Chromosoma (Berl.) 59, 179-191 (1977)
CHROMOSOMA 9 by Springer-Verlag 1977
Mechanisms for Sister Chromatid Exchanges and Their Relation to the Production of Chromosomal Aberrations Hatao Kato Department of Cytogenetics, National Institute of Genetics, Mishima, Shizuoka-ken, 41l Japan
Abstract. By taking advantage of the fact that fluorescent light (FL) induces
strand breaks only in bromodeoxyuridine(BrdU)-substituted DNA, and that those breaks eventually lead to the formation of sister chromatid exchanges (SCEs), the response of SCEs to FL was studied carefully in Chinese hamster chromosomes in which, out of four DNA strands, BrdU-substitution had occurred either in one or three strands. The FL-induced SCE frequency did not differ greatly between these two types of chromosomes. However, when they were submitted to caffeine treatment, a drastic increase in the frequency was detected in the trifilarly-substituted chromosomes while a significant decrease occurred in the unifilarly-substituted chromosomes. Based on these results, a working hypothesis was developed that the SCE can arise by at least two different mechanisms, one operating at replicating points probably utilizing the machinery of DNA replication, and the other acting only in the post-replicational DNA portion, probably in a similar fashion as assumed in a general model of crossing over in the eukaryote. These dual mechanisms may account for the discrepancy encountered in the explanations of the induction of SCEs by various exogenous agents as well as spontaneous SCEs. The present study also showed that some, but clearly not all, of chromatid deletions are the outcome of the failure to complete SCEs arising through these mechanisms. Introduction
Exposure of cells to ultraviolet (UV) light results in a manyfold increase in the frequency of sister chromatid exchanges (SCEs) (Rommelaere et al., 1973; Kato, 1973), whereas X-irradiation causes only a slight increment in their frequency in spite of its remarkable efficiency at producing chromosomal aberrations Acknowledgement. Contribution No. 1097 from the National Institute of Genetics, Japan. This work was supported in part by a grant from the Ministry of Education, Science, and Culture of Japan
180
H. Kato
(Marin and Prescott, 1964; Gatti et al., 1974; Perry and Evans, 1975). Many of chemical agents are also capable of inducing SCEs (Kato, 1974b; Latt, 1974; Kihlman, 1974; Perry and Evans, 1975), and their modes of action in inducing SCEs differ considerably from agent to agent. Circumstantial evidence suggests strongly a close relation between a postreplicational de novo synthesis repair (Lehmann, 1972) and the UV-induced SCEs (Kato, 1973, 1974a, b; Wolff et al., 1974), but this relation cannot be extended to the explanation of the genesis of SCEs induced by X-rays. Furthermore, the yield of spontaneous SCEs seems to be independent of the ability of cells to perform postreplicational as well as excision repair process (Wolff et al., 1975). These contradictory features with respect to the induction of SCEs may be accounted for by an assumption that the mechanism involved in the formation of SCEs is by no means unitary (Kato, 1974b). It has been known that exposure of cells to flourescent light (FL) causes breaks in DNA strands substituted with bromodeoxyuridine (BrdU) (Ben-Hur and Elkind, 1972), and that these breaks eventually lead to the formation of SCEs (Ikushima and Wolff, 1974; Kato, 1974d; Wolff and Perry, 1974). In this experimental system we can obtain, by controlling the labeling conditions, chromatids with a known number of BrdU-labeled DNA strands and hence a known number of break-bearing strands following photolysis by FL. This is clearly advantageous in analyzing in detail the process involved in the SCE formation. In addition, a unifilarly-substituted chromatid can be distinguished unequivocally from a bifilarly-substituted and unsubstituted chromatids after staining with flourescent dyes (Latt, 1973; Kato, 1974c) or Giemsa (Perry and Wolff, 1974; Korenberg and Freedlender, 1974). In the work reported here, the FL-induced SCE formation was studied carefully in chromosomes in which DNA strands were substituted with BrdU at various levels. Results were consistent with the postulation that at least two different mechanisms are involved in the induction of SCEs. The genesis of chromatid aberrations will be also discussed in relation to the mechanisms of SCEs.
Experimental Rationale The previous work has shown that FL induces SCEs even in the chromosome where only one of four DNA strands contains BrdU (Kato, 1974d). Admitting that strand breaks are inducible only in BrdU-substituted DNA portion by FL, this finding seems to imply that only one single-strand break is sufficient to initiate a process leading to the SCE. There may be two alternative pathways that enable such a process: 1) A free end of a single-strand break induced in the BrdU-substituted DNA in a chromatid would pair with the complementary sequence in a partially denatured region of an unbroken DNA duplex in its sister chromatid. This pairing would then induce a single-strand break in the latter and lead to the formation of a Holliday structure (Holliday, 1964), resulting finally in an SCE. The initial step may be as shown in Figure 1, I. This process is essentially similar to a model of recombination proposed by Meselson and
Mechanism of SCEs and Their Relation to Chromosomal Aberrations
A f
181
B
m ~.-n
s
m ~ - . - ~
4.-
"~
,"
III
~
4, I
9
,
4.-
-
4
9
Fig. 1A and B. Schematic illustration of the labeling patterns of chromosomes at the second postlabeling S phase and possible steps (I, II and III) of the initiation of the formation of sister chromatid exchange following introduction of strand breaks by photolysis in BrdU-substituted DNA. A and B are tentatively designated as a "pulse-labeled (PL)" and "continuously-labeled (CL)" chromosomes, respectively. Thick lines indicate BrdU-substituted DNA strands and thin lines unsubstituted strands. Discontinuities in thick lines show the sites of phogolysis. In Ilia, the site of a break in the BrdU-substituted parental strand is shown, for convenience sake, at juxtaposition with another break in the co-paralM nascent strand. For explanation for each step, see text. Encircled regions in the step II are considered to be the site of the exchange initiation, though the exact process remains to be elucidated
Radding (1975). 2) The existence of two single-strand breaks is a prerequisite to the initiation of the SCE formation process and, in the given chromosome, one of the breaks would be provided in a discontinuously growing D N A strand at a replication fork, thus the exchange site being restricted to the replicating point (Fig. 1, II). A test for these possibilities was made on the basis of a rationale as follows: If two sets of cultures are labeled with an identical concentration of BrdU for either one or two rounds of cell cycle, the labeling patterns of D N A in the chromosome at the 2nd S phase would be as shown in Figure 1, A and B, respectively. For convenience the chromosome A will tentatively be referred to as a "pulse-labeled ( P L ) " chromosome and B as "continuously-labeled (CL)" one. It seems probable that the amount of BrdU incorporated into a nascent strand during the 2nd S phase could be less than (probably half of) that contained in the parental strand, for it has been shown that cells begin to utilize the de novo synthesized thymidine pool and salvage nucleotide pools to about equal extent during the 2nd replication cycle (Kuebbing and Werner, 1975). Thus the number of breaks which will be caused by FL-illumination in the BrdU-substituted parental strand would be about the same as the sum of the numbers of breaks induced in the two newly-formed strands in the CL chromosomes. This means that the duplicated region of the CL chromosome would
182
H. Kato
b e a r b r e a k s in three out o f strands, the n u m b e r being n o t three times b u t a b o u t twice as m a n y as t h a t in the d u p l i c a t e d region o f the P L c h r o m o s o m e . Therefore, p r o v i d e d t h a t the S C E f o r m a t i o n is to be i n i t i a t e d a c c o r d i n g to the first possibility, the incidence o f SCEs in the C L c h r o m o s o m e w o u l d be at least twice t h a t in the P L c h r o m o s o m e , a s s u m i n g t h a t each b r e a k possesses a n equal o p p o r t u n i t y to initiate one exchange process as shown in F i g u r e 1, I A , B. O n the other h a n d , if the exchange site is restricted to the replicating p o i n t as s h o w n in Fig. 1, II, A a n d B, the exchange f r e q u e n c y w o u l d be the same for b o t h types o f c h r o m o s o m e s , since it w o u l d vary as a function o f the n u m b e r o f r e p l i c a t i o n forks b u t n o t o f the n u m b e r o f b r e a k s c a r r i e d by the c h r o m o s o m e . F u r t h e r m o r e , if t w o b r e a k s need to exist for the initiation o f the process l e a d i n g to the S C E f o r m a t i o n , a n d if t h o s e l o c a t e d in the duplic a t e d D N A region also p a r t i c i p a t e in the S C E f o r m a t i o n (Fig. l, [II a or b), the exchange frequency in the C L c h r o m o s o m e s w o u l d be a p p r e c i a b l y higher t h a n t h a t in the P L c h r o m o s o m e s . T h e steps I I I a or b are similar to m o d e l s p r o p o s e d for crossing over by H o l l i d a y (1964) a n d W h i t e h o u s e (1963), respectively. O b v i o u s l y these steps do n o t occur in the P L c h r o m o s o m e .
Material and Methods Chinese hamster cells growing exponentially, D-6, were labeled with 1.0 ixg/ml BrdU in two ways: To obtain the PL chromosome cells were labeled for 5 h, washed with and incubated further for 21 h in fresh medium containing 20 gg/ml thymidine. To obtain the CL chromosome BrdU was present in culture medium continuously for 26 h before fixation for chromosome preparation. Six h before fixation (during the 2nd S) both cultures were exposed for 5, 10 and 20 min respectively to FL emitted from a 20W blue light bulb (National) located 5 cm above the bottom of culture dishes. When cells were treated with caffeine, the drug (1 mM) was added to culture medium immediately after FL illumination and was present till the harvest. Chromosome preparations were processed following 1 h-pretreatment of cells with 0.05 gg/ml colcemid and stained with 0.125 mg/ml acridine orange (Kato, 1974c). SCEs and chromosomal aberrations were scored in the largest chromosome, No. 1, using flourescence microscopy. To study unscheduled uptake of 3H-thymidine by non-S phase cells following FL illumination, cultures were treated with 50 gg/ml BrdU for 16 h (one round of cell cycle), washed twice with fresh medium and replenished with prewarmed medium containing 5 gCi/ml 3H-thymidine (spec. act., 12.6 Ci/mM). Immediately thereafter the cultures were exposed for 15, 30 and 60 rain respectively to FL under the same condition as described above. Incubation with the isotope was terminated 2 h later. Ceils were then fixed with a 3:1 mixture of methanol-acetic acid, coated with autoradiographic emulsion (Sakura, NR-M2) and developed after a 4 week-exposure. In each sample 150 non-S nuclei were scored. Large-sized nuclei showing a sign of condensation of chromatin (probably G2 nuclei) were omitted from the score.
Results T h e frequency o f SCEs i n d u c e d b y F L in the C L c h r o m o s o m e s was slightly higher t h a n t h a t in the P L c h r o m o s o m e , b u t the difference was n o t statistically significant (Fig. 2). This result seems to be consistent with the s e c o n d possibility (Fig. 1, step II). A t least in the present e x p e r i m e n t a l system the p r o c e s s as s h o w n in F i g u r e 1, I does n o t seem to occur. H o w e v e r , the t e n d e n c y t o w a r d
Mechanism of SCEs and Their Relation to Chromosomal Aberrations
183
30
S o
2o
/// A ///
0.3~ ID_
5 0.2~
0.1 C)
5
10 FL( min )
15
20
Fig. 2. Frequencies of SCEs and chromatid aberrations in the PL ( - - - ) and the CL ( - - ) chromosomes. Illumination of cells with FL was carried out 6 h before harvest (during the 2nd post-labeling S). Each point on the curves was determined based on 150 No. 1 chromosomes. Thick line, SCEs; thin line, chromatid aberrations
the slightly higher yield of SCEs in the CL chromosomes than in the PL ones was observed repeatedly in three different esperiments. It is feasible that, in addition to the step II, the step III (Fig. I, IIIa or b) might be also involved in the SCE formation in the CL chromosomes, though its contribution is minimal. On the other hand, the incidence of chromosomal aberrations differed greatly between the two types of chromosomes. At 20 rain-illumination, the incidence of chromatid type aberrations as a whole was about 2.7 times higher in the CL chromosomes than in the PL chromosomes; the frequency of chromatid deletions was about 4 times and gaps about 1.9 times higher in the former than the latter. This shows that strand breaks induced in the post-replicational DNA portion do contribute to the production of chromosomal aberrations.
Effects of Caffeine Rather perplexing is the finding that only a small fraction of SCEs can arise from the step III (Fig. 1, III), since this process is essentially identical to that in models for recombination in eukaryotes (Whitehouse, 1963; Holliday, 1964), and since the situation developed in the duplicated DNA region of the CL chromosome appears to suffice for the initiation of such a process. A possible explanation for the results may be, therefore, that repair of FL-induced DNA
184
H. Kato
Table 1, Frequencies of sister chromatid exchanges and chromatid aberrations in BrdU-substituted Chinese hamster No. 1 chromosomes following exposure to FL for 20 rain Treatment
SCEs (mean _+s.e.)
Chromatid aberrations (mean _+s.e.)
PL
CL
PL
CL
0.41_+0.05 0.56_+0.06 3.11 +0.14 1.73+_0.11
0.47_+0.06 0.65-+0.07 3.74_+0.16 5.34+__0.19
0.007_+0.006 0.033_+0.015 0.127-+0.029 0.153-+0.032
0.013_+0.009 0.040_+0.016 0.260-+0.042 0.593_+0.063
0.52_+0.06 0.51_+0.06
0.073_+0.022 0.060_+0.020
0.227_+0.039 0.307_+0.045
7 h before harvest (S) Control Caffeine FL FL + caffeine ~
2.5 h before harvest (G2) FL FL + caffeine
0.56-+0.06 0.43_+0.05
a Caffeine (1 mM) was added to culture medium immediately after FL illumination and present continuously till harvest. Each value was determined based on 150 chromosomes
damage might occur rapidly leaving a slender chance for two broken ends to interact and initiate an exchange process. This was examined by seeing if the given repair process could be disturbed by a repair inhibitor and if the SCE frequency could be affected thereby. In Table 1 are presented results of such an examination. The incidence of FL-induced SCEs was actually affected by post-illumination treatment of cells with caffeine. However, its effects differed greatly between the two types of chromosomes. In the PL chromosomes caffeine caused a significant reduction in the SCE frequency whereas it increased markedly the SCE frequency in the CL chromosomes. Caffeine treatment alone caused neither rise nor fall in the SCE frequency in the non-illuminated controls. These results suggests strongly that strand breaks made in the post-replicational DNA regions of the CL chromosomes do initiate the SCE formation if appropriate conditions are provided. The reduction in the SCE frequency in the PL chromosomes might be due to the fact that the initiation of the SCE process in these chromosomes is restricted to the replicating point and the binding of caffeine to these regions would prevent the formation of SCEs. Regardless of caffeine post-treatment, FL illumination of cells at G2 did not affect the SCE frequency to any greater or lesser extent in both types of chromosomes.
Unscheduled DNA Synthesis To obtain some insights into the nature of DNA damage caused by FL and its repair process, unscheduled uptake of 3H-thymidine by cells with BrdUsubstituted DNA after FL illumination was studied autoradiographically, Silver grains were detected in the nuclei of FL-treated non-S nuclei as well as in the S-phase nuclei. The number of grains on the non-S nuclei varied considerably from cell to cell (Fig. 3). Even in cells exposed to FL for 60 min the grain
Mechanism of SCEs and Their Relation to Chromosomal Aberrations
5-
20-.
185
0 rain
[
100 15rain
-6 20IOE 0
I
20-
L~
30min
100 I0
~ ,
F-
10
--~-L~ i
i
r..1
r.q r...t i...1 min
20
&O
30 Number of groins
50
60
q 7O
Fig. 3. Unscheduled uptake of 3H-thymidine by non-S phase cells following FL illumination. Cultures were treated with 50 gg/ml BrdU for 16 h (one round of cell cycle), washed twice with fresh medium and incubated in prewarmed medium containing 5 txCi/ml 3H-thymidine (spec. act. 12.6 Ci/ raM) for 2 h. Cultures were exposed to FL immediately after the onset of incubation with the isotope
Table 2. Effect of caffeine post-treatment of the unscheduled uptake of 3H-thymidine by non-S cells following exposure to FL Post-treatment
None 1 m M caffeine"
FL dose (rain) 0
15
30
60
7.37 6.40
11.55 6.70
14.55 7.88
27.55 20.65
a Caffeine was present during the illumination and 3H-labeling period, which lasted 2 h. Each figure indicates the average number of silver grains scored in 150 non-S nuclei. Other experimental conditions were the same as those described in Figure 3
PL CL
F L + caffeine"
PL CL
FL +caffeine ~
240 150
150 140
150 50
186 150
Number of cells scored
102 (0.43) 161 (1,07)
61 (0.41) 168 (l.20)
122 (0.82) 177 (3.54)
113 (0.61) 195 (1.30)
N u m b e r of aberrations scored b (per cell)
0/14 3/34
0/7 0/34
2/19 21/44
6/16 11/47
0 8.8
0 0
10,5 47.7
37.5 23.4
1/33 2/56
0/10 0/43
7/28 27/54
17/36 15/58
q.0 3.6
0 0
25.0 50.0
47,2 25.9
%
2/47 1/58
0/36 3/75
4.3 1.7
0 4.0
11.9 25.0
5/42 17/68
25,6 25.9
10/39 15/58
%
Single locus
Nondispl,
Displaced
%
C h r o m a t i d gaps
C h r o m a t i d deletions
2/8 5/13
2/8 3/16
12/33 4/11
10/22 13/32
Isolocus
25.0 38.5
25.0 18.8
36.4 36.4
45.5 40.6
%
Caffeine (l m M ) was added to culture m e d i u m immediately after FL illumination b Only chromatid deletions and gaps are listed. C h r o m a t i d exchanges and isochromatid deletions were also observed b u t they constituted a very small fraction of aberrations, so that they were omitted from the score. Nondisplaced deletions a n d gaps were difficult to distinguish, but if the discontinuity of stained material in a chromatid was not convincing, ~t was scored as a gap. As for isolocus gaps, only those where a connection could be clearly seen were scored, since if the space between the ends of densely stained material was too large, the exchange of label at the lesion might occur during the swelling and spreading of cells for metaphase preparation
PL CL
FL
2,5 h before harvest (G2)
PL CL
Type of label
FL
7 h before harvest (S)
Treatment
Table 3. Relative frequencies of chromatid deletions and gaps showing exchange of label at break point
Mechanism of SCEs and Their Relationto ChromosomalAberrations
187
numbers in many nuclei were within the range of the non-illuminated control level. These features seem to be somewhat different from those encountered in the unscheduled DNA synthesis following ultraviolet irradiation. Yet, the average number of grains was found to increase steadily with increasing dose of FL, indicating that the uptake of 3H-thymidine by the non-S nuclei was in fact ascribed to repair of FL-induced DNA damage. When caffeine was present during the illumination and labeling period, the incorporation of the isotope was inhibited to an appreciable degree, especially at low FL doses (Table 2). It is known that when BrdU-substituted DNA is exposed to FL or ultraviolet light, uracil is produced by debromination of BrdU, which ultimately results in a single-strand break (Smith and Hanawalt, 1969; Smets and Cornelis, 1971). In Chinese hamster cells this break seems to be rejoined within 60 min as judged from a sedimentation profile of FL-illuminated DNA in alkaline sucrose gradients (Ben-Hur and Elkind, 1972). Smets and Cornelis (1971) have postulated that this repair process might involve the following steps; excision of uracil, repair incorporation of thymidine, and closure of the break. The present results are in agreement with their postulation. Caffeine may bind to the site of damage and interfere with this excision-type repair, although this contrasts with the fact that caffeine does not exert its inhibitory effect on excision repair of ultraviolet-induced primidine dimers in Chinese hamster cells (Wolff and Scott, 1969).
Chromosomal Aberrations Chromatid deletions induced by FL often showed an exchange of label at break points suggesting that they could have arisen from the failure to complete the SCE process. In Table 3 are presented relative frequencies of chromatid deletions and gaps that could be positively identified as incomplete SCEs. The frequencies fluctuated depending on the labeling conditions of chromosomes, the time of FL illumination and the presence or absence of caffeine. Thus, when illuminated for 7 h before harvest (during the 2nd S) about 45% deletions were found to be associated with SCEs in the PL chromosomes while about 25% association was noted in the CL chromosomes. Caffeine treatment reversed this situation; about 18% in the PL and 50% in the CL chromosomes.
Discussion
Mechanisms of SCEs Results obtained in this study seem to suggest that the SCE can be derived through at least two different pathways, the initial steps of which might be those as shown in Figure 1, II and III respectively. Processes involved in these pathways are highly speculative and meant to be taken as a working hypothesis.
188
H. Kato
The first pathway (Fig. 1, II) needs only one FL-induced single-strand break for its initiation. Strand breaks utilized in this pathway might be either those induced in the pre-replicational DNA region near the replicating point and approached by the fork before being closed by repair synthesis, or those induced directly at the replicating point. The exact molecular process that would occur at this region remains to be elucidated. Nevertheless, one of the possibilities may be that a free end of a single-strand break would somehow accomplish exchange with the discontinuously replicating DNA strand probably with the aid of the machinery of DNA replication. Temporal detachment of histones from the replicating point may facilitate such an interaction. Thus, strictly, this pathway would require two breaks for its initiation, one being induced by FL and the other occurring spontaneously, both residing in co-parallel strands. The results as shown in Figure 2 indicate that the majority of FLinduced SCEs arise through this pathway irrespective of whether chromosomes are unifilarly-substituted or trifilarly-substituted with BrdU. Other circumstantial evidence for this pathway is as follows: 1) The site of SCEs detectable in a metaphase chromosome coincides sharply with the chromosomal region which has been actively engaging in DNA synthesis at the time of FL illumination (Kato, 1974d). 2) Treatment of Chinese hamster cells with fluorodeoxyuridine (FdU) reduces the rate of DNA replication to a small percentage of the control while it seems to cause accumulation of replicating replicons during the treatment (Amaldi etal. 1972). In fact, in spite of the extremely low rate of replication, the number of cells that engage in replication after 6 h- FdU treatment reaches about 1.5 times that in the untreated control cell population. FL illumination of cells at this stage results in the SCE induction at about 1.5 times higher frequency than that obtained in cells untreated with FrdU, suggesting an intimate correlation between the site of SCEs and the replicating point (Kato, in preparation). The second type of pathway (Fig. 1, III a or b) would become available only when normal repair of FL-induced DNA damage is disturbed somehow, and would function only in the post-replicational DNA regions, probably in a similar fashion to that assumed in models of meiosis or gene conversion (Holliday, 1964; Whitehouse, 1963). It requires the presence of two single-strand breaks of staggered or juxtaposition in two sister chromatids, and eventual displacement of a free end of a single strand at the break point, thereby enabling the interaction of two DNA molecules. As described above the event that would follow the initial steps in both pathways is considered to be a generation of a heteroduplex (or a Holliday structure). However, this structure might soon be transformed into a doublestrand exdhange after cleavage of the non-crossing strands by a endonucleolytic attack (Sobell, 1973), since there is strong evidence against the SCE to be a single-strand exchange (Taylor, 1958; Wolff and Perry, 1975). Whether these two types of pathways occur in general in the SCE formation is yet uncertain. Inefficiency of X-rays to induce SCEs (Marin and Prescott, 1964; Gatti et al., 1974; Perry and Evans, 1975) may be attributed to an extensive operation of the first pathway by taking into consideration the short-lived nature of X-ray-induced DNA damage (Painter and Young, 1972) and the site specificity
Mechanism of SCEs and Their Relationto ChromosomalAberrations
189
of this pathway. Spontaneous SCEs may also arise from this pathway. When the replication fork reaches the site of spontaneous DNA damage, e.g., apurinic sites (Lindahl and Andersson, 1972; Lindahl and Nyberg, 1972; Verly et al., 1973), it may incorporate the lesion into an exchange process either by mistake or as a way of rescueing the damage (Kato, 1974a). On the other hand, ultraviolet-induced SCEs may arise from the second pathway. It has been known that the presence of pyrimidine dimers causes upon DNA replication a formation of gaps in the newly formed strand opposite the dimers (Rupp and HowardFlanders, 1968; Rupp et al., 1971; Lehmann, 1972). In mammalian cells these gaps, averaging 1000 nucleotide long, supposedly exist for some time and are somehow filled in by de novo DNA synthesis (Lehmann, 1972; Buhl et al., 1972; Fujiwara, 1972). It seems plausible that before a gap is closed, a free end at the gap be denatured to interact with a similarly displaced free end in the sister molecule.
Relation of SCEs to Chromatid Aberrations
To interpret results presented in Table 3, it seems possible to draw the following picture : There might exist two distinct types of chromatid deletions, one arising from the partial failure in the SCE event thus having a switch of label at the break point, and the other being derived from unrejoined strand breaks thus lacking in an associated exchange of sister chromatids. In the PL chromosomes, about a half of chromatid deletions would result from incomplete SCEs initiated at the replicating points and the rest being derived from unrejoined strand breaks induced in the duplicated regions. In the CL chromosomes, although chromatid deletions having label switch at break points would arise in a similar manner as in the PL chromosomes, a frequential production of SCE-independent aberrations in the duplicated DNA regions would lower the relative frequency of deletions with associated SCEs. On the other hand, caffeine posttreatment enhances the initiation of the SCE formation in the duplicated regions in the CL chromosomes so enhancing the production of aberrations due to incomplete SCEs. The relative frequency of such deletions is thus expected to be much higher than that obtained without caffeine treatment. In the PL chromosomes, caffeine interferes with the SCE formation (Table 1), thereby causing a decrease in the number of the SCE-associated aberrations. Caffeine may also inhibit repair of breaks in the post-replicational DNA regions (Table 2) so that the production of aberrations independent on the SCE process would be enhanced. Thus, the relative frequency of chromatid deletions associated with SCEs is expected to be lower than that obtained without caffeine treatment. FL illumination at G2 cannot induce SCEs (Table 1), so that chromatid deletions produced are without associated SCEs indicating that they arise from unrejoined strand breaks. Previous workers have examined the presence or absence of SCEs at break points of X-ray-induced chromatid breaks, and have reached a conclusion that many of chromatid deletions are simple deletions arising from single lesions and some are incomplete exchange between sister chromatids (Heddle and Body-
190 c o t e , 1 9 6 9 ; H e d d l e et al., 1970; W o l f f a n d B o d y c o t e ,
H. Kato 1975). R e s u l t s o b t a i n e d
in the present study seem to be compatible with their findings.
References Amaldi, F., Carnevali, F., Leoni, L., Mariotti, D.: Replicon origins in Chinese hamster cell DNA. I. Labeling procedure and preliminary observations. Expl. Cell Res. 74, 367-374 (1972) Ben-Hur, E., Elkind, M.M.: Damage and repair of DNA in 5-bromodeoxyuridine-labeled Chinese hamster cells exposed to fluorescent light. Biophys. J. 12, 636-647 (1972) Buhl, S.N., Setlow, R.B., Regan, J.D. : Steps in DNA chain elongation and joining after ultra-violet irradiation of human cells. Int. J. Radiat. Biol. 22, 417-424 (1972) Fujiwara, Y.: Characteristics of DNA synthesis following ultraviolet light irradiation in mouse L cells. Postreplication repair. Exp. Cell Res. 75, 483-489 (1972) Gatti, M., Pimpinelli, S., Olivieri, G.: The frequency and distribution of isolabelling in Chinese hamster chromosomes after exposure to X-rays. Mutation Res. 23, 229-238 (1974) Heddle, J.A., Bodycote, D.J.: On the formation of chromosomal aberrations. Mutation Res. 9, 117 126 (1970) Heddle, J.A., Whissell, D,, Bodycote, D.J. : Changes in chromosome structure induced by radiations: a test of the two chief hypotheses. Nature (Lond.) 221, 1158-1160 (1969) Holliday, R.: A mechanism for gene conversion in fungi. Geuet. Res. (Camb.) 5, 282 304 (1964) Ikushima, T., Wolff, S. : Sister chromatid exchanges induced by light flashes to 5-bromodeoxyuridineand 5-iodedeoxyuridine-substituted Chinese hamster chromosomes. Exp, Cell Res. 87, 15-19 (1974) Kato, H.: Induction of sister chromatid exchanges by UV light and its inhibition by caffeine. Exp. Cell Res. 82, 383-390 (1973) Kato, H. : Photoreactivation of sister chromatid exchanges induced by ultraviolet irradiation. Nature (Lond 0 249, 552-553 (1974a) Kato, H. : Induction of sister chromatid exchanges by chemical mutagens and its possible relevance to DNA repair. Exp. Cell Res. 85, 239-247 (1974b) Kato, H. : Spontaneous sister chromatid exchanges detected by a BUdR-labelling method. Nature (Lond,) 251, 70-72 (1974c) Kato, H.: Possible role of DNA synthesis in formation of sister chromatid exchanges. Nature (Loud.) 252, 739-741 (1974d) Kihlman, B.A.: Sister chromatid exchanges in Vicia faba. II. Effects of thiotepa, caffeine and 8-ethoxycaffeine on the frequency of SCEs. Chromosome (Berl.) 51, 11 18 (1975) Korenberg, J.R., Freedlender, E.F.: Giemsa technique for the detection of sister chromatid exo changes. Chromosoma (Berl.) 48, 355-360 (1974) Kuebbing, D., Werner, R.: A model for compartmentation of de novo and salvage thymidine nucleotide pools in mammalian cells. Proc. nat. Acad. Sci. (Wash.) 72, 3333 3336 (1975) Latt, S.A.: Microfluorometric detection of deoxyribonucleic acid replication in human metaphase chromosomes. Proc. nat. Acad. Sci. (Wash.) 70, 3395-3399 (1973) Latt, S.A. : Sister chromatid exchanges, indices of human chromosome damage and repair: Detection by fluorescence and induction by mitomycin C. Proc. nat. Acad. Sci. (Wash.) 71, 3162-3166 (1974) Lehmann, A.R. : Postreplication repair of DNA in ultraviolet-irradiated mammalian cells. J. molec. Biol. 66, 319-337 (1972) Lindahl, T., Andersson, A. : Rate of chain breakage at apurinic sites in double-stranded deoxyribonucleic acid. Biochemistry 11, 3618 3623 (1972) Lindahl, T., Nyberg, B.: Rate of depurination of nativ deoxyribonucleic acid. Biochemistry 11, 3610-3618 (1972) Marin, G. Prescott, D M . : The frequency of sister chromatid exchanges following exposure to varying doses of H3-thymidine or X-rays. J. Cell Biol. 21, 159-167 (1964) Meselson, M.S., Radding, C.M.: A general model for genetic recombination. Proc. nat. Acad. Sci. (Wash.) 72, 358-361 (1975) Painter, R.B., Young, B.R.: Repair replication in mammalian cells .after X-irradiation. Mutation Res. 14, 225-235 (1972)
Mechanism of SCEs and Their Relation to Chromosomal Aberrations
191
Perry, P., Evans, H.J. : Cytological detection of mutagen-carcinogen exposure by sister chromatid exchange. Nature (Lond.) 258, 121 125 (1975) Perry, P., Wolff, S. : New Giemsa method for the differential staining of sister chromatids. Nature (Lond.) 251, 156-158 (1974) Rommelaere, J., Suskind, M., Errera, M.: Chromosome and chromatid exchanges in Chinese hamster cells. Chromosoma (Berl.) 41, 243 257 (1973) Rupp, W.D., Howard-Flanders, P. : Discontinuities in the DNA synthesized in an excision-defective strain of Escherichia coli following ultraviolet irradiation. J. molec. Biol. 31, 291-304 (1968) Rupp, W.D., Wilde III, C.E., Reno, D.L., Howard-Flanders, P. : Exchanges between DNA strands in ultraviolet-irradiated Escherichia coli. J. molec. Biol. 61, 25M4 (1971) Smets, L.A., Cornelis, J.J.: Repairable and irrepairable damage in 5-bromouracil-substituted DNA exposed to ultra-violet radiation. Int. J. Radiat. Biol. 19, 445 457 (i971) Smith, K.C., Hanawalt, P.C.: Molecular photobiology. New York and London: Academic Press 1969 Sobell, H.M.: Symmetry in protein-nucleic acid interaction and its genetic implications. Advanc. Genet. 17, 411o490 (1973) Taylor, J.H.: Sister chromatid exchanges in tritum-labeled chromosomes. Genetics 43, 515-529 (1958) Verly, W.G., Paquette, Y., Thibodeau, L.: Nuclease for DNA apurinic sites may be involved in the maintenance of DNA in normal cells. Nature (Lond.) New Biol. 244, 67-69 (1973) Whitehouse, H.L.K.: A theory of crossing-over by means of hybrid deoxyribonucleic acid. Nature (Lond.) 199, 1034-1040 (1963) Wolff, S., Bodycote, J.: The induction of chromatid deletions in accord with the breakage-andreunion hypothesis. Mutation Res. 29, 85-91 (1975) Wolff, S., Bodycote, J., Painter, R.B.: Sister chromatid exchanges induced in Chinese hamster cells by UV irradiation of different stages of the cell cycle: The necessity for cells to pass through S. Mutation Res. 25, 73-81 (1974) Wolff, S., Bodycote, J., Thomas, G.H., Cleaver, J.E.: Sister chromatid exchange in xeroderma pigmentosum cells that are defective in DNA excision repair or post-replication repair. Genetics 81, 349-355 (I975) Wolff, S., Perry, P.: Differential Giemsa staining of sister chromatids and the ~study of sister chromatid exchanges without autoradiography. Chromosoma (Berl.) 48, 341-353 (1974) Wolff, S., Perry, P.: Insights on chromosome structure from sister chromatid exchange ratios and the lack of both isolabelling and heterolabelling as determined by the FPG technique. Exp. Ceil Res. 93, 23-30 (1975) Wolff, S., Scott, D. : Repair of radiation-induced damage to chromosomes. Independence of known DNA dark repair mechanisms. Exp. Cell Res. 55, 9 16 (1969)
Received February 9 November 1, 1976 / Accepted March 4, 1976 by J.H. Taylor Ready for press November 1, 1976
CHROMOSOMA 9 by Springer-Verlag 1977
Mechanisms for Sister Chromatid Exchanges and Their Relation to the Production of Chromosomal Aberrations Hatao Kato Department of Cytogenetics, National Institute of Genetics, Mishima, Shizuoka-ken, 41l Japan
Abstract. By taking advantage of the fact that fluorescent light (FL) induces
strand breaks only in bromodeoxyuridine(BrdU)-substituted DNA, and that those breaks eventually lead to the formation of sister chromatid exchanges (SCEs), the response of SCEs to FL was studied carefully in Chinese hamster chromosomes in which, out of four DNA strands, BrdU-substitution had occurred either in one or three strands. The FL-induced SCE frequency did not differ greatly between these two types of chromosomes. However, when they were submitted to caffeine treatment, a drastic increase in the frequency was detected in the trifilarly-substituted chromosomes while a significant decrease occurred in the unifilarly-substituted chromosomes. Based on these results, a working hypothesis was developed that the SCE can arise by at least two different mechanisms, one operating at replicating points probably utilizing the machinery of DNA replication, and the other acting only in the post-replicational DNA portion, probably in a similar fashion as assumed in a general model of crossing over in the eukaryote. These dual mechanisms may account for the discrepancy encountered in the explanations of the induction of SCEs by various exogenous agents as well as spontaneous SCEs. The present study also showed that some, but clearly not all, of chromatid deletions are the outcome of the failure to complete SCEs arising through these mechanisms. Introduction
Exposure of cells to ultraviolet (UV) light results in a manyfold increase in the frequency of sister chromatid exchanges (SCEs) (Rommelaere et al., 1973; Kato, 1973), whereas X-irradiation causes only a slight increment in their frequency in spite of its remarkable efficiency at producing chromosomal aberrations Acknowledgement. Contribution No. 1097 from the National Institute of Genetics, Japan. This work was supported in part by a grant from the Ministry of Education, Science, and Culture of Japan
180
H. Kato
(Marin and Prescott, 1964; Gatti et al., 1974; Perry and Evans, 1975). Many of chemical agents are also capable of inducing SCEs (Kato, 1974b; Latt, 1974; Kihlman, 1974; Perry and Evans, 1975), and their modes of action in inducing SCEs differ considerably from agent to agent. Circumstantial evidence suggests strongly a close relation between a postreplicational de novo synthesis repair (Lehmann, 1972) and the UV-induced SCEs (Kato, 1973, 1974a, b; Wolff et al., 1974), but this relation cannot be extended to the explanation of the genesis of SCEs induced by X-rays. Furthermore, the yield of spontaneous SCEs seems to be independent of the ability of cells to perform postreplicational as well as excision repair process (Wolff et al., 1975). These contradictory features with respect to the induction of SCEs may be accounted for by an assumption that the mechanism involved in the formation of SCEs is by no means unitary (Kato, 1974b). It has been known that exposure of cells to flourescent light (FL) causes breaks in DNA strands substituted with bromodeoxyuridine (BrdU) (Ben-Hur and Elkind, 1972), and that these breaks eventually lead to the formation of SCEs (Ikushima and Wolff, 1974; Kato, 1974d; Wolff and Perry, 1974). In this experimental system we can obtain, by controlling the labeling conditions, chromatids with a known number of BrdU-labeled DNA strands and hence a known number of break-bearing strands following photolysis by FL. This is clearly advantageous in analyzing in detail the process involved in the SCE formation. In addition, a unifilarly-substituted chromatid can be distinguished unequivocally from a bifilarly-substituted and unsubstituted chromatids after staining with flourescent dyes (Latt, 1973; Kato, 1974c) or Giemsa (Perry and Wolff, 1974; Korenberg and Freedlender, 1974). In the work reported here, the FL-induced SCE formation was studied carefully in chromosomes in which DNA strands were substituted with BrdU at various levels. Results were consistent with the postulation that at least two different mechanisms are involved in the induction of SCEs. The genesis of chromatid aberrations will be also discussed in relation to the mechanisms of SCEs.
Experimental Rationale The previous work has shown that FL induces SCEs even in the chromosome where only one of four DNA strands contains BrdU (Kato, 1974d). Admitting that strand breaks are inducible only in BrdU-substituted DNA portion by FL, this finding seems to imply that only one single-strand break is sufficient to initiate a process leading to the SCE. There may be two alternative pathways that enable such a process: 1) A free end of a single-strand break induced in the BrdU-substituted DNA in a chromatid would pair with the complementary sequence in a partially denatured region of an unbroken DNA duplex in its sister chromatid. This pairing would then induce a single-strand break in the latter and lead to the formation of a Holliday structure (Holliday, 1964), resulting finally in an SCE. The initial step may be as shown in Figure 1, I. This process is essentially similar to a model of recombination proposed by Meselson and
Mechanism of SCEs and Their Relation to Chromosomal Aberrations
A f
181
B
m ~.-n
s
m ~ - . - ~
4.-
"~
,"
III
~
4, I
9
,
4.-
-
4
9
Fig. 1A and B. Schematic illustration of the labeling patterns of chromosomes at the second postlabeling S phase and possible steps (I, II and III) of the initiation of the formation of sister chromatid exchange following introduction of strand breaks by photolysis in BrdU-substituted DNA. A and B are tentatively designated as a "pulse-labeled (PL)" and "continuously-labeled (CL)" chromosomes, respectively. Thick lines indicate BrdU-substituted DNA strands and thin lines unsubstituted strands. Discontinuities in thick lines show the sites of phogolysis. In Ilia, the site of a break in the BrdU-substituted parental strand is shown, for convenience sake, at juxtaposition with another break in the co-paralM nascent strand. For explanation for each step, see text. Encircled regions in the step II are considered to be the site of the exchange initiation, though the exact process remains to be elucidated
Radding (1975). 2) The existence of two single-strand breaks is a prerequisite to the initiation of the SCE formation process and, in the given chromosome, one of the breaks would be provided in a discontinuously growing D N A strand at a replication fork, thus the exchange site being restricted to the replicating point (Fig. 1, II). A test for these possibilities was made on the basis of a rationale as follows: If two sets of cultures are labeled with an identical concentration of BrdU for either one or two rounds of cell cycle, the labeling patterns of D N A in the chromosome at the 2nd S phase would be as shown in Figure 1, A and B, respectively. For convenience the chromosome A will tentatively be referred to as a "pulse-labeled ( P L ) " chromosome and B as "continuously-labeled (CL)" one. It seems probable that the amount of BrdU incorporated into a nascent strand during the 2nd S phase could be less than (probably half of) that contained in the parental strand, for it has been shown that cells begin to utilize the de novo synthesized thymidine pool and salvage nucleotide pools to about equal extent during the 2nd replication cycle (Kuebbing and Werner, 1975). Thus the number of breaks which will be caused by FL-illumination in the BrdU-substituted parental strand would be about the same as the sum of the numbers of breaks induced in the two newly-formed strands in the CL chromosomes. This means that the duplicated region of the CL chromosome would
182
H. Kato
b e a r b r e a k s in three out o f strands, the n u m b e r being n o t three times b u t a b o u t twice as m a n y as t h a t in the d u p l i c a t e d region o f the P L c h r o m o s o m e . Therefore, p r o v i d e d t h a t the S C E f o r m a t i o n is to be i n i t i a t e d a c c o r d i n g to the first possibility, the incidence o f SCEs in the C L c h r o m o s o m e w o u l d be at least twice t h a t in the P L c h r o m o s o m e , a s s u m i n g t h a t each b r e a k possesses a n equal o p p o r t u n i t y to initiate one exchange process as shown in F i g u r e 1, I A , B. O n the other h a n d , if the exchange site is restricted to the replicating p o i n t as s h o w n in Fig. 1, II, A a n d B, the exchange f r e q u e n c y w o u l d be the same for b o t h types o f c h r o m o s o m e s , since it w o u l d vary as a function o f the n u m b e r o f r e p l i c a t i o n forks b u t n o t o f the n u m b e r o f b r e a k s c a r r i e d by the c h r o m o s o m e . F u r t h e r m o r e , if t w o b r e a k s need to exist for the initiation o f the process l e a d i n g to the S C E f o r m a t i o n , a n d if t h o s e l o c a t e d in the duplic a t e d D N A region also p a r t i c i p a t e in the S C E f o r m a t i o n (Fig. l, [II a or b), the exchange frequency in the C L c h r o m o s o m e s w o u l d be a p p r e c i a b l y higher t h a n t h a t in the P L c h r o m o s o m e s . T h e steps I I I a or b are similar to m o d e l s p r o p o s e d for crossing over by H o l l i d a y (1964) a n d W h i t e h o u s e (1963), respectively. O b v i o u s l y these steps do n o t occur in the P L c h r o m o s o m e .
Material and Methods Chinese hamster cells growing exponentially, D-6, were labeled with 1.0 ixg/ml BrdU in two ways: To obtain the PL chromosome cells were labeled for 5 h, washed with and incubated further for 21 h in fresh medium containing 20 gg/ml thymidine. To obtain the CL chromosome BrdU was present in culture medium continuously for 26 h before fixation for chromosome preparation. Six h before fixation (during the 2nd S) both cultures were exposed for 5, 10 and 20 min respectively to FL emitted from a 20W blue light bulb (National) located 5 cm above the bottom of culture dishes. When cells were treated with caffeine, the drug (1 mM) was added to culture medium immediately after FL illumination and was present till the harvest. Chromosome preparations were processed following 1 h-pretreatment of cells with 0.05 gg/ml colcemid and stained with 0.125 mg/ml acridine orange (Kato, 1974c). SCEs and chromosomal aberrations were scored in the largest chromosome, No. 1, using flourescence microscopy. To study unscheduled uptake of 3H-thymidine by non-S phase cells following FL illumination, cultures were treated with 50 gg/ml BrdU for 16 h (one round of cell cycle), washed twice with fresh medium and replenished with prewarmed medium containing 5 gCi/ml 3H-thymidine (spec. act., 12.6 Ci/mM). Immediately thereafter the cultures were exposed for 15, 30 and 60 rain respectively to FL under the same condition as described above. Incubation with the isotope was terminated 2 h later. Ceils were then fixed with a 3:1 mixture of methanol-acetic acid, coated with autoradiographic emulsion (Sakura, NR-M2) and developed after a 4 week-exposure. In each sample 150 non-S nuclei were scored. Large-sized nuclei showing a sign of condensation of chromatin (probably G2 nuclei) were omitted from the score.
Results T h e frequency o f SCEs i n d u c e d b y F L in the C L c h r o m o s o m e s was slightly higher t h a n t h a t in the P L c h r o m o s o m e , b u t the difference was n o t statistically significant (Fig. 2). This result seems to be consistent with the s e c o n d possibility (Fig. 1, step II). A t least in the present e x p e r i m e n t a l system the p r o c e s s as s h o w n in F i g u r e 1, I does n o t seem to occur. H o w e v e r , the t e n d e n c y t o w a r d
Mechanism of SCEs and Their Relation to Chromosomal Aberrations
183
30
S o
2o
/// A ///
0.3~ ID_
5 0.2~
0.1 C)
5
10 FL( min )
15
20
Fig. 2. Frequencies of SCEs and chromatid aberrations in the PL ( - - - ) and the CL ( - - ) chromosomes. Illumination of cells with FL was carried out 6 h before harvest (during the 2nd post-labeling S). Each point on the curves was determined based on 150 No. 1 chromosomes. Thick line, SCEs; thin line, chromatid aberrations
the slightly higher yield of SCEs in the CL chromosomes than in the PL ones was observed repeatedly in three different esperiments. It is feasible that, in addition to the step II, the step III (Fig. I, IIIa or b) might be also involved in the SCE formation in the CL chromosomes, though its contribution is minimal. On the other hand, the incidence of chromosomal aberrations differed greatly between the two types of chromosomes. At 20 rain-illumination, the incidence of chromatid type aberrations as a whole was about 2.7 times higher in the CL chromosomes than in the PL chromosomes; the frequency of chromatid deletions was about 4 times and gaps about 1.9 times higher in the former than the latter. This shows that strand breaks induced in the post-replicational DNA portion do contribute to the production of chromosomal aberrations.
Effects of Caffeine Rather perplexing is the finding that only a small fraction of SCEs can arise from the step III (Fig. 1, III), since this process is essentially identical to that in models for recombination in eukaryotes (Whitehouse, 1963; Holliday, 1964), and since the situation developed in the duplicated DNA region of the CL chromosome appears to suffice for the initiation of such a process. A possible explanation for the results may be, therefore, that repair of FL-induced DNA
184
H. Kato
Table 1, Frequencies of sister chromatid exchanges and chromatid aberrations in BrdU-substituted Chinese hamster No. 1 chromosomes following exposure to FL for 20 rain Treatment
SCEs (mean _+s.e.)
Chromatid aberrations (mean _+s.e.)
PL
CL
PL
CL
0.41_+0.05 0.56_+0.06 3.11 +0.14 1.73+_0.11
0.47_+0.06 0.65-+0.07 3.74_+0.16 5.34+__0.19
0.007_+0.006 0.033_+0.015 0.127-+0.029 0.153-+0.032
0.013_+0.009 0.040_+0.016 0.260-+0.042 0.593_+0.063
0.52_+0.06 0.51_+0.06
0.073_+0.022 0.060_+0.020
0.227_+0.039 0.307_+0.045
7 h before harvest (S) Control Caffeine FL FL + caffeine ~
2.5 h before harvest (G2) FL FL + caffeine
0.56-+0.06 0.43_+0.05
a Caffeine (1 mM) was added to culture medium immediately after FL illumination and present continuously till harvest. Each value was determined based on 150 chromosomes
damage might occur rapidly leaving a slender chance for two broken ends to interact and initiate an exchange process. This was examined by seeing if the given repair process could be disturbed by a repair inhibitor and if the SCE frequency could be affected thereby. In Table 1 are presented results of such an examination. The incidence of FL-induced SCEs was actually affected by post-illumination treatment of cells with caffeine. However, its effects differed greatly between the two types of chromosomes. In the PL chromosomes caffeine caused a significant reduction in the SCE frequency whereas it increased markedly the SCE frequency in the CL chromosomes. Caffeine treatment alone caused neither rise nor fall in the SCE frequency in the non-illuminated controls. These results suggests strongly that strand breaks made in the post-replicational DNA regions of the CL chromosomes do initiate the SCE formation if appropriate conditions are provided. The reduction in the SCE frequency in the PL chromosomes might be due to the fact that the initiation of the SCE process in these chromosomes is restricted to the replicating point and the binding of caffeine to these regions would prevent the formation of SCEs. Regardless of caffeine post-treatment, FL illumination of cells at G2 did not affect the SCE frequency to any greater or lesser extent in both types of chromosomes.
Unscheduled DNA Synthesis To obtain some insights into the nature of DNA damage caused by FL and its repair process, unscheduled uptake of 3H-thymidine by cells with BrdUsubstituted DNA after FL illumination was studied autoradiographically, Silver grains were detected in the nuclei of FL-treated non-S nuclei as well as in the S-phase nuclei. The number of grains on the non-S nuclei varied considerably from cell to cell (Fig. 3). Even in cells exposed to FL for 60 min the grain
Mechanism of SCEs and Their Relation to Chromosomal Aberrations
5-
20-.
185
0 rain
[
100 15rain
-6 20IOE 0
I
20-
L~
30min
100 I0
~ ,
F-
10
--~-L~ i
i
r..1
r.q r...t i...1 min
20
&O
30 Number of groins
50
60
q 7O
Fig. 3. Unscheduled uptake of 3H-thymidine by non-S phase cells following FL illumination. Cultures were treated with 50 gg/ml BrdU for 16 h (one round of cell cycle), washed twice with fresh medium and incubated in prewarmed medium containing 5 txCi/ml 3H-thymidine (spec. act. 12.6 Ci/ raM) for 2 h. Cultures were exposed to FL immediately after the onset of incubation with the isotope
Table 2. Effect of caffeine post-treatment of the unscheduled uptake of 3H-thymidine by non-S cells following exposure to FL Post-treatment
None 1 m M caffeine"
FL dose (rain) 0
15
30
60
7.37 6.40
11.55 6.70
14.55 7.88
27.55 20.65
a Caffeine was present during the illumination and 3H-labeling period, which lasted 2 h. Each figure indicates the average number of silver grains scored in 150 non-S nuclei. Other experimental conditions were the same as those described in Figure 3
PL CL
F L + caffeine"
PL CL
FL +caffeine ~
240 150
150 140
150 50
186 150
Number of cells scored
102 (0.43) 161 (1,07)
61 (0.41) 168 (l.20)
122 (0.82) 177 (3.54)
113 (0.61) 195 (1.30)
N u m b e r of aberrations scored b (per cell)
0/14 3/34
0/7 0/34
2/19 21/44
6/16 11/47
0 8.8
0 0
10,5 47.7
37.5 23.4
1/33 2/56
0/10 0/43
7/28 27/54
17/36 15/58
q.0 3.6
0 0
25.0 50.0
47,2 25.9
%
2/47 1/58
0/36 3/75
4.3 1.7
0 4.0
11.9 25.0
5/42 17/68
25,6 25.9
10/39 15/58
%
Single locus
Nondispl,
Displaced
%
C h r o m a t i d gaps
C h r o m a t i d deletions
2/8 5/13
2/8 3/16
12/33 4/11
10/22 13/32
Isolocus
25.0 38.5
25.0 18.8
36.4 36.4
45.5 40.6
%
Caffeine (l m M ) was added to culture m e d i u m immediately after FL illumination b Only chromatid deletions and gaps are listed. C h r o m a t i d exchanges and isochromatid deletions were also observed b u t they constituted a very small fraction of aberrations, so that they were omitted from the score. Nondisplaced deletions a n d gaps were difficult to distinguish, but if the discontinuity of stained material in a chromatid was not convincing, ~t was scored as a gap. As for isolocus gaps, only those where a connection could be clearly seen were scored, since if the space between the ends of densely stained material was too large, the exchange of label at the lesion might occur during the swelling and spreading of cells for metaphase preparation
PL CL
FL
2,5 h before harvest (G2)
PL CL
Type of label
FL
7 h before harvest (S)
Treatment
Table 3. Relative frequencies of chromatid deletions and gaps showing exchange of label at break point
Mechanism of SCEs and Their Relationto ChromosomalAberrations
187
numbers in many nuclei were within the range of the non-illuminated control level. These features seem to be somewhat different from those encountered in the unscheduled DNA synthesis following ultraviolet irradiation. Yet, the average number of grains was found to increase steadily with increasing dose of FL, indicating that the uptake of 3H-thymidine by the non-S nuclei was in fact ascribed to repair of FL-induced DNA damage. When caffeine was present during the illumination and labeling period, the incorporation of the isotope was inhibited to an appreciable degree, especially at low FL doses (Table 2). It is known that when BrdU-substituted DNA is exposed to FL or ultraviolet light, uracil is produced by debromination of BrdU, which ultimately results in a single-strand break (Smith and Hanawalt, 1969; Smets and Cornelis, 1971). In Chinese hamster cells this break seems to be rejoined within 60 min as judged from a sedimentation profile of FL-illuminated DNA in alkaline sucrose gradients (Ben-Hur and Elkind, 1972). Smets and Cornelis (1971) have postulated that this repair process might involve the following steps; excision of uracil, repair incorporation of thymidine, and closure of the break. The present results are in agreement with their postulation. Caffeine may bind to the site of damage and interfere with this excision-type repair, although this contrasts with the fact that caffeine does not exert its inhibitory effect on excision repair of ultraviolet-induced primidine dimers in Chinese hamster cells (Wolff and Scott, 1969).
Chromosomal Aberrations Chromatid deletions induced by FL often showed an exchange of label at break points suggesting that they could have arisen from the failure to complete the SCE process. In Table 3 are presented relative frequencies of chromatid deletions and gaps that could be positively identified as incomplete SCEs. The frequencies fluctuated depending on the labeling conditions of chromosomes, the time of FL illumination and the presence or absence of caffeine. Thus, when illuminated for 7 h before harvest (during the 2nd S) about 45% deletions were found to be associated with SCEs in the PL chromosomes while about 25% association was noted in the CL chromosomes. Caffeine treatment reversed this situation; about 18% in the PL and 50% in the CL chromosomes.
Discussion
Mechanisms of SCEs Results obtained in this study seem to suggest that the SCE can be derived through at least two different pathways, the initial steps of which might be those as shown in Figure 1, II and III respectively. Processes involved in these pathways are highly speculative and meant to be taken as a working hypothesis.
188
H. Kato
The first pathway (Fig. 1, II) needs only one FL-induced single-strand break for its initiation. Strand breaks utilized in this pathway might be either those induced in the pre-replicational DNA region near the replicating point and approached by the fork before being closed by repair synthesis, or those induced directly at the replicating point. The exact molecular process that would occur at this region remains to be elucidated. Nevertheless, one of the possibilities may be that a free end of a single-strand break would somehow accomplish exchange with the discontinuously replicating DNA strand probably with the aid of the machinery of DNA replication. Temporal detachment of histones from the replicating point may facilitate such an interaction. Thus, strictly, this pathway would require two breaks for its initiation, one being induced by FL and the other occurring spontaneously, both residing in co-parallel strands. The results as shown in Figure 2 indicate that the majority of FLinduced SCEs arise through this pathway irrespective of whether chromosomes are unifilarly-substituted or trifilarly-substituted with BrdU. Other circumstantial evidence for this pathway is as follows: 1) The site of SCEs detectable in a metaphase chromosome coincides sharply with the chromosomal region which has been actively engaging in DNA synthesis at the time of FL illumination (Kato, 1974d). 2) Treatment of Chinese hamster cells with fluorodeoxyuridine (FdU) reduces the rate of DNA replication to a small percentage of the control while it seems to cause accumulation of replicating replicons during the treatment (Amaldi etal. 1972). In fact, in spite of the extremely low rate of replication, the number of cells that engage in replication after 6 h- FdU treatment reaches about 1.5 times that in the untreated control cell population. FL illumination of cells at this stage results in the SCE induction at about 1.5 times higher frequency than that obtained in cells untreated with FrdU, suggesting an intimate correlation between the site of SCEs and the replicating point (Kato, in preparation). The second type of pathway (Fig. 1, III a or b) would become available only when normal repair of FL-induced DNA damage is disturbed somehow, and would function only in the post-replicational DNA regions, probably in a similar fashion to that assumed in models of meiosis or gene conversion (Holliday, 1964; Whitehouse, 1963). It requires the presence of two single-strand breaks of staggered or juxtaposition in two sister chromatids, and eventual displacement of a free end of a single strand at the break point, thereby enabling the interaction of two DNA molecules. As described above the event that would follow the initial steps in both pathways is considered to be a generation of a heteroduplex (or a Holliday structure). However, this structure might soon be transformed into a doublestrand exdhange after cleavage of the non-crossing strands by a endonucleolytic attack (Sobell, 1973), since there is strong evidence against the SCE to be a single-strand exchange (Taylor, 1958; Wolff and Perry, 1975). Whether these two types of pathways occur in general in the SCE formation is yet uncertain. Inefficiency of X-rays to induce SCEs (Marin and Prescott, 1964; Gatti et al., 1974; Perry and Evans, 1975) may be attributed to an extensive operation of the first pathway by taking into consideration the short-lived nature of X-ray-induced DNA damage (Painter and Young, 1972) and the site specificity
Mechanism of SCEs and Their Relationto ChromosomalAberrations
189
of this pathway. Spontaneous SCEs may also arise from this pathway. When the replication fork reaches the site of spontaneous DNA damage, e.g., apurinic sites (Lindahl and Andersson, 1972; Lindahl and Nyberg, 1972; Verly et al., 1973), it may incorporate the lesion into an exchange process either by mistake or as a way of rescueing the damage (Kato, 1974a). On the other hand, ultraviolet-induced SCEs may arise from the second pathway. It has been known that the presence of pyrimidine dimers causes upon DNA replication a formation of gaps in the newly formed strand opposite the dimers (Rupp and HowardFlanders, 1968; Rupp et al., 1971; Lehmann, 1972). In mammalian cells these gaps, averaging 1000 nucleotide long, supposedly exist for some time and are somehow filled in by de novo DNA synthesis (Lehmann, 1972; Buhl et al., 1972; Fujiwara, 1972). It seems plausible that before a gap is closed, a free end at the gap be denatured to interact with a similarly displaced free end in the sister molecule.
Relation of SCEs to Chromatid Aberrations
To interpret results presented in Table 3, it seems possible to draw the following picture : There might exist two distinct types of chromatid deletions, one arising from the partial failure in the SCE event thus having a switch of label at the break point, and the other being derived from unrejoined strand breaks thus lacking in an associated exchange of sister chromatids. In the PL chromosomes, about a half of chromatid deletions would result from incomplete SCEs initiated at the replicating points and the rest being derived from unrejoined strand breaks induced in the duplicated regions. In the CL chromosomes, although chromatid deletions having label switch at break points would arise in a similar manner as in the PL chromosomes, a frequential production of SCE-independent aberrations in the duplicated DNA regions would lower the relative frequency of deletions with associated SCEs. On the other hand, caffeine posttreatment enhances the initiation of the SCE formation in the duplicated regions in the CL chromosomes so enhancing the production of aberrations due to incomplete SCEs. The relative frequency of such deletions is thus expected to be much higher than that obtained without caffeine treatment. In the PL chromosomes, caffeine interferes with the SCE formation (Table 1), thereby causing a decrease in the number of the SCE-associated aberrations. Caffeine may also inhibit repair of breaks in the post-replicational DNA regions (Table 2) so that the production of aberrations independent on the SCE process would be enhanced. Thus, the relative frequency of chromatid deletions associated with SCEs is expected to be lower than that obtained without caffeine treatment. FL illumination at G2 cannot induce SCEs (Table 1), so that chromatid deletions produced are without associated SCEs indicating that they arise from unrejoined strand breaks. Previous workers have examined the presence or absence of SCEs at break points of X-ray-induced chromatid breaks, and have reached a conclusion that many of chromatid deletions are simple deletions arising from single lesions and some are incomplete exchange between sister chromatids (Heddle and Body-
190 c o t e , 1 9 6 9 ; H e d d l e et al., 1970; W o l f f a n d B o d y c o t e ,
H. Kato 1975). R e s u l t s o b t a i n e d
in the present study seem to be compatible with their findings.
References Amaldi, F., Carnevali, F., Leoni, L., Mariotti, D.: Replicon origins in Chinese hamster cell DNA. I. Labeling procedure and preliminary observations. Expl. Cell Res. 74, 367-374 (1972) Ben-Hur, E., Elkind, M.M.: Damage and repair of DNA in 5-bromodeoxyuridine-labeled Chinese hamster cells exposed to fluorescent light. Biophys. J. 12, 636-647 (1972) Buhl, S.N., Setlow, R.B., Regan, J.D. : Steps in DNA chain elongation and joining after ultra-violet irradiation of human cells. Int. J. Radiat. Biol. 22, 417-424 (1972) Fujiwara, Y.: Characteristics of DNA synthesis following ultraviolet light irradiation in mouse L cells. Postreplication repair. Exp. Cell Res. 75, 483-489 (1972) Gatti, M., Pimpinelli, S., Olivieri, G.: The frequency and distribution of isolabelling in Chinese hamster chromosomes after exposure to X-rays. Mutation Res. 23, 229-238 (1974) Heddle, J.A., Bodycote, D.J.: On the formation of chromosomal aberrations. Mutation Res. 9, 117 126 (1970) Heddle, J.A., Whissell, D,, Bodycote, D.J. : Changes in chromosome structure induced by radiations: a test of the two chief hypotheses. Nature (Lond.) 221, 1158-1160 (1969) Holliday, R.: A mechanism for gene conversion in fungi. Geuet. Res. (Camb.) 5, 282 304 (1964) Ikushima, T., Wolff, S. : Sister chromatid exchanges induced by light flashes to 5-bromodeoxyuridineand 5-iodedeoxyuridine-substituted Chinese hamster chromosomes. Exp, Cell Res. 87, 15-19 (1974) Kato, H.: Induction of sister chromatid exchanges by UV light and its inhibition by caffeine. Exp. Cell Res. 82, 383-390 (1973) Kato, H. : Photoreactivation of sister chromatid exchanges induced by ultraviolet irradiation. Nature (Lond 0 249, 552-553 (1974a) Kato, H. : Induction of sister chromatid exchanges by chemical mutagens and its possible relevance to DNA repair. Exp. Cell Res. 85, 239-247 (1974b) Kato, H. : Spontaneous sister chromatid exchanges detected by a BUdR-labelling method. Nature (Lond,) 251, 70-72 (1974c) Kato, H.: Possible role of DNA synthesis in formation of sister chromatid exchanges. Nature (Loud.) 252, 739-741 (1974d) Kihlman, B.A.: Sister chromatid exchanges in Vicia faba. II. Effects of thiotepa, caffeine and 8-ethoxycaffeine on the frequency of SCEs. Chromosome (Berl.) 51, 11 18 (1975) Korenberg, J.R., Freedlender, E.F.: Giemsa technique for the detection of sister chromatid exo changes. Chromosoma (Berl.) 48, 355-360 (1974) Kuebbing, D., Werner, R.: A model for compartmentation of de novo and salvage thymidine nucleotide pools in mammalian cells. Proc. nat. Acad. Sci. (Wash.) 72, 3333 3336 (1975) Latt, S.A.: Microfluorometric detection of deoxyribonucleic acid replication in human metaphase chromosomes. Proc. nat. Acad. Sci. (Wash.) 70, 3395-3399 (1973) Latt, S.A. : Sister chromatid exchanges, indices of human chromosome damage and repair: Detection by fluorescence and induction by mitomycin C. Proc. nat. Acad. Sci. (Wash.) 71, 3162-3166 (1974) Lehmann, A.R. : Postreplication repair of DNA in ultraviolet-irradiated mammalian cells. J. molec. Biol. 66, 319-337 (1972) Lindahl, T., Andersson, A. : Rate of chain breakage at apurinic sites in double-stranded deoxyribonucleic acid. Biochemistry 11, 3618 3623 (1972) Lindahl, T., Nyberg, B.: Rate of depurination of nativ deoxyribonucleic acid. Biochemistry 11, 3610-3618 (1972) Marin, G. Prescott, D M . : The frequency of sister chromatid exchanges following exposure to varying doses of H3-thymidine or X-rays. J. Cell Biol. 21, 159-167 (1964) Meselson, M.S., Radding, C.M.: A general model for genetic recombination. Proc. nat. Acad. Sci. (Wash.) 72, 358-361 (1975) Painter, R.B., Young, B.R.: Repair replication in mammalian cells .after X-irradiation. Mutation Res. 14, 225-235 (1972)
Mechanism of SCEs and Their Relation to Chromosomal Aberrations
191
Perry, P., Evans, H.J. : Cytological detection of mutagen-carcinogen exposure by sister chromatid exchange. Nature (Lond.) 258, 121 125 (1975) Perry, P., Wolff, S. : New Giemsa method for the differential staining of sister chromatids. Nature (Lond.) 251, 156-158 (1974) Rommelaere, J., Suskind, M., Errera, M.: Chromosome and chromatid exchanges in Chinese hamster cells. Chromosoma (Berl.) 41, 243 257 (1973) Rupp, W.D., Howard-Flanders, P. : Discontinuities in the DNA synthesized in an excision-defective strain of Escherichia coli following ultraviolet irradiation. J. molec. Biol. 31, 291-304 (1968) Rupp, W.D., Wilde III, C.E., Reno, D.L., Howard-Flanders, P. : Exchanges between DNA strands in ultraviolet-irradiated Escherichia coli. J. molec. Biol. 61, 25M4 (1971) Smets, L.A., Cornelis, J.J.: Repairable and irrepairable damage in 5-bromouracil-substituted DNA exposed to ultra-violet radiation. Int. J. Radiat. Biol. 19, 445 457 (i971) Smith, K.C., Hanawalt, P.C.: Molecular photobiology. New York and London: Academic Press 1969 Sobell, H.M.: Symmetry in protein-nucleic acid interaction and its genetic implications. Advanc. Genet. 17, 411o490 (1973) Taylor, J.H.: Sister chromatid exchanges in tritum-labeled chromosomes. Genetics 43, 515-529 (1958) Verly, W.G., Paquette, Y., Thibodeau, L.: Nuclease for DNA apurinic sites may be involved in the maintenance of DNA in normal cells. Nature (Lond.) New Biol. 244, 67-69 (1973) Whitehouse, H.L.K.: A theory of crossing-over by means of hybrid deoxyribonucleic acid. Nature (Lond.) 199, 1034-1040 (1963) Wolff, S., Bodycote, J.: The induction of chromatid deletions in accord with the breakage-andreunion hypothesis. Mutation Res. 29, 85-91 (1975) Wolff, S., Bodycote, J., Painter, R.B.: Sister chromatid exchanges induced in Chinese hamster cells by UV irradiation of different stages of the cell cycle: The necessity for cells to pass through S. Mutation Res. 25, 73-81 (1974) Wolff, S., Bodycote, J., Thomas, G.H., Cleaver, J.E.: Sister chromatid exchange in xeroderma pigmentosum cells that are defective in DNA excision repair or post-replication repair. Genetics 81, 349-355 (I975) Wolff, S., Perry, P.: Differential Giemsa staining of sister chromatids and the ~study of sister chromatid exchanges without autoradiography. Chromosoma (Berl.) 48, 341-353 (1974) Wolff, S., Perry, P.: Insights on chromosome structure from sister chromatid exchange ratios and the lack of both isolabelling and heterolabelling as determined by the FPG technique. Exp. Ceil Res. 93, 23-30 (1975) Wolff, S., Scott, D. : Repair of radiation-induced damage to chromosomes. Independence of known DNA dark repair mechanisms. Exp. Cell Res. 55, 9 16 (1969)
Received February 9 November 1, 1976 / Accepted March 4, 1976 by J.H. Taylor Ready for press November 1, 1976
