[CANCER RESEARCH 33, 1780 1784, July 1973]
Relationship between Growth and Radiosensitivity in the P388
Murine Leukemia1
John W. Harris, Brian Shon,2 and Juanito Meneses
Laboratory of Radiobiology. University of California, San Francisco, California 94122
SUMMARY
inoculation of IO6cells into male DBA/2 mice (The Jackson
Laboratories, Bar Harbor, Maine) weighing approximately
20 g. Whenever possible, blood-free tumors were selected
for passage, but despite this precaution tumors became
increasingly bloody by the 6th day of growth. Assay of the
inoculum required a TD503 confirmed a figure of 2 to 3
cells in DBA/2 mice (3). However, despite similar TD60's
Growth of the P388 lymphocytic leukemia as an ascites
tumor in DBA/2 mice is characterized by lengthening of
the mean cell-cycle time from 8.5 hr on Day 1 to 22.0 hr on
Day 5. This increase reflects increases in S phase from 6.2 to
14.8 hr, in G¡from 0.2 to 4.2 hr, and in G2 from 1.6 to 3.0
hr; the greatest change thus occurs in G,. The proportion of this tumor is more virulent than the P388 that we obtained
cells in S decreases from 70% on Day 2 to < 10% on Day 8, some years ago from Dr. J. Belli and used in earlier exper
the last day of growth. This decrease is reversed 16 hr after iments (12). An inoculum of IO6cells from the tumor used
transplantation of IO6cells to fresh hosts. Since the mitotic in the present experiments kills >90% of the hosts on the
index does not change during the 16-hr lag period, the cells 7th postinoculation day and all of the mice by the 8th
that surge into S between the 16th and 48th hr after day, whereas Belli and Andrews (3) report that their P388
transplantation probably originate in G! (or G0).
did not kill the hosts until the 12th day.
Assay of Tumor Growth and Labeling with 3H-Labeled
The characteristics of the P388 tumor make it a useful
model for investigating the relationships between cell kinetic Thymidine. Growth curves were constructed by washing the
parameters and sensitivity to X-radiation or chemothera- peritoneal cavities of tumor-bearing mice repeatedly with up
peutic drugs. We find that P388 cells collected from late- to 50 ml of heparinized 0.9% NaCl solution. These wash
plateau-phase tumors and irradiated in air in vitro are fluids were diluted in volumetric flasks and cell concentra
more radioresistant than are cells harvested from exponen
tions were determined with a Model B Coulter counter.
tially growing tumors.
Coulter counts were checked against hemocytometer counts
and differential counts on stained smears. Viability was
INTRODUCTION
tested by the standard nigrosin exclusion test (14).
P388 cells were labeled by i.p. injection of 3H-labeled
The P388 murine lymphocytic leukemia has been used in
numerous investigations of cytocidal agents, including ion thymidine (Schwarz/Mann, Orangeburg, N. Y.) at 0.1 to 1
izing radiations (3-6, 12, 18) and chemicals (1). It is par /¿Ci/g;specific activity ranged from 3.7 to 17 Ci/mmole,
ticularly advantageous for these studies because it grows depending on the experiment. (The lowest specific activity
and dose were used for labeled mitosis and repeated-labeling
as an ascites and because survival can be assayed accurately experiments.) Twenty min after injection of 3H-labeled
by classical in vivo titration techniques (3). Despite its
cells were collected and washed, resuspended in a
widespread use, however, the P388 has not yet received a thymidine,
few drops of cell-free ascites fluid, and smeared on clean
complete cytokinetic analysis. Since the therapeutic efficacy
of "cell-cycle-specific" cytocidal agents varies with the slides. These were then dried, fixed in absolute methanol,
relative proportion of cells in resistant and sensitive phases and processed for autoradiography as described elsewhere
of the cell cycle, we analyzed the growth kinetics of this (11).
X-irradiation. Cells were harvested from donors of appro
tumor and correlated changes in cytokinetic parameters priate age, washed once, resuspended in Tris-buffered 0.9%
during growth with the in vitro radiosensitivity of cells
NaCl solution (pH 7.0), and irradiated at room temperature
harvested from exponential and late-plateau-phase tumors.
in a Petri dish. X-irradiation was performed with a G.E.
Maxitron unit at 300 kVp and 20 ma with 2-mm copper
filtration; the dose rate was 500 rads/min. After irradiation,
MATERIALS AND METHODS
10s cells were injected i.p. into fresh mice, which were
Tumor. A P388 tumor, obtained from Dr. J. Brennan sacrificed for tumor cell count at appropriate intervals. Cell
(University of Pennsylvania), was passaged weekly by i.p. survival was determined by extrapolating the growth curves
'Work performed under the auspices of the U. S. Atomic Energy back to the ordinate.
Commission.
"Supported in part by USPHS Medical Student Summer Research
Fellowship FR05355-09.
Received October 17, 1972; accepted April 10, 1973.
1780
Calculation of Cytokinetic
Parameters. Cell-cycle limes
3The abbreviation used is: TD5(),50% tumor takes.
CANCER RESEARCH
VOL. 33
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1973 American Association for Cancer Research.
Growth and Radiosensitivity of P388 Leukemia
were determined by the labeled mitosis method (19) using
the 50% points on the curve. The growth fraction4 and cell
loss factor were computed by standard formulations (2, 16,
17, 21, 23).
RESULTS
Growth of the P388 Leukemia as an Ascites Tumor in
DBA/2 Mice. Sixteen hr after inoculation of IO6P388 cells
into DBA/2 mice, the number of cells in the peritoneal
cavity began to increase (Chart 1/1); the population-dou
bling time was 8.3 hr between Day 2 and Day 3. The growth
curve plateaued at 4 to 5 x 10* cells/mouse and the host
usually died on the 7th day postinoculation. The initial lag
in the growth curve (Chart 1/4) does not reflect the presence
of dead cells in the inoculum since nigrosin exclusion tests
and TD60 assays indicated that essentially 100% of the
inoculated 7-day cells were viable and tumorigenic.
As shown in Chart \B, the 7-day tumor contained few S
cells (approximately 12%). However, when cells from a
tumor this age were transplanted into fresh mice there was a
lag of at least 16 hr and then a sustained rise in the
proportion of S cells. By Day 2, approximately 70% of the
cells were in S phase. The level remained high for a day or
so and then fell again as the tumor entered plateau phase.
The experimental data suggest a 2nd plateauing of the
labeling index between Days 4 and 5, but this was ignored in
constructing the curve in Chart Iß.The finding of an
extended lag period in the labeling index after transplanta
tion is reminiscent of our earlier results with Ehrlich ascites
tumors (11).
One possible explanation for the age-related lag period is
that the precursors for DNA synthesis are more severely
depleted in old tumors than in young ones. We are
investigating this possibility and can report that injection of
deoxyribonucleotides (9 nmoles each of deoxyadenosine,
deoxycytosine, and deoxyguanosine) directly into 7-day
P388 tumors has no effect on the rate of thymidine
incorporation, at least within the 1st 3 hr, in contrast to a
report for Ehrlich ascites tumors (8).
Labeled mitosis curves initiated 1 and 5 days after
inoculation of 10" cells indicate that the cycling cells, which
constituted only about one-third of the population on Day 1
(see below), had a mean cycle time of 8.5 hr on Day 1 and 22
hr on Day 5 (Chart 2). This lengthening of the cell cycle
during growth is caused by a doubling of S and G2 and
about a 20-fold increase in G, from 0.2 to 4.2 hr (Table 1).
This pattern is typical of many ascites tumors (9, 10, 13,
24).
From the data in Charts 1 and 2 we computed a growth
fraction of 0.36 for the Day 1 tumor (Table 1). The cellcycle time of the proliferating cells in this population was
8.5 hr and the essential identity of this figure and the
'Growth fraction is defined according to the standard mathematical
formulations used to calculate it (17). Since "noncycling" cells in ascites
tumors may be delayed in G, rather than entering a true G0 compartment
(7, 13), we will use "growth fraction" in the general sense, without
intending to imply that the "noncycling" cells necessarily halt completely
or enter a true G0.
u
o
tu
uj
«O
s?
DAY
Chart I. Population size (A) and proportion (B) of cells labeled by a
pulse of 3H-labeled thymidine during growth of the P388 leukemia in adult
DBA/2 mice. "H-Labeled thymidine was injected i.p. 30 min before
sacrifice. Each point shown with S.E. represents a minimum of 3 animals.
z
UJ
U
25
8
12
16
10
15
20
25
HOURS
Chart 2. Labeled mitosis curves for P388 tumors on the 1st and 5th
days of growth in adult DBA/2 mice. The inoculum was 10" cells: each
point represents a single mouse.
population-doubling time on Day 2 (8.3 hr) indicates that
the growth fraction rose to unity before the 48th hr after
transplantation. The labeling data in Chart \B suggest that
this occurred between Hr 16 and 48. As the population be
gan to enter plateau phase, the growth fraction declined
again, falling to 0.78 on Day 5 and to 0.36 on Day 7 (Table
1). This analysis is substantiated by repeated-labeling data,
JULY 1973
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1973 American Association for Cancer Research.
1781
J. W. Harris, B. Shon, and J. Meneses
Table 1
Cywkinelic parameters during growth of Ã-heP388 leukemia as an osciles tumor in DBA/2 mice
ofgrowth"135»Cell
Day
No.2.4
lossfactor000.35
ft
'C8.522.0To,0.24.2rs6.214.8To,1.63.0Growthfraction0.360.930.78Cell
10"57x
10'280x
x 10"Population-doublingtime(hr)19.211.336.0T
' Inoculum,
10' cells.
* Examination of the labeled mitosis curve for 5-day cells (Chart 2) shows that the trough in the curve occurs
in the region of 40 to 50% labeled cells and that the data do not precisely define the curve after the 15th hr.
Therefore, the durations of TVand 7",;,in the 5-day tumor must be regarded as only approximate (cf. Réf.
22).
wherein 100% of the cells in a 2-day tumor were labeled
within 1 cell-cycle time when 3H-labeled thymidine was
administered every 2 to 4 hr (unpublished data), whereas
5- and 7-day tumors reached a labeling index of only 78%
within 1 cell-cycle time and only 85% even after extended
exposure to 3H-labeled thymidine (Chart 3).
These and other cytokinetic parameters are summarized
in Table 1. The cell loss factor of 35% on Day 5 probably
reflects migration of tumor cells out of the peritoneal cavity
rather than cell death. This is supported by the observation
of Belli and Andrews (3) that there were P388 cells in spleen,
lymph nodes, and liver after the 3rd day of growth (3).
Relationship
between Tumor Growth Stage and
Radiosensitivity. Since the labeling index of the P388 falls
sharply between Days 3 and 7 (Chart 4) and, since S cells
are usually more radioresistant than G, cells (at least late
Gìcells), we investigated the radiation sensitivity of cells
harvested from tumors at these 2 extremes of age.
These experiments show that the response of P388 cells to
1000 rads of X-irradiation in vitro has a clear relationship to
the age of the tumor from which they were harvested (Chart
4). Cells harvested 3 days after inoculation of IO5 cells
(3H-labeled thymidine index, 70%) were approximately 2.7
times more sensitive than those harvested after 7 days
(3H-labeled thymidine index, 30 to 40%) and 7.5 times
more sensitive than those harvested 7 days after inocula
tion of 10" cells (3H-labeled thymidine index, 10%). In con
trast to this result at 1000 rads, 3- and 7-day cells appar
ently do not differ in their response to a lower dose (550
rads). This apparent discrepancy between the responses
to high and low doses is not unexpected; 550 rads is on
the shoulder of the survival curve for these cells (3), and
even large differences in sensitivity that are reflected in
slope changes are often indiscernible in this region (cf. Fig.
1 of Ref. 5), whereas 1000 rads is well down on the linear
portion of the curve.
DISCUSSION
The P388 lymphocytic leukemia growing as an ascites
tumor in DBA/2 mice exhibits many of the characteristics
that are common to other ascites tumors (e.g., Refs. 9, 13,
16, and 24): the cell cycle elongates during growth (primar
ily because of lengthening of GO, the growth fraction
declines, and the cell loss factor increases (at least in late
growth). After transplantation the tumor exhibits a lag
1782
TdR- H INJECTED
A4
EVERY:
HOURS
•¿
8 HOURS
16
24
32
40
48
HOURS
Chart 3. Repeated labeling of P388 leukemia cells with sH-labeled
thymidine (TdR-3H) between the 5th and 7th days of growth in adult
DBA/2 mice. 3H-Labeled thymidine was given every 4 or 8 hr, beginning
on the 5th day of growth. Each point represents the mean of 2 to 5 mice.
period of about 16 hr before cells reenter the cell cycle and
synthesize DNA. We previously hypothesized (11) that this
lag period reflects the time needed for resynthesis of some
macromolecule (e.g., a protein) that has decayed below a
critical concentration during prolonged residence in plateau
phase. We know little about this process or the nature of
these hypothetical molecules, except that the process appar
ently uses preexisting mRNA and is at least partially
oxygen dependent (11).
Shortening of the P388 cell cycle after transplantation
precedes the increase in the growth fraction. Indeed, the
growth fraction is actually higher in early plateau phase
(Day 5) than during the 1st day of growth, even though the
cell cycle time of the proliferating cells in the old tumor is
2.5 times greater than in the young one. This analysis
indicates that cell-cycle time and the growth fraction change
independently of each other in the P388, despite their
coordinated changes in some tumors (13).
Perhaps the most significant result of these experiments is
CANCER
RESEARCH
VOL. 33
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1973 American Association for Cancer Research.
Growth and Radiosensitivity of P388 Leukemia
radiosensitivity of cells grown for 1 or 7 days in vivo. This
disagreement is only apparent, however; Belli and Andrews'
400
growth curve and the fact that their animals did not die until
Day 12 both indicate that their study encompassed only
exponential and very-early-plateau-phase cells (3). Our
cytokinetic data suggest that only minimal differences in the
distribution of cells within the cell cycle would be expected
to occur during this interval. In our study, which compares
exponential and late-plateau-phase cells, a difference in
radiosensitivity is evident (Chart 4). Since these lateplateau-phase cells may be typical of those found within
large solid tumors, we are extending our studies. Prelimi
nary results suggest that the "biochemical status" of
late-plateau-phase cells, rather than their physical position
in the cell cycle, is responsible for their radioresistance.
200
f~
100
80
o
&
60
40
20
O
oc
ni
CD
S
CONTROL
10
8
ó
550
RADS
/
1000 RADS
Z
ACKNOWLEDGMENTS
l
0.8
0.6
0.4
We acknowledge with thanks the expert assistance of N. Allen.
8.3 1.4 0.5 0.2
PERCENT SURVIVAL
l
3456
8
DAY
Chart 4. Response of 3-day and 7-day P388 leukemia cells to Xirradiation. Cells were irradiated in vitro under aerobic conditions and then
injected (10s) into unirradiated DBA/2 mice. Euch point represents 3 to 19
animals; bars, standard errors. The control intercept (0.48 x IO6cells) is
taken as 100% survival since all cells in the control inoculum were viable.
A, control; •¿,
3 days after inoculum of IO5;O, 7 days after inoculum of
IO5;and A, 7 days after inoculum of 10*.
the observation that cells harvested from late-plateau-phase
tumors are more radioresistant than cells from exponen
tially growing tumors (Chart 4). Since the validity of this
conclusion rests on the assumption that extrapolation of the
growth curves is a valid measure of cell survival in this
tumor [it is not for all tumors (e.g., Ref. 20)], it should be
emphasized that the survival percentage obtained for cells in
exponential growth (Chart 4) is nearly identical to that
estimated by the TD50 technique (3, 4, 12).
Our finding that cells from late-plateau-phase tumors are
significantly more resistant to irradiation in air than are
cells from exponential tumors is consistent with results
obtained by Hahn and Bagshaw (10) with Chinese hamster
cells in vitro. They observed that cells accumulated in
plateau-phase Gt by growth in low-serum medium were
more resistant to X-rays than were exponentially growing
cultures ( 10). Chinese hamster cells blocked in an analogous
GÌ
(or GO) position by anoxia are also significantly more
radioresistant than normal G, cells (15). Taken together,
these data suggest that cells deprived of oxygen and/or
essential nutrients stop cycling at some point after mitosis
and before S phase and that in this "quasi-Gi" (or G0) state
they are more resistant to X-radiation than are normal G i
cells. The implications of this resistance for the radioresponsiveness of tumors containing appreciable numbers of these
cells are obvious.
In apparent contrast to our results, earlier studies of the
P388 by Belli and Andrews (3) disclosed no difference in the
JULY
REFERENCES
1. Apple, M. A., Ludwig, F. C., and Greenberg, D. M. Selective Cancer
Growth Inhibition in Mice by Dihydroxypropanol without Con
comitant Inhibition of Bone Marrow or Other Normal Tissue. Oncol
ogy, 24:210-222, 1970.
2. Barrett, J. C. A Mathematical Model of the Mitotic Cycle and Its
Application to the Interpretation of Percentage Labeled Mitoses
Data. J. Nati. Cancer Inst., 57:443 450, 1966.
3. Belli, J. A., and Andrews, J. R. Relationship between Tumor Growth
and Radiosensitivity. J. Nati. Cancer Inst., 31:689 703. 1963.
4. Belli, J. A., Dicus. G. J., and Bonté,F. J. Radiation Response of
Mammalian Tumor Cells. I. Repair of Sublethal Damage In Vivo.
J. Nati. Cancer Inst., 38: 673 682, 1967.
5. Berry, R. J. Hypoxie Protection and Recovery in Tumour Cells Ir
radiated at Low Dose-rates and Assessed in Vivo. Brit. J. Radio!.,
41: 921 -926, 1968.
6. Berry, R. J., and Stedeford, J. B. H. Reproductive Survival of Mam
malian Cells after Irradiation at Ultrahigh Dose Rates: Further Ob
servations and Their Importance for Radiotherapy. Brit. J. Radiol.,
45: 171 177, 1972.
7. Burns, F. J., and Tannock, I. F. On the Existence of a G0-Phase in the
Cell Cycle. Cell Tissue Kinet., 3: 321 334, 1970.
8. Cory, J. G., and Whitford, T. W. Ribonucleotide ReducÃ-aseand DNA
Synthesis in Ehrlich Ascites Tumor Cells. Cancer Res., 32:
1301 1306, 1972.
9. Frindel, E., Valieron, A. J., Vassort, F., and Tubiana, M. Proliferation
Kinetics of an Experimental Ascites Tumour of the Mouse. Cell Tissue
Kinet., 2: 51 65, 1969.
10. Hahn, G. M., and Bagshaw, M. A. Serum Concentration: Effects on
Cycle and X-ray Sensitivity of Mammalian Cells. Science, 151:
459-461, 1966.
11. Harris, J. W., Meyskens, F., and Patt, H. M. Biochemical Studies of
Cytokinetic Changes during Tumor Growth. Cancer Res., 30:
1937 1946, 1970.
12. Harris, J. W., and Phillips, T. L. Radiobiological and Biochemical
Studies of Thiophosphate Radioprotective Compounds Related to
Cysteamine. Radiation Res., 46: 362-379, 1971.
13. Lala, P. K. Cytokinetic Control Mechanisms in Ehrlich Ascites
Tumour Growth. In: Effects of Radiation on Cellular Proliferation
and Differentiation, pp. 463-474. Vienna: International Atomic
Energy Agency, 1968.
1973
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1973 American Association for Cancer Research.
1783
J. W. Harris, B. Shon, and J. Meneses
14. Kaltenbach, J. P., Kaltenbach, M. H., and Lyons, W. B. Nigrosin as a
Dye for Differentiating Live and Dead Ascites Cells. Exptl. Cell Res.,
15: 112 117, 1958.
15. Koch, C. J., Kruuv, J., Frey, H. E., and Snyder, R. A. Plateau Phase
in Growth Induced by Hypoxia. Intern. J. Radiation Biol., 23: 67-74,
1973.
16. Lala, P. K., and Patt, H. M. Cytokinetic Analysis of Tumor Growth.
Proc. Nati. Acad. Sei. U. S., 56: 1735-1742, 1966.
17. Mendelsohn, M. L. Autoradiographic Analysis of Cell Proliferation in
Spontaneous Breast Cancer of C3H Mouse. III. The Growth Fraction.
J. Nati. Cancer Inst., 28: 1015 1019, 1962.
18. Moroson, H., Schmid, M., and Furlan, M. Enhancement of Murine
Ascites Tumor Cell Killing with X-Irradiation by Thiol Binding
Agents. Radiation Res., Jo: 571-579, 1968.
1784
19. Quastler, H., and Sherman, F. G. Cell Population Kinetics in the
Intestinal Epithelium of the Mouse. Exptl. Cell Res., 17: 420-438,
1959.
20. Révész,
L. Effect of X-Irradiation on the Growth of the Ehrlich
Ascites Tumor. J. Nati. Cancer Inst., 15: 1691-1701, 1955.
21. Steel, G. G. Cell Loss from Experimental Tumours. Cell Tissue Kinet.,
/. 193 207, 1968.
22. Steel, G. G. The Cell Cycle in Tumours: An Examination of Data
Gained by the Technique of Labelled Mitoses. Cell Tissue Kinet., 5:
87 100, 1972.
23. Steel, G. G., and Lamerton, L. F. The Growth Rate of Human
Tumours. Brit. J. Cancer, 20: 74-86, 1966.
24. Tannock, I. F. A Comparison of Cell Proliferation Parameters in
Solid and Ascites Ehrlich Tumors. Cancer Res., 29: 1527-1534, 1969.
CANCER
RESEARCH
VOL. 33
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1973 American Association for Cancer Research.
Relationship between Growth and Radiosensitivity in the P388
Murine Leukemia
John W. Harris, Brian Shon and Juanito Meneses
Cancer Res 1973;33:1780-1784.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/33/7/1780
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1973 American Association for Cancer Research.