84, the contents of which are incorporated herein by reference). Moreover, the application potential of broad host range transposable elements is not restricted to insect species: warmer-derived transformation vectors have been shown to integrate stably into the germ-line of the nematode, Caenorhabditis elegans (see Bessereau, J.-L., Wright, A., Williams, D.C., Schuske, K., Davis, M.W. & Jorgensen, E.M. (2001). Mobilization of a Drosophila transposon in the Caenorhabditis elegans germ line. Nature 413, 70-74, the contents of which are incorporated herein by reference), the zebrafish, Danio rerio (see Fadool J.M., Haiti, D.L. & Dowling, J.E. (1998). Transposition of the mariner element from Drosophila mauritiana in Zebrafish. Proc. Natl. Acad. Sci. USA 95, 5182-5186, the contents of which are incorporated herein by reference) and chicken, Gallus spp. (see Sherman, A., Dawson, A., Mather, C, Gilhooley, H., Li, Y., Mitchell, R., Finnegan, D. & Sang, H. (1998). Transposition of the Drosophila element mariner into the chicken germ line. Nat. Biotechnol. 16, 1050-1053, the contents of which are incorporated herein by reference).

In order to follow germ-line transformation success, both species-specific and species- independent transformation markers have been established (see Horn, C, Schmid, B.G.M., Pogoda, F.S. & Wimmer, E.A. (2002). Fluorescent transformation markers for insect transgenesis. Insect Biochem. Mol. Biol. 32, 1221-1235, the contents of which are incorporated herein by reference). Species-independent markers consist of a combination of a promoter sequence which is phylogenetically conserved and a gene for a fluorescent protein placed under control of such a promoter (for example, GFP [green fluorescing protein] and derivatives thereof, or DsRed [Discosoma species red fluorescing protein] (see Chalfie, M. Tu, Y., Euskirchen, G., Ward, W., Prasher, D.C. (1994). Green fluorescent protein as a marker for gene expression. Science 263, 802-805, and Matz, M.V., Fradkov, A.F., Labas, Y.A., Savitsky, A.P., Zaraisky, A.G., Markelov, M.L., Lukyanov, S.A. (1999). Fluorescent proteins from nonbioluminescent Anthozoa species. Nat. Biotechnol. 17: 969-973, the contents of which are incorporated herein by reference). Species-independent markers are advantageous over species-specific markers because they are directly applicable to different insect species (and other organisms). The polyubiquitin-promoter (see Patent Cooperation Treaty PCT WO 01/14537 Al and Handler, A.M. & Harrell, R.A. (1999). Germline transformation of Drosophila melanogaster with the piggyBac transposon vector. Insect Mol. Biol. 8, 449-457, the contents of which are incorporated herein by reference) as well as the "3xP3"-promoter (see Patent Cooperation Treaty PCT WO 01/12667 Al and Berghammer, A.J., Klingler, M., & Wimmer, E.A. (1999). A universal marker for transgenic insects. Nature 402, 370-371 , the contents of which are incorporated herein by reference) linked to genes for fluorescent proteins have been used most widely for this purpose.

A transposon-independent technology aiming at targeting a gene-of-interest/transgene into the genome of cells relies on the principle of site-specific recombination. This is possible by using a recombinase enzyme and corresponding DNA target sites that are heterospecific. The steps are: First, incorporating into the genome by transposon-mediated transformation, a DNA cassette that is flanked by heterospecific recombinase target sites and contains a marker system for positive-negative selection. Second, recombinase-mediated targeting into the marked genomic locus the gene-of-interest, which is located within a plasmid and is flanked by the same heterospecific recombinase target sites. This principle has been described as RMCE or recombinase-mediated cassette exchange (see European Patent No. EP 0 939 120 Al and Baer, A. & Bode, J. (2001). Coping with kinetic and thermodynamic barriers: RMCE, an efficient strategy for the targeted integration of transgenes. Curr Opin Biotechnol.12, 473-480 and Kolb, A.F. (2002). Genome engineering using site-specific recombinases. Cloning Stem Cells. 4, 65- 80, the contents of which are incorporated herein by reference). The functionality of DNA cassette exchange systems has been demonstrated in different cell lines (comprising also murine embryonic stem cells) using the FLP-recombinase enzyme and heterospecific FRT target sites (see Schlake, T. & Bode, J. (1994). Use of mutated FLP recognition target (FRT) sites for the exchange of expression cassettes at defined chromosomal loci. Biochemistry 33, 12746-12751, and Seibler, J., Schϋbeler, D., Fiering, S, Groudine, M. & Bode, J. (1998). DNA cassette exchange in ES cells mediated by Flp recombinase: an efficient strategy for repeated modification of tagged loci by marker- free constructs. Biochemistry 37, 6229-6234, and European Patent No. EP 0 939 120 Al, the contents of which are incorporated herein by reference) as well as using the Cre-recombinase enzyme and heterospecific loxP target sites (see Kolb, A.F. (2001). Selection-marker- free modification of the murine beta-casein gene using a lox2272 [correction of lox2722] site. Anal Biochem. 290, 260-271.26), the contents of which are incorporated herein by reference ). However, RMCE has not been applied to genetically modified invertebrate organisms thus far.

Limitations of Prior Art / Improvements over Prior Art

Transposon-based plasmid vectors have proven to be efficient tools for producing genetically modified insects for research purposes, but so far only on a small laboratory scale. However, the mobile nature of DNA transposable elements will be disadvantageous when scaling up the production/rearing of genetically modified insects. Owing to potential re-mobilization, the stability of genomic transgene integrations cannot be assured and, connected to this issue, concerns relating to the safety of release of such genetically modified insects will be raised. Stability of genomic transgene integrations in large industrial scale

The current state-of-the-art provides, typically, for random transposon vector integrations into the host genome. While this may be advantageous for functional genomics studies that use vector integrations to cause random mutations (e.g. for transposon-tagging and enhancer trapping), it is typically disadvantageous for the creation of transgenic strains for applied use where high fitness levels and optimal transgene expression are desired. This results from integrations that create mutations by insertion into genomic sites that eliminate or disrupt normal gene function that negatively effect viability, reproduction, or behavior. Genomic position effects also influence expression of transgenes, typically causing decreased expression and or mis-expression of genes of interest and markers so that transformants may not be easily identified, and the desired transgene expression for application is not achieved. Thus, most transformation experiments require the screening of multiple transformant strains for optimal fitness and transgene expression, and often such strains cannot be identified. An important improvement over the current state-of-the-art would be an efficient and routine system to target transgene integrations to specific and defined genomic sites that are known not to disrupt normal gene function and whose position effects are limited or well characterized. Transgene integrations that negatively effect host strain fitness and reproduction also confer a selective disadvantage to the transformed organism in a population relative to wild type organisms. Thus, a selective advantage is provided to non-transformed organisms or transformants that have lost or relocated the transgene due to a re-mobilization event. Re- mobilization requires the activity of a transposase enzyme corresponding to, and acting upon, the transposon sequences flanking the genomic transgene. Although the transposase used for germ-line transformation usually is not encoded by the host species' genome, transposase introduction by symbiotic or infectious agents is possible, and cross-reactivity to related transposase enzymes that are genomically encoded cannot be excluded. Such cross-reactivities have been reported between the transposable elements Hermes, from Musca domestica, and hobo, from Drosophila melanogaster, that caused significant instablity of Hermes-flanked transgenes in λøbo-containing Drosophila strains (see Sundararajan, P., Atkinson, P.W. & O'Brochta, D.A. (1999). Transposable element interactions in insects: crossmobilization of hobo and Hermes. Insect Mol. Biol. 8, 359-368, the contents of which are incorporated herein by reference). It should be noted that well-characterized families of transposable elements contain multiple members and the cross-reactivity of them is largely unknown to date (e.g. the mariner/Tel superfamily (see Ηartl, D.L., Lohe, A.R. & Lozovskaya, E.R. (1997). Modem thoughts on an ancyent marinere: function, evolution, regulation. Annu. Rev. Genet. 31, 337- 358, the contents of which are incorporated herein by reference)). For these reasons, a transformation technology that excludes the possibility of transgene re-mobilization events a priori will provide a higher standard of transgene stability and will be superior to currently available technology.

Transgene instability resulting from vector remobilization will have several negative consequences. The first is loss or change in desired transgene expression. Secondly, strain breakdown will result after relocated transgenes can segregate freely in meiosis and selection pressure acts against transgene-carrying chromosomes. Research results on the stability of transgene insertions in insects, reared at an industrial scale, have not been reported thus far. However, data for insect strains selected by classical Mendelian genetics and carrying translocations are available (see Franz, G., Gencheva, E. & Kerremans, Ph. (1994). Improved stability of genetic sex-separation strains for the Mediterranean fruit fly, Ceratitis capitata. Genome 37, 72-82, the contents of which are incorporated herein by reference). When reared at an industrial scale, such translocation strains, constructed for the Mediterranean fruit fly (see Franz, G., Gencheva, E. & Kerremans, Ph. (1994). Improved stability of genetic sex-separation strains for the Mediterranean fruit fly, Ceratitis capitata. Genome 37, 72-82, the contents of which are incorporated herein by reference) suffered from instability. Recombination events causing reversion of the selected recessive trait were observed at a frequency of 10^"3 - 10^" (see Franz, G. (2002). Recombination between homologous autosomes in medfly (Ceratitis capitata) males: type-1 recombination and the implications for the stability of genetic sexing strains. Genetica 116, 73-84, the contents of which are incorporated herein by reference). Because the recessive trait conferred a selective disadvantage to the individual insect, such reversion events caused strain breakdown rapidly. Most interestingly, these events were not observed at a small laboratory scale and therefore were not anticipated. As strain breakdown during a continuous industrial production of those insects is not acceptable, major research efforts have been made to improve the situation. Cuπently a laborious (and expensive) but efficient manual detection system for quality control has been implemented (see Fisher, K. & Caceres, C. (2000). A filter rearing system for mass reared medfly, S. 543-550 in Area-wide control of fruit flies and other insect pests, Ed.: Tan, K.H., Penerbit Universiti Sains Malaysia, Penang, Malaysia, the contents of which are incorporated herein by reference) and allows the successful production of this translocation strain at a scale of 10⁶ - 10⁷ individuals per week (see Franz, G. (2002). Recombination between homologous autosomes in medfly (Ceratitis capitata) males: type-1 recombination and the implications for the stability of genetic sexing strains. Genetica 116, 73-84, the contents of which are incorporated herein by reference). Safety aspects concerning release of genetically modified insects

Another important concern for remobilization is the potential for lateral transmission of the transgene into unintended host strains or species. Many industrial applications of insect transgene technology will include the release of genetically modified insects into the environment (e.g. the Sterile Insect Technique). Therefore, aspects of biosafety and ecological risk assessment will be of fundamental importance. Biosafety includes minimizing the risk of unintended transgene transmission from the host to other procaryotic or eucaryotic species during rearing or after release into the field. Horizontal gene transfer cannot be excluded per se, because the mechanisms of nucleic acid exchange between species are not sufficiently investigated to date. While most transposon vectors have their transposase source eliminated and are not self-mobilizable, functional autonomous transposons can be transmitted among species horizontally, and transposase may be provided to the vector by associated organisms or by a related enzyme in the host species. Thus, the risk for transgene vector re-mobilization by a transposase-mediated event can be most definitively eliminated when transposon sequences, required for germ-line transformation, are removed from the genomic integration after the transformation process. Systems disclosed in this patent application contribute to risk minimization by introducing techniques for transposon sequence removal. It is probable that, in the future, procedures to remove such sequences, and therefore to assure a higher standard of biosafety, will become an obligate precondition for permission by regulatory organizations for release of transgenic organisms. In fact, there are sound prospects that such systems will set the safety standards and will become normative which in turn demonstrates the commercial potential of the invention.

Brief Summary of the Invention

The strategy: Post-transformational immobilization of transgenes

Disadvantages stated in the previous section show the need for novel germ-line transformation systems that enable the stable integration of transgenes/genes-of-interest. The challenge is to develop a transformation method that prevents re-mobilization of transgenes which have been incorporated into the genome. The strategy disclosed in this patent application is to remove the intact transposon parts (containing transposase-recognition sites) following the transformation procedure (i.e. post-transformational). Three variants of this invention are disclosed as embodiments. These variants allow (i) modification of transgene DNA, (ii) post- transformational inactivation of at least one of the transposon parts and (iii) inactivation of at least one of the two transposon recognition sites required for re-mobilization by physical deletion from the genome.

The first embodiment disclosed has been termed "excision-competent stabilization vectors" (Fig. 1). This embodiment comprises a transformation vector that, in addition to currently applied vectors that contain solely a TransposonLl half side and TransposonRl half side (now refeπed to as TransposonLl and Rl), contains an additional internally-positioned TransposonL half side (refeπed to as TransposonL2 in Fig. 1) placed in-between the original Transposon LI and Rl sides. L and R half sides are placed in the normal, or same, terminal inverted repeat orientation to one another as found in the original transposable element. Marker genes that can be distinguished from one another are placed in-between TransposonLl and TransposonL2 and in-between TransposonL2 and TransposonRl. The steps of transformation are as follows. First, the transformation procedure is carried out according to the current state-of-the-art germline transformation technology that will result in individuals transformed by one of two possible events with this vector. One possible event is the integration of TransposonLl and TransposonRl and all intervening DNA including the two marker genes, TransposonL2, and other genes of interest. The second possible event is integration of TransposonL2 and TransposonRl and all intervening DNA including the marker gene. For the purposes of this embodiment, only individuals transformed with TransposonLl and TransposonRl, which are identified by expression of the two marker genes, are conserved for further experimentation. The internal vector containing TransposonL2 and TransposonRl, within TransposonLl and TransposonRl, is then re-mobilized by introduction of a source of transposase derived from mating to a jumpstarter strain having a genomic transposase gene, or physical injection of the transposase DNA, RNA, or protein into embryos. Deletion by transposon excision of the TransposonL2 and TransposonRl half sides is identified by loss of the intervening marker gene. The remaining TransposonLl half-side, with the downstream marker gene and genes-of- interest, is identified by the single marker gene phenotype and verified by sequencing of amplified DNA. This remaining TransposonLl half side, marker gene and genes-of-interest should be incapable of re-mobilization by transposase in the absence of the requisite TransposonRl half side.

The second embodiment disclosed has been termed "conditional excision-competent transformation vectors" (Fig. 4). This embodiment comprises a modified excision-competent transformation vector that contains a transposonR2 half-side in an inverted orientation, relative to the Rl half side, with R2 also flanked by recombinase target sites in inverted orientation. In this configuration, only the TransposonLl and Rl half-sides can integrate by transposition, and remobilization of the TransposonLl and R2 half-sides can only occur after a recombinase- mediated inversion between the recombinase target sites. This modification will facilitate the stabilization process, by transposon LI and R2 half-side deletion, for those excision-competent transformation vectors and/or host species where the primary transposition is highly favored or limited to the internal TransposonLl and R2 half-sides if R2 was in a normal orientation.

A similar result is achieved by the third embodiment which has been termed "RMCE with subsequent transposon deletion" (Fig. 5). Completely new in this embodiment is a DNA targeting strategy. The ultimate germ-line transformation process is conducted as a recombinase-mediated process, instead of a transposase-mediated process, into an existing (and pre-defined) genomic target site. This involves the RMCE principle, i.e. a site-specific recombinase recognizes heterospecific DNA target sites and exchanges DNA-cassettes between a RMCE-acceptor and a RMCE-donor (step 1 in Fig. 5). The success of this cassette exchange is indicated by the exchange of the acceptor target marker gene (e.g. ECFP, see Fig. 7) by the donor vector marker gene (e.g. EYFP, see Fig. 7). It is important to stress that only the coding region of the transformation marker genes is exchanged, not the promotor regions (which are not present in the RMCE-donor plasmid). The advantage of this promoter-free exchange is that side-reactions, which involve non-targeted integration of the donor into the genome, will not be recognized. Most important to this first step of cassette exchange, is a "homing DNA sequence" that is present in both the RMCE-acceptor and the RMCE-donor and is identical in both functional parts. The homing DNA sequence functions to significantly enhance the cassette exchange efficiency. The principle of stably integrating a gene-of-interest via a RMCE strategy into the genome of an invertebrate organism is completely novel and extends previously described RMCE-technology (see European Patent No. EP 0 939 120 and Schlake, T. & Bode, J. (1994). Use of mutated FLP recognition target (FRT) sites for the exchange of expression cassettes at defined chromosomal loci. Biochemistry 33, 12746-12751, and Seibler, J., Schubeler, D., Fiering, S, Groudine, M. & Bode, J. (1998). DNA cassette exchange in ES cells mediated by Flp recombinase: an efficient strategy for repeated modification of tagged loci by marker-free constructs. Biochemistry 37, 6229-6234, and European Patent No. EP 0 939 120, the contents of which are incorporated herein by reference) to invertebrate organisms. Because the RMCE-acceptor also carries a transposon half-side (Transposon Rl in Fig. 5), a fully remobilizable internal transposon is reconstituted after a successful RMCE reaction. This reconstituted transposon is subsequently physically deleted from the organism's genome by the action of a transposase (step 2 in Fig. 5 and Fig. 7) exactly as described for the first embodiment. In conclusion, the gene-of-interest is only flanked by one transposon half side end and hence is immobilized because it does not provide a complete substrate for transposase-mediated mobilization.

Brief Description of the Figures

For a fuller understanding of the nature and objects of the present invention, reference should be made by the following detailed description taken with the accompanying figures, in which: Fig. 1 shows a protocol for integration and re-mobilization for stabilized vector creation; Fig. 2 shows a diagram of stabilization vector pBac (LI -PUbDsRedl-L2-3xP3-ECFP-Rl } ; Fig. 3 shows a PCR analysis and verification of pBac{Ll-PUbDsRedl-L2-3xP3-ECFP-Rl} vector integration in line F34 and L2-3xP3-ECFP-Rl remobilization in line F34-1M; Fig. 4 shows the principle of "conditional excision competent transformation vectors"; Fig. 5 shows the principle of "RMCE with subsequent transposon deletion"; Fig. 6 shows an embodiment of the principle as shown in Fig. 4

Fig. 7 shows an embodiment of the principle as shown in Fig. 5: Stabilized vector creation by

RMCE ;

Fig. 8 shows a diagram of RMCE acceptor vector;

Fig. 9 shows molecular analysis of RMCE acceptor and RMCE donor transgenic lines and

PCR analysis of transgene mobilization;

Fig. 10 shows a diagram of a final RMCE donor vector for transgene stabilization;

Fig. 11 shows the approximate sequence of the vector shown in Fig. 2;

Fig. 12 shows the approximate sequence of the vector shown in Fig. 8; and

Fig. 13 shows the approximate sequence of the vector shown in Fig. 10.

Detailed Description of the Invention

Embodiment 1 : Excision-competent stabilization vectors

The experimental steps for the method are described in Figure 1 , and the structure of the excision competent transformation vector, pBac{Ll-PUbDsRedl-L2-3xP3-ECFP-Rl }, is described in Figure 2. Integration and re-mobilization of the vector was verified by PCR and sequence analysis described in Figure 3. pBac{Ll-PUbDsRedl-L2-3xP3-ECFP-Rl} was constructed based on the transposable element "piggyBac" (see United States Patent No. US 6,218,185, the contents of which are incorporated herein by reference). Conventional piggyBac-based transformation vectors (see WO 01/14537 and WO 01/12667, the contents of which are incorporated herein by reference) typically contain piggyBαc- aλi sides or parts thereof, including 5' piggyBac terminal sequences (refeπed to as piggyBac ) and 3' piggyBac terminal sequences (refeπed to as piggyBacK), which flank a transformation marker gene and a cloning site to insert the genes-of-interest. (see Handler, A.M., 2001. A cuπent perspective on insect gene transfer, insect Biochem. Mol. Biol., 31, 111-128, the contents of which are incorporated herein by reference.) For vectors that are not autonomously transpositionally active, the transposase gene is partially deleted or interrupted by marker genes or genes-of- interest, thereby mutating the transposase. Non-autonomous vectors require an independent source of functional transposase for mobilization resulting in transposition. In contrast to conventional vectors, pBac{Ll-PUbDsRedl-L2-3xP3-ECFP-Rl} is provided with an additional piggyBacL half side (refeπed to as L2 half side) that is in the same orientation as the LI half side, and positioned internal to the piggyBac LI and Rl half sides. In this orientation, transposition can occur utilizing the LI and Rl half sides, or the internal L2 and Rl half sides. In addition, pBac{Ll-PUbDsRedl-L2-3xP3-ECFP-Rl} contains a unique Kasl restriction endonuclease site in the piggyBacTλ region that can be used to insert genes of interest. In order to follow the primary transformation integration event of the LI and Rl half-sides and to distinguish it from integration of L2 and Rl half-sides, independent transformation marker genes are placed in-between the two half-side pairs. In pBac {LI -PUbDsRedl -L2-3xP3 -ECFP- Rl}, the PUbDsRedl (see WO 01/14537, the contents of which are incorporated herein by reference) marker is placed in-between the LI and L2 half sides, and the 3xP3-ECFP (see WO 01/12667, the contents of which are incorporated herein by reference) marker is placed in- between the L2 and Rl half sides. pBac(Ll-PUbDsRedl-L2-3xP3-ECFP-Rl):

A 3.7-kb AflTH-Aflll fragment from pB[PUbDsRedl], containing 0.7kb of piggyBac LI half- side DNA and adjacent 5' insertion site DNA and the polyubiquitin:DsRedl DNA gene, was blunted by Klenow-mediated nucleotide fill-in reaction and isolated by agarose gel purification. The blunted fragment was ligated into the Mscl site of pXL-BacII-3xP3-ECFP. Plasmids having the 3xP3-ECFP and polyubiquitin:DsRedl reading frames in opposite orientation were selected. phspBac transposase helper plasmid:

For germline transformation experiments, the helper phspBac was (see PCT WO 01/14537, the contents of which are incorporated herein by reference). Experimental steps of the transgene immobilization process: a) Germ-line transformation with pBac{Ll-PUbDsRedl-L2-3xP3-ECFP-Rl}

The ρBac{Ll-PUbDsRedl-L2-3xP3-ECFP-Rl} vector was integrated into the Drosophila genome of the white eye w[m] strain by coinjection with the phspBac helper plasmid into pre- blastoderm embryos. Using conventional piggyBac-mediated germ-line transformation methods (see United States Patent No. US 6,218,185 and WO 01/14537, the contents of which are incorporated herein by reference), seven putative Gl transformant lines expressing only the 3xP3-ECFP marker were observed and discarded. One Gl male fly exhibited both thoracic expression of DsRed and eye expression of ECFP, and it was backcrossed to w[m] females to create a line designated as F34. Transformation by an intact pBac{Ll-PUbDsRedl-L2-3xP3- ECFP-R1} vector by /?tggy5αc-mediated transformation in F34 was confirmed by sequencing of internal PCR products and inverse PCR products, derived from F34 genomic DNA, which provided the insertion site DNA sequence (see below). b) piggyBac transposase-induced excision of piggyBacUl and piggyBacRl Transformed individuals identified and confirmed to have the marker genes 3xP3-ECFP and PUbDsRedl were backcrossed to w[m] flies for two generations. The presence of both markers solely in female progeny from F34 parental males indicated X-chromosome sex- linkage for the primary integration. F34 flies were mated as transgene heterozygotes to a piggyBac jumpstarter strain (w+/Y;pBac/pBac;+/+) having a homozygous P-element-mediated integration of an hsp70-τegu\ated piggyBac transposase gene into chromosome 2 and marked with the wild type white+ allele. Larval and pupal offspring of these matings were heat shocked at 37°C for 60 minutes every second day until adult emergence to promote transposase gene expression. Male and female progeny of these matings were screened, with those carrying the transposase gene (red eye pigmentation) and expressing the fluorescent protein markers, PUb-DsRedl and 3xP3-ECFP, being mated to w[m] individuals. Ten matings of 4 to 5 appropriately marked females to w[m] males and 18 matings of 2 to 3 marked males to w[m] females were set up. Progeny from these matings were screened for expression of PUb-DsRedl and the absence of 3xP3-ECFP, which would indicate loss by remobilization of the piggyBacL! and piggyBacRl half sides with the intervening 3xP3-ECFP marker DNA. Progeny expressing only DsRed 1 fluorescence were detected at an approximate frequency of 2% of all flies screened. A single white eye male (lacking the transposase gene) and expressing DsRedl, and not ECFP, was outcrossed to w[m] females with the resultant line designated as F34-1M. c) molecular analysis of the vector integration before and after remobilization The pBac{Ll-PUbDsRedl-L2-3xP3-ECFP-Rl} integration into the F34 Drosophila genome was initially identified by phenotypic expression of the DsRed and ECFP marker genes and verified by PCR amplication of transformant DNA using primers internal to the vector sequence (see Fig. 3). Genomic insertion site DNA flanking the integration was obtained by inverse PCR of the piggyBac l 5'-end half side using the 122R and 139F primers, in outward orientation, to F34 genomic DNA digested with Mspl endonuclease and circularized by ligation. The 5 'end insertion site sequence was compared by BLAST analysis to the Drosophila Genome Sequence Database, and consistent with segregation analysis, was found to be homologous to sequence found on the X-chromosome at locus 9B4. The database sequence was used to derive the piggyBacRl 3 '-end insertion site, and the 197F and 196R PCR primers were created to genomic insertion site DNA at the 5' and 3 '-end flanking sequences, respectively. The genomic primers were then used to amplify and sequence DNA that spans the pBac {LI -PUbDsRedl -L2-3xP3-ECFP-Rl } integration in F34, to further verify it as a primary intact piggyBac vector integration. The 197F and 196R primers were then used for PCR of F34-1M genomic DNA, which confirmed remobilization of the L2-PUbDsRedl-Rl internal vector DNA in F34. Further verification of the vector integration and subsequent remobilization was achieved by sequencing of PCR products obtained with primers 196 and 197 in combination with primers to internal vector DNA described in Figure 3. In all cases, positive PCR results yielded sequences consistent with a primary integration of pBac{Ll- PUbDsRedl-L2-3xP3-ECFP-Rl} in F34, and remobilization of the L2-PUbDsRedl-Rl sequence in F34-1M flies. PCR products were not obtained in F34-1M flies using primers to the L2-PUbDsRedl-Rl sequence consistent with its deletion from the genomic DNA after remobilization.

Embodiment 2: Conditional excision-competent transformation vectors The structure of the conditional excision-competent transformation vector, pBac_STBL, as well as the experimental steps are depicted schematically in Figures 4 and 6. pBac_STBL is based on the transposable element "piggyBac" (see United States Patent No. US 6,218,185, the contents of which are incorporated herein by reference) and is a modified version of pBac{Ll- PUbDsRedl-L2-3xP3-ECFP-Rl}. In pBac-STBL the internal transposon half-side (R2) is a duplication of the piggyBac 3 '-end, and it is in reverse, or opposite, orientation to Rl. In addition, it is flanked in upstream and downstream positions by FRT (FLP recombinase target) sites in opposite directions that create an inversion by recombination in the presence of FLP recombinase (see Figs. 4 and 6). Therefore, in this vector, only the piggyBacLl and Rl half sides and intervening DNA can integrate, but re-mobilization of piggyBacRl together with piggyBacLl or piggyBacRl should not be possible. Mobilization of piggyBacRl and LI is only possible after FRT recombination.

In addition, pBac STBL contains unique cloning sites for the rare octamer-specific restriction enzymes Ascl and Fsel. pBac_STBL is equipped with two separable transformation marker genes (see WO 01/12667, the contents of which are incorporated herein by reference), which are located upstream of the Ascl/Fsel cloning sites (3xP3-EYFP; Fig. 6) and downstream of the FRT-sites (3xP3-DsRed; Fig. 6), respectively. In the following, the details of pBac STBL plasmid construction starting from plasmid vectors already published are disclosed: _PSL-3xP3-DsRedaf:

A 0.8 kb Sall-Notl fragment from pDsRedl-1 (Clontech, Palo Alto, CA) is cloned into the plasmid pSL-3xP3-EGFPaf (see WO 01/12667, the contents of which are incorporated herein by reference) previously digested with Sall-Notl. Thereby, the EGFP (0.7 kb) open reading frame was replaced by the DsRed (0.8 kb) open reading frame. pSLfaFRTfa:

The FRT sequence (90 bp) is prepared by Satl-Aspl 8 restriction of pSL>AB> and cloned into the plasmid pSLfall80fa previously digested with Xhol-Aspll . The FRT sequence coπesponds to the substrate of the FLP recombinase:

TTGAAGTTCCTATTCCGAAGTTCCTATTCTCTAGAAAGTATAGGAACTTCAGAG CGCTTTTGAAGCT pSL-3xP3-DsRed-FRT:

A 1.0 kb EcoRl-BsiΨl fragment from pSL-3xP3-DsRedaf (containing the DsRed-ORF under 3xP3 promoter control) is cloned into pSLfaFRTfa previously digested with EcoRl-AspHS. pSL-3xP3-DsRed-FRT-FRT:

The PCR amplification product of the FRT sequence (template: pSL>AB>; Primers: CH_FRT_F 5'-GAGCTTAAGGGTACCCGGGGATCTTG-3' and CH FRT t 5'-GACTAGTCGATATCTAGGGCCGCCTAGCTTC-3') is digested with Bfrl-Spel and cloned into pSL-3xP3-DsRed-FRT previously digested with Bfrl-Spel. Both FRT sequences are oriented in opposite directions. pSL-3xP3-DsRed-FRT-pBacR2-FRT:

The piggyBac 3' sequence (refeπed to as: piggyBacRl) is prepared as a 1.3 kb Hpal-EcoRV fragment from the plasmid p3E1.2 (see United States Patent No. US 6,218,185, the contents of which are incorporated herein by reference) and cloned into the plasmid pSL-3xP3-DsRed- FRT-FRT previously cut with EcoRV. The piggyBacRl insertion with an orientation opposite to the DsRed-ORF is chosen (the EcoRV cloning site is restored at the 5'end of the insertion). pBac STBL:

A 2.7 kb EcoRl-Bfrl fragment (both restriction sites filled in by Klenow reaction) from pSL- 3xP3-DsRed-FRT-pBacR2-FRT is cloned into pBac-3xP3-ΕYFPaf (see WO 01/12667, the contents of which are incorporated herein by reference) previously cut with BgHl (Klenow fill- in reaction). The insertion with an opposite orientation of the DsRed- and EYFP-ORFs is chosen. This final plasmid contains piggyBacRl in opposite orientation to piggyBacRl (Fig. 6). phspBac transposase helper plasmid:

For germline transformation experiments, the helper phspBac is used (see PCT WO 01/14537, the contents of which are incorporated herein by reference). Experimental steps of the transgene immobilization process (Fig. 4 and Fig. 6) a) Germline transformation of pBac STBL (step 1 in Fig. 4 and Fig. 6)

DNA-sequences included in the plasmid pBac_STBL within the ends of piggyBacLl and piggyBacRl are integrated into the Drosophila genome by piggyBac-mediated germline transformation (see United States Patent No. US 6,218,185 and WO 01/14537, the contents of which are incorporated herein by reference). Similar constructs incorporating genes-of- interests inserted at the unique cloning sites would be treated in the same way. b) FLP recombinase induced inversion (step 2 in Fig. 4 and Fig. 6)

Genomic integrations of the pBac_STBL transgene are identifiable by both EYFP and DsRed eye fluorescence (see WO 01/12667, the contents of which are incorporated herein by reference). Following the identification of transgenic founder individuals (and to establish Drosophila strains carrying the transgene in the homozygous state), the inversion of the piggyBacRl sequence is carried out. This is performed by crossing in the strain beta2t-E P that expresses FLP-recombinase during spermatogenesis. Alternatives of step 2 in Fig. 6 include crossing in hsp70-FLP and bsRXR-strains, respectively, or microinjection of a FLP- recombinase encoding plasmid, e.g. pKhsp82-FLP (into preblastoderm embryos of homozygous transgenic pBac_STBL lines). Though the inversion event cannot be detected by the marker genes included into pBac STBL, a statistical equilibrium of original and inverted orientation of the piggyBacRl sequence can be assumed. Thus, the inversion process is detected by testing several independent sublines by sequencing of vector PCR products to identify sublines having undergone piggyBacRl inversion. c pissvBac transposase induced deletion (step 3 in Fig.4 and Fig. 6)

Strains with inverted piggyBacRl sequence are crossed to piggyBac transposase expressing strains (refeπed to as jumpstarter). Different lines of the Drosophila strain Her{3xP3-ECFP, alphaltub-p/ggvRαcKlO} are available for this step. Progeny from this cross expressing both EYFP/DsReD (indicating the presence of pBac_STBL) and ECFP (indicating the presence of the jumpstarter) are crossed out in single male setups. d) Identification of immobilized transgene DNA

ECFP^" progeny (selection against the jumpstarter) of single male crossings are analyzed for both the presence of EYFP fluorescence and the absence of DsRed fluorescence. Individuals putatively containing a transposon deletion event should show EYFP but absence of DsRed fluorescence and can be analyzed further. By inverse PCR, the transposon deletion can be molecularly confirmed and stability of the potentially immobilized transgene insertion can be assessed by challenging the transgene insertion with piggyBac transposase. Embodiment 3: RMCE with subsequent transposon deletion

The RMCE-acceptor plasmid, pBac{3xP3-FRT-ECFP-linotte-FRT3} (Fig. 8), is & piggyBac- based transformation vector that was provided additionally with a DNA exchange cassette.

This cassette consists of two heterospecific FRT sites (refeπed to as FRT and FRT3 equivalent to F and F3 (published in European Patent No. EP 0 939 120 Al, the contents of which are incorporated herein by reference)) in parallel orientation.

European Patent No. EP 0 939 120 Al (see page 2, line 50 to page 3, line 6) teaches the technology of the RMCE reaction:

"Recombinases such as FLP and Cre have emerged as powerful tools to manipulate the eucaryotic genome (Kilby, N.J., Snaith, M.R., Murray, J.A.H. (1993). Site-specific recombinases: tools for genome engineering. Trends Genet. 9, 413-421, and Sauer B. (1994). Site-specific recombination: developments and applications. Cuπ. Opin. Biotechnol. 5, 521-527, the contents of which are incorporated by reference herein). These enzymes mediate a recombination between two copies of their target sequence and have mainly been used for deletions. We show here that FLP-RMCE can be applied to introduce secondary mutations at a locus which has been previously tagged by a positive/negative selectable marker, and that these secondary mutations can be produced without depending on a selectable marker on the incoming DNA. FLP-RMCE utilizes a set of two 48 bp FLP target sites, in this case wild type (F) and F3, a mutant that was derived from a systematic mutagenesis of the 8 bp spacer localized between the FLP binding elements (see Schlake T., Bode, J. (1994). Use of mutated FLP recognition target (FRT) sites for the exchange of expression cassettes at defined chromosomal loci. Biochemistry 33, 12746- 12751, the contents of which are incorporated by reference herein). FLP effects recombination between the F3/F3 couple which is as efficient as between the wild type sites (F/F) but it does not catalyze recombination between a F/F3 pair (Seibler J., Bode J. (1997). Double-reciprocal crossover mediated by FLP- recombinase: a concept and an assay. Biochemistry 36, 1740-1747, the contents of which are incorporated by reference herein). Thereby FLP-RMCE enables the specific exchange of an expression cassette in the genome which is flanked by a F3-site on one end and a F-site on the other for an analogous cassette comprising virtually any sequence which is provided on a plasmid in a single step without the need of introducing a positive selectable marker. Nothing else in the genome is altered and no plasmid sequences are inserted. In contrast to approaches using a single recombination site the targeting product is stable even under the permanent influence of the recombinase unless it is exposed to an exchange plasmid (Seibler J., Bode J. (1997). Double-reciprocal crossover mediated by FLP-recombinase: a concept and an assay. Biochemistry 36, 1740-

1747, the contents of which are incorporated by reference herein). The system can be used to analyze the function of either a gene product or of regulatory sequences in ES-cells or of the derived transgenic mice." (citations added)

In the present invention, FRT and FRT3 flank the ECFP open reading frame and a "homing sequence". As a "homing sequence", the 1.6 kb H dIII fragment of the Drosophila linotte locus was chosen (see Taillebourg, E. & Dura, J.M. (1999). A novel mechanism for P element homing in Drosophila. Proc. Natl. Acad. Sci. USA 96, 6856-6861, the contents of which are incorporated herein by reference. This particular sequence has been described to act as "bait" for homing of identical/homologous DNA sequences by a process called "para-homologous pairing". We have shown previously that the positioning of the FRT site between the 3xP3 promoter and the start codon of the ECFP open reading frame does not interfere with expression of the 3xP3-ECFP gene (see PCT WO 01/12667, the contents of which are incorporated herein by reference). The RMCE donor plasmid, pSL-FRT-EYFP-pBacR2-3xP3-

DsRed-linotte-FRT3 (Fig. 10), contains the DNA cassette to be recombined in. The donor cassette comprises the two heterospecific FRT sites (FRT and FRT3) flanking the EYFP open reading frame (promoter- free), a piggyBacRl 3 '-half side sequence, the transformation marker gene 3xP3-DsRed and the homing sequence from the linotte locus (identical to the linotte sequence in the RMCE acceptor). The RMCE donor plasmid is a derivative of the plasmid pSLfal l80fa (see Patent Cooperation Treaty PCT WO 01/12667 Al), which does not contain any transposon sequences. Asc lFsel cloning sites have been incorporated to ease the insertion of gene(s)-of-interest upstream of the piggyBacRl sequence.

In the following, the details of the RMCE plasmids construction starting from plasmid already published are disclosed:

Construction of the RMCE acceptor plasmid (Fig. 8): pSL-3xP3-FRT-ECFPaf:

A 90 bp Sall-Asp7 IS fragment from the plasmid pSL>AB> containing the FRT sequence was cloned into the plasmid pSL-3xP3-ECFPaf (see Patent Cooperation Treaty PCT WO 01/12667, the contents of which are incorporated herein by reference) previously digested with Satl- AspHS. The FRT sequence coπesponds to the substrate of the FLP recombinase: TTGAAGTTCCTATTCCGAAGTTCCTATTCTCTAGAAAGTATAGGAACTTCAGAG

CGCTTTTGAAGCT pBac(3xP3-FRT-ECFPafl :

A 1.3 kb EcoRI-(blunted by Klenow fill in reaction)-Nr«I fragment from the plasmid pSL-

3xP3-FRT-ΕCFPaf was cloned into the plasmid p3E1.2 previously digested with Hpal. pBac(3xP3-FRT-ECFP-linotte-FRT3L final RMCE acceptor plasmid:

The plasmid pBac{3xP3-FRT-ECFPaf} was digested with Ascl-BgHl, and the following sequences were cloned into the linearized vector: i.) the Ascl-Aspl IS cut PCR amplification product of the 1.6 kb H dIII genomic linotte fragment. As a template, genomic DΝA of Drosophila melanogaster, strain OregonR, was chosen and as primers:

CΗ ioFwd (5'-TTGGCGCGCCAAAAGCTTCTGTCTCTCTTTCTG-3') and

CΗ_lioRev (5'-CGGGGTACCCCAAGCTTATTAGAGTAGTATTCTTC-3') and ii.) the AspllS-BgϊTl cut PCR amplifcation product of the FRT3 sequence (mutagenic PCR).

As a template, the plasmid pSL>AB> was chosen and as primers:

CH_F3Fwd (5'-TTGGCGCGCCAAGGGGTACCCGGGGATCTTG-3') und

CH_F3Rev (5'-

CCGCTCGAGCGGAAGATCTGAAGTTCCTATACTATTTGAAGAATAG-3').

The FRT3 sequence coπesponds to the F3 sequence (European Patent No. EP 0 939 120 Al):

TTGAAGTTCCTATTCCGAAGTTCCTATTCTtcAaAtAGTATAGGAACTTCAGAGCG

C

The diagram of this final RMCE acceptor vector is shown in Fig. 8.

Construction of the RMCE donor plasmid (Fig. 10) pSL-3xP3-FRT-EYFPaf:

Construction was analogous to pSL-3xP3-FRT-ECFPaf, but into the plasmid pSL-3xP3-

EYFPaf (see WO 01/12667, the contents of which are incorporated herein by reference). pSL-FRT-EYFPaf:

The 3xP3 promoter sequence was deleted from the plasmid pSL-3xP3-FRT-EYFPaf by digestion with EcoRI-RαwHI, filling-in by Klenow enzyme reaction and finally religating the blunted plasmid. pSL-FRT-EYFP-linotte-FRT3 :

A 1.7 kb Ascl-BgHl (both sites blunted by Klenow fill-in reaction) fragment from pBac{3xP3-

FRT-ECFP-linotte-FRT3} was cloned into the plasmid pSL-FRT-EYFPaf previously digested with Nrul. The orientation with maximal distance of the FRT and FRT3 sites was chosen. pBac{3xP3-DsRedafl:

A 1.2 kb EcoRI (site blunted by Klenow fill-in reaction)-NrwI fragment from the plasmid pSL-

3xP3-DsRedaf was cloned into the plasmid p3Ε1.2 (see United States Patent No. US

6,218,185, the contents of which are herein incorporated by reference) previously digested with Bglll-(site blunted by Klenow fill-in reaction)-HpαI. pSL-FRT-EYFP-linotte-FRT3-3xP3-DsRed:

A 1.25 kb EcoRI-(site blunted by Klenow fill-in reaction)- Nrwl fragment from pSL-3xP3-

DsRedaf was cloned into the plasmid pSL-FRT-ΕYFP-linotte-FRT3 previously digested with

Spel (site blunted by Klenow fill-in reaction). pSL-FRT-EYFP-pBacR-3xP3-DsRed-linotte-FRT3. final RMCE donor plasmid:

A 2.5 kb scl-(site blunted by Klenow fill-in reaction)- EcoRN fragment from pBac{3xP3-

DsRedaf} was cloned into the plasmid pSL-FRT-ΕYFP-linotte-FRT3 previously cut with

EcoRI (site blunted by Klenow fill-in reaction).

The diagram of this final RMCΕ acceptor vector is shown in Fig. 10.

FLP recombinase plasmid source: p¥ihsp82-FLP:

A 1.1 kb Asp718-Xbal (sites blunted by Klenow fill-in reaction) fragment from the plasmid pFL124 containing the FLP recombinase ORF and the 3' transcriptional terminator from the adh gene was cloned into the plasmid pKAsp<°2) previously cut with BamKl (site blunted by

Klenow fill-in reaction). phspBac transposase helper plasmid:

For germ-line transformation experiments, the helper phspBac was used (see PCT WO

01/14537 Al).

DΝA cassette exchange by RMCΕ is highly efficient in Drosophila melanogaster

Practical application of RMCΕ-based gene targeting and germline transformation (e.g. for the purpose of immobilizing transgenes) will depend strongly on the efficiency of the DΝA cassette exchange. This efficiency should be in the range observed with conventional transposon-mediated germline transformation systems that allow the isolation of several transgenic founder individuals among 1,000 - 10,000 progeny screened. Previous experiments involving DΝA cassette exchange have been performed only using cell culture and stringent selection conditions. Therefore the efficiency of such a system in an invertebrate organism such as Drosophila is hard to predict. Hence, a pilot experiment was performed. An intermediate of the RMCE donor plasmid, pSL-FRT-EYFP-linotte-FRT3 and the FLP recombinase expression vector pKhsp82-FLP were co-injected into pre-blastoderm embryos of a Drosophila melanogaster acceptor strain. These embryos cany the RMCE acceptor transgene vector (Fig. 8) integrated by piggyB ac-mediated germ-line transformation, in a homozygous state. The final concentration of the plasmids in the injection mix was 500 ng/μl (RMCE donor plasmid) and 300 ng/μl (pKhsp82-FLP). Altogether, around 3,000 Drosophila embryos were injected, coπesponding to ten times the number necessary for a conventional piggyB ac-mediated germ-line transformation. Successful exchange of the acceptor by the donor cassette was indicated by the change in the eye fluorescence from ECFP to EYFP (in FI individuals). Results documenting the frequency of such exchange events are given in Table 1 :

Acceptor Injected Male Injection F Feerrttiille Male Vials with EYFP-pos.

Line Embryos Survivors Inj. ϊ /. and ECFP-neg. progeny

M4.II ECFP 750 121 70 22

M7.III ECFP 750 138 72 17

M8.II ECFP 600 68 54 12

M9.II ECFP 750 123 109 27

Table 1 : Results of the RMCE experiment in Drosophila with the donor plasmid pSL-FRT- EYFP-linotte-FRT3. Acceptor lines (II: second, III: third chromosomal homozygous, ECFP fluorescence) used for microinjection, number of injected embryos, male and fertile male injection survivors and the number of vials containing EYFP-positive progeny are given. EYFP-positive founder males resulting from targeting events were bred to homozygosity and established as stocks (refeπed to as "M4.II EYFP", "M7.III EYFP", "M8.II EYFP" and "M9.II EYFP", respectively). Segregation analysis (genetic mapping of transgene integrations) indicated for all four lines that the chromosomal localization of the donor and acceptor transgene is identical.

We define the DNA cassette exchange frequency as a percentage of fertile Fj vials producing EYFP-positive progeny. With this definition, the frequency of RMCE events is 25% on average coπesponding well to the germ-line transformation frequency usually observed with piggyBac, Hermes or os-based vectors mDrosophila). This experiment demonstrates that, with the particular design of RMCE-vectors, the process of cassette exchange is highly efficient in an invertebrate organism such as Drosophila. Molecular characterization of RMCE events and integration site analysis a) Genomic integration site of donor and acceptor transgenes

The exchange of eye fluorescence from ECFP to EYFP suggests that the donor cassette (carrying the promotor-free eyfp gene) integrated at the locus of the acceptor transgene (providing the 3xP3 promoter). Therefore, the genomic integration sites of the acceptor transgene in the acceptor line and of the donor transgene in the coπesponding donor line should be identical. To identify genomic integration sites, inverse PCR experiments were carried out for acceptor and donor Drosophila lines. To recover DNA sequences flanking piggyBac insertions, inverse PCR was performed. The purified fragments were directly sequenced for the 5' junction with primer CH_PLSeq 5'-CGGCGACTGAGATGTCC-3'. The obtained sequences were used in BLAST searches against the Drosophila Genome Sequence Database. For the 5' junction, genomic DNA sequence identity could be confirmed for three acceptor/donor pairs (Table 2).

Acceptor line Location of insert Identical for coπesponding donor line?

Chromosome armgenomic scaffold position

M4.II ECFP 2L AE003662.3 204692 yes (M4.II EYFP)

M7.III ECFP 3L AE003558.3 171057 yes (M7.III EYFP)

M8.II ECFP 2L AE003618.2 15414 yes (M8.II EYFP)

M9.II ECFP 2L AE003662.3 15805 nd.

Table 2: Genomic integration sites of the acceptor transgene pBac{3xP3-FRT-ECFP-linotte-

FRT3} in four Drosophila lines used for RMCE targeting. Sequence numbers and nucleotide positions refer to the Release 3 sequence of the Drosophila Genome Sequence Database. For three coπesponding RMCE donor lines, integration sites could be confirmed to be identical. nd.: not determined

Interestingly, the acceptor line M9.II ECFP was found to cany the acceptor transgene integrated at the Drosop α-endogenous linotte locus (integration position coπesponds to bp

1185). This suggests that "para-homologous pairing" of the linotte sequences included in the acceptor plasmid to the homologous genomic sequence occuπed, further verifying the homing phenomenon. b) Southern Analysis To further verify at the molecular level that the donor transgene targeted the acceptor locus via an RMCE mechanism, Southern analysis on genomic DNA of the four acceptor and the four donor lines was performed. Rstl was chosen as an indicative restriction digest and a probe hybridizing to gfp-hased transformation marker genes (hybridizing to both ECFP and EYFP) was selected (Fig. 9). Only one strong hybridization signal was present in all acceptor lines which is consistent with a single integration of the acceptor transgene. The expected pattern of DNA-DNA hybridization, 2.4 kb for the acceptor transgene and 1.6 kb for the donor transgene, was detected for all four lines for each transgene (Fig. 9). Additionally, a ~6 kb hybridization signal was detected only in RMCE donor lines. As this signal might indicate the presence of the complete donor vector, further Southern experiments (using probes against the pUC plasmid backbone sequences) were carried out. The presence of pUC sequence in the donor lines could be confirmed (data not shown) pointing toward an integration of the entire donor vector in the four donor lines analyzed.

In summary, three lines of evidence let us infer that targeting of the RMCE acceptor locus by the RMCE donor vector took place: i) the exchange in eye color fluorescence from ECFP (acceptor) to EYFP (donor), ii) the identity of genomic DNA sequence flanking the piggyBac transgene integration in corresponding acceptor and donor lines, and iii) DNA hybridization signals in accordance with expectations for the exchange of the ecfp to the eyfp open reading frame.

Recombination occurs by cassette exchange via FRT and FRT3

The recombinase-mediated cassette exchange mechanism requires a double recombination event (see European Patent No. EP 0 939 120, the contents of which are incorporated herein by reference). Because the Southern analysis suggests that in the pilot RMCE experiments single recombination events caused integration of the entire donor plasmid, we analyzed in more detail whether the RMCE mechanism, which has not been established for an invertebrate organism, can occur in Drosophila. To this end, we modified the donor construct to include a 3xP3-DsRed marker gene downstream to the FRT3 sequence (pSL-FRT-EYFP-linotte-FRT3- 3xP3-DsRed). This vector configuration allows the separation of RMCE events: 1) double cross-over via FRT and FRT3 sites resulting in ECFP to EYFP eye fluorescence exchange

2) single recombination events (via FRT site) resulting in ECFP to EYFP and DsRed eye fluorescence exchange

3) single recombination events (via FRT3 site) resulting in ECFP to DsRed (and ECFP) eye fluorescence exchange

For the targeting experiment, the acceptor line M4.II ECFP (Tablel) was selected for further testing. FI individuals with ECFP to EYFP exchange indicating targeting were observed at a frequency of 13.1 %:

Embryos injected: 750 single GO male founders: 109

Fertile GO male founders: 84

Setups producing EYFP-fluorescing FI progeny: 11

The eleven setups yielding EYFP-fluorescing individuals were analyzed for the occuπence of double and single recombination events (Table 3).

Setup# Phenotype of individual flies double recombination single FRT rec. single FRT3 rec.

EYFP⁺ EYFP⁺, DsRed⁺ DsRed⁺, ECFP⁺ established Stocks

(DsRed^", ECFP^") (ECFP ) (EYFP^")

1 1 1 0

2 1 0 0 Rl

3 2 6 0

4 4 0 0 R2

5 3 0 0 R3

6 3 0 0 R4

7 1 10 0 R5 (EYFP⁺,DsRed⁺)

8 13 26 0 R6 (EYFP⁺,DsRed⁺)

9 1 2 0

10 11 3 0

11 1 0 0

Table 3: Phenotypic analysis of FI progeny from GO male founders of the acceptor line M4.II ECFP injected with the donor pSL-FRT-EYFP-linotte-FRT3-3xP3-DsRed. Double and single recombination events are indicated by differential analysis of eye fluorescence for ECFP, EYFP and DsRed.

Five out of eleven setups produced progeny showing EYFP but lacking DsRed (and ECFP) fluorescence. This phenotype is consistent with targeting via double recombination with only sequences between FRT and FRT3 being exchanged. However, single recombination events via FRT were also observed, in contrast to no single recombinations via FRT3. The results indicate that recombinase mediated cassette exchange is mechanistically feasible in an invertebrate organism (the vinegar fly Drosophila melanogaster) and, by applying a simple eye fluorescence marker scheme, double recombination events can be selected for. Experimental steps of the transgene immobilization process (Fig. 5 and Fig. 7) The previous results demonstrate that recombinase mediated targeting of genomic DNA loci is possible in an invertebrate organism like Drosophila. As depicted in Fig. 5, the RMCE strategy can be further employed for the purpose of post-transformational transgene immobilization. The general procedure consists of two steps. In the first step, a transformation vector containing the gene of interest, a transposon half-side (TransposonR2 in Fig. 5) and an additional marker gene is used as the RMCE donor to target the RMCE acceptor line (i.e. RMCE acceptor vector (Fig.8) genomically integrated). By a single or double recombination event, an 'internal' piggyBac transposon comprising both half-sides (piggyBacLl and piggyBacR2 in Fig. 5) is reconstituted. In a second step, transposase activity is introduced to remobilize the 'internal' transposon by selecting for individuals lacking the additional marker gene as demonstrated in embodiment 1. In the following section we provide data that prove this principle: Stepl: Targeted DNA cassette exchange (RMCE. Step 1 in Fig. 5 and Fig. 7) The final donor plasmid, pSL-FRT-EYFP-pBacR2-3xP3-DsRed-linotte-FRT3 (Fig. 10, in the following refeπed to as "final RMCE donor") contains, in-between the FRT and FRT3 sites, a cassette with: (i) a promotor-free eyfp ORF, (ii) the piggyBacRl (3' end) transposon sequence, (iii) the transformation marker 3xP3-DsRed, and (iv) the homing sequence from the Drosophila linotte locus (see Taillebourg, E. & Dura, J.M. (1999). A novel mechanism for P element homing in Drosophila. Proc. Natl. Acad. Sci. USA 96, 6856-6861, the contents of which are incorporated herein by reference). Derivatives of the final RMCE donor vector carrying additional DNA sequences (genes-of-interest) can be constructed by insertion into the unique Ascl and Fsel cloning sites which are located upstream of the piggyBacRl transposon sequence (Fig. 10).

Microinjection of the final RMCE donor was carried out using the Drosophila acceptor line M4.II ECFP (Table 2). This line carries the acceptor transgene pBac{3xP3-FRT-ECFP- linotte-FRT3} in the homozygous state. Embryos were injected under the conditions described previously. Single GO founder males were crossed out and progeny (generation FI) were screened for the presence of both EYFP fluorescence and DsRed fluorescence (see Fig. 7). Targeting (i.e. individuals with ECFP to EYFP exchange) were observed at a frequency of 22.2 %.

Embryos injected: 750 single GO male founders: 178

Fertile GO male founders: 158

Setups producing EYFP and DsRed fluorescing FI progeny: 34

In total, 91 female and 62 male individuals were obtained which consistently showed an EYFP and DsRed eye fluorescence phenotype. Moreover, in these individuals ECFP fluorescence was absent as expected for recombination events. Though the exact mechanism (single versus double recombination) was not investigated for individuals from this targeting experiment, the previous pilot experiments suggest a significant fraction of double recombination events resulting from cassette exchange via FRT and FRT3 sites.

The results confirm a high efficiency of the gene targeting system disclosed in this embodiment, which is comparable to 'conventional' transposon-mediated germ-line transformation, at least for the vinegar fly Drosophila. In particular, the efficiency did not decrease significantly due to the interruption of the linotte sequence in the final donor plasmid or the increased size (2.6 kb compared to previous "pilot" donor vector) of the final donor plasmid (Fig. 10). This suggests that recombinants can also be generated with derivatives of the final donor plasmid caπying additional gene(s)-of-interest.

Step 2: piεsvBac transposase induced transposon deletion of a targeted vector (Step 2 in Fig. 5 and Fig. 7)

Successful re-mobilization of the reconstituted piggyBac transposon is indicated by loss of

DsRed fluorescence. Progeny lacking the sequence between piggyBacRl and piggyBacLl exclusively express EYFP fluorescence (see Fig. 7).

To examine whether the reconstituted internal piggyBac transposon vector can be re-mobilized by piggyBac transposase activity, individuals of generation FI with EYFP and DsRed eye fluorescence were crossed to the following /?tggy5αc-expressing jumpstarter lines:

(1) line Her{3xP3-ECFP; abib-piggyBac) M6.II, refeπed to as "HerM6"

(2) line Her{3xP3-ECFP; atub-piggyBac) M10.III, refeπed to as "HerMlO"

(3) line Mi{3xP3-DsRed; hspl 0-piggyBac} M5.II, refeπed to as "MiM5"

Progeny (generation F2) carrying both the final RMCE donor and the jumpstarter transgenes were crossed individually to non-transgenic Drosophila and progeny from these crosses (generation F3) were analyzed for the presence of individuals carrying EYFP but lacking DsRed eye fluorescence (Table 4). Js HerM6 HerMlO MiM5

Setup EYFP⁺ DsRed^" EYFP⁺ DsRed^" EYFP⁺ DsRed^"

1 73 0 62 0 38 32

2 67 0 57 1 42 0

3 47 0 48 0 56 1

4 53 0 52 0 68 3

5 36 0 34 0 48 5

6 61 1 48 1 37 0

7 50 1 55 0 38 1

8 40 0 52 0 71 5

9 39 0 55 0 41 0

10 86 0 43 0 72 2

11 53 0 40 0 49 0

12 57 0 71 0 30 0

13 17 0 52 0 46 0

14 58 1 66 0 48 1

15 65 2 56 0 41 0

16 54 2 55 0 54 0

17 55 2 51 0 53 0

18 54 0 43 1 66 2

19 63 1 18 0 56 1

20 78 0 38 1 63 1

Sum: 1106 10 996 4 1017 25

Table 4: Phenotypic analysis for piggyBac transposon remobilization events. Progeny from single crosses of males caπying both final RMCE donor and jumpstarter transgenes (Js) to non-transgenic Drosophila virgin females were analyzed for individuals showing EYFP eye fluorescence but lacking DsRed eye fluorescence. Such a phenotype is consistent with a deletion of the internally reconstituted piggyBac transposon (Fig.7).

Depending on the jumpstarter line employed, the frequency of remobilization ranged from

0.4% (HerMlO) to 2.5% (MiM5). This indicates that the reconstituted internal piggyBac transposon vector can be remobilized efficiently, and the combination of different fluorescence markers allows the straightforward identification of remobilization events. Finally, the physical deletion of the reconstituted piggyBac transposon could be verified at a molecular level by PCR analysis (Fig. 9): Utilizing a primer pair binding to genomic region flanking to the RMCE acceptor transgenic line M4.II (primer M4.II Rev) and to piggyBacLl sequences (primer pBL-R), the deletion of piggyBacLl could be confirmed (compare PCR amplification products for acceptor line M4.II and immobilized lines #7 and #8 in Fig. 9). Moreover, utilizing a primer pair binding to genomic region flanking to the RMCE acceptor transgenic line M4.II (primer M4.II Rev) and to the linotte sequence (primer lioFwd) the truncation of the immobilized transgene could be confirmed (Fig, 9). The piggyBac remobilization event can be further confirmed by DNA sequencing over the genomic DNA to transgene DNA junction. In conclusion, our data provide a proof-of-principle for the strategy of transgene immobilization by "RMCE with subsequent transposon deletion" in an invertebrate organism (Drosophila melanogaster). Advantages of the Invention over the Prior Art

The major advantage of the novel transformation systems disclosed in this patent application is the possibility to physically delete transposon DNA following the germ-line transformation process, in addition to targeting transgene integrations into predefined target sites. In this way, transposase-mediated mobilization or cross-mobilization of the genes-of-interest are excluded mechanistically and random genomic integrations are eliminated. In contrast to conventional germ-line transformation technology, our systems provide enhanced stability to the transgene insertion. Furthermore, DNA sequences required for the modification (e.g. transformation marker genes, transposase or recombinase target sites) are, to a large extent, removed from the genome after the final experimental step (step 2 in Fig. 1, step 3 in Fig. 4 and step 2 in Fig. 5). The final transgene insertion does not contain DNA sequences encoding complete target sites for the recombinases or transposases employed during the process, thereby eliminating the possibility for instability generated by these processes.

The RMCE technology, which is disclosed in this patent application for invertebrate organisms (exemplified in Drosophila melanogaster) represents an extremely versatile tool with application potential far beyond the goal of transgene immobilization. RMCE makes possible the targeted integration of DNA cassettes into a specific genomic DNA locus. This locus is pre-defined by the integration of the RMCE acceptor plasmid and can be characterized prior to a targeting experiment. In addition to the expected expression properties of the transgenes (including strength of expression, stage-specificity, tissue-specificity, and sex-specificity), the genomic environment of the transgene integration can have a significant effect on the level and tissue-specificity of expression. Therefore, suitable loci for integrations can be pre-selected before performing a gene targeting experiment according to the requirements specific for the experimental setup, and in addition, host strains with optimal fitness may be selected. Moreover, multiple cassette exchange reactions can be performed in a repetitive way, i.e. an acceptor cassette in a particular invertebrate strain with a specific genetic makeup can be repetitively exchanged by multiple donor cassettes. Furthermore, several different transgenes can be placed exactly at the same genomic locus. This allows for the first time the ability to eliminate genomic positional effects and to comparatively study the biological effects of different transgenes.

The particular embodiments of the invention are highly flexible. The functionality of systems disclosed is neither dependent on the particular transposable elements used in the embodiments, nor on the particular transformation marker genes used in the embodiments, nor on the particular site-specific recombination system used in the embodiments, nor on the particular homing sequence used in embodiment 3. Finally, all embodiments have broad general application potential in vertebrate and invertebrate organisms that are subject to transposon-mediated transformation or recombinase-mediated recombination, and fluorescent protein marking systems.

Claims

1. A method for producing a heritable integration of a transgene within a genome of a somatic or germ line cell of an invertebrate organism, the method comprising: providing a first DNA cassette within said genome, wherein said first cassette comprises a first flanking transposon half side, a second flanking transposon half side, and an internal transposon half side, wherein said internal transposon half side and said first flanking transposon half side form a pair of excisable transposon half-sides, and wherein said first cassette further comprises said transgene in-between the internal transposon half side and said second flanking transposon half side; and mobilizing said excisable transposon half-sides.

2. The method of claim 1, wherein said internal transposon half side and said second flanking transposon half side are TransposonL half sides, and wherein said first flanking transposon half side is a TransposonR half side.

3. The method of claim 1, wherein said internal transposon half side and said second flanking transposon half side are TransposonR half sides, and wherein said first flanking transposon half side is a TransposonL half side.

4. The method of claim 1, wherein said excisable transposon half-sides and coπesponding transposase enzyme are from a transposable element, wherein said transposable element has terminal inverted sequences, and wherein said transposable element transposes via a DNA-mediated process.

5. The method of claim 1, wherein said first DNA cassette further comprises a first selectable marker gene located between said internal transposon half side and said first flanking transposon half side, and a second selectable marker gene located between said internal transposon half side and said second flanking transposon half side, and wherein said first and second selectable marker genes are phenotypically distinguishable.

6. The method of claim 5, wherein said first and second marker genes are, in either order, any combination of marker genes producing distinguishable fluorescent or other visible dominant phenotypes.

7. The method of claim 5 wherein said first and second marker genes are, in either order, a combination of the transformation marker genes PUbDsRedl and 3xP3-ECFP.

8. The method of claim 1 , wherein said internal transposon half side is provided in reverse orientation, wherein said excisable transposon is formed by inversion of said internal transposon half side relative to said first flanking transposon half side, wherein said internal transposon half side further comprises flanking recombinase sites, and wherein said inversion is catalyzed by a site-specific recombinase.

9. The method of claim 8, wherein said recombinase sites are FRT sites in opposite or reverse orientation.

10. The method of claim 1, wherein said excisable transposon is mobilized by a source of transposase coπesponding to said excisable transposon to render the remaining genomic DNA immobilizable.

11. A method for targeting a heritable integration of a transgene within a genome of a somatic or germ line cell of an invertebrate organism, said method comprising: integrating a first DNA cassette within said genome by transposase-mediated integration of flanking transposon half sides, wherein said first cassette comprises a wild- type/non-mutated or a mutated target site of a site-specific recombinase at one end and a mutated target site of said site-specific recombinase at an other end, wherein said recombinase target sites are heterospecific, and wherein said target sites flank marker gene DNA and additional DNA sequence, and exchanging said first DNA cassette for a second DNA cassette by a site-specific recombinase enzyme that catalyzes a DNA recombination reaction via homospecific recombinase target sites.

12. The method of claim 11, wherein said site-specific recombinase is FLP recombinase, and wherein said recombinase target sites are FRT sites or mutated derivatives of said FRT sites.

13. The method of claim 11, wherein said site-specific recombinase is Cre recombinase, and wherein said recombinase target sites are loxP sites or mutated derivatives of said loxP sites.

14. The method of claim 11, wherein said first cassette comprises one site-specific recombinase target site placed in-between a marker gene coding region and a promoter DNA that regulates its expression.

15. The method of claim 11, wherein said first cassette comprises a homing sequence to enhance pairing to said site-specific recombinase target sites in said second cassette.

16. The method of claim 11, wherein said homing sequence comprises a DNA sequence hybridizing to a Drosophila linotte locus.

17. The method of claim 11, wherein said second cassette comprises said heterospecific site-specific recombinase target sites.

18. The method of claim 17, wherein said second cassette comprises a marker gene coding region lacking a promoter for regulating its expression, and wherein, following the exchange of said first DNA cassette to said second cassette, said marker gene is placed under the control of said promoter derived from said first cassette.

19. The method of claim 17, wherein said second cassette comprises the same homing sequence as said first cassette within said recombinase target sites.

20. The method of claim 17, wherein said second cassette has a transposon half side in- between said recombinase target sites with phenotypically distinguishable marker genes on either side, wherein one of said marker genes lacks a promoter.

21. The method of claim 17, wherein site-specific recombinase mediated insertion occurs between a coding region of said second cassette and an operable promoter of a selectable marker gene of said first cassette.

22. The method of claim 20, wherein said internal transposon half side is excisable with a flanking transposon half side, and wherein said excisable transposon is mobilized by a source of transposase corresponding to said excisable transposon to render the remaining genomic DNA immobilizable.

23. An invertebrate organism comprising the heritable transgene produced according to claim 1.

24. An invertebrate organism comprising the heritable transgene produced according to claim 11.

25. A method for producing a heritable integration of a transgene within a genome of a somatic or germ line cell of an organism, the method comprising: providing a first DNA cassette within said genome, wherein said first cassette comprises a first flanking transposon half side, a second flanking transposon half side, and an internal transposon half side, wherein said internal transposon half side and said first flanking transposon half side form a pair of excisable transposon half-sides, and wherein said first cassette further comprises said transgene in-between the internal transposon half side and said second flanking transposon half side; and mobilizing said excisable transposon half-sides.

26. The method of claim 25, wherein said internal transposon half side and said second flanking transposon half side are TransposonL half sides, and wherein said first flanking transposon half side is a TransposonR half side.

27. The method of claim 25, wherein said internal transposon half side and said second flanking transposon half side are TransposonR half sides, and wherein said first flanking transposon half side is a TransposonL half side.

28. The method of claim 25, wherein said excisable transposon half-sides and coπesponding transposase enzyme are from a transposable element, wherein said transposable element has terminal inverted sequences, and wherein said transposable element transposes via a DNA-mediated process.

29. The method of claim 25, wherein said first DNA cassette further comprises a first selectable marker gene located between said internal transposon half side and said first flanking transposon half side, and a second selectable marker gene located between said internal transposon half side and said second flanking transposon half side, and wherein said first and second selectable marker genes are phenotypically distinguishable.

30. The method of claim 29, wherein said first and second marker genes are, in either order, any combination of marker genes producing distinguishable fluorescent or other visible dominant phenotypes.

31. The method of claim 29, wherein said first and second marker genes are, in either order, a combination of the transformation marker genes PUbDsRedl and 3xP3-ECFP.

32. The method of claim 25, wherein said internal transposon half side is provided in reverse orientation, wherein said excisable transposon is formed by inversion of said internal transposon half side relative to said first flanking transposon half side, wherein said internal transposon half side further comprises flanking recombinase sites, and wherein said inversion is catalyzed by a site-specific recombinase.

33. The method of claim 32, wherein said recombinase sites are FRT sites in opposite or reverse orientation.

34. The method of claim 25, wherein said excisable transposon is mobilized by a source of transposase coπesponding to said excisable transposon to render the remaining genomic DNA immobilizable.

35. A method for targeting a heritable integration of a transgene within a genome of a somatic or germ line cell of an organism, said method comprising: integrating a first DNA cassette within said genome by transposase-mediated integration of flanking transposon half sides, wherein said first cassette comprises a wild- type/non-mutated or a mutated target site of a site-specific recombinase at one end and a mutated target site of said site-specific recombinase at an other end, wherein said recombinase target sites are heterospecific, and wherein said target sites flank marker gene DNA and additional DNA sequence, and exchanging said first DNA cassette for a second DNA cassette by a site-specific recombinase enzyme that catalyzes a DNA recombination reaction via a homospecific recombinase target site.

36. The method of claim 35, wherein said site-specific recombinase is FLP recombinase, and wherein said recombinase target sites are FRT sites or mutated derivatives of said FRT sites.

37. The method of claim 39, wherein said site-specific recombinase is Cre recombinase, and wherein said recombinase target sites are loxP sites or mutated derivatives of said loxP sites.

38. The method of claim 35, wherein said first cassette comprises one site-specific recombinase target site placed in-between a marker gene coding region and a promoter DNA that regulates its expression.

39. The method of claim 35, wherein said first cassette comprises a homing sequence to enhance pairing to said site-specific recombinase target sites in said second cassette.

40. The method of claim 35, wherein said homing sequence comprises a DNA sequence hybridizing to a Drosophila linotte locus.

41. The method of claim 35, wherein said second cassette comprises said heterospecific site-specific recombinase target sites.

42. The method of claim 41, wherein said second cassette comprises a marker gene coding region lacking a promoter for regulating its expression, and wherein, following the exchange of said first DNA cassette to said second cassette, said marker gene is placed under the control of said promoter derived from said first cassette.

43. The method of claim 41, wherein said second cassette comprises the same homing sequence as said first cassette within said recombinase target sites.

44. The method of claim 41, wherein said second cassette has a transposon half side in- between said recombinase target sites with phenotypically distinguishable marker genes on either side, wherein one of said marker genes lacks a promoter.

45. The method of claim 41, wherein site-specific recombinase mediated insertion occurs between a coding region of said second cassette and an operable promoter of a selectable marker gene of said first cassette.

46. The method of claim 44, wherein said internal transposon half side is excisable with a flanking transposon half side, and wherein said excisable transposon is mobilized by a source of transposase coπesponding to said excisable transposon to render the remaining genomic DNA immobilizable.

47. An organism comprising the heritable transgene produced according to claim 25.

48. An organism comprising the heritable transgene produced according to claim 35.