Chromosome terminal deletion formation in Chinese hamster ovary cells

P"

!

Fundamental and Molecular Mechanisms of Mutagenesis

l

ELSEVIER

Mutation Research 311 (1994) 125-131

Chromosome terminal deletion formation in Chinese hamster ovary cells J e f f r e y L. S c h w a r t z a,b,., B.A. S e d i t a a, N i c h o l a s L a f f e l y a, D a v i d J. G r d i n a a,b a Centerfor Mechanistic Biology and Biotechnology, Argonne National Laboratory, 9700 South CassAvenue, Argonne, IL 60439-4833, USA b Department of Radiation and Cellular Oncology, The University of Chicago, Chicago, IL 60637, USA Received 7 March 1994; revision received 20 May 1994; accepted 25 May 1994

Abstract To investigate the fate of unrejoined DNA double-strand breaks, the frequency of 6°Co y-ray- and restriction-enzyme-induced terminal chromosome deletions, a marker of unrejoined breaks, was determined in CHO-K1 and in xrs-5 cells. The xrs-5 cell is a DNA double-strand break repair-deficient derivative of CHO-K1. Terminal deletion frequency was small in both CHO-K1 and xrs-5 cells when cells were irradiated or treated with restriction enzyme while in the G 1 phase of the cell cycle. In contrast, previous studies have shown that treatment of cells in G 2 leads to large deletion frequencies, especially in xrs-5 cells. Cell cycle analyses show large G 2 blocks in irradiated xrs-5 cells with only partial recovery over a 24-96-h period. These results suggest that most CHO cells with unrejoined breaks are blocked in G 2 and, therefore, do not contribute to chromosome mutation frequencies. The small frequencies of terminal deletions that are found in these cells may reflect either an inefficiency in the G 2 checkpoint mechanism or, perhaps, a modification of broken ends that allows passage through G 2. Keywords: Chromosome aberration induction; DNA repair mutants; Chinese hamster ovary cells; Gamma rays; Restriction enzymes

1. Introduction

* Corresponding author. Tel. (708) 252-3919; Fax (708) 2523387. The submitted manuscript has been authored by a contractor of the U.S. Government under contract No. W-31-109-ENG38. Accordingly, the U.S. Government retains a nonexclusive, royalty-free license to publish or reproduce the published form of this contribution, or allow others to do so, for U.S. Government purposes.

The ability to rejoin DNA double-strand breaks is an essential function in mammalian cells. Most biochemical or cytogenetic assays suggest that mammalian cells can rejoin over 90% of induced breaks (Schwartz et al., 1988; Iliakis and Pantelias, 1990), and deficiencies in DNA doublestrand break rejoining are often associated with acute radiation sensitivity (Jeggo, 1990; Illiakis, 1991). Unrejoined breaks are lethal and poten-

0027-5107/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0027-5107(94)00123-M

126

J.L. Schwartz et aL /Mutation Research 311 (1994) 125-131

tially recombinogenic lesions. In this study, we focused on the question of what is the fate of these unrejoined breaks. Do they lead to lethality through terminal deletion formation and loss during the cell cycle, or are they signals for other types of cellular events as well? To try and address this question, we used chromosome terminal deletion formation as a marker of unrejoined breaks (Savage, 1975). Terminal chromosome deletions are defined as acentric fragments of a size greater than the width of a chromosome, to distinguish them from small interstitial deletions, and fragments that are not associated with any obvious asymmetric exchange process, specifically dicentrics or centric rings, whose production also results in acentric fragments. The acentric fragments produced in asymmetrical exchanges differ from true terminal deletions in that they have telomers at their ends. Terminal deletion formation following both y-ray and restriction-enzyme exposure was studied in two Chinese hamster ovary (CHO) cell lines, CHO-K1 and its radiosensitive derivative xrs-5. The xrs-5 cell line is defective in its ability to rejoin D N A double-strand breaks (Jeggo and Kemp, 1983; Kemp et al., 1984; Jaffe et al., 1990).

2. Materials and methods

Both cell lines were maintained as asynchronous exponentially growing populations in a - M E M medium containing 25 mM HEPES, 10% fetal calf serum, and 2 mM L-glutamine. Cultures were incubated at 37°C in a 5% CO 2 atmosphere. For y-ray exposures, cells were suspended in 15-ml polypropylene tubes at a final density of 1 × 10 6 cells/ml in a total volume of 5 ml. The y-ray source was a Gamma Beam 650 Irradiator (Atomic Energy of Canada). Cells were irradiated with 6°Co y-rays at a dose rate of about 0.5 G y / m i n . Following irradiation, cells were cultured for 18-24 h prior to cytogenetic analysis. To ensure that only cells in the G 1 phase of the cell cycle during irradiation were being analyzed, cells were first synchronized with mitotic shake-off (Terasima and Tolmach, 1961; Schwartz, 1986). After each shake-off, a sample of cells was fixed

in 3:1 methanol:acetic acid, and the fraction of mitotic cells determined. At least 500 cells were used for this determination. Only shake-offs showing greater than 90% mitotic cells were used for the studies. The methods used to introduce restriction enzymes into cells were modified from Winegar et al. (1989). Briefly, synchronized G~ cells were washed twice in phosphate-buffered sucrose (7 mM KH2PO4, 1 mM MgCl 2, 272 mM sucrose, pH 7.4) and then resuspended in phosphatebuffered sucrose at a concentration of 107 cells/ml. A volume of 0.8 ml of cell suspension was added to a 0.4-cm electroporation cuvette (Bio-Rad) along with 20 units of Sau3AI restriction enzyme (New England Bio Labs) or an equivalent volume of buffer, and the cuvette was inverted to mix the contents. Each cuvette was electroporated at 300 V and 125 p~F on Bio-Rad Gene Pulser electroporation apparatus. The time constants for all samples were 19.7-21.9 ms. Each cuvette was placed on ice for 5 min following electroporation, and then the cell sample was added to a flask containing 10 ml of complete warmed medium. After 1-2 h incubation at 37°C, the medium was removed, flesh medium was added back, and cells were cultured for a further 20 h. Colcemid (2 × 10 - 7 M for CHO-K1 cells and 2 × 10 -9 M for xrs-5 cells) was added for the final 2 h of culture. Cells were harvested and slides were prepared by routine methods (Shadley et al., 1991). Slides were stained with a 5% Giemsa solution and analyzed for either chromosome-type or chromatid-type aberrations (Savage, 1975). Terminal deletions were defined as described in the Introduction. Two to three independent determinations were made. Fifty cells were counted for each determination. For some experiments, cells were cultured in medium supplemented with 1 /xM bromodeoxyuridine (BrdU) to label replicating cells. To identify labeled cells, slides were stained with an antibody against BrdU-containing DNA, as described in Pinkel et al. (1985). Slides were dehydrated through a series of washes in 70%, 80%, 90%, and 100% ethanol and then air dried. Chromosomal D N A was denatured by heating 5 min

J.L. Schwartz et al. / Mutation Research 311 (1994) 125-131

2.0

I A

127

a l 2.0

1.S

1.5

>-

1 .O

1.0

E

o.s.

/

o.5

/ O o.o~

/

2

4

6

/

o

2

DOSE(Gy)

4

6

DOSE(Gy)

Fig. l. Chromosome aberration frequency in CHO-K1 (A) and xrs-5 (B) cells irradiated in O1. Interstitial deletions (e), dicentrics and centric rings (o), and terminal deletions (<>) are shown. Results are mean + SEM of two to three independent experiments.

at 80°C in phosphate-buffered saline containing 0.5% Tween-20 (Sigma). While slides were wet, monoclonal mouse anti-BrdU (Sigma) was added to the slides and they were incubated for 30 min at 37°C in a humidified atmosphere, after which they were washed three times in Tween-20. Slides were then incubated for 30 min at 37°C in a humidified atmosphere with rabbit anti-mouse FITC (Sigma). Finally, the slides were washed three times with phosphate-buffered saline supplemented with 0.5% bovine serum albumin and then counterstained with 4,6-diamidino-2-phenyl indole (DAPI, Sigma) diluted 1:4 with anti-fade (Molecular Probes) solution. Cell cycle progression after radiation exposure was determined as described previously (Grdina and Sigdestad, 1992). Briefly, cells were cultured for 18-24 h prior to irradiation with 2.5, 5, or 7.5 Gy y-rays. At regular intervals after exposure,

individual cultures were trypsinized and fixed in 70% ethanol. They were then stained in 2 / z g / m l DAPI, and the fluorescence histogram was accumulated on a Partec PAS II cytometer. The distribution of cells throughout the cell cycle was calculated using software designed by Phoenix Flow Systems.

3. Results

The 6°Co y-ray-induced aberration frequency in CHO-K1 cells in G 1 is shown in Fig. 1 and Table 1. The chromosomal sensitivity of these cells has been previously reported (Kemp and Jeggo, 1986; Darroudi and Natarajan, 1987; Bryant et al., 1987), although terminal deletion frequency was not distinguished in these reports. As expected, irradiation of cells in G 1 resulted in

Table 1 Chromosome aberration induction in CHO-K1 cells treated in G 1 Treatment

0.0 Gy 2.0 Gy 4.0 Gy 6.0 Gy Buffer Sau3AI

y-rays y-rays y-rays y-rays

Total number

Aberrant

Chromatid-type

of cells

cells (%)

CD/ISO

100 100 100 100

4 25 51 74

100 130

4 32

Chromosome-type

EXCH

ISO/CD

D+ R

ID

TD

6 4 8 3

0 0 0 0

0 0 0 0

3 21 48 102

2 23 47 180

0 4 2 14

2 11

0 0

0 0

0 8

2 102

0 6

CD/ISO, chromatid and isochromatid deletion; EXCH, inter- and intra-chromosome exchange; ISO/CD, chromatid-isochromatid exchange; D + R, dicentrics and eentric rings; ID, interstitial deletions; TD, terminal deletions.

128

J.L. Schwartz et al. /Mutation Research 311 (1994) 125-131

almost exclusively chromosome-type aberrations. Most of the aberrations were either interstitial deletions, dicentrics, or rings. Terminal deletions comprised less than 10% of the total n u m b e r of aberrations. The '/-ray-induced aberration frequency in xrs5 cells irradiated in Gt is shown in Fig. 1 and Table 2. The xrs-5 cells were approximately twice as sensitive as CHO-K1 cells to ,/-ray-induced aberrations. These results are similar to those reported by Bryant et al. (1987) but are much lower than what Darroudi and Natarajan (1987) reported. Most of the aberrations seen in xrs-5 cells were of the chromosome type, but, as others have noted (Kemp and Jeggo, 1986; Darroudi and Natarajan, 1987; Bryant et al., 1987), chromatidtype aberrations were also induced in these irradiated cells in G 1. Approximately 2 5 - 4 0 % of the aberrations were of the chromatid type. About 70% of the cells showing chromatid-type aberrations also contained chromosome-type aberrations suggesting that the appearance of chromatid-type aberrations did not necessarily reflect contamination with non-G 1 cells. As was seen with the CHO-K1 cells, most of the induced chromosome-type aberrations were interstitial deletions, dicentrics, or rings. While the terminal deletion frequency was higher in xrs-5 than C H O K1 cells, it still comprised less than 15% of the total aberration frequency. Cells were electroporated with 20 units of S a u 3 A I . S a u 3 A I is a restriction enzyme that recognizes the sequence G A T C and produces a 4base 5' overhang. Electroporation alone had little

effect on aberration frequency by increasing baseline frequency in CHO-K1 cells from 4% to 6% and in xrs-5 ceils from 13% to 18% (Tables 1 and 2). T r e a t m e n t with 20 units of S a u 3 A I resulted in increases in aberration frequency in both C H O K1 and xrs-5 cells. The xrs-5 cell line was only slightly more sensitive to aberration induction with S a u 3 A I than was CHO-K1. Most of the aberrations induced in CHO-K1 cells with S a u 3 A I were chromosome-type aberrations (Table 1). Less than 10% of the total were chromatid-type aberrations. This low chromatidtype aberration frequency may reflect either S / G 2 contamination of our cultures or some residual enzyme activity that persists into S phase. The bulk of the chromosome-type aberrations induced with S a u 3 A I in CHO-K1 cells were interstitial deletions. Few dicentrics or terminal deletions were seen. Terminal deletions made up 5.5% of the total chromosome-type aberrations induced in CHO-K1 cells. Both chromosome- and chromatid-type aberrations were induced with S a u 3 A I in xrs-5 cells (Table 2). Chromatid-type aberrations made up about 38% of the total. This is similar to what is seen in ,/-ray-irradiated samples and has also been reported by Bryant et al. (1987) and Darroudi and Natarajan (1989). As with CHO-K1 cells, the bulk of the chromosome-type aberrations induced with S a u 3 A I in xrs-5 cells were interstitial deletions. Terminal deletions made up 9.5% of the total and dicentrics and centric rings made up less than 20% of the total chromosometype aberrations induced in xrs-5 cells.

Table 2 Chromosome aberration induction in xrs-5 cells treated in G 1 Treatment Total number Aberrant Chromatid-type of cells Cells (%) CD/ISO EXCH

ISO/CD

Chromosome-type D+R ID

TD

0.0 Gy y-rays 1.0 Gy y-rays 2.0 Gy T-rays 4.0 Gy T-rays

100 250 200 100

13 24 40 66

13 46 39 93

0 6 7 28

0 8 28 40

0 61 82 68

0 60 J 19 147

0 7 26 30

Buffer

100 100

18 58

12 38

0 0

0 19

2 17

6 67

0 8

Sau3A

CD/ISO, chromatid and isochromatid deletion; EXCH, inter- and intra-chromosome exchange; ISO/CD, chromatid-isochromatid exchange; D + R, dicentrics and centric rings; ID, interstitial deletions; TD, terminal deletions.

J.L. Schwartz et aL / Mutation Research 311 (1994) 125-131

I DO

9O 80

_z

70

V-Y

60 ~

o

v" /O

50~ 40:

uJ 0

30~

uJ

20-

13-

9

•

,2" \ • k 0 " 0 -""0 -\v/'v' 0"0

0-0-0-0-0-0.0.0.0-0

100

\

I

13

~

5

I

'

I

'

;

'.

I

'

I

'

I

,

35 10 15 20 25 30 TIME (HR) AFrER GAMMA IRRADIATION

100

B

90

8o

!_,, ._A.._,t_a

z 7o tJ

#aw

40

so°°

o. o.O" i" °---°---o

4030

°/

20

~-~o~

.

.

.

@,.0..0 -0": 0

.

O----O

.-o-O~o~O~ °

I0 0

I

0

5

f

I

I

J

I

I

I0 15 20 25 30 55 TIME (HR) AFTER GAMMA IRRADIATION

40

Fig. 2. Percentage of cells in G 2 as a function of time after exposure of CHO-K1 (A) or xrs-5 (B) cells to 10 Gy ( • ) , 7.5 Gy (A), 5 Gy (e), or 2.5 Gy ( * ) 3,-rays as compared with control cells (o).

According to Weibezahn et al. (1985), xrs-5 ceils show a large radiation-induced G 2 delay. This was confirmed with an analysis of cell cycle progression after y-ray exposure (Fig. 2). Exposure of CHO-K1 cells to 5 or 10 Gy y-rays resulted in a transient G2 delay (Fig. 2A). The G 2 peak was about 50% for 5 Gy and 70% for 10 Gy exposure. The delay lasted about 5 h for 5 Gy exposure and 10 h for 10 Gy exposure. In contrast, exposure of xrs-5 ceils to 2.5, 5, or 7.5 Gy y-rays resulted in 70-80% of the population being blocked in G 2 (Fig. 2B), and the majority of these cells remain blocked in G 2 for over 24 h.

129

Similar results were seen following exposure to Sau3A (unpublished observations). The cell cycle analysis suggested that most irradiated xrs-5 cells remain blocked in G 2 for over 24 h. To determine whether these blocked cells ever get through G 2, we harvested mitotic cells 48, 72, and 96 h after a 8 Gy (CHO-K1) or 4 Gy (xrs-5) y-ray exposure. To insure that only cells that had been arrested in G 2 were analyzed, 24 h after irradiation, 1 /zM BrdU was added to cultures to label cycling cells. Mitotic cells were stained for BrdU-containing DNA and only cells containing no BrdU were analyzed. These represent cells that had been blocked in G 2 and were entering mitosis for the first time since irradiation. The frequency of these previously blocked but now cycling cells was low. The frequency of unlabeled CHO-K1 metaphases was 6.4% at 48 h and 3.3% at 72 h. For xrs-5 cells, the frequency of unlabeled metaphases was 6.8% at 48 h, 2% at 72 h, and 3.3% at 96 h. Aberration frequency in these previously blocked cells was very high with individual cells usually showing too many aberrations to score accurately. Thus most of the irradiated xrs-5 cells that block in G 2 remain blocked for more than 96 h.

4. Discussion Terminal chromosome deletions are thought to represent unrejoined chromosome breaks. Their frequency is usually very low. Presumably, this reflects the fact that most mammalian cells rejoin over 90% of induced DNA double-strand breaks~ In this study, we report that for both CHO-K1 and xrs-5, a mutant derivative of CHOK1 with defective ability to rejoin DNA doublestrand breaks, most of the y-ray-induced and Sau3AI-induced chromosome-type aberrations are interstitial deletions, dicentrics, or rings. Terminal deletions comprise between 2% and 8% of the total chromosome-type aberrations induced in CHO-K1 ceils and between 5% and 12% of the total in xrs-5 cells. While the terminal deletion frequency was slightly higher in xrs-5 than in CHO-K1 cells, one might have predicted much higher terminal deletion frequencies based on

130

J.L. Schwartz et aL /Mutation Research 311 (1994) 125-131

D N A repair studies (Kemp et al., 1984; Jaffe et al., 1990), which usually suggest that xrs-5 cells can rejoin only about 50% of induced D N A double-strand breaks. Iliakis et al. (1992) have reported that the levels of unrejoined breaks seen in xrs-5 cells are dose dependent and are much smaller at lower doses. However, even their studies would suggest larger levels of terminal deletions than what were found in our studies. One reason so few terminal deletions are seen, even in xrs-5 cells, is that these unrejoined chromosome breaks are the signal for G 2 delay, and cells with unrejoined breaks remain blocked in G 2 as long as these breaks remain open. For xrs-5 cells, with their defect in double-strand break rejoining, the G 2 block can be permanent. By blocking cells with unrejoined breaks in G 2, terminal chromosome deletion frequency and overall aberration frequency will be low because these cells with unrejoined breaks do not contribute to aberration frequency which is measured in mitotic cells. This conclusion is supported by the following observations. First, while there are only two-fold differences between CHO-K1 and xrs-5 cells in aberration sensitivity when G~-irradiated cells are compared (Fig. 1), when the comparison is made between the two cell lines exposed in G2, after the G2 checkpoint, the differences are much greater and closer to that seen for cell killing (Shadley et al., 1991). Second, cell flow cytometric studies show large blocks in G 2 for exposed xrs-5 cells (Fig. 2) that, in contrast to CHO-K1 cells, are maintained for very long periods of time. Even measurements made 96 h after irradiation fail to show any evidence for these cells breaking through the G 2 block in any great numbers. Thus most of these Gz-blocked xrs-5 cells apparently die in G 2. There is no evidence for any apoptotic process occurring. The block is D N A double-strand break induced as it is found with restriction enzymes as well as with ionizing radiation (unpublished observations). As mentioned in the Introduction, true terminal deletions differ from acentric fragments produced by recombination processes in that they lack telomeres. In yeast, the presence of unrejoined D N A double-strand breaks and lack of a telomere is a signal to block cells in G 2 (Weinert

and Hartwell, 1988; Sandell and Zakian, 1993). Mammalian cells appear to act like yeast in this regard. An unrejoined break signals for a G 2 block. In the yeast system described by Sandell and Zakian (1993), the G 2 arrest produced by the loss of a telomere was transient. Most cells recovered to continue replicating and segregating the chromosome for as many as 10 cell divisions. Even more surprising was the observation that the G 2 delay was seen only in the first cell division. The results suggest that either the broken end is modified or some other process in the cell is modified so that the broken end does not continue to signal for a G 2 delay. If a similar process exists in mammalian ceils, it is obviously defective in xrs-5 cells as most irradiated xrs-5 cells remain blocked in G 2. It has recently been reported (Getts and Stamato, 1994) that there is an absence in xrs-5 cells of a Ku-like D N A endbinding activity which renders ends resistant to exonuclease digestion. Continued exonuclease digestion of broken ends may prevent end modification as well as repair and thereby may explain the permanent G 2 block in irradiated xrs-5 cells. In conclusion, these studies on terminal chromosome deletion induction in CHO-K1 and xrs-5 cells suggest that unrejoined breaks serve as signals for G 2 block and, therefore, do not contribute to terminal deletion frequency. The small frequencies of terminal deletions that are found in these cells probably reflect a modification of broken ends to permit passage through G 2.

Acknowledgements This work was supported by a Faculty Research Award from the American Cancer Society (J.L.S.), grant CA-37435 from the NIH (D.J.G), and U.S. Department of Energy contract No. W-31-109-ENG-38 and grant No. DE-FG0288ER60661 (J.L.S.). N.L. was a participant in the Fall 1993 Science and Engineering Research Semester program. The authors would like to thank Jane Perrin for her help with flow cytometry.

J.L. Schwartz et al. / Mutation Research 311 (1994) 125-131

References Bryant, P.E., D.A. Birch and P.A. Jeggo (1987) High chromosomal sensitivity of Chinese hamster xrs-5 cells to restriction endonuclease induced DNA double-strand breaks, Int. J. Radiat. Biol., 52, 537-554. Darroudi, F. and A.T. Natarajan (1987) Cytological characterization of Chinese hamster ovary X-ray-sensitive mutant cells xrs 5 and xrs 6: I. Induction of chromosomal aberrations by X-irradiation and its modulation with 3-aminobenzamide and caffeine, Mutation Res., 177, 133-148. Darroudi, F. and A.T. Natarajan (1989) Cytological characterization of Chinese hamster ovary X-ray-sensitive mutant cells xrs 5 and xrs 6: IV, Study of chromosomal aberrations and sister-chromatid exchanges by restriction endonucleases and inhibitors of DNA topoisomerase II, Mutation Res., 212, 137-148. Getts, R.C. and T.D. Stamato (1994) Absence of a Ku-like DNA end-binding activity in the xrs double-strand DNA repair-deficient mutant, J. Biol. Chem., in press. Grdina, D.J., and C.P. Sigdestad (1992) Chemical protections and cell-cycle effects on radiation-induced mutagenesis, Cell Prolif., 25, 23-29. Iliakis, G. (1991) The role of DNA double-strand breaks in ionizing radiation-induced killing of eukaryotic cells, Bioassays, 13, 641-648. Iliakis, G. and R. Okayasu (1990) Radiosensitivity throughout the cell cycle and repair of potentially lethal damage and DNA double-strand breaks in an X-ray-sensitive CHO mutant, Int. J. Radiat. Biol., 57, 1195-1211. Iliakis, G.E. and G.E. Pantelias (1990) Production and repair of chromosome damage in an X-ray sensitive CHO mutant visualized and analyzed in interphase using the technique of premature chromosome condensation, Int. J. Radiat. Biol., 57, 1213-1223. Iliakis, G., R. Mehta and M. Jackson (1992) Level of DNA double-strand break rejoining in Chinese hamster xrs-5 cells is dose-dependent: implications for the mechanism of radiosensitivity, Int. J. Radiat. Biol., 61,315-321. Jaffe, D.R., D. Haraf, J.L. Schwartz, R.R. Weichselbaum and A.M. Diamond (1990) Radioresistant derivatives of an X-ray-sensitive CHO cell line exhibit distinct patterns of sensitivity to DNA-damaging agents, Carcinogenesis, 11, 1265-1269. Jeggo, P.A. (1990) Studies on mammalian mutants defective in rejoining double-strand breaks in DNA, Mutation Res., 239, 1-16.

131

Jeggo, P.A. and L.M. Kemp (1983) X-ray-sensitive mutants of Chinese hamster ovary cell line: isolation and cross-sensitivity to other DNA-damaging agents, Mutation Res., 112, 313-327. Kemp, L.M. and P.A. Jeggo (1986) Radiation-induced chromosome damage in x-ray-sensitive mutants (xrs) of the Chinese hamster ovary cell line, Mutation Res., 166, 255263. Kemp, L.M., S.G. Sedgwick and P.A. Jeggo (1984) X-ray sensitive mutants of Chinese hamster ovary cells defective in DNA double strand break rejoining, Mutation Res., 132, 189-196. Pinkel, D., L.H. Thompson, J.W. Gray and M. Vanderlaan (1985) Measurement of sister chromatid exchanges at very low bromodeoxyuridine substitution levels using monoclonal antibody in Chinese hamster ovary cells, Cancer Res., 45, 5795-5798. Sandell, L.L. and V.A. Zakian (1993) Loss of a yeast telomere: arrest, recovery and chromosome loss, Cell, 19, 729739. Savage, J.R.K. (1975) Classification and relationships of induced chromosomal structural changes, J. Med. Genet., 12, 103-122. Schwartz, J.L. (1986) Analysis of bromodeoxyuridine-induced single and twin sister chromatid exchanges in tetraploid Chinese hamster ovary cells, Chromosoma, 93, 409-412. Schwartz, J.L., J. Rotmensch, S.M. Giovanazzi, M.B. Cohen and R.R. Weichselbaum (1988) Faster repair of DNA double-strand breaks in radioresistant human tumor cells, Int. J. Radiat. Oncol. Biol. Phys., 15, 907-912. Shadley, J.D., J.L. Whitlock, J. Rotmensch, R.W. Atcher, J. Tang and J.L. Schwartz (1991) The effects of radon daughter alpha-particle irradiation in K1 and xrs-5 CHO cell lines, Mutation Res., 248, 73-83. Terasima, T. and L.J. Tolmach (1961) Changes in X-ray sensitivity of HeLa cells during the division cycle, Nature, 190, 1210-1211. Weibezahn, K.F., H. Lohrer and P. Herrlich (1985) Doublestrand break repair and G 2 block in Chinese hamster ovary cells and their radiosensitive mutants, Mutation Res., 145, 177-183. Weinert, T. and L.H. Hartwell (1988) The rad9 gene controls the cell cycle response to DNA damage in Saccharomyces cerevisiae, Science, 241, 317-322. Winegar, R.A., J.W. Phillips, J.H. Youngblom and W.F. Morgan (1989) Cell electroporation is a highly efficient method for introducing restriction endonucleases into cells, Mutation Res., 225, 49-53.