The “recall effect” in radiotherapy: Is subeffective, reparable damage involved?

In, .l Rudmm, Onrolo~y Bml Printed I” the U.S.A. All nghts

Phpy Vol. reserved.

18. pp.

689-695 Copyright

0360.3016/90 $3.00 + .OO cc’ 1990 Pergamon Press plc

l Special Feature

THE “RECALL IS SUBEFFECTIVE, HIROSHI

KITANI,

KARIN

Department

PH.D.,

LINDQUIST,

EFFECT” IN RADIOTHERAPY: REPARABLE DAMAGE INVOLVED?

TOSHIFUMI MS.

AND

KOSAKA,

M.S.,

M. M. ELKIND,

TETSUO PH.D.,

FUJIHARA,

PH.D.,

M.S., M.M.E.

of Radiology and Radiation Biology, Colorado State University, Fort Collins, CO 80523

It has been proposed that lethal mutations among the progeny of a surviving cell could he the basis for the recall effect when chemotherapy is applied subsequent to the repair of normal-tissue injury resulting from a course of radiation therapy. Because radiotherapy is usually multifractionated, the possibility exists that repair of heritable injury of this type could occur between fractions as is the case for sublethal damage. To examine this possibility, the endpoint small-colony formation was used-an endpoint which integrates the effects of a number of radiationinduced aberrancies including lethal mutations-and low-dose-rate irradiation. It was found that, even after net surviving fractions comparable to those sought in radiotherapy were reached, little damage remained expressible as a deficiency in the size of the colony generated from a surviving cell. We conclude that the damage expressible as a lethal mutation is reparable and therefore the recall effect must be attributed to some other cellular mechanism. Recall effect, Memory effect, Remembered dose. Lethal mutations, Repair, Sublethal damage, Small-colony mation. Low-dose-rate irradiation.

for-

would be incapable of further proliferation leading to an enhanced response to the second therapy. Because conventional radiotherapy is multifractioned-in the course of which it has been amply documented that surviving cells, like the clonogenic cells of the skin (24), repair sublethal damage (8)-the question arises: Is the damage which is responsible for the production of lethal mutations modifiable by repair processes which may take effect during successive interfraction intervals? We have examined this question in the following way. First, we have used irradiation at very low dose rates to optimize the ability of cells to effect repair including the repair of sublethal damage. Second, we have applied an endpoint which is sensitive to the production of lethal mutations but which, at the same time, can be influenced by essentially all other factors that may affect the number of progeny that are derived from a colony-forming and therefore a surviving cell. The endpoint in question is radiation-induced, small-colony formation. It is well known that ionizing radiation, as well as many other cytotoxic agents, gives rise to colonies of reduced size ( 10, 11, 15, 16, 19,23), and that the reduction in size decreases with decreasing dose rate ( 10, 15). At the same

INTRODUCTION

The “recall” or “memory effect” when conventional radiotherapy is combined with chemotherapy (2-5, 17), (e.g., eryrthema or desquamation in a previously irradiated field) is a well accepted phenomenon particularly in reference to skin located in the irradiated field. One hypothesis for this observation is that the initial course of therapy depletes the stem cells in question and that the killing of additional stem cells during a second course hastens the onset of a natural process of senescence of these cells ( 13). Whether or not the “remembered dose” when both therapies are radiotherapy involves the same mechanism is not known. In a study of the production of lethal mutations, Seymour et al. (18) reported that sterile progeny may be found among cells which survive a dose of irradiation (e.g., among cells in a colony derived from a single surviving cell, non-colony forming cells are found). They proposed that lethal mutations could be the cellular basis for the recall effect because surviving cells could pass lethal defects along to their descendents. Hence, even though a tissue would appear to be fully reconstituted after a course of radiation therapy, significant proportions of stem cells

Reprint requests to: M. M. Elkind, Ph.D., MS., M.M.E.

Accepted

Acknowledgements-This research was supported by the U.S. Department of Health and Human Welfare, Public Health Service, via Grant No. CA 47497 from the National Cancer Institute. 689

for publication

19 July 1989.

690

I.J. Radiation

Oncology

l Biology 0 Physics

level of survival. the distribution appears to be similar for different radiations (16, 23). With increasing dose, the distribution becomes progressively more skewed towards smaller colony diameters. an effect which may be partially offset in a given cell line by giant cell formation (22). A small colony can be the net effect of a number of factors, such as, a delay in division, an increase in the intermitotic interval, the fusion of daughter cells, and/or the generation of sterile progeny. This last effect would be scored as a lethal mutation if the viability of the cells comprising the offspring of a surviving cell is determined by colony formation, and then only if the division-incompetent cell does not lyse. Because the likelihood of lysis increases with time, a sterile cell formed early in the development of a clone from a single survivor may not be scored as a lethal mutation late in the period of colony formation, although its effect would be detected nevertheless via the small-colony endpoint. However, it is not likely that the other factors have a significant influence on the development ofsmall colonies. Radiation-induced division delay (6, 25) is a small fraction of the total time which ordinarily elapses before colonies are scored. lntermitotic times do not appear to be significantly lengthened judging from the weak dependence on dose of the rate of DNA synthesis of those cells still cycling after exposure (12. 25). and the insensitivity to dose of intermitotic intervals (scored by time-lapse microphotography) among those progeny lines of a surviving colonyforming cell which continue to generate offspring (20. 2 1). As for the fusion of daughter cells. this is an event of low probability which also usually results in the termination of a progeny line (20. 21). Thus, although small-colony formation is an endpoint that integrates the contributions of cellular damage from several potential sources-all of which might contribute to the defective reconstitution of a tissue after a radiation experience-it is likely that the loss of progeny lines due to the induction of sterile progeny, or lethal mutations, plays a principal role. To begin with, the ability of a sterile cell to persist in a clonal population could depend upon the quality of the radiation, but also, could depend on specific and characteristic properties of the cells in question. For example, if proliferation normally is limited by cell-to-cell contact (e.g., as do cells which usually form confluent layers of viable, but nondividing cells) then it would seem reasonable that the morphological persistence of an inviable cell in a developing clone could reflect signals received by it from neighboring cells. By suppressing the metabolic pathways which normally lead to growth and division. these signals could also suppress those processes that are degradative and that culminate in lysis. The influence on colony size of the latter situation would be similar to that if lysis did occur, but on the other hand, the persistence of sterile cells would be required to score lethal mutations. To determine whether or not the various effects on cell growth and division which could result in the defective reconstitution of an irradiated population of cells. as in

March

1990. Volume

IX. Number

3

skin tissue, are reparable, we have measured the dependence on dose rate of an integrator of these effects. smallcolony formation. By resorting to low dose rates, total doses were accumulated which were sufficiently high to make it clear that in addition to sublethal damage, other forms ofdamage were repaired during extended exposure intervals. METHODS

AND

MATERIALS

Repair-competent V79 Chinese hamster cells were grown attached in dishes or in flasks, as described in earlier publications (e.g.. ref. 7 and citations therein). Briefly. cells were cultured, irradiated. and assayed for colony formation in the alpha modification of Eagle’s Minimum Essentials Medium plus 10% fetal calf serum. For irradiation with high-dose-rate “‘Cs y-rays, appropriate numbers of cells were inoculated into T-25 flasks containing 10 ml of medium to yield about 100 colonies after a minimum of 7 days of incubation at 37.5”C in an atmosphere of 2% CO? in air. Following the formation of colonies, cells were stained with methylene blue. In some experiments. irradiated cultures. corresponding to surviving fractions below 0. I, were refed with 10 ml of medium on the 2nd. 3rd, or 4th day after irradiation. Survival curves were constructed, normalizing the plating efficiency (PE) of the controls to 100%. and standard errors were computed as in past work (7).

Cells prepared for a low-dose-rate experiment wcrc harvested from cultures actively growing in 90 mm petri dishes. Roller bottles (670 cm’) were inoculated with 3.4 X IO6 cells in 200 ml of medium and incubated at 37.5”C while they were rotated at 0.8 rpm. Cells to be exposed to low-dose-rate ‘j’Cs y-rays were inoculated and grown in the same way. Every 2 days. control cells and cells irradiated for 47.5 hr were suspended with a trypsin solution, counted, diluted for plating efficiency or survival measurements, and reinoculated into other roller bottles to continue the incubation of control or irradiated cells. At selected intervals, from a suspension of control or irradiated cells, T-25 flasks were inoculated with graded numbers of cells prior to their exposure to 13’Cs y-rays at a high dose rate for the measurement of a survival curve as outlined above. Hi&dmt+rale irradiation Single, unattached cells at room temperature were exposed in T-25 flasks to ‘j’Cs -y-rays from above at a dose rate of 80.3 cGy/min delivered by a 6000 Ci source. The dose rate was estimated using a thimble chamber* placed at the same position at which the cells were irradiated. Low-dme-rate irradiution Irradiations were started promptly after cells were inoculated into two roller bottles while the bottles were ro-

Recall effect and reparable damage 0 H.

tated at 0.8 rpm and kept at 375°C. The temperature of the incubator, the rotation of the bottles, and the irradiation were remotely and continuously monitored to insure that the system performed as required during successive 2 day periods. Using a roller bottle which had been cut in half, thermoluminescence chips were taped onto the inside surface toward the neck, at the middle, and toward the bottom of the same roller bottle, 24 chips per periphery at each location. The two halves of the bottle were then taped together and rotated during exposures for calibration. The chips in turn were calibrated against a thimble chamber* and their readings were averaged. For irradiations at 1.85 cGy/hr, the aperture of a 600 Ci 13’Cs panoramic source was covered by 1.2 cm of lead. As for control cells growing in similar roller bottles at 37S”C, irradiated cells that were harvested periodically were used to measure high dose rate survival curves as outlined above.

KITAN

691

el al

V79-B3lOH ‘Ws Y-rays DOSE,

Gy,“‘Cs

REFED ABCD

(-,------\

To estimate the size of colonies following a given dose and a given period of incubation, after staining the relative diameters of colonies were measured using a colony counter and Area Totalizer.+ This optical electronic apparatus automatically counts all colonies above a given diameter setting. Thus, spectra of relative sizes of colonies on the surface of a flask were rapidly and reproducibly characterized.

RESULTS

Figure 1 shows composite data to illustrate: the dosedependent skewing of the distribution of colony diameters toward small colonies (i.e., toward 0.5-1.0 mm dia., left column in each histogram); the lack of an effect on the colony-size diameters due to refeeding (i.e., the histograms at the lower right side of the figure); and the survivals which result from refeeding after high doses. These are all plotted relative to the single-dose, single-cell survival curve (upper right panel) determined from colonies stained 7- 12 days after irradiation at a high dose rate (see Table 1). The distributions in Figure 1 were all determined after 7 days of incubation. A 7-day period of incubation was chosen because it is an adequate length of time to insure that colonies from unirradiated cells are easily storable but still not long enough to obscure differences in diameter caused by prolonged incubation. Typically, when surviving fractions smaller than about 0.2 are to be determined, the inoculum is increased relative to the number of cells

* Victoreen IOOR. ’ Biotran III Automatic Colony Counter and Area Totalizer, Model C I 12, New Brunswick Scientific Co., Edison, N.J., USA.

DIA. OF COLONIES:

A=0.5-l.Omm; C=l.5-ZOmm;

B=l.O-1.5 mm; D= >2.0mm

Fig. 1. A test of the influence of a medium change on the single dose survival curve of V79 Chinese hamster cells. The distribution of colony diameters is plotted for the doses shown all delivered at a high dose rate, 80.3 cGy/min. For the doses 12, 16. and 20 Gy, the 10 ml of medium which the flasks initially contained was replaced with 10 ml of fresh medium on either the 2nd, 3rd, or 4th day after exposure. All colonies were stained on the 7th day. The survival data in the upper right panel show that refeeding did not produce significant proportions of satellite colonies. The continuous curve in this panel is based upon the survival-curve parameters for control cells in Table I. The intervals of colony diameters which were used to characterize the distributions are indicated at the bottom.

used for zero dose in order to maintain approximately a uniform level of precision with dose in the estimation of surviving fraction. This practice means that below 0.2 surviving fraction appreciably increasing numbers of cells are inoculated and are killed per flask with increasing dose. The survival curves whose parameters are in Table 1 were determined in this way. If the diameters of colonies are

1.

692

Days

J. Radiation Oncology 0

Biology

0 Physics

Table 1. Survival parameters

(80.3 cGy/min)

Accumulated dose, Gy

P.E.

Low dose rate cGy/hr

34 41

0 0

34 41

1.85 1.85

0 0 14.9 18.0

P.E. = Plating efficiency of the single-cell population:

March

1990,

Volume 18. Number 3

of the cells used for the data in Figures 2 and 3

D,,, +- S.E., Gy

D,,f + S.E., Gy

n f S.E.

17.9 88.1 I

1.37 Ik 0.55

2.18 i 0.06

8.0 -t 1.2

74.1 69.3

7.18 +- 0.42

2.17 -t 0.05

10.3 + 3.1

D,, = Reciprocal of the slope of the initial part of the survival curve; Do,T

= Reciprocal of the slope of the final part of the survival curve: n = Extrapolation number. Note: The survival curves at 34 and 41 days were indistinguishable and, hence, the data were pooled in each case.

to be compared at a fixed time after irradiation, one has to account for the possibility that the distribution may have been contributed to by the progressive depletion of the medium due to the increasing numbers of killed cells in each flask with increasing dose. The results in Figure 1 show that refeeding had essentially no effect either on the dose-dependent distribution of colony diameters or on the surviving fraction after 12, 16, and 20 Gy.

6.84 cGy/hr. The data in Figure 4 compare the effect on colony size of high doses delivered at different dose rates, that is, 20 Gy at a high dose rate (left column), and three even larger doses at 1.85, 4.89, and 6.84 cGy/hr (right column). Although the latter distributions have not returned fully to that of unirradiated cells, it is evident that they are much more similar to the controls than to the distribution after 20 Gy at 4820 cGy/hr.

Pop&&ion growth and plating eficiencies at low dose rates

Dependence qfplating eficiency on dose rate

If small colony formation, whatever its cellular/molecular basis may be, is not reparable, then when cells are irradiated at a low dose rate their growth curve should depart increasingly with time from that of unirradiated cells. However, at a dose rate of 1.85 cGy/hr (Fig. 2), the growth curve is very close to that of unirradiated cells. After about 41 days of growth, a change in the slopes of both curves is apparent. At that time, a new batch of medium was introduced into the experiment. Whether or not the new medium was responsible for these changes, it is evident that up to an accumulated dose of about 1800 cGy there is little difference in the rates of growth of the two populations. Normalizing both sets of data by the plating efficiencies as a function of time, (see Fig. 3), does not significantly alter the relative positions of the curves (data not presented). Table 1 shows the survival parameters at a high dose rate for control cells and for cells that survived irradiation at a low dose rate. These data indicate that up to 4 1 days, low-dose-rate irradiation did not alter the survival properties of cells chronically irradiated at 1.85 cGy/hr. The results in Figure 3 show that the plating efficiency of the chronically irradiated cells relative to that of control cells did not change in the 74 days during which they were exposed. It is likely, therefore, that a steady-state, small proportion of sterile progeny was present throughout the exposure period suggesting that sterile progeny lines were generated continuously to replace cells which lysed during the extended irradiation.

ofcolony sizes, low-dose-rate irradiation Experiments similar to those at 1.85 cGy/hr (e.g., Figs. 2 and 3) were performed using dose rates from 0.52 to

Distributions

In Figure 5, the results of several experiments showing the plating efficiency (PE) dependence on dose rate determined after successive 47.5-day intervals are summarized. Equating the PE of unirradiated cells to lOO%, the

TOTAL

DOSE,

IO 1

12oF)

Gy

20 I

30 I

_I

0 Oe .

V79-B3lOH o unlrradlated l 1.85 cGy/hr

0

IO

20

30

DAYS

40

50

60

70

OF GROWTH

Fig. 2. The growth of a population of V79 Chinese hamster cells irradiated at I .85 cGy/hr with “‘Cs y-rays, for 41.5 hr out of every 48 hr, compared to the growth of unirradiated cells. The dashed curve is an extrapolation of a single line through the points for the first 41 days. An example of a distribution of colony diameters from this experiment is in Figure 4.

Recall effect and reparable

damage 0 H. KITANI el al.

693

V79-B3lOH 137Cs Y-rays 6o

34.3

Gy M35cGy/

40

0 20Gy

(4820

cGy/hr)

50 5 i

40’ 20

30

40

/

I

50

60

ABCD

30.2

0

60 Colony 40

50

60

DAYS

70

80

data show that at 0.52 cGy/hr, the PE was not decreased, but that at higher dose rates PE decreased progressively to 52%. Because at 1.85 cGy/hr initially no reduction in growth rate was observed (Fig. 2), as expected no change in growth rate was observed at a lower dose rate, 0.52 cGy/hr (data not presented). Hence, because the PE was loo%, lethal mutations could not have been produced at the latter dose rate even though a total of 680 cGy was accumulated during 54 days. More quantitative indications of the repair of the damage leading to small colonies, which is suggested by the distributions in Figure 4, are summarized in Table 2. For low net surviving fractions and high total doses, the percentage of large colonies, that is, larger than 1.5 mm at 7 days, was appreciably greater for low dose rates than it was after 20 Gy at a high dose rate even when larger total doses at low dose rates were considered.

As reviewed in the Introduction, brief exposures to radiation produce a shift in the distribution of colony sizes toward smaller diameters (Fig. 1). At reduced dose rates, however, the distributions are shifted back toward larger sizes which are closer to those of unirradiated cells. Because small colonies result from the integration of several processes, including the production of sterile cells or lethal mutations, the shift in the distributions toward that of unirradiated cells, as shown in Figure 4 and Table 2, indicates that repair relative to these processes has occurred.

25.9 Gy(6.84cGyI

20 0

ABCD

Fig. 4. Distributions of colony diameters of V79 Chinese hamster cells 7 days after the doses and dose rates shown of 13’Cs y-rays. Details as for Figure I.

It is well known that the repair of sublethal damage during a protracted exposure (8) (see also citations in ref. 9) results in a net survival considerably greater than the same dose delivered during a brief exposure. (The net survivals following low-dose-rate exposures summarized in Table 2 are considerably higher than the survivals which can be estimated from the high-dose-rate survival curves,

E z

I

I

120

Total dose t

1

* 5

DISCUSSION

ABCD

40

A=0.5-I mm B=l-1.5mm C=1.5-2mm D= >2mm

OF GROWTH

Fig. 3. The dependence of plating efficiency on time for the eight sampling points in Figure 2. The horizontal lines are drawn at the respective average values and the uncertainties of the average plating efficiencies are standard deviations. For individual points, the uncertainties are standard errors.

diometers

Gy(4.89cGy/

V79-B3lOH

2 w >

90Gy

67Gy

40-

‘%

r

rays

20-

% =

cc

0

0

I I

1 2

3

4

DOSE

RATE,

cGy/hr

5

6

7

Fig. 5. The dependence of the steady-state plating efficiency of V79 Chinese hamster cells irradiated at the dose rates shown and to the total doses indicated. The plating efficiency data for 1.85 cGy/hr are derived from Figure 3 and are relative to the average value for unirradiated cells which was normalized to 100%. Uncertain ties are standard errors.

1. J. Radiation Oncology

694

0 Biology 0 Physics

Table 2. Plating efficiencies Dose rate cGy/hr. 0.0 0.52 1.85 4.89 5.28 6.84 4820

PE, %, steady-state 100 100 84 49 52 52

March

1990. Volume

18, Number

3

and total doses for the data in figures 4 and 5

Tot. dose Gy (Fig. 4)

Colonies >I.5 mm, YO

S”

34 30

1.6 x IO-’ 9.4 x 10-5

25.9 20

5.3 x 10-3 1.2 x 10-3

83 ND 77 61 ND 69 5

PE = Plating efficiency relative to unirradiated cells, Figure 5; S, = S”, where S is the steady-state 5 and n is the number of 2-day growth periods; ND = Not determined. Note: Colony-size distributions were determined at 7 days after plating.

Table 1). However, one might argue that because survival can be expected to increase as the dose rate is decreased, so too one should expect the distribution of colony sizes to become less skewed toward small diameters with decreasing dose rate. A survival curve, which has negative curvature. indicates that killing results from a damage accumulation process, and ample experimental evidence has shown that such damage is reparable. What has not been demonstrated heretofore is that the several factors which may contribute to a reduced colony size, like lethal mutations, are closely connected to lethality resulting from damage accumulation. From the survivals in Table 2, one may also question how tight is the coupling between survival and colony size. For the dose rates 1.85, 4.89, and 6.84 cGy/hr, the percentages of colonies greater than I .5 mm diameter (Fig. 4) are close to the value for unirradiated cells even though the net survivals were low. Furthermore, as is evident in Figure 2, at 1.85 cGy/hr the number of doublings effected by the irradiated cells after 74 days is within 6% of the controls. Lethal mutations could not have significantly decreased the growth rate of that population even though the total dose was 34 Gy. The data reported by Seymour et al., (18) and by Mothersill and Seymour (14) indicate that the dose-dependence of the loss of proliferative ability, which they scored as lethal mutations, is a curve without a shouder. One might infer from the foregoing that subeffective damage is not involved. Thus, even though conventional radiotherapy is multifractionated, these observations suggest that lethal mutations should be present in reconsti-

Tot. dose, Gy (Fig. 5) 0 6.8 34 67 70 90

surviving

S”

1.6 1.0 I.0 1.1

fraction

1.0 1.0 x 10m3 x 10-9 x 10-9 x 10-x

from Figure

tuted tissue. However, more recently Born and Trott (1) and Gorgojo and Little (written personal communication, August 1988) using several different cell lines, both observed shoulders on the curves described above. Indeed, little effect due to lethal mutations was measurable at a dose of about 2 Gy in the latter studies nor in earlier work with V79 Chinese hamster cells (7). This suggests that the accumulation of subeffective damage can be involved in the induction of lethal mutations and hence the possibility arises that during protracted exposures the repair of such damage might occur. In view of the large total doses which were accumulated in the various experiments of this study, as in Figures 4 and 5 and Table 2, it is evident that considerable repair occurred of subeffective damage related to cell growth and division as well as to lethality. Because the irradiations were extended over prolonged periods, total doses and survival levels comparable to what are used and sought, respectively, in clinical practice were achieved. For example, at 4.89 cGy/hr a net surviving fraction of 1 X 10m9 was reached after a total dose of 67 Gy (Table 2). Thus, in view of the repair relative to colony size that is evident in Figure 4, it appears entirely likely that considerable repair of subeffective damage which leads to lethal mutations also occurred. Although the properties of cells differ, our results suggest that if the recall effect (or the remembered dose) in radiotherapy is due to the propagation of lethal damage, very considerable repair nonetheless must occur during the course of multifractionated therapy. Alternatively, our data suggest that the recall effect is not closely coupled to lethal mutations.

REFERENCES Born, R.; Trott, K. R. Clonogenicity of the progeny of surviving cells after irradiation. Int. J. Radiat. Biol. 53:319330; 1988. Cassady, J. R.; Richter, M. P.; Pior, A. J.; Jaffe, N. Radiationadriamycin interactions: preliminary clinical observations. Cancer 36:946-949; 1975. D’Angio, G. J. Clinical and biological studies of actinomycin

D and roentgen irradiation. Am. J. Roentgenol. 87:106109; 1962. 4. D’Angio, G. J.; Farber, S.; Maddock, C. L. Potentiation of X-ray effects by actinomycin D. Radiology 73: 175-177: 1959. 5. Danjoux, C. E.; Catton, G. E. Delayed complications in colorectal cancer treated by combination radiotherapy and

Recall effect and reparable damage 0 H. KITANI cf al.

6.

7.

8.

9.

IO.

I I.

12.

13.

14.

15.

5-Iluorourocil. Int. J. Radiat. Oncol. Biol. Phys. 2: 18 l- 184; 1977. Elkind. M. M.: Han, A.; Volz. K. W. Radiation response of mammalian cells grown in culture. IV. Dose dependence of division delay and postirradiation growth of surviving and nonsurviving Chinese hamster ceils. J. Nat. Cancer Inst. 30:705-721: 1963. Elkind. M. M.; Ngo. F. Q. H.: Hill. C. K.; Jones, C. Do lethal mutations influence radiation transformation frequencies? Int. J. Radiat. Biol. 53:84Y-859: 1988. Elkind. M. M.; Sutton, H. A. Radiation response of mammalian cells grown in culture. I. Repair of X-ray damage in surviving Chinese hamster cells. Radiat. Res. 13:556-593; 1960. Elkind. M. M.: Whitmore. G. F. The radiobiology of cultured mammalian cells. New York: Gordon and Breach Science Publishers; 1967. Fox. M.; Gilbert. C. W. Continuous irradiation of a murine lymphoma line P388F in vitro model for human cancer. Int. J. Radiat. Biol. I1:339-347: 1966. Fox. M.: Nias. A. H. W. The influences of recovery from sublethal damage on the response of cells to protracted irradiation at low dose-rate. Curr. Topics Radiat. Res. 7:72103: 1970/71. Hahn. G. M.; Little. J. B. Plateau-phase cultures of mammalian cells: an in v//m model for human cancer. Curr. Topics Radiat. Res. 8:39-83; 1972. Hellman. S.: Botnick. L. E. Skin cell depletion. an explantion of the late cIfects of cytotoxins. Int. J. Radiat. Oncol. Biol. Phys. 2:181-184: 1977. Mothersill, C.; Seymur. C. The influence of lethal mutations on the quantification of radiation transformation frequencies. Int. J. Radial. Biol. 5 1:723-729; 1987. Nias. A. H. W. Clone size analysis: a parameter in the study

16.

17. 18.

19.

20.

21.

22. 23.

24.

25.

695

of cell population kinetics. Cell Tissue Kinet. 1: 153-165; 1968. Nias, A. H. W.; Gilbert, C. W.: Lajtha, L. G.; Lange, C. W. Clone-size analysis in the study of cell growth following single or during continuous irradiation. Int. J. Radiat. Biol. 9: 275-290; 1965. Pinkle, D. Actinomycin D in childhood cancer. Pediatrics 23:342-347; 1959. Seymour, C. B.; Mothersill, C.: Alper, T. High yields of lethal mutations in somatic mammalian cells that survive ionizing radiation. Int. J. Radiat. Biol. 50: I67- 179: 1986. Sinclair, W. K. X-ray-induced heritable damage (small-colony formation) in cultured mammalian cells. Radiat. Res. 21:584-61 I; 1964. Thompson, L. H.; Suit, H. D. Proliferation kinetics of Xirradiated mouse L cells studied with time-lapse photography. I. Experimental methods and data analysis. Int. J. Radiat. Biol. 13:39 l-397: 1967. Thompson, L. H.; Suit, H. D. Proliferation kinetics of Xirradiated mouse L cells studied with time-lapse photography. II. Int. J. Radiat. BioI. 15:347-362; 1969. Tolmach, L. J.; Marcus, P. I. Development of X-ray induced giant HeLa cells. Exp. Cell Res. 20:350-360: 1960. Westra, A.: Barendsen, G. W. Proliferation characteristics of cultured mammalian cells after irradiation with sparsely and densely ionizing radiations. Int. J. Radiat. Biol. Il:477485: 1966. Whithers, H. R. Recovery and repopulation in r[ro by mouse skin epithelial cells during fractionated radiotherapy. Radiat. Res. 321227-239; 1967. Whitmore. G. F.; Stanners, C. P.; Till, J. E.: Gulyas. S. Nucleic acid synthesis and the division cycle in X-irradiated L-strain mouse cells. Biochem. Biophys. Acta. 47:66-77; 1961.