Effects of Different Ionizing Radiations on Human Cells in Tissue Culture: IV.

Modification
of Radiation Damage
Author(s): G. W. Barendsen and H. M. D. Walter
Source: Radiation Research, Vol. 21, No. 2 (Feb., 1964), pp. 314-329
Published by: Radiation Research Society
Stable URL: http://www.jstor.org/stable/3571569 .
Accessed: 14/10/2014 11:26

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

Radiation Research Society is collaborating with JSTOR to digitize, preserve and extend access to Radiation
Research.

http://www.jstor.org

This content downloaded from 192.84.147.234 on Tue, 14 Oct 2014 11:26:53 AM

All use subject to JSTOR Terms and Conditions
RADIATION RESEARCH 21, 314-329 (1964)

Effects of Different Ionizing Radiations on Human

Cells in Tissue Culture
IV.Modification
of RadiationDamage

G. W. BARENDSEN AND H.. M . WALTER

Radiobiological Institute of the Organizationfor Health Research TNO,
Rijswijk (Z.H.), Netherlands

INTRODUCTION
During the last few years accurate quantitative analysis of the reproductive
capacity of mammalian cells in tissue culture, as introduced by Puck et al., has
been applied to investigate various problems in cellular radiobiology (1-3).
In previous publications results of experiments with kidney cells of human
origin were described, in which inhibition of clone formation was studied after
irradiation with a-particles, deuterons, P-particles, and X-rays (4-6). Important
differences were found between dose-survival curves obtained with densely and
with sparsely ionizing radiations.
Modification of radiation damage in biological systems by various treatments
has been investigated most thoroughly with radiations of low average LET. Ex-
periments described in this paper show that densely and sparsely ionizing radiations
differ markedly in the degree in which damage to the reproductive capacity of
human cells can be influenced by various conditions and treatments. The time
interval between plating of the cells and irradiation, fractionation of the dose of
radiation, anoxia, and the presence of various chemicals have been found to modify
the quantitative response of cells to X-rays significantly, whereas with densely
ionizing a-particles from Po210much smaller changes in response were observed.
MATERIALS AND METHODS
In all experiments kidney cells of human origin (T1 cells) were used. The
culture medium, preparation of single-cell suspensions, irradiation techniques, and
growth conditions have been described previously (4, 5).1 The number of cells
1In part of the experiments a 200-kv X-ray machine was used, and in later experiments a
250-kvp X-ray machine. Both radiations had an HVL of about 2.0 mm Cu. The dose rate with
the 250-kvp X-rays was about 200 rads/min. The survival curves obtained with standard con-
ditions were identical for both radiations.
314

This content downloaded from 192.84.147.234 on Tue, 14 Oct 2014 11:26:53 AM

All use subject to JSTOR Terms and Conditions
 RBE OF RADIATIONS STUDIED ON HUMAN CELLS 315

which 14 days after irradiation have developed into clones comprising more than
fifty cells has been taken to represent the number of surviving cells. Plating
efficiencies of unirradiated controls ranged from 70 to 110% of the values deter-
mined from counting the "single-cell" suspension in a hemocytometer before dilu-
tion and plating. The initial suspensions obtained by trypsinization of stock
cultures contained about 106 cells/ml and were diluted to 200 cells/ml.
Percentage survival is given relative to controls treated in exactly the same
way except for the irradiation. In each experiment the mean value is derived of
at least three and usually four or more dishes, each of which has received the same
dose of radiation and treatment. Standard deviations were calculated from the
variation between the dishes of such a group. If this value was smaller than the
square root of the total number of clones counted per group of dishes, the latter
value was used.
In order to measure the effect of anoxia, the medium was removed from the
culture dishes after the cells had attached to the bottom. Air or nitrogen was then
passed over the cells during a time interval of 10 minutes which was found sufficient
to reach equilibrium (7). The oxygen concentration in the nitrogen was measured
with a Hersch cell (8).
EXPERIMENTAL RESULTS

Effects of Time Interval between Plating and Irradiation

In the experiments described in previous publications, a schedule was strictly
adhered to, in which the cells were taken out of the incubator 4 hours after plating
and allowed to attain room temperature during at least 10 minutes. After these
4 hours more than 99% of the cells had attached to the bottom. The dishes were
then irradiated within 2 hours. Control experiments had shown that at the be-
ginning and at the end of this period the radiosensitivity was the same within
experimental errors; i.e., the dose-survival curves were identical.
In the more involved experiments to be described in this paper it was necessary
to extend this period from 2 to 4, 8, and sometimes 12 hours. Therefore the effect
of the interval between plating and irradiation was investigated. During this
interval the cells were maintained at 37?C. The combined results of four ex-
periments are given in Fig. 1. In each experiment a control curve at 4 hours after
plating was included and was found to be the same within experimental errors.
From these results it may be concluded that in experiments with Po210a-particles
the time between plating and irradiation is important only if the interval is in-
creased to more than 28 hours. Consequently a period of at least 20 hours is
available for reproducible experiments.
Quite different results were obtained with 250-kvp X-rays. Immediately after
plating the survival percentage at a given dose is lowest; i.e., the cells are most
sensitive. During the first few hours after plating the sensitivity decreases rapidly,

This content downloaded from 192.84.147.234 on Tue, 14 Oct 2014 11:26:53 AM

All use subject to JSTOR Terms and Conditions
 100
Ln \ \"
LA
o \\ \

oU

C
o, -

? ,\\\\12 3I
0
\ 1 4 5\ 6 97 10
o

1 2 3 ^ 5 6 7 8
dose in(rod x 1
FIG. 1. Inhibition of clone formation as a function of the radiation dose i
time interval between plating of the cells and irradiation. Curves 1, 2, 3, 4
curves obtained with Po"21a-radiation administered 4, 28, 52, 76, and 100 ho
respectively. Curves 6 to 15: survival curves obtained with 250-kvp X-rays a
4, 8, 12, 20, 28, 52, 76, and 100 hours after plating, respectively.

This content downloaded from 192.84.147.234 on Tue, 14 Oct 2014 11:26:53 AM

All use subject to JSTOR Terms and Conditions
 RBE OF RADIATIONSSTUDIED ON HUMAN CELLS 317

and appreciably higher survival percentages are observed if the irradiation takes
place, for instance, 4 hours after plating as compared with 2 hours after plating
(see Fig. 1, curves 7 and 8).
This difference between a-particles and X-rays with respect to the time interval
between plating and irradiation is more clearly demonstrated in Fig. 2, which
is derived from Fig. 1. The change in survival percentage is here plotted as a
function of the interval between plating and irratiation for a dose of 150 rads of
a-radiation (curve 1) and a dose of 550 rads of 250-kvp X-radiation (curve 2)
both of which at 4 hours after plating result in about 12% survival.
In addition to the time interval between plating and irradiation, it was of
interest to study the effect of the temperature at which the cells are maintained
before and after irradiation. Therefore cells were kept at room temperature (20?
+ 2?C) during 4 hours, either before or after irradiation, and compared with cells
kept at 37?C. Although the general trend was the same in four experiments, the

0
L.

c.
O 70-

60 /

s~
u 50,
o 50. 7 J /
a- 40, /

O40- if /2

3 doses of radiation if administered4 hours after la

both

20'
E
E __I J j
o1 10' 2 3 0 5 7 8 hours

10 i20 30 40 50 60 70 80 90 100 hours

time intervol between ploting and irradiation
FIG.2. Effect of the intervalbetweenplatingand irradiationof cells on the capacityfor
clone formationafter (1) 150radsof a-radiationfrom Po21and (2) 550 rads of 250-kvpX-
radiation.These doseswere chosenbecausethe percentageof survivalis about the same for
bothdosesof radiationif administered4 hoursafterplating.

This content downloaded from 192.84.147.234 on Tue, 14 Oct 2014 11:26:53 AM

All use subject to JSTOR Terms and Conditions
318 BARENDSEN AND WALTER

effects varied in magnitude, and therefore the data could not be pooled. The re-
sults of one experiment are given in Table I. In this experiment groups of dishes
were either irradiated immediately after plating (Table I, group 1), kept at 20?C
during 4 hours between plating and irradiation (Table I, group 2), or kept at 37?C
during 4 hours between plating and irradiation (Table I, group 3). In each of
these groups two types of postirradiation treatment were employed. The cells
were either kept at 20?C during 4 hours and subsequently transferred to the in-
cubator at 37?C (Table I, groups lb, 2b, and 3b), or the dishes were transferred
to the incubator at 37?C immediately after irradiation (Table I, groups la, 2a, and
3a). Unirradiated controls for all groups were the same within the statistical
variation, and the average is given as 100%. From the results the following con-
clusions can be drawn:
1. Cells kept at 20?C during 4 hours between plating and irradiation are more
sensitive than cells kept at 37?C; i.e., lower survival percentages after doses of
350 rads and 700 rads are found in group 2a as compared with group 3a, and in
group 2b as compared with group 3b.
2. A significant effect of postirradiation treatment of the cells is found in all
groups. Cells maintained at 20?C during 4 hours after irradiation show lower sur-
vival percentages as compared with cells maintained at 37?C during 4 hours after

TABLE I
SURVIVAL OF CULTURED CELLS AFTER 350 RADS AND AFTER 700 RADS OF 250-KVP
X-RADIATION IN RELATION TO THE TEMPERATURE OF THE CELLS BEFORE
AND AFTER IRRADIATION

Percentagesurvival
Treatmentafter plating
Controls 350 rads 700 rads

1. Irradiation immediately after plat-

ing and:
a. After irradiation 4 hours at 37?C 104.6 - 4.2 30.6 d 2.3 3.8 ? 0.5
b. After irradiation 4 hours at 20?C 96.9 dt 4.0 8.9 ? 1.2 0.2 i 0.1

2. Irradiation 4 hours after plating; in

this interval the cells were kept at
20?C
a. After irradiation 4 hours at 37?C 101.7 - 4.2 30.7 = 2.3 3.2 - 0.7
b. After irradiation 4 hours at 20?C 93.2 = 4.0 17.4 ? 1.7 0.3 -4 0.1

3. Irradiation 4 hours after plating; in

this interval the cells were kept at
37?C
a. After irradiation 4 hours at 37?C 98.1 1 4.1 36.7 =- 2.5 5.6 - 0.9
b. After irradiation 4 hours at 20?C 101.8 i 4.2 21.9 -i 2.0 2.5 - 0.6

This content downloaded from 192.84.147.234 on Tue, 14 Oct 2014 11:26:53 AM

All use subject to JSTOR Terms and Conditions
 RBE OF RADIATIONSSTUDIED ON HUMAN CELLS 319

irradiation. Survival percentages after 350 rads and 700 rads are higher in group
la as compared with lb, in group 2a as compared with 2b, and in group 3a as com-
pared with 3b.
3. The difference between the effects of postirradiation treatments at 37?C and
20?C on the survival irradiated cells is more important if the cells are irradiated
immediately after plating or are kept at 20?C during 4 hours between plating and
irradiation, as compared with cells kept at 37?C during 4 hours between plating
and irradiation.
It may be concluded that for reproducible experiments with X-rays it is impor-
tant to adhere strictly to a schedule in which cells are stored in the incubator for a
given number of hours and after irradiation are replaced in the incubator as soon as
possible. With a-particles no significant changes in sensitivity due to temperature
differences were observed.

Fractionation of Radiation Doses in Relation to LET

Repair of sublethal damage induced by X-rays was first demonstrated by frac-
tionation experiments with Chinese hamster cells by Elkind and Sutton (2) and
found in our human cell line as well (3). With densely ionizing a-particles, howevei,
no differences in response due to fractionation of the dose have been detected (3).
In order to investigate this difference further, experiments have been carried out
in which cells were irradiated with doses of either a-radiation from Po210or 250-
kvp X-radiation in three equal fractions, administered 4, 8, and 12 hours after
plating, and in four parts, administered 4, 7, 10, and 13 hours after plating, In
the intervals the cells were maintained at 37?C. The results obtained with intervals
of 4 hours, in which the highest recovery effect was found with X-rays, are given
in Table II and Fig. 3. The data obtained were corrected for the decrease in sensi-
tivity of the cells with the time after plating, as described in the discussion. Figure
3 shows that with the a-radiation no effect of fractionation is observed, whereas
the results obtained with 250-kvp X-rays show a clear effect of fractionation at
higher doses. At a total dose of 100 rads of 250-kvp X-rays, however, administered
in three equal fractions, no difference with the single exposure was found; i.e., no
recovery at this low dose was apparent.
It would have been of great importance to extend these experiments to frac-
tionation schedules with smaller dose increments and longer intervals. Two ex-
periments were carried out with four fractions each separated by 3 hours, but with
250-kvp X-rays the effect of this fractionation schedule was at high doses smaller
than that obtained with three fractions separated by 4 hours. At low doses again
no effect of fractionation was found. The smaller effect at high doses is presumably
due to a less complete recovery in 3 hours as compared with the 4-hour intervals.
An increase of the time interval between the first and the last radiation increment
beyond 8 or 9 hours was not possible with our system because the correction men-

This content downloaded from 192.84.147.234 on Tue, 14 Oct 2014 11:26:53 AM

All use subject to JSTOR Terms and Conditions
320 BAiRENDSEN AND WALTER

TABLE II
SURVIVAL DATA ON SINGLE AND FRACTIONATED EXPOSURES TO PO210 a-PARTICLES
AND 250-KVP X-RAYS

Percentagesurvival
Dose Single Fractionated Per cent Fractionated
exposure
DoseoSingle exposurea correctionb exposure
(uncorrected) (corrected)

Irradiation in air
50 rads of a-radiation 49.0 i- 2.8 0 49.0 ? 2.8
100 rads of a-radiation 24.2 i 2.3 22.8 - 2.0 0 22.8 ? 2.0
200 rads of a-radiation 5.1 i 0.5 5.5 ? 0.4 0 5.5 ? 0.4

100 rads of X-radiation 81.5 ? 3.1 78.1 - 2.7 1.4 77.0 f- 2.7
300 rads of X-radiation 39.0 ? 2.3 53.1 ?= 2.5 4.3 50.8 ? 2.6
600 rads of X-radiation 7.8 -0.7 19.9 ? 1.3 12.3 17.5 ? 1.5
900 rads of X-radiation 1.3 ?t0.2 5.3 ? 0.4 23.6 4.0 - 0.5
1000 rads of X-radiation 0.44 ? 0.08 5.8 ? 0.5 2.80 4.2 - 0.6

Irradiation in nitrogen
100 rads of a-radiation 27.8 i 3.1 0 27.8 ? 3.1
200 rads of a-radiation 8.6 ? 0.6 8.0 - 0.8 0 8.0 4i 0.8
300 rads of a-radiation 2.5 ? 0.4 2.3 + 0.5 0 2.3 ? 0.5

300 rads of X-radiation 76.3 ? 4.4 79.7 ? 3.5 1.9 78.2 ?t 3.5
600 rads of X-radiation 53.0 ? 3.7 55.0 ? 2.8 3.7 53.0 ? 2.9
1000 rads of X-radiation 25.9 ? 2.5 40.1 ? 2.9 7.3 37.2 i- 3.2
1600 rads of X-radiation 4.2 ? 0.5
1750 rads of X-radiation 10.3 ?t 0.6 18.0 8.3 i- 0.7
2050 rads of X-radiation 4.9 ? 0.5 23.2 3.8 ? 0.6
2200 rads of X-radiation 0.64 i 0.14
a Total dose fractionated in three
equal parts, administered at 4, 8, and 12 hours after
plating.
b To be
applied only to fractionated exposures (see Discussion).

tioned earlier for the differences in survival as a function of the time interval be-
tween plating and irradiation became unduly large, introducing too large an
uncertainty in the results.

Effects of Fractionation of Different Radiations on Anoxic Cells

Anoxia has been shown to protect many biological systems from effects of X-rays.
Equilibration of our T1 cells with different mixtures of 02 and N2 has been found
to influence the sensitivity of these cells to 250-kvp X-rays significantly, whereas
the sensitivity to a-particles changed very little (see Fig. 3; compare curves 1
and 2, curves 3 and 6) (7). As has been described elsewhere, with very pure nitrogen,

This content downloaded from 192.84.147.234 on Tue, 14 Oct 2014 11:26:53 AM

All use subject to JSTOR Terms and Conditions
 RBE OF RADIATIONS STUDIED ON HUMAN CELLS 321

rInr

UI ^'^

10'

c O \

o 1

0,1
0 5 10 15 i0 25
dose in (rod x 100)
FIG. 3. Effects of single and fractionated exposuresof cultured human cells to Po2'0a-particles
and 250-kvp X-rays, with cells in equilibrium with air and with nitrogen. Curve 1, survival
after a-irradiation of cells in equilibrium with air; closed triangles correspondto single expo-
sures, and open triangles correspondto exposure with the total doses fractionated in three equal
parts with time intervals of 4 hours. Curve 2, survival after a-irradiationof cells in equilibrium
with nitrogen; closed circles correspondto single exposures, and open circles correspondto ex-
posures with the total doses fractionated in three equal parts administered with time intervals
of 4 hours. Curve 3, survival after 250-kvp X-irradiation; single exposures, cells in equilibrium
with air. Curve 4, survival after 250-kvp X-irradiation; total doses fractionated in three equal
parts administeredwith time intervals of 4 hours, cells in equilibrium with air. Curve 5, curve
indicating initial slope of survival curve 3. Curve 6, survival after 250-kvp X-irradiation; single
exposures, cells in equilibrium with nitrogen. Curve 7, survival after 250-kvp X-irradiation;
total dose fractionated in three equal parts administered with intervals of 4 hours, cells in
equilibriumwith nitrogen. Curve 8, curve indicating initial slope of survival curve 6.

dose reduction factors of 2.6 and 1.15 were obtained with 250-kvp X-rays and a-
particles, respectively (7). The change of the reduction factor as a function of the
oxygen concentration in the cell environment has also been discussed in detail
previously (7).
In the experiments described here, nitrogen was taken from a cylinder which
contained about 0.1% 02. With this nitrogen a dose-reduction factor of only 2.3
was obtained for 250-kvp X-rays by comparison of curves 3 and 6 in Fig. 3, whereas
with a-particles a dose-reduction factor of 1.15 was found by comparison of
curves 1 and 2.
Doses of 250-kvp X-rays or Po210 a-radiation were administered in three equal

This content downloaded from 192.84.147.234 on Tue, 14 Oct 2014 11:26:53 AM

All use subject to JSTOR Terms and Conditions
322 BARENDSENAND WALTER

fractions with intervals of 4 hours at 4, 8, and 12 hours after plating; during each
irradiation the cells were in equilibrium with nitrogen. In the intervals medium
was added to the dishes, and the cells were maintained at 37?C in the incubator
flushed with air and 3% C02. The results of these experiments are given in Table
II and represented by curves 2 and 7 of Fig. 3. Curve 2 shows that in cells equili-
brated with nitrogen no significant effect of fractionation is observed with a-radia-
tion from Po210. With 250-kvp X-rays, however, comparison of curves 6 and 7
of Fig. 3 shows that in the high dose range a considerable effect of fractionation is
observed even in anoxic cells. At a dose of 300 rads, however, no effect of frac-
tionation is found.

Effects of Cysteamine and Glycerol

A considerable number of substances have been found to protect different biologi-
cal systems from the action of ionizing radiations. The action of various com-
pounds on radiation-induced inhibition of clone formation has recently been
investigated by Vos et al., using T1 cells, cultured in the same Melinex culture
dishes as described by us (9-11). They achieved the best protection with relatively
high concentrations of cysteamine, and therefore we selected this compound to
investigate differences in protection due to variation of the LET. In Fig. 4 the
results are given of experiments in which cysteamine was added to the culture
medium at a concentration of 25 mM 30 minutes before irradiation. After irradia-
tion the medium was replaced immediately with normal medium without cyste-
amine. After a minute this medium was again removed and replaced. This washing
procedure was repeated three times, after which the cells were allowed to grow
and multiply during the standard time of 14 days at 37?C. The untreated controls
(curves 2 and 5 of Fig. 4) showed slightly higher survival percentages than the
curves obtained in other experiments without the washing procedure, as shown
by curves 1 and 4, which are given for comparison. The unirradiated controls
gave the same plating efficiencies for cysteamine-treated and untreated cultures.
The slight difference does not influence the conclusion to be drawn from Fig. 4 that
the protective action of 25 mM cysteamine is much more pronounced with 250-
kvp X-rays as compared with a-radiation from Po210.From Fig. 4 a dose-reduction
factor of 3.7 can be calculated for 250-kvp X-radiation, and a factor of 1.2 for
a-particles from Po210.
Another compound investigated for its protective action against effects of
a-radiation and X-radiation is glycerol. Glycerol has been shown to protect yeast
cells equally well against effects of densely ionizing particles as well as against
effects of X-rays (12). Furthermore, Vos and Kaalen (10) found with 200-kvp
X-rays a considerable protective effect of glycerol in tissue culture cells. Experi-
ments were therefore carried out in which glycerol to a final concentration of 10%

This content downloaded from 192.84.147.234 on Tue, 14 Oct 2014 11:26:53 AM

All use subject to JSTOR Terms and Conditions
 RBE OF RADIATIONSSTUDIED ON HUMAN CELLS 323

0
C

1-

015 10 15 20 25 30 35
dose in(rad x 100)

FIG. 4. Effects of a-radiation from Po210 and 250-kvp X-rays on cultured human cells treated
with 25 mM cysteamine during 30 minutes prior to irradiation. Curves 1 and 4 obtained with
a-radiation and 250-kvp X-rays, respectively; no treatment. Curves 2 and 5 obtained with a-
radiation and 250-kvp X-rays, respectively; cells received fresh medium 30 minutes before
irradiation and were washed four times after irradiation. Curves 3 and 6 obtained with a-radia-
tion and 250-kvp X-rays, respectively; cells received fresh medium with 25 mM cysteamine
during 30 minutes before irradiation and were washed four times after irradiation.

by weight was added to the growth medium 10 minutes before irradiation and re-
moved by washing immediately after irradiation. The treatment with glycerol
did not affect the plating efficiency of unirradiated cultures. The results of these
experiments are given in Fig. 5. It is clear that the same difference between Po210
a-radiation and 250-kvp X-rays is found with regard to the protective effect of
glycerol as was observed with cysteamine. That is, the protective action is much
larger for the sparsely ionizing X-rays as compared with the densely ionizing
a-particles. It is probably noteworthy, however, that the dose-reduction factor
found with X-rays is only 2.0 for 10% glycerol as compared with 3.7 for 25 mM
cysteamine, whereas with a-radiation a dose-reduction factor of 1.3 is observed
for glycerol as compared with 1.2 for cysteamine.

This content downloaded from 192.84.147.234 on Tue, 14 Oct 2014 11:26:53 AM

All use subject to JSTOR Terms and Conditions
324 BARENDSEN AND WALTER

100-

C0
U
,- 10-
0

m i1 \-1 2

0.

_ 1-
0

E
C

S\\~DISCUSSION

0.1 .. ...

dose in(rad
dOSe x 100)
;n(rad X
FIG. 5. Effects of a-radiation from Po210and 250-kvp X-rays on the capacity for clone forma-
tion of cultured human cells treated with 10% by weight glycerol in the culture fluid. Curves
1 and 2 obtained with a-irradiationof untreated and treated cells, respectively. Curves 3 and 4
obtained with 250-kvp X-irradiation of untreated and treated cells, respectively.

DISCUSSION
a. The effect of different time intervals between plating and irradiation on the
survival of cells may be due, in principle, to two factors-namely, to a change in
radiosensitivity with time after plating, and to an increase in cell number due to
multiplication of cells before irradiation. From curve 1 of Fig. 2 it may be deduced
that with a-radiation no change in sensitivity is observed during the first 28 hours,
after which an increase of the survival percentage is found. This increase is most
likely due to multiplication, because at 28 hours more than one generation time

This content downloaded from 192.84.147.234 on Tue, 14 Oct 2014 11:26:53 AM

All use subject to JSTOR Terms and Conditions
 RBE OF RADIATIONSSTUDIED ON HUMAN CELLS 325

(approximately 22 hours) has elapsed. This assumption is also in agreement with

curves 3, 4, and 5 of Fig. 1, which strongly suggest a multiple-hit character.
Curve 2 of Fig. 2 shows that, after a sharp increase during the first 4 hours, the
sensitivity to 250-kvp X-rays changes less rapidly, until starting between 20 and
40 hours after plating a more pronounced increase is found. For the greater part
the change in sensitivity during the first hours after plating cannot be due to cell
multiplication, because it is much more rapid than multiplication could account
for. At 550 rads of X-rays, for instance, a difference of a factor of 2 is found be-
tween the percentages of survival obtained with irradiation at 4 hours after plating,
as compared with 2 hours after plating. As an average generation time of about 22
hours was found for these cells in log phase, multiplication could account for an
increase by at most 10%.
An explanation of the large increase may be that the sensitivity to X-rays is
influenced by the trypsinization procedure used to prepare the single-cell sus-
pension (5). A "recovery" from this treatment during the period after plating
might result in a decrease in the sensitivity to X-rays, while the sensitivity to
a-particles might well remain constant. As discussed in a previous paper, results
obtained with a-radiation may be explained by assuming that one densely ionizing
particle, passing through a sensitive structure which takes up a large part of the
nucleus, is sufficient to produce inhibition of clone formation (5). According to
our results this does not change even if the cell is irradiated as much as 28 hours
after plating.
With X-rays some evidence that the trypsinization plays a part was obtained
by comparison of the sensitivity of cells treated with trypsin during 5 and 15
minutes. The plating efficiencies were the same, but the longer treatment resulted
in a higher sensitivity immediately after plating. Eight hours after plating, how-
ever, both groups of cells showed within experimental errors the same percentages
of survival at equal doses. It is clear that besides trypsinization other factors
which may influence the metabolic conditions must play a part. One of these
factors may be the phase of the cells with respect to the division cycle. It may
be assumed that the trypsinization and plating procedure, which is carried out at
room temperature (18? to 22?C), results in a partial synchronization of the cells.
Changes in sensitivity to X-rays during the division cycle have been described
for HeLa cells with the same criterion of clone formation (13). Other factors,
such as the attachment of cells to the bottom of the dishes, may also be important.
With regard to the increase in the number of clones after X-irradiation at more
than 30 hours after plating, as given in curve 2 of Fig. 2, the most likely cause is
multiplication of the cells, as discussed earlier for a-radiation. The assumption
that multiplication plays a major part here is in agreement with the fact that the
curves 1 and 2 for a-particles and X-rays, respectively, increase in this region in
approximately the same way.

This content downloaded from 192.84.147.234 on Tue, 14 Oct 2014 11:26:53 AM

All use subject to JSTOR Terms and Conditions
326 BARENDSENAND WALTER

From the results of irradiation at different times after plating it is clear that, if
experiments take more than a few hours to be completed, a correction has to be
applied for the change in sensitivity during the first 20 hours and for the effect of
multiplication during longer time intervals. In our experiments on fractionation,
irradiations were carried out at 4, 8, and 12 hours after plating, and therefore in
Table II the corrections applied for the effects of doses administered at 8 and 12
hours are given. As the effect of the time interval could not be measured with
sufficient accuracy at low doses, the assumption is made that in the first 20 hours
the effectiveness of a given dose is reduced by a fixed percentage depending only
on the time interval. Thus curves 9 and 10 in Fig. 1 are shifted to the right with re-
spect to curve 8 by 6.5% and 15% of the dose, respectively. Of a total dose of 900
rads, for instance, 300 rads is administered at 4 hours after plating and does not
have to be corrected. The effect of the second fraction of 300 rads at 8 hours is
corrected by 19.5 rads, corresponding to a decrease in survival by 6.8%. The
third fraction is corrected by 45 rads corresponding to a decrease in survival by
15.8%. The result is a total correction of 23.6%. In this way all correction factors
given in Table II are calculated. They are small for low doses, but significant at
higher doses.
b. The results obtained with fractionated exposures given in Fig. 3 and Table II
show that at a total dose of 100 rads no difference with the single exposure is found.
At a total dose of 300 rads, repair is observed only to the extent that the survival
percentage is very close to the corresponding value of curve 5 of Fig. 3, which is
drawn as the tangent to curve 3 in the low-dose region. As discussed in detail else-
where, accurate measurements of the survival curve obtained with 250-kvp X-rays
have provided evidence that the slope of the curve does not approach zero at very
low doses (3). The assumption was made that the damage from X-radiation is
produced partly by a "single-event" type of action, i.e., by the passage of a single
ionizing particle, partly by a "multiple event" or cumulative type of action (3).
It was further suggested that fractionation of the dose in sufficiently small parts,
spaced by intervals long enough to allow for maximum repair of sublethal damage,
would affect only the "multiple-event" part of the damage, but would leave the
"single-event" part of the damage unaltered. As a consequence the survival curve
obtained with 250-kvp X-rays obtained under such conditions of fractionated
exposure should be exponential.
On the basis of this hypothesis, curve 5 of Fig. 3 represents the exponential
survival curve to be expected if the "multiple-event" part of the damage were
completely eliminated by repair processes. From Fig. 3 it can be seen that up to
300 rads curves 4 and 5 are indeed the same within experimental errors; at total
doses of 600, 900, and 1000 rads of 250-kvp X-rays, however, the individual incre-
ments of dose are so large that the "multiple-event" part of the damage plays a
part even if maximum repair had been allowed to occur in the intervals.

This content downloaded from 192.84.147.234 on Tue, 14 Oct 2014 11:26:53 AM

All use subject to JSTOR Terms and Conditions
 RBE OF RADIATIONS STUDIED ON HUMAN CELLS 327

Figure 3 shows further that with anoxic cells the pattern of response to frac-
tionated X-irradiation is the same as compared with oxygenated cells. At a rela-
tively low dose of 300 rads no effect of fractionation is found, and at doses up to
1000 rads the observed values of curve 7 closely follow curve 8, which is drawn
as the tangent to curve 6 in the low-dose region.
c. The results obtained with cysteamine and glycerol show that densely ionizing
a-particles produce damage which is much less reduced than damage from sparsely
ionizing X-rays. The differences between effects of densely and sparsely ionizing
radiations are generally assumed to be due to differences in the spatial distribution
of ionizations. The a-particles used in the experiments described here have an
average LET of about 140 kev/, and produce ionizations closely spaced along
the tracks of the particles. About 20% of the energy is dissipated through S-rays
which all have less than 3 kev of energy and relatively high LET's as compared
with fast electrons. In a previous paper of this series it was shown that the effec-
tiveness per a-particle for the inhibition of clone formation of T1 cells increases
rapidly with increasing LET between 60 and 100 kev/u/. Furthermore these parti-
cles with an LET in excess of 60 kev/u give rise to exponential survival curves, and
no effect of fractionation is observed (6). It was concluded that inhibition of clone
formation is produced by a high local energy density; i.e., a certain minimum
amount of energy has to be dissipated within a small volume to produce damage
(6). The observed results on modification of radiation damage from Po210 a-parti-
cles might now be explained by the assumption that along the greater part of the
tracks of these particles much more energy is dissipated within the small volumes
traversed than is required for inhibition of clone formation. The fact that protective
compounds and anoxia are ineffective if administered after irradiation indicates
that these factors do not act through a repair mechanism but by reducing the ex-
tent of the primary damage. If, however, from an excess amount of energy dis-
sipated by a-particles from Po210 a certain percentage is rendered ineffective,
enough damage may be left to cause the same effect, and consequently the effi-
ciencies of these particles will not be reduced to a great extent. If this interpretation
is valid it is even possible to estimate from the relation of "effectiveness per parti-
cle" versus LET, given in Part III of this paper (6), that 25 mM cysteamine,
for instance, reduces the relative effectiveness of a-particles with an LET of 140
kev/,u (0.86) by a factor of 1.20 to the effectiveness of particles with an LET of
about 100 kev/, (0.71); i.e., it renders ineffective 40/140 or about one-third of
the energy primarily dissipated. In the same way the effect of anoxia on the
damage produced by a-particles can be explained by assuming that about 20 to
25% of the energy is rendered ineffective, whereby the effective LET is reduced
from 140 kev/,u to about 110 kev/u of tissue, and the effectiveness per particle by
a factor of 1.15.
For the explanation of the large variations in effectiveness of 250-kvp X-radia-

This content downloaded from 192.84.147.234 on Tue, 14 Oct 2014 11:26:53 AM

All use subject to JSTOR Terms and Conditions
328 BARENDSENAND WALTER

tion resulting from various treatments, we have to consider the spatial distribution
of the ionizations produced by this radiation. The mean LET of the fast electrons
generated by these X-rays is much lower than the LET of a-particles (14). As a
consequence, the type of locally concentrated energy deposition required for
damage to the reproductive capacity of the cells is produced very inefficiently by
these sparsely ionizing radiations as compared with the densely ionizing a-particles.
Furthermore, in a relatively great part of the locations where a sufficient amount of
damage is produced, this amount will not exceed the required minimum to a great
extent. As a consequence a small reduction of the damage primarily produced
might reduce by a comparatively large factor of 2 or 3 the number of locations at
which sufficient damage is left to result in inhibition of clone formation of a cell.
This may explain in a qualitative way the observed large reduction in the effi-
ciency of 250-kvp X-rays produced by anoxia and chemical protectors (14).
SUMMARY
Modification by various treatments of radiation-induced damage in kidney cells
of human origin has been investigated with 250-kvp X-rays and a-radiation from
Po210.The capacity for clone formation as a function of the radiation dose was
found to be influenced significantly by the time interval between plating and
irradiation, the temperature at which the cells were maintained before and after
irradiation, fractionation of the dose, and the presence of oxygen, cysteamine, and
glycerol. The modification of the response of the cells was shown to be much more
important after irradiation with sparsely ionizing 250-kvp X-radiation than with
densely ionizing a-radiation from Po210.An explanation of the differences between
the effects of these radiations has been given on the basis of the differences in
spatial distribution of the energy deposition.
RECEIVED:June 20, 1962

REFERENCES
Action of X-rays on mammalian cells. J. Exptl. Med. 103,
1. T. T. PUCKand P. I. MARCUS,
653-666 (1956).
2. M. M. ELKINDand H. SUTTON,Radiation response of mammalian cells grown in culture. I.
Repair of X-ray damage in surviving hamster cells. Radiation Res. 13, 556-593 (1960).
3. G. W. BARENDSEN, Dose-survival curves of human cells in tissue culture irradiated with
alpha-, beta-, 20 kV X- and 200 kV X-radiation. Nature 193, 1153-1155(1962).
4. G. W. BARENDSEN
and T. L. J. BEUSKER,Effects of different ionizing radiations on human
cells in tissue culture. I. Irradiationtechniques and dosimetry. Radiation Res. 13, 832-840
(1960).
5. G. W. BARENDSEN,
T. L. J. BEUSKER,A. J. VERGROESEN,
and L. BUDKE,Effects of different
ionizing radiations on human cells in tissue culture. II. Biological experiments. Radiation
Res. 13, 841-849 (1960).
6. G. W. BARENDSEN, H. M. D. WALTER, J. F. FOWLER,and D. K. BEWLEY,Effects of different
ionizing radiations on human cells in tissue culture. III. Experiments with cyclotron ac-
celerated a-particles and deuterons. Radiation Res. 18, 106-119 (1963).

This content downloaded from 192.84.147.234 on Tue, 14 Oct 2014 11:26:53 AM

All use subject to JSTOR Terms and Conditions
 RBE OF RADIATIONS STUDIED ON HUMAN CELLS 329

7. G. W. BARENDSEN, Damage to the reproductive capacity of human cells in tissue culture by

ionizing radiations of different linear energy transfer. In The Initial Effects of Ionizing
Radiations on Cells (R. J. C. Harris, ed.), pp. 183-194.Academic Press, London, 1961.
8. P. HERSCH, Galvanic oxygen recorder.Instr. Pract. 11, 937-942 (1957).
9. 0. Vos, L. BUDKE,and A. J. VERGROESEN, Protection of tissue culture cells against ionizing
radiation. I. The effect of biological amines, disulfide compounds and thiols. Intern J.
Radiation Biol. 5, 543-557 (1962).
10. O. Vos and M. C. A. C. KAALEN, Protection of tissue culture cells against ionizing radiation.
II. The activity of hypoxia, dimethyl sulphoxide, dimethyl sulphone, glycerol and cys-
teamine at room temperatureand at -196?C. Intern. J. Radiation Biol. 5, 609-621 (1962).
11. A. J. VERGROESEN,L. BUDKE,and O. Vos, Protection of tissue culture cells against ionizing
radiation. III. The influence of anoxia on the radioprotection of tissue culture cells by
cysteamine. Intern. J. Radiation Biol. 6, 117-126 (1963).
12. T. R. MANNEY, T. BRUSTAD, and C. A. TOBIAS, Effects of glycerol and of anoxia on the radio-
sensitivity of haploid yeasts to densely ionizing particles. University of California Report
UCRL-10181,(1962).
13. T. TERASIMA
and L. J. TOLMACH,
Changes in X-ray sensitivity of HeLa cells during the
division cycle. Nature 190, 1210-1211(1961).
14. G. W. BARENDSEN, Modification of radiation damage in relation to LET. Proceedings of the
Conference on Physical Factors Modifying Response to Radiation, New York Academy
of Sciences, November 29-December 1, 1962.In press.

This content downloaded from 192.84.147.234 on Tue, 14 Oct 2014 11:26:53 AM

All use subject to JSTOR Terms and Conditions