Site-specific recombination for the modification of transgenic strains of the Mediterranean fruit fly Ceratitis capitata

Abstract

Insect transgenesis is mainly based on the random genomic integration of DNA fragments embedded into non-autonomous transposable elements. Once a random insertion into a specific location of the genome has been identified as particularly useful with respect to transgene expression, the ability to make the insertion homozygous, and lack of fitness costs, it may be advantageous to use that location for further modification. Here we describe an efficient method for the modification of previously inserted transgenes by the use of the site-specific integration system from phage phiC31 in a tephritid pest species, the Mediterranean fruit fly Ceratitis capitata. First, suitable transgenic strains with randomly integrated attP landing sites within transposon-based vectors were identified by molecular and functional characterization. Second, donor plasmids containing an attB site, with additional markers, and transposon ends were integrated into attP sites by phiC31 integrase-mediated recombination. Third, transposase-encoding ‘jumpstarter’ strains were created and mated to transgenic strains resulting in the postintegrational excision of transposon ends, which left stably integrated transgene insertions that could not be remobilized. This three-step integration and stabilization system will allow the combination of several transgene-encoded advantageous traits at evaluated genomic positions to generate optimized strains for pest control that minimize environmental concerns.

The development of techniques for the genetic manipulation of insect genomes using transposable elements has a strong impact on our understanding of a wide range of biological processes (1). Moreover, insect transgenesis provides powerful tools that have the potential to improve current pest management strategies (2). For a number of economically or medically important species, transgenesis is nowadays exploited to develop strains with different features which may improve the efficacy of existing biocontrol methods such as the sterile insect technique (SIT) (3–7). The SIT is a powerful pest control strategy involving mass rearing of the target pest insect, sterilization, and release of males over the target area. These compete with the wild-type males for mating with wild-type females, thereby causing a reduction in the pest population size (8). The efficiency of this method may take great advantage from the availability of transgenic modifications conferring different useful features such as fluorescent marking (5, 7, 9), transgenic sexing (3), dominant lethality (4, 10), or reproductive sterility (6, 11). Such transgenes are routinely inserted into the genome by the use of transposons as gene vectors, which integrate randomly. This often leads to position effects, which impact on the transgene functionality (6), and/or disrupts gene structure due to insertional mutagenesis, which can cause recessive lethality (12, 13). These phenomena often negatively impact the overall fitness and reliability of the transgenic strain (14). However, once a fit and functional transgenic strain has been generated and characterized, it would be desirable to take advantage of such an innocuous genomic integration site to manipulate or replace the existing transgene and to introduce additional transgenes to the same genomic position. For this purpose, a site-specific integration system can be used, of which some components need to be already integrated with the initial transposition event. A proven method for generating efficient, unidirectional, and stable integrations is the phiC31-mediated site-specific integration system (15). This approach was successfully used to generate Drosophila melanogaster transgenic strains (16) and to stably integrate DNA fragments larger than 100 kb at specific attachment P (attP) sites into the genome of this model species (17). Moreover, the phiC31-mediated integration system has been shown functional in a vector-disease species, the yellow fever mosquito Aedes aegypti (18).

Also for the economically important insect pest, the Mediterranean fruit fly (medfly) Ceratitis capitata (Diptera: Tephritidae), site-specific recombination will be a valuable tool for the optimization of the currently available and characterized transgenic strains. In this study, we describe a combined approach to manipulate and stabilize medfly transgenic strains, using two strategies in a tephritid species: (i) the phiC31-mediated integration system to site-specifically modify previously integrated transgenes; and (ii) piggyBac-transposase expressing medfly ‘jumpstarter’ strains as a tool for efficient remobilization of transposon vectors. This will allow the generation of integrations of a transgene by simple crossings or the excision of inverted terminal repeats (ITRs) leading to the stabilization of specific transgene insertions. Due to the design of the integrating constructs, additional transposable ITRs can be added to the ones introduced by the initial transposon-mediated germline transformation. Adding such ITRs will allow subsequent directional deletion of specific transgene segments, generating, for example, 5′ ITR-free strains. After such a stabilization process, the resulting strains should be improved in terms of potential ecological concerns related to the release of transgenic insects in biological control programs. Genetically stable strains have key importance for field applications such as SIT programs, in which strains have to fulfill additional qualifications regarding transgene stability in mass rearing and release compared to strains used in small populations for research. In addition, the site-specific integration will permit the addition of further functional transgenes at a preevaluated genomic position, allowing the development of composite transgenic individuals able to express several advantageous traits.

Results

The transgene modification system described here is based on the combination of two technologies for medfly: (i) A method for site-specific integration via co-injection of capped mRNA integrase from phage phiC31 and attB-containing DNA plasmids into transgenic embryos carrying an attP landing site in the genome. (ii) The use of medfly jumpstarter strains that can provide a transposase source by interbreeding, resulting in remobilization or stabilization of recombined transgenic insertions by ITR-mediated excision (Fig. 1).

Fig. 1.

Three-step strategy for transgene stabilization by site-specific integration and subsequent removal of piggyBac ITRs. The original transposon-mediated randomly integrated insertion of a transgene cassette carries one pair of piggyBac ITRs (5′and 3′), an attP site, the transgene of interest and one fluorescent marker (M1). (A) By co-injection of phiC31 integrase mRNA and a plasmid containing a 3′ piggyBac ITR, an attB site, an additional transgene (AT) as well as a different fluorescent marker (M2), the plasmid is site-specifically integrated at the attP landing site. The recombined insertion then carries one additional 3′ piggyBac end, two fluorescent markers, the two transgenes, the plasmid backbone (ampR), as well as the hybrid sites attR and attL. (B) The exposure to a transposase source encoded by a jumpstarter strain excises the piggyBac cassette including 5′end, attR, M2, plasmid backbone, and 3′end. This generates a 5′end-free combination of transgenes.

Development and Test of piggyBac Jumpstarter Strains in Medfly.

To test medfly piggyBac jumpstarter strains as a transposase source for the remobilization or excision of ITRs, we generated the three jumpstarter strains jump_1, jump_3, and jump_4 by transformation with a Minos-based vector (AH370) containing an hsp70-regulated piggyBac transposase gene and the medfly we⁺ marker. Medfly transformants were generated with this construct in the we,wp (19) host strain and selected by red eye pigmentation, with resulting G1 progeny inbred to generate homozygous transgenic progeny. By crossing and inbreeding the strains sl1-tTA_F1m1 or sl1-tTA_M6m1 (both marked with PUbDsRed; 3) with strain we,wp, two jumpstarter-tester strains were obtained that carry the red fluorescent marker DsRed homozygously under control of the D. melanogaster polyubiquitin promoter (PUb) (5, 7, 9) in a white eye background. Subsequently, the three jumpstarter strains were independently crossed in four repetitions to the two tester strains sl1-tTA_F1m1_we,wp and sl1-tTA_M6m1_we,wp. Progeny from all 24 crossings were independently backcrossed to we,wp and their progeny screened for new patterns of red fluorescence (Fig. 2).

Fig. 2.

Activity of piggyBac jumpstarter strains. A male from the homozygous medfly strains sl1-tTA_F1m1_we,wp or sl1-tTA_M6m1_we,wp (in both lines the transgenes are marked with PUbDsRed) was mated in four repetitions (R1–R4) to four females of the three independent homozygous jumpstarter strains jump_1 (B1–B8), jump_3 (C1–C8), or jump_4 (D1–D8), respectively (jumpstarter transgenes are marked with we⁺). G1 males were outcrossed to we,wp females. Their progeny was scored for successful remobilization events, indicated by red fluorescent patterns different from those of the heterozygous control adults from sl1-tTA_F1m1 x we,wp (A1 and A2) and sl1-tTA_M6m1 x we,wp (A3-A4) (dorsal view: A1 and A3; ventral view: A2 and A4). Flies were observed with the DsRedwide filter.

All jumpstarter strains successfully remobilized piggyBac-mediated insertions with a similar efficiency as indicated by new expression patterns of the PUb-DsRed marker. Between 10–20% of all screened progeny showed between one and eight new patterns in respect to the original one. As jump_1 did not create as clearly identifiable new patterns as jump_3 and jump_4 (jump_3 created e.g., fluorescently marked eyes) (Fig. 2, C8), jump_3 was selected for the remobilization-stabilization experiments described below.

attP Landing Sites and attB Integration Constructs.

Site-specific integration was tested at attP landing sites from two molecularly, as well as functionally, well characterized and critically evaluated transgenic strains generated by piggyBac-mediated germline transformation: (i) the strain carrying the effector gene insertion TREhs43-hid^Ala-5_F1m2 was used to establish a functional embryonic lethality system to generate reproductive sterility. One of the resulting lethality lines, #67, was shown to cause 100% conditional embryonic lethality and males of this strain were highly competitive relative to wild-type medfly males in laboratory and field cage tests (6); (ii) the strain carrying insertion 1260_F-3_m-1 has been evaluated to potentially improve monitoring of pest management programs by fluorescent sperm marking and showed no disadvantages in preliminary laboratory competitiveness assays (7). Both (i) and (ii) transgene insertions are single transposon integrations as proven by Southern blots and inverse PCR (6, 7) and therefore contain a single 52-bp attP recombination site that can serve as the landing site for attB-containing plasmids (20).

The strain TREhs43-hid^Ala-5_F1m2 carries an EGFP marker driven by PUb (6), leading to EGFP expression in the adult thorax (Fig. 3). The strain 1260_F-3_m-1 carries two different markers: turboGFP driven by the C. capitata β2 tubulin promoter and DsRed driven by PUb, leading to tGFP expression in the testes and DsRed expression in the adult thorax, respectively (2) (Fig. 3).

Fig. 3.

Stabilization procedure. The original genomic (A), the integrated (B) and the stabilized situation (C) together with the fluorescence marker expression of an adult male from each strain (ventral view) are shown. All flies were observed under cold light and the filter sets DsRed and EYFP, as indicated for each picture. attB-containing plasmids were integrated by site-specific recombination at the attP landing sites of the original strains TREhs43-hid^Ala-5_F1m2 and 1260_F-3_m-1 (A), generating the integrated strains int_TREhs43-hid^Ala-5_F1m2 and int_1260_F-3_m-1 (B). By crossing adults from (B) to the jumpstarter strain jump_3, crossing their progeny to we,wp and screening by fluorescence microscopy, we developed the stabilized strains stab_TREhs43-hid^Ala-5_F1m2 and stab_1260_F-3_m-1 (C).

Since critical evaluation of the strains TREhs43-hid^Ala-5_F1m2 and 1260_F-3_m-1 indicated that their integration sites were particularly useful with respect to transgene expression, the ability to make the insertion homozygous, and the lack of fitness costs (2, 6), they were used in this study to test whether their insertions could be further modified by the use of the phiC31 integration system. For integrase-mediated germline transformation two plasmids containing a 51-bp attB recombination site (20) and an additional 3′ piggyBac ITR were generated: pSLaf_3′pBac-attB_PUb-DsRed_af (#1252) and pSLaf_3′pBac->-attB_PUb-EGFP_af (#1255) carrying a red fluorescent (DsRed) or a green fluorescent marker (EGFP) under the control of the PUb promoter (21, 22), respectively (Fig. 3). When designing these plasmids, we paid particular attention to the orientation and arrangement of the different components: the specific placement of the plasmid backbone including its antibiotic resistance gene between the attB and the 3′ piggyBac ITR makes it possible that this part is removed again by the stabilizing excision (Figs. 1 and 3).

Site-Specific Integration into attP Landing Sites.

When we performed site-specific integration experiments into the attP landing sites of the strains TREhs43-hid^Ala-5_F1m2 (6) and 1260_F-3_m-1 (7), we tested for the highest integration frequencies by injecting different ratios of attB-containing plasmid DNA and capped integrase mRNA (23). Four different co-precipitations (DNA/capped-mRNA in ng/μL: (i) 200/100, (ii) 200/300, (iii) 200/600, and (iv) 500/400) were prepared for each attB-containing plasmid (#1252 and #1255). The co-precipitations of plasmid #1252 were then independently injected into 200 homozygous medfly embryos of the strain TREhs43-hid^Ala-5_F1m2, resulting in 26, 49, 52, and 65 hatched larvae of which 16, 34, 38, and 39 survived to adulthood, respectively. The four different co-precipitations of plasmid #1255 were independently injected into 250 medfly embryos from the strain 1260_F-3_m-1, resulting in 120, 107, 104, and 93 hatched larvae of which 77, 64, 63, and 61 survived to adulthood, respectively. Adults obtained after injections were sorted by sex and backcrossed to wild-type flies in two independent crossings and their offspring were screened for fluorescence expression patterns. In both injection series, transgene integrations were obtained from injections at ratio (iii), 200 ng/μL DNA to 600 ng/μL capped mRNA, which therefore seems to be the most reliable ratio for successful phiC31-mediated site-specific integration in medfly. Ten G₁ flies from injections of plasmid #1252 into TREhs43-hid^Ala-5_F1m2 embryos showed both the original green and the integrated red fluorescence in the body. All 10 flies were from one of the two crossings. One G₁ fly from injections of plasmid #1255 into 1260_F-3_m-1 embryos showed the green fluorescence in the testes, original red fluorescence in the thorax, and newly integrated green fluorescence also in the thorax (Fig. 3).

Initially, site-specific integrations were identified by the phenotypic expression of the respective fluorescent markers. Since PUb-mediated fluorescence can vary significantly dependent on the particular genomic insertion site (Fig. 2), the identity of fluorescent patterns of the previously integrated transformation marker (EGFP or DsRed) and the new integration marker (DsRed or EGFP, respectively) indicates close proximity in the genome which suggests site-specific integration. For both integration targets, transformed individuals showed the integrated fluorescent markers in a similar pattern as the original transformation marker (Fig. 3). The stable DsRed marker (5, 7, 9) showed additional fluorescent expression in the adult, because the EGFP marker loses its intensity in the abdomen during the first hours after eclosion. The two integrated strains, int_TREhs43-hid^Ala-5_F1m2 and int_1260_F-3_m-1, were then independently made homozygous by inbreeding recombinant adult flies and screening for homozygous progeny by fluorescence intensity. Since we did not detect any different fluorescent patterns upon integration, we have no evidence for non-specific off-target integrations.

Molecular Characterization of Site-Specific Integration Events.

To molecularly confirm the site-specific integrations of attB-containing plasmids, strains int_TREhs43-hid^Ala-5_F1m2 and int_1260_F-3_m-1 were characterized by PCR and DNA sequencing. Fragments covering the attR site were amplified from int_TREhs43-hid^Ala-5_F1m2 and int_1260_F-3_m-1 (196- and 560-bp predicted fragment sizes, respectively) by using the primer pair mfs-372/mfs-373. Fragments encompassing the attL site were amplified from int_TREhs43-hid^Ala-5_F1m2 and int_1260_F-3_m-1 (185- and 535-bp predicted fragment sizes, respectively) by using the primer pair mfs-330/mfs-360 (Fig. 1). All DNA fragments were cloned and sequencing confirmed the successful integration of the attB-containing plasmids into the attP site of the medfly strains (Dataset S1 and Dataset S2). Integration-generated attR and attL sites were identical to the respective sequences generated after integration in Drosophila (16).

Excision of piggyBac ITRs and Stabilization of Transposon-Vector Insertions.

As the attB-containing plasmids were designed to carry an additional 3′ piggyBac ITR, their integration into the attP site generated a recombinant genomic transgene with three piggyBac ITRs, as shown in Figs. 1 and 3. Remobilization of the 5′ with either one of the 3′ piggyBac ITRs, duplicated in tandem, should therefore be possible. As described previously (24), remobilization with the internal 3′ ITR sequence should result in stabilization of the remaining 3′ ITR and proximal vector sequences. To assay for this, the homozygous jumpstarter strain jump_3 was crossed separately to the homozygous strains int_TREhs43-hid^Ala-5_F1m2 and int_1260_F-3_m-1. Progeny were sexed and outcrossed to the we,wp strain to identify jumpstarter-free individuals by loss of eye pigmentation (see Fig. S1 for details on the crossing scheme). Excision or transposition events caused by the transposase source were detected in the G₂ progeny by loss or change of fluorescent patterns. In the G₂ progeny of the cross (i) int_TREhs43-hid^Ala-5_F1m2 × jump_3, we identified 3% of flies showing only the green, but not red fluorescent thorax/legs (Fig. 3). Five percent of the G₂ progeny obtained from the cross (ii) int_1260_F-3_m-1 × jump_3 showed red, but not green fluorescent thorax/legs (Fig. 3). This indicated that these flies lost the integrated fluorescent markers by successful excision of part of the integrated cassette including the plasmid backbone flanked by the 5′ and a 3′ piggyBac end, but retained the original transformation marker and, expectedly, also the transgene(s) (Figs. 1 and 3). Independently observed loss and/or change of fluorescent patterns indicate the occurrence of complete excision as well as several combinations of transposition events, which vary widely due to the presence of three piggyBac ITRs in the integrated strains (see Fig. S2).

From crosses (i) and (ii), one G₂ white eyed fly verified for the loss of the integrated marker was outcrossed to we,wp, their progeny screened again for individuals with the correct fluorescent pattern and white eyes, indicating the absence of the transposase source, and finally inbred. By this procedure we generated the homozygous stabilized strains stab_TREhs43-hid^Ala-5_F1m2 and stab_1260_F-3_m-1 (Fig. 3).

Molecular Characterization of Stabilized Medfly Strains.

To molecularly verify the controlled excision of part of the site-specifically integrated DNA construct, PCRs with a forward primer in the flanking genomic region and a reverse primer within the transgene were used. Regions including genomic DNA down to the attL site were amplified from stab_TREhs43-hid^Ala-5_F1m2 and stab_1260_F-3_m-1 (460- and 708-bp fragment sizes, respectively) by using the primer pair mfs-383/mfs-360 and mfs-380/mfs-382, respectively (Fig. 1). Both fragments were subcloned and sequencing confirmed the successful and precise piggyBac excision of the integration marker and ITRs, leaving the transgene and original transformation marker intact in its original genomic position (Dataset S1 and Dataset S2). In addition, we performed a Southern blot on genomic DNA of the original, integrated, and stabilized strains that further indicates the correct site-specific integration and precise piggyBac excision (see Fig. S3). Moreover, this Southern blot also indicates that there are no additional non-specific off-target integrations in the genome of the stabilized lines.

Stability of 5′ piggyBac ITR-Free Strains.

To assay for the stability of the remaining part of the transgene insertions after controlled ITR-excision, we separately crossed the strains stab_TREhs43-hid^Ala-5_F1m2 and stab_1260_F-3_m-1 to jump_3. As a control, their respective original versions (TREhs43-hid^Ala-5_F1m2 and 1260_F-3_m-1) were also crossed to jump_3. The progeny of these four crossings were then mated to the we,wp strain and their offspring separately scored according to the fluorescence expression patterns. No new fluorescent pattern was detected in three independent repetitions with strain stab_TREhs43-hid^Ala-5_F1m2 out of a total of 37,058 progeny or with strain stab_1260_F-3_m-1 out of a total of 36,802 progeny (Table 1). On the contrary, the control with the original strains crossed to jump_3 always generated new fluorescent patterns (in total 36 progeny out of 627 for TREhs43-hid^Ala-5_F1m2 and 81 progeny out of 807 for 1260_F-3_m-1). Moreover, whereas in the control less than 50% of the progeny showed fluorescence due to excision events, in the stabilized strains 50% of the progeny remained fluorescent which suggests that no excision events took place. This indicates that the presence of only one ITR (3′ piggyBac end) (Figs. 1 and 3) in the stabilized strains does not mediate excisions or transpositions in C. capitata similar to what has been shown in D. melanogaster (24). Furthermore, when propagating the homozygous stabilized strains over many generations, we have no indication of transgene loss. These results support the conclusion that the transgenic insertions were successfully stabilized and were no longer a substrate for further transposase-mediated remobilization events.

Table 1.

Verification of transgene stability

	Strains	TREhs43-hidAla5_F1m2 (C)			stab_TREhs43- hidAla5_F1m2			1260_F-3_m-1 (C)			stab_1260_ F-3_m-1
	Replication	I	II	III	I	II	III	I	II	III	I	II	III
pattern	WT flies	149	160	172	6206	6168	6200	245	186	146	6176	6198	6074
	Original	43	36	31	6130	6204	6150	44	71	34	6128	6096	6130
	New (different)	18 (2)	9 (3)	9 (3)	0	0	0	36 (6)	26 (4)	19 (3)	0	0	0
	total n	627			37058			807			36802

The numbers of screened G₂ progeny from the strains TREhs43-hid^Ala5_F1m2, stab_TREhs43-hid^Ala5_F1m2, 1260_F-3_m-1, and stab_1260_F-3_m-1 after crossing them to jump_3 and outcross progeny to we,wp are shown. Numbers of transgenic and non-transgenic (WT) progeny were recorded. All fluorescent expression patterns different (New) from the initial pattern (Original) generated after the crossing to jump_3 represent a remobilization event which occurred in the parental adult. The number of new different patterns is given in brackets. Original strains used as remobilization controls are marked with (C). Note that the number of fluorescent progeny is less than 50% in the non-stabilized strains and about 50% in the stabilized strains, which indicates a lack of excision in the stabilized strains.

Discussion

The established piggyBac jumpstarter strains can be used to generate new integrations of transgenes without the need for re-injecting vector and helper plasmids to perform germline transformation. Since position effects can greatly affect the usability of transgene insertions (3), the remobilization of transgenes by jumpstarter strains enables the quick generation of many new genomic insertions that then can be tested for transgene effectiveness and the ability to become homozygous as well as potential fitness costs.

The successful application of the phiC31-mediated site-specific integration system will confer great flexibility to medfly transformation technology by: (i) enabling the combination of different transgenic systems in this pest species at preevaluated genomic positions; (ii) facilitating the stabilization of transgenic insertions; and (iii) potentially permitting the integration of large DNA fragments as it has been reported for D. melanogaster (17). Using site-specific systems, researchers should be able to create transposon-free cassette exchange systems for replacement of original transgenes with any desired transgenic construct and to more efficiently combine different systems. This will enable the creation of transgenic strains carrying diverse transgenes in tandem (Fig. 1), which will be inherited by their progeny together. For example, a combination of a conditional embryonic lethality (6), a sperm marking (7), and a sexing system (3, 25) could be put into effect in medfly with the described strains. Deleting piggyBac ends could then further increase the stability of such strains. Having all these systems in one genomic position, proven to be non-compromising and effective, should improve the sterilization, sexing and monitoring processes of SIT programs. We would expect that the generated stabilized lines show the same fitness as their predecessors, since the stabilization procedure only removes part of the transposon construct that has not been there in the wild-type situation. Nevertheless, the fitness of the stabilized lines needs to be verified by large scale field cage competition assays before an eventual field application.

The presence of attP landing sites in the genome of already characterized transgenic strains will enable the insertion of different transgenes at precise positions for comparative analysis. This advance should greatly facilitate the detailed study of medfly regulatory elements such as enhancers, silencers, and insulators or the function of medfly protein variants by the use of expression systems (26).

Transgenic technology relying on transposon-based vectors has proven to be a powerful tool for the genetic manipulation of insect genomes, which can be tested and applied to the improvement of current environment-friendly pest control methods (1). However, a fundamental requirement for a safe release of genetically modified insects into the wild is the development of systems inert to any potential mobilizing transposase source present in the environment. Therefore the development of non-autonomous transposons whose ends can be deleted after integration to cause effective immobilization of the inserted transgene is essential to ultimately taking transgenic improvements of pest management programs from the lab to the field (27). Here we demonstrate that it is possible to generate 5′ piggyBac ITR-free insertions in a major pest species and that these modified insertions are inert to the piggyBac transposase, which might be inadvertently present in the transgenic host strain. This represents an important step forward toward the safe field use of transgenic insects.

Methods

Medfly Strains.

The medfly strains we,wp (phenotype white eyes, white pupae) and Egll (WT) were received from Gerald Franz (IAEA). Both WT and transgenic medfly strains were reared under standard laboratory conditions (28).

Plasmid Construction.

Each of the pBac-jumpstarter strains contains the pMi{Ccwhite+; hspBac} (AH_370) generated by an exchange of the 3×P3-DsRed from pMi{3xP3-DsRed; hsp70-piggyBac} (29) for the medfly white gene (30). The attB-containing constructs pSLaf_3′pBac-attB_PUb-DsRed_af (#1252) and pSLaf_3′pBac->-attB_PUb-EGFP_af (#1255) were generated by ligating the BglII/AflII cut fragment PUb-DsRed (3.0 kb) and PUb-EGFP (3.2 kb) from #1200 and #1201 (7), in the BglII/AflII cut vectors pSLaf_3′pBac-attB_af (#1250) and pSLaf_3′pBac->-attB_af (#1251), respectively. We created #1250 or #1251 by cloning hybridized primers mfs-205 (GATCCTGCGGGTGCCAGGGCGTGCCCTTGGGCTCCCCGGGCGCGTACTCCACCTCACA)/mfs-206 (GATCTGTGAGGTGGAGTACGCGCCCGGGGAGCCCAAGGGCACGCCCTGGCACCCGCAG) in the BamHI/BglII site of pSLaf_3′pBac_af or pSLaf_3′pBac->_af, respectively. The pSLaf_3′pBac->_af plasmid was generated by ligating a 0.4-kb gypsy element, amplified by PCR on pBac{3xP3-DsRed>af} (31) with the primer pair mfs-197 (CAGTGGGCCCGGTACCCTATTCGCAAAAAC)/mfs-198 (ACTCGGATCCGGCTAAATGGTATGGCAAG) and subsequent digestion with restriction enzymes ApaI/BamHI, in the ApaI/BamHI cut pSLaf_3′pBac_af. To generate pSLaf_3′pBac_af, a 1.3-kb 3′pBac EcoRV/HpaI fragment from p3E1.2 was ligated into the SmaI cut pSLfa1180fa (32).

Medfly Germline Transformation.

Jumpstarter Strains.

Medfly transformation using Minos transposable elements marked by hsCcw (33) to generate piggyBac jumpstarter strains was performed using standard methods (34) into the we,wp strain.

Site-Specific Integration.

The capped phiC31 integrase mRNA was transcribed from the BamHI cut plasmid pcDNA3.1-phiC31 (23) according to the protocol of the mMESSAGE mMACHINE T7 kit (Ambion). phiC31 capped integrase mRNA was co-precipitated with attB plasmids #1252 or #1255, and injected in 200–250 embryos of from the strains TREhs43-hid^Ala-5_F1m2 or 1260_F-3_m-1, respectively. All G₀ adults were sexed and mated to 20 WT flies of the opposite sex. The G1 progeny were screened for the presence of fluorescence by epifluorescence microscopy using the stereomicroscope LEICA MZ 16 FA and Filter sets DsRedwide (Ext. 546/12; Emm. 605/75), DsRed (Ext. 545/30; Emm. 620/60), and EYFP (Ext. 500/20; Emm. 535/30). Recombinant G1 individuals were bred to homozygosity.

Molecular Characterization.

For the molecular characterization, touchdown PCRs were performed (2 min at 94 °C; 6 cycles of 30 s at 94 °C, 45 s at 70 °C (−2 °C each cycle), 1.5 min at 72 °C; 27 cycles of 30 s at 94 °C, 45 s at 56 °C, 1.5 min at 72 °C; and 5 min at 72 °C) on genomic DNA from the original strains TREhs43-hid^Ala-5_F1m2 and 1260_F-3_m-1 and from the integrated strains int_TREhs43-hid^Ala-5_F1m2 and int_1260_F-3_m-1 using the BD Advantage 2 PCR Kit (BD Biosciences). To verify the genomic positions of the transgenes in TREhs43-hid^Ala-5_F1m2 the primer pair mfs-383 (GTCTGGACTGTAAGGTTGGTGATTA)/mfs-360 (GCCTAGCGACCCTACGCCCCCAACTGAG) was used, while primer pair mfs-380 (GGCCCACATTTGTCTAACTTCT)/mfs-229 (CAGTGACACTTACCGCATTGACAAGCACGCCTCAC) was used for 1260_F-3_m-1. To verify the attR integration site in the integrated strains the primer pair mfs-373 (ATCTTGACCTTGCCACAGAGGACTATTAGAG)/mfs-372 (TTGAGCTCGAGATCTGTGAGGTGGAGTACG) was used. To verify the attL integration site of int_TREhs43-hid^Ala-5_F1m2 the primer pair mfs-330 (GCTCATCGACTTGATATTGTCCGACAC)/mfs-360 was used, while for int_1260_F-3_m-1 the primer pair mfs-330/mfs-382 (ATTGACCCTACGCCCCCAACTGAG) was used. PCR products were cloned into pCRII vectors (Invitrogen) and sequenced.

Touchdown PCRs (2 min at 94 °C; 5 cycles of 30 s at 94 °C, 45 s at 68 °C (−2 °C each cycle), 1 min at 68 °C; 27 cycles of 30 s at 94 °C, 45 s at 58 °C, 1 min at 68 °C; and 2 min at 68 °C) for the molecular characterization of stabilized strains were performed on genomic DNA from the strains stab_TREhs43-hid^Ala-5_F1m2 and stab_1260_F-3_m-1 using the BD Advantage 2 PCR Kit (BD Biosciences). To verify the remobilization of the integrated markers, the 5′ piggyBac end and one 3′ piggyBac end, the primer pair mfs-383/mfs-360 was used for the strain stab_TREhs43-hid^Ala-5_F1m2 and mfs-380/mfs-382 for the strain stab_1260_F-3_m-1. PCR products were cloned into pCRII vectors (Invitrogen) and sequenced.

piggyBac ITR Excision Assay.

Four medfly strains were used to perform the excision assay: int_TREhs43-hid^Ala-5_F1m2, int_1260_F-3_m-1, jump_3, and we,wp. The integrated strains int_TREhs43-hid^Ala-5_F1m2 and int_1260_F-3_m-1 carry both the genetically linked red and green fluorescent markers as homozygous alleles. jump_3 carries in we background the genetically linked medfly hsCcw⁺ marker gene (25) and the piggyBac transposase as homozygous alleles. To excise the integration marker together with the transposable ends, the homozygous strains int_TREhs43-hid^Ala-5_F1m2 and int_1260_F-3_m-1 (Fig. 3) were separately crossed to the homozygous jumpstarter strain jump_3. From both crossings progeny heterozygous for both markers and the piggyBac transposase was outcrossed to we,wp. Their progeny was screened for green and red fluorescent expression patterns. White eyes indicated the absence of the genetically linked hsCcw marker and piggyBac transposase gene. Successful excision or remobilization events were identified by the loss or the appearance of new fluorescent expression patterns of the PUb-mediated fluorescence, respectively.

Stability Test.

To check the stability of the transgenic insertions of stab_TREhs43-hid^Ala-5_F1m2 and stab_1260_F-3_m-1, 50 males from each of these two strains were crossed to 50 homozygous jump_3 females in three independent repetitions. Of their progeny, 250 virgin females were then crossed to 250 we,wp males. In parallel, at a lower scale, the same experimental design was applied to the two original strains TREhs43-hid^Ala-5_F1m2 and 1260_F-3_m-1 as a control: 10 males from each of the original strains were crossed to 10 homozygous jump_3 females and 10 virgin females of their progeny were then crossed to 10 we,wp males. About 12,350 progeny from each independent crossing of stabilized lines and between 200–325 progeny from each independent crossing of the original lines were screened for fluorescence patterns. The numbers of flies showing (i) the WT phenotype, (ii) the original fluorescence pattern, (iii) a different fluorescence pattern and (iv), the number of new patterns generated by the transgene remobilization events were scored.