- Email: [email protected]
Evidence that most radiation-induced HPRT mutants are generated directly by the initial radiation exposure
Mutation Research 426 Ž1999. 23–30
Evidence that most radiation-induced HPRT mutants are generated directly by the initial radiation exposure Edith A. Leonhardt ) , Maxine Trinh, Kenneth Chu, William C. Dewey Radiation Oncology Research Laboratory, UniÕersity of California San Francisco, 1855 Folsom St., MCB-200, San Francisco, CA 94103, USA Received 16 November 1998; received in revised form 1 March 1999; accepted 3 March 1999
Abstract Radiation-induced HPRT mutants are generally assumed to arise directly from DNA damage that is misrepaired within a few hours after X-irradiation. However, there is the possibility that mutations result indirectly from radiation-induced genomic instability that may occur several days after the initial radiation exposure. The protocols that commonly employ a 5–7 day expression period to allow for expression of the mutant phenotype prior to replating for selection of mutants would not be able to discriminate between mutants that occurred initially and those that arose during or after the expression period. To address this question, we performed a fluctuation analysis in which synchronous or asynchronous populations of human bladder carcinoma cells were treated with single doses of X-irradiation. For comparison, radiation was delivered during the expression period, either from an initial dose of 1.0 Gy followed by two 1.0 Gy doses separated by 24 h or from disintegrations resulting from I 125dU incorporated into DNA. The mutation frequency observed at the time of replating was used to calculate the average number of mutants in the initial irradiated culture by assuming that the mutants were induced directly at the time of irradiation. Then, this average number was used to calculate the fraction of the irradiated cultures that would be predicted by a Poisson distribution to have zero mutants. There was reasonably good agreement between the predicted poisson distribution and the observed distribution for the cultures that received single doses. Moreover, as expected, when cultures were irradiated during the expression period, the fraction of the cultures having zero mutants was significantly less than that predicted by a Poisson distribution. These results indicate that most radiation-induced HPRT mutations are induced directly by the initial DNA damage, and are not the result of radiation-induced instability during the 5–7 day expression period. q 1999 Elsevier Science B.V. All rights reserved. Keywords: hprt; Mutation; X-irradiation
1. Introduction The induction of mutations at the HPRT locus has been studied extensively because HPRT mutants
) Corresponding author. Tel.: q1-415-476-1076; Fax: q1-415476-9069; E-mail: [email protected]
lacking hypoxanthine phosphoribosyl transferase ŽHPRTase. can be selected easily using the drug 6-Thioguanine Ž6-TG.. If the wild-type HPRT protein is not present, the drug 6-TG is not incorporated into the DNA, and thus the mutant cell survives in the presence of the toxic drug. However, if the cell is wild type for the HPRT locus, 6-TG is incorporated into the DNA, and the wild-type cell dies w1–4x. On
0027-5107r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 2 7 - 5 1 0 7 Ž 9 9 . 0 0 0 8 0 - 9
24
E.A. Leonhardt et al.r Mutation Research 426 (1999) 23–30
the other hand, if the cells are plated into HAT medium, the wild-type cells survive, while the mutants at the HPRT locus do not. This selection against the HPRT mutants occurs because aminopterin has blocked endogenous purine synthesis, and without HPRTase, exogenous hypoxanthine cannot be utilized w3–7x. In most experimental protocols studying mutations at the HPRT locus, cells are first treated with a mutagen and then incubated for several days. This period of incubation is termed the expression period. The theory behind the expression period is that at the point of mutagenesis, wild type HPRT protein still exists in the cell, thus making it phenotypically wild type for the hprt gene, although it is genotypically an HPRT mutant w8,9x. Therefore, a period of time must elapse between the time that the mutagenic event occurs and the time when selection for mutants begins. This period of time allows the cells to degrade the wild type HPRT protein that remains. Many studies have shown the need for an expression period by demonstrating that the frequency of mutants selected increases as the duration of the expression period increases, with a maximum frequency usually after 3 days w9–12x. For EJ30 cells we determined that ; 4 days were required for the EJ30 cells to reach a constant mutant frequency w13x. There is a possibility, however, that the increase in mutants observed as the expression period is lengthened could be due at least in part to mutations induced during the expression period. In fact, treatment of irradiated cultures with HAT medium during the expression period to kill off any mutants that were induced directly at the time of irradiation, led the investigators to conclude that as many as 50% of the HPRT mutants were induced indirectly during andror after the expression period w14x. Furthermore, delayed HPRT mutations have been selected several weeks after irradiation w15x, and are molecularly distinct from those selected about a week after irradiation. These delayed mutations are thought to result from genomic instability caused by the initial radiation damage. Thus, if a considerable amount of genomic instability were occurring during the expression period, a significant fraction of the mutants observed after the expression period might not be directly induced, but instead might be due to genomic instability.
In the present study, we attempted to test the hypothesis that a significant fraction of the radiation-induced HPRT mutants are delayed mutations induced during the expression period. This was done by carrying out a fluctuation analysis to determine if the fraction of the irradiated cultures not having any mutants could be described by a Poisson distribution. If the mutants were induced directly by the initial radiation insult, the observed fraction with no mutants should correspond to the fraction calculated by a Poisson distribution to have no mutants. If, on the other hand, most of the mutants were induced during the expression period, the observed fraction with no mutants should be significantly less than the fraction calculated by a Poisson distribution w5,16x.
2. Methods and materials 2.1. Cell lines We used a human male bladder carcinoma cell line, EJ30-15, which was subcloned from the original cell line EJ30 Žalso known as MGH-U1. w13,17x. Cells were cultured without nucleosides at 378C with 5% CO 2 in alpha MEM medium ŽGibco., containing 4.5% bovine calf serum ŽHyClone Laboratories., 0.5% fetal bovine serum ŽGemini Bio Products., 2 mM glutamine, and the following antibiotics: neomycin Ž0.1 grl., streptomycin sulfate Ž0.1 grl., and potassium penicillin G Ž0.07 grl.. 2.2. Cell synchrony, cell plating and irradiation for mutant selection Synchronous mitotic cells Ž88–98% in or near mitosis. were obtained and irradiated in a previous study w13x. Briefly, for mutant selection, the synchronous mitotic cells were plated into 10–50 T75 flasks per group at a density of 1.1 = 10 5 cellsrflask Ž1.5 = 10 3 cellsrcm2 . for the unirradiated cells and 1.6 = 10 5 Ž2.2 = 10 3 cellsrcm2 . cellsrflask for those to be irradiated. Then, the cells were incubated at 378C either for 3 h Žin G1 . or 14 h Žin S phase. before they were irradiated Ž4 Gyrmin.. To determine initial plating efficiencies, with and without irradiation, 50–200 synchronous cells were plated
E.A. Leonhardt et al.r Mutation Research 426 (1999) 23–30
into T25 flasks and incubated for 10 days to allow for colony formation. The initial plating efficiencies were 35.1 " 7.7% and 6.1 " 1.9% for 3 and 6 Gy, respectively, during G1 and 35.3 " 6.2% and 8.8 " 2.3% for 3 and 6 Gy, respectively, during S w13x. For mutant selection, the T-75 flasks plated above were incubated at 378C for 5–7 days to allow the mutant HPRT phenotype to be expressed. Due to the large number of individual flasks processed concurrently w13x, the cells were harvested over a period of 3 days. Unirradiated cells were harvested 5 days after the synchronous cells had been plated. The cells irradiated in G1 which had a division delay of 4–7 h w18x, were harvested after 6 days, and the cells irradiated in S phase, which had a division delay of 7–22 h w18x were harvested after 7 days. 2.3. Plating and irradiating asynchronous cells for mutant selection Asynchronous cells were plated into 20 T75 flasks per treatment group, with the number of cells plated into each flask for mutant selection ranging from 1.5 = 10 4 to 1 = 10 5. After the initial plating, the cells were incubated for 14 h and then irradiated either with a single dose Ž0, 1.5, 3, or 6 Gy., or with an initial 1.0-Gy dose followed 24 h later by two more 1.0-Gy doses separated by 24 h. For one group, three 2.5-Gy doses were used. The initial plating efficiencies, determined as described above, were 71.5%, 38.0%, or 11.7% for single doses of 1.5, 3, or 6 Gy, respectively, 54.5% for three 1.0-Gy doses, and 23.7% for three 2.5-Gy doses. For mutant selection, the unirradiated groups Ž1 = 10 5 cellsrT75. were incubated for 5 days after plating; the irradiated groups with 1 = 10 5 or 3 = 10 4 cells were incubated for 6 days; and the groups with 1.5 = 10 4 cells for 7 days. After these expression periods Ž3.5 or 4.5 days after the last dose., the cells were trypsinized from the flasks, and the harvested cells were plated for mutant selection as described below. 2.4. Inducing mutations during the expression period with I 12 5dU incorporated into the DNA Synchronous mitotic cells plated into 2 T75 flasks Ž1.8 = 10 6 cellsrflask. were treated with 4 mgrml
25
of aphidicolin for 22 h at 378C to arrest the cells at the G1rS border. After a 22-h incubation at 378C, the cells were washed with Hanks balanced salt solution ŽHBSS. two times followed by one wash with fresh medium to remove the aphidicolin. Then, the medium was replaced with drug free alpha MEM medium. Next, I 125 dU was added at a concentration of 0.014 mCirml Ž2000 CirmM., and the cells were incubated at 378C for 1.5 h. After the 1.5-h incubation, the cells were washed three times with 2.5 ml of cold alpha MEM medium containing 10 mgrml unlabeled thymidine. The cells were then incubated for 15 min at 378C. Next, the cells were trypsinized from the flasks, and 1 = 10 5 cells were plated into each of 18 T75 flasks. Also, 2 = 10 5 cells were used for gamma scintillation counting and autoradiography; 85% of the cells were labeled with 0.016 disintegrationsrmin per labeled cell, and ; 96% of the radioactivity was in the acid insoluble fraction Ždata not shown.. These cells, pulse-labeled with I 125 dU during early S phase when the HPRT gene was replicating w19,20x, were incubated for 7 days. During this 7-day expression period, HPRT mutations were induced from DNA double strand breaks ŽDSBs. that resulted from I 125 disintegrations that occurred in the chromosomal domain containing the hprt gene. About 23 DSBs should have resulted from the estimated 23 disintegrations that occurred in the DNA during the 7 days. These DSBs did not reduce the plating efficiency below the value of 0.60 observed for unirradiated cells. For mutant selection, cells were trypsinized from the 18 individual flasks after the 7-day incubation period, and replated as described below, except with 3.3 = 10 5 cellsrT75. 2.5. Selection of HPRT mutants with 6-TG For each of the 10–50 individual T75 flasks in a treatment group, the number of cells harvested after the 5–7 day expression period was determined by Coulter counting ŽCoulter Electronics.. Then, the cells from each T75 flask were replated into four separate T75 flasks Žreferred to as a set., with 1 = 10 5 cellsrflask Ž1.45 = 10 3 cellsrcm2 . w13x. The cells were allowed to attach at 378C for 2 h before 6-TG ŽARCOS. was added at a concentration of 10 mgrml Ž60 mM. to select for HPRT mutants. After 3 days,
26
E.A. Leonhardt et al.r Mutation Research 426 (1999) 23–30
the medium containing toxic dead cells was removed from the flasks and replaced with fresh medium containing 10 mgrml 6-TG and 1.25 mgrml Fungizone. Care was taken during the change of medium to not disturb the growing colonies. Most importantly, the critical density of 1 = 10 5 or 4.4 = 10 3 cellsrcm2 w13x was not exceeded either during the expression period or after replating, respectively, because metabolic cooperation would occur between wild-type and mutant cells above these critical densities. In all cases, the cells treated with 6-TG were incubated for 3.5 weeks before the number of colonies were counted for each set of four flasks. To determine the plating efficiencies, cells from each treatment group also were replated into T25 flasks without 6-TG. The average plating efficiencies for three individual X-irradiated synchronous experiments were: 54.1% for unirradiated cells, 30.5% for 3 Gy during G1 , 7.0% for 6 Gy during G1 , 39.7% for 3 Gy during S, and 36.7% for 6 Gy during S. For the asynchronous cells, the plating efficiencies at the time of replating were 60.3%, 36.9%, or 26.1% for single doses of 1.5, 3, or 6 Gy, respectively, 38.7% for three 1.0-Gy doses, and 22.3% for three 2.5-Gy doses. From these data, the frequency of mutants per viable cell at the time of replating was determined for each group of flasks. Also, the fraction of the cultures that did not contain one or more mutants per culture was determined. A culture was scored as having zero mutants if the set of four flasks plated from the initial culture did not contain a 6-TG resistant colony.
3. Results From our experiments, we determined that most radiation-induced HPRT mutants were produced directly at the time of irradiation, with very few if any produced indirectly during or after the expression period. The rationale of the experiments was to determine if the distribution of HPRT mutants induced by irradiating replicate cultures could be described by a Poisson distribution w16x. If this were the case, the fraction of the cultures having zero mutants would be equal to eyx , where x is the mean number of mutants per culture induced at the time of irradiation. On the other hand, if the mutants were
produced indirectly during the expression period, the fraction of the cultures having zero mutants would be significantly less than eyx , or conversely, the fraction having G 1 mutant would be greater than 1 y eyx . This would occur because as the number of cells in a culture increases during the expression period, the probability of a mutant appearing in the culture would also increase if the mutants were produced during the expression period. In Fig. 1, the fraction of cultures which did not contain any mutants was plotted against the frequency of mutantsrinitial culture Ži.e., x .. The re-
Fig. 1. The frequency of flasks containing zero mutants was approximated by a Poison distribution Žopen symbols. when the cells were irradiated once immediately before the expression period. However, if the irradiation was delivered during the expression period, the frequency of flasks containing zero mutants did not fit a Poison distribution Žsolid symbols.. The solid line Ž . represents the calculated Poisson distribution Žsee text.. The unirradiated cell populations Ž0 Gy. are designated by Žcircle with diagonal slash from left to right. for the synchronous populations and by Žtriangle with diagonal slash from left to right. for the asynchronous populations. The open symbols represent cell populations which had been irradiated with a single dose of irradiation: synchronous populations treated in G1 with 3 or 6 Gy Ž` or right half of a circle, respectively., synchronous populations treated in S phase with 3 or 6 Gy ŽI or left half of a circle, respectively., and asynchronous cell populations treated with 1.5, 3 or 6 Gy Ž\, ^, or e, respectively.. The solid symbols represent cell populations which received radiation doses during the expression period; these populations were either labeled with I 125 dU Žsix-pointed star. or irradiated with three equal X-ray doses of 1.0 or 2.5 Gy delivered 24 h apart Ž' or %, respectively..
E.A. Leonhardt et al.r Mutation Research 426 (1999) 23–30
sults demonstrate that when the cells were irradiated before the expression period Žopen symbols., the
27
fraction of the cultures in each irradiated group that had zero mutants clustered around the theoretical
Table 1 Data used to generate Fig. 1 Cell populations and treatments Žinitial cells plated= 10 4 . a
Mean no. of mutantsrculture for those with G 1b
Mutant frequency =10y5 and Ž x . c
Ratio of variance to the mean for number of mutants, with standard deviationsd
Fraction of cultures with zero mutants
Fraction of cell population replated into the sets of four flasks
Synchronous cells 0 Gy Ž11–5.3. 0 Gy Ž11–5.8. 0 Gy Ž11–3.9. 0 Gy Ž11–2.9. 0 Gy Ž11–6.9. 3 Gy G1 Ž16–3.3. 3 Gy G1 Ž16–4.0. 3 Gy G1 Ž16–7.8. 3 Gy S Ž16–4.3. 3 Gy S Ž16–3.6. 3 Gy S Ž16–7.8. 6 Gy G1 Ž16–1.0. 6 Gy G1 Ž16–0.5. 6 Gy G1 Ž16–1.7. 6 Gy S Ž16–1.4. 6 Gy S Ž16–0.8. 6 Gy S Ž16–2.4.
3.7 " 3.8 1.5 " 1.0 3.2 " 2.5 3.8 " 3.4 4.5 " 0.7 7.7 " 5.2 12.4 " 8.9 12.7 " 6.2 5.4 " 6.6 7.0 " 8.1 3.9 " 4.9 6.1 " 4.6 7.9 " 4.4 6.1 " 4.6 14.4 " 13.1 25.4 " 24.1 9.5 " 10.2
0.24 Ž0.12. 0.18 Ž0.10. 0.17 Ž0.07. 0.17 Ž0.05. 0.11 Ž0.08. 7.2 Ž2.35. 15.1 Ž6.00. 6.4 Ž5.09. 2.4 Ž1.00. 3.0 Ž1.08. 1.1 Ž0.87. 8.2 Ž0.81. 31.5 Ž1.43. 11.9 Ž2.00. 2.9 Ž0.42. 10.7 Ž0.83. 2.8 Ž0.68.
1.9 " 1.8 1.1 " 0.1 1.6 " 1.3 1.1 " 0.1 0.7 " 0.04 1.3 " 1.3 1.6 " 0.9 1.0 " 0.7 1.3 " 0.5 1.5 " 0.8 1.1 " 0.8 1.0 " 0.3 1.3 " 0.8 1.1 " 0.8 1.4 " 1.6 0.7 " 0.3 1.0 " 0.6
0.81 0.75 0.88 0.88 0.95 0 0 0 0.74 0.36 0.19 0.63 0.45 0.08 0.75 0.8 0.33
0.09 0.08 0.23 0.19 0.15 0.11 0.18 0.08 0.14 0.09 0.08 0.80 0.43 0.43 0.36 0.37 0.13
Asynchronous cells 0 Gy Ž10–7.2. 0 Gy Ž10–6.9. 0 Gy Ž10–8.2. 1.5 Gy Ž1.5–1.1. 1.5 Gy Ž3.0–2.2. 3.0 Gy Ž1.5–0.6. 3.0 Gy Ž3.0–1.1. 6.0 Gy Ž10–1.2. 3 = 1 Gy Ž1.5–0.8. 3 = 1 Gy Ž1.5–0.8. 3 = 1 Gy Ž3.0–1.6. 3 = 1 Gy Ž3.0–1.7. 3 = 2.5 Gy Ž10–2.4. I 125 Ž10–6.4.
2.3 " 2.5 7.0 " 8.5 3.5 " 3.5 8.8 " 9.4 6.9 " 5.4 12.1 " 12.3 24.2 " 24.5 38.1 " 37.8 7.8 " 13.8 8.6 " 9.0 5.6 " 5.7 6.7 " 5.3 16.4 " 11.8 3.4 " 3.3U
0.14 Ž0.10. 0.24 Ž0.17. 0.14 Ž0.11. 0.92 Ž0.10. 1.6 Ž0.35. 3.0 Ž0.17. 11.6 Ž1.33. 34.6 Ž4.05. 2.8 Ž0.22. 3.3 Ž0.28. 2.2 Ž0.36. 4.2 Ž0.71. 18.4 Ž4.36. 1.5 Ž0.85.
0.8 " 0.4 2.8 " 2.5 0.6 " 0.6 0.8 " 0.3 1.1 " 0.7 1.3 " 1.3 1.7 " 1.6 1.4 " 1.5 1.5 " 0.8 1.0 " 0.4 1.5 " 0.8 1.4 " 0.8 1.0 " .7 0.9 " 0.6
0.8 0.9 0.9 0.75 0.45 0.6 0.35 0.05 0.4 0.3 0.45 0.15 0 0.06
0.04 0.08 0.10 0.34 0.30 0.39 0.38 0.28 0.31 0.34 0.30 0.29 0.15 0.16
a
For the synchronous cells, the point in the cell cycle when the cells were irradiated is noted along with the single dose of X-rays delivered. For the asynchronous cells, the single or multiple doses of radiation are listed. For both synchronous and asynchronous populations, the initial number of cells plated = 10 4 is shown in parentheses, with the second number representing the number of clonogenic cells. b The average number of mutantsrculture was calculated by using only the sets of four flasks that had at least one 6-TG resistant colony. standard deviations are indicated. The values for single doses of irradiation are generally larger than the values for multiple doses and I 125 ŽU the values were divided by 3.3 since 3.3 times as many cells were replated in this group., when mutantsrculture are plotted vs. mutant frequencies. c Mutants frequencies per viable cell determined for each group at the time of replating are tabulated. The values of x, i.e., the frequency of mutantsrinitial culture are shown in parentheses. d In the cultures which had mutants, the ratio was determined by obtaining the variance for the number of mutants in each set of four flasks and dividing that number by the mean. These numbers were then averaged to determine the ratio of the variance to the mean for all the sets for each group. Standard deviations are indicated.
E.A. Leonhardt et al.r Mutation Research 426 (1999) 23–30
28
line Žsolid. predicted by the Poisson distribution calculated as described above Žregression coefficient of 0.85.. The value for x was calculated by the equations shown below: x s Ž mutantsrinitial clonogenic cell . = Ž clonogenic cellsrinitial culture. . We assumed that Žmutantsrinitial clonogenic cell. s Žmutantsrclonogenic cells at replating.. In other words, we assumed that wild-type and mutant cells multiplied at the same rate during the expression period Ždata not shown and observed for HPRT and TK mutants by Jacobson and Morgan, w21x.. 1 Then, x s Žmutantsrclonogenic cell at replating. =Žclonogenic cellsrinitial culture.. See Table 1 for a listing of these values. The reasonably good agreement between the expected and observed fractions of cultures with zero mutants suggests that the HPRT mutants were induced directly, i.e., that they were present at the time the cultures were plated for expression of the mutant phenotype. Furthermore, these conclusions can be reached for cells irradiated in G1 or S phase, or as asynchronous cultures. However, when mutants were induced during the expression period by having I 125 dU incorporated into the DNA Žsix-pointed star., only 0.05 of the cultures had zero mutants, compared with 0.42 that would be predicted by a Poisson distribution. Also, when asynchronous cells were irradiated with 1.0 Gy initially followed 24 h later by two 1.0-Gy doses during the expression period, the fractions of cultures with zero mutants Žsolid symbols. were statistically lower Ž p s 0.05, least squares analysis. than the fractions of cultures with zero mutants in the populations irradiated with a single
1
Although a few HPRT mutants isolated from irradiated populations have been observed to multiply slower than wild-type cells Ždata not shown and Ref. w22x., there is no reason to believe that the few slowly multiplying HPRT mutants will be any more abundant than other slowly multiplying cells that have been isolated from irradiated populations w23x. However, if the HPRT mutants multiplied slower than the wild-type cells, i.e., if Žmutantsrinitial clonogenic cell. ) Žmutantsrclonogenic cells at replating., the data points in Fig. 1 should be shifted to the right, which would cause the points observed for the single doses to cluster above the theoretical line predicted by the Poisson distribution.
dose. If we had increased the expression period beyond 3.5 to 4.5 days after the last dose, we would not have expected any change in mutant frequency since at ; 4 days we have observed a constant mutant frequency w13x. ŽIn fact, if the mutant frequency would have increased with a longer expression time, the fraction of cultures with zero mutants would be even lower.. Finally, when the cultures, for all treatment groups, were replated after the expression period into four individual flasks for each culture, the distribution of mutants within the four flasks was not significantly different from a Poisson distribution; i.e., the variance was approximately equal to the mean number of mutants observed in the particular culture Žsee ratios in Table 1.. Thus, all of these results strongly suggest that most of the mutants were induced initially at the time of irradiation and were not a product of genomic instability during or shortly after the expression period.
4. Discussion Our conclusions that HPRT mutants are induced directly at the time of irradiation differ from those reached by Loucas and Cornforth w14x who concluded that at least 50% of the mutants are produced during or shortly after the expression period. They also studied human EJ-30 cells as we did, but they exposed the irradiated cells to HAT medium during the expression period to eliminate any HPRT mutants that would have expressed the mutant HPRT phenotype during the expression period. They concluded that most of the mutants were produced late during or after the expression period, because after the HAT medium was removed, the frequency of mutants in the cells replated into 6-TG medium was almost as great following the HAT treatment as when the HAT treatment was omitted. However, this assumes that the turnover of the HPRT protein and synthesis of mutant protein during the expression period would not be reduced appreciably by the HAT treatment. Another implicit assumption is that the intracellular pools of nucleosides that might compete with incorporation of 6-TG into DNA are not altered by the HAT treatment; there is evidence that this assumption may not be true w6x. Also, Hofer et.
E.A. Leonhardt et al.r Mutation Research 426 (1999) 23–30
al. w24x showed that treating CHO cells with HAT for 8 days had dramatic effects subsequently on dose response curves for cell killing and micronuclei that resulted from I 125 dU incorporated into DNA. Furthermore, these unexplained qualitative and quantitative effects persisted for as long as 30 days after termination of the HAT treatment. In addition, in experiments that quantify the frequencies of HPRT mutants, the cells must be maintained below a critical density. Above the critical density, metabolic cooperation between the wild-type and mutant cells will cause the observed mutant frequency to be lower than the true frequency. This critical density must not be exceeded either during the expression period or after the cells are replated into 6-TG w3,9,10x. We have determined that the critical density for EJ-30 cells is 1 = 10 5 cellsrcm2 before replating and 4.4 = 10 3 after replating, and in all of our experiments, the maximum cell densities were never exceeded. There is no mention of cell densities during the expression period in the experiments of Loucas and Cornforth w14x. Possibly, the delayed increase in mutant frequencies observed between 8 and 16 days after irradiation for cells treated with HAT, compared with the decline in mutant frequencies between 8 and 16 days for cells not treated with HAT could be caused by any or all of the factors mentioned above. Thus, to confirm the conclusions from the HAT experiments that at least 50% of HPRT mutations result from latent effects of radiation-induced damage, the possible complications mentioned above should be excluded. Also, the molecular spectra of the mutations, with and without HAT treatment, should be determined because 30– 70% of radiation-induced HPRT mutations consist of large deletions involving all or part of the hprt gene Žsee references in Ref. w13x. whereas, delayed mutations that have been observed several weeks or months after irradiation consist primarily of point mutations or small deletions w15x. In summary, our fluctuation analysis presents strong evidence that most if not all radiation-induced HPRT mutants are induced directly at the time of irradiation as the DNA damage is being repairedr misrepaired. Due to the experimental design of the fluctuation analysis, mutations resulting from the initial insult, whether manifested immediately or within a few days, would not affect the distribution.
29
This is different from damage due to genomic instability which has been detected several weeks or months after the initial insult. Furthermore, the frequency of radiation-induced HPRT mutants was the same when CHO cells were replated for mutant selection as when they were not replated Žtermed in situ. w10x. In both cases, metabolic cooperation was avoided by maintaining the cells below a critical density. If mutants had been produced during the expression period as microcolonies were forming, we would expect the fraction of microcolonies having a mutant in the in situ method to be higher than the fraction of the cells having a mutant in the replating method. The fact that the in situ and replating methods gave very similar mutant frequencies supports the conclusion reached from our fluctuation analysis that HPRT mutants are induced directly by X-irradiation.
Acknowledgements We would like to thank Barbara Docklova for her invaluable assistance on the project, and Dr. Norman Albright for his help with the statistical analysis. This research was supported in part by Research Grant CA31808 from the National Institute of Health.
References w1x J.A. Nelson, J.W. Carpenter, L.M. Rose, D.J. Adamson, Mechanisms of action of 6-thioguanine, 6-mercaptopurine, and 8-azaguanine, Cancer Res. 35 Ž1975. 2872–2878. w2x E.R. Nestmann, R.L. Brillinger, J.P.W. Gilman, C.J. Rudd, S.H.H. Swierenga, Recommended protocols based on a survey of current practice in genotoxicity testing laboratories: II. Mutation in Chinese hamster ovary, V79 Chinese hamster lung and L5178Y mouse lymphoma cells, Mutat. Res. 246 Ž1991. 255–284. w3x R. DeMars, Genetic studies of HG-PRT deficiency and the Lesch–Nyhan syndrome with cultured human cells, Fed. Proc. 30 Ž1971. 944–955. w4x B.R. Migeon, Studies of skin fibroblasts from 10 families with HGPRT deficiency, with reference to X-chromosome inactivation, Am. J. Hum. Genet. 23 Ž1971. 199–210. w5x J.H. Carver, W.C. Dewey, L.E. Hopwood, X-ray-induced mutants resistant to 8-azaguanine: II. Cell cycle dose response, Mutat. Res. 34 Ž1976. 465–480.
30
E.A. Leonhardt et al.r Mutation Research 426 (1999) 23–30
w6x V.M. Riccardi, J.W. Littlefield, Adaptation in Lesch–Nyhan cells exposed to aminopterin, Exp. Cell Res. 74 Ž1972. 417–422. w7x P.J. Benke, N. Herrick, A. Hebert, Transport of hypoxanthine in fibroblasts with normal and mutant hypoxanthine–guanine phosphoribosyltransferase, Biochem. Med. 8 Ž1973. 309–323. w8x W.G. Thilly, J.G. Deluca, H. Hoppe IV, B.W. Penman, Phenotypic lag and mutation to 6-thioguanine resistance in diploid human lymphoblasts, Mutat. Res. 50 Ž1978. 137–144. w9x E.H.Y. Chu, H.V. Malling, Mammalian cell genetics: II. Chemical induction of specific locus mutations in Chinese hamster cells in vitro, Proc. Natl. Acad. Sci. U.S.A. 61 Ž1968. 1306–1312. w10x J.H. Carver, W.C. Dewey, L.E. Hopwood, X-ray-induced mutants resistant to 8-azaguanine: I. Effects of cell density and expression time, Mutat. Res. 34 Ž1976. 447–464. w11x A.J. Grosovsky, J.B. Little, Effect of growth rate on phenotypic expression of 6-thioguanine resistance in human diploid fibroblasts, Mutat. Res. 110 Ž1983. 163–170. w12x J.P. O’Neill, A.W. Hsie, Phenotypic expression time of mutagen-induced 6-thioguanine resistance in Chinese hamster ovary cells ŽCHOrHGPRT system., Mutat. Res. 59 Ž1979. 109–118. w13x E.A. Leonhardt, M. Trinh, H.B. Forrester, R.T. Johnson, W.C. Dewey, Comparisons of the frequencies and molecular spectra of HPRT mutants when human cancer cells were X-irradiated during G1 or S phase, Radiat. Res. 148 Ž1997. 548–560. w14x B.D. Loucas, M.N. Cornforth, Postirradiation growth in HAT medium fails to eliminate the delayed appearance of 6-thioguanine-resistant clones in EJ30 human epithelial cells, Radiat. Res. 149 Ž1998. 171–178. w15x J.B. Little, H. Nagasawa, T. Pfenning, H. Vetrovs,
w16x w17x
w18x
w19x
w20x
w21x w22x
w23x
w24x
Radiation-induced genomic instability: delayed mutagenic and cytogenetic effects of X-rays and alpha particles, Radiat. Res. 148 Ž1997. 299–307. S.E. Luria, M. Delbruck, Mutation of bacteria from virus sensitivity to virus resistance, Genetics 28 Ž1943. 491–511. S. Squires, R.T. Johnson, A.R.S. Collins, Initial rates of DNA incision in UV-irradiated human cells differences between normal, xeroderma pigmentosum and tumour cells, Mutat. Res. 95 Ž1982. 389–404. E.A. Leonhardt, M. Trinh, H.B. Forrester, W.C. Dewey, Persistent decrease in viability as a function of X-irradiation of human bladder carcinoma cells in G1 or S phase, Radiat. Res. 149 Ž1998. 343–349. M. Schmidt, B.R. Migeon, Asynchronous replication of homologous loci on human active and inactive X chromosomes, Proc. Natl. Acad. Sci. U.S.A. 87 Ž1990. 3685–3689. A. Grossmann, V.M. Maher, J.J. McCormick, The frequency of mutants in Human fibroblasts UV-irradiated at various times during S-phase suggests that genes for thioguanineand diphtheria toxin-resistance are replicated early, Mutat. Res. 152 Ž1985. 67–76. B.S. Jacobson, T.L. Morgan, Growth rates of hprt and tk mutant CHO cell lines, Mutat. Res. 344 Ž1995. 141–145. S.L. Nelson, I.M. Jones, J.C. Fuscoe, K. Burkhart-Schultz, A.J. Grosovsky, Mapping the end points of large deletions affecting the hprt locus in human peripheral blood cells and cell lines, Radiat. Res. 141 Ž1995. 2–10. W.K. Sinclair, X-ray-induced heritable damage Žsmall-colony formation. in cultured mammalian cells, Radiat. Res. 21 Ž1964. 584–611. K.G. Hofer, X. Lin, S.-P. Bao, DNA damage, micronucleus formation, and cell death from 125 I decays in DNA, Acta Oncologica 35 Ž1996. 825–832.
Evidence that most radiation-induced HPRT mutants are generated directly by the initial radiation exposure Edith A. Leonhardt ) , Maxine Trinh, Kenneth Chu, William C. Dewey Radiation Oncology Research Laboratory, UniÕersity of California San Francisco, 1855 Folsom St., MCB-200, San Francisco, CA 94103, USA Received 16 November 1998; received in revised form 1 March 1999; accepted 3 March 1999
Abstract Radiation-induced HPRT mutants are generally assumed to arise directly from DNA damage that is misrepaired within a few hours after X-irradiation. However, there is the possibility that mutations result indirectly from radiation-induced genomic instability that may occur several days after the initial radiation exposure. The protocols that commonly employ a 5–7 day expression period to allow for expression of the mutant phenotype prior to replating for selection of mutants would not be able to discriminate between mutants that occurred initially and those that arose during or after the expression period. To address this question, we performed a fluctuation analysis in which synchronous or asynchronous populations of human bladder carcinoma cells were treated with single doses of X-irradiation. For comparison, radiation was delivered during the expression period, either from an initial dose of 1.0 Gy followed by two 1.0 Gy doses separated by 24 h or from disintegrations resulting from I 125dU incorporated into DNA. The mutation frequency observed at the time of replating was used to calculate the average number of mutants in the initial irradiated culture by assuming that the mutants were induced directly at the time of irradiation. Then, this average number was used to calculate the fraction of the irradiated cultures that would be predicted by a Poisson distribution to have zero mutants. There was reasonably good agreement between the predicted poisson distribution and the observed distribution for the cultures that received single doses. Moreover, as expected, when cultures were irradiated during the expression period, the fraction of the cultures having zero mutants was significantly less than that predicted by a Poisson distribution. These results indicate that most radiation-induced HPRT mutations are induced directly by the initial DNA damage, and are not the result of radiation-induced instability during the 5–7 day expression period. q 1999 Elsevier Science B.V. All rights reserved. Keywords: hprt; Mutation; X-irradiation
1. Introduction The induction of mutations at the HPRT locus has been studied extensively because HPRT mutants
) Corresponding author. Tel.: q1-415-476-1076; Fax: q1-415476-9069; E-mail: [email protected]
lacking hypoxanthine phosphoribosyl transferase ŽHPRTase. can be selected easily using the drug 6-Thioguanine Ž6-TG.. If the wild-type HPRT protein is not present, the drug 6-TG is not incorporated into the DNA, and thus the mutant cell survives in the presence of the toxic drug. However, if the cell is wild type for the HPRT locus, 6-TG is incorporated into the DNA, and the wild-type cell dies w1–4x. On
0027-5107r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 2 7 - 5 1 0 7 Ž 9 9 . 0 0 0 8 0 - 9
24
E.A. Leonhardt et al.r Mutation Research 426 (1999) 23–30
the other hand, if the cells are plated into HAT medium, the wild-type cells survive, while the mutants at the HPRT locus do not. This selection against the HPRT mutants occurs because aminopterin has blocked endogenous purine synthesis, and without HPRTase, exogenous hypoxanthine cannot be utilized w3–7x. In most experimental protocols studying mutations at the HPRT locus, cells are first treated with a mutagen and then incubated for several days. This period of incubation is termed the expression period. The theory behind the expression period is that at the point of mutagenesis, wild type HPRT protein still exists in the cell, thus making it phenotypically wild type for the hprt gene, although it is genotypically an HPRT mutant w8,9x. Therefore, a period of time must elapse between the time that the mutagenic event occurs and the time when selection for mutants begins. This period of time allows the cells to degrade the wild type HPRT protein that remains. Many studies have shown the need for an expression period by demonstrating that the frequency of mutants selected increases as the duration of the expression period increases, with a maximum frequency usually after 3 days w9–12x. For EJ30 cells we determined that ; 4 days were required for the EJ30 cells to reach a constant mutant frequency w13x. There is a possibility, however, that the increase in mutants observed as the expression period is lengthened could be due at least in part to mutations induced during the expression period. In fact, treatment of irradiated cultures with HAT medium during the expression period to kill off any mutants that were induced directly at the time of irradiation, led the investigators to conclude that as many as 50% of the HPRT mutants were induced indirectly during andror after the expression period w14x. Furthermore, delayed HPRT mutations have been selected several weeks after irradiation w15x, and are molecularly distinct from those selected about a week after irradiation. These delayed mutations are thought to result from genomic instability caused by the initial radiation damage. Thus, if a considerable amount of genomic instability were occurring during the expression period, a significant fraction of the mutants observed after the expression period might not be directly induced, but instead might be due to genomic instability.
In the present study, we attempted to test the hypothesis that a significant fraction of the radiation-induced HPRT mutants are delayed mutations induced during the expression period. This was done by carrying out a fluctuation analysis to determine if the fraction of the irradiated cultures not having any mutants could be described by a Poisson distribution. If the mutants were induced directly by the initial radiation insult, the observed fraction with no mutants should correspond to the fraction calculated by a Poisson distribution to have no mutants. If, on the other hand, most of the mutants were induced during the expression period, the observed fraction with no mutants should be significantly less than the fraction calculated by a Poisson distribution w5,16x.
2. Methods and materials 2.1. Cell lines We used a human male bladder carcinoma cell line, EJ30-15, which was subcloned from the original cell line EJ30 Žalso known as MGH-U1. w13,17x. Cells were cultured without nucleosides at 378C with 5% CO 2 in alpha MEM medium ŽGibco., containing 4.5% bovine calf serum ŽHyClone Laboratories., 0.5% fetal bovine serum ŽGemini Bio Products., 2 mM glutamine, and the following antibiotics: neomycin Ž0.1 grl., streptomycin sulfate Ž0.1 grl., and potassium penicillin G Ž0.07 grl.. 2.2. Cell synchrony, cell plating and irradiation for mutant selection Synchronous mitotic cells Ž88–98% in or near mitosis. were obtained and irradiated in a previous study w13x. Briefly, for mutant selection, the synchronous mitotic cells were plated into 10–50 T75 flasks per group at a density of 1.1 = 10 5 cellsrflask Ž1.5 = 10 3 cellsrcm2 . for the unirradiated cells and 1.6 = 10 5 Ž2.2 = 10 3 cellsrcm2 . cellsrflask for those to be irradiated. Then, the cells were incubated at 378C either for 3 h Žin G1 . or 14 h Žin S phase. before they were irradiated Ž4 Gyrmin.. To determine initial plating efficiencies, with and without irradiation, 50–200 synchronous cells were plated
E.A. Leonhardt et al.r Mutation Research 426 (1999) 23–30
into T25 flasks and incubated for 10 days to allow for colony formation. The initial plating efficiencies were 35.1 " 7.7% and 6.1 " 1.9% for 3 and 6 Gy, respectively, during G1 and 35.3 " 6.2% and 8.8 " 2.3% for 3 and 6 Gy, respectively, during S w13x. For mutant selection, the T-75 flasks plated above were incubated at 378C for 5–7 days to allow the mutant HPRT phenotype to be expressed. Due to the large number of individual flasks processed concurrently w13x, the cells were harvested over a period of 3 days. Unirradiated cells were harvested 5 days after the synchronous cells had been plated. The cells irradiated in G1 which had a division delay of 4–7 h w18x, were harvested after 6 days, and the cells irradiated in S phase, which had a division delay of 7–22 h w18x were harvested after 7 days. 2.3. Plating and irradiating asynchronous cells for mutant selection Asynchronous cells were plated into 20 T75 flasks per treatment group, with the number of cells plated into each flask for mutant selection ranging from 1.5 = 10 4 to 1 = 10 5. After the initial plating, the cells were incubated for 14 h and then irradiated either with a single dose Ž0, 1.5, 3, or 6 Gy., or with an initial 1.0-Gy dose followed 24 h later by two more 1.0-Gy doses separated by 24 h. For one group, three 2.5-Gy doses were used. The initial plating efficiencies, determined as described above, were 71.5%, 38.0%, or 11.7% for single doses of 1.5, 3, or 6 Gy, respectively, 54.5% for three 1.0-Gy doses, and 23.7% for three 2.5-Gy doses. For mutant selection, the unirradiated groups Ž1 = 10 5 cellsrT75. were incubated for 5 days after plating; the irradiated groups with 1 = 10 5 or 3 = 10 4 cells were incubated for 6 days; and the groups with 1.5 = 10 4 cells for 7 days. After these expression periods Ž3.5 or 4.5 days after the last dose., the cells were trypsinized from the flasks, and the harvested cells were plated for mutant selection as described below. 2.4. Inducing mutations during the expression period with I 12 5dU incorporated into the DNA Synchronous mitotic cells plated into 2 T75 flasks Ž1.8 = 10 6 cellsrflask. were treated with 4 mgrml
25
of aphidicolin for 22 h at 378C to arrest the cells at the G1rS border. After a 22-h incubation at 378C, the cells were washed with Hanks balanced salt solution ŽHBSS. two times followed by one wash with fresh medium to remove the aphidicolin. Then, the medium was replaced with drug free alpha MEM medium. Next, I 125 dU was added at a concentration of 0.014 mCirml Ž2000 CirmM., and the cells were incubated at 378C for 1.5 h. After the 1.5-h incubation, the cells were washed three times with 2.5 ml of cold alpha MEM medium containing 10 mgrml unlabeled thymidine. The cells were then incubated for 15 min at 378C. Next, the cells were trypsinized from the flasks, and 1 = 10 5 cells were plated into each of 18 T75 flasks. Also, 2 = 10 5 cells were used for gamma scintillation counting and autoradiography; 85% of the cells were labeled with 0.016 disintegrationsrmin per labeled cell, and ; 96% of the radioactivity was in the acid insoluble fraction Ždata not shown.. These cells, pulse-labeled with I 125 dU during early S phase when the HPRT gene was replicating w19,20x, were incubated for 7 days. During this 7-day expression period, HPRT mutations were induced from DNA double strand breaks ŽDSBs. that resulted from I 125 disintegrations that occurred in the chromosomal domain containing the hprt gene. About 23 DSBs should have resulted from the estimated 23 disintegrations that occurred in the DNA during the 7 days. These DSBs did not reduce the plating efficiency below the value of 0.60 observed for unirradiated cells. For mutant selection, cells were trypsinized from the 18 individual flasks after the 7-day incubation period, and replated as described below, except with 3.3 = 10 5 cellsrT75. 2.5. Selection of HPRT mutants with 6-TG For each of the 10–50 individual T75 flasks in a treatment group, the number of cells harvested after the 5–7 day expression period was determined by Coulter counting ŽCoulter Electronics.. Then, the cells from each T75 flask were replated into four separate T75 flasks Žreferred to as a set., with 1 = 10 5 cellsrflask Ž1.45 = 10 3 cellsrcm2 . w13x. The cells were allowed to attach at 378C for 2 h before 6-TG ŽARCOS. was added at a concentration of 10 mgrml Ž60 mM. to select for HPRT mutants. After 3 days,
26
E.A. Leonhardt et al.r Mutation Research 426 (1999) 23–30
the medium containing toxic dead cells was removed from the flasks and replaced with fresh medium containing 10 mgrml 6-TG and 1.25 mgrml Fungizone. Care was taken during the change of medium to not disturb the growing colonies. Most importantly, the critical density of 1 = 10 5 or 4.4 = 10 3 cellsrcm2 w13x was not exceeded either during the expression period or after replating, respectively, because metabolic cooperation would occur between wild-type and mutant cells above these critical densities. In all cases, the cells treated with 6-TG were incubated for 3.5 weeks before the number of colonies were counted for each set of four flasks. To determine the plating efficiencies, cells from each treatment group also were replated into T25 flasks without 6-TG. The average plating efficiencies for three individual X-irradiated synchronous experiments were: 54.1% for unirradiated cells, 30.5% for 3 Gy during G1 , 7.0% for 6 Gy during G1 , 39.7% for 3 Gy during S, and 36.7% for 6 Gy during S. For the asynchronous cells, the plating efficiencies at the time of replating were 60.3%, 36.9%, or 26.1% for single doses of 1.5, 3, or 6 Gy, respectively, 38.7% for three 1.0-Gy doses, and 22.3% for three 2.5-Gy doses. From these data, the frequency of mutants per viable cell at the time of replating was determined for each group of flasks. Also, the fraction of the cultures that did not contain one or more mutants per culture was determined. A culture was scored as having zero mutants if the set of four flasks plated from the initial culture did not contain a 6-TG resistant colony.
3. Results From our experiments, we determined that most radiation-induced HPRT mutants were produced directly at the time of irradiation, with very few if any produced indirectly during or after the expression period. The rationale of the experiments was to determine if the distribution of HPRT mutants induced by irradiating replicate cultures could be described by a Poisson distribution w16x. If this were the case, the fraction of the cultures having zero mutants would be equal to eyx , where x is the mean number of mutants per culture induced at the time of irradiation. On the other hand, if the mutants were
produced indirectly during the expression period, the fraction of the cultures having zero mutants would be significantly less than eyx , or conversely, the fraction having G 1 mutant would be greater than 1 y eyx . This would occur because as the number of cells in a culture increases during the expression period, the probability of a mutant appearing in the culture would also increase if the mutants were produced during the expression period. In Fig. 1, the fraction of cultures which did not contain any mutants was plotted against the frequency of mutantsrinitial culture Ži.e., x .. The re-
Fig. 1. The frequency of flasks containing zero mutants was approximated by a Poison distribution Žopen symbols. when the cells were irradiated once immediately before the expression period. However, if the irradiation was delivered during the expression period, the frequency of flasks containing zero mutants did not fit a Poison distribution Žsolid symbols.. The solid line Ž . represents the calculated Poisson distribution Žsee text.. The unirradiated cell populations Ž0 Gy. are designated by Žcircle with diagonal slash from left to right. for the synchronous populations and by Žtriangle with diagonal slash from left to right. for the asynchronous populations. The open symbols represent cell populations which had been irradiated with a single dose of irradiation: synchronous populations treated in G1 with 3 or 6 Gy Ž` or right half of a circle, respectively., synchronous populations treated in S phase with 3 or 6 Gy ŽI or left half of a circle, respectively., and asynchronous cell populations treated with 1.5, 3 or 6 Gy Ž\, ^, or e, respectively.. The solid symbols represent cell populations which received radiation doses during the expression period; these populations were either labeled with I 125 dU Žsix-pointed star. or irradiated with three equal X-ray doses of 1.0 or 2.5 Gy delivered 24 h apart Ž' or %, respectively..
E.A. Leonhardt et al.r Mutation Research 426 (1999) 23–30
sults demonstrate that when the cells were irradiated before the expression period Žopen symbols., the
27
fraction of the cultures in each irradiated group that had zero mutants clustered around the theoretical
Table 1 Data used to generate Fig. 1 Cell populations and treatments Žinitial cells plated= 10 4 . a
Mean no. of mutantsrculture for those with G 1b
Mutant frequency =10y5 and Ž x . c
Ratio of variance to the mean for number of mutants, with standard deviationsd
Fraction of cultures with zero mutants
Fraction of cell population replated into the sets of four flasks
Synchronous cells 0 Gy Ž11–5.3. 0 Gy Ž11–5.8. 0 Gy Ž11–3.9. 0 Gy Ž11–2.9. 0 Gy Ž11–6.9. 3 Gy G1 Ž16–3.3. 3 Gy G1 Ž16–4.0. 3 Gy G1 Ž16–7.8. 3 Gy S Ž16–4.3. 3 Gy S Ž16–3.6. 3 Gy S Ž16–7.8. 6 Gy G1 Ž16–1.0. 6 Gy G1 Ž16–0.5. 6 Gy G1 Ž16–1.7. 6 Gy S Ž16–1.4. 6 Gy S Ž16–0.8. 6 Gy S Ž16–2.4.
3.7 " 3.8 1.5 " 1.0 3.2 " 2.5 3.8 " 3.4 4.5 " 0.7 7.7 " 5.2 12.4 " 8.9 12.7 " 6.2 5.4 " 6.6 7.0 " 8.1 3.9 " 4.9 6.1 " 4.6 7.9 " 4.4 6.1 " 4.6 14.4 " 13.1 25.4 " 24.1 9.5 " 10.2
0.24 Ž0.12. 0.18 Ž0.10. 0.17 Ž0.07. 0.17 Ž0.05. 0.11 Ž0.08. 7.2 Ž2.35. 15.1 Ž6.00. 6.4 Ž5.09. 2.4 Ž1.00. 3.0 Ž1.08. 1.1 Ž0.87. 8.2 Ž0.81. 31.5 Ž1.43. 11.9 Ž2.00. 2.9 Ž0.42. 10.7 Ž0.83. 2.8 Ž0.68.
1.9 " 1.8 1.1 " 0.1 1.6 " 1.3 1.1 " 0.1 0.7 " 0.04 1.3 " 1.3 1.6 " 0.9 1.0 " 0.7 1.3 " 0.5 1.5 " 0.8 1.1 " 0.8 1.0 " 0.3 1.3 " 0.8 1.1 " 0.8 1.4 " 1.6 0.7 " 0.3 1.0 " 0.6
0.81 0.75 0.88 0.88 0.95 0 0 0 0.74 0.36 0.19 0.63 0.45 0.08 0.75 0.8 0.33
0.09 0.08 0.23 0.19 0.15 0.11 0.18 0.08 0.14 0.09 0.08 0.80 0.43 0.43 0.36 0.37 0.13
Asynchronous cells 0 Gy Ž10–7.2. 0 Gy Ž10–6.9. 0 Gy Ž10–8.2. 1.5 Gy Ž1.5–1.1. 1.5 Gy Ž3.0–2.2. 3.0 Gy Ž1.5–0.6. 3.0 Gy Ž3.0–1.1. 6.0 Gy Ž10–1.2. 3 = 1 Gy Ž1.5–0.8. 3 = 1 Gy Ž1.5–0.8. 3 = 1 Gy Ž3.0–1.6. 3 = 1 Gy Ž3.0–1.7. 3 = 2.5 Gy Ž10–2.4. I 125 Ž10–6.4.
2.3 " 2.5 7.0 " 8.5 3.5 " 3.5 8.8 " 9.4 6.9 " 5.4 12.1 " 12.3 24.2 " 24.5 38.1 " 37.8 7.8 " 13.8 8.6 " 9.0 5.6 " 5.7 6.7 " 5.3 16.4 " 11.8 3.4 " 3.3U
0.14 Ž0.10. 0.24 Ž0.17. 0.14 Ž0.11. 0.92 Ž0.10. 1.6 Ž0.35. 3.0 Ž0.17. 11.6 Ž1.33. 34.6 Ž4.05. 2.8 Ž0.22. 3.3 Ž0.28. 2.2 Ž0.36. 4.2 Ž0.71. 18.4 Ž4.36. 1.5 Ž0.85.
0.8 " 0.4 2.8 " 2.5 0.6 " 0.6 0.8 " 0.3 1.1 " 0.7 1.3 " 1.3 1.7 " 1.6 1.4 " 1.5 1.5 " 0.8 1.0 " 0.4 1.5 " 0.8 1.4 " 0.8 1.0 " .7 0.9 " 0.6
0.8 0.9 0.9 0.75 0.45 0.6 0.35 0.05 0.4 0.3 0.45 0.15 0 0.06
0.04 0.08 0.10 0.34 0.30 0.39 0.38 0.28 0.31 0.34 0.30 0.29 0.15 0.16
a
For the synchronous cells, the point in the cell cycle when the cells were irradiated is noted along with the single dose of X-rays delivered. For the asynchronous cells, the single or multiple doses of radiation are listed. For both synchronous and asynchronous populations, the initial number of cells plated = 10 4 is shown in parentheses, with the second number representing the number of clonogenic cells. b The average number of mutantsrculture was calculated by using only the sets of four flasks that had at least one 6-TG resistant colony. standard deviations are indicated. The values for single doses of irradiation are generally larger than the values for multiple doses and I 125 ŽU the values were divided by 3.3 since 3.3 times as many cells were replated in this group., when mutantsrculture are plotted vs. mutant frequencies. c Mutants frequencies per viable cell determined for each group at the time of replating are tabulated. The values of x, i.e., the frequency of mutantsrinitial culture are shown in parentheses. d In the cultures which had mutants, the ratio was determined by obtaining the variance for the number of mutants in each set of four flasks and dividing that number by the mean. These numbers were then averaged to determine the ratio of the variance to the mean for all the sets for each group. Standard deviations are indicated.
E.A. Leonhardt et al.r Mutation Research 426 (1999) 23–30
28
line Žsolid. predicted by the Poisson distribution calculated as described above Žregression coefficient of 0.85.. The value for x was calculated by the equations shown below: x s Ž mutantsrinitial clonogenic cell . = Ž clonogenic cellsrinitial culture. . We assumed that Žmutantsrinitial clonogenic cell. s Žmutantsrclonogenic cells at replating.. In other words, we assumed that wild-type and mutant cells multiplied at the same rate during the expression period Ždata not shown and observed for HPRT and TK mutants by Jacobson and Morgan, w21x.. 1 Then, x s Žmutantsrclonogenic cell at replating. =Žclonogenic cellsrinitial culture.. See Table 1 for a listing of these values. The reasonably good agreement between the expected and observed fractions of cultures with zero mutants suggests that the HPRT mutants were induced directly, i.e., that they were present at the time the cultures were plated for expression of the mutant phenotype. Furthermore, these conclusions can be reached for cells irradiated in G1 or S phase, or as asynchronous cultures. However, when mutants were induced during the expression period by having I 125 dU incorporated into the DNA Žsix-pointed star., only 0.05 of the cultures had zero mutants, compared with 0.42 that would be predicted by a Poisson distribution. Also, when asynchronous cells were irradiated with 1.0 Gy initially followed 24 h later by two 1.0-Gy doses during the expression period, the fractions of cultures with zero mutants Žsolid symbols. were statistically lower Ž p s 0.05, least squares analysis. than the fractions of cultures with zero mutants in the populations irradiated with a single
1
Although a few HPRT mutants isolated from irradiated populations have been observed to multiply slower than wild-type cells Ždata not shown and Ref. w22x., there is no reason to believe that the few slowly multiplying HPRT mutants will be any more abundant than other slowly multiplying cells that have been isolated from irradiated populations w23x. However, if the HPRT mutants multiplied slower than the wild-type cells, i.e., if Žmutantsrinitial clonogenic cell. ) Žmutantsrclonogenic cells at replating., the data points in Fig. 1 should be shifted to the right, which would cause the points observed for the single doses to cluster above the theoretical line predicted by the Poisson distribution.
dose. If we had increased the expression period beyond 3.5 to 4.5 days after the last dose, we would not have expected any change in mutant frequency since at ; 4 days we have observed a constant mutant frequency w13x. ŽIn fact, if the mutant frequency would have increased with a longer expression time, the fraction of cultures with zero mutants would be even lower.. Finally, when the cultures, for all treatment groups, were replated after the expression period into four individual flasks for each culture, the distribution of mutants within the four flasks was not significantly different from a Poisson distribution; i.e., the variance was approximately equal to the mean number of mutants observed in the particular culture Žsee ratios in Table 1.. Thus, all of these results strongly suggest that most of the mutants were induced initially at the time of irradiation and were not a product of genomic instability during or shortly after the expression period.
4. Discussion Our conclusions that HPRT mutants are induced directly at the time of irradiation differ from those reached by Loucas and Cornforth w14x who concluded that at least 50% of the mutants are produced during or shortly after the expression period. They also studied human EJ-30 cells as we did, but they exposed the irradiated cells to HAT medium during the expression period to eliminate any HPRT mutants that would have expressed the mutant HPRT phenotype during the expression period. They concluded that most of the mutants were produced late during or after the expression period, because after the HAT medium was removed, the frequency of mutants in the cells replated into 6-TG medium was almost as great following the HAT treatment as when the HAT treatment was omitted. However, this assumes that the turnover of the HPRT protein and synthesis of mutant protein during the expression period would not be reduced appreciably by the HAT treatment. Another implicit assumption is that the intracellular pools of nucleosides that might compete with incorporation of 6-TG into DNA are not altered by the HAT treatment; there is evidence that this assumption may not be true w6x. Also, Hofer et.
E.A. Leonhardt et al.r Mutation Research 426 (1999) 23–30
al. w24x showed that treating CHO cells with HAT for 8 days had dramatic effects subsequently on dose response curves for cell killing and micronuclei that resulted from I 125 dU incorporated into DNA. Furthermore, these unexplained qualitative and quantitative effects persisted for as long as 30 days after termination of the HAT treatment. In addition, in experiments that quantify the frequencies of HPRT mutants, the cells must be maintained below a critical density. Above the critical density, metabolic cooperation between the wild-type and mutant cells will cause the observed mutant frequency to be lower than the true frequency. This critical density must not be exceeded either during the expression period or after the cells are replated into 6-TG w3,9,10x. We have determined that the critical density for EJ-30 cells is 1 = 10 5 cellsrcm2 before replating and 4.4 = 10 3 after replating, and in all of our experiments, the maximum cell densities were never exceeded. There is no mention of cell densities during the expression period in the experiments of Loucas and Cornforth w14x. Possibly, the delayed increase in mutant frequencies observed between 8 and 16 days after irradiation for cells treated with HAT, compared with the decline in mutant frequencies between 8 and 16 days for cells not treated with HAT could be caused by any or all of the factors mentioned above. Thus, to confirm the conclusions from the HAT experiments that at least 50% of HPRT mutations result from latent effects of radiation-induced damage, the possible complications mentioned above should be excluded. Also, the molecular spectra of the mutations, with and without HAT treatment, should be determined because 30– 70% of radiation-induced HPRT mutations consist of large deletions involving all or part of the hprt gene Žsee references in Ref. w13x. whereas, delayed mutations that have been observed several weeks or months after irradiation consist primarily of point mutations or small deletions w15x. In summary, our fluctuation analysis presents strong evidence that most if not all radiation-induced HPRT mutants are induced directly at the time of irradiation as the DNA damage is being repairedr misrepaired. Due to the experimental design of the fluctuation analysis, mutations resulting from the initial insult, whether manifested immediately or within a few days, would not affect the distribution.
29
This is different from damage due to genomic instability which has been detected several weeks or months after the initial insult. Furthermore, the frequency of radiation-induced HPRT mutants was the same when CHO cells were replated for mutant selection as when they were not replated Žtermed in situ. w10x. In both cases, metabolic cooperation was avoided by maintaining the cells below a critical density. If mutants had been produced during the expression period as microcolonies were forming, we would expect the fraction of microcolonies having a mutant in the in situ method to be higher than the fraction of the cells having a mutant in the replating method. The fact that the in situ and replating methods gave very similar mutant frequencies supports the conclusion reached from our fluctuation analysis that HPRT mutants are induced directly by X-irradiation.
Acknowledgements We would like to thank Barbara Docklova for her invaluable assistance on the project, and Dr. Norman Albright for his help with the statistical analysis. This research was supported in part by Research Grant CA31808 from the National Institute of Health.
References w1x J.A. Nelson, J.W. Carpenter, L.M. Rose, D.J. Adamson, Mechanisms of action of 6-thioguanine, 6-mercaptopurine, and 8-azaguanine, Cancer Res. 35 Ž1975. 2872–2878. w2x E.R. Nestmann, R.L. Brillinger, J.P.W. Gilman, C.J. Rudd, S.H.H. Swierenga, Recommended protocols based on a survey of current practice in genotoxicity testing laboratories: II. Mutation in Chinese hamster ovary, V79 Chinese hamster lung and L5178Y mouse lymphoma cells, Mutat. Res. 246 Ž1991. 255–284. w3x R. DeMars, Genetic studies of HG-PRT deficiency and the Lesch–Nyhan syndrome with cultured human cells, Fed. Proc. 30 Ž1971. 944–955. w4x B.R. Migeon, Studies of skin fibroblasts from 10 families with HGPRT deficiency, with reference to X-chromosome inactivation, Am. J. Hum. Genet. 23 Ž1971. 199–210. w5x J.H. Carver, W.C. Dewey, L.E. Hopwood, X-ray-induced mutants resistant to 8-azaguanine: II. Cell cycle dose response, Mutat. Res. 34 Ž1976. 465–480.
30
E.A. Leonhardt et al.r Mutation Research 426 (1999) 23–30
w6x V.M. Riccardi, J.W. Littlefield, Adaptation in Lesch–Nyhan cells exposed to aminopterin, Exp. Cell Res. 74 Ž1972. 417–422. w7x P.J. Benke, N. Herrick, A. Hebert, Transport of hypoxanthine in fibroblasts with normal and mutant hypoxanthine–guanine phosphoribosyltransferase, Biochem. Med. 8 Ž1973. 309–323. w8x W.G. Thilly, J.G. Deluca, H. Hoppe IV, B.W. Penman, Phenotypic lag and mutation to 6-thioguanine resistance in diploid human lymphoblasts, Mutat. Res. 50 Ž1978. 137–144. w9x E.H.Y. Chu, H.V. Malling, Mammalian cell genetics: II. Chemical induction of specific locus mutations in Chinese hamster cells in vitro, Proc. Natl. Acad. Sci. U.S.A. 61 Ž1968. 1306–1312. w10x J.H. Carver, W.C. Dewey, L.E. Hopwood, X-ray-induced mutants resistant to 8-azaguanine: I. Effects of cell density and expression time, Mutat. Res. 34 Ž1976. 447–464. w11x A.J. Grosovsky, J.B. Little, Effect of growth rate on phenotypic expression of 6-thioguanine resistance in human diploid fibroblasts, Mutat. Res. 110 Ž1983. 163–170. w12x J.P. O’Neill, A.W. Hsie, Phenotypic expression time of mutagen-induced 6-thioguanine resistance in Chinese hamster ovary cells ŽCHOrHGPRT system., Mutat. Res. 59 Ž1979. 109–118. w13x E.A. Leonhardt, M. Trinh, H.B. Forrester, R.T. Johnson, W.C. Dewey, Comparisons of the frequencies and molecular spectra of HPRT mutants when human cancer cells were X-irradiated during G1 or S phase, Radiat. Res. 148 Ž1997. 548–560. w14x B.D. Loucas, M.N. Cornforth, Postirradiation growth in HAT medium fails to eliminate the delayed appearance of 6-thioguanine-resistant clones in EJ30 human epithelial cells, Radiat. Res. 149 Ž1998. 171–178. w15x J.B. Little, H. Nagasawa, T. Pfenning, H. Vetrovs,
w16x w17x
w18x
w19x
w20x
w21x w22x
w23x
w24x
Radiation-induced genomic instability: delayed mutagenic and cytogenetic effects of X-rays and alpha particles, Radiat. Res. 148 Ž1997. 299–307. S.E. Luria, M. Delbruck, Mutation of bacteria from virus sensitivity to virus resistance, Genetics 28 Ž1943. 491–511. S. Squires, R.T. Johnson, A.R.S. Collins, Initial rates of DNA incision in UV-irradiated human cells differences between normal, xeroderma pigmentosum and tumour cells, Mutat. Res. 95 Ž1982. 389–404. E.A. Leonhardt, M. Trinh, H.B. Forrester, W.C. Dewey, Persistent decrease in viability as a function of X-irradiation of human bladder carcinoma cells in G1 or S phase, Radiat. Res. 149 Ž1998. 343–349. M. Schmidt, B.R. Migeon, Asynchronous replication of homologous loci on human active and inactive X chromosomes, Proc. Natl. Acad. Sci. U.S.A. 87 Ž1990. 3685–3689. A. Grossmann, V.M. Maher, J.J. McCormick, The frequency of mutants in Human fibroblasts UV-irradiated at various times during S-phase suggests that genes for thioguanineand diphtheria toxin-resistance are replicated early, Mutat. Res. 152 Ž1985. 67–76. B.S. Jacobson, T.L. Morgan, Growth rates of hprt and tk mutant CHO cell lines, Mutat. Res. 344 Ž1995. 141–145. S.L. Nelson, I.M. Jones, J.C. Fuscoe, K. Burkhart-Schultz, A.J. Grosovsky, Mapping the end points of large deletions affecting the hprt locus in human peripheral blood cells and cell lines, Radiat. Res. 141 Ž1995. 2–10. W.K. Sinclair, X-ray-induced heritable damage Žsmall-colony formation. in cultured mammalian cells, Radiat. Res. 21 Ž1964. 584–611. K.G. Hofer, X. Lin, S.-P. Bao, DNA damage, micronucleus formation, and cell death from 125 I decays in DNA, Acta Oncologica 35 Ž1996. 825–832.