. Copwght
Oxi-30 I h/92 $5 00 t .W) 1~ lY92 Pergamon Press Ltd.
??Biology Original Contribution
THE RADIATION DOSE-RESPONSE RELATIONSHIP IN A HUMAN GLIOMA XENOGRAFT AND AN EVALUATION OF THE INFLUENCE OF GLUTATHIONE DEPLETION BY BUTHIONINE SULFOXIMINE E. C. HALPERIN, 0. Departments
M.D.,’ D. M. BRIZEL,
W. GRIFFITH,
PH.D.,~
M.D.,’
D. D. BIGNER,
G. HONORE, M.D.,
PH.D.~-~
PH.D.,’
PH.D.,’
AND H. S. FRIEDMAN~‘~
of ‘Radiation Oncology, *Pathology, 3Pediatrics, and the 4Preuss Laboratory University
M. R. SONTAG,
for Brain Tumor Medical Center, Durham, NC; and the ‘Department of Biochemistry, Cornell University Medical Center, New York, NY
Research,
Duke
We have used an extensively characterized human glioma cell line in an athymic mouse model to evaluate new therapeutic approaches for human supratentorial high grade gliomas. The tumor, D-54MG, is a subline of a human anaplastic glioma. Eight days after homozygous nu/nu BALB/c athymic mice received intracranial (IC) injections of a tumor homogenate, the whole brain was irradiated with either single fractions of 4,8,9, and 12 Gy or twice daily fractions, separated by least 6 hr, of 2.28 Gy X 2 or 7.53 Gy X 2. To evaluate whether or not glutathione depletion influenced animal survival, animals at each dose level received either intraperitoneal (IP) buthionine sulfoximine (BSO) alone or I.P. BSO plus BSO in the drinking water. There was a stepwise prolongation of animal survival with increasing doses of external beam radiation. Mean survival in 9 of the 10 control groups (8-12 animals per group) ranged from 14.1 to 18.8 days. Mean survival ranged from 15.3 to 22.5 days at 4 Gy, 25 to 30 days at 8 Gy, 22.3 to 29.7 days at 9 Gy, and 32.9 to 33.6 days at 12 Gy single dose irradiation. At 2.28 Gy X 2 split dose irradiation mean survival was 29.3 days, for 7.53 Gy X 2 mean survival was over 47 days. The data for single fraction irradiation fit a linear regression line (r = 0.908) of mean animal survival = (1.22 [dose in Gy] + 16.7) days. Tumor GSH levels were decreased with all BSO dosing regimens tested. The most aggressive regimen (I.P. BSO + oral BSO for 5 days), reduced tumor GSH to 6.2% of control. Increased survival in irradiated glutathione depleted mice versus mice receiving radiation alone was not seen. Brain neoplasms, Buthionine Sulfoximine, Glutathione,
Radiation therapy.
INTRODUCTION
Agents which lower GSH levels in cells and tissues have greatly facilitated studies of the inter-relationships of ionizing radiation and GSH ( 12). The compound buthionine sulfoximine (BSO) is able to deplete intracellular GSH levels (23-24). We have reported that following intraperitoneal (I.P.) administration of BSO in athymic nude mice previously injected intracranially with an anaplastic glioma cell line, depletion of tumor GSH occurs without any decrease of GSH levels in the contralateral normal brain (50). Such experiments suggested that BSO-mediated radiosensitization via GSH depletion, if it exists, might be confined to an intracranial tumor without affecting the radiation tolerance of normal brain. Following establishment of an external beam radiation therapy dose response curve for a human glioma xenograft model, we investigated the influence of BSO on radiosensitivity. To
The treatment of humans with supratentorial high grade gliomas is highly unsatisfactory. The long term survival for individuals afflicted with glioblastoma multiforme is less than 5% while that for individuals with anaplastic astrocytoma is approximately 20% (6, 28, 33, 55). Hopefully, rational therapeutic strategies based upon experimental models may improve these results. One such strategy is to influence tumor cell radiosensitivity by depletion of intracellular glutathione (GSH). It is clear that GSH affords protection against a wide variety of cytotoxic agents. Previous in vitro studies have shown that GSH depletion by itself results in mild enhancement of cellular radiosensitivity (3, 11, 13, 17, 25, 36-40,45,47, 53-54).
work was supported by NIH grants CA 44640, CA 11898, NS 20023, NS 00958, and DK 26912 as well as American Cancer Society grant CH-403 and a Career Development Award (ECH) and Bristol Meyer’s grant lOO-RI 8. Accepted for publication 22 January 1992.
Reprint requests to: Edward C. Halperin, M.D., Department of Radiation Oncology, Box 3085, Duke University Medical Center, Durham, NC 277 10. Acknowledgements-Cheryl Wallace, Donna Wimberley, and Ruth Aultman typed the manuscript; Ann Tamariz edited it. Dr. Scott Clegg assisted in the linear regression analysis. This 103
104
I. J. Radiation Oncology 0 Biology 0 Physics
our knowledge, this represents the first evaluation of BSO and external beam radiotherapy in a human glioma xenograft model. METHODS
AND MATERIALS
Animals Homozygous nu/nu BALB/c athymic mice at least 6 weeks old were used for all experiments. The animals were from the independent breeding colony maintained at Duke University. They were maintained as previously described (46). Xenograji establishment The tumor line, D-54MG, is the Duke University subline of the human anaplastic glioma line A- 172. The morphology, biochemistry, antigenicity, and karyotype of D54MG have been extensively investigated (4, 5, 22). Its sensitivity in athymic mice to a variety of chemotherapeutic agents has been determined (47). The xenografts were grown subcutaneously in athymic mice in a manner previously described (7). Tumor homogenate was prepared with a 60-mesh tissue cytosieve. Xenograji implantation Tumor was washed through the cytosieve with minimal essential medium and the cells were concentrated by centrifugation. The resulting homogeneous tumor suspension was mixed with an equal volume of 7% methyl cellulose* and the mixture was loaded into a 250-ul Hamilton syringe. Animals were given injections of 5 ul of tumor suspension into the right cerebral hemisphere with a 27 gauge needle equipped with a sleeve allowing 4.5 mm penetration. Animals were checked daily for neurological symptoms and death. Control animals received tumor injections and no therapy. Radiation therapy technique Following anesthesia with I.P. pentobarbital (50 mg/ kg) the animals were irradiated to the entire head with a single dorsal photon field from a 4 MeV linear accelerator+. The dose was prescribed to the skin surface between the eyes of the animals. The animals were placed on a 30 X 30 X 10 cm polystyrene block to provide scattering material, and 1 cm bolus was placed on the skin to ensure full dose. The target to skin distance was 80 cm. For single fraction radiotherapy, total doses of 4, 8, 9, and 12 Gy at the prescription point were administered at 200 cGy/min. For two-fraction irradiation, doses equivalent to 4 Gy X 1 and 12 Gy X 1 were determined using a model of E/a = nd (I + d/a//3) with E = equivalent dose, n = number of fractions, d = dose/fraction, and a and p being the factors of the linear-quadratic model of radiation killing.
* Fluka, Switzerland. +Clinic 4, Varian Associates, Palo Alto, CA.
Volume 24. Number I. 1992
We assumed that acutely reacting tumor had an a/B = 10 (18, 19). Thus, 4 Gy X 1 was equivalent to 2.28 Gy X 2 with fractions separated by at least 6 hr and 12 Gy X 1 was equivalent to 7.53 Gy X 2, also with fractions separated by at least 6 hr. BSO regimen/therapy Animals were irradiated 8 days following I.C. injection of the tumor homogenate. In those animals receiving BSO, the I.P. injections of the drug began on Day 3 following tumor injection. In experiments 1 through 4 D,L-BSO (22.2 mg/ml, molecular weight 222.3)* dissolved in normal saline, was given by I.P. injection (2.5 mmol/kg) starting five days prior to irradiation. Doses of D,L-BSO were given at 0, 12, 24, 48, 72, and 96 hr in experiments 1 through 4. In experiments 5 and 6, L-BSO (synthesized as previously described, (24)) at the same concentration, was administered by I.P. injection, starting 3 days prior to irradiation, every 12 hr for seven doses and was also available to the animals prior to and following irradiation in acidified drinking water (pH 3.0) at a concentration of 20 mM. Since only the L isomer is active, the L-BSO regimen represented an increase in both the number of injections and in the active dose per injection. Irradiation took place 12 hr after the last injection. In experiments 7 through 10, L-BSO was administered every 12 hr for 2 days prior to irradiation in addition to being available in the animals’ drinking water. Survival Animals were examined each day. Survival was measured from the date of I.C. injection of the tumor homogenate until the date of death. Death was due to localized tumor growth without distant metastases. Gluthathione assay GSH levels in I.C. xenografts (Day 12 after implantation) were measured as previously described (50) by the method of Tietze (52) using procedures and solutions described by Griffith (23). Groups of five animals with I.C. D-54MG xenografts were treated with D,L-BSO at the above dose schedules for 6 I.P. doses and compared to previously published values using L-BSO for 4 or 7 I.P. doses plus concomitant availability in drinking water (33). Control GSH levels were measured in 10 xenografts from untreated animals. Statistical evaluation Statistical significance was assessed by the Wilcoxon rank sum test for median survival (56). The single fraction radiation dose response data was fitted, using a linear regression analysis, to a best fit line y = mx + b.
f Sigma Chemical Company, St. Louis, MO.
105
Radiation response and BSO in a human glioma cell line 0 E. C. HALPERINet al.
Glutathione
BSO regimen*
(umol/g)+
0.48 + 0.14 (74.7)* 0.15 + 0.04 (92.3)” 0.12 -+ 0.12 (93.8)”
DL-BSO (5 mmol/kg X 6) L-BSO (2.5 mmol/kg X 4) L-BSO (2.5 mmol/kg X 7)
in all experiments, but in no case was the effect of BSO as a single agent statistically significant. The use of single fraction external beam radiation therapy produced a stepwise increase in animal survival. The median survival in the five experimental groups receiving 4 Gy ranged from 15 to 22.5 days. In the two experimental groups receiving 8 Gy the median survival was 28 days. Median survival for the four groups receiving 9 Gy ranged from 22 to 29 days. The median survivals for the two groups receiving 12 Gy was 32 and 33 days. (Tables 24). The single fraction radiation dose-response data in Figure 1 shows the trend towards increased survival with increasing dose. The mean survival data are moderately well fitted (r = 0.908) by a regression analysis to a y = mx + b line with y = [ 1.22 (dose in Gy) + 16.771 days. Fractionated irradiation significantly increased survival compared to control animals. (Table 4). The median survival for 2.28 Gy X 2 was similar to those obtained for 4 Gy X 1 (2 1 days v. 15-22 days). The survival of animals receiving 7.53 Gy X 2 was longer than those receiving 12 Gy X 1. The groups are not, however, comparable because of the unusually long survival of the control animals in experiment ten. For each of the radiation doses tested, comparative experiments were done following either D,L-BSO or L-BSO mediated GSH depletion. In experiments 2 through 4, using a BSO dose regimen producing a modest depletion of GSH levels in the I.C. xenografts, no appreciable increase in the median survival of the animals was seen. Administration of BSO regimens effecting greater depletions of GSH in the xenografts (experiments 5 through 10) also failed to produce results which reproducibly showed BSO-mediated radiosensitization. In general, pretreatment of mice with BSO resulted in no significant increase in survival in animals subsequently treated with irradiation compared to mice receiving radiotherapy alone. In portions of some experiments (experiments 4-8) mice receiving BSO had slight increases in mean or median survival compared to controls or mice receiving radiation alone. These observations were not confirmed by the majority of experiments,
imal survival
Table 1. D-54 MG xenograft glutathione values following administration of BSO
* DL-BSO was given by I.P. injection for 6 doses (5 mmol/ kg) at 0. 12, 24, 48, 72, and 96 h. L-buthionine-SR-sulfoximine was given by I.P. injection for 4 or 7 doses (2.5 mmol/kg) at 12h intervals, with concomitant availability in acidified drinking water (pH 3.0) at a concentration of 20 mM. Values shown for comparitive analysis (33). 7 Mean + S.D. Number in parenthesis, depletion as a percent of control from five measurements. t Control glutathione 1.89 AZ0.13 umol/g. 4Control glutathione 1.94 + 0.09 umol/g (33).
RESULTS The GSH concentration (mean + S.D.) in D-54MG glioma xenografts 12 days after tumor inoculation was 1.89 r 0.13 umol/g wet weight (Table I), similar to pre-
viously published values (33). Treatment with D,L-BSO (six doses) resulted in depletion of GSH to 0.48 f 0.14 umol/g wet weight. Previous studies have shown that treatment with LBSO (four doses I.P. plus administration in drinking water) resulted in depletion of glutathione levels to 0.15 + 0.04 umol/g wet weight. More extensive treatment with further L-BSO (seven I.P. doses plus administration in drinking water) resulted in depletion of GSH levels to 0.12 +- 0.12 umol/g wet weight (6.2% of controls) (33). In 9 of the 10 experiments, using between 8 and 12 animals per experimental group, there was excellent reproducibility of the survival of control animals. The mean survival ranged from 14.1 t 3 days to 18.8 +- 1.5 days while the median survival ranged between 14.5 and 19 days. (Tables 2-4). In experiment 10, survival of controls was prolonged (mean 34.3 k 2.6 days)-perhaps because of cells being used from an earlier serial passage. The use of either D,L-BSO or L-BSO alone slightly prolonged an-
Table 2. Survival in days of animals receiving radiotherapy alone or radiotherapy plus I.P. D,L-BSO 1
2
Experiment number
Mean
Median
Mean
Therapy control BSO
18.6 -t 2.6 20.6 I!I 2.5
19 22
18.5 + 2.2 21.3 ? 7.4
3 Median
Mean
18 20
17.1 + 20.2 f
32* 29.5*
25 16.2 33.6 31.5
4 Median
2.3 2.4
17 20.5
t- 8 + 9 -+ 3.4 f 10
28* 9.5 33* 33*
4Gy 4 Gy/BSO
8 GY 8 Gy/BSO 12Gy 12 Gy/BSO * p < 0.05 v. control. n = 8-12 animals for each value shown.
32.9 f 4.3 30.1 + 7.7
Mean 18.8 21.3 22.5 22.3 30 25.6
* + f + f +
1.5 5.5 3.6 2.7 6.1 8.8
Median 19 20 22 23 28* 29
106
I.J.
Radiation
Oncology
0 Biology
0 Physics
Volume
Table 3. Survival in days of animals receiving radiotherapy alone or radiotherapy plus seven doses of I.P. L-BSO plus LBSO in drinking
Mean
Therapy control BSO 4Gy 4 Gy/BSO 9Gy 9 Gy/BSO
14.1 14.5 15.3 17.8 22.6 27.5
+ * f + f +
6 Median
1.3 1.8 0.5 1.1 2.5 1.9
Mean 17.3 17.2 20.8 21.2 28.1 22.1
14.5 15 15 18*,+ 22* 27*,+
* p < 0.05 v. control. ‘p < 0.05 v. equal radiation dose without n = 7-l 1 animals for each value shown.
f f + f & I?
Median 1.7 3.1 1.8 4.2 3.8 9.5
17.5 18 21 21* 29* 27
BSO.
DISCUSSION Radiation therapy prolongs the survival of humans with supratentorial high grade gliomas (6, 28, 34, 55). There is some evidence of a clinical radiation dose response curve with an improvement in survival in individuals who receive greater than 55 Gy although this finding is not universally confirmed (55). Our experiments, with a human glioma xenograft model, demonstrate a linear single fraction radiation dose response curve. We have shown stepwise prolongation of animal survival with single fractions of radiation ranging from 4 to 12 Gy. Our relatively small number of data points does not allow us to completely exclude the possibility that the dose response curve is sigmoid. Radiation damage to DNA may include single and double strand breaks in the sugar-phosphate backbone of the molecule, alteration or loss of bases, formulation of cross-links between the DNA strands, or cross-links between the DNA and chromosomal proteins. It has been suggested that unrepaired or incorrectly repaired doublestrand breaks are the critical lesions (29, 42). Changes in
Table 4. Survival in days of animals receiving fractionated radiotherapy plus four doses of I.P. L-BSO in drinking 7 Experiment
number
Therapy control BSO 4Gy 4 Gy/BSO 2.28 x 2 = 4.56 Gy 2.28 x 2 = 4.56 Gy/BSO 9Gy 9 Gy/BSO 7.53 x 2 = 15.06 Gy 7.53 x 2 = 15.06 Gy/BSO
16.3 17.2 20.9 22.5
+ I + +
1.2 1.9 3.7 1.9
22.3 +- 5.6 23.7 +- 5.4
* p < 0.05 v. control. n = lo-12 animals for each value shown.
alone or fractionated water
8 Median
Mean
I. 1992
nuclear and cytoplasmic GSH levels may influence cellular radioresponsiveness (15, 16. 30). A large number of potential mechanisms for this phenomenon exist. GSH. the predominant non-protein intracellular thiol. has the potential to eliminate primary or secondary radiationinduced free radicals with hydrogen atom donation via one-electron reactions. GSH depletion may thus affect radiosensitivity because of changes in the ability of the cells to detoxify radiation-induced free radicals ( 1. 2. 1 i12, 14, 20. 27, 34-35, 4 l. 50). As a radioprotector via chemical reduction of radiation-induced radicals, GSH competes with oxygen and electron affinic hypoxic cell sensitizers for fixation and repair of these lesions. This competition model of thiol depletion may only apply at low oxygen concentrations because of reaction rate kinetics (8, 9). Thus, GSH depletion by BSO may increase the efficacy of hypoxic cell sensitizers. Nitroimidazoles. when incubated for a prolonged period of time at 37°C under hypoxic conditions, tend to be preferentially cytotoxic towards cells that are oxygen deficient. These compounds also tend to reduce the level of GSH and other nonprotein sulfhydryls in the cell. As a consequence of this thiol depletion. cells may attain heightened sensitivity to radiation and certain chemotherapy agents-particularly alkylating agents (27). BSO might increase the radiosensitizing effects of nitroimidazoles by furthering thiol depletion. /I? vitv experiments have tended to show radiosensitization under conditions of extreme thiol depletion in combination with a nitroimidazole sensitizer and high doses of radiation ( 10, 2 I. 26,42. 57). Very recently. BSO has been shown to accentuate the radiosensitization of SR-2508 in a fractionated radiation model with clinically relevant doses (32). Glutathione may also be involved in the repair of critically damaged DNA sites. It is known. for example. that GSH depletion slows the repair of DNA-protein cross links produced by radiation (40). This inhibition of repair could explain the therapeutic significance of compounds that
water
5 Experiment number
24. Number
17 17 22.5 22.5
23 22
Mean 16.5 16.8 18.7 20.1
? f f f
1.7 1.6 2.8 2.3
29.7 * 16.7 25.3 + 4.6
radiotherapy
9
10
Median
Mean
Median
Mean
Median
16.5 17 19.5 20*
16.5 + 2.2 17 _+ 2.9
16 18
34.3 +- 2.6 34.9 I!Z4.2
34 33
29.3 +- 25.5 20.5 + 3.1
21* 22*
> 47 50.2 f 6.8
49* 50*
25* 25*
Radiation response and BSO in a human glioma cell line 0 E. C. HALPERINd al.
Dose
[Gy]
Fig. 1. Mean survival in days of single fraction irradiated nude mice with human glioma xenografts. Fitted to linear regression line survival = (1.22 [dose in Gy] + 16.77) days (r = 0.908).
intracellular GSH in the post-radiation period ( 12). Glutathione may also influence the activity of enzymes responsible for dealing with radiation damage, chemotherapy damage, and the transport of chemotherapeutic agents across membranes (49). Glutathione is also involved in the detoxification of radiation-induced H202 or organoperoxides. It should be noted that changes in GSH level, or in the activity of GSH-dependent enzymes, are expected to affect radiation damage or repair only if the alteration is sufficiently large. If the GSH level or enzyme activity greatly exceeds the amount required, substantial decreases may be without significant biological effect. One recent study, for example, found no effect of selenite-induced variation in glutathione peroxidase activity on radiation-induced cell killing or DNA strand breakage (45). Other possible mechanisms of radioprotection by GSH include direct restoration of altered DNA, reduction of DNA peroxyl radicals, DNA peroxides, or lipid peroxides, restoration by GSH of free radical altered radioprotectors (i.e., vitamin C or E), and/or maintenance of DNA-associated protein sulthydryls in the reduced form. We have demonstrated, in previous experiments, that administration of BSO to athymic mice bearing I.C. human glioma xenografts results in highly selective depletion of GSH in neoplastic tissue versus contralateral normal brain (37). Furthermore, treatment of mice bearing I.C. human medulloblastoma or glioma xenografts with melphalan plus BSO resulted in a significant increase in me-
deplete
107
dian survival over that produced by melphalan alone (20). Additionally, treatment of mice bearing intracranial D54MG xenografts with interstitial brachytherapy (I’*’ seeds) plus BSO produced a significant increase in median survival over that produced by brachytherapy alone (34). The current studies provide evidence that an assortment of BSO dosage regimens, given prior to external beam therapy, did not reproducably influence the survival of nu/nu BALB/c athymic mice with human glioma xenografts. These results are consistent with in vitro studies reporting low radiosensitization following BSO-mediated GSH depletion (3, 11, 13, 38-41, 4849, 5 1). In contrast to our previous studies using brachytherapy in GSH depletion (33) our current study did not demonstrate consistent increases in median survival in BSO-pretreated mice with human glioma xenografts receiving external beam therapy. Differences in the influence of BSO on animal survival, as a function of the form of radiation (single fraction teletherapy vs brachytherapy), may be explicable if an important influence of BSO on mammalian tumors is via inhibition of GSH-mediated repair of lowdose-rate radiation damage. If this were the case, BSOmediated GSH depletion might be more likely to favorably influence survival in experiments using fractionated rather than single dose teletherapy. Kramer ef al., however, have recently demonstrated that BSO treatment, by itself, did not effect RIF-1 or MCA tumor growth delay in C3H/ Sed mice following fractionated teletherapy (32). The lack of influence of BSO in our b.i.d. experiments (Table 4) supports Kramer’s findings. The finding that the radiosensitizing effects of BSO during brachytherapy are significant suggests the need to elucidate mechanisms of action which pertain during continuous low dose rate irradiation but not with fractionated external beam. It remains possible, of course, that the differing radioresponsiveness between brachytherapy and teletherapy may be due to our tumor xenografts being heterogeneous with respect to radiation response and that this heterogeneity, or at least some of it, is maintained during serial transplantation (43, 44). In summary, we have reported the development of a human glioma xenograft model which has a relatively reproducible duration of survival in nude mice. We have further shown a single fraction external beam radiation dose response curve in this model. Finally, we have shown that the administration of I.P. and oral BSO does not consistently influence the duration of animal survival.
REFERENCES Bernstein, M.; Gutin, P. H.; Deen, D. F.; Weaver, K. A.; Levin, V. A., Barcellos, M. H. Radiosensitization of the RIF- 1 murine flank tumor by desmethylmisonidazole (Ro 05 9963) during interstitial radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. 8:487-490; 1982. Biaglow, J. E.; Varnes, M. E.; Epp, E. R.; Clark, E. P.; Tuttle, S. W.; Held, K. D. Role of glutathione in the aerobic ra-
diation response. Int. J. Radiat. Oncol. Biol. Phys. 16: 13 1 l1314;1989. 3. Biaglow, J. E.; Varnes, M. E.; Tubble, S. W.; Oleinick, N. L.; Glazier, K.; Clark, E. P.; Epp, E. R.; Dethlefsen, L. A.: The effect of L-buthionine sulfoximine on the aerobic radiation response of A549 human lung carcinoma cells. Int. J. Radiat. Oncol. Biol. Phys. 12: 1139-l 142; 1986.
108
I. J. Radiation Oncology 0 Biology ??Physics
4. Bigner, D. D.; Bigner, S. H.; Ponten, B.; Westermark, B.; Mahaley, Jr. M. S.; Ruoslahti, E.; Herschman, H.; Eng, L. F.; Wikstrand, C. J. Heterogeneity of genotypic and phenotypic characteristics of fifteen permanent cell lines derived from human glioma. J. Neuropath. Exp. Neurology. 40: 201-229;1981. 5. Bigner, S. H.; Mark, J.; Bigner, D. D. Chromosomal composition of four permanent cultured cell lines derived from human gliomas. Cancer Genet. Cytogenet. 10:335349;1983. 6. Bloom, H. J. G. Intracranial tumors: response and resistance to therapeutic endeavors, 1970- 1980. Int. J. Radiat. Oncol. Biol. Phys. 8: 1083-I I 13. 7. Bullard, D. E.; Schold, S. C.; Bigner, S. H.; Bigner, D. D. Growth and chemotherapeutic response in athymic mice of tumors arising from human glioma-derived cell lines. J. Neuropath. Exp. Neurol. 40:4 10-427; 198 1. 8. Bump, E. A.; Brown, J. M. Role of glutathione in the radiation response of mammalian cells in vitro and in viva. Pharmac. Ther. 47: 117-I 36; 1990. 9. Bump, E. A.; Yu, N. Y.; Brown, J. M. Radiosensitization of hypoxic cells by depletion of intrace. Science 127:554555;1982. 10. Bump, E. A.; Yu, N. Y.: Taylor, Y. T.; Brown, J. M.; Travis, E. L.; Boyd, M. R. Radiosensitization and chemosensitization by diethylmaleate. In: Nygaard, O., Simic, M., eds. Radioprotectors and anticarcinogens. New York: Academic Press; 1983:297-323. biochemical modification of 1I. Clark, E. P. Thiol-induced chemo- and radioresponses. Int. J. Radiat. Oncol. Biol. Phys. 12:1121-l 126;1986. 12. Clark, E. P.; Epp, E. R.; Biaglow, J. E.; Morse-Gaudio, M.; Zachgo, E.: Glutathione depletion, radiosensitization, and misonidazole potentiation in hypoxic Chinese hamster ovary cells by buthionine sulfoximine. Radiat. Res. 98:370380;1984. 13. Dethlefsen, L. A.: Biaglow, J. E.; Peck, V. M.: Ridinger, D. N. Toxic effects of extended glutathione depletion by buthionine sulfoximine on murine mammary carcinoma cells. Int. J. Radiat. Oncol. Biol. Phys. 12: 1157-l 160;1986. 14. Durand, R. E. Roles of thiols in cellular radiosensitivity. Int. J. Radio]. Oncol. Biol. Phys. 10:1235-1238;1984. 15. Edgren, M. R. Nuclear glutathione and oxygen enhancement of radiosensitivity. Int. J. Radiat. Oncol. Biol. Phys. 51:36; 1987. depletion of 16. Edgren, M. R.; Revesz, L. Compartmentalised glutathione in cells treated with buthionine sulfoximine. Br. J. Radiat. 60:723-724;1987. 17. Evans, J. W.; Taylor, Y. C.; Brown, J. M. The role of glutathione and DNA strand repair in determining the shoulder of the radiation survival curve. Br. J. Cancer 49:49-53;1984. and therapeutic gain. In: G. G. 18. Fowler, J. F. Fractionation Steel, G. E. Adams, and A. Horwich, editors. The biological basis of radiotherapy, 2nd edition. Amsterdan: Elsevier; 1989:181-208. of brachytherapy. In A. A. 19. Fowler, J. The radiobiology Martinez, C. G. Orton, and R. F. Mould, editors. Brachytherapy: HDR and LDR. Columbia: Nucleotron Corporation; 1990:121-137. 20. Friedman, H. S.; Colvin, 0. M.; Griffith, 0. W.; Lippitz, B.; Elion, G. B.; Schold, S. C.; Hilton, J.; Bigner, D. D. Increased melphalan activity in intracranial human medulloblastoma and glioma xenografts following buthionine sulfoximine-mediated glutathione depletion. JNCI 8 1:524527;1989. 21. Fu, K. K.; Hurst, S.; Begg, A. C.; Brown, J. M.; The effects of misonidazole during continuous low dose rate irradiation. Cancer Clin. Trials 3:257-26;1980.
Volume 24. Number 1, lYY2 22. Giard, D. J.; Aaronson, S. A.; Todaro, G. J.; Amstein, P.; Kersey, J. H.: Dosile, H.; Parks, W. P. In vifrocultivation of human tumors: establishment of cell lines derived from a series of solid tumors. JNCI 1417-l 1423;1973. 23. Griffith, 0. W. Determination of glutathione and glutathione-disulfide using glutathione reductase and 2-vinylpyridine. Anal. B&hem. 106:207-212:1980. 24. Griffith, 0. W. Mechanism of action, metabolism and toxicity of buthionine sulfoximine and its higher homologs. Potent inhibitors ofglutathione biosynthesis. J. Biol. Chem. 257:137O&l3712;1982. 25. Griffith, 0. W.; Meister, A. Potent and specilic inhibition ofglutathione synthesis by buthionine sulfoximine (S-n-butylhomocystein sulfoximine). J. Biol. Chem. 254:75587560: 1979. 26. Guichard, M.; Lespinasse. F.; Malaise, E. P. Influence ot buthionine sulfoximine and misonidazole on glutathione level radiosensitivity of human tumor xenografts. Radiat. Res. 105:l 15-125:1986. 27. Hall, E. J.; Cox, J. D. Physical and biologic basis of radiation therapy. In: Moss, W. T., Cox, J. D., eds. Radiation oncology: rationale, technique, results, 6th edition. St. Louis: C. V. Mosby Company; 1987: l-57. 28. Halperin. E. C.; Burger, P. C.; Bullard, D. E. The fallacy of the localized supratentorial malignant ghoma. Int. J. Radiat. Oncol. Biol. Phys. 15:.505-509:1988. 29. Hill, R. P. Cellular basis of radiotherapy. In: Tannock, I. F., Hill, R. P., eds. The basic science of oncology. Tarrytown, NY: Pergamon Press; 1987:237-255. 30. Jones, E. L.: Douple, E. B. The effect of in viva GSH depletion on thermosensitivity, radiosensitivity and thermal radiosensitization. Int. J. Hyperther. 6:951-955;1990. 31. Kinsella. T. J.: Dobson, P. P.; Russo, A.: Mitchell, J. B.: Fornace, Jr., A. J. Modulation of x-ray damage by SR-2508 f buthionine sulfoximine. Int. J. Radiat. Oncol. Biol. Phys. 12:l 127-l 130:1986. 32. Kramer, R. A.; Sable, M.: Howes, A. E.; Montaya. V. P. The effect ofglutathione (GSH) depletion in vivoby buthionine sulfoximine (BSO) on the radiosensitization ofSR 2508. Int. J. Radiat. Oncol. Biol. Phys. 16:1325-1330;1989. 33. Leibel, S. A.; Sheline, G. E. Radiation therapy for neoplasms of the brain. J. Neurosurg. 66: I-22; 1987. 34. Lippitz, B. E.; Halperin, E. C.; Griffith, 0. W.; Colvin, 0. M.: Honore, G.: Ostertag, C. B.; Bigner, D. D.; and Friedman. H. S. L-buthionine-sulfoximine mediated radiosensitization in experimental intersistial radiotherapy of intracerebral D-54 MC glioma xenografts in athymic mice. Neurosurg. 26:255-260;1990. 35. Loh, S. N.; Dethlefsen, L. A.; Newton, G. L.; Aguilera, J. A.; and Fahey, R. C. Nuclear thiols: technical limitations on the determination of endogenous nuclear glutathionine and the potential importance of sulhydryl proteins. Radiat. Res. 121:98-196;1990. 36. Meister, A. Metabolism and transport of glutathione and other gamma-glutamyl compounds; functions of glutathione: biochemical physiological, toxicological, and clinical aspects. In: Larsson, A., Orrenius, S.. Holmgren, A.. Mannervick. B., eds. Functions of glutathione: biochemical. physiological, toxicological, and clinical aspects. New York: Raven Press: 1983. A. 1.; Rojas, A.: Smith, K. A.; Soranson, 37. Minchington. J. A.; Shrieve, D. C.; Jones, N. R.; Bremer, J. C. Glutathione depletion in tissues after administration of buthionine sulfoximine. Int. J. Radiat. Oncol. Biol. Phys. IO: 12611264;1984. 38. Mitchell, J. B.; Cook, J. A.; DeGraff, W.; Glatstein, E.; Russo, A. Keynote address: glutathione modulation in can-
Radiation response and BSO in a human glioma cell line 0 E. C. HALPERIN et a/.
39.
40.
41.
42.
43.
44. 45.
46.
47.
48.
cer treatment: will it work? Int. J. Radiat. Oncol. Biol. Phys. 16:1289-1296;1989. Mitchell, J. B.; Russo, A.; Biaglow, J. B.; McPherson, S. Cellular glutathione depletion by diethyl maleate or buthionine sulfoximine: no effect of glutathione depletion on the oxygen enhancement ratio. Radiat. Res. 96:422-428; 1982. Oleinick, N. L.; Liang-Yan, X.; Friedman, L. R.; Donahue, L. L.; Biaglow, J. E. Inhibition of radiation-induced DNAprotein cross-link repair by glutathione depletion with LButhionine Sulfoximine. NC1 Monogr. 6:225-229;1988. Oliver, J. M.; Spielberg, S. P.; Pearson, C. B.; Schulman, J. D. Microtubule assembly and function in normal and glutathione synthetase-deficient polymorphonuclear leukocytes. J. Immunol. 12O:l 181-1 186;1978. Rauth, A. M. Radiation carcinogenesis. In: Tannack, I. F., Hill, R. P., eds. The basic science of oncology. N.Y.: Pergamon Press; 1987: 106- 124. Rockwell, S.; Moulder, J. E. Hypoxic fractions of human tumors xenografted into mice: a review. Int. J. Radiat. Oncol. Biol. Phys. 19: 197-202; 1990. Rofstad, E. K. Human tumor xenografts in radiotherapeutic research. Radiother. Oncol. 3:35-46;1985. Sandstrom, B. E.; Carlsson, J.; Markland, S. L. Seleniteinduced variation in glutathione peroxidase activity of three mammalian cell lines: no effect on radiation-induced cell killing or DNA strand breakage. Radiat. Res. I 17:3 1832$1989. Schold, Jr. S. C.; Bullard, D. E.; Bigner, S. H.; Jones, T. R.; Bigner, D. D. Growth, morphology and serial transplantation of anaplastic human gliomas in athymic mice. J. NeuroOncol. 1:5-14;1983. Schold, Jr., S. C.; Friedman, H. S.; Bigner, D. D. Therapeutic profile of the human ghoma line D-54 MG in athymic mice. Cancer Treat. Rep. 7 1:849-850;1987. Shrieve, D. C.; Denekamp, J.; Minchington, A. I. Effects of glutathione depletion by glutathione sulfoximine on radiosensitization by oxygen and misonidazole in vitro. Radiat. Res. 102:283-294; 1985.
109
49. Shrieve, D. C.; Harris, J. W. Effects of glutathione depletion buthionine sulfoximine on the sensitivity of EMT6/SF cells to chemotherapy agents or X radiation. Int. J. Radiat. Oncol. Biol. Phys. 12:l 171-l 174;1986. 50. Skapek, S. X.; Calvin, 0. M.; Griffith, 0. W.; Groothius, D. R.; Colapinto, E. V.; Yisheng, L.; Hilton, J.; Elion, G.; Bigner, D. D.; Friedman, H. S. Buthionine sulfoximinemediated depletion of glutathione in intracranial human glioma-derived xenografts. Biochem. Pharm. 37:43 134317;1988. 51. Sun, J. D.; Ragsdale, S. S.; Benson, J. M.; Henderson, R. F. Effects of the long-term depletion of reduced glutathione in mice administered L-buthionine-S, R-sulfoximnie. Fund. Appl. Toxicol. 5:9 13-9 19; 1985. 52. Tietze, F. Enzymatic method for quantitative determination of nanogram amounts of total and oxidized glutathione: applications to mammalian blood and other tissues. Anal. Biochem. 27:502-522;1969. 53. Varnes, M. E.; Biaglow, J. E.; Roizin-Towle, L.; Hall, E. J. Depletion of intracellular GSH and NPSH by buthionine sulfoximine and diethyl maleate: factors that influence enhancement of aerobic radioresponse. Int. J. Radiat. Oncol. Biol. Phys. 10: 1229- 1233; 1984. 54. Vos, 0.; van der Schans, G. P.; Roos-Verhey, W. S. Reduction of intracellular glutathione content and radiosensitivity. Int. J. Radiat. Oncol. Biol. Phys. 50: 155-165;1986. 55. Walker, M. D.; Strike, T. A.; Sheline. G. E. An analysis of dose-effect relationship in the radiotherapy of malignant ghomas. Int. J. Radiat. Oncol. Biol. Phys. 5:17251731:1979. 56. Wilcoxon, F. Individual comparisons Biomet. Bull. 1:80-83;1945.
by ranking methods.
57. Yu, N. Y.; Brown, J. M. Depletion of glutathione in vivo as a method of improving the therapeutic ratio of misonidazole and SR 2508. Int. J. Radiat. Oncol. Biol. Phys. IO: 1265%1269;1984.
Oxi-30 I h/92 $5 00 t .W) 1~ lY92 Pergamon Press Ltd.
??Biology Original Contribution
THE RADIATION DOSE-RESPONSE RELATIONSHIP IN A HUMAN GLIOMA XENOGRAFT AND AN EVALUATION OF THE INFLUENCE OF GLUTATHIONE DEPLETION BY BUTHIONINE SULFOXIMINE E. C. HALPERIN, 0. Departments
M.D.,’ D. M. BRIZEL,
W. GRIFFITH,
PH.D.,~
M.D.,’
D. D. BIGNER,
G. HONORE, M.D.,
PH.D.~-~
PH.D.,’
PH.D.,’
AND H. S. FRIEDMAN~‘~
of ‘Radiation Oncology, *Pathology, 3Pediatrics, and the 4Preuss Laboratory University
M. R. SONTAG,
for Brain Tumor Medical Center, Durham, NC; and the ‘Department of Biochemistry, Cornell University Medical Center, New York, NY
Research,
Duke
We have used an extensively characterized human glioma cell line in an athymic mouse model to evaluate new therapeutic approaches for human supratentorial high grade gliomas. The tumor, D-54MG, is a subline of a human anaplastic glioma. Eight days after homozygous nu/nu BALB/c athymic mice received intracranial (IC) injections of a tumor homogenate, the whole brain was irradiated with either single fractions of 4,8,9, and 12 Gy or twice daily fractions, separated by least 6 hr, of 2.28 Gy X 2 or 7.53 Gy X 2. To evaluate whether or not glutathione depletion influenced animal survival, animals at each dose level received either intraperitoneal (IP) buthionine sulfoximine (BSO) alone or I.P. BSO plus BSO in the drinking water. There was a stepwise prolongation of animal survival with increasing doses of external beam radiation. Mean survival in 9 of the 10 control groups (8-12 animals per group) ranged from 14.1 to 18.8 days. Mean survival ranged from 15.3 to 22.5 days at 4 Gy, 25 to 30 days at 8 Gy, 22.3 to 29.7 days at 9 Gy, and 32.9 to 33.6 days at 12 Gy single dose irradiation. At 2.28 Gy X 2 split dose irradiation mean survival was 29.3 days, for 7.53 Gy X 2 mean survival was over 47 days. The data for single fraction irradiation fit a linear regression line (r = 0.908) of mean animal survival = (1.22 [dose in Gy] + 16.7) days. Tumor GSH levels were decreased with all BSO dosing regimens tested. The most aggressive regimen (I.P. BSO + oral BSO for 5 days), reduced tumor GSH to 6.2% of control. Increased survival in irradiated glutathione depleted mice versus mice receiving radiation alone was not seen. Brain neoplasms, Buthionine Sulfoximine, Glutathione,
Radiation therapy.
INTRODUCTION
Agents which lower GSH levels in cells and tissues have greatly facilitated studies of the inter-relationships of ionizing radiation and GSH ( 12). The compound buthionine sulfoximine (BSO) is able to deplete intracellular GSH levels (23-24). We have reported that following intraperitoneal (I.P.) administration of BSO in athymic nude mice previously injected intracranially with an anaplastic glioma cell line, depletion of tumor GSH occurs without any decrease of GSH levels in the contralateral normal brain (50). Such experiments suggested that BSO-mediated radiosensitization via GSH depletion, if it exists, might be confined to an intracranial tumor without affecting the radiation tolerance of normal brain. Following establishment of an external beam radiation therapy dose response curve for a human glioma xenograft model, we investigated the influence of BSO on radiosensitivity. To
The treatment of humans with supratentorial high grade gliomas is highly unsatisfactory. The long term survival for individuals afflicted with glioblastoma multiforme is less than 5% while that for individuals with anaplastic astrocytoma is approximately 20% (6, 28, 33, 55). Hopefully, rational therapeutic strategies based upon experimental models may improve these results. One such strategy is to influence tumor cell radiosensitivity by depletion of intracellular glutathione (GSH). It is clear that GSH affords protection against a wide variety of cytotoxic agents. Previous in vitro studies have shown that GSH depletion by itself results in mild enhancement of cellular radiosensitivity (3, 11, 13, 17, 25, 36-40,45,47, 53-54).
work was supported by NIH grants CA 44640, CA 11898, NS 20023, NS 00958, and DK 26912 as well as American Cancer Society grant CH-403 and a Career Development Award (ECH) and Bristol Meyer’s grant lOO-RI 8. Accepted for publication 22 January 1992.
Reprint requests to: Edward C. Halperin, M.D., Department of Radiation Oncology, Box 3085, Duke University Medical Center, Durham, NC 277 10. Acknowledgements-Cheryl Wallace, Donna Wimberley, and Ruth Aultman typed the manuscript; Ann Tamariz edited it. Dr. Scott Clegg assisted in the linear regression analysis. This 103
104
I. J. Radiation Oncology 0 Biology 0 Physics
our knowledge, this represents the first evaluation of BSO and external beam radiotherapy in a human glioma xenograft model. METHODS
AND MATERIALS
Animals Homozygous nu/nu BALB/c athymic mice at least 6 weeks old were used for all experiments. The animals were from the independent breeding colony maintained at Duke University. They were maintained as previously described (46). Xenograji establishment The tumor line, D-54MG, is the Duke University subline of the human anaplastic glioma line A- 172. The morphology, biochemistry, antigenicity, and karyotype of D54MG have been extensively investigated (4, 5, 22). Its sensitivity in athymic mice to a variety of chemotherapeutic agents has been determined (47). The xenografts were grown subcutaneously in athymic mice in a manner previously described (7). Tumor homogenate was prepared with a 60-mesh tissue cytosieve. Xenograji implantation Tumor was washed through the cytosieve with minimal essential medium and the cells were concentrated by centrifugation. The resulting homogeneous tumor suspension was mixed with an equal volume of 7% methyl cellulose* and the mixture was loaded into a 250-ul Hamilton syringe. Animals were given injections of 5 ul of tumor suspension into the right cerebral hemisphere with a 27 gauge needle equipped with a sleeve allowing 4.5 mm penetration. Animals were checked daily for neurological symptoms and death. Control animals received tumor injections and no therapy. Radiation therapy technique Following anesthesia with I.P. pentobarbital (50 mg/ kg) the animals were irradiated to the entire head with a single dorsal photon field from a 4 MeV linear accelerator+. The dose was prescribed to the skin surface between the eyes of the animals. The animals were placed on a 30 X 30 X 10 cm polystyrene block to provide scattering material, and 1 cm bolus was placed on the skin to ensure full dose. The target to skin distance was 80 cm. For single fraction radiotherapy, total doses of 4, 8, 9, and 12 Gy at the prescription point were administered at 200 cGy/min. For two-fraction irradiation, doses equivalent to 4 Gy X 1 and 12 Gy X 1 were determined using a model of E/a = nd (I + d/a//3) with E = equivalent dose, n = number of fractions, d = dose/fraction, and a and p being the factors of the linear-quadratic model of radiation killing.
* Fluka, Switzerland. +Clinic 4, Varian Associates, Palo Alto, CA.
Volume 24. Number I. 1992
We assumed that acutely reacting tumor had an a/B = 10 (18, 19). Thus, 4 Gy X 1 was equivalent to 2.28 Gy X 2 with fractions separated by at least 6 hr and 12 Gy X 1 was equivalent to 7.53 Gy X 2, also with fractions separated by at least 6 hr. BSO regimen/therapy Animals were irradiated 8 days following I.C. injection of the tumor homogenate. In those animals receiving BSO, the I.P. injections of the drug began on Day 3 following tumor injection. In experiments 1 through 4 D,L-BSO (22.2 mg/ml, molecular weight 222.3)* dissolved in normal saline, was given by I.P. injection (2.5 mmol/kg) starting five days prior to irradiation. Doses of D,L-BSO were given at 0, 12, 24, 48, 72, and 96 hr in experiments 1 through 4. In experiments 5 and 6, L-BSO (synthesized as previously described, (24)) at the same concentration, was administered by I.P. injection, starting 3 days prior to irradiation, every 12 hr for seven doses and was also available to the animals prior to and following irradiation in acidified drinking water (pH 3.0) at a concentration of 20 mM. Since only the L isomer is active, the L-BSO regimen represented an increase in both the number of injections and in the active dose per injection. Irradiation took place 12 hr after the last injection. In experiments 7 through 10, L-BSO was administered every 12 hr for 2 days prior to irradiation in addition to being available in the animals’ drinking water. Survival Animals were examined each day. Survival was measured from the date of I.C. injection of the tumor homogenate until the date of death. Death was due to localized tumor growth without distant metastases. Gluthathione assay GSH levels in I.C. xenografts (Day 12 after implantation) were measured as previously described (50) by the method of Tietze (52) using procedures and solutions described by Griffith (23). Groups of five animals with I.C. D-54MG xenografts were treated with D,L-BSO at the above dose schedules for 6 I.P. doses and compared to previously published values using L-BSO for 4 or 7 I.P. doses plus concomitant availability in drinking water (33). Control GSH levels were measured in 10 xenografts from untreated animals. Statistical evaluation Statistical significance was assessed by the Wilcoxon rank sum test for median survival (56). The single fraction radiation dose response data was fitted, using a linear regression analysis, to a best fit line y = mx + b.
f Sigma Chemical Company, St. Louis, MO.
105
Radiation response and BSO in a human glioma cell line 0 E. C. HALPERINet al.
Glutathione
BSO regimen*
(umol/g)+
0.48 + 0.14 (74.7)* 0.15 + 0.04 (92.3)” 0.12 -+ 0.12 (93.8)”
DL-BSO (5 mmol/kg X 6) L-BSO (2.5 mmol/kg X 4) L-BSO (2.5 mmol/kg X 7)
in all experiments, but in no case was the effect of BSO as a single agent statistically significant. The use of single fraction external beam radiation therapy produced a stepwise increase in animal survival. The median survival in the five experimental groups receiving 4 Gy ranged from 15 to 22.5 days. In the two experimental groups receiving 8 Gy the median survival was 28 days. Median survival for the four groups receiving 9 Gy ranged from 22 to 29 days. The median survivals for the two groups receiving 12 Gy was 32 and 33 days. (Tables 24). The single fraction radiation dose-response data in Figure 1 shows the trend towards increased survival with increasing dose. The mean survival data are moderately well fitted (r = 0.908) by a regression analysis to a y = mx + b line with y = [ 1.22 (dose in Gy) + 16.771 days. Fractionated irradiation significantly increased survival compared to control animals. (Table 4). The median survival for 2.28 Gy X 2 was similar to those obtained for 4 Gy X 1 (2 1 days v. 15-22 days). The survival of animals receiving 7.53 Gy X 2 was longer than those receiving 12 Gy X 1. The groups are not, however, comparable because of the unusually long survival of the control animals in experiment ten. For each of the radiation doses tested, comparative experiments were done following either D,L-BSO or L-BSO mediated GSH depletion. In experiments 2 through 4, using a BSO dose regimen producing a modest depletion of GSH levels in the I.C. xenografts, no appreciable increase in the median survival of the animals was seen. Administration of BSO regimens effecting greater depletions of GSH in the xenografts (experiments 5 through 10) also failed to produce results which reproducibly showed BSO-mediated radiosensitization. In general, pretreatment of mice with BSO resulted in no significant increase in survival in animals subsequently treated with irradiation compared to mice receiving radiotherapy alone. In portions of some experiments (experiments 4-8) mice receiving BSO had slight increases in mean or median survival compared to controls or mice receiving radiation alone. These observations were not confirmed by the majority of experiments,
imal survival
Table 1. D-54 MG xenograft glutathione values following administration of BSO
* DL-BSO was given by I.P. injection for 6 doses (5 mmol/ kg) at 0. 12, 24, 48, 72, and 96 h. L-buthionine-SR-sulfoximine was given by I.P. injection for 4 or 7 doses (2.5 mmol/kg) at 12h intervals, with concomitant availability in acidified drinking water (pH 3.0) at a concentration of 20 mM. Values shown for comparitive analysis (33). 7 Mean + S.D. Number in parenthesis, depletion as a percent of control from five measurements. t Control glutathione 1.89 AZ0.13 umol/g. 4Control glutathione 1.94 + 0.09 umol/g (33).
RESULTS The GSH concentration (mean + S.D.) in D-54MG glioma xenografts 12 days after tumor inoculation was 1.89 r 0.13 umol/g wet weight (Table I), similar to pre-
viously published values (33). Treatment with D,L-BSO (six doses) resulted in depletion of GSH to 0.48 f 0.14 umol/g wet weight. Previous studies have shown that treatment with LBSO (four doses I.P. plus administration in drinking water) resulted in depletion of glutathione levels to 0.15 + 0.04 umol/g wet weight. More extensive treatment with further L-BSO (seven I.P. doses plus administration in drinking water) resulted in depletion of GSH levels to 0.12 +- 0.12 umol/g wet weight (6.2% of controls) (33). In 9 of the 10 experiments, using between 8 and 12 animals per experimental group, there was excellent reproducibility of the survival of control animals. The mean survival ranged from 14.1 t 3 days to 18.8 +- 1.5 days while the median survival ranged between 14.5 and 19 days. (Tables 2-4). In experiment 10, survival of controls was prolonged (mean 34.3 k 2.6 days)-perhaps because of cells being used from an earlier serial passage. The use of either D,L-BSO or L-BSO alone slightly prolonged an-
Table 2. Survival in days of animals receiving radiotherapy alone or radiotherapy plus I.P. D,L-BSO 1
2
Experiment number
Mean
Median
Mean
Therapy control BSO
18.6 -t 2.6 20.6 I!I 2.5
19 22
18.5 + 2.2 21.3 ? 7.4
3 Median
Mean
18 20
17.1 + 20.2 f
32* 29.5*
25 16.2 33.6 31.5
4 Median
2.3 2.4
17 20.5
t- 8 + 9 -+ 3.4 f 10
28* 9.5 33* 33*
4Gy 4 Gy/BSO
8 GY 8 Gy/BSO 12Gy 12 Gy/BSO * p < 0.05 v. control. n = 8-12 animals for each value shown.
32.9 f 4.3 30.1 + 7.7
Mean 18.8 21.3 22.5 22.3 30 25.6
* + f + f +
1.5 5.5 3.6 2.7 6.1 8.8
Median 19 20 22 23 28* 29
106
I.J.
Radiation
Oncology
0 Biology
0 Physics
Volume
Table 3. Survival in days of animals receiving radiotherapy alone or radiotherapy plus seven doses of I.P. L-BSO plus LBSO in drinking
Mean
Therapy control BSO 4Gy 4 Gy/BSO 9Gy 9 Gy/BSO
14.1 14.5 15.3 17.8 22.6 27.5
+ * f + f +
6 Median
1.3 1.8 0.5 1.1 2.5 1.9
Mean 17.3 17.2 20.8 21.2 28.1 22.1
14.5 15 15 18*,+ 22* 27*,+
* p < 0.05 v. control. ‘p < 0.05 v. equal radiation dose without n = 7-l 1 animals for each value shown.
f f + f & I?
Median 1.7 3.1 1.8 4.2 3.8 9.5
17.5 18 21 21* 29* 27
BSO.
DISCUSSION Radiation therapy prolongs the survival of humans with supratentorial high grade gliomas (6, 28, 34, 55). There is some evidence of a clinical radiation dose response curve with an improvement in survival in individuals who receive greater than 55 Gy although this finding is not universally confirmed (55). Our experiments, with a human glioma xenograft model, demonstrate a linear single fraction radiation dose response curve. We have shown stepwise prolongation of animal survival with single fractions of radiation ranging from 4 to 12 Gy. Our relatively small number of data points does not allow us to completely exclude the possibility that the dose response curve is sigmoid. Radiation damage to DNA may include single and double strand breaks in the sugar-phosphate backbone of the molecule, alteration or loss of bases, formulation of cross-links between the DNA strands, or cross-links between the DNA and chromosomal proteins. It has been suggested that unrepaired or incorrectly repaired doublestrand breaks are the critical lesions (29, 42). Changes in
Table 4. Survival in days of animals receiving fractionated radiotherapy plus four doses of I.P. L-BSO in drinking 7 Experiment
number
Therapy control BSO 4Gy 4 Gy/BSO 2.28 x 2 = 4.56 Gy 2.28 x 2 = 4.56 Gy/BSO 9Gy 9 Gy/BSO 7.53 x 2 = 15.06 Gy 7.53 x 2 = 15.06 Gy/BSO
16.3 17.2 20.9 22.5
+ I + +
1.2 1.9 3.7 1.9
22.3 +- 5.6 23.7 +- 5.4
* p < 0.05 v. control. n = lo-12 animals for each value shown.
alone or fractionated water
8 Median
Mean
I. 1992
nuclear and cytoplasmic GSH levels may influence cellular radioresponsiveness (15, 16. 30). A large number of potential mechanisms for this phenomenon exist. GSH. the predominant non-protein intracellular thiol. has the potential to eliminate primary or secondary radiationinduced free radicals with hydrogen atom donation via one-electron reactions. GSH depletion may thus affect radiosensitivity because of changes in the ability of the cells to detoxify radiation-induced free radicals ( 1. 2. 1 i12, 14, 20. 27, 34-35, 4 l. 50). As a radioprotector via chemical reduction of radiation-induced radicals, GSH competes with oxygen and electron affinic hypoxic cell sensitizers for fixation and repair of these lesions. This competition model of thiol depletion may only apply at low oxygen concentrations because of reaction rate kinetics (8, 9). Thus, GSH depletion by BSO may increase the efficacy of hypoxic cell sensitizers. Nitroimidazoles. when incubated for a prolonged period of time at 37°C under hypoxic conditions, tend to be preferentially cytotoxic towards cells that are oxygen deficient. These compounds also tend to reduce the level of GSH and other nonprotein sulfhydryls in the cell. As a consequence of this thiol depletion. cells may attain heightened sensitivity to radiation and certain chemotherapy agents-particularly alkylating agents (27). BSO might increase the radiosensitizing effects of nitroimidazoles by furthering thiol depletion. /I? vitv experiments have tended to show radiosensitization under conditions of extreme thiol depletion in combination with a nitroimidazole sensitizer and high doses of radiation ( 10, 2 I. 26,42. 57). Very recently. BSO has been shown to accentuate the radiosensitization of SR-2508 in a fractionated radiation model with clinically relevant doses (32). Glutathione may also be involved in the repair of critically damaged DNA sites. It is known. for example. that GSH depletion slows the repair of DNA-protein cross links produced by radiation (40). This inhibition of repair could explain the therapeutic significance of compounds that
water
5 Experiment number
24. Number
17 17 22.5 22.5
23 22
Mean 16.5 16.8 18.7 20.1
? f f f
1.7 1.6 2.8 2.3
29.7 * 16.7 25.3 + 4.6
radiotherapy
9
10
Median
Mean
Median
Mean
Median
16.5 17 19.5 20*
16.5 + 2.2 17 _+ 2.9
16 18
34.3 +- 2.6 34.9 I!Z4.2
34 33
29.3 +- 25.5 20.5 + 3.1
21* 22*
> 47 50.2 f 6.8
49* 50*
25* 25*
Radiation response and BSO in a human glioma cell line 0 E. C. HALPERINd al.
Dose
[Gy]
Fig. 1. Mean survival in days of single fraction irradiated nude mice with human glioma xenografts. Fitted to linear regression line survival = (1.22 [dose in Gy] + 16.77) days (r = 0.908).
intracellular GSH in the post-radiation period ( 12). Glutathione may also influence the activity of enzymes responsible for dealing with radiation damage, chemotherapy damage, and the transport of chemotherapeutic agents across membranes (49). Glutathione is also involved in the detoxification of radiation-induced H202 or organoperoxides. It should be noted that changes in GSH level, or in the activity of GSH-dependent enzymes, are expected to affect radiation damage or repair only if the alteration is sufficiently large. If the GSH level or enzyme activity greatly exceeds the amount required, substantial decreases may be without significant biological effect. One recent study, for example, found no effect of selenite-induced variation in glutathione peroxidase activity on radiation-induced cell killing or DNA strand breakage (45). Other possible mechanisms of radioprotection by GSH include direct restoration of altered DNA, reduction of DNA peroxyl radicals, DNA peroxides, or lipid peroxides, restoration by GSH of free radical altered radioprotectors (i.e., vitamin C or E), and/or maintenance of DNA-associated protein sulthydryls in the reduced form. We have demonstrated, in previous experiments, that administration of BSO to athymic mice bearing I.C. human glioma xenografts results in highly selective depletion of GSH in neoplastic tissue versus contralateral normal brain (37). Furthermore, treatment of mice bearing I.C. human medulloblastoma or glioma xenografts with melphalan plus BSO resulted in a significant increase in me-
deplete
107
dian survival over that produced by melphalan alone (20). Additionally, treatment of mice bearing intracranial D54MG xenografts with interstitial brachytherapy (I’*’ seeds) plus BSO produced a significant increase in median survival over that produced by brachytherapy alone (34). The current studies provide evidence that an assortment of BSO dosage regimens, given prior to external beam therapy, did not reproducably influence the survival of nu/nu BALB/c athymic mice with human glioma xenografts. These results are consistent with in vitro studies reporting low radiosensitization following BSO-mediated GSH depletion (3, 11, 13, 38-41, 4849, 5 1). In contrast to our previous studies using brachytherapy in GSH depletion (33) our current study did not demonstrate consistent increases in median survival in BSO-pretreated mice with human glioma xenografts receiving external beam therapy. Differences in the influence of BSO on animal survival, as a function of the form of radiation (single fraction teletherapy vs brachytherapy), may be explicable if an important influence of BSO on mammalian tumors is via inhibition of GSH-mediated repair of lowdose-rate radiation damage. If this were the case, BSOmediated GSH depletion might be more likely to favorably influence survival in experiments using fractionated rather than single dose teletherapy. Kramer ef al., however, have recently demonstrated that BSO treatment, by itself, did not effect RIF-1 or MCA tumor growth delay in C3H/ Sed mice following fractionated teletherapy (32). The lack of influence of BSO in our b.i.d. experiments (Table 4) supports Kramer’s findings. The finding that the radiosensitizing effects of BSO during brachytherapy are significant suggests the need to elucidate mechanisms of action which pertain during continuous low dose rate irradiation but not with fractionated external beam. It remains possible, of course, that the differing radioresponsiveness between brachytherapy and teletherapy may be due to our tumor xenografts being heterogeneous with respect to radiation response and that this heterogeneity, or at least some of it, is maintained during serial transplantation (43, 44). In summary, we have reported the development of a human glioma xenograft model which has a relatively reproducible duration of survival in nude mice. We have further shown a single fraction external beam radiation dose response curve in this model. Finally, we have shown that the administration of I.P. and oral BSO does not consistently influence the duration of animal survival.
REFERENCES Bernstein, M.; Gutin, P. H.; Deen, D. F.; Weaver, K. A.; Levin, V. A., Barcellos, M. H. Radiosensitization of the RIF- 1 murine flank tumor by desmethylmisonidazole (Ro 05 9963) during interstitial radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. 8:487-490; 1982. Biaglow, J. E.; Varnes, M. E.; Epp, E. R.; Clark, E. P.; Tuttle, S. W.; Held, K. D. Role of glutathione in the aerobic ra-
diation response. Int. J. Radiat. Oncol. Biol. Phys. 16: 13 1 l1314;1989. 3. Biaglow, J. E.; Varnes, M. E.; Tubble, S. W.; Oleinick, N. L.; Glazier, K.; Clark, E. P.; Epp, E. R.; Dethlefsen, L. A.: The effect of L-buthionine sulfoximine on the aerobic radiation response of A549 human lung carcinoma cells. Int. J. Radiat. Oncol. Biol. Phys. 12: 1139-l 142; 1986.
108
I. J. Radiation Oncology 0 Biology ??Physics
4. Bigner, D. D.; Bigner, S. H.; Ponten, B.; Westermark, B.; Mahaley, Jr. M. S.; Ruoslahti, E.; Herschman, H.; Eng, L. F.; Wikstrand, C. J. Heterogeneity of genotypic and phenotypic characteristics of fifteen permanent cell lines derived from human glioma. J. Neuropath. Exp. Neurology. 40: 201-229;1981. 5. Bigner, S. H.; Mark, J.; Bigner, D. D. Chromosomal composition of four permanent cultured cell lines derived from human gliomas. Cancer Genet. Cytogenet. 10:335349;1983. 6. Bloom, H. J. G. Intracranial tumors: response and resistance to therapeutic endeavors, 1970- 1980. Int. J. Radiat. Oncol. Biol. Phys. 8: 1083-I I 13. 7. Bullard, D. E.; Schold, S. C.; Bigner, S. H.; Bigner, D. D. Growth and chemotherapeutic response in athymic mice of tumors arising from human glioma-derived cell lines. J. Neuropath. Exp. Neurol. 40:4 10-427; 198 1. 8. Bump, E. A.; Brown, J. M. Role of glutathione in the radiation response of mammalian cells in vitro and in viva. Pharmac. Ther. 47: 117-I 36; 1990. 9. Bump, E. A.; Yu, N. Y.; Brown, J. M. Radiosensitization of hypoxic cells by depletion of intrace. Science 127:554555;1982. 10. Bump, E. A.; Yu, N. Y.: Taylor, Y. T.; Brown, J. M.; Travis, E. L.; Boyd, M. R. Radiosensitization and chemosensitization by diethylmaleate. In: Nygaard, O., Simic, M., eds. Radioprotectors and anticarcinogens. New York: Academic Press; 1983:297-323. biochemical modification of 1I. Clark, E. P. Thiol-induced chemo- and radioresponses. Int. J. Radiat. Oncol. Biol. Phys. 12:1121-l 126;1986. 12. Clark, E. P.; Epp, E. R.; Biaglow, J. E.; Morse-Gaudio, M.; Zachgo, E.: Glutathione depletion, radiosensitization, and misonidazole potentiation in hypoxic Chinese hamster ovary cells by buthionine sulfoximine. Radiat. Res. 98:370380;1984. 13. Dethlefsen, L. A.: Biaglow, J. E.; Peck, V. M.: Ridinger, D. N. Toxic effects of extended glutathione depletion by buthionine sulfoximine on murine mammary carcinoma cells. Int. J. Radiat. Oncol. Biol. Phys. 12: 1157-l 160;1986. 14. Durand, R. E. Roles of thiols in cellular radiosensitivity. Int. J. Radio]. Oncol. Biol. Phys. 10:1235-1238;1984. 15. Edgren, M. R. Nuclear glutathione and oxygen enhancement of radiosensitivity. Int. J. Radiat. Oncol. Biol. Phys. 51:36; 1987. depletion of 16. Edgren, M. R.; Revesz, L. Compartmentalised glutathione in cells treated with buthionine sulfoximine. Br. J. Radiat. 60:723-724;1987. 17. Evans, J. W.; Taylor, Y. C.; Brown, J. M. The role of glutathione and DNA strand repair in determining the shoulder of the radiation survival curve. Br. J. Cancer 49:49-53;1984. and therapeutic gain. In: G. G. 18. Fowler, J. F. Fractionation Steel, G. E. Adams, and A. Horwich, editors. The biological basis of radiotherapy, 2nd edition. Amsterdan: Elsevier; 1989:181-208. of brachytherapy. In A. A. 19. Fowler, J. The radiobiology Martinez, C. G. Orton, and R. F. Mould, editors. Brachytherapy: HDR and LDR. Columbia: Nucleotron Corporation; 1990:121-137. 20. Friedman, H. S.; Colvin, 0. M.; Griffith, 0. W.; Lippitz, B.; Elion, G. B.; Schold, S. C.; Hilton, J.; Bigner, D. D. Increased melphalan activity in intracranial human medulloblastoma and glioma xenografts following buthionine sulfoximine-mediated glutathione depletion. JNCI 8 1:524527;1989. 21. Fu, K. K.; Hurst, S.; Begg, A. C.; Brown, J. M.; The effects of misonidazole during continuous low dose rate irradiation. Cancer Clin. Trials 3:257-26;1980.
Volume 24. Number 1, lYY2 22. Giard, D. J.; Aaronson, S. A.; Todaro, G. J.; Amstein, P.; Kersey, J. H.: Dosile, H.; Parks, W. P. In vifrocultivation of human tumors: establishment of cell lines derived from a series of solid tumors. JNCI 1417-l 1423;1973. 23. Griffith, 0. W. Determination of glutathione and glutathione-disulfide using glutathione reductase and 2-vinylpyridine. Anal. B&hem. 106:207-212:1980. 24. Griffith, 0. W. Mechanism of action, metabolism and toxicity of buthionine sulfoximine and its higher homologs. Potent inhibitors ofglutathione biosynthesis. J. Biol. Chem. 257:137O&l3712;1982. 25. Griffith, 0. W.; Meister, A. Potent and specilic inhibition ofglutathione synthesis by buthionine sulfoximine (S-n-butylhomocystein sulfoximine). J. Biol. Chem. 254:75587560: 1979. 26. Guichard, M.; Lespinasse. F.; Malaise, E. P. Influence ot buthionine sulfoximine and misonidazole on glutathione level radiosensitivity of human tumor xenografts. Radiat. Res. 105:l 15-125:1986. 27. Hall, E. J.; Cox, J. D. Physical and biologic basis of radiation therapy. In: Moss, W. T., Cox, J. D., eds. Radiation oncology: rationale, technique, results, 6th edition. St. Louis: C. V. Mosby Company; 1987: l-57. 28. Halperin. E. C.; Burger, P. C.; Bullard, D. E. The fallacy of the localized supratentorial malignant ghoma. Int. J. Radiat. Oncol. Biol. Phys. 15:.505-509:1988. 29. Hill, R. P. Cellular basis of radiotherapy. In: Tannock, I. F., Hill, R. P., eds. The basic science of oncology. Tarrytown, NY: Pergamon Press; 1987:237-255. 30. Jones, E. L.: Douple, E. B. The effect of in viva GSH depletion on thermosensitivity, radiosensitivity and thermal radiosensitization. Int. J. Hyperther. 6:951-955;1990. 31. Kinsella. T. J.: Dobson, P. P.; Russo, A.: Mitchell, J. B.: Fornace, Jr., A. J. Modulation of x-ray damage by SR-2508 f buthionine sulfoximine. Int. J. Radiat. Oncol. Biol. Phys. 12:l 127-l 130:1986. 32. Kramer, R. A.; Sable, M.: Howes, A. E.; Montaya. V. P. The effect ofglutathione (GSH) depletion in vivoby buthionine sulfoximine (BSO) on the radiosensitization ofSR 2508. Int. J. Radiat. Oncol. Biol. Phys. 16:1325-1330;1989. 33. Leibel, S. A.; Sheline, G. E. Radiation therapy for neoplasms of the brain. J. Neurosurg. 66: I-22; 1987. 34. Lippitz, B. E.; Halperin, E. C.; Griffith, 0. W.; Colvin, 0. M.: Honore, G.: Ostertag, C. B.; Bigner, D. D.; and Friedman. H. S. L-buthionine-sulfoximine mediated radiosensitization in experimental intersistial radiotherapy of intracerebral D-54 MC glioma xenografts in athymic mice. Neurosurg. 26:255-260;1990. 35. Loh, S. N.; Dethlefsen, L. A.; Newton, G. L.; Aguilera, J. A.; and Fahey, R. C. Nuclear thiols: technical limitations on the determination of endogenous nuclear glutathionine and the potential importance of sulhydryl proteins. Radiat. Res. 121:98-196;1990. 36. Meister, A. Metabolism and transport of glutathione and other gamma-glutamyl compounds; functions of glutathione: biochemical physiological, toxicological, and clinical aspects. In: Larsson, A., Orrenius, S.. Holmgren, A.. Mannervick. B., eds. Functions of glutathione: biochemical. physiological, toxicological, and clinical aspects. New York: Raven Press: 1983. A. 1.; Rojas, A.: Smith, K. A.; Soranson, 37. Minchington. J. A.; Shrieve, D. C.; Jones, N. R.; Bremer, J. C. Glutathione depletion in tissues after administration of buthionine sulfoximine. Int. J. Radiat. Oncol. Biol. Phys. IO: 12611264;1984. 38. Mitchell, J. B.; Cook, J. A.; DeGraff, W.; Glatstein, E.; Russo, A. Keynote address: glutathione modulation in can-
Radiation response and BSO in a human glioma cell line 0 E. C. HALPERIN et a/.
39.
40.
41.
42.
43.
44. 45.
46.
47.
48.
cer treatment: will it work? Int. J. Radiat. Oncol. Biol. Phys. 16:1289-1296;1989. Mitchell, J. B.; Russo, A.; Biaglow, J. B.; McPherson, S. Cellular glutathione depletion by diethyl maleate or buthionine sulfoximine: no effect of glutathione depletion on the oxygen enhancement ratio. Radiat. Res. 96:422-428; 1982. Oleinick, N. L.; Liang-Yan, X.; Friedman, L. R.; Donahue, L. L.; Biaglow, J. E. Inhibition of radiation-induced DNAprotein cross-link repair by glutathione depletion with LButhionine Sulfoximine. NC1 Monogr. 6:225-229;1988. Oliver, J. M.; Spielberg, S. P.; Pearson, C. B.; Schulman, J. D. Microtubule assembly and function in normal and glutathione synthetase-deficient polymorphonuclear leukocytes. J. Immunol. 12O:l 181-1 186;1978. Rauth, A. M. Radiation carcinogenesis. In: Tannack, I. F., Hill, R. P., eds. The basic science of oncology. N.Y.: Pergamon Press; 1987: 106- 124. Rockwell, S.; Moulder, J. E. Hypoxic fractions of human tumors xenografted into mice: a review. Int. J. Radiat. Oncol. Biol. Phys. 19: 197-202; 1990. Rofstad, E. K. Human tumor xenografts in radiotherapeutic research. Radiother. Oncol. 3:35-46;1985. Sandstrom, B. E.; Carlsson, J.; Markland, S. L. Seleniteinduced variation in glutathione peroxidase activity of three mammalian cell lines: no effect on radiation-induced cell killing or DNA strand breakage. Radiat. Res. I 17:3 1832$1989. Schold, Jr. S. C.; Bullard, D. E.; Bigner, S. H.; Jones, T. R.; Bigner, D. D. Growth, morphology and serial transplantation of anaplastic human gliomas in athymic mice. J. NeuroOncol. 1:5-14;1983. Schold, Jr., S. C.; Friedman, H. S.; Bigner, D. D. Therapeutic profile of the human ghoma line D-54 MG in athymic mice. Cancer Treat. Rep. 7 1:849-850;1987. Shrieve, D. C.; Denekamp, J.; Minchington, A. I. Effects of glutathione depletion by glutathione sulfoximine on radiosensitization by oxygen and misonidazole in vitro. Radiat. Res. 102:283-294; 1985.
109
49. Shrieve, D. C.; Harris, J. W. Effects of glutathione depletion buthionine sulfoximine on the sensitivity of EMT6/SF cells to chemotherapy agents or X radiation. Int. J. Radiat. Oncol. Biol. Phys. 12:l 171-l 174;1986. 50. Skapek, S. X.; Calvin, 0. M.; Griffith, 0. W.; Groothius, D. R.; Colapinto, E. V.; Yisheng, L.; Hilton, J.; Elion, G.; Bigner, D. D.; Friedman, H. S. Buthionine sulfoximinemediated depletion of glutathione in intracranial human glioma-derived xenografts. Biochem. Pharm. 37:43 134317;1988. 51. Sun, J. D.; Ragsdale, S. S.; Benson, J. M.; Henderson, R. F. Effects of the long-term depletion of reduced glutathione in mice administered L-buthionine-S, R-sulfoximnie. Fund. Appl. Toxicol. 5:9 13-9 19; 1985. 52. Tietze, F. Enzymatic method for quantitative determination of nanogram amounts of total and oxidized glutathione: applications to mammalian blood and other tissues. Anal. Biochem. 27:502-522;1969. 53. Varnes, M. E.; Biaglow, J. E.; Roizin-Towle, L.; Hall, E. J. Depletion of intracellular GSH and NPSH by buthionine sulfoximine and diethyl maleate: factors that influence enhancement of aerobic radioresponse. Int. J. Radiat. Oncol. Biol. Phys. 10: 1229- 1233; 1984. 54. Vos, 0.; van der Schans, G. P.; Roos-Verhey, W. S. Reduction of intracellular glutathione content and radiosensitivity. Int. J. Radiat. Oncol. Biol. Phys. 50: 155-165;1986. 55. Walker, M. D.; Strike, T. A.; Sheline. G. E. An analysis of dose-effect relationship in the radiotherapy of malignant ghomas. Int. J. Radiat. Oncol. Biol. Phys. 5:17251731:1979. 56. Wilcoxon, F. Individual comparisons Biomet. Bull. 1:80-83;1945.
by ranking methods.
57. Yu, N. Y.; Brown, J. M. Depletion of glutathione in vivo as a method of improving the therapeutic ratio of misonidazole and SR 2508. Int. J. Radiat. Oncol. Biol. Phys. IO: 1265%1269;1984.
