- Email: [email protected]
Transmission of radiation-induced rearrangements through cell divisions
Mutation Research,
198 (1988) 191-198
191
Elsevier MTR 04505
Transmission of radiation-induced rearrangements through cell divisions Walid A1-Achkar
1,3, L a u r e
Sabatier 1 a n d B e r n a r d D u t r i l l a u x 1,2
I Commissariat ~ l'EnergieAtomique, IPSN, Ddpartement de Protection Sanitaire, D.P.S./S.P.E./L.G.E., P.O. Box 6, 92265 Fontenay-aux-Roses (France), 2 lnstitut Curie, Section de Biologie, UA 620 du C.N.R.S., 26, rue d'Ulm, 75231 Paris Cedex 05 (France), and 3 Commissariat ~ l'Energie Atomique, P.O. Box 6091, Damas (Syria)
(Received7 April 1987) (Revisionreceived17 July 1987) (Accepted 23 July 1987)
Keywords:
Radiation-induced rearrangements; Transmission; Cell divisions; BrdU incorporation technique; Heat denaturation; Acridine orange staining.
Summary A BrdU incorporation technique associated with heat denaturation, acridine orange staining and UV irradiation, was applied to G0-irradiated lymphocyte cultures. This made it possible both to obtain an R-banding, and to estimate the number of divisions undergone by each cell in mitosis since irradiation. Cell survival and slowing down of the cell cycle could be distinguished. The frequency of various types of rearrangements, and their association was studied at each cell division. It is shown that the loss of cells carrying chromosomal" rearrangements is determined by several parameters such as the presence of dicentric or multicentric chromosomes and above all the association of several rearrangements in the same cell.
The fact that the rate of radiation-induced chromosomal aberrations depends on the received dose and on the time between irradiation and observation is well known. By the use of BrdU incorporation techniques, it can be shown that irradiation slows down the cell cycle besides decreasing cell survival. This last effect seems to be partly related to the formation of chromosomal rearrangements, mainly dicentric and multi-break anomalies. It has been proposed that the frequency
Correspondence: Dr. Walid A1-Achkar, Commissariat ~ FEnergie Atomique, P.O. Box 6091, Damas (Syria).
of dicentrics is divided by two at each cell generation (Carrano, 1973; Scott and Lyons, 1979; Van Buul and Natarajan, 1980; Dutrillaux et al., 1985) but this estimate was not based on the result of direct experiments and may result from various parameters (Bender and Brewen, 1969; Steffen and Michalonski, 1973). This is why, in order to study cell survival, a technique has been developed in our laboratory to analyse, for each mitosis, both its R-banding and the number of cell cycles that have occurred since irradiation. The method used consists of a BrdU incorporation followed by heat denaturation, acridine orange staining and UV irradiation (Dutrillaux et al., 1974). The results obtained clearly show that the decrease of so-called non-transmissible rearrangements cannot be attributed to cell death only.
0027-5107/88/$03.50 © 1988 ElsevierScience Publishers B.V. (BiomedicalDivision)
192
Material and methods
Whole-blood samples from 3 healthy donors were irradiated in the G O phase by 6°Co ~,-rays, at a dose rate of 0.5 G y / m i n . 4 series of experiments were performed with a dose of 2 Gy. Culture medium TC 199 was used supplemented with 25% human serum, phytohaemagglutinin and BrdU (final concentration 10 #g/ml). Mitoses were harvested following a 2-h colchicine treatment, 72 or 96 h after initiating the cultures. Slides prepared according to our usual method (Dutrillaux
and Couturier, 1981) were treated with Earle's BSS at pH 5.3 during 40 min, and then stained with acridine orange. As previously described (Dutrillaux et al., 1974), the chromosome staining progressively changes upon UV irradiation during microscopic observation. For the first minute there is a typical R-banding of both chromatids and this time is sufficient to take a first photograph. Afterwards, the R-banding progressively disappears, each chromatid becomes homogeneous, and the staining reveals the degree of substitution with BrdU: green for none or 1-strand substitution,
Fig. 1. Mitoses from h u m a n lymphocyte exhibiting a typical R-banding (a and c) after BrdU incorporation, heat treatment and acridine orange staining. The same mitoses (b and d) exhibit after about 1 min UV irradiation, a differential staining of their chromatids, indicating the number of divisions having occurred " i n vitro": 2 for a and 3 for d.
193
and orange for 2-strand substitution. A second photograph is taken, and it is easy to determine the number of cell cycles which occurred in vitro after y-irradiation (Fig. 1). Results and discussion
more than 3 divisions after irradiation. For instance, mitoses with chromosomal rearrangements are significantly less frequent in a sample of cells which have undergone more than 3 divisions than in a sample of cells which have undergone 1-3 divisions (X2= 6.53, 8 = 1, data from Table 1).
(1) Effect of radiation and of the induction of chromosomal rearrangements on the rapidity of the cell cycle In 72-h cultures, we compared the frequencies of 1st-, 2nd- and 3rd-generation mitoses in irradiated and control samples (Table 1). For this estimate, a single donor was used, in order to avoid interindividual variation (Crossen and Morgan, 1977). The frequency of 2nd-generation mitoses is quite similar in both samples, whereas the frequency of lst-generation mitoses is higher in the irradiated cultures and that of 3rd-generation mitoses is higher in control cultures. This fits with the fact that irradiation slows down the cell cycle (Lloyd et al., 1977; Vulpis et al., 1978; Purrott et al., 1980; Dutrillaux et al., 1985). Among irradiated mitoses, we differentiated between those with normal and those with abnormal karyotypes. The latter tend to be more frequent (37 versus 26) among lst-division mitoses, and less frequent (30 versus 39) among 3rd-generation mitoses, but the differences are not significant (X2= 3.42, ~ = 1). In 96-h cultures (Table 1), a similar effect is observed in the irradiated sample which is due to the low frequencies of cells having undergone
60
ca) 0
40
20.
I
I
I
1
2
3
1
2
3
60
40
TABLE 1 N U M B E R O F 1st-, 2nd-, 3rd- A N D > 3rd-DIVISION MITOSES IN 72- A N D 96-h C U L T U R E S A F T E R 2 Gy I R R A D I A T I O N A N D IN N O N - I R R A D I A T E D CELLS Data obtained from 1 donor. The sample of irradiated cells was subdivided in relation to the presence or absence of structural rearrangements. Culture time (h):
72
Division:
1st 2nd
3rd
1st 2nd
Control
19
47
76
12
22
80 86
Irradiated cultures Normal mitoses Abnormal mitoses
62 25 37
69 32 37
69 39 30
20 11 9
29 12 17
109 42 51 29 58 13
20.
96 3rd
>3rd
>3
Fig. 2. Percentages of lst-, 2nd-, 3rd- and > 3rd-division mitoses in control and irradiated 72-h (a) and 96-h (b) cultures. White circles, controls; black circles, irradiated cultures; white triangles, normal mitoses in irradiated cultures; black triangles, abnormal mitoses in irradiated cultures.
194 The data o b t a i n e d for the 3 donors from 72-h cultures are shown in Fig. 2a. This enlarged sample confirms the results described above, with a decrease of 3rd-division mitoses after irradiation. This is also due to the decrease of mitoses carrying rearranged chromosomes. F o r instance, the mitoses with rearranged chromosomes are very significantly less frequent in the sample of cells having u n d e r g o n e 3 divisions t h a n in that having u n d e r gone 1 division (X 2 = 12.9, 8 = 1, data from T a b l e 2). The data o b t a i n e d from 96-h cultures of 3 d o n o r s are given in Fig. 2b. The lack of mitoses with c h r o m o s o m e r e a r r a n g e m e n t s after more t h a n 3 divisions b y c o m p a r i s o n with n o r m a l cells is also very significant (X 2 = 12.5, 8 = 1, calculated from data shown in Fig. 2b).
The frequencies of metaphases with reciprocal translocations, deletions a n d of r e a r r a n g e m e n t s between more t h a n 2 breaks are similar i n lst-, 2nd- a n d 3rd-division mitoses. I n v e r s i o n s a n d rings tend to decrease, b u t the samples are too small to be informative. The frequencies of dicentrics progressively decrease from 0.34 in l s t - to 0.24 in 2nd- a n d 0.08 in 3rd-division mitoses. This is c o m p a t i b l e with a 50% r e d u c t i o n per generation. I n the d i s t r i b u t i o n of r e a r r a n g e m e n t s observed in 96-h cultures (Table 2), the complex rearrangem e n t s are most frequent in l s t - g e n e r a t i o n metaphases. The frequency of dicentrics also decreases i n relation to cell divisions, b u t this decrease is lower t h a n 50% per generation. The frequencies of other r e a r r a n g e m e n t s vary w i t h o u t obvious relationship to the n u m b e r of cell divisions.
(2) Distribution of the types of chromosomal rearrangements in relation to the number of cell divisions and to the culture time
(3) Distribution of the numbers of rearrangements, and of presumed corresponding breaks in relation to the number of cell divisions
The d i s t r i b u t i o n of the various types of chrom o s o m a l r e a r r a n g e m e n t s in 72-h cultures is given in T a b l e 2.
The results presented in Fig. 3a a n d b, show a progressive decrease of the rate of r e a r r a n g e m e n t s a n d of breaks in relation to the n u m b e r of cell
TABLE 2 DISTRIBUTION OF THE VARIOUS TYPES OF REARRANGEMENTS PER MITOSIS IN RELATION TO THE CELL CYCLE IN 72-h AND 96-h CULTURES n, observed numbers and f, frequency per mitosis; trcp, reciprocal translocations; inv, inversions; del, deletions; crea, complex rearrangements; dic, dicentrics; r, rings; cte, chromatid exchanges; Nrea, total number of rearrangements; Nb, numbers of detected breakpoints; Mrea, mean number of rearrangements/cell; Mb, mean number of breakpoints/cell. Culture time (h):
72
Cell generation:
1st
96
N analysed mitoses: 128 N normal mitoses: 45
2nd
3nd
Total
1st
2nd
3rd
> 3rd
Total
172 77
100 59
400 181
37 14
54 17
160 69
49 33
300 133
Rearrangements
n
f
n
trcp inv del crea dic r cte
33 9 27 14 44 6 1
0.26 33 0.07 2 0.21 38 0.10 22 0.34 42 0.04 5 0.007 2
Nrea Nb
134 263
144 285
Mrea Mb
1.04 2.05
0.83 1.65
f
n
0.19 19 0.01 2 0.22 18 0.13 11 0.24 8 0.02 1 0.01 0
f
nt
f
0.19 0.02 0.18 0.11 0.08 0.01 0
85 13 83 47 94 12 3
0.21 5 0.03 1 0.21 10 0.12 11 0.24 9 0.03 1 0.01 0
0.14 7 0.02 6 0.27 19 0.29 10 0.24 11 0.02 1 0 0
337 668
37 82
54 95
128 233
21 34
240 444
0.84 1.67
1 2.21
1 1.75
0.8 1.45
0.42 0.69
0.8 1.48
59 120 0.59 1.2
n
f
n
f
n
f
n
0.13 0.11 0.35 0.18 0.20 0.01 0
33 6 48 16 22 3 0
0.21 4 0.03 1 0.30 10 0.10 3 0.13 3 0.01 0 0 0
f
n t
f
0.08 0.02 0.20 0.06 0.06 0 0
49 14 87 40 45 5 0
0.16 0.04 0.29 0.13 0.15 0.01 0
195
(a)
.5
I
I
I
I
1
2
3
>3
cells carrying rearranged chromosomes. The comparison of the curves also indicates that the complexity of rearrangements progressively decreases with cell divisions. For instance, in 96-h cultures, the ratio of the number of breaks per cell to the number of rearrangements per cell decreases from 2.21 in 1st division to 1.54 after more than 3 divisions. This effect is even more pronounced in 72-h cultures.
(4) Analysis of cells with dicentric chromosomes
(b)
2
1.5
.5
I
i
I
I
1
2
3
>3
Fig. 3. Average number of rearrangements (a) and of breakpoints (b) per mitosis in relation to the number of the divisions (1 to > 3). Black squares, 72-h cultures; black triangles, 96-h cultures.
divisions after irradiation. This decrease is more regular in 72-h than in 96-h cultures. There are more rearrangements and breaks in 96- than in 72-h cultures, for equivalent numbers of cell cycles, except for lst-division cells. This contrasts with the fact that there are on the whole fewer anomalies in 96- than in 72-h cultures, and it shows again the tendency for a slowing down of
In 72-h cultures, the frequency of dicentrics decreases in relation to the number of cell divisions, and this decrease is more marked between the 2nd and the 3rd than between the 1st and the 2nd cycles (Fig. 4a). The frequency of cells carrying only 1 dicentric remains stable from 1st to 2nd but there is a 50% decrease from the 2nd to the 3rd cell generation. The situation is different for cells carrying other rearrangement(s) in addition to the dicentric, whose frequency largely decreases in 2nd- and becomes very low in 3rd-generation cells. In 96-h cultures, the frequency of dicentrics regularly decreases from 1st to more than 3 generations, from 0.24 to 0.06. However, the frequency of cells carrying a dicentric only remains quite stable, whereas that of cells carrying additional rearrangement(s) progressively decreases (Fig. 4b). This indicates that the loss of dicentrics is not only related to their poor transmissibility but also to the eventual presence of other chromosomal rearrangements. Thus it is for a large part the multiplicity of lesions which determines the loss of dicentric carrier cells, since the data obtained result both from cell death and the slowing down of the cell cycle. However, comparing 2nd-generation mitoses from 72-h cultures to 3rd-generation mitoses from 96-h cultures, the frequency of dicentrics decreases from 0.24 to 0.13, which is compatible with a 50% reduction in one cell generation. This loss may result either from the instability of the dicentrics at anaphase, or from the imbalance caused by the frequent loss or duplication of the acentric fragments. Thus we have calculated the frequency of acentrics in cells carrying a dicentric but no other rearrangements. In lst-generation mitoses, all possess either 1 acentric fragment resulting from the fusion of the 2 deleted
196
(5) Analysis of reciprocal translocations (a)
A n=94
40.
34 30.
B
C
20. 10.
2
3
1
2
3
1
2
A n=4s
30
3
In 72-h cultures, the frequency of reciprocal translocations decreases from 1st to 2nd, and remains stable from 2nd to 3rd generations (Fig. 5a). However, the frequency of cells carrying a translocation only slightly increases from 1st to 3rd generation, whereas that of cells carrying additional rearrangement(s) decreases. In 96-h cultures (Fig. 5b), the variations of the frequency of translocations in relation to the cell cycle may be due to the small size of the samples. Nevertheless, there is a good conservation of translocations, as shown by the increased frequency in 3rd-division mitoses. This increase may be partly due to the loss of cells with dicentrics, which increases the relative frequency of cells with other rearrangement(s).
[b) (a) 8
C
20
A n= a s
30
13
13,,5 ~
10
1
2
3 >3
2_.~_e
9.2
1
2
3 >3
1 2
6
e
20. 3 >3
10.
Fig. 4. Distribution of dicentrics in 72-h (a) and 96-h (b) cultures per 100 mitoses, in relation to the number of cell divisions (1 to > 3). (A) In the whole sample, (B) in mitoses carrying 1 dicentric only, (C) in mitoses carrying 1 dicentric plus another rearrangement at least.
chromosomes (13/16) or the 2 non-reassociated acentrics (3/16). In 2nd generation 50% of the mitoses (24) possess either 1 or 2 acentrics, and in 3rd generation (20), this percentage is 35. These results are compatible with a random segregation of acentrics, half of them being eliminated at each generation. As can be expected, acentrics are not duplicated in 1st-generation ceils. They are found duplicated in about 15% of both 2nd- and 3rd-generation cells. This rate of duplication is much lower than the rate of loss. It is also noteworthy that the loss of dicentrics and that of acentrics do not seem to be correlated. This indicates that the deletions due to the loss of acentrics are not the major cause of elimination of cells carrying dicentrics.
1
23
1
23
1
23
(b)
A n=41) 20
10
B
C
n=28
n=18
2o
13.S12,9 ~
1
2
lo.8
3>3
1
11.~ • 2
3 >3
7.3 ?.4 1
2
3 >3
Fig. 5. Distribution of reciprocal translocations in 72-h (a) and 96-h (b) cultures per 100 mitoses, in relation to the number of cell divisions (1 to > 3). (A) In the whole sample, (B) in mitoses carrying 1 reciprocal translocation only, (C) in mitoses carrying 1 reciprocal translocation plus another rearrangement at least.
197
The analysis of deletions is not easy, because they are heterogeneous. On the one hand, it is not always possible to differentiate the 1-break terminal from intercalary deletions. On the other hand, in 2nd-and-more-generation mitoses, it is likely that some of the rearrangements interpreted as deletions result in fact from the breakage of unstable rearrangements like dicentrics. On the whole, the frequency of deletions remains fairly stable (Table 2).
50% at each generation. Thus, the loss of cells with dicentric chromosomes seems to be mainly due to mechanical reasons at anaphase but not to the monosomies. As can be expected, translocations, and probably also inversions, are fairly well maintained, especially when they are not associated with other anomalies, which lets us suppose that translocations induced by low doses of irradiation are better transmitted than those induced by higher doses. This should be taken into consideration for the calculation of genetic risk.
(7) Complex rearrangements
Acknowledgements
This category of rearrangements resulting from more than 2 breaks is still more heterogeneous than the deletions. Their frequency is fairly stable in 72-h cultures, but progressively decreases with the cell generations in 96-h cultures. This decrease is about one third per generation (Table 2).
We wish to thank Catherine Luccioni for the revision of the manuscript. This research was supported by Euratom (contract No. BI 6 149 F).
(6) Analysis of deletions
(8) Other rearrangements The frequency of inversions, rings and chromatid exchanges is too small to allow a valuable analysis. The results obtained are compatible with a progressive decrease of the frequencies of rings, possibly by a reduction of 50% per generation. Inversions are more conserved. Chromatid exchanges are rare, and may be poorly or not induced by radiations at G O phase. In conclusion, our results confirm the notion that radiation-induced rearrangements influence cell survival, and possibly also the rapidity of the cell cycle. These two parameters which depend on dose are closely involved making it impossible to separate them without the use of a BrdU incorporation technique. As was shown by previous reports (Carrano, 1973; Scott and Lyons, 1979; Van Buul and Natarajan, 1980; Beek, 1981; Dutrillaux et al., 1985), there is a decrease of the frequency of dicentrics compatible with a reduction of 50% at each cell division. However, we could show that this decrease is largely due to the loss of cells carrying several rearrangements, including one or more dicentrics. This shows that cell death is due to several factors, non-transmissible rearrangements being one of them. The loss of dicentric carrier cells is poorly related to the monosomy resulting from the loss of acentric fragments, although they are also lost at a rate compatible with
References Beek, B. (1981) Cell proliferation and chromosomal damage in human leukocytes; dicentrics and premature chromosome condensations in first, second and third mitoses after Xirradiation, Hum. Grnrt., 57, 75-77. Bender, M.A, and J.G. Brewen (1969) Factors influencing chromosome aberration yields in the human peripheral leukocyte system, Mutation Res., 8, 383-399. Carrano, A.V. (1973) Chromosome aberrations and radiationinduced cell death, I. Transmission and survival parameters of aberrations, Mutation Res., 17, 341-353. Crossen, P.E., and W.F. Morgan (1977) Analysis of human lymphocyte cell cycle time in culture measured by sister chromatid differential staining, Exp. Cell. Res., 104, 453-457. Dutrillaux, B., and J. Couturier (1981) La Pratique de l'Analyse Chromosomique, Masson, Paris. Dutrillaux, B., A.M. Fosse, M. Prieur and J. Lejeune (1974) Analyse des 6changes dans les cellules somatiques humaines, Chromosoma, 48, 327-340. Dutrillaux, B., E. Viegas-P&tuignot, M. Prod'homme and M. Sportes (1985) Distribution of various radiation-induced chromosomal rearrangements in relation to the dose and sampling time, Mutation Res., 152, 197-203. Lloyd, D.C., G.W. Dolphin, R.J. Purrott and P.A. Tipper (1977) The effect of X-ray induced mitotic delay on chromosome aberration yields in human lymphocytes, Mutation Res., 42, 401-412. Purrott, R.J., N. Vulpis and D.C. Lloyd (1980) The use of harlequin staining to measure delay in the human lymphocytes cell cycle induced by in vitro X-irradiation, Mutation Res., 69, 275-282. Scott, D., and C.Y. Lyons (1979) Homogeneous sensitivity of human peripheral blood lymphocytes to radiation-induced chromosome damage, Nature (London), 278, 756-758.
198 Steffen, J., and A. Michalowski (1973) Heterogeneous chromosomal radiosensitivity of phytohaemagglutinin-stimulated human blood lymphocytes in culture, Mutation Res., 17, 367-376. Van Buul, P.P.W., and A.T. Natarajan (1980) Chromosomal radiosensitivity of human leukocytes in relation to sampling time, Mutation Res., 70, 61-69.
Vulpis, N., R.J. Purrott and D.C. Lloyd (1978) The use of harlequin staining to assess radiation effects on the in vitro cell cycle kinetics of human lymphocytes, Mutation Res., 51, 145-148.
198 (1988) 191-198
191
Elsevier MTR 04505
Transmission of radiation-induced rearrangements through cell divisions Walid A1-Achkar
1,3, L a u r e
Sabatier 1 a n d B e r n a r d D u t r i l l a u x 1,2
I Commissariat ~ l'EnergieAtomique, IPSN, Ddpartement de Protection Sanitaire, D.P.S./S.P.E./L.G.E., P.O. Box 6, 92265 Fontenay-aux-Roses (France), 2 lnstitut Curie, Section de Biologie, UA 620 du C.N.R.S., 26, rue d'Ulm, 75231 Paris Cedex 05 (France), and 3 Commissariat ~ l'Energie Atomique, P.O. Box 6091, Damas (Syria)
(Received7 April 1987) (Revisionreceived17 July 1987) (Accepted 23 July 1987)
Keywords:
Radiation-induced rearrangements; Transmission; Cell divisions; BrdU incorporation technique; Heat denaturation; Acridine orange staining.
Summary A BrdU incorporation technique associated with heat denaturation, acridine orange staining and UV irradiation, was applied to G0-irradiated lymphocyte cultures. This made it possible both to obtain an R-banding, and to estimate the number of divisions undergone by each cell in mitosis since irradiation. Cell survival and slowing down of the cell cycle could be distinguished. The frequency of various types of rearrangements, and their association was studied at each cell division. It is shown that the loss of cells carrying chromosomal" rearrangements is determined by several parameters such as the presence of dicentric or multicentric chromosomes and above all the association of several rearrangements in the same cell.
The fact that the rate of radiation-induced chromosomal aberrations depends on the received dose and on the time between irradiation and observation is well known. By the use of BrdU incorporation techniques, it can be shown that irradiation slows down the cell cycle besides decreasing cell survival. This last effect seems to be partly related to the formation of chromosomal rearrangements, mainly dicentric and multi-break anomalies. It has been proposed that the frequency
Correspondence: Dr. Walid A1-Achkar, Commissariat ~ FEnergie Atomique, P.O. Box 6091, Damas (Syria).
of dicentrics is divided by two at each cell generation (Carrano, 1973; Scott and Lyons, 1979; Van Buul and Natarajan, 1980; Dutrillaux et al., 1985) but this estimate was not based on the result of direct experiments and may result from various parameters (Bender and Brewen, 1969; Steffen and Michalonski, 1973). This is why, in order to study cell survival, a technique has been developed in our laboratory to analyse, for each mitosis, both its R-banding and the number of cell cycles that have occurred since irradiation. The method used consists of a BrdU incorporation followed by heat denaturation, acridine orange staining and UV irradiation (Dutrillaux et al., 1974). The results obtained clearly show that the decrease of so-called non-transmissible rearrangements cannot be attributed to cell death only.
0027-5107/88/$03.50 © 1988 ElsevierScience Publishers B.V. (BiomedicalDivision)
192
Material and methods
Whole-blood samples from 3 healthy donors were irradiated in the G O phase by 6°Co ~,-rays, at a dose rate of 0.5 G y / m i n . 4 series of experiments were performed with a dose of 2 Gy. Culture medium TC 199 was used supplemented with 25% human serum, phytohaemagglutinin and BrdU (final concentration 10 #g/ml). Mitoses were harvested following a 2-h colchicine treatment, 72 or 96 h after initiating the cultures. Slides prepared according to our usual method (Dutrillaux
and Couturier, 1981) were treated with Earle's BSS at pH 5.3 during 40 min, and then stained with acridine orange. As previously described (Dutrillaux et al., 1974), the chromosome staining progressively changes upon UV irradiation during microscopic observation. For the first minute there is a typical R-banding of both chromatids and this time is sufficient to take a first photograph. Afterwards, the R-banding progressively disappears, each chromatid becomes homogeneous, and the staining reveals the degree of substitution with BrdU: green for none or 1-strand substitution,
Fig. 1. Mitoses from h u m a n lymphocyte exhibiting a typical R-banding (a and c) after BrdU incorporation, heat treatment and acridine orange staining. The same mitoses (b and d) exhibit after about 1 min UV irradiation, a differential staining of their chromatids, indicating the number of divisions having occurred " i n vitro": 2 for a and 3 for d.
193
and orange for 2-strand substitution. A second photograph is taken, and it is easy to determine the number of cell cycles which occurred in vitro after y-irradiation (Fig. 1). Results and discussion
more than 3 divisions after irradiation. For instance, mitoses with chromosomal rearrangements are significantly less frequent in a sample of cells which have undergone more than 3 divisions than in a sample of cells which have undergone 1-3 divisions (X2= 6.53, 8 = 1, data from Table 1).
(1) Effect of radiation and of the induction of chromosomal rearrangements on the rapidity of the cell cycle In 72-h cultures, we compared the frequencies of 1st-, 2nd- and 3rd-generation mitoses in irradiated and control samples (Table 1). For this estimate, a single donor was used, in order to avoid interindividual variation (Crossen and Morgan, 1977). The frequency of 2nd-generation mitoses is quite similar in both samples, whereas the frequency of lst-generation mitoses is higher in the irradiated cultures and that of 3rd-generation mitoses is higher in control cultures. This fits with the fact that irradiation slows down the cell cycle (Lloyd et al., 1977; Vulpis et al., 1978; Purrott et al., 1980; Dutrillaux et al., 1985). Among irradiated mitoses, we differentiated between those with normal and those with abnormal karyotypes. The latter tend to be more frequent (37 versus 26) among lst-division mitoses, and less frequent (30 versus 39) among 3rd-generation mitoses, but the differences are not significant (X2= 3.42, ~ = 1). In 96-h cultures (Table 1), a similar effect is observed in the irradiated sample which is due to the low frequencies of cells having undergone
60
ca) 0
40
20.
I
I
I
1
2
3
1
2
3
60
40
TABLE 1 N U M B E R O F 1st-, 2nd-, 3rd- A N D > 3rd-DIVISION MITOSES IN 72- A N D 96-h C U L T U R E S A F T E R 2 Gy I R R A D I A T I O N A N D IN N O N - I R R A D I A T E D CELLS Data obtained from 1 donor. The sample of irradiated cells was subdivided in relation to the presence or absence of structural rearrangements. Culture time (h):
72
Division:
1st 2nd
3rd
1st 2nd
Control
19
47
76
12
22
80 86
Irradiated cultures Normal mitoses Abnormal mitoses
62 25 37
69 32 37
69 39 30
20 11 9
29 12 17
109 42 51 29 58 13
20.
96 3rd
>3rd
>3
Fig. 2. Percentages of lst-, 2nd-, 3rd- and > 3rd-division mitoses in control and irradiated 72-h (a) and 96-h (b) cultures. White circles, controls; black circles, irradiated cultures; white triangles, normal mitoses in irradiated cultures; black triangles, abnormal mitoses in irradiated cultures.
194 The data o b t a i n e d for the 3 donors from 72-h cultures are shown in Fig. 2a. This enlarged sample confirms the results described above, with a decrease of 3rd-division mitoses after irradiation. This is also due to the decrease of mitoses carrying rearranged chromosomes. F o r instance, the mitoses with rearranged chromosomes are very significantly less frequent in the sample of cells having u n d e r g o n e 3 divisions t h a n in that having u n d e r gone 1 division (X 2 = 12.9, 8 = 1, data from T a b l e 2). The data o b t a i n e d from 96-h cultures of 3 d o n o r s are given in Fig. 2b. The lack of mitoses with c h r o m o s o m e r e a r r a n g e m e n t s after more t h a n 3 divisions b y c o m p a r i s o n with n o r m a l cells is also very significant (X 2 = 12.5, 8 = 1, calculated from data shown in Fig. 2b).
The frequencies of metaphases with reciprocal translocations, deletions a n d of r e a r r a n g e m e n t s between more t h a n 2 breaks are similar i n lst-, 2nd- a n d 3rd-division mitoses. I n v e r s i o n s a n d rings tend to decrease, b u t the samples are too small to be informative. The frequencies of dicentrics progressively decrease from 0.34 in l s t - to 0.24 in 2nd- a n d 0.08 in 3rd-division mitoses. This is c o m p a t i b l e with a 50% r e d u c t i o n per generation. I n the d i s t r i b u t i o n of r e a r r a n g e m e n t s observed in 96-h cultures (Table 2), the complex rearrangem e n t s are most frequent in l s t - g e n e r a t i o n metaphases. The frequency of dicentrics also decreases i n relation to cell divisions, b u t this decrease is lower t h a n 50% per generation. The frequencies of other r e a r r a n g e m e n t s vary w i t h o u t obvious relationship to the n u m b e r of cell divisions.
(2) Distribution of the types of chromosomal rearrangements in relation to the number of cell divisions and to the culture time
(3) Distribution of the numbers of rearrangements, and of presumed corresponding breaks in relation to the number of cell divisions
The d i s t r i b u t i o n of the various types of chrom o s o m a l r e a r r a n g e m e n t s in 72-h cultures is given in T a b l e 2.
The results presented in Fig. 3a a n d b, show a progressive decrease of the rate of r e a r r a n g e m e n t s a n d of breaks in relation to the n u m b e r of cell
TABLE 2 DISTRIBUTION OF THE VARIOUS TYPES OF REARRANGEMENTS PER MITOSIS IN RELATION TO THE CELL CYCLE IN 72-h AND 96-h CULTURES n, observed numbers and f, frequency per mitosis; trcp, reciprocal translocations; inv, inversions; del, deletions; crea, complex rearrangements; dic, dicentrics; r, rings; cte, chromatid exchanges; Nrea, total number of rearrangements; Nb, numbers of detected breakpoints; Mrea, mean number of rearrangements/cell; Mb, mean number of breakpoints/cell. Culture time (h):
72
Cell generation:
1st
96
N analysed mitoses: 128 N normal mitoses: 45
2nd
3nd
Total
1st
2nd
3rd
> 3rd
Total
172 77
100 59
400 181
37 14
54 17
160 69
49 33
300 133
Rearrangements
n
f
n
trcp inv del crea dic r cte
33 9 27 14 44 6 1
0.26 33 0.07 2 0.21 38 0.10 22 0.34 42 0.04 5 0.007 2
Nrea Nb
134 263
144 285
Mrea Mb
1.04 2.05
0.83 1.65
f
n
0.19 19 0.01 2 0.22 18 0.13 11 0.24 8 0.02 1 0.01 0
f
nt
f
0.19 0.02 0.18 0.11 0.08 0.01 0
85 13 83 47 94 12 3
0.21 5 0.03 1 0.21 10 0.12 11 0.24 9 0.03 1 0.01 0
0.14 7 0.02 6 0.27 19 0.29 10 0.24 11 0.02 1 0 0
337 668
37 82
54 95
128 233
21 34
240 444
0.84 1.67
1 2.21
1 1.75
0.8 1.45
0.42 0.69
0.8 1.48
59 120 0.59 1.2
n
f
n
f
n
f
n
0.13 0.11 0.35 0.18 0.20 0.01 0
33 6 48 16 22 3 0
0.21 4 0.03 1 0.30 10 0.10 3 0.13 3 0.01 0 0 0
f
n t
f
0.08 0.02 0.20 0.06 0.06 0 0
49 14 87 40 45 5 0
0.16 0.04 0.29 0.13 0.15 0.01 0
195
(a)
.5
I
I
I
I
1
2
3
>3
cells carrying rearranged chromosomes. The comparison of the curves also indicates that the complexity of rearrangements progressively decreases with cell divisions. For instance, in 96-h cultures, the ratio of the number of breaks per cell to the number of rearrangements per cell decreases from 2.21 in 1st division to 1.54 after more than 3 divisions. This effect is even more pronounced in 72-h cultures.
(4) Analysis of cells with dicentric chromosomes
(b)
2
1.5
.5
I
i
I
I
1
2
3
>3
Fig. 3. Average number of rearrangements (a) and of breakpoints (b) per mitosis in relation to the number of the divisions (1 to > 3). Black squares, 72-h cultures; black triangles, 96-h cultures.
divisions after irradiation. This decrease is more regular in 72-h than in 96-h cultures. There are more rearrangements and breaks in 96- than in 72-h cultures, for equivalent numbers of cell cycles, except for lst-division cells. This contrasts with the fact that there are on the whole fewer anomalies in 96- than in 72-h cultures, and it shows again the tendency for a slowing down of
In 72-h cultures, the frequency of dicentrics decreases in relation to the number of cell divisions, and this decrease is more marked between the 2nd and the 3rd than between the 1st and the 2nd cycles (Fig. 4a). The frequency of cells carrying only 1 dicentric remains stable from 1st to 2nd but there is a 50% decrease from the 2nd to the 3rd cell generation. The situation is different for cells carrying other rearrangement(s) in addition to the dicentric, whose frequency largely decreases in 2nd- and becomes very low in 3rd-generation cells. In 96-h cultures, the frequency of dicentrics regularly decreases from 1st to more than 3 generations, from 0.24 to 0.06. However, the frequency of cells carrying a dicentric only remains quite stable, whereas that of cells carrying additional rearrangement(s) progressively decreases (Fig. 4b). This indicates that the loss of dicentrics is not only related to their poor transmissibility but also to the eventual presence of other chromosomal rearrangements. Thus it is for a large part the multiplicity of lesions which determines the loss of dicentric carrier cells, since the data obtained result both from cell death and the slowing down of the cell cycle. However, comparing 2nd-generation mitoses from 72-h cultures to 3rd-generation mitoses from 96-h cultures, the frequency of dicentrics decreases from 0.24 to 0.13, which is compatible with a 50% reduction in one cell generation. This loss may result either from the instability of the dicentrics at anaphase, or from the imbalance caused by the frequent loss or duplication of the acentric fragments. Thus we have calculated the frequency of acentrics in cells carrying a dicentric but no other rearrangements. In lst-generation mitoses, all possess either 1 acentric fragment resulting from the fusion of the 2 deleted
196
(5) Analysis of reciprocal translocations (a)
A n=94
40.
34 30.
B
C
20. 10.
2
3
1
2
3
1
2
A n=4s
30
3
In 72-h cultures, the frequency of reciprocal translocations decreases from 1st to 2nd, and remains stable from 2nd to 3rd generations (Fig. 5a). However, the frequency of cells carrying a translocation only slightly increases from 1st to 3rd generation, whereas that of cells carrying additional rearrangement(s) decreases. In 96-h cultures (Fig. 5b), the variations of the frequency of translocations in relation to the cell cycle may be due to the small size of the samples. Nevertheless, there is a good conservation of translocations, as shown by the increased frequency in 3rd-division mitoses. This increase may be partly due to the loss of cells with dicentrics, which increases the relative frequency of cells with other rearrangement(s).
[b) (a) 8
C
20
A n= a s
30
13
13,,5 ~
10
1
2
3 >3
2_.~_e
9.2
1
2
3 >3
1 2
6
e
20. 3 >3
10.
Fig. 4. Distribution of dicentrics in 72-h (a) and 96-h (b) cultures per 100 mitoses, in relation to the number of cell divisions (1 to > 3). (A) In the whole sample, (B) in mitoses carrying 1 dicentric only, (C) in mitoses carrying 1 dicentric plus another rearrangement at least.
chromosomes (13/16) or the 2 non-reassociated acentrics (3/16). In 2nd generation 50% of the mitoses (24) possess either 1 or 2 acentrics, and in 3rd generation (20), this percentage is 35. These results are compatible with a random segregation of acentrics, half of them being eliminated at each generation. As can be expected, acentrics are not duplicated in 1st-generation ceils. They are found duplicated in about 15% of both 2nd- and 3rd-generation cells. This rate of duplication is much lower than the rate of loss. It is also noteworthy that the loss of dicentrics and that of acentrics do not seem to be correlated. This indicates that the deletions due to the loss of acentrics are not the major cause of elimination of cells carrying dicentrics.
1
23
1
23
1
23
(b)
A n=41) 20
10
B
C
n=28
n=18
2o
13.S12,9 ~
1
2
lo.8
3>3
1
11.~ • 2
3 >3
7.3 ?.4 1
2
3 >3
Fig. 5. Distribution of reciprocal translocations in 72-h (a) and 96-h (b) cultures per 100 mitoses, in relation to the number of cell divisions (1 to > 3). (A) In the whole sample, (B) in mitoses carrying 1 reciprocal translocation only, (C) in mitoses carrying 1 reciprocal translocation plus another rearrangement at least.
197
The analysis of deletions is not easy, because they are heterogeneous. On the one hand, it is not always possible to differentiate the 1-break terminal from intercalary deletions. On the other hand, in 2nd-and-more-generation mitoses, it is likely that some of the rearrangements interpreted as deletions result in fact from the breakage of unstable rearrangements like dicentrics. On the whole, the frequency of deletions remains fairly stable (Table 2).
50% at each generation. Thus, the loss of cells with dicentric chromosomes seems to be mainly due to mechanical reasons at anaphase but not to the monosomies. As can be expected, translocations, and probably also inversions, are fairly well maintained, especially when they are not associated with other anomalies, which lets us suppose that translocations induced by low doses of irradiation are better transmitted than those induced by higher doses. This should be taken into consideration for the calculation of genetic risk.
(7) Complex rearrangements
Acknowledgements
This category of rearrangements resulting from more than 2 breaks is still more heterogeneous than the deletions. Their frequency is fairly stable in 72-h cultures, but progressively decreases with the cell generations in 96-h cultures. This decrease is about one third per generation (Table 2).
We wish to thank Catherine Luccioni for the revision of the manuscript. This research was supported by Euratom (contract No. BI 6 149 F).
(6) Analysis of deletions
(8) Other rearrangements The frequency of inversions, rings and chromatid exchanges is too small to allow a valuable analysis. The results obtained are compatible with a progressive decrease of the frequencies of rings, possibly by a reduction of 50% per generation. Inversions are more conserved. Chromatid exchanges are rare, and may be poorly or not induced by radiations at G O phase. In conclusion, our results confirm the notion that radiation-induced rearrangements influence cell survival, and possibly also the rapidity of the cell cycle. These two parameters which depend on dose are closely involved making it impossible to separate them without the use of a BrdU incorporation technique. As was shown by previous reports (Carrano, 1973; Scott and Lyons, 1979; Van Buul and Natarajan, 1980; Beek, 1981; Dutrillaux et al., 1985), there is a decrease of the frequency of dicentrics compatible with a reduction of 50% at each cell division. However, we could show that this decrease is largely due to the loss of cells carrying several rearrangements, including one or more dicentrics. This shows that cell death is due to several factors, non-transmissible rearrangements being one of them. The loss of dicentric carrier cells is poorly related to the monosomy resulting from the loss of acentric fragments, although they are also lost at a rate compatible with
References Beek, B. (1981) Cell proliferation and chromosomal damage in human leukocytes; dicentrics and premature chromosome condensations in first, second and third mitoses after Xirradiation, Hum. Grnrt., 57, 75-77. Bender, M.A, and J.G. Brewen (1969) Factors influencing chromosome aberration yields in the human peripheral leukocyte system, Mutation Res., 8, 383-399. Carrano, A.V. (1973) Chromosome aberrations and radiationinduced cell death, I. Transmission and survival parameters of aberrations, Mutation Res., 17, 341-353. Crossen, P.E., and W.F. Morgan (1977) Analysis of human lymphocyte cell cycle time in culture measured by sister chromatid differential staining, Exp. Cell. Res., 104, 453-457. Dutrillaux, B., and J. Couturier (1981) La Pratique de l'Analyse Chromosomique, Masson, Paris. Dutrillaux, B., A.M. Fosse, M. Prieur and J. Lejeune (1974) Analyse des 6changes dans les cellules somatiques humaines, Chromosoma, 48, 327-340. Dutrillaux, B., E. Viegas-P&tuignot, M. Prod'homme and M. Sportes (1985) Distribution of various radiation-induced chromosomal rearrangements in relation to the dose and sampling time, Mutation Res., 152, 197-203. Lloyd, D.C., G.W. Dolphin, R.J. Purrott and P.A. Tipper (1977) The effect of X-ray induced mitotic delay on chromosome aberration yields in human lymphocytes, Mutation Res., 42, 401-412. Purrott, R.J., N. Vulpis and D.C. Lloyd (1980) The use of harlequin staining to measure delay in the human lymphocytes cell cycle induced by in vitro X-irradiation, Mutation Res., 69, 275-282. Scott, D., and C.Y. Lyons (1979) Homogeneous sensitivity of human peripheral blood lymphocytes to radiation-induced chromosome damage, Nature (London), 278, 756-758.
198 Steffen, J., and A. Michalowski (1973) Heterogeneous chromosomal radiosensitivity of phytohaemagglutinin-stimulated human blood lymphocytes in culture, Mutation Res., 17, 367-376. Van Buul, P.P.W., and A.T. Natarajan (1980) Chromosomal radiosensitivity of human leukocytes in relation to sampling time, Mutation Res., 70, 61-69.
Vulpis, N., R.J. Purrott and D.C. Lloyd (1978) The use of harlequin staining to assess radiation effects on the in vitro cell cycle kinetics of human lymphocytes, Mutation Res., 51, 145-148.