Induction of chromosome aberrations in unirradiated

International Journal of Radiation Biology

ISSN: 0955-3002 (Print) 1362-3095 (Online) Journal homepage: http://www.tandfonline.com/loi/irab20

Induction of chromosome aberrations in unirradiated chromatin after partial irradiation of a cell nucleus G. Ludwików, Yun Xiao, R. A. Hoebe, N. A. P. Franken, F. Darroudi, J. Stap, C. H. Van Oven, C. J. F. Van Noorden & J. A. Aten To cite this article: G. Ludwików, Yun Xiao, R. A. Hoebe, N. A. P. Franken, F. Darroudi, J. Stap, C. H. Van Oven, C. J. F. Van Noorden & J. A. Aten (2002) Induction of chromosome aberrations in unirradiated chromatin after partial irradiation of a cell nucleus, International Journal of Radiation Biology, 78:4, 239-247, DOI: 10.1080/09553000110110086 To link to this article: http://dx.doi.org/10.1080/09553000110110086

Published online: 03 Jul 2009.

Submit your article to this journal

Article views: 17

View related articles

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=irab20 Download by: [UVA Universiteitsbibliotheek SZ]

Date: 09 May 2017, At: 04:30

int. j. radiat. biol 2002, vol. 78, no. 4, 239± 247

Induction of chromosome aberrations in unirradiated chromatin after partial irradiation of a cell nucleus ´ W†, YUN XIAO†§, R. A. HOEBE†, N. A. P. FRANKEN‡, F. DARROUDI§, G. LUDWIKO J. STAP†, C. H. VAN OVEN†, C. J. F. VAN NOORDEN† and J. A. ATEN†* (Received 13 July 2001; accepted 26 October 2001) Abstract. Purpose : It is generally accepted that chromosom e exchanges in irradiated cells are formed through interactions between separate DNA double-strand breaks (DSB). Here we tested whether nonirradiated DNA participates in the formation of chromosom e aberrations when complex DNA DSB are induced elsewhere in the nucleus. Materials and methods: Synchronized Chinese hamster cells containing an X chromosome with a late replicating q arm (Xq domain) were labelled with 1 2 5 I-iododeoxyuridine ( 1 2 5 IdUrd) in a period of S-phase when the vast majority of the X q domain was not replicating. DNA damage from 1 2 5 I decay was accumulated at the G1/S border while the cells were stored in liquid nitrogen. Decay of 125 I induced DSB in the immediate vicinity of the 125 I atom. Chromosome aberrations involving what is essentially the 125 I-free Xq domain were scored at the Ž rst mitosis after cell thawing. As a positive control, cells were treated with 125 IdUrd at a later period in S-phase when the Xq domain replicates, yielding a labelled Xq domain. Results : The 1 25 I-free X q domain exhibited chromosome aberrations (exchanges and fragments). The frequency of these aberrations was linearly dependent on the number of 1 2 5 I decays elsewhere in the cell nucleus. The eYciency of formation of chromosom e aberrations by the 1 2 5 I-free Xq domain was approximately half of that observed in the 1 2 5 I-labelled Xq domain. Conclusions : The involvement of the 1 2 5 I-free Xq domain in chromosom e aberrations suggests that DNA not damaged by the decay of incorporated 1 2 5 I can interact with damaged DNA, indicating the existence of an alternative pathway for the formation of chromosome aberrations.

1. Introduction There is considerable current interest in the mechanisms by which chromosome aberrations are generated. An early model (Serebrovski 1929) proposed that aberrations may result from contact between chromosomes followed by rupture when the attached chromosomes break away from each other. Soon, the ‘contact Ž rst’ model was rejected on the basis of *Author for correspondence; e-mail: [email protected] †Center for Microscopical Research, Department of Cell Biology and Histology, and ‡Department of Radiotherapy, Academic Medical Center, University of Amsterdam, PO Box 22 700, 1100 DE, Amsterdam, The Netherlands. §Medical Genetic Centre, Department of Radiation Genetics and Chemical Mutagenesis, Leiden University Medical Centre, Leiden, The Netherlands.

new experimental data. Six decades ago the classical ‘breakage-and-reunion’ model for the formation of chromosome exchanges (e.g. interchanges) was developed based on the assumption that two separate breaks on diVerent chromosomes interact (Sax 1941, Lea 1946). Later, these lesions were identiŽ ed as DNA double-strand breaks (DSB) (Bryant 1984, Natarajan 1990). It was assumed that chromosome aberrations result from errors in the DSB repair process. The classical model is interpreted in terms of the non-homologous end-joining repair of DSB ( Jeggo 1998). Twenty years ago the classical model was challenged by an alternative model postulating that chromosome exchanges were formed through interaction of a single DSB with undamaged DNA (Chadwick and Leenhouts 1978). In their model the authors introduced homologous recombination repair to explain the involvement undamaged DNA in the repair of DSB (Resnick 1976). Experimental evidence has been provided for the formation of chromosome aberrations in a way predicted by the classical model. This was demonstrated for DSB induced by c-radiation (Cornforth 1990) and bleomycin (Wang et al. 1997). Several studies testing the alternative model showed that DSB-free DNA was not involved in the formation of chromosome exchanges. During homologous recombination repair of DSB induced by the rare-cutting endonuclease I-Sce1, formation of chromosome exchanges was suppressed (Richardson et al. 1998). Moreover, chromosomes containing DSB induced by c-radiation produced no, or very few, exchanges with undamaged chromosomes (Cornforth 1990). However, the predominantly linear dose–response relationship observed for the formation of chromosome aberrations induced by ultrasoft X-rays suggests that the alternative pathway does produce chromosome exchanges (Thacker et al. 1986, Goodhead 1989, Goodhead et al. 1993, GriYn et al. 1998). DSB induced by ultrasoft X-rays are often complex as they are accompanied locally by other local DNA damage and thus might diVer in structure, processing

International Journal of Radiation Biology ISSN 0955-3002 print/ISSN 1362-3095 online © 2002 Taylor & Francis Ltd http://www.tandf.co.uk/journals DOI: 10.1080/0955300011011008 6

240

G. Ludwiko´w et al.

and biological consequences from DSB induced by c-rays, bleomycin and endonucleases. In the present study, we investigated whether complex DNA damage induced by 125 I decay could engage chromatin without DSB in the formation of chromosome aberrations. In cell nuclei labelled with 125 IdUrd, low-energy Auger electrons emitted during the decay of 125 I are absorbed locally and induce complex DSB conŽ ned to the DNA at the incorporation site (Charlton and Booz 1981, Martin and Haseltine 1981, Charlton 1986). We incorporated 125 I into a restricted region of the cell nucleus and examined a selected ‘125 I-free’ chromosome domain for the formation of chromosome aberrations. 2. Materials and methods 2.1. Cell culture and cell synchronization The HA-1 cell line (Yang et al. 1966) was grown in Eagle’s minimal essential medium with Eagle’s salts (MEM, Gibco Brl, Breda, The Netherlands), supplemented with 10% foetal calf serum (Gibco Brl) and antibiotics in a 2% CO2 atmosphere at 37ß C. Synchronization of cells was performed by incubating asynchronous cell cultures with 1.5 mm hydroxyurea (Sigma, St Louis MO, USA) for 16 h. After washing in prewarmed PBS (3 3 ) and further culturing of the cells in normal culture medium, the distribution of cells throughout the cell cycle was investigated at diVerent post-synchronization times by DNA  ow cytometry. Nuclei were prepared according to Vindelov et al. (1983). 2.2. Labelling and detection of chromosome replication Labelling and detection of chromosome replication were performed as described by Ludwikow et al. (2000). Brie y, 10 mm 5-iodo-2¾ -deoxyuridine (IdUrd; Sigma) was added for 20 min to asynchronous cultures and to synchronous cultures at 15 min or at 3 h after release from the synchronization block. After labelling, cells were washed with prewarmed PBS (33 ) and re-incubated in normal medium. Mitotic cells were collected up to 12 h after labelling, at 1.5–2-h intervals, using colcemid (0.1 mg mlÕ 1 ). Slides with metaphase spreads were prepared according to standard cytogenetical procedures. Metaphase spreads were denatured for 2 min in 0.07 N NaOH. Following washing in PBS supplemented with 0.05% Tween 20 (PBT) and pre-incubation with bovine serum albumin (Sigma) (1:1 in PBS), the slides were incubated with mouse anti-BrdUrd antibodies that cross-react with incorporated IdUrd (BectonDickinson, Mountain View, CA, USA) (1:4 in PBT)

for 30 min at room temperature. Slides were washed in Tris high salt buVer for 7 min followed by washing in PBT for 10 min. After pre-incubation with normal goat serum (Dako, Glostrup, Denmark) (1:1 in PBS) for 5 min, Texas red (TXR)-conjugated goat antimouse antibodies ( Jackson, West Grove, PA, USA) (1:100 in PBT) were applied for 30 min. The slides were then stained with DAPI (0.1 mg mlÕ 1 in PBS) for 10 min followed by PBT washing, drying and Ž nally mounted in Vectashield mounting medium (Vector, Burlingame, CA, USA). Analysis of replication timing of Xq during S-phase was performed as described by Savage et al. (1984), based on the kinetics of its replication signal. One hundred metaphases were studied for each sample and two observations were recorded for each metaphase: (1) presence or absence of replication staining, and (2) localization of replication staining on the Xq . To estimate the background 125 IdUrd incorporation in Xq during early S-phase, staining methods were used that result in anti-IdUrd  uorescence signals proportional to the amount of incorporated IdUrd (for a review, see Dolbeare 1995). The  uorescence intensity of replication label on the q-arm (with no detectable replication signals) was compared with the  uorescence intensity on the p-arm (with strong replication signal). Images of DAPI (genomic staining) and TXR (staining of replication label) signals on metaphase cells were registered with a cooled CCD camera (HISIS24; Lambert Instruments). The DAPI image was used to construct a mask for the TXR image. After subtraction of local background, the average TXR intensities were calculated for the p-arm stained for replication and the unstained q-arm. Image processing was performed using SCIL-Image software (University of Amsterdam and TNO-TPD). 2.3. Metaphase preparations after selective incorporation of 125 IdUrd Cells were incubated with 3.7–7.5 kBq mlÕ 1 5-[ 125 I]iodo-2¾ -deoxyuridine (125 IdUrd, speciŽ c activity 74 TBq mmÕ 1 ; Amersham, Amersham, UK) for 20 min either at 15 min or 3 h after release from the hydroxyurea-induced synchronization block. After washing with PBS (33 ), the cells were incubated in normal culture medium up to 10 h postsynchronization and synchronized again with hydroxyurea for 16 h (1.5 mm). Subsequently, the cells were trypsinized, resuspended in culture medium and counted. Of the cell suspension, 1 ml was taken to measure 125 I-activity concentration (Gamma counter, Packard, Groningen, The Netherlands) and to calculate dose-rates in the cells. The remaining cells were

Chromosome aberrations in unirradiated chromatin after partial irradiation of the nucleus

241

frozen in liquid nitrogen in culture medium with 20% DMSO. The dose-rate at the time of freezing was 0.2–0.6 125 I decay cellÕ 1 hÕ 1 . Cells were stored in liquid nitrogen for 2–40 days. After thawing, the cells were plated and cultured for 12–48 h, until Ž rst mitoses appeared, depending on the accumulated amounts of 125 I decay. After 6 h of colcemid treatment, the cells were Ž xed and slides with metaphase spread were prepared by the standard air-dry method. 2.4. Analysis of chromosomal aberrations formed by the Xq Chromosomal aberrations involving the Xq were analysed by  uorescence in situ hybridization (FISH) (Pinkel et al. 1986) using a Chinese hamster chromosome-speciŽ c probe (Xiao et al. 1996). Chromosome aberrations involving the Xq domain were scored and classiŽ ed as exchanges and excess fragments. Complex aberrations were evaluated as combinations of simple exchanges. 3. Results 3.1. Theoretical design of the experiment The strategy of our experiments is presented in Ž gure 1. In the interphase cell nucleus, chromosomes and their subregions are concentrated in separate non-intermingling domains (Manuelidis 1985, Visser and Aten 1999, Cremer and Cremer 2001). These chromosome subregions replicate during diVerent periods of S-phase (Savage et al. 1984). We took advantage of the diVerences in replication timing to select a chromosome domain that replicated during late S-phase (e.g. the Xq of HA-1 cells). This selected domain was visualized by  uorescence in situ hybridization (FISH) (Ž gure 1A). DNA damage was induced by 125 I decay. By applying 125 IdUrd in early or in late S-phase, we prevented or promoted induction of DSB in the selected chromosome domain, whereas in both cases other chromosome domains always incorporated 125 IdUrd (Ž gure 1B, C). Thus, there were always chromosome domains surrounding the selected domain that contained DSB. To accumulate damage in the same phase of the cell cycle, cells were blocked at the border between G1 and S-phase of the cell cycle following 125 IdUrd incorporation and frozen in liquid nitrogen. After accumulation of damage, the selected chromosome domain was analysed for chromosome aberrations. 3.2. Xq of HA-1 cell The HA-1 cell line, derived from CHO (Chinese hamster ovary) cells, contains only one of the two X chromosomes characteristic of female mammalian

Figure 1. Scheme of the experiment designed to test whether non-irradiated chromatin participates in the formation of chromosome aberrations. (A) In the three-dimensional interphase nucleus (column n), a selected chromatin domain (green) is surrounded by other chromatin domains (blue). In metaphase cells (column m), the target domain is visualized by  uorescence in situ hybridization. (B) In ( positive) control experiments, the target domain and other chromatin domains incorporate 1 2 5 IdUrd (radioactive sign). As a result of 1 2 5 I-induced complex DNA damage at the sites of 1 2 5 I incorporation, the target domain interacts with other domains (column n, two arrows pointing in opposite directions). Chromosome aberrations formed by the 1 2 5 I-labelled target domain are detected in metaphase cells (column m, arrows). (C) The target domain does not incorporate 1 2 5 IdUrd, while surrounding domains do incorporate 1 2 5 IdUrd. The 125 IdUrd-free target domain can be involved in chromosome aberrations exclusively through DSB induced in domains elsewhere in the nucleus (columns n, m, arrows).

cells. The Xq arm is composed of transcriptionally inactive heterochromatin. The FISH probe used paints the Xq only (Ž gure 2). By labelling asynchronous cell populations for 20 min with the replication marker IdUrd followed by immuno uorescence staining of metaphases sampled up to 12 h after labelling, we found that there was no detectable early S-phase replication activity on Xq (Ž gure 3A). The Xq replication signal was observed in later S-phase. Figure 3B

242

G. Ludwiko´w et al.

Figure 4. Flow cytometric analysis of DNA distributions. HA-1 cells were analysed at 15 min (A) or 3 h (B) after release from the hydroxyurea-induced synchronization block, respectively. Figure 2. Detection of the Xq chromosome domain by  uorescence in situ hybridization. The FISH signal is shown in green (artiŽ cial colour). Xq is indicated by an arrow and is shown at a higher magniŽ cation in the inset. The digital micrographs were processed using Adobe Photoshop and Microsoft Powerpoint.

Figure 3. Replication timing of Xq . In micrographs (A) and (B), the replication signals on metaphase chromosomes are red (artiŽ cial colour); Xq is underlined and is shown at a higher magniŽ cation in the inset (upper right corner) ; the q (Xq ) and p arms are indicated. (A) Metaphase labelled in early S-phase showing no detectable replication signal in the Xq domain. (B) Metaphase labelled in late S-phase showing replication signals in the entire Xq .

shows the late replication signal covering the entire Xq . Treatment with 1.5 mm hydroxyurea for 16 h accumulated the cells at the G1/S border. After release from the hydroxyurea-induced block, cells entered the S-phase as a cohort. The distribution of cells in the cell cycle at 15 min and 3 h post-synchronization is presented in Ž gure 4A, B. Pulse-labelling with the replication label IdUrd at 15 min after release from the synchronization block resulted in unlabelled Xq domains in > 95% of the cells that were replicating at that time (Ž gure 5A). Cultures that had been pulselabelled with the replication label at 3 h after release from the synchronization block showed labelled Xq in 70%, on average, of all replicating cells (Ž gure 5B).

Figure 5. Fractions of replication-labelled mitoses. Fractions of labelled mitoses with (circles) and without replication signals on Xq (squares) obtained from cell cultures incubated with the replication label at 15 min (A) or 3 h (B) after release from the synchronization block, respectively. The x-axis indicates the time interval between labelling and the collection of metaphases.

3.3. Chromosome aberrations formed by the 125 and I-labelled Xq

125

I-free Xq

Four individual experiments were performed to investigate chromosome aberrations formed by the 125 I-free Xq and 125 I-labelled Xq (table 1). Each experiment consisted of at least Ž ve doses. More than 100 metaphases were analysed for each dose point. To exclude selectively incorporation of 125 IdUrd from the Xq domain while labelling chromatin domains surrounding the Xq domain, the cells were incubated with 125 IdUrd at 15–20 min after release from the hydroxyurea block. Alternatively, to incorporate 125 I into both the Xq domain and surrounding chromatin domains, the cells were incubated with 125 IdUrd at 3 h after release from the synchronization block. We found that the 125 I-free Xq formed chromosome exchanges and excess of fragments. Examples of the 125 I-free Xq aberrations detected by FISH are shown in Ž gure 6A–C. The frequencies of aberrations formed by the 125 I-free Xq and 125 I-labelled Xq (Ž gure 6D) are summarized in Ž gure 7. Exchanges and fragments formed by either the 125 I-free Xq or

Chromosome aberrations in unirradiated chromatin after partial irradiation of the nucleus Table 1.

Chromosome aberrations formed by the

125

I-free X q and

125

243

I-labelled Xq .

X q aberrations* 125

I decay per cell

Cells scored

Total

Exchanges

Excess fragments

125

IdUrd-free Xq (experiment 1) 0 12 25 50 75 200 125 IdUrd-free Xq (experiment 2) 0 52 58 81 102 125 IdUrd-labelled Xq (experiment 1) 12 30 50 80 100 125 IdUrd-labelled Xq (experiment 2) 0 23 48 51 67 77

1200 380 680 1030 900 100

4 7 20 55 76 26

(0.003) (0.018) (0.029) (0.053) (0.084) (0.26)

720 550 1053 675 315

6 35 106 96 67

(0.008) (0.064) (0.10) (0.14) (0.21)

300 200 210 100 300

15 21 16 20 99

(0.50) (0.105) (0.076) (0.20) (0.33)

430 100 194 123 150 100

5 16 41 36 47 33

(0.012) (0.16) (0.21) (0.29) (0.31) (0.33)

3 4 18 38 49 19

(0.003) (0.011) (0.026) (0.037) (0.054) (0.19)

1 3 2 17 27 7

(0.001) (0.008) (0.003) (0.017) (0.030) (0.07)

9 15 12 14 65

(0.030) (0.075) (0.057) (0.14) (0.22)

6 6 4 6 34

(0.02) (0.03) (0.02) (0.06) (0.11)

*Values per cell are in parentheses. 125

I-labelled Xq were approximately linearly dependent on the number of 125 I decays per cell. The 125 Ilabelled Xq were approximately twice as frequently involved in chromosome aberrations as the 125 I-free Xq , as indicated by the slopes of the curves (table 2). Complex aberrations were 4–5% and excess fragments 30–40% of all aberrations, on average. 4. Discussion The formation of chromosome aberrations by chromosome domains free of 125 I indicates that chromatin without radiation damage can be involved in chromosome aberrations when elsewhere in the cell nucleus DSB are induced. The near-linear doserelationships seen here between 125 I decays and DNA aberrations provide an additional indication that chromosome aberrations can be caused by a single damage event. The approximately twofold lower frequency of chromosome exchanges involving the 125 I-free Xq domain in comparison with that of the 125 I-labelled Xq domain suggests that damaged DNA

was involved in recombinational events with undamaged DNA. We considered the 125 I-free Xq domain to be free of radiation-induced DSB. This assumption was based on microdosimetry of 125 I and our current understanding of the arrangement of chromosomes in the interphase nucleus. Although damage to 125 Ilabelled DNA caused by low-energy Auger electrons is very severe, involving several base pairs, its spatial range is limited to a few nanometers from the site of 125 I incorporation. As chromosome domains within the interphase nucleus apparently form mutually exclusive units (Manuelides 1985, Visser and Aten 1999, Cremer and Cremer 2001), it seems unlikely that 125 I incorporated into a certain domain could introduce complex DNA damage in another domain. Therefore, the 125 I-free Xq domain is also free of complex DSB. Because the 125 I-free Xq domain may have incorporated small amounts of 125 IdUrd in early S-phase that were not detected visually, we quantitatively analysed immuno uorescently stained metaphases of cells labelled during early S-phase. As we

G. Ludwiko´w et al.

244

Figure 6. Examples of chromosom e aberrations formed by the 1 2 5 I-free X q domain (A–C) and the 1 2 5 I-labelled Xq domain (D). (A) Complete dicentric, (B) deletion, (C) complex exchange, (D) chromosom e aberration with three interstitial exchanges.

Figure 7. Relationship between frequency of chromosome aberrations formed by Xq and the number of 1 2 5 I decays per cell for the 1 25 I-free Xq domain (squares) and 1 2 5 I-labelled Xq domain (circles). (A) Total aberrations (exchanges plus excess of fragments), (B) chromosome exchanges. Table 2. Slopes of dose–response curves between 1 2 5 I decays and the frequency of chromosome aberrations formed by 125 I-labelled Xq and 1 2 5 I-free Xq . Aberration Total Exchanges

125

I-labelled Xq 3.53 10Õ 1.83 10Õ

Values are aberrations per decay.

3 3

125

I-free Xq

1.53 10Õ 1.03 10Õ

3 3

found that the IdUrd background signal in the late replicating Xq domain was approximately 50 times weaker than the positive IdUrd signal in the early replicating p-arm of the X chromosome (data not shown), it was concluded that traces of 125 IdUrd incorporated into Xq during early S-phase could have only induced negligible amounts of complex DSB in Xq . In addition to the large number of low-energy Auger electrons released during 125 I decay that induce local complex DNA damage, some Auger electrons of higher energy are produced. Moreover, 7% of the 125 I decays proceed by emission of c-rays. These types of radiation induce simple DSB with a frequency of 0.17 DSB for each 125 I decay (Charlton and Humm 1988) randomly distributed throughout the nucleus and may induce simple DSB in 125 I-free domains. As estimated from its relative length, Xq contains approximately 3% of the genomic DNA and has a probability of 0.005 per 125 I decay of receiving one of these random DSB. When 200 decays occur in the nucleus, which is the highest dose applied in this study, an average of only one random DSB is induced in Xq . Thus, the 125 I-free Xq domain can be considered as a domain that, by comparison with the rest of the hamster genome, is essentially free of DSB. Inhibition of DNA synthesis through interference

Chromosome aberrations in unirradiated chromatin after partial irradiation of the nucleus with metabolism of DNA or its precursors can block closure of DNA strand gaps produced during replication, providing an opportunity for conversion into double-strand gaps and chromosome aberration formation (Galloway et al. 1998). In our experiments, essentially all cells were synchronized at the border between G1 and S-phase for exposure to 125 I-decay. Therefore, chromosome aberrations resulting from inhibition of DNA synthesis may have been formed in the period directly following the release of the cells from the hydroxyurea block, aVecting chromatin replicating at the very beginning of S-phase. However, we analysed the formation of chromosome aberrations in the late-replicating Xq domain. Replication of this domain started several hours after removal of the blocking agent. It seems improbable that inhibition of DNA synthesis by hydroxyurea at the border between G1 and S-phase could have contributed to the formation of chromosome aberrations in the late replicating Xq domain. This is conŽ rmed by the low chromosome aberration frequencies observed in Xq in cells exposed to small numbers of decays. The involvement of chromatin without DSB in chromosome aberrations supports the alternative model for the formation of chromosome aberration proposed by Chadwick and Leenhouts (1978). According to this model reciprocal recombination between DNA with a DSB (the recipient) and undamaged DNA (the donor) that serves as a template leads to diVerent types of chromosome rearrangements. We have observed that the undamaged Xq was not only involved in exchanges with other chromosomes, but also in chromosome fragments. This suggests that the reciprocal recombination between damaged and undamaged DNA may result in discontinuity in the participating intact DNA molecule. The higher frequency of chromosome aberrations formation involving the 125 I-labelled Xq domain as compared with the 125 I-free Xq domain may result from its double role as recipient and donor in the reciprocal recombination process, in contrast to the 125 I-free Xq that, devoid of DSB, can serve only as a donor. Theoretically, DSB in the 125 I-labelled Xq domain can have formed chromosome aberrations by endjoining with other DSB. However, the linear dose– response relationship with 125 I decays and the double aberration frequency suggest that the same mechanism was involved in both cases. Investigating the formation of chromosome exchanges by carbon K ultrasoft X-rays, GriYn et al. (1998) observed a predominantly linear dosedependence of the yield of simple exchanges. These X-ray photons produce a single photoelectron with a track length of < 7 nm. The low energy and short

245

range imply that the probability of a track producing DSB in two diVerent chromosomes is practically zero. The authors indicated that if the classical model would apply, all exchanges should be multitrack events and have at least a dose-squared relationship. Therefore, they suggested that a single lesion introduced by ultrasoft X-rays can lead to a simple exchange, in line with the proposition that damaged DNA may interact with undamaged DNA. Moreover, Cucinotta et al. (2000) predicted a linear dose– response for simple chromosome exchanges formed in this way, i.e. by recombination between DSB and undamaged DNA, when they performed detailed computer studies simulating activity of a repair enzyme–DNA complex and competition between rejoining pathways. When Hofer and Bao (1995) and Ludwiko´w et al. (1996) studied the formation of micronuclei and chromosome aberrations in 125 IdUrd-labelled cells, they noticed a strong dependence of the response on the interval between the labelling pulse and freezing of cells for exposure to 125 I decay. Ludwiko´w et al. indicated that the observed change from low-LET-type action towards high-LET eVects could be related to homologous recombination repair activity involving interaction between damaged and undamaged DNA. The present study and that of GriYn et al. (1998) are in contrast with other studies of the mechanism of formation of chromosome aberrations. Cornforth (1990) did not Ž nd substantial formation of chromosome exchanges between undamaged chromosomes and irradiated chromosomes in metaphases from fusion nuclei produced by fusing untreated cells with cells exposed to 60 Co radiation. Richardson et al. (1998) and Richardson and Jasin (2000) analysed cells for chromosome aberrations at homologous loci in two chromosomes, one or both containing a cleavage site for the rare-cutting I-Sce1 endonuclease. In cells with a I-Sce1 cleavage site in one of the two loci, introduction of a DSB did not lead to chromosome translocations at the locus. On the other hand, when cleavage sites were introduced in both loci, reciprocal translocations were observed frequently. Cornforth (1990), Richardson et al. (1998) and Richardson and Jasin (2000) demonstrated that chromosome aberrations were formed as predicted by the classical model rather than the alternative mechanism. However, their studies diVer from ours with respect to the experimental methods used and the quality of DSB involved. Our system provides irradiated and unirradiated chromosome domains in one nucleus, obviating the need of cell fusion as was used by Cornforth (1990). In contrast to Richardson and Jasin (2000) who only could evaluate cell clones that

246

G. Ludwiko´w et al.

were selected for the presence of a functional neo+ gene, we analysed all mitotic cells after induction of damage. Moreover, DSB induced by 125 I are probably qualitatively very diVerent from DSB induced by either 60 Co in the experiments of Cornforth (1990) or the I-Sce1 endonuclease in the experiments of Richardson et al. (1998) and Richardson and Jasin (2000). 125 I decay produces a high concentration of energy that causes multiple damage in base pairs immediately adjacent to the disintegrating atom. The heavily damaged DNA may require, and even activate, speciŽ c repair mechanisms (Goodhead 1989, Goodhead et al. 1993, Ward 1994) that are diVerent from those engaged in the repair of DSB induced by 60 Co or the I-Sce1 endonuclease. Therefore, the pathway for the formation of chromosome aberrations may well depend on the nature of DSB implying that the alternative mechanism might be activated when DNA is subjected to complex damage. The classical model for the formation of chromosome exchanges, based on interactions between two DSB through non-homologous end-joining, depends on the proximity and density of DSB in the cell nucleus. Based on interaction between a single DSB and undamaged DNA, the alternative model represents a greater hazard for cells and organisms than the classical model, particularly at low levels of damage. Acknowledgements The authors thank Professor A. T. Natarajan, Dr A. E. Visser for their helpful comments, Dr G. C. Li, Memorial Sloan-Kettering Cancer Center, New York, for providing the HA-1 cell line, and the Swedish National Protection Institute, the Dutch Interuniversity Research Institute of Radiopathology and Radiation Protection (IRS) Grant No. 9.0.16, the European Community Grant No. FIGH-CT19999-00011, the John and Augusta Persson Foundation and the Maurits and Anna de Kock Foundation for Ž nancial support. References Bryant, P. E., 1984, Enzymatic restriction of mammalian cell DNA using Pvu II and Bam H1: evidence for the doublestrand break origin of chromosomal aberrations. International Journal of Radiation Biology, 46, 57–65. Chadwick, K. H. and Leenhouts, H. P., 1978, The rejoining of DNA double-strand breaks and a model for the formation of chromosom e rearrangements. International Journal of Radiation Biology, 33, 517–529. Charlton, D. E., 1986, The range of high LET eVects from 125 I decays. Radiation Research, 107, 163– 171. Charlto n, D. E. and Booz, J., 1981, A Monte Carlo treatment of the decay of 1 25 I. Radiation Research, 87, 10–23.

Charlton, D. E. and Humm, J. L., 1988, A method of calculating initial DNA strand breakage following the decay of incorporated 1 2 5 I. International Journal of Radiation Biology, 53, 353–365. Cornforth, M. N., 1990, Testing the notion of the one-hit exchange. Radiation Research, 121, 21–27. Cremer, T . and Cremer , C., 2001, Chromosom e territories, nuclear architecture and gene regulation in mammalian cells. Nature Reviews Genetics, 2, 292–301. Cucinotta, F. A., Nikjoo, H., O ’ Neill, P . and Goodhead, D. T., 2000, Kinetics of DSB rejoining and formation of simple chromosom e exchange aberrations. International Journal of Radiation Biology, 76, 1463– 1474. Dolbeare , F., 1995, Bromodeoxyuridine: a diagnostic tool in biology and medicine, Part I: Historical perspectives, histochemical methods and cell kinetics. Histochemical Journal, 27, 339– 369. Galloway, S. M., Miller, J. E., Armstrong, M. J., Bean, C. L., Skopek, T. R. and Nichols, W. W., 1998, DNA synthesis inhibition as an indirect mechanism of chromosome aberrations: comparison of DNA-reactive and non-DNA-reactive clastogens. Mutation Research, 400, 169– 186. Goodhead, D. T., 1989, The initial physical damage produced by ionizing radiations. International Journal of Radiation Biology, 56, 623– 634. Goodhead, D. T., T hacker , J. and Cox, R. 1993, EVects of radiation of diVerent qualities on cells: molecular mechanism of damage and repair. International Journal of Radiation Biology, 63, 543– 556. Griffin, C. S., H ill, M. A., P apworth, D. G., Townsend, K. M. S. Savage, J. R. K. and Goodhead, D. T., 1998, EVectiveness of 0. 28 keV carbon K ultrasoft X-rays at producing simple and complex chromosome exchanges in human Ž broblasts in vitro detected using FISH. International Journal of Radiation Biology, 73, 591– 598. Hofer, K. G. and Bao, S.-P., 1995, Low-LET and high-LET radiation action of 1 2 5 I decays in DNA: eVect of cysteamine on micronucleus formation and cell killing. Radiation Research, 141, 183– 192. Jeggo, P. A., 1998, IdentiŽ cation of genes involved in repair of DNA double-strand breaks in mammalian cells. Radiation Research, 150, S80–S91. Lea, D. E., 1946, Actions Of Radiations On Living Cells (Cambridge: Cambridge University Press). Ludwiko w, G., H oebe, R. A., Franken, N. A. P ., Stap, J., van O ven, C. H. and Aten, J. A. 2000, A method for the selective irradiation of part of a genome. Acta Oncologica, 39, 693–698. Ludwiko w, G., Hofer, K. G., Bao, S.-P . and Ludwiko w, F., 1996, The eVect of 1 2 5 I decay at diVerent stages of S-phase on survival, expression of micronuclei and chromosome aberrations in CHO cells. International Journal of Radiation Biology, 70, 177– 187. Manuelidis, L. 1985, Individual interphase chromosome domains revealed by in situ hybridization. Human Genetics, 71, 288–193. Martin, R. F. and Haseltine, W. A., 1981, Range of radiochemical damage to DNA with decay of iodine-125. Science, 213, 896–898. Natarajan, A. T., 1990, Trends in Chromosome Research (Berlin: Springer), pp. 119–124. P inkel, D., Straume, T . and Gray, J. W., 1986, Cytogenetic analysis using quantitative, high-sensitivity,  uorescence

Chromosome aberrations in unirradiated chromatin after partial irradiation of the nucleus hybridization. Proceedings of the National Academy of Sciences, USA, 83, 2934–2938. Resnick, M. A., 1976, The repair of double-strand breaks in DNA; a model involving recombination. Journal of Theoretical Biology, 59, 97–106. Richardson, C. and Jasin, M., 2000, Frequent chromosomal translocations induced by DNA double-strand breaks. Nature, 405, 697– 700. Richardson, C., Moynahan, M. E. and Jasin, M., 1998, Double-strand break repair by interchromosomal recombination: suppression of chromosomal translocations. Genes and Development, 12, 3831– 3842. Savage, J. R. K., P rasad, R. and P apworth, D. G., 1984, Subdivision of S-phase and its use for comparative purposes in cultured human cells. Journal of Theoretical Biology, 111, 355– 367. Sax, K., 1941, Types and frequencies of chromosomal aberrations induced by X-rays. Cold Spring Harbor Symposium on Quantitative Biology, 9, 93–103. Serebrovski , A. S., 1929, A general scheme for the origin of mutations. American Naturalist, 63, 374– 378. Thacker , J., Wilkinson, R. E. and Goodhead, D. T., 1986, The induction of chromosome exchange aberrations by

247

carbon ultrasoft X-rays in V79 hamster cells. International Journal of Radiation Biology, 49, 645–656. Vindelov, L. L., Christensen, I. J., Keiding, N., SpangThomsen, M. and Nissen, N. T., 1983, Long-term storage of samples for  ow cytometric DNA analysis. Cytometry, 3, 317– 322. Visser, A. E. and Aten, J. A., 1999, Chromosome s as well as chromosomal subdomains constitute distinct units in interphase nuclei. Journal of Cell Science, 112, 3353–3360. Wang, P ., Zhou, R.-H., Zou, Y., Jackson-Cook, C. K. and P ovirk, L. F., 1997, Highly conservative reciprocal translocations formed by apparent joining of exchanged DNA double-strand breaks ends. Proceedings of the National Academy of Sciences, USA, 94, 120018–12023. Ward, J. F., 1994, The complexity of DNA damage: relevance to biological consequences. International Journal of Radiation Biology, 66, 427– 432. Xiao, Y., Slijepcevic, P ., Arkestein, G., Darroudi, F. and Natarajan A. T., 1996, Development of DNA libraries speciŽ c for Chinese hamster chromosom e 3, 4, 9, 10, X and Y by DOP-PCR. Cytogenetics Cell Genetics, 75, 57–62. Yang, S.-J., Hahn, G. M. and Bagshaw, M. A., 1966, Chromosom e aberrations induced by thymidine. Experimental Cell Research, 42, 130– 135.