Rearrangements of the DNA in Carbon Ion-Induced Mutants of Arabidopsis thaliana

Corresponding author: Naoya Shikazono, Watanuki-machi 1233, Takasaki, Gunma, 370-1292, Japan., naoya{at}taka.jaeri.go.jp (E-mail)

Communicating editor: C. S. GASSER

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

To elucidate the nature of structural alterations in plants, three carbon ion-induced mutations in Arabidopsis thaliana, gl1-3, tt4(C1), and ttg1-21, were analyzed. The gl1-3 mutation was found to be generated by an inversion of a fragment that contained GL1 and Atpk7 loci on chromosome 3. The size of the inverted fragment was a few hundred kilobase pairs. The inversion was found to accompany an insertion of a 107-bp fragment derived from chromosome 2. The tt4(C1) mutation was also found to be due to an inversion. The size of the intervening region between the breakpoints was also estimated to be a few hundred kilobase pairs. In the case of ttg1-21, it was found that a break occurred at the TTG1 locus on chromosome 5, and reciprocal translocation took place between it and chromosome 3. From the sequences flanking the breakpoints, the DNA strand breaks induced by carbon ions were found to be rejoined using, if present, only short homologous sequences. Small deletions were also observed around the breakpoints. These results suggest that the nonhomologous end-joining (NHEJ) pathway operates after plant cells are exposed to ion particles.

STRUCTURAL alterations of DNA, such as deletions, produce valuable null mutations for plant genetics. Structural alterations of the DNA encompassing a single gene are of considerable interest because one can directly clone a gene from a plant carrying the mutation using genomic subtraction or representative difference analysis (RDA; SUN et al. 1992 ; LISITSYN et al. 1993 ). Thus, elucidation of the size and nature of the structural alterations induced is essential not only for the characterization of the mutation but for the application of such molecular techniques. Alternatively, analyses on the nature of structural alterations of the DNA in plants may provide important insights into DNA repair mechanisms.

Recent molecular analyses have revealed that X-ray and fast neutron irradiation and T-DNA integration could induce DNA rearrangements in the plant genome. BRUGGEMANN et al. 1996 reported that 13 out of 18 long hypocotyl 4 (hy4) homozygous mutants of Arabidopsis induced by fast neutrons carried deletions >5 kbp. It was also shown that complex rearrangements involving deletions, inversions, or insertions occurred in X-ray- and fast neutron-induced tt3 and tt5 mutants (SHIRLEY et al. 1992 ). LAUFS et al. 1999 found a 26-cM inversion bordered by two T-DNAs in a direct orientation. Nonhomologous end joining (NHEJ) of double-strand breaks was suggested to be involved in the formation of these rearrangements (SHIRLEY et al. 1992 ; LAUFS et al. 1999 ).

Characteristics of NHEJ include the use of short homologies for rejoining, the occurrence of a short deletion at the broken site, and the presence of filler DNAs as well as direct or inverted repeats at the rejoined site. NHEJs were also observed at the rejoined sites of restriction enzyme-digested plasmids in tobacco protoplasts (GORBUNOVA and LEVY 1997 ), at the rejoined sites of an integrated marker gene in tobacco cells that was digested with I-SceI (SALOMON and PUTCHA 1998 ), at the integration or rearranged sites in transgenic rice and Arabidopsis after direct gene transfer (TAKANO et al. 1997 ; SAWASAKI et al. 1998 ; KOHLI et al. 1999 ), and at the border of T-DNA insertion sites (GHEYSEN et al. 1991 ; MAYERHOFER et al. 1991 ; LAUFS et al. 1999 ).

The killing and mutagenic effects of ion particles in higher plants have been studied extensively using various kinds of plants (see SMITH 1972 for review). These studies demonstrated that plant cells were most efficiently inactivated or mutagenized per unit dose when the linear energy transfer (LET) of the ion particle was 100–200 keV/µm. Similar results were obtained with other cell types such as bacteria and mammalian cells (KRAFT et al. 1992 ; GOODHEAD 1995 ). LETs in this range are high and cannot be obtained with X rays or fast neutrons. A biologically important characteristic of ion particle irradiation is that it produces dense ionization within a localized area in the irradiated cell. This has been postulated to produce "clustered damage" on the DNA and to lead to an enhanced cell inactivation and mutation induction (GOODHEAD 1995 ). Therefore, rearrangements are expected to be more frequent after ion particle irradiation than after X-ray or fast neutron irradiation. After a rearrangement occurs, sequencing of the affected area can provide valuable information on the underlying DNA repair process that led to the rearrangement. Furthermore, as a result of such clustered damage, the sizes of deletions or inversions induced by ion particles may be smaller than those induced by low-LET radiation. If this is true, structural alterations induced by ion particles may provide new alleles that show unique phenotypes, and these alleles or phenotypes could be useful in plant genetics.

Only a few studies have examined the structural alterations of DNA induced by ion particles using plants as materials. MEI et al. 1994 found that a rearrangement had occurred in the genome of an argon ion-induced dwarf mutant of rice by Southern blot analysis using random genomic clones as probes. Previously, we analyzed the characteristics of the carbon ion-induced mutations by PCR and Southern blot analyses and found that, out of four mutants, one had a total gene deletion, two contained inversions and/or translocations, and one had a point-like mutation (SHIKAZONO et al. 1998B ). The results of these studies suggested that ion particles frequently induce DNA rearrangements in plants. However, to our knowledge, there have been no reports on the analysis of the molecular nature of rearrangements at the sequence level after ion particle irradiation. In our study, DNA fragments from carbon ion-induced Arabidopsis mutants, which were found to contain rearrangements (SHIKAZONO et al. 1998B ), were cloned and subjected to sequence analysis.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Plant material:
Dry seeds of Arabidopsis thaliana ecotype Columbia (Col) were exposed to carbon ions (LET = 113 keV/µm) as described by TANAKA et al. 1997 . The dry seeds were irradiated with a dose of 150 Gy, which was at the shoulder of the survival curve. Irradiated M1 seeds (26,200 lines) were sown and selfed to obtain 104,088 M2 seeds. From the screening of mutants and from the complementation test, six lines of gl1 mutant, six lines of tt4 mutant, and four lines of ttg1 mutant were isolated. The gl1, tt4, and ttg1 alleles, each containing a rearrangement, were designated gl1-3, tt4(C1), and ttg1-21, respectively, and were subjected to analysis (SHIKAZONO et al. 1998B ).

Extraction of the genomic DNA and the cloning of DNA fragments:
Genomic DNA was extracted from 3- or 4-week-old M4 plants by slightly modifying the procedure described by MURRAY and THOMPSON 1980 .

For the analysis of gl1-3, a genomic library was constructed by ligating HindIII-digested genomic DNA to pUC19. DNA fragments were screened from this library by colony hybridization. Colony blots were prepared on positively charged nylon membranes (Hybond-N⁺; Amersham, Arlington Heights, IL). Digoxigenin-labeled probe I and probe II (Fig 1A), which were used for cloning fragments from gl1-3, were synthesized by PCR using primer 1, 5'-GAATGAGAATAAGGAGAAGAGATGAA-3', and primer 2, 5'-GGCAGTGATGAACAATGACGGTGG-3'; and primer 3, 5'-ATGTTGAGTACTGCCTTTAG-3', and primer 4, 5'-CCATGATCCGAAGAGACTAT-3', respectively. In the present study, amplification was carried out at 94° for 3 min, followed by 50 cycles of 94° for 1 min, 50° for 1 min, and 72° for 2 min. At the end of the 50 cycles, the samples were incubated at 72° for another 10 min for complete extension. The probe for cloning the fragment containing the 107 bp from Col was synthesized by PCR using primer 5, 5'-AAGCTACGAGACCGGTAGAA-3', and primer 6, 5'-TTAGGCTTAGACTATGTAATTTGAT-3'.

View larger version (23K): In this window In a new window Download PPT slide   Figure 1. Cloning and sequencing of fragments containing the GL1 gene sequence from wild type (Col) and gl1-3. (A) A schematic representation of the HindIII fragment containing the GL1 gene from wild type (Col). Exons are shown as solid boxes. Probes used to clone fragments from gl1-3 are shown below as thick solid lines at the corresponding location. The break that occurred in gl1-3 locates between probes I and II. Bg, BglII; E, EcoRI; H, HindIII; Sc, ScaI; Xb, XbaI. (B) A fragment from gl1-3 cloned by probe I. (C) A fragment from gl1-3 cloned by probe II.

Hybridization was conducted according to the protocol of the manufacturer (Boehringer Mannheim, Indianapolis). After hybridization, filters were washed once with 2x SSC, 0.1% SDS and then once with 0.1x SSC, 0.1% SDS at 68°. Hybridized probes were detected by a color reaction using nitro blue tetrazolium chloride and 5-bromo-4-chloro-3-indolyl-phosphate (Boehringer Mannheim).

For the analysis of the rejoined sites of tt4(C1) and ttg1-21, thermal asymmetric interlaced (TAIL) PCR was carried out as described by LIU et al. 1995 . The specific primers DSP1, 5'-CTGCAGGCATCTTGGCTATTGG-3'; DSP2, 5'-TTGGCACTGCTAACCCTGAGAA-3'; and DSP3, 5'-CTACTACTTCCGCATCACCAACAG-3', and the arbitrary degenerate primer AD1, 5'-(A/T)GTGNAG(A/T)ANCANAGA-3', were used to amplify the downstream sequence of the break in tt4(C1); and the specific primers USP1, 5'-ACGGAAGGACGGAGACCAAG-3'; USP2, 5'-CTTCACTGCCGCTTCTTTGC-3'; and USP3, 5'-CGATGTCCTGTCTGGTGTCC-3', and the arbitrary degenerate primer AD2, 5'-NGTCGA(G/C)(A/T)GANA(A/T)GAA-3', were used to amplify the upstream sequence of the break in tt4(C1). The specific primers DSP1, 5'-AGCCTCTCCCGAATCTCTCCTT-3'; DSP2, 5'-CTCCTTCGAGCATCCTTATCCT-3'; and DSP3, 5'-AGCATCCTTATCCTCCAACAAA-3', and the arbitrary degenerate primer AD3, 5'-GTNCGA(G/C)(A/T)CANA(A/T)GTT-3', were used to amplify the downstream sequence of the break in ttg1-21; and the specific primers USP1, 5'-CGAATATCGAGAATCACAACCTTA-3'; USP2, 5'-TTAGAATCCATCAAAATCGTAGCC-3'; and USP3, 5'-TTCGGCTCTACATCGTTCCAATCG-3', and the arbitrary degenerate primer AD2, 5'-NGTCGA(G/C)(A/T)GANA(A/T)GAA-3', were used to amplify the upstream sequence of the break in ttg1-21.

Sequencing:
Nucleotide sequences of both DNA strands of the cloned fragments were determined by the dideoxynucleotide chain termination method (SANGER et al. 1977 ) with an automated sequencer (ABI Prism 377 and 310 Genetic analyzer; Perkin-Elmer, Norwalk, CT). The assembly and editing of the DNA sequences were performed with the Sequencher (Takara, Japan) and Autoassembler program (Perkin-Elmer). Homologous sequences were obtained from the database using the BLAST program at the Arabidopsis Information Resource (http://www.arabidopsis.org/blast/). Analyses of the sequences of the cloned fragments were performed with the GENETYX MAC program (version 10.0; Software Development Co., Japan).

Mapping:
DNA fragments were mapped by the cleaved amplified polymorphic sequences (CAPS) method developed by KONIECZNY and AUSUBEL 1993 . In brief, DNA from F₂ plants derived from a cross between Landsberg erecta (Ler) and Col were amplified by PCR. PCR was carried out at 94° for 3 min, followed by 50 cycles of 94° for 1 min, 50° for 1 min, and 72° for 2 min. At the end of the 50 cycles, the samples were incubated at 72° for another 10 min for complete extension. The primers for mapping the GL1 locus were primers 3 and 4. The primers for mapping the 107-bp fragment were primer 7, 5'-TTCAAATATCTAATCCAAACAA-3', and primer 8, 5'-ATATTAAAAAGGACCAAGAGTT-3'. The primers for mapping Atpk7 were primer 9, 5'-TTATGTTTGCTTTTGAGGTTTG-3', and primer 10, 5'-CGGTGACAGACGGTTACTTTAG-3'. The amplified fragment at the GL1 locus and the fragment at the Atpk7 locus were digested with TaqI for 1 hr at 65°, and the amplified fragment containing the 107 bp was digested with BfaI for 1 hr at 37°. The reaction mixtures were subjected to agarose gel electrophoresis to observe the polymorphisms. The linkage analysis was done using the MAPL mapping program, which was downloaded from http://peach.ab.a.u-tokyo.ac.jp/~ukai/mapl1.htm.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We have isolated a gl1-3 mutant from carbon ion-mutagenized M2 plants and further found, by Southern blot and PCR analyses, that a break occurred within exon 3 of the GL1 gene and then a rearrangement took place in this mutant (SHIKAZONO et al. 1998A , accession no. AB006078; SHIKAZONO et al. 1998B ). As a first step to analyze the structural alteration of gl1-3 in more detail, the HindIII fragment containing the GL1 gene was cloned from the wild type (Col; Fig 1A). This fragment was sequenced and was found to be 12,367 bp in length. PCR-amplified probes that hybridize to the upstream and downstream regions of the break were designated probe I and probe II, respectively (Fig 1A). The results of cloning and sequencing of the hybridizing fragments for each probe in the mutant are schematically shown in Fig 1B and Fig C. The EcoRI fragment cloned by probe I was 10,209 bp and the HindIII fragment cloned by probe II was 7230 bp in length. From the homology search, it was found that the sequence upstream of the break at the third exon of GL1 was rejoined to the region containing the Atpk7 gene (HAYASHIDA et al. 1992 , accession no. D10910). The sequence downstream of the break at the third exon of GL1 was found to be rejoined to the region downstream of the Atpk7 gene, with an unknown 107-bp sequence inserted in between these two regions. No point mutations were found in these fragments cloned from gl1-3.

Southern blot analysis using the 107-bp fragment as a probe detected a single HindIII fragment (data not shown), indicating that this fragment is not a repetitive sequence. The hybridizing HindIII fragment was cloned and then sequenced (Fig 2A). The clone was 3047 bp in length and contained the complete 107-bp sequence. To locate this 3047-bp HindIII fragment, a marker for the CAPS method was generated within this region. A BfaI fragment of 561 bp was specific to Col, and BfaI fragments of 400 and 160 bp in length were specific to Ler. Using this polymorphism, the 107-bp fragment was found to lie on chromosome 2 between markers PHYB and COP1 (Fig 2B). PHYB and COP1 map at 34.4 cM and at 63.3 cM (http://www.arabidopsis.org/maps.html), respectively. The location of the 107-bp fragment was further confirmed from the physical map position of the sequenced BAC F8N16 (accession no. AC005727), in which a sequence identical to the 3047-bp HindIII fragment was found.

View larger version (15K): In this window In a new window Download PPT slide   Figure 2. Characterization of the inserted 107-bp fragment. (A) Cloning and sequencing of the HindIII fragment containing the 107-bp sequence from the wild type (Col). The 107-bp region is shown as a solid box. Primers of the CAPS marker for this region are shown as solid arrowheads. E, EcoRI; H, HindIII; Xb, XbaI. (B) Linkage of the 107-bp region to other CAPS markers. Forty-six F2 plants were subjected to the mapping experiment.

To determine whether this 107-bp fragment was cut out from the original site and then inserted between the downstream region of GL1 and the downstream region of the Atpk7 gene in the gl1-3 mutant, PCR was carried out using the same primer set for amplifying the CAPS marker of the 107-bp fragment. It was found that the amplified fragment from the gl1-3 mutant was 100 bp smaller than that from the wild type, which is consistent with the assumption of the transfer of this 107-bp region in gl1-3 (data not shown). Sequencing these PCR products confirmed that 119 bp, which contained the 107-bp insertion sequence, was deleted from this region (Fig 3A). The sequence of the deleted region indicated that the broken ends recombined with 1 bp of homology. Imperfect inverted repeats were found at the rejoined site.

View larger version (55K): In this window In a new window Download PPT slide   Figure 3. Nucleotide sequence of the rearranged sites in carbon ion-induced mutants. Short homologous sequences involved in rejoining are shown as gray letters. Boxed sequences were deleted during the rejoining process. Potential topoisomerase I recognition sequences (thick lines) and cleavage sites (arrowheads) are shown above the sequence. Direct repeats are shown as black horizontal arrows, whereas imperfect direct and/or inverted repeats are shown as gray horizontal arrows below the sequence. (A) Nucleotide sequence of the rearranged site in gl1-3. (B) Nucleotide sequence of the rearranged site in tt4(C1). (C) Nucleotide sequence of the rearranged site in ttg1-21.

Since the location of the Atpk7 gene was unknown, a novel CAPS marker for the Atpk7 locus was generated. After PCR and digestion with TaqI, two fragments, 430 and 290 bp in length, were observed in Col, while a single undigested fragment was found in Ler. Out of 46 F₂ plants, no recombination was observed between the Atpk7 and the GL1 marker. It is shown that GL1 locates at 48.4 cM on chromosome 3 (http://www.arabidopsis.org/maps.html). The result of mapping suggests that Atpk7 lies near GL1 on chromosome 3. Consistent with this conclusion, the Atpk7 gene was recently placed on the BAC MMJ24 (accession no. AB-025626). The position of the Atpk7 gene is indicated to be a few hundred kilobase pairs north of GL1 (http://www.arabidopsis.org/maps.html).

Nucleotide sequences of the rejoined sites in gl1-3 are shown in Fig 3A. Comparing the sequence of the rejoined site of the upstream sequence of the GL1 and the Atpk7 gene with the sequences of these genes in the wild type demonstrated that 4 bp of homology (TGAT) was involved when DNA strand breaks were rejoined. In the case of the rejoined site of the downstream sequence of the GL1 gene, the 107-bp fragment, and the downstream region of the Atpk7 gene, it was found that homologies of 1 bp (A) and 3 bp (CAA) were used to rejoin these fragments. Further, 10 bp and 7 bp were deleted during this rejoining process around the breakpoints at the GL1 and the Atpk7 gene, respectively. One base pair and 7 bp were also deleted from the ends of the fragment derived from chromosome 2. Potential topoisomerase I cleavage sites were found at or around the breakpoints. At the rejoined sites of the 107-bp insertion, imperfect inverted repeats were found.

Analyses of the rejoined site were also done with carbon ion-induced tt4(C1) and ttg1-21 mutants. From PCR analysis, the tt4(C1) and ttg1-21 mutants were found to have a break within the first and the second exon of their genes, respectively, suggesting that a translocation or an inversion took place in each mutant (SHIKAZONO et al. 1998B ). To analyze the structural alteration of these mutations in more detail, the rejoined sites were cloned by TAIL PCR and were sequenced.

In the case of tt4(C1), 1.3-kbp and 1-kbp fragments containing the upstream and downstream regions of the TT4 gene (accession no. M20308), respectively, were cloned by TAIL PCR (Fig 4B). Sequencing of these fragments revealed that the broken ends of the TT4 gene were rejoined to broken ends generated at the fourth exon of a putative gene on BAC F18O22 (accession no. AL163817; Fig 3B and Fig 4B). The position of the breakpoint was estimated to be a few hundred kilobase pairs south of the TT4 gene (http://www.arabidopsis.org/maps.html). Comparison of the nucleotide sequences flanking the rejoined sites with those of the wild type revealed that 1 bp was deleted in both regions during the rejoining process. Further, the broken site at the TT4 gene contained a potential topoisomerase I cleavage site. There was an imperfect direct repeat at the rejoined site containing the downstream region of the TT4 gene. The rejoined sites revealed that no homologous end was used and that no point mutations occurred during the rejoining process.

View larger version (58K): In this window In a new window Download PPT slide   Figure 4. Structural alterations induced by carbon ions. (A) Structural alterations in gl1-3. The map of these regions was determined by sequencing and the CAPS method. Exons are shown as black boxes. The insertion of the 107-bp sequence is represented by a gray box. The relative positions of the centromere of chromosome 3 are shown as CEN3. From the physical map of Arabidopsis, it was found that the Atpk7 gene lies a few hundred kilobase pairs north of the GL1 gene. It should be noted that, since the direction of transcription of the two genes is not known, it is equally likely that the direction of transcription of both genes is away from the centromere. (B) Structural alterations in tt4(C1). Exons are shown as solid boxes. Regions cloned by TAIL PCR are shown below at the rejoined sites as thick solid lines. It was found that the putative gene lies a few hundred kilobase pairs south of the TT4 gene. The relative positions of the centromere are shown as CEN5. The direction of the TT4 gene shown is based on the fact that the direction of transcription of the putative gene is toward the centromere. (C) Structural alterations in ttg1-21. Exons are shown as solid boxes. Regions cloned by TAIL PCR are shown below at the rejoined sites as thick solid lines. The relative positions of the centromeres of chromosomes 3 and 5 are shown as CEN3 and CEN5, respectively. The direction of the TTG1 gene shown is based on the fact that the direction of transcription of the putative gene on chromosome 3 is away from the centromere, and the fact that a dicentric chromosome leads to lethality.

In the case of ttg1-21, 600-bp and 700-bp fragments containing the upstream and downstream regions of the TTG1 gene (accession no. AJ133743), respectively, were cloned by TAIL PCR (Fig 4C). Sequencing of these fragments revealed that the broken ends of the TTG1 gene were rejoined to broken ends generated at the first intron of a putative gene on BAC F1C9 (accession no. AC011664; Fig 3C and Fig 4C). The TTG1 gene and this putative gene lie at 50 cM on chromosome 5 and 6 cM on chromosome 3, respectively (http://www.arabidopsis.org/maps.html). A homology of 1 bp was found at the rejoined site of the upstream region of TTG1, whereas no homologous end was found at the rejoining site of the downstream region of the TTG1. There were imperfect direct repeats at the rejoined site containing the upstream region of the TTG1 and direct and inverted repeats at the rejoined site containing the downstream region of the TTG1. It was also revealed that 3 bp were deleted at TTG1 and that 2 bp (TA) were inserted at one of the rejoined sites, and that no point mutations were induced during the rejoining process.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

A sequence analysis of the carbon ion-induced mutations was carried out in our study. The structural alteration of the mutations is summarized in Fig 4. In the case of gl1-3, from the fact that both the upstream and downstream regions of the GL1 gene were rejoined to fragments at the Atpk7 locus and the fact that the Atpk7 locus locates quite near to GL1 on chromosome 3, it was concluded that an inversion took place in this mutant (Fig 4A). The size of the inverted fragment was a few hundred kilobase pairs. Furthermore, it was demonstrated that a 107-bp fragment from chromosome 2 was inserted at the rejoining site of the downstream region of the GL1 gene and the 3' downstream region of the Atpk7 gene. In the case of tt4(C1), since the location of TT4 is 29.5 cM on chromosome 5, and the position of the rejoined site is a few hundred kilobase pairs south of the TT4 gene, it was concluded that a fragment of a few hundred kilobase pairs in size was inverted in this mutant (Fig 4B). Similarly, since the breakpoint within TTG1 and the other breakpoint locate around 50 cM on chromosome 5 and 6 cM on chromosome 3, respectively, this mutant seems to contain a reciprocal translocation between chromosomes 3 and 5 (Fig 4C). It has been reported that low-LET radiation, such as X-ray and fast neutron irradiation, could simultaneously cause deletion, inversion, and/or insertion in the plant genome (SHIRLEY et al. 1992 ). Our study showed that similar large and complex structural alterations could be induced by ion particles. Since high-LET radiation is much more mutagenic than low-LET radiation, ion particle irradiation can be a powerful tool for inducing genetic rearrangements in plants.

Rejoining of double-strand breaks via NHEJ was assumed to operate at the broken ends after X-ray and fast neutron irradiation of Arabidopsis seeds (SHIRLEY et al. 1992 ). Several characteristics of NHEJ were also found from the sequence of the rejoining site after carbon ion irradiation. Short homologies of 1–4 bp were involved in 5 of 8 rejoined sites and small deletions were present at 7 of 8 breakpoints (Fig 3). Involvement of short homologies and/or deletions has been reported at a similar frequency during the rejoining process in plant cells (GHEYSEN et al. 1991 ; MAYERHOFER et al. 1991 ; GORBUNOVA and LEVY 1997 ; TAKANO et al. 1997 ). Direct repeats and imperfect inverted repeats were observed at the rejoined site in our study (Fig 3), indicating that these sequences may have played some role in rejoining the DNA ends. Insertions of 2 and 107 bp were also observed after carbon ion exposure (Fig 3). This is consistent with previous findings that 30–40% of the sites rejoined via NHEJ in plant cells are accompanied by filler DNAs (MAYERHOFER et al. 1991 ; GHEYSEN et al. 1991 ; GORBUNOVA and LEVY 1997 ; TAKANO et al. 1997 ). Consequently, these results strongly suggest that the NHEJ pathway of double-strand breaks operates after plant cells are exposed to both low- and high-LET radiation.

The underlying mechanisms of the formation of DNA rearrangements are largely unknown in plants. Potential topoisomerase I cleavage sites (5' G/C-A/T-T 3'; BEEN et al. 1984 ) were observed around the breakpoints in this study (Fig 3). This result suggests the involvement of topoisomerase I in generating DNA rearrangements. The involvement of topoisomerase I in DNA rearrangements was also implied by sequence analysis after -irradiation of human fibroblast cells (FORRESTER and RADFORD 1998 ) and after particle bombardment of Arabidopsis (SAWASAKI et al. 1998 ). It has been reported that topoisomerase I cleavage was stimulated by the presence of a nick or a gap on the opposite strand several bases downstream of the cleavage site (POURQUIER et al. 1997 ; KINGMA and OSHEROFF 1998 ). Since ion particles would frequently produce nicks or gaps, cleavage of topoisomerase I would have often resulted in a double-strand break and this might have served as a template for NHEJ. ZHU and SCHIESTL 1996 reported that the overexpression of topoisomerase I in yeast led to a 6- to 12-fold increase of nonhomologous integration of transformed DNA compared with that in yeast with repression or deletion of topoisomerase I, suggesting that topoisomerase I plays some role in integration. Further study is needed to elucidate the initial DNA damage and the subsequent rejoining process that leads to large structural changes such as deletion, inversion, and translocation in plants.

Since high-LET radiation would cause dense ionization within a localized area in the nucleus, the damage specifically induced by high-LET radiation, referred to as "locally multiply damaged sites" or "clustered damage," is assumed to occur (WARD 1994 ; GOODHEAD 1995 ). It seems plausible to hypothesize that, due to this type of damage, some base changes occur near the broken end after the cells are exposed to ion particles. However, no such changes were observed in our study. This result could be explained by assuming that, even though there were some sites of base damage near the broken ends, these damaged sites were removed by exonuclease activity during the rejoining process. Since deletions of several base pairs are observed at broken ends, this explanation seems plausible. Alternatively, it is possible that such mutations are rarely detected simply because the clustered damage is mainly nonreparable, resulting in cell death.

It was suggested from the efficient induction of large and complex structural alterations that ion particle irradiation is a powerful tool for inducing valuable null mutants for plant genetics. However, it remains unclear whether there is any difference between the nature of the mutations induced by high- and low-LET radiation in plants. It was reported that the distribution of double-strand breaks induced after high-LET radiation was nonrandom both theoretically and experimentally in mammalian cells on scales of kilobase to megabase pairs. (HOLLY and CHATTERJEE 1996 ; LOBRICH et al. 1996 ; RYDBERG 1996 ; SACHS et al. 1998 ). It seems plausible to assume that the characteristics of the structural alterations of the genome induced by high-LET radiation would reflect these nonrandom clustered distributions of the damage. Although no clustered mutations on a scale of 10 bp were found in our study, it seems possible that clustered structural alterations on scales of kilo- or megabase pairs are frequently induced by ion particles. Consistent with this assumption, the intragenic deletions containing multiple exons at the HPRT locus in CHO-K1 cells were reported to be more frequent after -particle exposure than after -irradiation (SCHWARTZ et al. 1994 ). Further, CHAUDHRY et al. 1996 found that the proportion of radon-induced mutants with relatively small deletions involving the TK locus alone or the TK locus with one or both of the two closest markers was significantly higher than in the case of X-ray-induced mutants. From these results, one may speculate that ion particle irradiation efficiently induces moderate-sized deletions (of several kilobases) in plants, which could be directly applied to cloning a gene using molecular techniques such as genomic subtraction or representative difference analysis (SUN et al. 1992 ; LISITSYN et al. 1993 ). Thus, it appears to be quite important to comparatively analyze the high- and low-LET radiation-induced mutations at the molecular level in plants and to clarify the character of high-LET radiation-induced mutations.

	FOOTNOTES

¹ Present address: Matsudo, Chiba, 270-0035, Japan.

	ACKNOWLEDGMENTS

We thank Dr. A. Sakamoto, Dr. T. Hirose, and Mr. Y. Hase for helpful suggestions and discussions. We also thank Dr. A. R. Walker for providing the name of the ttg1 allele.

Manuscript received June 2, 2000; Accepted for publication October 2, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

BEEN, M. D., R. R. BURGESS, and J. J. CHAMPOUX, 1984  Nucleotide sequence preference at rat liver and wheat germ type I DNA topoisomerase breakage sites in duplex SV40 DNA. Nucleic Acids Res. 12:3097-3114[Abstract/Free Full Text].

BRUGGEMANN, E., K. HANDWERGER, C. ESSEX, and G. STORZ, 1996  Analysis of fast neutron-generated mutants at the Arabidopsis thaliana HY4 locus. Plant J. 10:755-760[Medline].

CHAUDHRY, M. A., Q. JIANG, M. RICANATI, M. F. HORNG, and H. H. EVANS, 1996  Characterization of multilocus lesions in human cells exposed to X radiation and radon. Radiat. Res. 145:31-38[Medline].

FORRESTER, H. B. and I. R. RADFORD, 1998  Detection and sequencing of ionizing radiation-induced DNA rearrangements using the inverse polymerase chain reaction. Int. J. Radiat. Biol. 74:1-15[Medline].

GHEYSEN, G., R. VILLARROEL, and M. C. MONTAGU, 1991  Illegitimate recombination in plants: a model for T-DNA integration. Genes Dev. 5:287-297[Abstract/Free Full Text].

GOODHEAD, D. T., 1995  Molecular and cell models of biological effects of heavy ion radiation. Radiat. Environ. Biophys. 34:67-72[Medline].

GORBUNOVA, V. and A. A. LEVY, 1997  Nonhomologous DNA end joining in plant cells is associated with deletions and filler DNA insertions. Nucleic Acids Res. 25:4650-4657[Abstract/Free Full Text].

HAYASHIDA, N., T. MIZOGUCHI, K. YAMAGUCHI-SHINOZAKI, and K. SHINOZAKI, 1992  Characterization of a gene that encodes a homologue of protein kinase in Arabidopsis thaliana.. Gene 121:325-330[Medline].

HOLLY, W. R. and A. CHATTERJEE, 1996  Clusters of DNA damage induced by ionizing radiation: formation of short DNA fragments. I. Theoretical modeling. Radiat. Res. 145:188-199[Medline].

KINGMA, P. S. and N. OSHEROFF, 1998  The response of eukaryotic topoisomerases to DNA damage. Biochim. Biophys. Acta 1400:223-232[Medline].

KOHLI, A., S. GRIFFITHS, N. PALACIOS, R. M. TWYMAN, and P. VAIN et al., 1999  Molecular characterization of transforming plasmid rearrangements in transgenic rice reveals a recombination hotspot in the CaMV 35S promoter and confirms the predominance of microhomology mediated recombination. Plant J. 17:591-601[Medline].

KONIECZNY, A. and F. M. AUSUBEL, 1993  A procedure for mapping Arabidopsis mutations using co-dominant ecotype-specific PCR-based markers. Plant J. 4:403-410[Medline].

KRAFT, G., M. KRAMER, and M. SCHOLZ, 1992  LET, track structure and models. Radiat. Environ. Biophys. 31:161-180[Medline].

LAUFS, P., D. AUTRAN, and J. TRAAS, 1999  A chromosomal paracentric inversion associated with T-DNA integration in Arabidopsis. Plant J. 18:131-139[Medline].

LIU, Y-G., N. MITSUKAWA, T. OOSUMI, and R. F. WHITTIER, 1995  Efficient isolation and mapping of Arabidopsis thaliana T-DNA insert junctions by thermal asymmetric interlaced PCR. Plant J. 8:457-463[Medline].

LISITSYN, N. A., J. A. SEGRE, and M. WIGLER, 1993  Cloning the differences between two complex genomes. Science 259:946-951[Abstract].

LÖBRICH, M., P. K. COOPER, and B. RYDBERG, 1996  Non-random distribution of DNA double strand breaks induced by particle irradiation. Int. J. Radiat. Biol. 70:493-503[Medline].

MAYERHOFER, R., Z. KONCZ-KALMAN, C. NAWARATH, G. BAKKEREN, and A. CRAMERI et al., 1991  T-DNA integration: a mode of illegitimate recombination in plants. EMBO J. 10:697-704[Medline].

MEI, M., H. DENG, Y. LU, C. ZHUANG, and Z. LIU et al., 1994  Mutagenic effects of heavy ion radiation in plants. Adv. Space Res. 10:363-372.

MURRAY, M. G. and W. F. THOMPSON, 1980  Rapid isolation of high molecular weight plant DNA. Nucleic Acids Res. 8:4321-4325[Abstract/Free Full Text].

POURQUIER, P., A. A. PILON, G. KOHLHAGEN, A. MAZUMDER, and A. SHARMA et al., 1997  Trapping of mammalian topoisomerase I and recombinations induced by damaged DNA containing nicks or gaps. J. Biol. Chem. 272:26441-26447[Abstract/Free Full Text].

RYDBERG, B., 1996  Clusters of DNA damage induced by ionizing radiation: formation of short DNA fragments. II. Experimental detection. Radiat. Res. 145:200-209[Medline].

SACHS, R. K., D. J. BRENNER, P. HAHNFELDT, and L. R. HLATKYS, 1998  A formalism for analysing large-scale clustering of radiation-induced breaks along chromosomes. Int. J. Radiat. Biol. 74:185-206[Medline].

SALOMON, S. and H. PUTCHA, 1998  Capture of genomic and T-DNA sequences during double-strand break repair in somatic plant cells. EMBO J. 17:6086-6095[Medline].

SANGER, F., S. NICKLEN, and A. R. COULSON, 1977  DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463-5467[Abstract/Free Full Text].

SAWASAKI, T., M. TAKAHASHI, N. GOSHIMA, and H. MORIKAWA, 1998  Structures of transgene loci in transgenic Arabidopsis plants obtained by particle bombardment: junction regions can bind to nuclear matrices. Gene 218:27-35[Medline].

SCHWARTZ, J. L., J. ROTMENSCH, J. SUN, J. AN, and Z. XU et al., 1994  Multiplex polymerase chain reaction-based deletion analysis of spontaneous, gamma ray- and alpha-induced hprt mutants of CHO-K1 cells. Mutagenesis 9:537-540[Abstract/Free Full Text].

SHIKAZONO, N., A. TANAKA, Y. YOKOTA, H. WATANABE, and S. TANO, 1998a  Nucleotide sequence of the GLABROUS1 gene of Arabidopsis thaliana ecotype Columbia. DNA Seq. 9:177-181[Medline].

SHIKAZONO, N., Y. YOKOTA, A. TANAKA, H. WATANABE, and S. TANO, 1998b  Molecular analysis of carbon ion-induced mutations in Arabidopsis thaliana.. Genes Genet. Syst. 73:173-179[Medline].

SHIRLEY, B. W., S. HANLEY, and H. M. GOODMAN, 1992  Effects of ionizing radiation on a plant genome: analysis of two Arabidopsis transparent testa mutations. Plant Cell 4:333-347[Abstract/Free Full Text].

SMITH, H. H., 1972 Comparative genetic effects of different physical mutagens in higher plants, Induced Mutations and Plant Improvement, pp. 75–93. International Atomic Energy Agency, Vienna.

SUN, T., H. M. GOODMAN, and F. M. AUSUBEL, 1992  Cloning the Arabidopsis GA1 locus by genomic subtraction. Plant Cell 4:119-128[Abstract/Free Full Text].

TAKANO, M., H. EGAWA, J. IKEDA, and K. WAKASA, 1997  The structures of integration sites in transgenic rice. Plant J. 11:353-361[Medline].

TANAKA, A., N. SHIKAZONO, Y. YOKOTA, H. WATANABE, and S. TANO, 1997  Effects of heavy ions on the germination and survival of Arabidopsis thaliana.. Int. J. Radiat. Biol. 72:121-127[Medline].

WARD, J. F., 1994  The complexity of DNA damage: relative to biological consequences. Int. J. Radiat. Biol. 70:427-432.

ZHU, J. and R. H. SCHIESTL, 1996  Topoisomerase I involvement in illegitimate recombination in Saccharomyces cerevisiae.. Mol. Cell. Biol. 16:1805-1812[Abstract].

This article has been cited by other articles:

				N. Shikazono, C. Suzuki, S. Kitamura, H. Watanabe, S. Tano, and A. Tanaka Analysis of mutations induced by carbon ions in Arabidopsis thaliana J. Exp. Bot., February 1, 2005; 56(412): 587 - 596. [Abstract] [Full Text] [PDF]

				A. Sakamoto, V. T. T. Lan, Y. Hase, N. Shikazono, T. Matsunaga, and A. Tanaka Disruption of the AtREV3 Gene Causes Hypersensitivity to Ultraviolet B Light and {gamma}-Rays in Arabidopsis: Implication of the Presence of a Translesion Synthesis Mechanism in Plants PLANT CELL, September 1, 2003; 15(9): 2042 - 2057. [Abstract] [Full Text] [PDF]

				N. Shikazono, Y. Yokota, S. Kitamura, C. Suzuki, H. Watanabe, S. Tano, and A. Tanaka Mutation Rate and Novel tt Mutants of Arabidopsis thaliana Induced by Carbon Ions Genetics, April 1, 2003; 163(4): 1449 - 1455. [Abstract] [Full Text] [PDF]

				A. Tanaka, A. Sakamoto, Y. Ishigaki, O. Nikaido, G. Sun, Y. Hase, N. Shikazono, S. Tano, and H. Watanabe An Ultraviolet-B-Resistant Mutant with Enhanced DNA Repair in Arabidopsis Plant Physiology, May 1, 2002; 129(1): 64 - 71. [Abstract] [Full Text] [PDF]

				A. Tanaka, Y. Kobayashi, Y. Hase, and H. Watanabe Positional effect of cell inactivation on root gravitropism using heavy-ion microbeams J. Exp. Bot., April 1, 2002; 53(369): 683 - 687. [Abstract] [Full Text] [PDF]