Mutagenesis and repair by low doses of α radiation in mammalian cells

Abstract

Low doses of α radiation in basements have been causally implicated in lung cancer. Previous studies have concentrated on high dose effects, for which no significant repair was found. In the present study, the methodology for measuring mutation by quantitating mitotic breaks and gaps was found to be applicable to G₂-phase Chinese hamster ovary cells irradiated with 10–50 cGy of α radiation. The mutation yield in such cells closely resembles that of γ irradiation. Caffeine, which inhibits repair, produces the same straight line increase of α and γ mutation yields plotted against the dose. In the absence of caffeine, the repair of α radiation lesions is almost twice as great as for γ radiation. Mitotic index changes substantiate these interpretations. It is proposed that the higher ion density associated with α radiation may result in fewer lesions being missed by the repair processes. The quantitation of chromosomal lesions for G₂ cells exposed to low doses of α radiation, γ radiation, or chemical mutagens in the presence and absence of caffeine is a rapid and reproducible methodology. Protection from mutational disease in a fashion similar to the use of sanitation for infectious disease appears practical.

Reports on the effects of α radiation of mammalian cells usually concentrate on high doses, where cell killing is extensive, and complex multihit chromosomal lesions are obtained (1–3). In this paper, we report results on the effects of doses less than 50 cGy, a dose region that should be easier to understand theoretically and that presumably is responsible for a significant amount of human lung cancer. We have earlier shown for γ radiation that high doses produce complex chromosomal rearrangements as opposed to the simple breaks and gaps from low doses (4).

In previous papers, a methodology for measuring chromosomal aberrations produced by physical and chemical agents in mammalian cells has been described (5, 6). The procedure is rapid, sensitive, and reproducible, and yields quantitative measurement of breaks and gaps in mitotic chromosomes resulting from mutagenesis in G₂ cells. The principle used consists of scoring microscopically identifiable breaks and gaps in mitotic chromosomes. In these chromosomes, condensation has reduced the overall length by approximately 20,000-fold, so that the corresponding increase in thickness renders each chromosome visible under the microscope. The condensation is accomplished by means of successive coiling, supercoiling, folding, and other highly specific molecular interactions that are attended by attachment of specific macromolecules, e.g., proteins, to designated points in the chromosome. Thus, even a small mutational lesion at an appropriate position, by preventing normal attachment of the supercoiling protein to the mutated DNA, can presumably prevent condensation at a given point resulting in an apparent discontinuity (break or gap) in the resulting mitotic figure. An example of a gap without a discontinuity is illustrated by the work of Kremer et al. (7), who demonstrated that the mitotic chromosomes of patients with fragile X disease, who display apparent gaps in their X chromosomes, actually have additional repetitive sequences but no real DNA discontinuities.

Our methodology measures genetic damage associated with chromosomal aberrations. These include gene mutations at points that prevent normal condensation in mitosis as well as actual chromosomal discontinuities. The method is highly sensitive, because DNA sites involved in chromosomal condensation constitute a considerable fraction of the entire genome. The large target size and high reproducibility of this method make it well suited for mutation quantitation. In this connection, it should be noted that, whereas mutation is usually considered to be an inherited change, lethal mutations are not inherited. A more logical definition of mutation would seem to be any nonphysiological change in genome structure. By this definition, the aberrations here measured would be mutations. G₂-phase cells offer means for rapid and reproducible scoring of chromosomal aberrations because of the DNA condensation process. They also are the seat of a DNA repair process that can be inhibited by caffeine under carefully controlled conditions (5). Finally, because the DNA is double in G₂, these cells present the maximum target size for mutagenic attack.

The region of 0–50 cGy was selected for study as representing the dose probably most involved in the development of lung cancer arising from α radiation in basements and from exposure to uranium by miners (8–10). High doses produce more mutations but also kill cells, so that cancer risk may be less than expected.

Our methodology quantitatively measures mutagenesis and mutation repair in mammalian cells exposed to mutagens. α radiation produces a much higher ionization and free radical density than does γ radiation. It becomes necessary, therefore, to understand what effect this will have on genetic damage and repair in mammalian cells. Moreover, low-dose α radiation originating from radon in basements of human dwellings has been implicated in a significant percentage of human lung cancer (8, 10), and α radiation originating from uranium has been associated with very high incidence of lung cancer in miners, particularly those who smoke cigarettes (11).

Methods and Materials

The very low penetrance of α radiation makes measurement of its mutational effects more difficult than the corresponding measurements with γ radiation. We replaced our standard cell, the AG09391 human immortalized lymphocyte, by a cell that would attach to a surface as a monolayer to ensure reasonably uniform radiation doses over the cell population. Chinese hamster ovary (CHO)-K1 cells can easily be plated as a monolayer so that a fairly uniform exposure is obtained. Cytologic mutational scoring was carried out in accordance with definitions described in ref. 12.

The exposure of the cells to α radiation was accomplished as follows: 2.5 × 10⁵ CHO-K1 cells were plated into 30-mm Mylar bottomed tissue culture dishes and incubated 24 hr before exposure to α particles. Such exposure was performed by using a ²³⁸Pu α particle collimated exposure system described earlier (13). The average energy of the α particles at the cell–Mylar surface was ≈3.5 MeV delivered at a dose rate of 3.51 cGy per second. The quantitation of mutagenesis and of the mitotic index was carried out with the CHO cell as previously described with the human lymphocyte (5). γ Radiation used for comparison studies was delivered from a ¹³⁷Cs source, with an energy of 1.17 MeV. Caffeine was used at a concentration of 1.0 mg/ml during the 2-hr period after irradiation to inhibit mutation repair as described (5, 6, 14). To accumulate mitotic figures, colcemid, at a final concentration of 0.05 μg/ml, was added to the cultures 1 hr before harvesting the cells. Collection and processing of the cells were done according to standard cytogenetic procedures (15).

Results

Mutation Yields of α and γ Radiation.

Fig. 1 presents the aberration yield for each of the radiations studied. A larger yield in the presence of caffeine is obtained with both radiations, demonstrating that caffeine prevents repair of α ray-induced mutations in a fashion generally similar to that previously described with γ rays (5, 6, 14). In Fig. 1, the failure of some of the curves to pass through the origin is not statistically significant. The quantitative data of Table 1 show that the aberrations in the presence of caffeine, i.e., without repair, are similar for γ and α radiation. In the absence of caffeine, the aberration yield is much greater for γ radiation.

Fig 1.

(a) Mutation yield, expressed as mitotic chromosome aberrations per 20 cells, of CHO cells irradiated in G₂ with α rays in the absence (hollow squares) (y = 0.146X − 0.701) and presence (solid squares) (y = 1.056X − 2.745) of caffeine. b presents the corresponding curves for γ radiation. The mutation yields in the presence of caffeine (solid squares) (y = 1.006X + 4.662) are very similar for the two types of radiation, but in the absence of caffeine (hollow squares) (y = 0.744X + 0.745), the mutation yield after γ radiation is distinctly greater than that after α. Presumably, cellular repair is more effective after α radiation.

Table 1.

Calculation of the percent of mutations repaired for each irradiation procedure, averaged from three experiments

Dose, cGy	α Radiation			γ Radiation
	Mutation yield		% Repair	Mutation yield		% Repair
	+ Caffeine	− Caffeine		+ Caffeine	− Caffeine
10	9	1	89	17	9	47
25	20	2	90	33	22	33
35	28	2	93	47	28	40
50	56	8	86	58	38	34

The mutation yield is expressed in aberrations per 20 cells. The percent repair is obtained by subtracting the yield in the absence of caffeine from that in its presence and dividing the result by the yield in the presence of caffeine.

Quantitation of Mutation Repair in G₂ for α and γ Radiation.

Table 1 shows the mutation yield in the presence and absence of caffeine for α and γ radiation of CHO cells, from which data the amount of mutation repair in the G₂ cells was calculated. In the absence of caffeine, the number of aberrations at each dose is significantly lower for α radiation, but in the presence of caffeine, where repair is suppressed, the two sets of data are much more similar. Indeed, the combined data fit the same straight line going through the origin (Fig. 2) despite the large difference in linear energy transfer of the two radiations. These results imply that the magnitude of the initial DNA damage is similar for α and γ radiation in this dose range, but that repair is more extensive, indeed approximately doubled, for α radiation. The data of Table 1 show that at these low doses, G₂ cells exhibit considerably more caffeine-inhibited DNA repair for the α- than for the γ-ray-induced lesions. Conceivably, this could be because the α ray lesions, more localized in each chromosome because of the higher ion density associated with this radiation, are more easily engaged by the repair complexes.

Fig 2.

Demonstration that the combined values of α and γ radiation produced lesions (three experiments each) yield a reasonably straight-line function of the dose that goes through the point (0, 0) (y = 1.017X + 0.987). Such a curve eliminates the uncertainty of low-dose extrapolation, which has been a source of great difficulty in the past.

Measurement of Mitotic Index.

As previously demonstrated (5, 14, 16), repair of mutated G₂ cells causes a lag in their reaching mitosis and therefore a drop in mitotic index. There is no such drop in the presence of caffeine, which prevents the repair process. Measurement of the depression in mitotic index in irradiated G₂ cells and its prevention by caffeine afford another means for assessing the extent of repair (16). Table 2 presents the change in mitotic index for each type of radiation, in the presence and absence of caffeine, respectively. The large drop in mitotic index for cells treated with a mutagen in G₂, plus the restoration of the higher value by the presence of caffeine, demonstrates a basic similarity in pattern between the actions of α and γ radiation. The data of Table 2 confirm our conclusion, drawn from measurement of mutational yield, that damage by low doses of α radiation is subject to repair and to an extent greater than that of γ radiation. These mitotic index figures form a specific indicator of mutagenesis that could be used as an even more rapid mutagen detection system or to test data obtained by other methodologies. That caffeine restores the mitotic index to the same or similar value as that of the unirradiated control culture is an indication that caffeine prevents all, or almost all, repair in these irradiated G₂ cells.

Table 2.

Comparison of the effect of α and γ radiations on the mitotic index of CHO cultures, in the presence and absence of repair-inhibiting concentrations of caffeine

cGy	α Radiation		γ Radiation
	Mitotic index − caffeine	Mitotic index + caffeine	Mitotic index − caffeine	Mitotic index + caffeine
0	(100  ± 19) %	(100  ± 16) %	(100  ± 9) %	(100  ± 23) %
25	(28  ± 6) %	(104  ± 25) %	(72  ± 9) %	(106  ± 8) %
50	(9  ± 4) %	(71  ± 4) %	(41  ± 16) %	(91  ± 3) %

In the absence of caffeine, the α radiation has a greater inhibitory action on the mitotic index than γ radiation. In both cases, addition of caffeine tends to restore the mitotic index toward its control value. The mitotic index was measured 2.0 hr after the irradiation, the last hour of which was in colcemid. The mitotic index is expressed as a percentage of that of the unirradiated control culture, which varied in different experiments between 3.0 and 5.

Discussion

Previous reports have described the lesions caused by α radiation as being poorly repaired (1, 17, 18). However, these papers describe the effects of high doses of radiation varying between 0.50 and 200 Gy. They monitor complex chromosomal lesions such as translocations, double-strand breaks, and rings, which are multiple hit events, and their observations are not confined to G₂ cells. At high doses, there is much cell killing, which greatly lessens the otherwise expected effect of these lesions on human disease. Most human exposure to α radiation appears to lie in the low-dose range, where few complex chromosomal lesions are obtained (9). We find that very few complex chromosomal lesions occur below 50 cGy, but many chromosomal breaks and gaps occur. Presumably these represent either single-chromosome breaks or mutations that prevent normal chromosome condensation at various positions along each chromosome, as we showed earlier with γ radiation (6, 19). These lesions can be repaired in G₂ cells by a process that is inhibited by caffeine for damage caused by α as well as γ radiation. The action of caffeine on the more complex lesions resulting from high doses of α radiation has not yet been defined.

Use of caffeine to inhibit repair in G₂ cells irradiated with low doses of radiation involves the following features: (i) the G₂ lag associated with repair is eliminated as would be expected; (ii) the mitotic index in irradiated cells is nearly restored by caffeine to the value of the unirradiated control, in agreement with expectation, because most repair is eliminated; (iii) a plot of the combined chromosomal lesions versus the radiation dose in the presence of caffeine in the interval between 0 and 50 cGy yields a straight line through the origin, as expected if the caffeine treatment has indeed eliminated repair in both instances (Fig. 2); and (iv) quantitation of the mitotic index in the presence and absence of caffeine is rapid and convenient, easily lending itself to human epidemiological studies, and can serve as a check on the direct counting of chromosomal lesions.

The data here presented indicate that exposure of G₂ mammalian cells to α radiation produces mutational effects similar to those resulting from γ radiation. The G₂ interval of the cell cycle is particularly convenient for mutation monitoring, and the large size of the DNA target makes it very sensitive. Further studies are necessary to determine cell sensitivity during the S and G₁ phases of the life cycle and to ascertain whether antimutagens might be able to afford protection. It is particularly necessary to analyze the dynamics and mechanism of action of combinations of mutagens, such as those involved in cigarette smoke plus α radiation, which are known to cause far greater health hazards than the sum of the effects of each agent alone (11, 20).

The molecular picture we have visualized for the G₂ repair process involves continuous traverse of the DNA by a monitoring molecular complex that stops when it encounters a recognizable structural change in the DNA. Repair enzymes of different kinds attach to the immobilized monitoring complex and carry out whatever DNA repair processes are appropriate. Thereafter, the aggregate dissociates from the DNA that continues the condensation process and enters mitosis (19). If sufficient unrepaired mutations remain, apoptosis can be initiated early in G₁, shortly after mitosis completion or in other points in the life cycle (21).

Thus the general picture that explains γ ray mutation and repair in G₂ cells appears applicable, at least qualitatively, to that for low doses of α radiation. Although a large number of DNA repair enzymes have already been identified, caffeine under the carefully controlled conditions used in our procedure appears to stop virtually all mutational repair in the G₂ phase. We interpret this to mean that caffeine prevents entry of the cell into the repair mode in which all of the different repair reactions are carried out, as visualized in Fig. 3.

Fig 3.

Visualization of the dynamics of the repair reaction, R. In the absence of mutation, the cells in G₂ proceed along the circular path, a. When mutated loci are present, these are detected by monitoring molecules that take the cell along path b to the repair loop in which the many different repair enzymes carry out their functions before restoring the cell to the completion of its G₂ journey. Consequently, the mutagenized cells are delayed in reaching mitosis so that a drop in the mitotic index is found. Caffeine prevents entry into the repair loop and so maintains the mitotic index at its control value.

Previous work has demonstrated that exposure of mammalian cells to α radiation can cause mutation even when the path of the α particle is confined to the cytoplasm (22, 23). Under these conditions the resulting ionization processes presumably cause formation of free radicals, which diffuse into the nucleus producing mutations. The data presented here would include the effects of both these primary and secondary actions due to α radiation.

By combining the A_L methodology (23) with that described here, it should be possible to determine the relationship between molecular changes in DNA structure and the breaks and gaps described. It will be especially important to determine the extent of any erroneous repair.

Our methodology for testing the radiation sensitivity of lymphocytes to identify persons who are at elevated risk for radiation damage can also be used for α radiation if steps are taken to allow for the low penetration of this radiation. α radiation, implicated as an important factor in human lung cancer, when combined with other factors like smoking, becomes an even greater threat to human health.

The results here described support the concept that mutational disease is preventable by identification and removal of mutagens from the environment. Similarly, agents like caffeine that inhibit repair must also be controlled. Finally, blood samples will identify persons at elevated risk for mutational disease, who can then be counseled to lead more hygienic lifestyles.

The low-dose region of exposure to α radiation considered in this paper appears to be important epidemiologically and can be quantitated simply and reproducibly. It would appear that these measurements may help control mutational disease in the same way that sanitation measures are important in control of infectious disease.

Abbreviations

• CHO, Chinese hamster ovary