•
•
Repair of Potentially Lethal Radiation Damage in Acute and Chronically Hypoxic Tumor Cells in Vivo 1
Radiation Biology
Muneyasu Urano, M.D., Naohumi Nesumi, M.D., Koichi Ando, D.D.S., Mrs. Sachiko Koike, and Naomi Ohnuma, M.D. The ability of animal tumor cells to repair potentially lethal damage was studied in vivo. Fifth-generation isotransplants of a spontaneous mouse squamous-cell carcinoma were irradiated under tourniquet-induced hypoxia or in air. Tumors were removed either immediately or 6 hours after irradiation and dose-response curves were determined by TD so assays. _Repair was attributed to cells in the hypoxic cell component for animals irradiated in air. Extensive repair was also noted for those irradiated under hypoxic conditions. Implications of these results are discussed. INDEX TERMS: Hypoxia. Neoplasms, experimental. Radiations, injurious effects, neoplastic • Radiobiology, cell and tissue studies
Radiology 118:447-451, February 1976
• potentially lethal radiation damage was first demonstrated by Philips and Tolmach in cultured mammalian cells (13). Later studies have been made extensively in stationary-phase cells (4, 5, 10, 11), and recently this kind of repair was observed in animal tumor cells in vivo (3, 6, 12). It is well known that tumors are composed of oxygenated (radiosensitive) and oxygen-deficient (radioresistant) tumor cells, the latter being a critical factor for tumor control by ionizing radiations (8). One study suggests that several postirradiation conditions characteristic of hypoxic cells in vivo might facilitate the repair of potentially lethal damage (5). If hypoxic tumor cells were capable of this kind of repair, it might have a practical application in a clinical situation such as fractionated radiotherapy. Therefore, experiments were carried out to examine the capability of two kinds of hypoxic tumor cells (17) to repair radiation damage. One type is chronically hypoxic cells and the other is acute hypoxic cells which are made hypoxic only at the time of irradiation.
R
EPAIR
OF
MATERIALS AND METHODS
Experimental animals were 8- 1O-week-old C3Hf /He mice which have been maintained in our Institute as specific pathogen-free (SPF) colonies at constant temperature and humidity. Animals were provided with sterilized Purina pellets and chlorinated water ad libitum. Fifth-generation isotransplants of a squamous-cell carcinoma which arose spontaneously in a C3Hf/He female mouse were used. The method of obtaining these tumor cells was similar to that described for C3H mouse mammary carcinoma (16). Briefly, first- to third-generation tumors were stored in liquid nitrogen and fourthgeneration tumors were grown in flank tissues of female mice. This tumor is designated as NR-S 1 by us and the cell cycle time was found to be approximately 18 hours. Tumor cell suspension was prepared as follows. Ani-
mats with fourth-generation isotransplants were sacrificed by cervical dislocation. Intact tumor tissue was excised, minced by scissors, and trypsinized at 37°C for 30 minutes in a flask which held 0.2 % trypsin (Difco, 1: 250) in Ca- and Mg-free Dulbecco's solution. It was then removed, put into crushed ice, and allowed to settle for 15 minutes. Approximately two-thirds of the supernatant was removed by a syringe, divided into several test tubes, and centrifuged at 1,600 rpm for 5 minutes. The sediment was resuspended with a sufficient amount of Hanks' medium (containing 5 % fetal calf serum) for transplantation and injected into the subcutaneous tissue of the right thigh. Viable tumor cells were counted in a hemocytometer by use of the trypan-blue staining method. Tumors were irradiated with 200-kVp x rays as described elsewhere (23). Physical factors were: h.v.1. 0.8 mm Cu, target-tumor distance 15 em, and dose rate approximately 700 rads/min. Irradiation was given in air or under hypoxic conditions produced by applying a heavy metal clamp at the proximal side of the tumor for 2 minutes before and during irradiation. Animals irradiated in air were breathing air at atmospheric pressure and without obstruction of the circulation of blood to the tumor. The animals were anesthetized by intraperitoneal administration of Nembutal (60 mg/kg) before irradiation. The response of well-oxygenated tumor cells to irradiation was determined as described in a previous paper (23). A single-cell suspension containing 10 7 viable tumor cells/ml was packed in a test tube, and O 2 was bubbled through at 3 I/min. for 20 minutes before and during irradiation. This suspension was irradiated at a TSD of 25 cm and a dose rate of 230 rads/min.; other physical factors were the same as for tumor irradiation. Cell survival was assayed by the TD so method, in which the number of tumor cells required for tumor regrowth in half the transplant sites is determined (21). Tu-
1 From the Division of Clinical Research, National Institute of Radiological Sciences, Chiba, Japan. Accepted for publication in August 1975. Supported in part by the Japanese Ministry of Education. elk
447
MUNEYASU URANO AND OTHERS
448
Hypoxia 2500rads
I
10
I
1/
24
Hours after Irradiation
Fig. 1. The increase of survival of NR-S1 tumor cells as a function of the time left in situ after. irradiation with 2,500 rads under hypoxic conditions. Open circles with vertical bars represent experi.. mental points with 95 % confidence limit. Average values are shown by solid circles at 6 and 24 hours.
c
.g
Dose; x K rads 2 3 4 1 O~_ _- - - - , r - - - - ' - - - - - - - i r - - - - - - - - - ,
IJ
~
u.
.~ 0
o Hypoxia .Air
:;
o Oxygen
Cl
'S
(./')
February 1976
radiated) cells. A dose-response curve was fitted to the data by computer and Do', the radiation dose required to reduce the surviving fraction to 1Ie on the linear portion of the dose-cell survival curve, and m, the extrapolation number, were determined (14). In a single TD 5 0 assay, approximately 10 male and 10 female mice were employed; 4-6 assays were usually performed simultaneously. The time required for half the irradiated tumors to regrow to a certain volume (TRT 50) was determined (22, 23). Three perpendicular diameters of each tumor were measured by caliper at least every other day. The tumors were assumed to have an ellipsoid shape, and the volume was calculated to be 1rabc/6 where a, b, and c are diameters. The logarithm of tumor volume was plotted as a function of time after irradiation. TRT was measured graphically for each tumor and TRT 50 was calculated by logit analysis. In the present study, TRT 50 to 500 mrn'' was analyzed and survival of tumor cells was estimated as follows. The difference between TRT 50 of irradiated tumors (TRT sox) and that of unirradiated tumors (TRT SON) should be equal to the product of the doubling time (Td ) of surviving tumor cells and the number of their doublings (N). Hence:
2
Cl
.s
-1
,
(TRT sox -
\ \ \
-2
\
=
N
If tumor cells are assumed to grow with a growth fraction of 1.0 without any delay following irradiation, Td is identical with cell generation time (Tg ) and the number of tumor cells produced within this time might be 2 N . Then the surviving fraction could be estimated as a reciprocal of 2 N • The accuracy of this type of assay will be discussed later.
~ ~ \~ \
\ \ \ \
l
TRT 'JON)/ T d
\
, \
Fig. 2. Dose-response curves of NR-S 1 tumor cells irradiated and assayed in vivo. Open and closed circles indicate cells irradiated under hypoxic conditions and in air. respectively. Dose-cell survival in cells irradiated in vitro in oxygen and assayed in vivo is represented by open squares. Vertical bars indicate the 95 % confidence limit.
mors were irradiated when they reached 8-9 mrn in diameter and were used to prepare a single-cell suspension containing a known number of viable tumor cells which was serially diluted two- to threefold by Hanks' medium into 6-8 doses. Each dilution was injected subcutaneously into both thighs. Recipients were given whole-body irradiation of 450 rads from a 137CS irradiator and were randomly arranged in groups 24 hours before transplantation. Tumor "takes" were examined by palpating the transplanted regions every 7 days. Tumors larger than 10 mm in diameter within 60 days after transplantation were scored as "takes" and TD 5 0 was calculated by the log it method. The surviving fraction was obtained from the ratio of TD 5 0 (control) to TD so (ir-
RESULTS
Tumors were irradiated with 2,500 rads under hypoxic conditions and excised for TD 5 0 assay at various times thereafter. Surviving fractions were progressively increased if tumors were allowed to remain in situ for more than 2 hours after a single irradiation dose and almost reached a plateau at 6 hours (Fig. 1). In fact, survival determined at 24 hours after irradiation was just comparable with the value obtained at 6 hours. Dose-response relations were examined for tumor cells irradiated in air or under hypoxic conditions and excised either immediately or 6 hours later. Dose-cell survival curves obtained immediately after irradiation are shown in Figure 2. Tumor cells irradiated under hypoxic conditions exhibited m of 2.3 and Do of 360 rads. The dose-response curve of tumor cells irradiated in air demonstrated that approximately 35 % of tumor cells were in the hypoxic fraction. In the same figure is a survival curve of well-oxygenated cells irradiated in vitro characterized by m of 10 and Do of 120 rads, indicating that the oxygen enhancement ratio (OER), if expressed as a ratio of Do (hypoxia) to Do (oxygenated), is 3.0, and
Vol. 118
449
REPAIR OF RADIATION DAMAGE IN TUMOR CELLS
that the extrapolation number is higher for oxygenated cells irradiated in vitro than for hypoxic cells irradiated in vivo. Dose-response curves of tumor cells removed 6 hours after a single irradiation exhibited an increase of Do (Fig. 3). This increase was characteristically observed in the resistant fraction if tumors were irradiated in air. It is of interest that the resistant fraction in this survival curve and that of tumor ceils removed immediately after irradiation in air could be extrapolated back to the same point on the ordinate. On the other hand, tumor cells irradiated under hypoxic conditions and left in situ for 6 hours exhibited the same increase in extrapolation number as that of Do if compared with those excised immediately after hypoxic irradiation. The extrapolation number was 5.0, i.e., approximately twice as large as that of tumor cells removed immediately after hypoxic irradiation, while the Do increased 1.25 times and was 450 rads. These results indicated that both acute and chronically hypoxic tumor cells were able to repair potentially lethal radiation damage, while the increase of the extrapolation number was exhibited only by the former. Regrowth curves of tumors receiving a single dose are illustrated in Figure 4, which demonstrates that tumors regrew rapidly soon after exposure without significant decrease in tumor volume. No tumor control was obtained by a dose of less than 5,500 rads under hypoxic conditions. Surviving fractions were calculated on the basis of TRT 50 analysis and form a concave curve as a function of radiation dose (Fig. 5). When dose-response curves of tumor cells obtained after the repair of potentially lethal damage (Fig. 3) were superimposed on the above curves, it is significant that the surviving fractions based on TRT 50 analysis were higher than those which might indicate that some other factor(s) than the repair of potentially lethal damage is still involved in volumetric changes of tumors following irradiation.
Dose; x K rads
Several factors which facilitate the repair of potentially lethal radiation damage in vitro have been studied extensively (4, 5, 10, 11, 13). In addition it has been demonstrated that both ascites and solid tumors are capable of this kind of repair (3, 6, 12) as well as repair of sublethal injury (2, 16). Survival of mouse fibrosarcoma cells was reported by Little et al. to be enhanced if tumors were allowed to remain in situ for several hours after irradiation; this was attributed to the tumor cells in the hypoxic fraction, i.e., chronically hypoxic cells (12). Our results agree with their findings. Both studies demonstrate that this repair is dose-dependent, i.e., chronically hypoxic cells left in situ following irradiation exhibited a larger Do than those excised immediately after treatment. These results are supported by the finding that density-inhibited cultured cells were capable of this kind of repair in vitro (5). The fact that this repair was
3
2
o
4
Hypoxia
• Air
-2
-3
-4 Fig. 3. Dose-response curves of NR-S 1 tumor cells left in situ for 6 hours after a single irradiation. Open and solid squares indicate cells irradiated under hypoxic conditions and in air, respectively. Thin lines are dose-response curves superimposed from Figure 2. Vertical bars represent the 95 % confidence limit.
mn,3
2000
3
1000
or
28
2J
500
I
I r .' .
• 1800 600raJS ..
0
24r 3000 .. L~~~~~
•
Repair of Potentially Lethal Radiation Damage in Acute and Chronically Hypoxic Tumor Cells in Vivo 1
Radiation Biology
Muneyasu Urano, M.D., Naohumi Nesumi, M.D., Koichi Ando, D.D.S., Mrs. Sachiko Koike, and Naomi Ohnuma, M.D. The ability of animal tumor cells to repair potentially lethal damage was studied in vivo. Fifth-generation isotransplants of a spontaneous mouse squamous-cell carcinoma were irradiated under tourniquet-induced hypoxia or in air. Tumors were removed either immediately or 6 hours after irradiation and dose-response curves were determined by TD so assays. _Repair was attributed to cells in the hypoxic cell component for animals irradiated in air. Extensive repair was also noted for those irradiated under hypoxic conditions. Implications of these results are discussed. INDEX TERMS: Hypoxia. Neoplasms, experimental. Radiations, injurious effects, neoplastic • Radiobiology, cell and tissue studies
Radiology 118:447-451, February 1976
• potentially lethal radiation damage was first demonstrated by Philips and Tolmach in cultured mammalian cells (13). Later studies have been made extensively in stationary-phase cells (4, 5, 10, 11), and recently this kind of repair was observed in animal tumor cells in vivo (3, 6, 12). It is well known that tumors are composed of oxygenated (radiosensitive) and oxygen-deficient (radioresistant) tumor cells, the latter being a critical factor for tumor control by ionizing radiations (8). One study suggests that several postirradiation conditions characteristic of hypoxic cells in vivo might facilitate the repair of potentially lethal damage (5). If hypoxic tumor cells were capable of this kind of repair, it might have a practical application in a clinical situation such as fractionated radiotherapy. Therefore, experiments were carried out to examine the capability of two kinds of hypoxic tumor cells (17) to repair radiation damage. One type is chronically hypoxic cells and the other is acute hypoxic cells which are made hypoxic only at the time of irradiation.
R
EPAIR
OF
MATERIALS AND METHODS
Experimental animals were 8- 1O-week-old C3Hf /He mice which have been maintained in our Institute as specific pathogen-free (SPF) colonies at constant temperature and humidity. Animals were provided with sterilized Purina pellets and chlorinated water ad libitum. Fifth-generation isotransplants of a squamous-cell carcinoma which arose spontaneously in a C3Hf/He female mouse were used. The method of obtaining these tumor cells was similar to that described for C3H mouse mammary carcinoma (16). Briefly, first- to third-generation tumors were stored in liquid nitrogen and fourthgeneration tumors were grown in flank tissues of female mice. This tumor is designated as NR-S 1 by us and the cell cycle time was found to be approximately 18 hours. Tumor cell suspension was prepared as follows. Ani-
mats with fourth-generation isotransplants were sacrificed by cervical dislocation. Intact tumor tissue was excised, minced by scissors, and trypsinized at 37°C for 30 minutes in a flask which held 0.2 % trypsin (Difco, 1: 250) in Ca- and Mg-free Dulbecco's solution. It was then removed, put into crushed ice, and allowed to settle for 15 minutes. Approximately two-thirds of the supernatant was removed by a syringe, divided into several test tubes, and centrifuged at 1,600 rpm for 5 minutes. The sediment was resuspended with a sufficient amount of Hanks' medium (containing 5 % fetal calf serum) for transplantation and injected into the subcutaneous tissue of the right thigh. Viable tumor cells were counted in a hemocytometer by use of the trypan-blue staining method. Tumors were irradiated with 200-kVp x rays as described elsewhere (23). Physical factors were: h.v.1. 0.8 mm Cu, target-tumor distance 15 em, and dose rate approximately 700 rads/min. Irradiation was given in air or under hypoxic conditions produced by applying a heavy metal clamp at the proximal side of the tumor for 2 minutes before and during irradiation. Animals irradiated in air were breathing air at atmospheric pressure and without obstruction of the circulation of blood to the tumor. The animals were anesthetized by intraperitoneal administration of Nembutal (60 mg/kg) before irradiation. The response of well-oxygenated tumor cells to irradiation was determined as described in a previous paper (23). A single-cell suspension containing 10 7 viable tumor cells/ml was packed in a test tube, and O 2 was bubbled through at 3 I/min. for 20 minutes before and during irradiation. This suspension was irradiated at a TSD of 25 cm and a dose rate of 230 rads/min.; other physical factors were the same as for tumor irradiation. Cell survival was assayed by the TD so method, in which the number of tumor cells required for tumor regrowth in half the transplant sites is determined (21). Tu-
1 From the Division of Clinical Research, National Institute of Radiological Sciences, Chiba, Japan. Accepted for publication in August 1975. Supported in part by the Japanese Ministry of Education. elk
447
MUNEYASU URANO AND OTHERS
448
Hypoxia 2500rads
I
10
I
1/
24
Hours after Irradiation
Fig. 1. The increase of survival of NR-S1 tumor cells as a function of the time left in situ after. irradiation with 2,500 rads under hypoxic conditions. Open circles with vertical bars represent experi.. mental points with 95 % confidence limit. Average values are shown by solid circles at 6 and 24 hours.
c
.g
Dose; x K rads 2 3 4 1 O~_ _- - - - , r - - - - ' - - - - - - - i r - - - - - - - - - ,
IJ
~
u.
.~ 0
o Hypoxia .Air
:;
o Oxygen
Cl
'S
(./')
February 1976
radiated) cells. A dose-response curve was fitted to the data by computer and Do', the radiation dose required to reduce the surviving fraction to 1Ie on the linear portion of the dose-cell survival curve, and m, the extrapolation number, were determined (14). In a single TD 5 0 assay, approximately 10 male and 10 female mice were employed; 4-6 assays were usually performed simultaneously. The time required for half the irradiated tumors to regrow to a certain volume (TRT 50) was determined (22, 23). Three perpendicular diameters of each tumor were measured by caliper at least every other day. The tumors were assumed to have an ellipsoid shape, and the volume was calculated to be 1rabc/6 where a, b, and c are diameters. The logarithm of tumor volume was plotted as a function of time after irradiation. TRT was measured graphically for each tumor and TRT 50 was calculated by logit analysis. In the present study, TRT 50 to 500 mrn'' was analyzed and survival of tumor cells was estimated as follows. The difference between TRT 50 of irradiated tumors (TRT sox) and that of unirradiated tumors (TRT SON) should be equal to the product of the doubling time (Td ) of surviving tumor cells and the number of their doublings (N). Hence:
2
Cl
.s
-1
,
(TRT sox -
\ \ \
-2
\
=
N
If tumor cells are assumed to grow with a growth fraction of 1.0 without any delay following irradiation, Td is identical with cell generation time (Tg ) and the number of tumor cells produced within this time might be 2 N . Then the surviving fraction could be estimated as a reciprocal of 2 N • The accuracy of this type of assay will be discussed later.
~ ~ \~ \
\ \ \ \
l
TRT 'JON)/ T d
\
, \
Fig. 2. Dose-response curves of NR-S 1 tumor cells irradiated and assayed in vivo. Open and closed circles indicate cells irradiated under hypoxic conditions and in air. respectively. Dose-cell survival in cells irradiated in vitro in oxygen and assayed in vivo is represented by open squares. Vertical bars indicate the 95 % confidence limit.
mors were irradiated when they reached 8-9 mrn in diameter and were used to prepare a single-cell suspension containing a known number of viable tumor cells which was serially diluted two- to threefold by Hanks' medium into 6-8 doses. Each dilution was injected subcutaneously into both thighs. Recipients were given whole-body irradiation of 450 rads from a 137CS irradiator and were randomly arranged in groups 24 hours before transplantation. Tumor "takes" were examined by palpating the transplanted regions every 7 days. Tumors larger than 10 mm in diameter within 60 days after transplantation were scored as "takes" and TD 5 0 was calculated by the log it method. The surviving fraction was obtained from the ratio of TD 5 0 (control) to TD so (ir-
RESULTS
Tumors were irradiated with 2,500 rads under hypoxic conditions and excised for TD 5 0 assay at various times thereafter. Surviving fractions were progressively increased if tumors were allowed to remain in situ for more than 2 hours after a single irradiation dose and almost reached a plateau at 6 hours (Fig. 1). In fact, survival determined at 24 hours after irradiation was just comparable with the value obtained at 6 hours. Dose-response relations were examined for tumor cells irradiated in air or under hypoxic conditions and excised either immediately or 6 hours later. Dose-cell survival curves obtained immediately after irradiation are shown in Figure 2. Tumor cells irradiated under hypoxic conditions exhibited m of 2.3 and Do of 360 rads. The dose-response curve of tumor cells irradiated in air demonstrated that approximately 35 % of tumor cells were in the hypoxic fraction. In the same figure is a survival curve of well-oxygenated cells irradiated in vitro characterized by m of 10 and Do of 120 rads, indicating that the oxygen enhancement ratio (OER), if expressed as a ratio of Do (hypoxia) to Do (oxygenated), is 3.0, and
Vol. 118
449
REPAIR OF RADIATION DAMAGE IN TUMOR CELLS
that the extrapolation number is higher for oxygenated cells irradiated in vitro than for hypoxic cells irradiated in vivo. Dose-response curves of tumor cells removed 6 hours after a single irradiation exhibited an increase of Do (Fig. 3). This increase was characteristically observed in the resistant fraction if tumors were irradiated in air. It is of interest that the resistant fraction in this survival curve and that of tumor ceils removed immediately after irradiation in air could be extrapolated back to the same point on the ordinate. On the other hand, tumor cells irradiated under hypoxic conditions and left in situ for 6 hours exhibited the same increase in extrapolation number as that of Do if compared with those excised immediately after hypoxic irradiation. The extrapolation number was 5.0, i.e., approximately twice as large as that of tumor cells removed immediately after hypoxic irradiation, while the Do increased 1.25 times and was 450 rads. These results indicated that both acute and chronically hypoxic tumor cells were able to repair potentially lethal radiation damage, while the increase of the extrapolation number was exhibited only by the former. Regrowth curves of tumors receiving a single dose are illustrated in Figure 4, which demonstrates that tumors regrew rapidly soon after exposure without significant decrease in tumor volume. No tumor control was obtained by a dose of less than 5,500 rads under hypoxic conditions. Surviving fractions were calculated on the basis of TRT 50 analysis and form a concave curve as a function of radiation dose (Fig. 5). When dose-response curves of tumor cells obtained after the repair of potentially lethal damage (Fig. 3) were superimposed on the above curves, it is significant that the surviving fractions based on TRT 50 analysis were higher than those which might indicate that some other factor(s) than the repair of potentially lethal damage is still involved in volumetric changes of tumors following irradiation.
Dose; x K rads
Several factors which facilitate the repair of potentially lethal radiation damage in vitro have been studied extensively (4, 5, 10, 11, 13). In addition it has been demonstrated that both ascites and solid tumors are capable of this kind of repair (3, 6, 12) as well as repair of sublethal injury (2, 16). Survival of mouse fibrosarcoma cells was reported by Little et al. to be enhanced if tumors were allowed to remain in situ for several hours after irradiation; this was attributed to the tumor cells in the hypoxic fraction, i.e., chronically hypoxic cells (12). Our results agree with their findings. Both studies demonstrate that this repair is dose-dependent, i.e., chronically hypoxic cells left in situ following irradiation exhibited a larger Do than those excised immediately after treatment. These results are supported by the finding that density-inhibited cultured cells were capable of this kind of repair in vitro (5). The fact that this repair was
3
2
o
4
Hypoxia
• Air
-2
-3
-4 Fig. 3. Dose-response curves of NR-S 1 tumor cells left in situ for 6 hours after a single irradiation. Open and solid squares indicate cells irradiated under hypoxic conditions and in air, respectively. Thin lines are dose-response curves superimposed from Figure 2. Vertical bars represent the 95 % confidence limit.
mn,3
2000
3
1000
or
28
2J
500
I
I r .' .
• 1800 600raJS ..
0
24r 3000 .. L~~~~~
