Post-integration behavior of a Mos1 mariner gene vector in Aedes aegypti

Insect Biochemistry and Molecular Biology 33 (2003) 853–863www.elsevier.com/locate/ibmb

Post-integration behavior of aMos1 mariner gene vector inAedes aegypti

Raymond Wilsonab, Jamison Orsettia, Andrew D. Klockoa, Channa Aluvihare,Edward Peckham, Peter W. Atkinsonc, Michael J. Lehaneb, David A. O’Brochtaa,∗

a Center for Biosystems Research, University of Maryland Biotechnology Institute, College Park, MD 20742-4450, USAb School of Biological Sciences, University of Wales, Bangor, Gwynedd LL57 2UW, UK

c Department of Entomology, University of California, Riverside, CA 92521, USA

Received 10 January 2003; accepted 3 March 2003

Abstract

The post-integration behavior of insect gene vectors will determine the types of applications for which they can be used. Transpo-son mutagenesis, enhancer trapping, and the use of transposable elements as genetic drive systems in insects requires transposableelements with high rates of remobilization in the presence of transposase. We investigated the post-integration behavior of theMos1mariner element in transgenicAedes aegypti by examining both germ-line and somatic transpositions of a non-autonomous elementin the presence ofMos1 transposase. Somatic transpositions were occasionally detected while germ-line transposition was onlyrarely observed. Only a single germ-line transposition event was recovered after screening 14,000 progeny. The observed patternsof transposition suggest thatMos1 movement takes place between the S phase and anaphase. The data reported here indicate thatMos1 will be a useful vector inAe. aegypti for applications requiring a very high degree of vector stability but will have limiteduse in the construction of genetic drive, enhancer trap, or transposon tagging systems in this species. 2003 Elsevier Ltd. All rights reserved.

1. Introduction

The Mos1 mariner transposable element is one of sixinsect transposable elements that can be used to createtransgenic insects. TheP and hobo elements from Dro-sophila have been used as vectors inD. melanogasterand related species within the genus (Blackman et al.,1989; Lozovskaya et al., 1996; Rubin and Spradling,1982) while Hermes, Minos, piggyBac and Mos1 havebeen used in a wide range in species in three Ordersof insects (Handler et al., 1998; Loukeris et al., 1995;O’Brochta and Atkinson, 1996). The active marinerelementMos 1, however, currently rivals most othertransposable elements in its host range and has been usedto create transgenic microbes, protozoans (Gueiros-Fihoand Beverley, 1997; Mamoun et al., 2000), insects(Lampe et al., 2000) and vertebrates (Fadool et al., 1998;Sherman et al., 1998). Consistent with its wide host

∗ Corresponding author.E-mail address: [email protected] (D.A. O’Brochta).

0965-1748/$ - see front matter 2003 Elsevier Ltd. All rights reserved.doi:10.1016/S0965-1748(03)00044-4

range as a germ-line transformation vector is the wide-spread distribution ofmariner-like elements in nature.Robertson (1993, 1995)and Robertson and MacLeod(1993)found that more than 15% of the 400 insect spec-ies they examined containedmariner-like elements.These elements have since been reported in plants, ver-tebrates (including humans), planaria, nematodes, mites,and centipedes (for review seePlasterk et al., 1999).Horizontal transfer ofmariner-like elements followed byelement amplification by transposition seems to be oneway in which Mos1 and mariner-like elements havebecome so widely distributed (Hartl et al., 1997).Together these data suggest that theMos1 transposableelement system is an autonomous recombination system,requiring few or no host factors for its movement. Invitro studies support this conclusion and have shown thatMos1 transposition can occur in the presence of onlyMos1 donor elements, target DNA, purifiedMos1 trans-posase and Mg+2 (Lampe and Robertson, 1996).

While much of what is known aboutMos1 seems toparallel the behavior of many other Class II transposableelements, there are some aspects of theMos1 system

854 R. Wilson et al. / Insect Biochemistry and Molecular Biology 33 (2003) 853–863

that remain enigmatic. The first notable and unexpectedcharacteristic of Mos1 was observed when the elementwas introduced into the germ-line of D. melanogaster.The widespread ability of mariner-like elements toinvade and transpose in foreign genomes is contrastedby the observed stability of Mos1-based gene vectors inD. melanogaster. Lidholm et al. (1993) created two inde-pendent transgenic lines of D. melangaster using aMos1-based gene vector containing an 11.9 kb insert ofexogenous DNA. They estimated an initial transform-ation frequency of approximately 5%, however, theywere only able to detect very low rates of subsequentmovement of the integrated elements in both the somaand the germ-line when Mos1 transposase was sub-sequently supplied in trans. Lidholm et al. (1993)reported somatic mosacism in 1% or less of the progenycontaining a Mos1 vector. In contrast, the whitepeach

mariner element, a complete and intact, yet non-auton-omous element, was highly active in the presence ofactive Mos1 transposase and resulted in somaticmosaicism in 100% of the progeny. Other examples ofMos1’s post-integration stability have also been reported(Lohe et al., 1995; Lozovsky et al., 2002). Lozovsky etal. (2002) reported testing a collection of six Mos1elements containing exogenous DNA inserts rangingfrom 1.3 to 11.9 kb in length and located at differentpositions within the element. These simple insertion vec-tors were introduced into D. melanogaster using aHermes-based vector and then subsequently tested fortheir abilities to excise and transpose in the presence offunctional Mos1 transposase. Of the four simple inser-tion vectors that permitted somatic movement to beassessed (SalW, SphW, ClaW, ClaY; Mos1 elementswith either the mini-white (W) or yellow (Y) geneinserted into a unique restriction site within the element)only one showed any evidence of somatic movement. Ofthe approximately 3500 progeny examined containingthe SalW vector only eight had evidence for somaticmosaicism (0.23%). Using more sensitive PCR-basedmethods revealed that Mos1 element movement in thesoma only resulted in excisions being detected in theSalW and SphW vector-containing insects but not ininsects containing the ClaW and ClaY vectors. It was notknown whether the relative stability of integrated Mos1vectors was a unique characteristic of the element in D.melanogaster or whether this was a host-independentcharacteristic of Mos1 (Lozovsky et al., 2002). Given theinterest in using transgenic insect technologies, includingthose based on Mos1, for the development of uniquegene-finding and analysis tools, as well as the creationand release of transgenic insects as biocontrol agents,questions regarding the post-integration behavior of allcurrent insect gene vectors are highly relevant (Atkinsonet al., 2001; Handler and James, 2000). Here we reportthe results of an experiment designed to test the post-integration behavior of Mos1 in the yellow fever mos-

quito, Aedes aegypti, and as in D. melanogaster post-integration transposition was very rare.

2. Materials and methods

2.1. Mosquitoes

Adult Ae. aegypti were maintained at 28 ± 1 °C and80 ± 5% relative humidity with 12 h cycles of light anddark and fed on 10% sucrose. Female mosquitoes wereroutinely permitted to blood-feed on Swiss Webstermice. Larvae were reared at 28 ± 1 °C and fed grounddog biscuits. The strain khw was used in all experimentsand is a white-eyed line containing a homozygousrecessive mutation in the kynurenine 3-monooxygenasegene (Bhalla, 1968). The mutant phenotype can becomplemented by expression of the kynurenine 3-monooxygenase gene (cinnabar) from D. melanogaster(Cornel et al., 1997; Coates et al., 1998; Jasinskiene etal., 1998).

2.2. Gene vectors

MhspRedcn: This vector was originally designed asan enhancer-trap reporter element containing functionalMos1 inverted repeats, approximately 0.3 kb of sub-ter-minal sequences of the right and left ends of Mos1, a4.7 kb genomic DNA fragment from D. melanogastercontaining the cinnabar gene and DsRed attached to aminimal promoter (Fig. 1A). The left and right end ofMos1 were amplified by PCR using pMOS 5’+ 3’ Kan-Rori (Coates et al., 1995) as a template and the primers5’ -TAG GGT ACC(KpnI) CCT GTG TAT ATA TGCGTA AGA AC-3’ and 5’ -CCT CTC GAG(XhoI) CCACTT TTG AAG CGT TGA AAC C-3’ for the left endand 5’ -AAT GCG GCC GC(NotI)C TAC CAA ATGATG TAG ATA GG-3’ and 5’ -CTC GAA TTC(EcoRI)GCG GCT TAC TCA CCA GAC CTG-3’ for the rightend. The left and right ends of Mos1 were inserted intopBCKS (Stratagene) as KpnI/XhoI and NotI/EcoRI frag-ments respectively to create pBCMarLR. The minimalpromoter region of the Drosophila hsp70 promoter wasalso cloned by PCR, using pKhspRB (a plasmid contain-ing 500 bp of the 5’ regulatory region of hsp70 beginningin the 5’ untranslated region) as a template and the pri-mers 5’ TTG CGG CCG CGG AGT ATA AAT AGAGGC GCT T-3’ and 5’ -TAC GGA TCC(BamHI) CACTTT ACT GCA GAT TGT TTA GC-3’ , and insertedas an EcoRI/BamHI fragment into pEYFP-NotI (pEYFP-1 (Clontech) with NotI site removed). The EYFP con-taining BamHI/XbaI fragment of this intermediate plas-mid was replaced with the 700bp DsRed –containingBamHI/XbaI fragment from pDsRed (Clontech). The1kb SacI/XhoI fragment from this plasmid containingDsRed and the minimal promoter was excised and

855R. Wilson et al. / Insect Biochemistry and Molecular Biology 33 (2003) 853–863

Fig. 1. Gene vectors. A) MhspRedcn. A Mos1 vector containing cinnabar and a DsRed-containing enhancer-reporter gene. B) pBac3xP3EGFP-MOS. A piggyBac vector containing an EGFP marker gene and the Mos1 transposase open reading frame (MOS transposase). The location andtranscriptional orientation of all genes are shown by the large arrows. The small arrows are transposable element sequences. piggyBac L, R refersto the left and right ends of piggyBac. Mar L, R refers to the left and right ends of Mos1. hsp70tata is a minimal promoter. SV40polyA is a 3’non-translated region containing a poly A addition signal. 3xP3 is an eye-specific promoter. hsp82 is the promoter region from the hsp82 gene ofD. pseudoobscura. Relevant restrictions sites and corresponding fragments expected on a southern blot probed with the probe indicated are shown.B : BamHI, N : NruI, S : SalI, X : XhoI. The location of PCR primers used in TE display and RT-PCR are shown.

inserted into the EcoRV/XhoI site of pBCMarLR, cre-ating pBCMhspRed. Finally, the SacII site in pUC19-DmCn (Warren et al., 1996–97) was replaced with anXhoI site, and the 4.7kb XhoI fragment inserted into theXhoI site of pBCMhspRed to yield the final plasmid,pBCMhspRedcn (Fig. 1A).

pBac3xP3EGFP-MOS: This piggyBac vector containsthe Mos1 transposase gene necessary for remobilizationof the reporter construct under the regulatory control ofthe D. pseudoobscura hsp82 promoter and an EGFPmarker gene. The D. pseudoobscura hsp82 promoter hasbeen shown to be active in the germ-line of Ae. aegypti(Coates et al., 1995). The 3.2kb EcoRI/PstI fragmentfrom pKhsp82MOS (Coates et al., 1995) was inserted inthe EcoRI/PstI site of pSLfa1180fa (Horn and Wimmer,2000). The 3.5kb AscI fragment from the resulting plas-mid was then inserted into the AscI site of pBac(3xP3-EGFPafm) (Horn and Wimmer, 2000) to create the finalplasmid (Fig. 1B).

2.3. Transformation

Mosquito transformations were essentially performedas described by Coates et al. (1998). To create Mos1transposition-reporter strains a mixture ofpBCMhspRedcn (0.3µg/µl) and pKhsp82MOS(0.2µg/µl) was injected into preblastoderm khw embryos.Similarly, to create the Mos1 transposase-expressinghelper line a mixture of pBac3xP3EGFP-MOS(0.3µg/µl) and pBac/hs�SST (0.2µg/µl) (Thibault et al.,1999) was injected into preblastoderm khw embryos.Injected embryos were heat-shocked at 41 °C for 1 h16–24 h after injection and reared to adulthood. G0

adults were sexed and small families consisting ofapproximately five G0 males or females were mated toapproximately 10–15 khw individuals of the opposite sex.Transformed G1 progeny were identified by their red eyecolor in the case of the transposition reporter lines andthe presence of EGFP-expression in the eyes in the caseof the transposase-expressing (helper) line.

2.4. Remobilization strategy

The transposition-reporter strains, cn7 and cn16, werecrossed to the helper strain, pBacMOS. The resultingheterozygotes (F1) were either heat-shocked daily as lar-vae for 1 h at 41 °C until pupation, or reared at a constant28 °C, and the adults crossed in small pools of approxi-mately ten individuals of one sex to approximately 20khw individuals of the opposite sex. Adults emergingfrom this cross (F2) were screened for changes in eyecolor (lighter or darker than the parental eye color) twodays after emergence. Adults arising from these crosseswith eye colors differing from the parental eye color ofheterozygous cn7 or cn16 were crossed to khw to amplifythe putative transposition event. Once eggs (F3) hadbeen collected the red-eyed F2 parental mosquito wassacrificed for analysis by TE display.

2.5. Transposable Element display

Transposable Element (TE) display is a DNA finger-printing method similar to AFLP (Vos et al., 1995) andwas developed to monitor transposable element move-ment (Casa et al., 2000; Van den Broeck et al., 1998;Guimond et al., 2003). Single mosquito genomic DNA

856 R. Wilson et al. / Insect Biochemistry and Molecular Biology 33 (2003) 853–863

preparations were prepared using Protocol 48 in Ash-burner (1989). For analysis of the Mos1 element-con-taining transgenic lines, 50–100 ng of genomic DNAwas digested at 37 °C with 4U BfaI (New EnglandBiolabs) in 40µl of reaction buffer (50 mM NaCl, 10mM Tris-HCl pH7.9, 10 mM MgCl2, 1 mM dithiothrei-tol (DTT), and 100 µg/ml BSA). After 3 h, 60 pmolesof BfaI adapters (consisting of a dimer of the oligonucle-otides MseIa 5’ -GAG TCC TGA GTA GCA G-3’ andMseIb 5’ -TAC TCA GGA CTC AT-3’ ), 1U T4 DNAligase, 50 mM DTT and 5 mM ATP were added and thetotal volume increased to 50 µl with buffer. Therestriction/ligation reaction was allowed to continueovernight and then diluted 3x with 0.1x TE buffer. Pre-selective PCR was performed using 5µl of therestriction/ligation mixture as template and MseIa andMOS187r (5’ -TGT CCG CGT TTG CTC TTT ATT CG-3’ ) primers. Reaction conditions included an initial dena-turation step of 3 min at 95 °C followed by 25 cyclesof 15 s at 95 °C, 1 min at 60 °C and 1 min at 72 °Cwith a final 5 min elongation at 72 °C. The PCR productswere diluted ten times with 0.1x TE and 5µl used as atemplate for the selective PCR with Mse1a and the Cy5-labelled MOSCy5::46r (5’ -ACA ATC GAT AAA TATTTA CGT TTG C-3’ ) primers. The reaction conditionsincluded an initial denaturation step of 3 min at 95 °C,followed by five cycles of touchdown PCR consisting ofa denaturation step of 15 s at 95 °C followed byannealing for 1 min at a temperature that was reducedby one degree on each successive cycle beginning at 64°C and an extension step of 1 min at 72 °C. The touch-down phase of the reaction was followed by 25 cyclesof 15 s at 95 °C, 1 min at 60 °C and 1 min at 72 °Cwith a final elongation of 5 min at 72 °C.

Analysis of the piggyBac element-containing trans-genic line (helper line) was done using a similar protocolexcept genomic DNA was cut with MspI; MspI adaptors(consisting of a dimer of the oligonucleotides MspIa 5’ -GAC GAT GAG TCC TGA G-3’ and MspIb 5’ -TACTCA GGA CTC GC-3’ ) were used for the ligation; thepre-selective PCR was run using the primers MspIa andpiggyL1 (5’ -TAT GAG TTA AAT CTT AAA AGTCAC G-3’ ) for the left junction and MspIa and piggyR1(5’ -GTT GAA TTT ATT ATT AGT ATG TAA GTG-3’ ) for the right junction with reaction conditions of 3min at 95 °C followed by 25 cycles of 15 s at 95 °C,30 s at 54 °C and 1 min at 72 °C, finishing with 5 minat 72 °C; the selective PCR was run using primersMsp1a and Cy5-labelled piggyL2Cy5 (5’ -CAG TGACAC TTA CCG CAT TAC AAG C-3’ ) for the left junc-tion and Msp1a and Cy5-labelled piggyR2Cy5 (5’ -ATATAC AGA CCG ATA AAA ACA CAT GCG) for theright junction with reaction conditions of an initial 3 mindenaturation at 95 °C followed by five cycles of 15 s at95 °C followed by five cycles of touchdown PCR con-sisting of a denaturation step of 15 s at 95 °C followed

by annealing for 30 s at a temperature that was reducedby one degree on each successive cycle beginning at 59°C and an extension step of 1 min at 72 °C. The touch-down phase of the reaction was followed by 25 cyclesof 15 s at 95 °C, 30 s at 54 °C, and 1 min at 72 °C,finishing with a five min elongation at 72 °C.

PCR reaction products were separated on an 8% poly-acrylamide denaturing gel and viewed, after drying on apiece of 3 MM filter paper, on a Storm 860 optical scan-ner (Molecular Dynamics) using the red excitation wave-length of 635 nm. Bands of interest were excised fromthe gel, re-amplified using the selective PCR protocol,cleaned using Wizard PCR preps (Promega), andsequenced.

2.6. RT-PCR

Pools of ten mosquitoes were stored at –80 °C priorto RNA extraction. RNA extraction was perform usingan RNEasy kit (Qiagen) according to the manufacturer’sspecifications and resuspended in 30µl dH2O. Prior toPCR 8 mosquito equivalents of total RNA (24 µl) wastreated with 3 µl (1 unit/µl) of RQ1 RNAse-free DNAse(Promega) in 1 x RQ1 DNAse reaction buffer at 37 °Cfor 90 min. Treated samples were phenol/chloroformextracted and the RNA precipitated with sodium acetateand 100% ethanol at –20 °C overnight. The precipitatedRNA was pelleted, dried and resuspended in 40 µl ofdH2O.

Complementary DNA was produced from the purifiedRNA essentially as described (Vizioli et al., 2000) ina 60 µl reaction consisting of 12 µl 5xMuLV ReverseTranscriptase buffer (Promega), 3 µl of 40 mM dNTPs(10 mM each), 0.4 µl of 60 µM oligo-dT2 primers (IDT),1 µl of 50 U/µl RNAsin (PanVera), 0.8 µl of 200U/µlMuLV Reverse Transcriptase (Promega), 20µl (fourmosquito equivalents) RNA, 21.8 µl dH2O. The reactionmixture was heated for 5 min at 72 °C and then incu-bated at 37 °C for 90 min. The resulting cDNA was usedas a template in the following PCR.

Mariner cDNA was detected by performing a 20 µlPCR reaction containing 1x PCR buffer II (AppliedBiosystems), 0.2 mM dNTP, 2.5 mM MgCl2, 1unitAmpliTaq DNA polymerase (Applied Biosystems), 24pmoles of the primers MOS300f (5’TGT GAA AACGTG TGA ACG GTG G3’ ) and MOS1040r (5’GCGAAT AGG TGG TAA TCG GAT GG-3’ ), 1 µl cDNA.The cycle conditions included an initial denaturation stepof 3 min at 95 °C, followed by followed by 36 cyclesof 95 °C for 15 s, 60 °C for 15 s and 72 °C for 1 minwith a final elongation step of 72 °C for 5 min. Cytoplas-mic actin cDNA was detected by performing PCR simul-taneously under the conditions as described above exceptwith primers Actinf (5’ATT AAG GAG AAG CTGTGC TAC GTC-3’ ) and Actinr (5’CAT ACG ATC AGCAAT ACC TGG G-3’ ) and the reactions were removed

857R. Wilson et al. / Insect Biochemistry and Molecular Biology 33 (2003) 853–863

after 22 cycles. All PCR reaction products were separ-ated by electrophoresis in agarose.

2.7. Southern blotting

Genomic DNA (15–30 ug) was digested with eitherBamHI, NruI, SalI, or XhoI and size-fractionated on a0.8% agarose gel. Size-fractionated DNA was trans-ferred by capillary blotting to Duralon-UV membranes(Stratagene) using 10x SSC, prehybridized in QuikHybhybridization solution (Stratagene) for 6 h at 65 °C andhybridized with a 32P-labelled probe at 65 °C overnight(Fig. 1). Filters were washed repeatedly in 2x SSC, 0.1%SDS at room temperature and finally in 0.1x SSC, 0.1%SDS at 60 °C.

Hybridization signals were detected using a Storm 860phosphoimager (Molecular Dynamics).

3. Results

Three transgenic lines were created in the course ofthis study. Two lines (cn7 and cn16) were created withthe Mos1 vector MhspRedcn (Fig. 2B). Each line con-tained a single copy of the Mos1 vector based on theresults of the inheritance patterns of the cinnabar marker(Table 2), TE display (Table 1), and Southern blot analy-sis (Fig. 3A). DNA sequence analysis of the Mos1 inte-gration sites showed that the elements inserted into TA

Fig. 2. Transgenic Lines. A) pBacMOS. B) cn7 and cn16. khw is thenon-transgenic host strain and Orlando is a wild-type strain. “Normalcn16” refers to the parental eye color phenotype seen within the cn16line. “Light eyed cn16” refers to a putative transposition event arisingfrom a cross in which cn16 was the element being mobilized. Thisinsect has an eye color phenotype distinct from the parental insect.

target sites as is characteristic of Mos1 elements. In ahomozygous khw background, expression of the D. mel-anogaster cinnabar gene in both lines resulted in pig-mentation levels in the eye that were less than thoseobserved in wild-type individuals, which is typical ofthis marker gene in Ae. aegypti (Fig. 2). Line-specificeye-color phenotypes are typical with eye-color markergenes and reflect the common phenomenon of position-dependent transgene expression. Some intra-strain vari-ation in eye color was also observed and this was great-est among newly eclosed adults. However, by two dayspost eclosion intra-strain variation in eye color was mini-mal. The structural integrity of the Mos1 sequencespresent within the integrated vectors was determined byamplifying, cloning and sequencing the 420 bp of Mos1DNA on the left end of the vector and the 680 bp ofMos1 DNA on the right end of the vector. The sequencesof eight independently cloned PCR products weredetermined and no mutations were found. Based on thestructure of the integrated Mos1 elements they appearedcompetent to undergo transposition.

A single transgenic line (pBacMOS) was created withthe piggyBac vector pBac3xP3EGFP-MOS. This linecontained a single integration of the vector into a TTAAtarget site based on TE display (data not shown), South-ern blot analysis (Fig. 3B) and the pattern of inheritanceof the 3xP3EGFP marker gene. As expected with the3xP3 promoter, pBacMOS individuals had EGFP markergene expression confined to the larval and adult brainand optic stalk (Fig. 2A). Based on inheritance patternsthe pBac3xP3EGFP-MOS integration event was not ontothe same linkage group as either of the events producingreporter lines cn7 and cn16. The Mos1 transposasecoding region was amplified from the genomic DNA ofthe pBacMOS line, cloned and sequenced to assess itsintegrity. The sequence of eight independently clonedPCR products was determined and no mutations werefound indicating that this transgene should function as asource of active Mos1 transposase. Transcripts of Mos1transposase were readily detected in total RNA of adultsfollowing reverse transcription and PCR using transpos-ase-specific primers (Fig. 4). In males, Mos1 transcriptswere detected in the absence of heat shock induction (41°C for one hour) while expression in females was detect-able at the end of a one hour heat-shock but not in theabsence of heat shock (Fig. 4). Expression in both malesand females could be detected for at least 2 hours postheat-shock.

The above data indicate that the Mos1 elements andthe transposase helper gene were intact and functional.We tested this directly by examining the ability of theMos1 elements in lines cn7 and cn16 to be remobilizedin the presence of Mos1 transposase expressed from thepBac3xP3EGFP-MOS transgene. Remobilizationactivity was measured in both somatic and germ-linetissue. To examine somatic remobilization, hetero-

858 R. Wilson et al. / Insect Biochemistry and Molecular Biology 33 (2003) 853–863

Table 1Integration site analysis

Left ITR Integration site BLAST Description

Starting elementscn7 TTGTACACCTGG TAAAAAATTAAAACTA NSSa

cn 16 TTGTACACCTGG TAAAACCAGTGTTTTA gb|AF208670 Aedes aegypti Pony-Aa-A13 MITE

Germ-line event8.46 TTGTACACCTGG TATGTCAAGCACCCTG gb|AF168419 DsRed in MhspRedCn

Somatic events1.1A TTGTACACCTGG TATCAAGGGCCCAGTA NSS1.2A TTGTACACCTGG TATATGAAATAAACCT NSS1.6A TTGTACACCTGG TACTTGCTTTAAAATA Many hits SV40 polyA in MhspRedCn1.9A TTGTACACCTGG TAAGAGTACCCACCTA NSS1.9B TTGTACACCTGG TATATAGATAGATAGA NSS1.11A TTGTACACCTGG TAGAGGATCCTCAGTA NSS1.11B TTGTACACCTGG TATACAATTCCTCTGC NSS1.11C TTGTACACCTGG TAAAAGTATGAAAAGA NSS1.12A TTGTACACCTGG TAGTTGGTGTTACCGT gb|AF107695 Aedes aegypti Feilai family of SINES1.12B TTGTACACCTGG TATACAAACAAAAGTT NSS1.18A TTGTACACCTGG TAAAGAAAAATAAAAA gb|U56245 cinnabar in MshpRedCn1.19A TTGTACACCTGG TATGTTTTCCATACAA NSS1.20A TTGTACACCTGG TAAAAGCAATATTCTC NSS2.2B TTGTACACCTGG TAATAGTCGGATATCT NSS2.3B TTGTACACCTGG TACAGTTGAAAAAGCA gb|AF208671 Aedes aegypti Pony-Aa-A14 MITEb

2.3C TTGTACACCTGG TGTACAAGTATGAAAT emb|X78906 mariner sequence in MhspRedCn

a NSS, no significant similarity.b MITE DNA begins approximately 50 bp from the integration site. DNA immediately flanking the integration site shows no similarity to

anything in the database.

Table 2Germ-line transposition experiments

Non-recombinants Putative transpositionsHeat shocka Crossb White eyes Red eyes Lighter eyes(%) Darker eyes(%)

Yes pBacMOS/cn7 × khw 4731 4902 22 (0.2) 102 (1.0)pBacMOS/cn16 × khw 3674 3511 14 (0.2) 13 (0.2)

No pBacMOS/cn7 × khw 3160 3168 14 (0.2) 0 (0)pBacMOS/cn16 × khw 2275 2442 3 (0.1) 24 (0.5)

Total 53 139TE display analysis 53 139Confirmed jumps 0 1

a Larvae were heatshocked at 41 °C for 1 h daily until pupation.b Crosses were performed so that transposition could be detected in both male and female germlines.

zygotes were created by mating individuals from theMos1-containing lines (cn7 and cn16) with those fromthe Mos1 transposase expressing line, pBacMOS. Adultred-eyed progeny with EGFP expression in the brainwere analyzed by isolating genomic DNA and per-forming Mos1-specific TE display. Putative somatictransposition events were recognized as PCR productswith a different molecular weight and lower abundancefrom the PCR product arising from the parental element.Somatic transposition events were confirmed by isolat-ing, reamplifying, and sequencing the resulting PCR pro-

duct. Transposition was confirmed by detecting the pres-ence of the Mos1 inverted terminal repeat and flankingDNA that was different from the flanking DNA foundassociated with the parental element (Table 1). Approxi-mately 50% of the individuals analyzed (n=30) had evi-dence of at least one somatic remobilization event. Threeof the 16 confirmed Mos1 transposition events were intoAe. aegypti sequences that have been previously clonedand sequenced. Two integrations were into copies of thetransposable element Pony, a previously characterizedMITE, while the third integration was into Feilai, a

859R. Wilson et al. / Insect Biochemistry and Molecular Biology 33 (2003) 853–863

Fig. 3. Southern blot analysis of transgenic lines. A) Mos1-vectorcontaining lines probed with the probe shown in Fig. 1A. B) piggyBacvector-containing line probed with the probe shown in Fig. 1B. Theposition and sizes in kilobasepairs of molecular weight standards areshown to the left of each blot. The position and sizes of detected bandsare shown to the right of each blot. B, N, S, X refer to genomic DNAdigested with the restriction enzymes BamHI, NruI, SalI and XhoIrespectively. khw is the non-transgenic host strain.

Fig. 4. Mos1 transposase transcript detection by reverse transcriptasePCR. Cytoplasmic actin transcripts were detected as a control. Adultmales and females were analyzed separately. Analysis was preformedon insects that did not receive a heat shock (no hs), on insects immedi-ately after receiving a heat shock for one hour (t = 0 h), on insects 1and 2 h after receiving a heat shock for 1 h (t = 1 h, t = 2 h). khw isthe non-transgenic host strain. RT- refers to a sample from femalesafter a 2 h heat shock that was not treated with reverse transcriptase.

SINE (Table 1). Three of the integration events were intosequences found in the Mos1 vector itself and representintegrations of the element into itself (Table 1, Fig. 5).With one exception all transpositions were into TA tar-get sites (Table 1) although we have not examined theright end of the integrated Mos1 elements to confirm thepresence of a direct duplication of the target site. Thesedata confirm that the individual components of the Mos1transposable element system, i.e., the non-autonomous

Fig. 5. Location and orientation of integrations within the Mos1 vec-tor. L, R refer to the left and right ends of the Mos1 element respect-ively. Somatic and germ-line refer to where the transposition eventswere recovered.

vectors present in cn7 and cn16 and the Mos1 transpos-ase present in the helper line, were functional.

Germ-line activity was examined by mating hetero-zygotes not used for the assessment of somatic activitydescribed above with individuals from the mutant linekhw (Table 2). Crosses were performed with four combi-nations of parental insects; heterozygous males(cn7/pBacMOS or cn16/pBac MOS) mated with khw

females and heterozygous females (cn7/pBacMOS orcn16/pBac MOS) mated with khw males. Experimentswere performed in which the heterozygous parentalinsects either had, or had not been heat-shocked dailyduring their larval development at 41 °C for one hour.Of the 14,023 red-eyed adult progeny recovered fromthese crosses 192 had eye color phenotypes that differedfrom the parental eye color (either lighter or darker).These were provisionally considered as transpositionevents and were further analyzed by TE display to testfor the presence of a Mos1 element in a new location.Only one of 192 progeny analyzed by TE display con-tained a Mos1 element in a new location indicating thatthere was some (1.4%) intra-strain variation in the eyecolor phenotype. We suspect the cause of this intra-strainvariation to be due, in part, to variable expression of thecinnabar transgene coupled with variation in thenutritional status of the insects. The eye color of the pro-geny containing the transposition event was darker thanthe parental eye color and the genotype was confirmedby PCR to be cn16; MOS-. Because the individual con-taining the transposition event was MOS-, i.e. it did notcontain the hsp82-MOS transposase gene, there was nopossibility for the recovered transposition event to havearisen in the soma. The germ-line event was similar toseveral events recovered from the soma in that it hadbeen inserted into the Mos1 vector (Table 1, Fig. 5).

4. Discussion

The Mos1 mariner transposable element has provento be an effective tool for the creation of transgenic Ae.aegypti and, based on the results of this study, theelement shows a very high degree of stability in the gen-ome of this species following integration. When Ae.aegypti were created that were heterozygous for both thecinnabar-marked Mos1 element and the Mos1 transpos-ase-expressing transgene, evidence for infrequent Mos1transposition was obtained. Only one germ-line transpo-sition event was recovered from over 14,000 progeny(0.01%) and few somatic transposition events weredetected. We are unable to estimate the frequency oftransposition in the soma but a comparison of the TEdisplay results obtained in this study with those reportedby Guimond et al. (2003) in their study of Hermes move-ment in the soma of D. melanogaster reveal somenotable differences. TE display of somatically active

860 R. Wilson et al. / Insect Biochemistry and Molecular Biology 33 (2003) 853–863

Hermes elements in D. melanogaster revealed many(approximately 50–100) somatic transposition events perindividual (see Fig. 2 in Guimond et al., 2003) while, inthis study of Mos1 movement, less than one event perindividual was observed suggesting that the rate ofsomatic movement is very low. The observed inactivityof Mos1 in the germ-line could not be attributed to theabsence of Mos1 transposase expression in theMhspRedcn/pBacMOS heterozygotes because thehsp82-MOS transposase transgene present in thepBacMOS helper line was the same transgene used as asource of functional Mos1 transposase during the cre-ation of the transgenic lines cn7 and cn16. Creatingtransgenic insects by microinjection of preblastodermembryos with vector and helper plasmids requires thetransient expression of the helper transgene containingthe transposase ORF in the germ-line. Therefore,because we successfully created transgenic Ae. aegyptiwith Mos1 using the hsp82-MOS helper gene at fre-quencies similar to those reported by others(approximately 2–5%), we conclude that hsp82-MOS isexpressed in the germ-line. Therefore, the transgenic linepBacMOS was a functional helper line and the stabilityof Mos1 in lines cn7 and cn16 was not due to the failureto express functional Mos1 transposase in the germ-line.

The stability we observed in Ae. aegypti was verysimilar to the reported stability of this element in D. mel-anogaster (Lidholm et al., 1993; Lohe et al., 1995;Lozovsky et al., 2002). Lozovsky et al. (2002) suggestthat Mos1 mobility is highly dependent upon criticalspacing of sub-terminal sequences and the inverted ter-minal repeats. All Mos1 vectors tested by Lozovsky etal. (2002) had highly reduced levels of mobility in vivorelative to an unmodified element. Only a vector con-sisting of two almost complete elements oriented asdirect repeats and flanking exogenous DNA could beremobilized at detectable levels. The observed levels ofactivity were much lower that those observed with unin-terrupted elements. The “critical spacing” hypothesis isconsistent with the observations of Lohe and Hartl(2002) who, based on the excision activity of internallydeleted Mos1 elements, concluded that there were threecritical regions located in the central part of the element.These critical regions are located at least 200 bp fromthe ends of the element. The presence of sequences withstructural significance located far from the terminalsequences has not been reported for other Class IIelements. While the “critical sequence/critical spacing”model is plausible based on data collected from experi-ments performed in vivo, it is somewhat inconsistentwith the findings of Tosi and Beverly (2000) who dem-onstrated that only 64 nucleotides of the left end and 33nucleotides of the right end of Mos1 are essential for invitro transposition activity of an approximately 1.1 kbvector. Furthermore, the rate of transposition of the mini-mal Mos1 element in vitro was only 2 fold less than

that of a vector containing essentially the complete Mos1element, suggesting that sequences flanking the terminiof Mos1 are not critical for function. One way to resolvethese apparently conflicting data is to suggest that hostfactors play an important role in the transposition pro-cess in vivo and influence the relative importance of cissequences in the Mos1 transposition process. The Mos1vector used in our experiments contained approximately300 bp from both ends of the Mos1 element. This vectorcontains only approximately 100 bp of the 350 bp “criti-cal region I” and 60 bp of the 150 bp “critical regionIII” and it completely lacks “critical region II” as definedby Lohe and Hartl (2002). Hence the inefficient mobiliz-ation of Mos1 vectors in Ae. aegypti observed in ourexperiments may be because of the lack of criticalsequences and/or sub-optimal spacing of the inverted ter-minal repeat-sequences.

It is also possible that the size of the vector contrib-uted to its reduced mobility. Class II transposableelements that are currently used as gene vectors aresensitive to the insertion of exogenous DNA. As thelength of the element increases mobility decreases. Forexample, a 6 kb Himar element, a mariner-like elementfrom Haematobia irritans, had only 10% of the transpo-sitional activity of a 1.5 kb element in vitro. How thesein vitro size requirements relate to size requirements invivo is not clear although the negative correlationbetween an element’s size and its activity seems gener-ally true.

What is somewhat puzzling about the Mos1 system isthe apparent discrepancy between the frequencies ofgerm-line transformation compared with the frequencyof mobilization of integrated vectors. Lozovsky et al.(2002) speculated that unspecified differences in theMos1 elements as a function of them being located onchromosomes versus plasmids might account for theapparent discrepancy in mobility properties within thesetwo contexts. Differences might include the degree ofsupercoiling and the state of compaction of the chroma-tin, both of which have been shown in some cases toinfluence transposable element movement (Cost et al.,2001; Wang and Harshey, 1994).

Until now the question of whether the post-integrationbehavior of Mos1 in D. melanogaster reflected a generalmobility characteristic of this element or whether it wasa reflection of a specific (D. melanogaster) host effectremained unresolved (Lozovsky et al., 2002). The resultsreported here support the hypothesis that post-integrationstability of Mos1 is a general characteristic of theelement and does not reflect a host specific influenceon the behavior of the element. Consequently we cananticipate a high degree of Mos1 gene vector stability inother insect species.

Finally, the pattern of transpositions observed in thisstudy may provide us with some insight into the mech-anism of Mos1 transposition. Mos1, like most other

861R. Wilson et al. / Insect Biochemistry and Molecular Biology 33 (2003) 853–863

transposable elements studied, increases in copy numberupon being introduced into a genome. Copy numbersrange from tens to ten thousands (reviewed by Hartl etal., 1997). How this element increases to such high copynumbers while undergoing conservative (cut-and-paste)transposition is not entirely clear. One mechanism is toemploy template-directed gap repair (TDGR) followingelement excision. TDGR following transposable elementexcision was first described in the D. melanogaster Pelement system (Gloor et al., 1991; Nassif et al., 1994)and involves repairing gaps created by the excision ofP elements by a DNA polymerase-dependent copyingmechanism that uses another copy of the element else-where in the genome as a template. Until recently therewas little evidence for TDGR following Mos1 excisionhowever, Lohe et al. (2000) recently estimated that 30%of all excisions of Mos1 in D. melanogaster are repairedby this mechanism. The data reported here suggest anadditional mechanism. We suggest the data presentedhere reflect mariner’s propensity to transpose during orafter S phase but before the end of mitosis. Of the trans-position events recovered and analyzed in this study,approximately 25% were instances where the elementtransposed into a copy of itself. However, in all experi-ments the element was in a heterozygous condition andin diploid cells mitosis is the only time when a secondcopy of the element would be present to serve as targetfor Mos1 transposition. Therefore, it appears that Mos1transposes during mitosis and this provides the elementwith the opportunity to replicate and increase in copynumber (Fig. 6). Transposition during mitosis has beenreported for the Ac/Ds element of maize based on pat-terns of transposition (Fedoroff, 1989; Greenblatt andBrink, 1963). Recently, Timakov et al. (2002) suggestedthat P element transposition in D. melanogaster occursduring meiosis. Because not all cells in an adult mos-quito are diploid it remains possible that the somatictransposition events into the Mos1 vector occurred in apolyploid cell in which there would be many copies ofthe Mos1 element to serve as targets. We think this hasnot occurred because the TE display method does not

Fig. 6. Transposition during mitosis accounts for the patterns oftransposition observed in this experiment and can lead to an increasein element copy number if the element transposes ahead of the repli-cation fork.

have a level of sensitivity permitting the genotyping ofa single cell in an adult. Furthermore, the germ-linetransposition event in which the Mos1 element was usedas a target only seems explicable if transpositionoccurred during mitosis since at no other time is thegerm-line polyploid. Transposition of the vector into acopy of itself on a sister chromatid may reflect either apropensity of the element to integrate locally or theinfluence of sequences contained within the vector in aprocess referred to as homing. Local hopping and hom-ing have been described for a number of Class II trans-posable elements (Tower et al., 1993; Hama et al., 1990).The data presented here does not permit us to distinguishbetween those possibilities.

Our results have implications for the use of Mos1 asa functional genomics tool. Various gene finding techno-logies that have been successfully developed for D. mel-anogaster have been based on the transposable elementsP, hobo and piggyBac (Bellen et al., 1989; Smith et al.,1993; Berg and Spradling, 1991; Cooley et al., 1988;Horn et al., 2003). A successful insertional mutagenesissystem based on mobilizing a single P element provedto be useful even though only 8% of the germ-linestested yielded a transposition event (Cooley et al., 1988).More efficient systems have been described in whichapproximately 40% or more of the germ-lines yielded atransposition event (Berg and Spradling, 1991; Horn etal., 2003). Smith et al. (1993) reported 4–16% of thegerm-lines tested yielded a hobo transposition event aspart of a hobo-based enhancer trap system. Consideringthe number of progeny expected from each cross it hasbeen estimated that between 2 and 10% of all progenycontain a unique transposition event and this representsa level of activity that should serve as a guide forassessing all future efforts to develop similar techno-logies in non-drosophilid species. The data presentedhere indicate that Mos1 is not a good candidate elementfor developing gene-finding tools such aspromoter/enhancer trapping and transposon tagging sys-tems in Ae. aegypti and perhaps other insects. On theother hand, if a high level of post-integration stability isdesired, then Mos1 is an appropriate element to considerin insects. The potential of this element to be lost byexcision or to transpose to new locations in the genomeis very low, even in the presence of functional Mos1transposase. The risk of being mobilized by a relatedtransposase (cross-mobilized) as has been described forhobo and Hermes (Sundararajan et al., 1999) or of beingtransferred horizontally by a transposition-dependentmechanism seems minimal and this is consistent withdirect tests of Mos1 cross mobilization in vitro (Lampeet al., 2001). Therefore, Mos1 vectors, as they are cur-rently configured and used, are essentially suicide vec-tors in insects in that they essentially become dysfunc-tional upon integration.

862 R. Wilson et al. / Insect Biochemistry and Molecular Biology 33 (2003) 853–863

Acknowledgements

This work was supported in part by grants from theNational Institutes of Health (AI45743) (D.A.O’B.) andThe Wellcome Trust (M.J.L.)

References

Ashburner, M., 1989. Drosophila: A Laboratory Manual. Cold SpringHarbor Laboratory Press, Cold Spring Harbor.

Atkinson, P.W., Pinkerton, A.C., O’Brochta, D.A., 2001. Genetictransformation systems in insects. Ann. Rev. Entomology 46,317–346.

Bellen, H.J., O’Kane, C.J., Wilson, C., Grossniklaus, U., Pearson,R.K., Gehring, W.J., 1989. P-element-mediated enhancer detection:a versatile method to study development in Drosophila. GenesDevelop. 3, 1288–1300.

Berg, C.A., Spradling, A.C., 1991. Studies on the rate and site-speci-ficity of P element transposition. Genetics 127, 515–524.

Blackman, R.K., Macy, M., Koehler, D., Grimaila, R., Gelbart, W.M.,1989. Identification of a fully-functional hobo transposable elementand its use for germ-line transformation of Drosophila. EMBO J8, 211–217.

Casa, A.M., Brouwer, C., Nagel, A., Wang, L., Zhang, Q., Kresovich,S., Wessler, S.R., 2000. The MITE family Heartbreaker (Hbr):Molecular markers in maize. Proc. Natl. Acad. Sci. USA 97,10083-10089.

Coates, C.J., Jasinskiene, N., Miyashiro, L., James, A.A., 1998. Mari-ner transposition and transformation of the yellow fever mosquito,Aedes aegypti. Proc. Natl. Acad. Sci. 95, 3748-3751.

Coates, C.J., Truney, C.L., Frommer, M., O’Brochta, D.A., Warren,W.D., Atkinson, P.W., 1995. The transposable element mariner canexcise in non-drosophilid insects. Molec. Gen. Genet. 249, 246–252.

Cooley, L., Kelly, R., Sprading, A., 1988. Insertional mutagenesis ofthe Drosophila genome with single P elements. Science 239,1121–1128.

Cornel, A.J., Benedict, M.Q., Salazar-Rafferty, C., Howells, A.J., Col-lins, F.H., 1997. Transient expression of the Drosophila melanogas-ter cinnabar gene rescues evy color in the white eye (WE) strainof Aedes aegypti. Insect. Biochem. Mol. Biol. 27, 993–997.

Cost, G.J., Golding, A., Schlissel, M.S., Boeke, J.D., 2001. TargetDNA chromatinization modulates nicking by L1 endonuclease.Nuc. Acid Res. 29, 573.

Fadool, J.M., Hartl, D.L., Dowling, J.E., 1998. Transposition of themariner element from Drosophila mauritiana in zebrafish. Proc.Natl. Acad. Sci. USA 95, 5182–5186.

Fedoroff, N.V., 1989. Maize transposable elements. In: Berg, D.E.,Howe, M.M. (Eds.), Mobile DNA. American Society for Micro-biology, Washington D.C., pp. 375–411.

Gloor, G.B., Nassif, N.A., Johnson-Schlitz, D.M., Preston, C.R.,Engels, W.R., 1991. Targeted gene replacement in Drosophila viaP element-induced gap repair. Science 253, 1110–1117.

Greenblatt, I.M., Brink, R.A., 1963. Transposition of Modulator inmaize into divided and undivided chromosome segments. Nature292, 860–862.

Gueiros-Fiho, F.J., Beverley, S.M., 1997. Trans-kingdom transpositionof the Drosophila element mariner within the protozoan Leish-mania. Science 276, 1716–1719.

Guimond, N., Bideshi, D.K., Pinkerton, A.C., Atkinson, P.W.,O’Brochta, D.A., 2003. Patterns of Hermes Transposition in Droso-phila melanogaster. Molec. Gen. Genet. 268, 779–790.

Hama, C., Ali, Z., Kornberg, T.B., 1990. Region-specific recombi-nation and expression are directed by portions of the Drosophilaengrailed promoter. Genes Dev. 4, 1079–1093.

Handler, A.M., James, A.A., 2000. Insect Transgenesis. CRC Press,Boca Raton.

Handler, A.M., McCombs, S.D., Fraser, M.J., Saul, S.H., 1998. Thelepidopteran transposon vector, piggyBac, mediates germ-linetransformation in the Mediterranean fruit fly. Proc. Natl. Acad. Sci.USA 95, 7520–7525.

Hartl, D.L., Lohe, A.R., Lozovskaya, E.R., 1997. Modern thoughtson an ancyent marinere: function, evolution, regulation. Ann. Rev.Genet. 31, 337–358.

Horn, C., Wimmer, E.A., 2000. A versatile vector set for animal trans-genesis. Dev. Genes Evol. 210, 630–637.

Horn, C., Offen, N., Nystedt, S., Haecker, U., Wimmer, E.A., 2003.piggyBac-based insertional mutagensis and enhancer detection asa tool for functional insect genomics. Genetics In press.

Jasinskiene, N., Coates, C.J., Benedict, M.Q., Cornel, A.J., Rafferty,C., Salazar-Rafferty, C., James, A.A., Collns, F.H., 1998. Stable,transposon mediated transformation of the yellow fever mosquito,Aedes aegypti, using the Hermes element from the housefly. Proc.Natl. Acad. Sci. 95, 3743–3747.

Lampe, D.J., Robertson, H.M., 1996. A purified mariner transposaseis sufficient to mediate transposition in vitro. EMBO J. 15,5470–5479.

Lampe, D.J., Walden, K.K., Robertson, H.M., 2001. Loss of transpos-ase-DNA interaction may underlie the divergence of mariner fam-ily transposable elements and the ability of more than one marinerto occupy the same genome. Molec. Biol. Evol. 18, 954–961.

Lampe, D.J., Walden, K.K.O., Sherwood, J.M., Robertson, H.M.,2000. Genetic engineering of insects with mariner transposons. In:Handler, A.M., James, A.A. (Eds.), Transgenic Insects: Methodsand Applications. CRC Press LLC, Boca Raton, 397.

Lidholm, D., Lohe, A., Hartl, D., 1993. The transposable element mari-ner mediates germline transformaton in Drosophila melanogaster.Genetics 134, 859–868.

Lohe, A.R., Hartl, D.L., 2002. Efficient mobilization of mariner invivo requires multiple internal sequences. Genetics 160, 519–526.

Lohe, A.R., Lidholm, D.A., Hartl, D.L., 1995. Genotypic effects,maternal effects and grand-maternal effects of immobilized deriva-tives of the transposable element mariner. Genetics 140, 183–192.

Lohe, A.R., Timmons, C., Beerman, I., Lozovskaya, E.R., Hartl, D.L.,2000. Self-inflicted wounds, template-directed gap repair and arecombination hotspot: Effects of the mariner transposase. Genetics154, 647–656.

Loukeris, T.G., Arca, B., Livadaras, I., Dialektaki, G., Savakis, C.,1995. Introduction of the transposable element Minos into the germline of Drosophila melanogaster. Proc. Natl. Acad. Sci. 92,9485–9489.

Lozovskaya, E.R., Nurminsky, D.I., Hartl, D.L., Sullivan, D.T., 1996.Germline transformation of Drosophila virilis mediated by thetransposable element hobo. Genetics 142, 173–177.

Lozovsky, E.R., Nurminsky, D., Wimmer, E.A., Hartl, D.L., 2002.Unexpected stability of mariner transgenes in Drosophila. Genetics160, 527–535.

Mamoun, C.B., Guzman, I.Y., Beverly, S.M., Goldberg, D.E., 2000.Transposition of the Drosophila element mariner within the humanmalaria parasite Plasmodium falciparum. Molec. Bio. Parasitology110, 405–407.

Nassif, N., Penney, J., Pal, S., Engels, W.R., Gloor, G,B., 1994.Efficient copying of nonhomologous sequences from ectopic sitesvia P-element-induced gap repair. Molec. Cell Biol. 14, 1613–1625.

O’Brochta, D.A., Atkinson, P.W., 1996. Transposable elements andgene transformation in non-drosophilid insects. Insect Biochem.Mole. Biol. 26, 739–753.

Plasterk, R.H., Izsvak, Z., Ivics, Z., 1999. Resident aliens: TheTc1/mariner superfamily of transposable elements. Trends Genet.15, 326–332.

863R. Wilson et al. / Insect Biochemistry and Molecular Biology 33 (2003) 853–863

Robertson, H.M., 1993. The mariner transposable element is wide-spread in insects. Nature 362, 241–245.

Robertson, H.M., 1995. The Tc1-mariner superfamily of transposonsin animals. J. Insect Physiology 41, 99–105.

Robertson, H.M., MacLeod, E.G., 1993. Five major subfamilies ofmariner transposable elements in insects, including the Mediter-ranean fruit fly, and related arthropods. Insect Molec. Biol. 2,125–139.

Rubin, G.M., Spradling, A.C., 1982. Genetic transformation of Droso-phila with transposable element vectors. Science 218, 348–353.

Sherman, A., Dawson, A., Mather, C., Gilhooley, H., Li, Y., Mitchell,R., Finnegan, D., Sang, H., 1998. Transposition of the Drosophilaelement mariner into the chicken germ line. Nat. Biotech. 16,1050–1053.

Smith, D., Wohlgemuth, J., Calvi, B.R., Franklin, I., Gelbart, W.M.,1993. hobo enhancer trapping mutagenesis in Drosophila revealsan insertion specificity different from P elements. Genetics 135,1063–1076.

Sundararajan, P., Atkinson, P.W., O’Brochta, D.A., 1999. Transpos-able element interactions in insects: Crossmobilization of hobo andHermes. Insect Molec. Biol. 8, 359–368.

Thibault, S.T., Luu, H.T., Vann, N., Miller, T.A., 1999. Precise exci-sion and transposition of piggyBac in pink bollworm embyos.Insect Molec. Biol. 8, 119–124.

Timakov, B., Liu, X., Turgut, I., Zhang, P., 2002. Timing and targetingof P-element local transposition in the male germline cells of Dro-sophila melanogaster. Genetics 160, 1011–1022.

Tosi, L.R.O., Beverly, S.M., 2000. Cis and trans factors affectingMos1 mariner evolution and transposition in vitro, and its potentialfor functional genomics. Nuc. Acid Res. 28, 784–790.

Tower, J., Karpen, G.H., Craig, N., Spradling, A.C., 1993. Preferentialtransposition of Drosophila P elements to nearby chromosomalsites. Genetics 133, 347–359.

Van den Broeck, D., Maes, T., Sauer, M., Zethof, H., De Keukeleire,P., D’Hauw, M., Van Montagu, M., Gerats, T., 1998. Transposondisplay identifies individual transposable elements in high copynumber lines. Plant J. 13, 121–129.

Vizioli, J., Bulet, P., Charlet, M., Lowenberger, C., Blass, C., Muller,H.M., Dimopoulos, G., Hoffmann, J., Kafatos, F.C., Richman,R.A., 2000. Cloning and analysis of a cecropin gene from themalaria vector mosquito, Anopheles gambiae. Insect Molec. Biol.9, 75–84.

Vos, P., Hogers, R., Bleeker, M., Reijans, M., van de Lee, T., Hornes,M., Frijters, A., Pot, J., Peleman, J., Kuiper, M., Zabeau, M., 1995.AFLP: a new technique for DNA fingerprinting. Nuc. Acid Res.23, 4407–4414.

Wang, Z., Harshey, R.M., 1994. Crucial role for DNA supercoiling inMu transposition: a kinetic study. Proc. Natl. Acad. Sci. USA 91,699–703.

Warren, W.D., Palmer, S., Howells, A.J., 1996-97. Molecular charac-terization of the cinnabar region of Drosophila melanogaster:identification of the cinnabar transcription unit. Genetica 98,249–262.