Circularized Ac/Ds Transposons: Formation, Structure and Fate · circularized transposons. The...

Circularized Ac/Ds Transposons: Formation, Structure and Fate

Vera Gorbunova and Avraham A. Levy

Plant Genetics Department, The Weizmnnn Institute of Science, Rehovot, 76100 Israel Manuscript received August 23, 1996

Accepted for publication January 7 , 1997

ABSTRACT The maize Ac/Ds transposable elements are thought to transpose via a cut-and-paste mechanism, but

the intermediates formed during transposition are still unknown. In this work we present evidence that circular Ac molecules are formed in plants containing actively transposing elements. In these circles, transposon ends are joined head-to-head. The sequence at the ends’ junction is variable, containing small deletions or insertions. Circles containing deleted Ac ends are probably unable to successfully reintegrate. To test the ability of circles with intact transposon ends to integrate into the genome, an artificial Ds circle was constructed by cloning the joined ends of Ac into a plasmid carrying a plant selectable marker. When such a circular Ds was introduced into tobacco protoplasts in the presence of Ar-trdnsposase, no efficient transposase-mediated integration was observed. Although a circular transposition intermediate cannot be ruled out, the findings of circles with deleted transposon ends and the absence of transposase-mediated integration of the circular Ds suggest that some of the joined-ends- carrying elements are not transposition intermediates, but rather abortive excision products. The formation of Ac circles might account for the previously described phenomenon of Ac-loss. The origin of Ar circles and the implications for models of Ac transposition are discussed.

T HE maize Ar/Ds elements were the first discovered transposons ( MCCLINTOCK 1948). Since then,

transposable elements of a similar type have been found in a wide range of prokaryotic and eukaryotic species. Extensive studies provided a detailed description of the transposition process for bacterial elements such as Tn 7 and TnlO (for reviews see SAEDLER and GIERL 1996), but the precise mechanism by which Ac/Ds elements move from donor to recipient site remains unclear.

Ac belongs to the DNA-DNA type of transposable elements (FINNEGAN 1992). It is flanked by 11-bp terminal inverted repeats and generates an 8-bp target site duplication upon insertion (MULL~ER-NEUMANN et al. 1984; POHLMAN et al. 1984; SUTTON et al. 1984). Ac encodes transposase, a protein required for transposition (KUNZE et al. 1987). Ds elements are considered mu- tated Ac’s that retain the terminal Ac regions required for transposition but contain either deletions or insertions in their internal part that abolish transposase pro- duction (for review see FEDOROFF 1989). Ac transposes autonomously, whereas Ds transposes only in the presence of Ac (MCCLINTOCK 1951) or Ac transposase. It has been shown that Ac is active in plant species other than maize (reviewed in KUNZE 1996), when introduced into the genome via Agrobacterium-mediated transformation. It was also demonstrated that, following protoplast transformation, Ds can transpose from extrachromosomal DNA vectors into plant chromosomes in the presence of transposase ( HOUBA-HERIN et al. 1994; SUCI- MOT0 et nl. 1994).

Corresponding author; Avraham A. Levy, Plant Genetics Department, The Weizmann Institute of Science, Rehovot, 76100 Israel, E-mail: Iple\[email protected]

Genetics 145: 1161-1169 (April, 1997)

It is thought that Ac transposes by a conservative (cut- and-paste) process in which the element excises from a donor site and reinserts into a recipient site (GREENBLATT and BRINK 1962; CHEN et al. 1992). Yet, the intermediates formed in the course of transposition are unknown. One possibility, proposed for Ac (GREENBL4TT 1984) and for Tarn3 (ROBBINS et al. 1989), is that Ac transposes without extrachromosomal intermediates. According to this model, one transposon end is cut first and the second transposon end is cut only after the first end is ligated to the recipient site. Alterna- tively, extrachromosomal intermediates might be formed during transposition. Extrachromosomal circular DNA molecules were identified for the Mu1 transposon of maize (SUNDARESAN and FREELING 1987), and extrachromosomal linear molecules were found for the Tat1 element of Arabidopsis thaliana (PELEMAN et al. 1991) as well as for Tc3 of Cnenorhabditis e lepns (VAN LUENEN et al. 1993). For the Tcl element of C. elegans, both linear and circular forms were detected (ROSE and SNUTCH 1984; RUAN and EMMONS 1984). However, the ability of extrachromosomal putative intermediates to integrate into the genome has not been tested for any eukaryotic transposon. By contrast, linear extrachromosomal species that were detected for the bacterial transposons Tn 7 (B’4INTON et al. 199 1 ) and Tn 10 ( HANI- FORD et al. 1991; BENJAMIN and KLECKNER 1992) were shown to integrate into the chromosome, as expected for true transposition intermediates.

In this work, we show that joined Ac ends are formed in plants containing active Ac elements. The sequence at the junction of the transposon ends is variable, containing small deletions or insertions. Several pieces of

1162 V. Gorbunova and A. A. Levy

evidence indicate that joined Ac ends are derived from circularized transposons. The ability of transposon circles to reintegrate into the genome was examined using a synthetic Ds circle transformed into plant protoplasts. Analysis of the transformed plants failed to identify insertions with the pattern characteristic of Ac transposition. The implications of these findings on the origin of Ac circles and on models for Ac transposition are discussed.

MATERIALS AND METHODS

Plant material: The transgenic tobacco plants used for detection of joined Ac ends consisted of T’L plants of Nicotiana tabacum var. Xanthi. T2 plants were obtained from independent T1 transformants, each one containing a single T-DNA insertion, as deduced from the segregation of T2 plants. Transformation was performed either with the Ac-carrying construct pAGS4411 (DOONER et al. 1991) or the DJ-carrying construct pAGS4081 (KELLER et al. 1993) by an Agrobacte- rium-mediated procedure (HORSCH et al. 1985). Constructs were kindly provided by H. DOONER. Transposition activity in these plants was tested on streptomycin-containing media as described (JONES rt al. 1989). For protoplast transformation experiments, two types of N. tabacum var. Samsun plants were used: either wild type or transgenic plants carrying the pMFOO6 construct (FRIDLENDER et al. 1996), containing the Ac transposase open reading frame (ORF) under the control of the cauliflower mosaic virus 35s promoter. Two maize lines were used as Ac/Ds sources. One line, given by the maize cooperative, carried Acwithin the Plocus (P,-VV/P,-W). The other line, given by V. WALBOT, carried a Ds2 element at the Bz2 locus and an active Ar element in the background as deduced by the aleurone variegation of the bz2: : Ds2 allele (THERES et al. 1987).

Plant DNA extraction: DNA was isolated from young leaves of tobacco and maize immature cobs as described (DELIA- PORTA et al. 1983). To extract DNA from protoplasts the same protocol was used with the following modifications: the volume of all buffers was scaled down to perform the procedure in an Eppendorf tube, and the isopropanol precipitation step was omitted.

PCR and IPCR, cloning of the PCR products and sequencing: All PCR amplifications were performed with a cycle of 1 min denaturation at 94”, 1 min of annealing at 55”, and 1 min of extension at 72” repeated 30 times. Sequencing was done by Sanger’s method using an Applied Biosystems 373A DNA sequencer. Forjoined ends detection, - 1 pg of genomic DNA was used as a template for two nested rounds ofamplifi- cation, as shown in Figure 1, with primers 2 (5“GGCGGC- GTGTGAATGTGT-3’) and 3 (5’-CTAAGTGATCCAGAT- GTGAGJ’) in the first round. In the second round, 1.5 pl of the reaction from the first round was used as a template for amplification with primers 1 (5’-TCCCGAATTAGAAAA- TACGGJ‘) and 4 (5’-GTGTGCTCCAGATTTATATG3‘). PCR products were cloned as follows: clone 3 (Table 1) was obtained by direct cloning of the PCR product into the pCRII TA cloning kit (Invitrogen); all other clones were obtained by digestion of the PCR products with BamHI and Pad, and cloning into the corresponding sites of the pNEB193 vector (New England Biolabs). The resulting plasmids were sequenced with primers 5 (5’-GTTTTTTACCTCGGGTCG3’) and 13 (5’-TTTCGTTTCCGTCCCCCAAGTTMTA-3’).

To amplify empty donor sites following Ds excision from pVG5, two nested rounds of PCR were done with standard T3 and T7 primers in the first round, and in the second

round, with two primers from the Ac-flanking region of the maize Waxy gene: W1 (5”GAGGCTGTACGTGGGGCGTTGC- 3’) and W2 (5”AGTGGATCCCCCGCGCGGCC-S’). Resulting PCR products were cloned directly into a pGEM-T vector (Pro- mega) and sequenced with the T7 primer.

For IPCR 1 pg of genomic DNA was digested with one-sixth unit of SnuSA, extracted with phenol-chloroform, ethanol pre- cipitated, and circularized in a total volume of 500 p1 with 10 units of T4 DNA ligase. After ethanol precipitation, one-half of the reaction was used to amplify the 5’ end of the transposon i n two nested rounds of PCR with primers 5 and 6 (5’- CCAAAACGGAACGGAAACGGG3‘) in the first round, and primers 7 (5”ACTAACAAAATCGGTTATACG-J‘) and 8 (5‘- AGTTTCCCGACCGTTTCACC-3’) in the second. The 3‘ end of the element was amplified with primers 9 (5“ACCGGT- AACGAAAACGAACGJ’) and 10 (5”TTTCGTTTCCGTCCC- GCAAGTTMTA-J‘), in the first round, and primers 11 (5’- CGGGATAAATACGGTAATCG-3’) and 12 (5”GAAAATGAA- AACGGTAGAGGS’) in the second round. IPCR products were cloned directly into the pGEM-T vector (Promega) and sequenced with the primers 7 and 12 for the 5’ and 3’ transposon ends, respectively.

Plasmids: Plasmid pVG2 was constructed as follows: the Ac- containing 4.8-kb KpnI-BamHI fragment of pAGS4411 (DOONER rt al. 1991) was subcloned into the corresponding sites of pBluescriptIIKS (Strategene), and Ac was converted into Ils by deletion of the internal 1.6-kb Hind111 Acfragment. Plasmid pNT804 (HOLIBA-HERIN ri al. 1990), the transposase producer, was kindly provided by Dr. N. HOURA-HERIN. Plas- mid pVG6 was derived from pNT804 by deleting a 3-kb BamHI fragment, containing the transposase ORF. The synthetic Ds circle (Figure 3A) was constructed as follows: the PCR fragment containing joined Ac ends (clone 3, Table 1) was subcloned from the pCRII vector (Invitrogen) into the pRWl plasmid using the KpnI and XbuI sites. pRWl is a pBluescript1- IKS (Stratagene) derivative containing the 2.3-kb ClaI-Sac11 fragment from pGA492 (AN 1987); the latter carries the plant kanamycin-resistance gene under the nopaline syntase promoter. pVG5 is schematically described in Figure 4A. It contains a Ds element cloned into the KpnI-BamHI sites of Blues- cript 11%. The Ds element contains Ac 5’ end, derived from an -0.3-kb KpnI-MscI fragment from pAGS4411, followed by the plant kanamycin-resistance gene of pRW1, a bacterial chloramphenicol resistance gene isolated as a 2.7-kb AccI fragment from pACYC184 (New England Biolabs), and as a 3’ Ac end, the PacI-BamHI fragment from pAGS4411.

Protoplast transformation and plant regeneration: Proto- plasts were isolated from sterile leaves (AVIV and GALUN 1985). Polyethylene glycol-mediated transformation was done as described (VARUI et al. 1990) with some modifications: plasmids were uncut and calf thymus DNA was not added. For stable transformation aliquots of 10” protoplasts were cotransformed with 5.4 pmol of Ds circle and either 2.7 pmol of pNT804 or 2.7 pmol of pVG6. After transformation, protoplasts were resuspended in 0.5 M sucrose-containing VKM medium (Awv and GAI.UN 1985) at a density of lo” per ml, and kept for 4 days in the dark, and 2 days at low light at 25”. The protoplasts were then transferred under light, and the sucrose concentration was gradually reduced to 0.2 M within 10 days. Sixteen days after transformation microcalli were em- bedded in 0.6% Seaplaque LMT agarose (Marine Colloids, Rockland, USA) in 5-cm Petri dishes. The agarose with e m bedded microcalli was cut into pieces, which were inverted upon transferring into two 5-cm Petri dishes, each with 6 ml of liquid containing equal volumes of MS (MuR’\sHI(:I< and SKOOC; 1962) and VKM and 50 pg/ml kanamycin. One week later the liquid medium was replaced by liquid MS with 100 pg/ml kanamycin. The medium was replaced each week with

Circularized Ac/D.s Transposons 1163

A 1 2 3 4 fi Chromosomal DNA

B 2;opL&3 tandem Acs

Chromosomal DNA

FIGURE 1 .-A PCR assay designed to detect the formation of joined Ac ends. The filled and empty arrow heads corre- spond to the 5' and 3' terminal inverted repeats of Ac, respectively. (A) A set of PCR primers (arrows) designed to amplify joined Ac ends. (R) Possible amplification templates for the primers 2 + 3 and 1 + 4.

a fresh one. After 2 months kanamycin-resistant calli were scored and transferred into solid MS medium supplemented with 2 pg/ml naphthaleneacetic acid and 0.5 pg/ml benzyl- aminopurine. After 1 month of growth, calli were transferred to regeneration medium (MS containing 2 pg/ml kinetin and 0.8 pg/ml indoleacetic acid).

Southern analysis of tobacco plants transformed with Ds circle: For Southern analysis, 5 pg of genomic DNA was digested with PucI and EcoRV, fractionated on 0.8% agarose gels, and transferred to nitrocellulose purchased from MSI. Hybridization was performed according to manufacturers in- structions. A 2.5kb XbuI-Sac11 fragment from the Ds circle plasmid, containing the kanamycin-resistance gene, was radio- labeled by the random priming method (FEINRERC and VO- GELSTEIN 1983) and used as a probe.

RESULTS

Joined Ac ends are formed in plants containing actively transposing elements: To test for the presence of Ac circles, an attempt was made to physically isolate such circles by CsCI-EtdBr gradient centrifugation and Southern blotting. Several DNA extractions were done from Accontaining tobacco leaves, and 200-500 pg of DNA, depending on the extraction, was run on a gradient. A band of the size expected for the circularized Ac was seen only in some of the gradients and was always at the limit of detection (data not shown). This a p proach was not practical for further analysis, therefore a PCR strategy was used to identify circular Ac molecules. Two nested pairs of PCR primers were constructed that anneal within Ac and prime DNA synthesis toward the ends of the element (Figure 1A). Thus, only molecules containing nearby or adjacent right and left ends of Ac could serve as amplification templates (Fig- ure 1B). Such molecules could be either extrachromosomal Ac circles or tandem genomic Als in direct orien- tation.

FIGURE 2.-Detection of joined Ac/Ds ends by PCR. An ethidium bromide-stained agarose gel displaying the PCR products obtained with the primers designed to amplify joined Acends (Figure 1A). Primers 2 and 3 were used in the first round of amplification, and 1.5 pI of the reaction mixture was then used as a template for the second round of amplification with primers 1 and 4. The arrow indicates -540-bp PCR products expected for a circularized Ac or Ds element. The template in lanes 1-10 was the genomic DNA extracted from leaves of tobacco Containing Ac (lanes 1-5) or Ds (lanes 6-10) elements. In lane 1 1 , the template was the genomic DNA extracted from maize containing actively transposing elements. Lanes 12 and 13 show the results of the experiment designed to detectjoined ends upon transient transformation of tobacco protoplasts with D.wontaining plasmid (pVG2). pVG2 was transformed into protoplasts that express stable transposase (lane 12) and into wild-type protoplasts (lane 13). Transformation was done by Ca"/polyethyleneglycol technique (VARDI et ul. 1990). After overnight incubation protoplast DNA was extracted and analyzed by PCR.

Transgenic tobacco: We used transgenic tobacco plants containing an Ac or Ds element cloned into the leader sequence of the streptomycin-resistance gene (constructs pAGS4411 or pAGS4081). To confirm that Ac was actively excising and that Ds was stable, progeny seeds of the transgenic plants were germinated on strep tomycincontaining media where excision can be moni- tored in cotyledons by the appearance of green (strep tomycin resistant) sectors on a white background (JONES et al. 1989). All progeny of D.wontaining plants had white cotyledons, demonstrating that the element was inactive. The progeny of Accontaining plants showed green sectors, as expected for an active element, with the exception of plant Ar-4 whose progeny were all white. In this plant Ac was still present as deter- mined by PCR (data not shown), hence the absence of green spots suggests that it became inactive.

Genomic DNA extracted from leaves of tobacco plants tested for Ac/Ds activity was used as a template for two nested rounds of PCR with primers 2 and 3 in the first round and primers 1 and 4 in the second (Fig- ure 1A). Figure 2 shows the results of two nested rounds of amplification: no PCR product was obtained from


Ds-containing plants (lanes 6-10). On the other hand, out of 17 Ac-containing plants tested, all but Ac-4 gave rise to PCR products of the size expected for circularized Ac (-540 bp) as shown for a sample of plants in lanes 1-5. The absence of the 540-bp PCR product with D.r, and with Ac-4, which apparently carries an inactive Ac, demonstrates that transposition activity is involved in the formation of joined Ac ends. Occasionally PCR products of variable sizes, bigger than 540 bp, were observed (data not shown). They might originate from tandem Ac copies that arose as a result of somatic nearby jumps. The major reaction product always had the size expected for joined Ac ends and is therefore likely to be derived from extrachromosomal Ac circles; tandem genomic Ac's would be expected to give rise to PCR products of various sizes depending on the length of the sequence between the two elements.

Maiz~: To test whether joined Ac or Ds ends are formed in maize, plants were grown from variegated kernels ( i . e . , kernels carrying actively transposing elements). One line carried Ds within the Bz2 locus (THERES et al. 1987), and another carried Ac within the P1 locus (GREENBLATT and BRINK 1962). DNA was extracted from immature cobs and amplified with the primers designed to detect joined Ac ends. Unlike in transgenic tobacco, PCR amplification yielded several products (Figure 2, lane l l ) , one of which had the size expected for joined Ac ends. The multiplicity of bands in maize was expected as various active and defective Ac/Ds elements are probably present in the genome.

Protoplusts: To rule out the possibility that joined Ac ends originate from tandem genomic Ac's, we used transient assays to test for the formation ofjoined ends in tobacco protoplasts. Protoplasts obtained from transgenic plants that express stable Ac transposase, but do not carry any functional Ac or Ds elements, were transformed with the Ds containing plasmid (pVG2). Wild-type protoplasts were used in the control treat- ment. Twelve hours after transformation total DNA was extracted from the protoplasts and tested by PCR for the presence of joined Ds ends. PCR products corresponding to joined Ds ends were observed only in DNA obtained from transposase containing cells (Figure 2, lanes 12 and 13). Formation of joined Ds ends in the transient assay, in plants that do not contain Ac or Ds in the genome, strongly supports previous indications that joined ends are derived from extrachromosomal transposon circles.

Structure of joined Ac/Ds ends: PCR fragments of "540 bp were excised from a gel and purified. Se- quences derived directly from the PCR products were readable only up to the junction point, indicating that the DNA was a mixture of different products. There- fore, PCR products were cloned before sequencing. The structure of 26 different clones, obtained from nine plants (eight tobacco and one maize plant), is shown in Table 1. No molecules were found with perfect

TABLE 1

Sequence at the junction of joined Ac ends

Conceptnal head-to-head joining of Ar ends

-CATCCTACTTTCATCCCTG TAGGGATGAAAACGGTC-

1 2 3 4" 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

Sequences of the PCR clones

-CATCCTACTTTCATCCCTG -CATCCTACTTTCATCCCTG -CATCCTACTTTCATCCCTG -CATCCTACTTTCATCCCTA -CATCCTACTTTCATCCCTG -CATCCTACTTTCATCCCTG -CATCCTACTTTCATCCCTG -CATCCTACTTTCATCCCTG -CATCCTACTTTCATCCCTG -CATCCTACTTTCATCCCTG -CATCCTACTTTCATCCCTG -CATCCTACTTTCATCCCTG -CATCCTACTTTCATCCCTG -CATCCTACTTTCATCCCTG -CATCCTACTTTCATCCCTG -CATCCTACTTTCATCCCTG -CATCCTACTTTCATC -CATCCTACTTTCATC -CATCCTACTTTCATCCCT -CATCCTACTTTC -C -deletion of 41bp -CATCCTACTTTCATCCC -CATCCTACTTTCATCCCTG -CATCCTACTTTCATCCCTG -CATCCTACTTTCATCCCTG

G TAGGGATGAAAACGGTC- C TAGGGATGAAAACGGTC- AC TAGGGATGAAAACGGTC- GC TAGGGATGAAAACGGTC- GC TAGGGATGAAAACGGTC- CT TAGGGATGAAAACGGTC- CC TAGGGATGAAAACGGTC- AA TAGGGATGAAAACGGTC- TTG TAGGGATGAAAACGGTC- CTAA TAGGGATGAAAACGGTC-

7 bp" TAGGGATGAAAACGGTC- 21 bp" TAGGGATGAAAACGGTC- 21 bpb TAGGGATGAAAACGGTC- 27 bp!' TAGGGATGAAAACGGTC- 6 4 bp" TAGGGATGAAAACGGTC- GC deletion of 3 3 b p - TAC TAGGGATGAAAACGGTC- T ATGAAAACGGTC-

GGGATGAAAACGGTC- GATGAAAACGGTC- ATGAAAACGGTC-

AACGGTC- TAGGGATGAAAACGGTC-

GGGATGAAAACGGTC- deletion of 29 bp-

AACGGTC-

Sequence5 flanking Ac in pAGS4411 construct

GTGGGGCGCGTTGCGTGACCAcGCGTGACCCGGCGTGACCCGGCCGCGCGGG.

"The bz2::L)s2 allele of maize, which generated sequence 4, contains a D . L ~ element with perfect TI%.

Clones 11-15 contain insertions of the indicated size; sequences are shown below: clone l l , AGCATAA, 12, CCCCGC- GCGGCCGGGTCACGC;13,TATATCTTACTATACTGGGAC; 14, GGTC4CG~~CGCGCCCCACGTACAGC; 15, CGAGTT-

CTTATTCTATGGACAAAGC. CATTCGGATGGATACCTCGAGTJTACAAAAGAAGGTAAA-

joined ends. All the sequences contained either two intact Ac ends with a short intervening insertion, or small deletions in one or both ends. Clones with the deletions from both sides (five sequences out of 26) gave an additional evidence in favor of extrachromosomal Ac circles. They could not arise from tandem Ac copies, because if one transposon inserts at the position adjacent or even within the second element, at least one of the transposon ends must remain intact. Mole- cules that had one end intact and a deletion in the other end (four sequences out of 26) could be interpre- ted either as transposition of Ac into itself near its termini, or as a circular Ac. Deleted circles are not thought to reintegrate successfully as all de novo Ds insertions found so far have perfect terminal inverted repeats, or in the case of Ac, imperfect terminal inverted repeats with one single mismatch at the 5' end. Sequences with intact Ar ends are of two types: those with insertions

(;ircda1izrtl ,-\c/Il.s Transposons 1 165

resembling the flanking sequences of the donor site i n the original construct (clones I , 2. .5, 12, 14, and 16), and those with insertions unrelated to the donor. Inser- tions resembling the flanking donor site could be caused in principle by tandem jumps, or alternativcly, flanking sequences might be carried by the circularized element as a result of the excision process. Insertions unrelated to the donor site could be derived from scc- onclan, transposition events, o r could be generated upon circularization. r l r circles with two intact tl-ansptr son ends might sene as transposition intermediates.

Are Ac/Ds circles functional transposition intermediates? To find, whether Ar/lls circles arc filnctional transposition intermediates, we tested their ability 1 0 reinsert into the genome. We constructed an artificial Ds circle (Figure 3.4) that carried intactjoincd /\e ends from clone 3 (Table 1) with >2OO bp of f\r seqllence flanking the junction from each side [this length o f subterminal regions is sufficient for transposition (COUPIASD v/ d . 1989)] and a plant selectable marker (kanamycin resistance). Srlch a Ils circle was cotransformed into tobacco protoplasts, with the transposase producing plasmid (pNT804), or with the control plasmid (pVG6) lacking the transposase ORF. The number of kanamycin-resist;lrlt calli was scored. The transposase sollrce was identical to that used i n experiments ~'11e1.e Ils was shown to transpose from a plasmid into the tobacco genome (HOI'RA-HEKIS r / nl. 1994).

Two types of events might lead to the formation o f kanamvcin-resistant calli: ( 1 ) random insertion, which should also occur in the absence o f transposase, and (2) insertion via the transposition pathway. Results of three independent experiments are summarized in Ta- ble 2. In a11 three experiments, transposasc increased the efficiency of transformation with 1 ) s circles.

To study whether 11s circles integrated via the transposition pathway, insertion patterns were analyzed. In the case of transposition, the circle is expected t o open precisely between the Ac ends and t o generate an X- hp target site duplication upon insertion. Plants were regenerated from kanamycin-resistallt calli, which were obtained by cotransformation of Ilscircle and the trans- posaseexpressing plasmid. Genomic DNA from 34 plants was digested with P,rcI and I.:CoRV enzymes that cut within the 11,s circle close to the.joined ends (Figme 3A) and hybridized with a probe to an internal part of the Ds circle. Integration of the 11,s circle via transposition, i.p., upon linearization at the ends junction, is expected to give a 5.2-kh hand in a Southern blot. Fig- ure 3R shows the Southern hlot of 18 plants out of 34 analyzed; the remaining 16 plants gave a similar hand pattern (data not shown). Most plants showed multiple hands with sizes corresponding to random plasmid integration; no prominent 5.2-kb band was obsenwl among the different plants.

Four plants that showed a hand of "5.2 kb and contained a low number of insertions (plant 14 in Figure

Flc;r'KI' ?.-Testing reinsrrtion of putative c i n " h - inter- mctliatrs into t h r grnomc*. (A) Sc1wnl;Itic rc.1~r~~sc~lt;~tioll of ;In artificial I)( circle containing intact joinrrl ,,\e mrls S C ~ I -

latrd hy two nuclrotitlrs (clonc 3, T a h l r 1) w i t h >!?OO hp 01' /\e suhtc~mlinal scvqwnccs: n 1 I l n h c v - s h c ~ l o w t h r ;~rro\vs corrc- spontl t o the positions i n ,.\e, starting nrlnlhcring from thr .5' rnd. Thc 1 ) s circlr ;dso conrains ;I kall;~mycin-rc.sist;lnec gcnr unclcr t h r n o / d i r 7 ~ syn/n.ct> promotrr t h a t eo11fc.r~ klnanlycin rcsist;uncc t o p I ; \ n t CCIIS (ktna"), scqwnccs rrquirc*cI for rc*pIi- cation i n /*,k/rr*riclrh tnli, ( : o l E l origin of~~plication (ori) . ant1 ;I hactcriaI anlpicillin rcsistancr gcnc* (AmpK). I'ritnrrs usrd in invcrsc P(:R ; I ~ C intlicatrtl ;IS snlall ;~rrows. Thrir sequmces ~ r r givcn i n ~ ~ ~ \ ~ ~ ~ ~ ~ ~ ~ ; \ ~ . ~ . \ ~ ~ ) \ I ~ : ~ ' ~ I O I ) S . (R) Souther11 hlot iIl1aly-

sis o f the transgrnic plants olminctl hy eot~;msforlnation of thc 1)s eirclc and pSlXO4. thr transpos;w prodrlcc*r. Plant gcnomic DNX was rligwrd w i t h /"I ;wcl I.hR\' enzymes. which e111 w i t h i n t h c I).\ circlc closc. t o t h r . j o i n c d rnds (see Figurc- :%;I), ;lntl hyhritlixctl w i t h ;I prohe consisting o f the k;lnamycin-r~sist;llle~~ gc.nc-. If t h c I)scirclr intcgratcas via transposition. ;x., is linc;lrizctl at the cnds.junction, a .5.!?-kh hand, intlicatrrl hy the arrow. is rxpc-ctcd t o apptm on the Soutllern h l o t .

3B, and another three that gave similar pattern o n the Southern blot) were chosen, and the regions flanking Ar ends i n these plants were analyzed by inverse PCR. Genomic DNA of thew plants was partially digested with S d M , circularized and amplified at both 11s bor- ders. Two nested pairs o l ' PCR primers were rlsetl that anneal within the 1 ) s circle, with one primer of each pair pointing toward the tl-ansposon end and thc second primer pointing toward the proximal part of 11s (see Figure 3A and \I.\vrmI..\I.s AM) WTHOIX). Mllen


TABLE 2

Number of kanamycin-resistant calli obtained in three independent experiments

Plasmids used for transformation

Ds circle + pNT804 Ds circle + pVG6 Experiment (+ transposase) (- transposase)

1 12 1 2 362 87 3 494 138

Values are number of kanamycin resistant calli. Dszcircle was cotransformed into tobacco protoplasts with a plasmid that expresses transposase (pNT804) or with a control plasmid (pVC6) that does not express transposase.

PCR products were cloned and sequenced, all of them were found to carry the same joined Ac ends as in the Ds circle, indicating that in these plants Ds circles integrated in a random (illegitimate) manner.

To confirm that experimental conditions were adequate to detect transposition, we tested whether a “standard” Ds element (not a Ds circle) cloned into a Blues- cript vector could excise in the presence of transposase. Plasmid pVG5, containing a Ds element with 2.7 kb of DNA (the vector) separating the element ends, was cotransformed into tobacco protoplasts, with the transposase-producing plasmid (pNT804), or with the control plasmid (pVG6) lacking the transposase OW. Twenty-four hours after transformation, protoplast DNA was extracted and used as a template to detect empty donor sites by PCR (see MATERIALS AND METH- ODS). A DNA fragment with the size expected for empty donor sites was found only in cotransformation with pNT804. This fragment was cloned and independent clones were sequenced. The sequences obtained (Fig- ure 4) contain typical Ac-excision footprints (SCOTT et nl. 1996). This suggests that the experimental conditions used were adequate for transposition to occur. Nevertheless, the LIS circle used in this experiment seemed not to be an efficient substrate for transposase- mediated integration.

DISCUSSION

Ac joined ends are derived from circular DNA molecules: .Joined Acends were detected, using a PCR assay, in maize and transgenic tobacco plants. Joined ends were found only in plants that contained actively transposing elements, which demonstrates that the transposition machinery was involved in their formation and argues strongly against the possibility of a PCR arti- fact. The following evidence indicates that the joined ends were derived from Ac or Ds circles, rather than from tandem genomic copies of the transposon. (1) Based on migration in agarose gels, the major amplification product had the size expected for precisejoining of the transposon ends, whereas tandem copies of the

A

Ac 2 1 A c5’

B Waxy

pVGS

Ex 1

Ex2

CGTTGCGTGACC

CGTTGCGTGACC; (Ds) GCGTGAG

CGlTGCGTGAC GC CGTGACG

C G T T a G FIGURE 4.-Excision of Ds from a plasmid in which Ac ends

are separated by 2.7 kb. (A) A schematic description of pVG5, a standard Ds element cloned into a Bluescript vector. The Ds element contains Ac terminal regions, from nucleotides 1 to 252 in the 5‘ end and from nucleotides 4300 to 4561 in Ac 3’ end. The Ds internal part consists of a plant kanamycin- resistance gene (KanaR) and of a bacterial chloramphenicol- resistance gene (CmR). The Bluescript vector sequences separating Ac ends are shown, including the ColElori origin of replication (on), and a bacterial ampicillin-resistance gene (AmpR). Primers used in PCR for excision footprints amplification (T3, T7, W1, W2) are indicated as small arrows. Their sequences are given in MATERIAI.~ AND METHODS. (B) Excision footprints obtained upon cotransformation of pVG5 with the transposase source (plasmid pNT804). The genomic sequence of the wild-type donor site flanking the Ds element in pVG5 is shown (Waxy). The unexcised Ds element flanked by the underlined host 8-bp duplication is shown (pVG5). Two sequences (Ex1 and Ex2) showing excision footprints are presented.

transposon were expected to give rise to PCR products of various sizes depending on the length of the sequence separating the two elements. It must be men- tioned that Ac was not reported to have a preference for transpositions within only a few bp from the original site. Among more than 15 reports corresponding to hundreds of transposon insertions mapped relative to the original site, only one event has been found where Ac transposed just 6 bp from its original position (ATHMA et nl. 1992). (2) Several sequences of the PCR products containing joined ends had deletions at both ends, which cannot be explained by tandem transposition. (3) Joined ends were formed 12 hr after protoplast transformation with Ds-containing plasmid, making the possibility of formation of tandem genomic elements very unlikely.

Circularized Ac/Ds Transposons 1167

Interpretation of the Ac end junctions: We have presented the structure of 26 amplification products containing joined Ac ends (Table l), obtained from nine plants (eight tobacco and one maize). Different clones were obtained from the same plant, indicating the pop- ulation of transposon circles is heterogeneous. No molecules were found with either perfect joined ends, or with an eight-nucleotide insertion. This is consistent with the observations that the type of excision that leaves the host duplication unchanged (SCOTT et al. 1996) and precise excision (BARAN et al. 1992) are rare. All the sequences contained either two intact Ac ends with a short insertion in between, or a few nucleotides deleted from one or both ends. Deletions at the transposon ends in Ac circles could have been caused by exonucleolytic attack that occurred before circularization or by imprecise excision. Additional nucleotides inserted between the transposon ends might have been derived from the flanking sequences as a result of imprecise excision. In this case, the absence of molecules with perfect ends joining would suggest that Ac is not excised by flush cleavages at the transposon ends but rather by some kind of a staggered cut, as previously proposed (SAEDLER and NEVERS 1985; COEN et al. 1989). Excision with staggered cuts was observed upon Tn7 transposition (BAINTON et al. 1991). Alternatively, additional nucleotides may reflect the use of “filler DNA” to repair deletion (WESSLER et al. 1990; NASSIF et al. 1994) or untemplated synthesis, similar to “N” bases at V(D)J junctions (GELLERT 1992).

Are Ac circles transposition intermediates? Although transposon circles have been previously described in eukaryotes, they have never been assayed for their ability to integrate. In this work, we tested the ability of a putative Ds circular transposition intermediate to reinsert into the genome. When Ds circles were transformed into tobacco protoplasts, transposase increased the efficiency of transformation. Analysis of the transformed plants, however, did not show efficient transposase-mediated integration, but rather patterns of illegitimate recombination typical of direct DNA transformation. We cannot exclude the possibility that circles used in our experiments could occasionally reinsert via transposition-mediated integration, since in plants with multiple insertions (Figure 3B), some inserts might be due to transposition events. However, only a minority of the integrations could be transposition-mediated based on the size of hybridizing fragments in S blots (Figure 3B). The effect of transposase on transformation efficiency might be explained by an overall stimulation of recombination activity in the cell, or by a role in targeting Ds circles into the nucleus.

Experimental conditions used here were probably adequate to detect transposase-mediated integration, because when similar transformation experiments were performed (HOUBA-HERIN et al. 1994) using Ds cloned into a plasmid in a conventional way ( i e . , with the

transposon ends separated by a long enough stretch of plasmid DNA), quite efficient integration was reported. Moreover, we confirmed that a standard Ds element could excise from a plasmid donor (Figure 4) under the experimental conditions used. Therefore one interpretation of the lack of reinsertion of a joined-ends- containing circle is that circular transposons are not transposition intermediates, but rather abortive products. Alternatively, it is possible that real intermediates are circular. Such intermediates might have perfect joined ends, or ends spaced by 8 bp corresponding to the host duplication. As we did not find such molecules, it might be implied that the real circular intermediates occur so transiently in the cell that they were not detected even though a sensitive PCR assay was used. An- other possibility is that circular intermediates are not able to reinsert in our system if, for example, a transposase-transposon complex needed for reinsertion can be formed only upon excision. In summary, we cannot rule out a circular intermediate in the process of Ac transposition. However, it is likely that some of the circles, such as the ones with deletions at both ends, are not able to reinsert.

Model for Ac transposition: A question raised by this work is how do circular Ac molecules fit into the models of transposition? One transposition model, which is still speculative, has been proposed for Ac (GREENBLATT 1984) and for the Tam3 element of Antirrhinum ma& (ROBBINS et al. 1989). According to this model, there is no full excision of the transposon, rather, transposition is a step by step process in which one end of the element is first excised and ligated to the recipient site and only after that the second end is cut (Figure 5B). In such a pathway, circles can be formed only if the transposition process is aberrant and, following cleavage of the first end of Ac (end A in Figure 5B), the second end (end B) is cleaved, yielding a fully excised element, and then the two free ends are joined. The resulting circles might be abortive products, as sug- gested here for some of the circles, or might reintegrate. In any case, our findings of circles suggests that at some point there is full excision of the element. An alternative model to the step by step pathway can pro- vide a simpler explanation for circle formation. Ac- cording to this model, Ac normally transposes via a full excision pathway (Figure 5C) with cleavage at both ends (C/A and B/D) and formation of a linear intermediate as was shown for prokaryotic transposons Tn 7 (BAINTON et al. 1991) and TnlO (HANIFORD P t al. 1991). Such linear molecule could transpose directly and/or be circularized (Figure 5C).

Circularization of extrachromosomal linear DNA molecules by nonhomologous end-joining has been shown for several species including mammalian (WOODWORTH-GUTAI 1981; WILSON et al. 1982; ROTH et al. 1985), avian (MILLER and TEMIN 1983) and plant cells (V. GORBUNOVA and A. A. LEVY, unpublished), and


A

B

a. I 1

I ‘* C D - EA B F

d. C.D

FIGURE 5.-Two models for conservative transposition. Thick lines represent transposon sequences; thin straight lines, donor site; thin wavy lines, recipient site; C/A and B/D, element-donor site junctions; E/F, recipient site; small arrows, sites of cleavage. (A) Precleavage complex consisting of donor and target DNA. (B) Mechanism of transposition according to ROHHINS cl nl. (1989): (a) transposition starts with a cleavage at one transposon end (C/A) and in the recipient site (E/F); (b) ligation of this transposon end to the recipient site (A + E); (c) the second transposon end (B/D) is cleaved and (d) ligated to the recipient site (F + B). (C) Transposition according to the model implying the formation of extrachromosomal intermediate: (a) transposition starts with the cleavages at both transposon ends (C/A and B/D) and recipient site (E/F); (b) only after the transposon is fully excised, its ends are ligated to recipient site (A + E and B + F ) ; (c) transposon circle formation is caused by the ligation of A and B.

is often associated with deletions and insertions reminis- cent of those shown here (Table 1). Hence, host-mediated circularization of a linear transposition intermediate could explain the origin of Ac circles and of the footprints at their endsjunction. Our attempts to isolate this linear intermediate have not been yet successful (data not shown), possibly because ofa very low amount of such molecules, as expected from their transient oc- currence in the cell.

Transposon circles were reported for eukaryotic DNA-DNA elements (ROSE and SNUTCH 1984; SUNDAM- SAN and FKEELING 1987). Interestingly, for the 7i.1 element of C. elegans, both circular and linear extrachromosomal forms were detected (ROSE and SNUTCH 1984; RUAN and EMMONS 1984). One may speculate that these circles were generated by a host-promoted conversion of linear intermediates into circles.

It has been found previously that -50% of individu- als that exhibit germinal excision do not carry a trans-

posed Ac. This phenomenon of “Ac loss” has been observed in maize (MCCLINTOCK 1956; GREENBLATT 1984; DOONEK and BELACHEW 1989), in tobacco (JONES et al. 1990) and in Arabidopsis (ALTMANN et al. 1992; DEAN et al. 1992). Meiotic segregation, transposition to the sister chromatid and excision without reinsertion were proposed as possible explanations for Ac loss. If circular transposons are abortive products, circularization of linear intermediates might be a mechanism that prevents an excised transposon to reinsert. Such mechanism might enable eukaryotic hosts to cure themselves from transposons.

We thank A. SE:I.L~ANOV for help and advice throughout this study, C. I,I(:HTI.:NSHTEIN and E. RL~BIN for critical reading of the manuscript, Y. AWVI and V. LWY for editing the manuscript, and D. BERLIN and V. PI:PKO for technical help. This work was supported by the Forch- heimer foundation and by Dr. CHUTICK’S doctoral fellowship to V.G.

LITERATURE CITED

.k \ rMANN, T., R. SCHMIDT and L. WILLMITZER, (1992) Establishment of a gene tagging system in Arubidopsis thaliana based on the maize transposable element Ac. Theor. Appl. Genet. 84: 371- 383.

AN, G., 1987 Binaty Ti vectors for plant transformation and promoter analysis. Methods Enzymol. 153: 292-305.

ATHMA, P., E. GROTEWOI.II and T. PETERSON, 1992 Insertional muta- genesis of the maize P gene by intragenic transposition of Ar. Genetics 131: 199-208.

AVIV, D., and E. GAIL’N, 1985 An zn vitro procedure to assign pig- ment mutations in Nicotiana to either the chloroplast or the nucleus. .J. Hered. 76: 135-136.

BAIWON, K., P. GAMAS and N. CRAIG, 1991 Tn7 transposition in

crated by staggered breaks in DNA. Cell 65: 805-616. v i l m proceeds through an excised transposon intermediate gen-

BARAN, G., C . EWT, T. BUKEAIJ and S. WESSLER, 1992 Molecular analysis of the maize wx-li3 allele indicates that precise excision of the transposabk Ar element is rare. Genetics 130: 377-384.

BENJAMIN, H. W., and N. KI.E(:I(NEK, 1992 Excision of 7 n 1 0 from the donor site during transposition occurs by flush doublestrand cleavages at the transposon termini. Proc. Natl. Acad. Sci. USA 89: 4648-4652.

<:HEN, J., I. M. ~ R E E K B I A T T and S . L. DEIJAPORTA, 1992 Molecular aualysis of Ac transposition and DNA replication. Genetics 130: 665-676.

COEN, E. S., T. P. ROBBINS, J. ALMEIDA, A. HUDSON and R. CARPENTER, 1989 <:onsequences and Mecharlisms of transposition in Anlir- rhinum rnaju, pp. 413-436 in Mobile DNA, edited by D. E. BERG atid M. M. Howl;. American Society for Microbiology, Washing- ton, DC.

COL~PIANI), G., C . PLUM, S. C:~lr27-TqE, A. POST and P. STAKL.IN(;EK, 1989 Sequences near the termini are required for transposition of the maize transposon Ac in transgenic tobhacco plants. Proc. Natl. head. Sci. USA 86: 9585-9388.

DFAN, C., C. S p I ) I N , T. PAGE, J. JONES and C. LISTER, (1992) Behav- iour of the maize transposable element Ar in Arubidopsis thuliana.

DEI.IAPORTA, S . L., J . Wool ) and J. B. HICKS, 1983 A plant DNA PlantJ. 2: 69-81.

mini-preparation: version 11. Plant Mol. Biol. Reporter 1: 19-21. DOONER, H. K., and A. BEIKIIEW, 1989 Transposition pattern of

thc maize transposable element Ac from the bz-m2(Ac) allele. (knetics 122: 447-457.

DOONEK, H. K., J. KI<I.I.F.R, E. HARPER and E. RAMTON, 1991 L‘ariable patterns of transposition of the maize element Artivator in tobacco. Plant Cell 3: 473-482.

FEI)OROFF, N. V., 1989 Maize transposable elements, pp. 375-41 1 in hlobik DNA, edited by D. E. BERG and M. M. H o w . American Societv for Microbiolow, Washington, DC.

FEINHER(;, A . P., and B. \’O~~F.l.STI~3N, 1983 A technique for radiola-

Circularized Ac/Ds Transposons 1169

beling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132: 6-13.

FINNEGAV, D. J., 1992 Transposable elements. Curr. Opin. Genet. Dev. 2: 861-867.

FRlDI.EXDER, M., K. HARRISON, J. D. G. JONES and A. A. LEW, 1996 Repression of the Ac transposase gene promoter by Ar transposase. PlantJ. 9: 911-917.

GEILEKT, M., 1992 Molecular analysis of V(D)J recombination. Annu. Rev. Genet. 22: 425-446.

GREENMATT, I. M., 1984 A chromosome replication pattern deduced from pericarp phenotypes resulting from movements of

471-485. the trdnsposdbk element, Modulator, in maize. Genetics 108:

GwxxwAm, 1. M., and R. A. BRINK, 1962 Twin mutations in medium variegated pericarp maize. Genetics 47: 489-501.

HAXIFORD, B. D., H. Mi. BEyJAMrN and N. KI.F.CKNER, 1991 Kinetics and structural analysis of a cleaved donor intermediate and a strand transfer intermediate in TnlO transposition. Cell 64: 171- 179.

HANIE’OKL), D., and N. KI.EC:KTVER, 1994 TnlO transposition in vivo: temporal separation of cleavages at the two transposon ends and

EMBO J. 1 3 3401-3411. roles of terminal basepairs subsequent to interaction of ends.

HORSCH, R. B. ,J . E. FRY, N. I,. HOFFMXYN, D. EICHHOLTZ, S. G. Ro- (XKS et aL, 1985 A simple and general method for transferring genes into plants. Science 227: 1229-1231.

HOURA-HERIN, N., D. BECKF,R, A. POST, A. LARONDELI~E and P. STAR- t.IN(;ER, 1990 Excision of a Dslike maize transposable element in transient assay in Petunia is enhanced by a truncated coding region of the transposable element Ac. Mol. Gen. Genet. 2 2 4 17-23.

HOUR..\-HERIN, N., M. D O M I N and J. PEDRON, 1994 Transposition of a Ds element from a plasmid into the plant genome in Nirotiana plumhaginofolia protoplast-derived cells. Plant J. 6: 55-66.

JONES, J. D. G., F. C. CARI14ND, P. M,U.IGA and H. K. DOONER, 1989 Visual detection of transposition of the maize element Activator (Ac) in tobacco seedlings. Science 244: 204-207.

JONES,J. D. G., F. C. CAMAND, E. LIM, E. RALSIOK and H. K. DOOXER, 1990 Preferential transposition of the maize element Activator to linked chromosomal locations in tobacco. Plant Cell 2: 701- 707.

KEI.I .ER, J., E. LIM and H. K. DOONER, 1993 Preferential transposition of Ar to linked sites in Arabidopsis. Theor. Appl. Genet. 86: 585-588.

KUNZE, R., 1996 The maize transposable element Activator ( A c ) , pp. 161-194 in Transposable Elements, edited by H. SAEIXER and A. GIEIU.. Springer, Berlin.

MARTIN, C., and C. LJSTER, 1989 Genome juggling by transposons: Tam3induced rearrangement3 in Antirrhinum majus. Dev. Genet. 10: 438-451.

M ~ I J N T O C K , B., 1948 Mutable loci in maize. Carnegie Inst. Wash- ington Year Book 47: 155-169.

MCCI.INTOCK, B., 1951 Chromosome organization and genic expression. Cold Spring Harbor Symp. Quant. Biol. 16: 13-47.

MCCI.INTOCK, B., 1956 Mutation in maize. Carnegie Inst. Wash. Year Book 5 5 323-332.

MII.I.ER, C. R, and H. M. TEMIN, 1983 High efficiency ligation and recombination of DNA fragments by vertebrate cells. Science 220: 606-609.

MUI.~.EK-NEUMANN, M., J. I. YODER and P. STARI.INC;EK, 1984 The DNA sequence of the transposable element Ac of &a mays L. Mol. Gen. Genet. 198: 19-24.

MURASHIGF., T., and F. SKOOC, 1962 A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiologia Plantarum 15: 473-497.

NASSIF, N., J. PENNEY, S. PAL, W . R. ENGELS and G. B. GLOOR, 1994 Efficient copying of nonhomologous sequences from ectopic sites via P-element-induced gap repair. Mol. Cell. Bid. 14: 1613- 1625.

PEIJMAN, J., B. C o m ~ , W. VAN CAMP, M. VAN MONTAGU and D. INZE, 1991 Transient occurence of an extrachromosomal DNA of an Arabidopsis lhaliana transposon-like element, Tall. Proc. Natl. Acad. Sci. USA 88: 3618-3622.

POHl.hlr\x, R. F., N. V. FEDOROFF and J. MESSING, 1984 The nucleotide sequence of the maize controlling element Arfioator. Cell 37: 635-643.

ROBBINS, T., R. CARP~NTER and E. S. COEN, 1989 A chromosome rearrangement suggests that donor and recipient sites are associated during Tam? transposition in Antzrrhinum majus. EMBO J. 8: 5-13.

ROSF., A. M., and T. P. SNUTCH, 1984 Isolation ofthe closed circular form of the transposable element Tr1 in Cacnorhahditis & p t . s . Nature 331: 485-486.

ROTH, D. B., T. N. PORTER and J. H. WII.SON, 1985 Mechanisms o f nonhomologous recombination in mammalian cells. Mol. Ccl. Biol. 5: 2599-2607.

R ~ A N , K., and S. W. EMMONS, 1984 Extrachromosomal copies of transposon ‘fcl in the nematode ~arnorhabditir ekgans. hoc . Natl. Acad. Sci. USA 81: 4018-4022.

SAEDIXR, H., and A . CIEIU., 1996 7rc~nsposc~ble Elummt.y. Springer, Berlin.

SAEDI.EK, H., and P. NEVERS, 1985 Transposition in plants: a molecular model. EMB0.J. 4: 585-590.

Scorr, L., D. LAFOE and C. F. WEII., 1996 Ajacent sequences influ- ence DNA repair accompanying transposon excision in maize. Genetics 142: 237-246.

SUGIMOTO, K.,Y. OTSLIM, S. SA11 and H. HIRO(:HIKA, 1994 Transposi- tion of the maize Ds element from a viral vector to the rise genome. Plant J. 5: 863-871.

SLINI)AFXSAN, V., and M. FREEI.IN(;, 1987 An extrachl-omosomal form of the Mu transposons of maize. Proc. Natl. Acad. Sci. USA 84: 4924-4928.

StTrON, h’. D., W. I>. GERIA(: l l , D. SCHWARTZ and W . J . PE.K:o(:K, 1984 Molecular analysis of 1l.s controlling element mutations at the Adhl locus of maize. Science 223: 1265-1268.

THERES, N., T. SCHEFLF, and P. STARLINGER, 1987 Cloning of the Bz2 locus of mays using the transposable element 11s as a gene tag. Mol. Gen. Genet. 209: 193-197.

Mobilisation of quiet, endogenous Tr3 transposons of C. tkgans by forced expression of Tc3 transposase. EMBO J. 1 2 2513-2520.

VAKL)I, A,, S. BI.EICl1MAN and D. A\lc, 1990 Genetic transformation of citrus protoplasts and regeneration of transgenic plants. Plant Sci. 69: 199-206.

WESSLER, S. R., A. TI\RPI.EY, M. PURL:(;GANAN, M. SPEI.1. and R. Om- G m , 1990 Filler DNA is associated with spontaneous deletions in maize. Proc. Natl. Acad. Sci. USA 87: 8731-8735.

WII.SON,J. H., P. BER(;EI‘ and J. PIPAS, 1982 Somatic cells efficiently join unrelated DNA segments end-to-end. Mol. Cell. Biol. 2: 1258-1269.

WOOl~~~OKTH-~~lTAl, M., 1981 Recombination in SV40-infected cells: viral DNA sequences at sites of circularization of transfect- ing linear DNA. Virology 109: 353-365.

VAN IllF,NF,N, H. G. A. M., s. D. COLI.OMS and R. H. A. PIASTERK, 1993

Communicating editor: V. SUNIIARESAN

Circularized Ac/Ds Transposons: Formation, Structure and Fate · circularized transposons. The...

Documents

Transcript of Circularized Ac/Ds Transposons: Formation, Structure and Fate · circularized transposons. The...