Intrachromosomal Homologous Recombination in Arabidopsis Induced by a Maize Transposon

Mol Gen Genet (2000) 263: 22±29 Ó Springer-Verlag 2000

ORIGINAL PAPER

Y.-L. Xiao á T. Peterson Intrachromosomal homologous recombination in Arabidopsis induced by a maize transposon

Received: 28 September 1999 / Accepted: 19 November 1999

Abstract In plants, the frequency of spontaneous in- Introduction trachromosomal homologous recombination is low. Here, we show that a maize transposable element greatly Transposable element sequences make up a large por- stimulates intrachromosomal homologous recombina- tion of many eukaryotic genomes, and hence may be able tion between direct repeat sequences in Arabidopsis. to in¯uence genetic recombination in a number of ways. Plants were transformed with a construct (GU-Ds-US) For example, transposon insertions may have a negative containing a Ds (Dissociation) transposable element in- impact on recombination; in heterozygous condition serted between two partially deleted GUS reporter gene they constitute sequence heterologies that may interrupt segments. Homologous recombination between the pairing or branch migration of Holliday junctions overlapping GUS fragments generates clonal sectors (Dooner and Martinez-Ferez 1997; Biswas et al. 1998). visible upon staining for GUS activity. Plants containing Alternatively, transposable elements may provide dis- the GU-Ds-US construct and a source of Ac (Activator) persed sequence homologies that can participate in ec- transposase showed an over 1000-fold increase in the topic recombination (Montgomery et al. 1991; Clegg incidence of recombination relative to plants containing et al. 1997; Caceres et al. 1999). Moreover, transposable the same construct but lacking transposase. Transposon- elements may a€ect recombination more directly induced recombination was observed in vegetative and through the action of transposon-encoded proteins that ¯oral organs, and several germinally transmitted events cleave and rejoin DNA during the transposition process. were recovered. Transposon-induced recombination The possible e€ects of transposable elements on re- appears to be a general phenomenon in plants, and thus combination have been investigated since their discovery may have contributed to genome evolution by inducing by McClintock (1949). In E. coli,Tn3 was reported to deletions between repeated sequences. promote general recombination in neighboring regions (Kondo et al. 1989). Also, Tn7 transposition can create Key words Direct repeat á Recombination á a hotspot for homologous recombination at the trans- Transposon á Arabidopsis á b-Glucuronidase position donor site (Hagemann and Craig 1993). Inter- molecular transposition of IS10 greatly stimulates, and is coupled to, homologous recombination between the donor and acceptor molecules at the transposition site Communicated by H. Saedler (Eichenbaum and Livneh 1995). In Drosophila, the P element can signi®cantly increase recombination fre- Y.-L. Xiao quencies in the male germline (Hiraizumi 1971; Kidwell Interdepartmental Genetics Program, and Kidwell 1976) and somatic cells (Sved et al. 1990). Department of Zoology and Genetics, 2288 Molecular Biology Building, The P element also promotes ecient gene conversion, Iowa State University, Ames, IA 50010, USA which enables site-speci®c gene replacement and can T. Peterson (&) induce a variety of associated chromosomal rearrange- Interdepartmental Genetics Program, ments (Engels et al. 1990; Gloor et al. 1991; Preston and Department of Zoology and Genetics, Engels 1996; Preston et al. 1996). In humans, it was and Department of Agronomy, found that two inherited peripheral neuropathies are 2206 Molecular Biology Building, Iowa State University, Ames, IA 50010, USA caused by a recombination hot spot located near a E-mail: [email protected] Mariner transposon-like element (Kiyosawa and Chance Tel.: +1-515-294-6345; Fax: +515-294-6755 1996; Reiter et al. 1996). In maize, early reports indi- 23 cated that the transposable element Activator (Ac)orDs Recovery of the germinally transmitted recombination events either reduced (McClintock 1953), or had no e€ect on Seeds were harvested from the siblings of plants of the genotype (Fradkin and Brink 1956), crossing over in the region (GU-Ds-US/-, sAc/-) that gave a high frequency of blue spots upon ¯anking the element. However, it has been observed that staining for GUS activity. A portion of the seeds from each of 18 Ac insertions can induce homologous recombination plants were grown on MS plates until two true leaves had devel- between directly repeated sequences in the maize P locus oped, at which stage the plants were stained for GUS activity. If (Athma and Peterson 1991). Ac may also destabilize a any uniform blue seedlings were detected, the remaining sibling seeds were planted in soil. At the 4-leaf stage, one leaf from each tandem duplication in the maize Bz locus (Dooner and plant was stained for GUS activity. In the case of those plants Martinez-Ferez 1997), albeit at a much lower frequency which gave a uniformly staining leaf, a second leaf was picked and than that observed for the maize P locus. In addition, Ac stained for GUS activity. In this way, we identi®ed uniformly is reported to induce low levels of somatic recombina- GUS-positive plants which were used as a source of DNA for Southern analysis of the recombination event. tion between ectopic sites in transgenic tobacco (Shalev and Levy 1997). The aim of this study was to test directly the ability of PCR and Southern hybridization a transposable element to induce recombination between Genomic DNA samples (Dellaporta et al. 1983) were ampli®ed by homologous repeat sequences in plants. For this pur- PCR for 35 cycles of denaturation at 94 °C for 30 s, annealing at pose, we examined transgenic Arabidopsis plants con- 57 °C for 30 s, and elongation for 1 min at 72 °C. The primer taining a recombination substrate composed of a maize sequences used were: P1, 5¢-GAAGACTCAGACTCAGACT-3¢; Ds transposable element inserted between overlapping G1, 5¢-GGTGGGAAAGCGCGTTACAAG-3¢; H3, 5¢-CGTCT- GGACCGATGGCTGTG-3¢; H2, 5¢-TTCGGGGCAGTCCTCG- segments of a bacterial gene (uidA) encoding b-glucu- G-3¢; H1, 5¢-GATGTAGGAGGGCGTGG-3¢; D1, 5¢-GATCCG- ronidase (GUS). The results indicate that activation of GTTCTCTCCAAATG-3¢; S1, 5¢-CTGTCTGGCTTTTGGCTG- the Ds insertion by Ac transposase supplied in trans TG-3¢;X,5¢-GGATATTCTGCAACCCTTCCCCTCC-3¢; and Y, causes a high level of recombination between the 5¢-CTCGCAGGTATGTTTGTCTC-3¢. ¯anking repeat sequences. Recombination may be PCR products were cloned into the pT7Blue T-vector (Nov- agen) and sequenced at the ISU DNA Sequencing and Synthesis stimulated in part by a double-strand break induced by Facility. Probe preparations and Southern hybridizations were transposon excision, and possibly by additional performed as described. unknown properties of the Ac transposase.

Results Materials and methods Detection of somatic recombination events Construction of the binary vector GU-Ds-US in transgenic plants

Vector pWS31 (Sundaresan et al. 1995), obtained from Dr. We inserted a Ds element between two partially over- V. Sundaresan, was digested with SalI to remove the GUS gene lapping, non-functional segments of the b-glucuronidase fragment, then religated to form pWS31Y, which contains a Ds element with NPTII as selection marker. Plasmid pWS31Y was gene (gus; Je€erson et al. 1987) to generate a binary cleaved with SacI to release the 4.7-kb band containing Ds,to plant transformation vector termed GU-Ds-US (Fig. 1). which SacI-PstI linker fragments (synthesized by the Sequencing The homologous direct repeat sequences are 618 bp in and Synthesis Facility at ISU) were ligated. Plasmid pGU.US length, and the distance between them is 6.3 kb, in- (Tinland et al. 1994) was cleaved with PstI and ligated to the Ds element with attached SacI-PstI linkers to generate GU-Ds-US. cluding the 4.7-kb Ds element. Recombination between Restriction enzyme digestion, ligation and plasmid preparation the direct repeats in the GU and US segments would were performed according to standard protocols and enzyme generate a functional GUS gene driven by a strong manufacturers' instructions (Sambrook et al. 1989). constitutive (CaMV 35S) promoter. A. thaliana ecotype Columbia was transformed with the GU-Ds-US con- Plant transformation and histochemical assays struct by vacuum in®ltration (Bechtold et al. 1993). The original transformed plants were allowed to self-polli- Arabidopsis thaliana ecotype Columbia was transformed by vac- nate, and progeny were analyzed by Southern hybrid- uum in®ltration (Bechtold et al. 1993) with Agrobacterium strain ization (data not shown). Three independent transgenic ASE (obtained from E. Meyerowitz) containing the binary vector GU-Ds-US, and transformed seeds were selected by growth on starter lines with single-copy GU-Ds-US transgene in- agar media containing 30 lg/ml kanamycin. Kanamycin-resistant sertions were selected for analysis. Plants homozygous lines were further screened by progeny testing and Southern for the GU-Ds-US transgene were crossed with lines analysis to identify three independent lines (DsI5, DsI6, DsII7) expressing the Ac transposase in the No-O (CS 8037 and that carried a single integrated GU-Ds-US transgene. Ac trans- posase lines (CS8037, CS8038, CS8045) were obtained from the CS 8038) and Landsberg (CS 8045) backgrounds. As a Arabidopsis Biological Resource Center (ABRC, Columbus, control, the transgenic GU-Ds-US plants were also Ohio). Plants were grown as described by Koncz et al. (1992). crossed with non-Ac, wild-type Arabidopsis No-O and Histochemical staining of leaf samples or whole plants at the full- Landsberg. The progeny plants were grown to the full- rosette stage was performed as described by Je€erson et al. (1987; rosette stage and stained with X-Gluc. Blue sectors, in- Swoboda et al. 1994). X-Gluc (5-bromo-chloro-3-indolyl-b-D- glucoronide) was obtained from Rose Scienti®c (Edmonton, dicative of restoration of the GUS coding sequence, were Alberta). observed in all plant parts examined, including roots, 24

Fig. 1 Structure of the GU-Ds-US construct and recombination lacks a Ds insertion between the GU and US segments products. The upper section shows a map of the GU-Ds-US construct (Swoboda et al. 1994; Chiurazzi et al. 1996). In the integrated into the plant genome. RB and LB, T-DNA right border and left border sequences, respectively; P, 35S promoter of CaMV; T, presence of the sAc transposase source, the average re- nopaline synthase terminator; GU and US indicate the partially combination frequency of the GU-Ds-US construct is overlapping GUS gene fragments; the triangle represents a Ds element increased by more than 1000 fold. containing the NPTII gene. The lower section shows a map of the intact GUS gene, restored by homologous recombination between the direct repeat sequences. Restriction sites for HindIII, XhoIandEcoRV are indicated. The ®lled bars indicate plant genome DNA, the arrows PCR analysis of somatic recombination events show the positions of primers used for PCR analysis. The dark bars below the maps indicate probes U and S used for Southern analysis We determined the structure of the GU-Ds-US transgene in the presence or absence of the Ac transposase source hypocotyls, cotyledons, petioles, leaves, siliques and (sAc) using PCR ampli®cation of genomic DNA seeds (Fig. 2). The GUS+ sectors ranged in size from (Fig. 3). A band predicted to arise from the restored less than 10 cells, to occasional large sectors that en- GUS gene was obtained using DNA from plants con- compassed 50% or more of an entire leaf (Fig. 2b). taining the GU-Ds-US construct together with sAc These results indicate that transposon-induced recom- (Fig. 3, lane 6), but not from plants lacking sAc (Fig. 3, bination events can occur in both vegetative and ¯oral lane 11). The PCR band derived from the presumptive tissues, and at most or all stages of plant development. recombination product was sequenced and found to As shown in Table 1, plants in the control group contain a precisely restored GUS gene (not shown). (genotype GU-Ds-US/),noAc) had few or no blue Similarly, a PCR product predicted to arise by excision sectors (average of 0.6 blue sectors per plant), whereas of Ds from the GU-Ds-US construct was found in plants plants in the experimental group commonly had hun- containing sAc (Fig. 3, lane 5), but not in plants lacking dreds of blue sectors (average of 700 blue sectors per sAc (Fig. 3, lane 10). Sequencing of the presumptive Ds plant). These results were obtained from three inde- excision product showed that it contained a typical Ds pendently transformed GU-Ds-US lines (DsI5, DsI6, excision footprint (data not shown). The non-transposed DsII7), in crosses with three di€erent lines expressing Ac Ds (4.7 kb) is too large to be ampli®ed under the PCR transposase (CS8045, CS8537, CS8538). Based on the conditions used here. average numbers of blue sectors found in control group plants, the calculated spontaneous recombination fre- quency of the GU-Ds-US transgene in the absence of Ac Molecular con®rmation of germinally transmitted transposase is in the range 1.99 ´ 10)7 to 5.40 ´ 10)8 recombination events events/genome. This frequency was determined by re- lating the number of recombination events (sectors) to To con®rm that restoration of GUS gene function the number of genomes present per plant at the time of occurred through homologous recombination, we ana- staining, as described by Swoboda et al. (1994). This lyzed the molecular structure of the transgene locus in value is similar to that reported previously for the rate of germinally stable GUS-positive plants. Plants carrying spontaneous recombination of a GU.US transgene sim- both the GU-Ds-US and sAc transgene loci were allowed ilar to the GU-Ds-US transgene used here, but which to self pollinate, and the progeny were screened for 25

)7 Fig. 2a±f GUS activity stain of plants transformed with GU-Ds-US somatic recombination as determined here (1.99 ´ 10 and crossed with plants bearing an Ac transposase source. a Frequent to 5.40 ´ 10)8 events/genome), and reported previously small blue sectors in leaves. b Blue sector comprising 50% of the leaf, by Swoboda et al. (1994) (10)6 to 10)7 events/genome). indicating an early recombination event. c Blue sectors in cotyledons; blue sector encompassing new leaves, indicating a recombination Four plants with uniform GUS expression were event in meristem. d Siliques containing blue seeds. e Isolated blue grown to maturity and analyzed by Southern hybrid- seeds. f Uniformly blue F3 seedlings ization. Hybridization with GUS-speci®c probe U (Fig. 1) detected two bands (7.3 kb and 4.7 kb) in HindIII- digested DNA from a plant of genotype GU-Ds-US/-, individual seedlings that showed uniform GUS activity. sAc/- (Fig. 4a, lane 2), whereas the GUS-speci®c probe Among the progeny of 18 self pollinated plants (14,314 detected only one band of 5.0 kb in DNA from two seedlings), six plants with uniform GUS staining were plants with uniform GUS expression (Fig. 4a, lanes 3 identi®ed (see Materials and methods). Thus, the fre- and 4). Similar results were obtained from XhoI-digested quency of germinally transmitted recombination events DNA hybridized with U probe (Fig. 4b). The U probe giving rise to GUS-positive plants is approximately detected one large band (12 kb) in a plant of genotype 4.2 ´ 10)4 events/seed. This value is approximately GU-Ds-US/-, sAc/- (Fig. 4b, lane 1), and a small band 1000-fold higher than the frequency of spontaneous (5.2 kb) in two plants with uniform GUS expression 26

Table 1 Frequencies of blue (GUS+) sectors in plants containing the GU-Ds-US transgene

GU-Ds-US linea Control lines (no Ac)b Ac transposase linesc Number of plants examined Number of blue spotsd

DsI5 Landsberg ± 9 3 No-O ± 10 11 CS8045 10 3310 CS8537 11 2467 CS8538 7 3152 DsI6 No-O ± 10 3 CS8537 11 11956 DsII7 No-O ± 10 7 CS8537 10 5465 CS8538 10 7984 a DsI5, DsI6 and DsII7 are independent GU-Ds-US transgenic lines transposase b Landsberg and No-O are control wild-type lines d Blue sectors were visualized under a dissecting microscope at 5´ c CS8045, CS8537 and CS8538 are three lines which express Ac magni®cation

ed using enzymes that cleave in the DNA ¯anking the transgene insertion. Genomic DNA from plants of ge- notype GU-Ds-US/-, sAc/- and the uniformly GUS+ plants were digested with EcoRV and hybridized with probe S, which is speci®c for the 3¢ end of the GU-Ds- US construct (Fig. 1). The S probe hybridizes to the same 4.2-kb band in the GU-Ds-US progenitor plants and the uniformly GUS+ plants (Fig. 4c). This result indicates that the two GUS+ plants could not have arisen by seed or pollen contamination, but did in fact originate by recombination at the GU-Ds-US transgene locus.

Fig. 3 PCR analysis of genomic DNA from GU-Ds-US transformant plants. Lane 1, MW standards (1 kb ladder, BRL); lane 2, Negative PCR control (primers G1 + H3, but using distilled water instead of Discussion plant DNA); lanes 3±7, DNA from plants of the genotype GU-Ds- US/-, sAc/-; lanes 8±12, DNA from plants of the genotype GU-Ds- US/-, no Ac. Lanes 3 and 8, primers H2 + D1; lanes 4 and 9, primers We show here that insertion of Ds, a maize transposon, G1 + H3; lanes 5 and 10, primers H1 + H2; lanes 6 and 11, primers can stimulate recombination between directly repeated P1 + S1; lanes 7 and 12, primers X and Y (speci®c for Ac)

Fig. 4a±c Southern analysis of recombination of the GU-Ds-US (Fig. 4b, lanes 3 and 4). These results are consistent with construct. a DNA was digested with HindIII and hybridized with recombination between the homologous regions of the probe U (see Fig. 1). Lane 1, DNA from wild type Arabidopsis;lane2, DNA from variegated GUS+ plants of the genotype GU-Ds-US/- GU-Ds-US construct, because there is no XhoI site in sAc/-; lanes 3 and 4, DNA from uniform GUS+ plants, derived by the construct. In addition, we detected a band (7.4 kb) recombination of the GU-Ds-US construct. b DNA was digested with expected from simple Ds excision in DNA from the XhoI and hybridized with probe U. Lane 1, DNA from variegated plant of genotype GU-Ds-US/GU-US, sAc/- (Fig. 4b, GUS+ plants of the genotype GU-Ds-US/- sAc/-; lane 2, DNA from lane 2). To verify that the recombinant GUS+ plants variegated GUS+ plants of the genotype GU-Ds-US/GU-US sAc/-; lanes 3 and 4, DNA from uniform GUS+ plants, derived by were derived from the original GU-Ds-US transform- recombination of the GU-Ds-US construct. c DNA was digested with ants, additional Southern hybridizations were perform- EcoRV and hybridized with probe S (see Fig. 1). Lanes as in a 27 sequences in transgenic Arabidopsis plants. Recombi- postulated mechanisms, including DSB repair (DSBR; nation events were detected by a gain-of-function assay Szostak et al. 1983), single-strand annealing (SSA; Lin based upon restoration of the reading frame encoding et al. 1984, 1990; Fishman-Lobell et al. 1992), or syn- b-glucuronidase (GUS;Je€erson et al. 1987).Theincrease thesis-dependent single-strand annealing (SDSA; Nassif in the recombination frequency requires the presence of et al. 1994). SSA is thought to be the major pathway a second transgene that expresses Ac transposase. causing deletion of the internal sequence between direct Recombination in the presence of the Ac transposase repeats in yeast (Fishman-Lobell et al. 1992; Osman occurred approximately 1000-fold more frequently than et al. 1996). Similarly, recombination between the the spontaneous recombination frequency for both so- homologous GUS segments in our experiments could be matic sectors and germinally transmitted events. The most easily explained by the SSA mechanism. Recently, somatic sectors of GUS+ cells are clearly of premeiotic an alternative model termed One-Sided Invasion (OSI) origin, whereas the uniformly GUS-positive seedlings was postulated as a major pathway for somatic recom- may have resulted from either premeiotic or meiotic bination events in plants (Belmasza and Chartrand 1994; recombination. Plants lack a germ line, and hence pre- Puchta et al. 1996; Puchta 1998). meiotic recombination events could be transmitted to On the other hand, the enhancement of recombina- the next generation if a GUS+ clonal sector contributes tion by the Ac/Ds transposons may be due not only to to the gametophytic lineage. the generation of a DSB by transposon excision. In the Induction of recombination is not unique to the Ac/ experiments reported here, we observed a greater than Ds transposon families, as the Mutator transposon also 1000-fold enhancement of recombination over the promotes recombination between direct repeat se- spontaneous frequency. This high recombination rate quences in the maize Knotted locus (Lowe et al. 1992). may re¯ect a heretofore unknown e€ect of Ac trans- Therefore, transposon-induced homologous recombi- posase, such as the ability to mediate the pairing of nation may be a general phenomenon in plants, and homologous sequences, to recruit host recombination possibly other organisms (Thompson-Stewart et al. factors, or both. Alternatively, because Ac/Ds transpo- 1994). All plant genomes studied to date contain trans- sition is thought to occur predominantly or exclusively poson-like elements and repetitive sequences, and many immediately following DNA replication (Chen et al. contain duplicated genes. Our results suggest that, in 1992), transposon-generated DSBs may occur at a par- addition to increasing genome size by generating local ticular time of the cell cycle in which recombination is sequence duplications (Zhang and Peterson 1999), enhanced relative to simple transposon excision. transposable elements can also reduce genome size by Transposon-induced recombination may provide a inducing deletions between directly repeated sequence useful way to enhance the frequency of certain homol- elements. ogy-dependent gene manipulations in plants. The The mechanism of transposon-induced recombina- Arabidopsis and maize genomes are currently being tion remains to be determined, but one possibility is that saturated with Ac/Ds insertions (Long et al. 1997), and recombination is stimulated by DNA double-strand it may be possible to induce recombination between breaks (DSBs) generated by transposon excision. Dou- endogenous sequence duplications that carry Ac/Ds ble-strand breaks are known to initiate recombination in insertions. Such recombination events would generate fungi (Szostak et al. 1983) and plants (Gorbunova and interstitial deletions that could simplify the analysis of Levy 1999). Somatic crossover between homologous multicopy gene clusters such as are frequently associated plant chromosomes can be induced by DNA-damaging with disease resistance genes (Welin et al. 1995; Capel agents such as gamma-radiation (Carlson 1974). Low et al. 1997; Takken et al. 1998). Transposon-induced doses of X-rays, gamma-rays and UV light can increase recombination might also be used to delete undesirable the intrachromosomal homologous recombination fre- sequences from integrated transgenes, or to delete mul- quency in plants with arti®cial substrates (Tovar and tiple transgene copies that are commonly generated in Lichtenstein 1992; Lebel et al. 1993; Puchta et al. 1995). many transformation protocols (Kwok et al. 1985; Klein Treatment with chemical agents that induce DSBs, such et al. 1989). Of greatest bene®t would be the develop- as methylmethanesulfonate (MMS) and mitomycin C, ment of an ecient gene-targeting system (Hohn and can also enhance recombination rates (Lebel et al. 1993; Puchta 1999; Vergunst and Hooykaas 1999). Targeted Puchta et al. 1995). Moreover, homologous recombi- gene disruption by homologous recombination in nation in plants is enhanced by in vivo induction of Arabidopsis was recently reported (Kempin et al. 1997), DNA double-strand breaks by a site-speci®c endonu- but gene targeting remains dicult in plants due to the clease (Puchta et al. 1993; Puchta 1998). In Arabidopsis, low frequency of homologous recombination. The gene generation of DSBs by HO endonuclease increased the targeting frequency via homologous recombination was frequency of somatic intrachromosomal homologous estimated at between 10)3 and 10)6 (Miao and Lam recombination about tenfold (Chiurazzi et al. 1996). 1995; Risseeuw et al. 1997). Here we unambiguously Therefore, transient DNA double strand breaks gener- demonstrate that Ac/Ds transposons can stimulate re- ated by transposon excision may be a critical initiator of combination between two homologous sequences in cis. recombination events. Following the appearance of a However, it remains to be determined whether trans- DSBs, recombination could occur via one of several poson-induced recombination can also enhance recom- 28 bination in trans, and thereby facilitate the development Hiraizumi Y (1971) Spontaneous recombination in Drosophila of a workable gene-targeting system in plants. melanogaster males. Proc Natl Acad Sci USA 68: 268±270 Hohn B, Puchta H (1999) Gene therapy in plants. Proc Natl Acad Sci USA 96: 8321±8323 Acknowledgements We thank Holger Puchta for the pGU.US Je€erson RA, Kavanagh TA, Bevan MW (1987) GUS fusions: plasmid and helpful suggestions, and V. Sundaresan for the pWS31 b-glucuronidase as a sensitive and versatile gene marker in plasmid. We also thank Dongying Wu and David Wright for their higher plants. EMBO J 6: 3901±3907 help. This work was supported by a grant from the Iowa Corn Kempin SA, Liljegren SJ, Block LM, Rounsley SD, Yanofsky MF, Promotion Board. Journal Paper No. J-18553 of the Iowa Agri- Lam E (1997) Targeted disruption in Arabidopsis. Nature 389: culture and Home Economics Experiment Station, Ames, Iowa, 802±803 Project No. 3297, and supported by Hatch Act and State of Iowa Kidwell MG, Kidwell JF (1976) Selection for male recombination funds. in Drosophila melanogaster. Genetics 84: 333±351 Kiyosawa H, Chance PF (1996) Primate origin of the CMT1A- REP repeat and analysis of a putative transposon-associated recombinational hotspot. Hum Mol Genet 5: 745±753 References Klein TM, Kornstein L, Sanford JC, Fromm ME (1989) Genetic transformation of maize cells by particle bombardment. Plant Athma P, Peterson T (1991) Ac induces homologous recombination Physiol 91: 440±444 at the maize P locus. Genetics 128: 163±173 Koncz C, Chua N-H, Schell J (1992) Methods in Arabidopsis Belmaaza A, Chartrand P (1994) One-sided invasion events in research. World Scienti®c, Singapore homologous recombination at double-strand breaks. Mutat Res Kondo K, Inokuchi H, Ozeki H (1989) The transposable element 314: 199±208 Tn3 promotes general recombination at the neighboring Bechtold N, Ellis J, Pelletier G (1993) In planta Agrobacterium- regions. Jpn J Genet 64: 417±434 mediated gene transfer by in®ltration of adult Arabidopsis Kwok WW, Nester EW, Gordon MP (1985) Unusual plasmid thaliana plants. CR Acad Sci 316: 1194±1199 DNA organization in an octopine crown gall tumor. Nucleic Biswas I, Yamamoto A, Hsieh P (1998) Branch migration through Acids Res 13: 459±471 DNA sequence heterology. J Mol Biol 279: 795±806 Lebel EG, Masson J, Bogucki A, Paszkowski J (1993) Stress- Caceres M, Ranz JM, Barbadilla A, Long M, Ruiz A (1999) induced intrahomologous recombination in plant somatic cells. Generation of a widespread Drosophila inversion by a trans- Proc Natl Acad Sci USA 90: 422±426 posable element. Science 285: 415±418 Lin FL, Skerle K, Sternberg N (1984) Model for homologous Capel J, Jarillo JA, Salinas J, Martinez-Zapater JM (1997) Two recombination during transfer of DNA into mouse L cells: role homologous low-temperature-inducible genes from Arabidopsis for DNA ends in the recombination process. Mol Cell Biol 4: encode highly hydrophobic proteins. Plant Physiol 115: 569± 1020±1034 576 Lin FL, Skerle K, Sternberg N (1990) Intermolecular recombina- Carlson PS (1974) Mitotic crossing-over in a higher plant. Genet tion between DNAs introduced into mouse L cells is mediated Res (Camb) 24: 109±112 by a nonconservative pathway that leads to crossover products. Chen J, Greenblatt IM, Dellaporta SL (1992) Molecular analysis Mol Cell Biol 10: 103±112 of Ac transposition and DNA replication. Genetics 130: 665± Long D, Goodrich J, Wilson K, Sundberg E, Martin M, 676 Puangsomlee P, Coupland G (1997) Ds elements on all ®ve Chiurazzi M, Ray A, Viret JF, Perera R, Wang X-H, Lloyd AM, Arabidopsis chromosomes and assessment of their utility for Signer ER (1996) Enhancement of somatic intrachromosomal transposon tagging. Plant J 11: 145±148 homologous recombination in Arabidopsis by the HO endonu- Lowe B, Mathern J, Hake S (1992) Active Mutator elements sup- clease. Plant Cell 8: 2057±2066 press the knotted phenotype and increase recombination at the Clegg MT, Cummings MP, Durbin ML (1997) The evolution of Kn1-0 tandem duplication. Genetics 132: 813±822 plant nuclear genes. Proc Natl Acad Sci USA 94: 7791±7798 McClintock B (1949) Mutation in maize. Carnegie Inst Washing- Dellaporta SL, Wood J, Hicks JB (1983) A plant DNA mini- ton Year Book 48: 142±154 preparation: version II. Plant Mol Biol Rep 1: 19±21 McClintock B (1953) Mutation in maize. Carnegie Inst Washing- Dooner HK, Martinez-Ferez IM (1997) Germinal excisions of the ton Year Book 52: 227±237 maize transposon Activator do not stimulate meiotic recombi- Miao ZH, Lam E (1995) Targeted disruption of the TGA3 locus in nation or homology-dependent repair at the bz locus. Genetics Arabidopsis thaliana. Plant J 7: 359±365 147: 1923±1932 Montgomery EA, Huang SM, Langley CH, Judd BH (1991) Eichenbaum Z, Livneh Z (1995) Intermolecular transposition of Chromosome rearrangement by ectopic recombination in IS10 causes coupled homologous recombination at the trans- Drosophila melanogaster: genome structure and evolution. position site. Genetics 140: 861±874 Genetics 129: 1085±1098 Engels WR, Johnson-Schlitz DM, Eggleston WB, Sved J (1990) Nassif N, Penney J, Pal S, Engels WR, Gloor GB (1994) E- High-frequency P element loss in Drosophila is homolog cient copying of nonhomologous sequences from ectopic sites dependent. Cell 62: 515±525 via P-element-induced gap repair. Mol Cell Biol 14: 1613± Fishman-Lobell J, Rudin N, Haber JE (1992) Two alternative 1625 pathways of double-strand break repair that are kinetically Osman F, Fortunato EA, Subramani S (1996) Double-strand separable and independently modulated. Mol Cell Biol 12: break-induced intrachromosomal recombination in the ®ssion 1292±1303 yeast Schizosaccharomyces pombe. Genetics 142: 341±357 Fradkin C-MW, Brink RL (1956) Crossing over in maize in the Preston CR, Engels WR (1996) P-element-induced male recombi- presence of the transposable factors Activator and Modulator. nation and gene conversion in Drosophila. Genetics 144: 1611± Genetics 41: 901±906 1622 Gloor GB, Nassif NB, Johnson-Schlitz DM, Preston CR, Engels Preston CR, Sved JA, Engels WR (1996) Flanking duplications and WR (1991) Targeted gene replacement in Drosophila via P deletions associated with P-induced male recombination in element-induced gap repair. Science 253: 1110±1117 Drosophila. Genetics 144: 1623±1638 Gorbunova V, Levy AA (1999) How plants make ends meet: DNA Puchta H (1998) Repair of genomic double-strand breaks in double-strand break repair. Trends Plant Sci 4: 263±269 somatic plant cells by one-sided invasion of homologous Hagemann AT, Craig NL (1993) Tn7 transposition creates a hot- sequences. Plant J 13: 331±339 spot for homologous recombination at the transposon donor Puchta H, Dujon B, Hohn B (1993) Homologous recombination in site. Genetics 133: 9±16 plant cells is enhanced by in vivo induction of double strand 29

breaks into DNA by a site-speci®c endonuclease. Nucleic Acids Swoboda P, Gal S, Hohn B, Puchta H (1994) Intrachromosomal Res 21: 5034±5040 homologous recombination in whole plants. EMBO J 13: 484± Puchta H, Swoboda P, Gal S, Blot M, Hohn B (1995) Somatic 489 intrachromosomal homologous recombination events in popu- Szostak JW, Orr-Weaver T, Rothstein R, Stahl F (1983) The lations of plant siblings. Plant Mol Biol 28: 281±292 double-strand-break repair model for recombination. Cell 33: Puchta H, Dujon B, Hohn B (1996) Two di€erent but related 25±35 mechanisms are used in plants for the repair of genomic double- Takken FL, Schipper D, Nijkamp HJ, Hille J (1998) Identi®cation strand breaks by homologous recombination. Proc Natl Acad and Ds-tagged isolation of a new gene at the Cf-4 locus of Sci USA 93: 5055±5060 tomato involved in disease resistance to Cladosporium fulvum Reiter LT, Murakami T, Koeuth T, Pentao L, Muzny DM, Gibbs race 5. Plant J 14: 401±411 RA, Lupski JR (1996) A recombination hotspot responsible for Thompson-Stewart D, Karpen GH, Spradling AC (1994) A two inherited peripheral neuropathies is located near a mariner transposable element can drive the concerted evolution of tan- transposon-like element. Nature Genet 12: 288±297 demly repetitious DNA. Proc Natl Acad Sci USA 91: 9042± Risseeuw E, Franke-van Dijk MEI, Hooykaas PJJ (1997) Gene 9046 targeting and instability of Agrobacterium T-DNA loci in the Tinland B, Hohn B, Puchta H (1994) Agrobacterium tumefaciens plant genome. Plant J 11: 717±728 transfers single-stranded transferred DNA (T-DNA) into Sambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning: the plant cell nucleus. Proc Natl Acad Sci USA 91: 8000± a laboratory manual. Cold Spring Harbor Laboratory Press, 8004 Cold Spring Harbor, NY Tovar J, Lichtenstein C (1992) Somatic and meiotic chromosomal Shalev G, Levy AA (1997) The maize transposable element Ac recombination between inverted duplications in transgenic induces recombination between the donor site and an homol- tobacco plants. Plant Cell 4: 319±322 ogous ectopic sequence. Genetics 146: 1143±1151 Vergunst AC, Hooykass PJJ (1999) Recombination in the plant Sundaresan V, Springer P, Volpe T, Haward S, Jones JDG, Dean genome and its application in biotechnology. Crit Rev Plant Sci C, Ma H, Martienssen R (1995) Patterns of gene action in plant 18: 1±31 development revealed by enhancer trap and gene trap trans- Welin BV, Olson A, Palva ET (1995) Structure and organization of posable elements. Gene Dev 9: 1797±1810 two closely related low-temperature-induced dhn/lea/rab-like Sved JA, Eggleston WB, Engels WR (1990) Germline and somatic genes in Arabidopsis thaliana. Plant Mol Biol 29: 391±395 recombination induced by in vitro modi®ed P elements in Zhang J, Peterson T (1999) Genome rearrangements by non-linear Drosophila melanogaster. Genetics 124: 331±337 transposons in maize. Genetics 153: 1403±1410