Differential radioprotection of cultured human diploid fibroblasts and fibrosarcoma cells by WR1065

0360-3016/92 $5.00 + 00 Copyright G 1992 Pergamon Press Ltd.

Inr J Radramn Oncology Biol Phys.. Vol. 24, pp. 7 13-7 19 Printed in the U.S.A. All nghts reserved.

??Biology Original Contribution DIFFERENTIAL RADIOPROTECTION OF CULTURED HUMAN DIPLOID FIBROBLASTS AND FIBROSARCOMA CELLS BY WR1065 XIAFANG

ZHANG,

M.D.,

PETER P. LAI, M.D., PH.D. AND YVONNE

Radiation Oncology Center, Mallinckrodt

C. TAYLOR,

PH.D.

Institute of Radiology, Washington University School of Medicine, St. Louis, MO

The present studies were performed to determine whether WR1065, the dephosphorylated, free-thiol active metabolite of WR2721, could provide differential radioprotection of normal and tumor cell lines in vitro and secondly to investigate potential mechanisms for the selective nature of the radioprotection at the cellular and molecular level. When 4 mM WR1065 was administered 30 min prior to and during irradiation, a protection factor of 1.9 was obtained in clonogenic assays performed with normal human diploid fibroblasts (AG1522) while no protection of fibrosarcoma cells (HTIOSO) was observed. Some radioprotection of fibrosarcoma cells was observed with higher drug concentrations (lo-40 mM), but the increase in survival was considerably less than the plateau level reached with the diploid fibroblasts (Ifold vs 24-fold at 6 Gy). The observation of such a selective effect in vitro with WR1065 indicates that differences in tissue-specific variables such as blood flow, pH, pOZ, and drug dephosphorylation cannot solely account for the selective nature of the radioprotection afforded by WR2721 in viva. Incubation of nucleoids with increasing concentrations of the DNA intercalating dye propidium iodide was used to titrate the ability of DNA to undergo supercoiling changes. The relaxation and rewinding of supercoiled DNA loops in isolated nucleoids serves as an indicator of both the presence of DNA damage and inherent differences in DNA loop characteristics. Fibrosarcoma cells had a much larger propidium iodide-relaxable DNA loop size than fibroblasts. The rewinding phase of the DNA supercoiling response is impaired by the presence of radiation-induced DNA strand breaks. Four mM WR1065 resulted in a significant reduction in the amount of rewinding inhibition observed after a dose of 10 Gy in diploid fibroblasts (protection factor = 1.43) but did not alter the response of irradiated fibrosarcoma cells. These results, indicating that WR1065 had a preferential radioprotective effect in vitro on both survival and the manifestation of DNA damage at the nucleoid level, are consistent with the hypothesis that cell type differences in chromatin organization and DNA-drug associations could play a role in the selective radioprotection. Radioprotection,

WR1065,

Nucleoid, DNA supercoiling, Cell kinetics.

INTRODUCIION

ficient tumor vascularity, reduced tumor oxygen tensions and hence less efficient thiol radioprotection, reduced uptake through altered drug transport mechanisms and/or tumor membrane properties, and altered ability to dephosphorylate WR2721 (1, 3,7, 10, 15,26,27) are among the reasons postulated for the differential protection of normal vs. tumor tissues. Uptake studies with WR2721 in vitro indicate that WR1065, the dephosphorylated, freethiol active metabolite of WR272 1, is the form that is transported into cells and responsible for radioprotection and that transport occurs through passive diffusion (1, 19). From the lack of a correlation between total drug absorbed and the magnitude of radioprotection observed in normal tissues (25,27), it appears that the basis for the

S-2-(3-aminopropylamino)ethylphosphorothioic acid (WR272 1) has been shown to protect many normal tissues in murine systems where dose modification factors around 2-3 have been observed (27). However, there is less agreement as to whether WR2721 provides any tumor radioprotection. Yuhas et al. (27) found that none of 12 murine tumors tested were protected by WR2721 and only 1 of 13 tumors showed drug uptake. However, tumor radioprotection has been noted in several other experimental studies (3, 7, 10, 25) where it appears that timing of the irradiation, radiation dose, tumor size, tumor type, and tumor oxygenation may all be important variables. De-

This work was presented at the 33rd annual meeting of the American Society for Therapeutic Radiology and Oncology,

gashikubo for his assistance in the flow cytometric analyses. This work was partially supported by a research grant from U.S. Biosciences, Blue Bell, PA and by a Radiation Oncology Institutional award for pilot studies in cancer-related research. Dr. Lai is supported by an American Cancer Society Clinical Oncology Career Development Award. Accepted for publication 7 May 1992.

Washington, DC, 4-8 November 199 1. Reprint requests to: Dr. Xiafang Zhang, Cancer Biology Section, Washington University School of Medicine, 4511 Forest Park Blvd., St. Louis, MO 63108. Acknowledgrments-The authors would like to thank Dr. Hi713

714

I. J. Radiation Oncology 0 Biology 0 Physics

selectivity of the radioprotection by aminothiols is very complex and may involve, in addition to the effects described above, differences in intracellular compartmentalization. Recently, many investigators have examined the radioprotective effects of WR1065 utilizing elution assays to measure DNA damage production and repair. WR 1065 effectively protected rodent cell lines (CHO, V79) against radiation-induced DNA single-strand (5, 14) and doublestrand break formation (13, 18) although protection factors were generally lower than those measured for cell survival. WR 1065 was found to reduce the rate of singlestrand break (SSB) rejoining presumably the consequence of additional breaks generated by H202 formation through the auto-oxidation of the drug during the repair incubation, but appeared not to alter double-strand break repair kinetics (14, 18). WR1065 also protects cells against the mutagenic and neoplastic effects of ionizing radiation (6, 8). If aminothiols protect DNA from radiation-induced damage by a free radical-scavenging mechanism, then the short halflives of these species would require that the drug be in very close proximity to the DNA to be effective. Equilibrium dialysis studies have demonstrated a preferential association of WR 1065 with DNA (20). The more efficient radioprotection by cationic thiols (WR 1065 and cysteamine) may be the result of counterion condensation and thus charge dependent (2 1,28). Studies of the incorporation of “C-WR1065 in vitro have indicated a rapid initial accumulation of the drug which plateaus after 30 min and that 10% of the drug remains tightly associated with nuclear components (9). The binding of 14C-WR1065 to intracellular DNA and nucleoproteins was found to be increased in cells that were irradiated after drug treatment and that DNA binding appeared to rapidly saturate (4). These results suggest that there may be a limited number of nuclear sites within DNA or nuclear matrix proteins with which the drug is tightly associated, however their identity remains unknown. In a study in which the ability of DNA to undergo supercoiling changes in isolated nucleoids was examined, WR1065 appeared to relax DNA loops and to increase the amount of ethidium bromide intercalation. Several studies have examined the effects of WR 1065 on cell cycle progression during and/or after drug treatment and have noted primarily an accumulation of cells in S- or G2-phase (4, 5, 12, 17). However, these studies have not examined the effects of drug treatment on postirradiation cell cycle progression to determine if there are alterations in the radiation-induced cell cycle delays. It is apparent from these studies that the mechanism of radioprotection and basis for selectivity of aminothiols remains to be defined. We postulate that the selectivity is caused by differences in DNA-drug associations that may

* Coriell Institute for Medical Research, Camden, NJ.

Volume 24, Number 4, 1992

be controlled by inherent differences in the chromatin organization of transformed/undifferentiated versus normal/differentiated cells. Assays of the DNA supercoiling response of human cell lines in our lab have shown that transformed or tumor cells and noncycling cells that are induced to proliferate tend to have increased DNA loop sizes (23). We have chosen to use an in vitro cell system and the free-thiol active metabolite WR1065 for our investigation of the basis for the selective radioprotection by aminothiols so that contributions from variables such as drug delivery rate, drug dephosphorylation, pH, and pOZ can be eliminated. The present results demonstrate that selective radioprotection can be observed in vitro with human normal and tumor cell lines thus providing a convenient system in which to investigate cellular and molecular mechanisms. METHODS

AND MATERIALS

Cell lines An apparently normal diploid fibroblast, AG 1522,* and a fibrosarcoma cell line, HT 1080,+were used in the present study. Cells were grown in alpha-MEM supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 pg/ml streptomycin as monolayers in 75 cm2 tissue culture flasks. Fibroblast cell stocks were passed once weekly at confluence. HT1080 cells were passed twice weekly and maintained constantly in exponential growth. All experiments were performed with exponentially growing cells. Drug preparation WR1065 was kindly supplied by the Drug Synthesis and Chemistry Branch, Division of Cancer Treatment, National Cancer Institute, Bethesda, MD. The drug was freshly prepared before each experiment by dissolving in Dulbecco’s phosphate-buffered saline (PBS) at a concentration of 1 M or 100 mM and diluted with cell culture medium to give the final desired concentrations. In experiments using drug concentrations over 4 mM, the drug was diluted into medium containing 25 mM HEPES buffer to better control the extracellular pH. The medium with WR1065 was filtered through a 0.45 pm Millipore filter prior to addition to cells. Radiation survival assay Exponentially growing cells were trypsinized and preplated in 60 mm tissue culture plates according to their expected radiation survival levels such that a yield of 5080 colonies would be obtained. After a 3-hr incubation to overcome the effects of trypsinization, cells were exposed to 4 mM WR1065 for 30 min and then irradiated with 220 KVp x-rays at a dose rate of 1 Gy/min. The

+ American Type Culture Collection, Rockville, MD.

Selective radioprotection by WRl065 0 X.

ZHANC

rt

715

al.

drug was removed after irradiation, the cells washed twice with PBS, and fresh medium added. All manipulations were performed at 37°C and the total WR1065 treatment time kept at 45 min. After a 1Cday (or lo-day, HT1080 cell line) incubation, dishes were stained with crystal violet and colonies containing > 50 cells were scored. Fluorescent halo assay The fluorescent halo assay was performed as previously described (22, 23). Exponentially growing cells were treated with WR1065 at 4 mM for 30 min. Drug treated or untreated cells were trypsinized, centrifuged, and resuspended at a density of - IO5 cells/ml in cold (4°C) PBS. Cells were irradiated in a Shepherd Mark I 13’Cs unit at a dose rate of 4.22 Gy/min. After irradiation on ice, nucleoids were prepared in a single step by mixing equal volumes of cell suspension and dye-lysis solution (2 M NaCI, 20 mM EDTA, 20 mM Tris-Base, pH 8.0, 2~ final desired propidium iodide (PI) concentration plus a cell-type specific concentration of the detergent TX100) and incubated in the dark at room temperature for 30 min. The TX-100 concentrations used were 0.0750.1% for AG1522 cells and 0.025-0.075% for HT1080 cells. Samples were viewed with an inverted fluorescence microscope using an excitation wavelength of 520-570 nm. The images were visualized by a silicone intensified target camera and monitor,* and measurements of nucleoid diameters (nuclear matrix core plus extruded DNA loops) and core sizes performed by a computer-linked image analyzerA as described previously (22, 23). A minimum of 25 particle diameters were measured at each PI concentration. Mean nucleoid dimensions were determined in three or more experiments for each cell line and/or condition tested, the data averaged and standard errors (SEM) calculated. Cell cycle kinetics At various times after irradiation and WR 1065 treatment, individual flasks of cells were trypsinized, the cell suspensions centrifuged (5 min, 200 X g, 4”C), and the cells fixed by resuspension in equal volumes (2.5 ml) of PBS and 95% ethanol. Cell samples were stored at 4-8°C until analysis. DNA staining with chromomycin A3 was performed by washing cell samples ( l-2 X 1O6cells total) once in Mg/Na buffer (50 mM NgCL, 150 mM NaCl) and resuspending cells in 0.5 ml Mg/Na buffer and 0.5 ml of the same buffer with 100 kg/ml chromomycin As. After 1 hr, clumps were removed by filtering and the samples run on a FACS 440 flow cytometer.** DNA histograms were analyzed by the method of Dean and Jett (2).

* Model 66, DAGE Electronics. DModel 3000 Image Analyzer, Image Technologies, Corp.

L

. . . WI,1065 4mM +X-w

1B

tiTlO60

10-p 0

\I

A

B 4

a RADIATION

0

FIBROSARCOMA

4

,

a

DOSE (Gvl

Fig. 1. The surviving fraction of AC1522 and HT1080 cells exposed to graded dose of x-rays with or without WR1065 treatment. Four mM WR1065 was present 30 min prior to and during the irradiation. Data points and error bars represent the mean f 1 SEM from four independent experiments.

RESULTS Radiation survival studies The data in Figure 1 illustrate the influence of 4 mM WR 1065 on the radiation response of AGl522 diploid fibroblasts and HT 1080 fibrosarcoma cells when administered 30 min prior to and during the irradiation. From the regression analysis of the data points beyond the survival curve shoulder region (dose 2 4 Gy), Do values of 1.67 Gy and 0.88 Gy were determined for AG1522 cells in the presence and absence of the drug, respectively. From the ratio of these values, a protection factor (PF) of 1.9 was calculated. In contrast, the results obtained with the fibrosarcoma cell line (Fig. 1B) indicate the absence of any radioprotection by 4 mM WR1065. A separate study was performed to examine the radioprotective effect of WR1065 at higher concentrations. A constant dose of 6 Gy was given to cells exposured to 440 mM WR 1065 30 min before and during the irradiation. Even at the highest concentration tested, there was little drug toxicity (plating efficiencies were > 70% of the untreated controls). The relative increase in post-irradiation survival of drug-treated cells is plotted in Figure 2 as a function of drug concentration. This ratio was calculated by dividing the surviving fraction from the combined treatment by that obtained for radiation alone after normalizing to the plating efficiency of the respective unirradiated controls (0, 4, 10, 20 and 40 mM WR1065). In the diploid fibroblast cell line, radioprotection increased significantly with increasing drug concentration up to 20 mM, at which point survival levels appeared to reach a plateau (Fig. 2). Some radioprotection of the fibrosarcoma

** Becton Dickinson.

716

I. J. Radiation Oncology 0 Biology0 Physics

WR1065 CONCENTRATION (mM)

Fig. 2. Radioprotection of AC1522 and HT1080 cells by WR1065 as a function of drug concentration. The relative increase in survival represents the ratio of the surviving fractions obtained with and without drug treatment for cells receiving 6 Gy. WR 1065 was present 30 min prior to and during the irradiation. Data points represented the average relative increase in survival from three independent experiments. Error bars indicate r+ 1 SEM.

cell line was obtained with higher drug concentrations (Fig. 2) but the survival increase was considerably less than the plateau levels reached with the diploid fibroblasts (3-fold vs 24-fold). These results indicate that radioprotection was not completely absent in the fibrosarcoma cell line, that the level of radioprotection was saturable in both cell lines, and that the differences between the tumor and normal cell line were maintained even at very high drug concentrations. DNA supercoiling measurements Figure 3 illustrates the effects of WR1065 and/or ionizing radiation on the DNA supercoiling response of both the human fibroblast and tumor cell line. At low PI concentrations (0.5-7.5 p/ml), the relaxation of negatively supercoiled looped domains results in an increase in nucleoid diameter. Further increases in PI concentration (7.5-50 pg/ml) cause a decrease in nucleoid diameter as a result of the rewinding of DNA loops in the opposite sense (Fig. 3, open circles). In this assay, differences in the relaxation and rewinding of supercoiled DNA loops in isolated nucleoids serve as an indicator of both the presence of DNA damage and inherent differences in DNA loop characteristics. A comparison of the control supercoiling response curves in Figure 3 (open symbols) at the point where DNA loops are maximally relaxed indicates that the fibrosarcoma cells appear to have an inherently larger DNA loop size. A 30 min treatment with 4 mM WR1065 did not significantly alter the DNA supercoiling response of either cell line (Fig. 3, solid circles). In irradiated cells, the presence of DNA strand breaks results in an inhibition of the rewinding phase of the supercoiling response (Fig. 3, open triangles). Four mM WR1065 greatly reduced the amount of rewinding inhibition produced by 10 Gy in the fibroblast cell line but did not significantly alter the effects of ionizing radiation

Volume 24,

Number 4, 1992

in the fibrosarcoma line (Fig. 4, solid triangles). Nucleoid core sizes were also measured in AG 1522 cells. The values obtained, 16.63 +- 0.08, 16.12 +- 0.31, 16.19 + 0.33, and 16.09 + 0.17 for untreated control, 10 Gy, WR 1065 alone, and WR 1065 plus 10 Gy, respectively, indicate that the WR 1065induced changes in the radiation response reflect only a reduction in DNA loop size. Such an effect would be consistent with the presence of fewer radiation-induced DNA strand breaks ( 16). WR 1065 at 20 mM did not produce any further reduction in the radiation-induced inhibition of DNA rewinding in the fibroblast line nor any significant alteration of the DNA supercoiling response of irradiated HT1080 cells (data not shown). These results again indicate a response that appears to be saturable. To quantitate protection factors in the fluorescent halo assay, increasing radiation doses were given to WR 1065treated AG1522 cells to determine the dose necessary to obtain the same degree of rewinding inhibition obtained with 10 Gy alone. The relative inhibition of DNA rewinding as a function of radiation dose is indicated in Figure 4 by the “excess halo diameter,” the sum of the differences in halo diameters between irradiated and unirradiated cells measured at PI concentrations 2 10 pg/ ml (the rewinding phase of the supercoiling response). From the ratio of doses required to induce the same effect, a PF of 1.43 was obtained. This is considerably less than the PF of 1.9 measured for cell survival. To simulate the same conditions used in the DNA supercoiling studies, cell survival curves were generated for cells irradiated in suspension on ice with ‘37Cs. The protection factor for cell survival, calculated in the same way as described above, was also 1.9. Thus, both irradiation conditions gave PF’s for cell survival that were greater than that measured in an assay that reflects the presence of DNA damage. Cell cycle kinetics Figure 5 indicates the fraction of AG1522 cells in the Gl, S and G2/M phases of the cell cycle as a function of

-

-z 1 E 40

Control

-.-.

??

WR 1065 4mM

.---.

WRl065

4mM + IOGy

L zz u E

30

g ” 2 20 0

20

40

0

20

40

PROPIDIUM IODIDE CONC (~g/ml]

Fig. 3. The DNA supercoiling response of nucleoids isolated from AC 1522 and HT 1080 cells. For the combined treatments cell were exposed to 4 mM WR1065 30 min before irradiation (13’Cs, 4°C). The data points represent the mean nucleoid diameters measured in at least four independent experiments. SEM values were < +- 1 pm and have not been included in the figure for clarity.

Selective radioprotection by WR1065 0 X.

Radiation alone ‘u___-_--------______~~~~~m /I:

/ @ 4

WRl065

I

+Rodiotm

I IO RADlATlON

I

-

11 I

I I

14 12 DOSE (Gy)

Fig. 4. The inhibition of DNA rewinding as a function of radiation dose for AG I522 fibroblasts irradiated with 13’Cs on ice immediately before cell lysis and nucleoid measurement. For each radiation dose indicated, the ‘excess halo diameter (EHD) was calculated from the sum of the differences between the nucleoid diameters of irradiated and control cells measured at 10. 20,35, and 50 pg/ml PI. Data points and error bars indicate the average EHD values t_ 1 SEM from three independent experiments. From the ratio of the doses required to produce the same increase in EHD, a protection factor of 1.43 was obtained. with and without WR1065 and their respective unirradiated controls. Panel B shows a large delayed accumulation of cells in S-phase following treatment with WR1065 alone (4 mM, 45 min) which is paralleled by a depletion of the G 1 population (Panel A). An increase in the S-phase fraction is typically observed at early times following irradiation due to the transient radiation-induced inhibition of DNA synthesis. This response is seen in Fig. 5B for cells irradiated with or without drug treatment (open and closed squares). At 12 hr postirradiation, a radiation induced G2 block is apparent from the increased G2/M fractions in Fig. 5C. WR1065 did not appear to dramatically alter the G2 delay. Thus, the only significant effect of WR 1065 on cell cycle progression is a dramatic S-phase accumulation apparent only in unirradiated cells. The striking feature of this effect is its occurrence at such long times after drug removal suggesting that some form of the drug or drug-related alteration may persist in cells for at least one cell cycle. WR 1065 did not alter the cell cycle distribution or radiation-induced cell cycle delays in the fibrosarcoma cell line. time for cells irradiated

ZHANG

717

et a/.

radioprotection of tumor tissue with sufficiently low drug concentrations. This feature would be clinically desirable since no escalation of tumor doses (by yet some unknown amount) would need to be applied to obtain the same degree of tumor control. The observation of such a selective effect in vitro with WR1065 indicates that differences in tissue-specific variables such as blood flow, pH, ~02, and drug dephosphorylation (3, 7, IO, 11, 15.27) cannot solely account for the selective nature of the radioprotection afforded by WR2721 in vitro. Consistent with the results of other assays of DNA damage (5, 13, 14, 18), larger protection factors for cell survival were obtained than for the radiation-induced inhibition of DNA supercoil rewinding. However, a strong correlation would not be expected if only a subcomponent of radiation-induced strand breaks was being affected or other types of DNA damage, such as crosslinks or base damage, have been altered. Our results (Fig. 4) did not indicate any effect of the drug alone on the DNA supercoiling response, which contrasts with the study by Vaughan et al. (24) in which an effect of WR1065 alone on the response of isolated nucleoids to DNA intercalating dyes was observed. However, this could reflect differences in the experimental

A

GI

90 -

60 50 -

??

WR1065+6Gy

DISCUSSION The results of the present study demonstrated a selective radioprotection by WR1065 of a human normal versus tumor cell line. Such differences have also been observed with other normal and transformed cell lines in our laboratory. While 4 mM WR1065 provided significant radioprotection of normal human lymphocytes (PF 2 2), it afforded no radioprotection of EBV-transformed lymphoblasts as determined by survival measurements (limited dilution assay, data not shown). Slight radioprotection of only one human tumor cell line (A549, PF - 1.2) has been observed so far at this drug concentration (data not shown). Thus, it would appear to be possible to avoid any

0

3

6 HOURS

9 POST

12

15

18

21

24

IRRADIATION

Fig. 5. Post-irradiation cell cycle distribution of AG1522 cells. Cells were given 6 Gy with or without treatment with 4 mM WR1065 and fixed at various times after irradiation for flow cytometric analysis of cell cycle parameters. Data shown is from a single representative experiment.

718

1. J. Radiation Oncology 0 Biology 0 Physics

methods and cell lines. Similar protection factors were obtained in this study with cells irradiated with the drug still present at 37°C and with cells irradiated on ice after drug removal. Since the majority of WR1065 appears to rapidly diffuse out of cells when resuspended (9, 19), it would appear that some more tightly bound form of the drug is responsible for radioprotection. The saturable nature of the magnitude of radioprotection observed as a function of drug concentration or treatment time (Fig. 2, refs. 14, 19) suggest that there may be a limited number of intracellular drug association sites that render the cell more radioresistant, a feature that could be cell line dependent. Consistent with the potential role of specific nuclear associations of the drug in radioprotection are the observations of a tightly associated form of WR1065 in isolated DNA, nuclei, nucleoids, and nucleoproteins at levels that often appear to saturate with time or drug concentration (4, 9) and the observation, in several studies (Fig. 5, refs. 12, 17), of cell cycle progression delays at

Volume 24, Number 4, 1992

long times after drug removal. The delayed accumulation of cells in S-phase could be the consequence of drug interactions with proteins involved in DNA metabolism. Our previous studies (22, 23) of the DNA supercoiling response of human cells differing in their proliferative activity, transformation state, and radiosensitivity have indicated inherent differences in DNA loop characteristics and nuclear protein composition. The results of the present study, indicating that WR1065 had a preferential radioprotective effect in vitro on both survival and the manifestation of DNA damage at the nucleoid level, are consistent with the hypothesis that cell type differences in chromatin organization and DNA-drug associations could play a role in the selective radioprotection. The cell lines used in this study and other human tumor and normal cell lines currently being examined in our lab should provide a convenient system in which to identify any celltype specific differences in the interaction of WR 1065 with components of the cell nucleus.

REFERENCES 1. Calabro-Jones, P. M.; Aguilera, J. A.; Ward, J. F.; Smoluk,

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

G. D.; Fahey, R. C. Uptake of WR-272 1 derivatives by cells in culture: Identification of the transported form of the drug. Cancer Res. 48:3634-3640;1988. Dean, P. N.; Jett, J. H. Mathematical analysis of DNA distributions derived from microfluorometry. J. Cell Biol. 60: 523-527; 1974. Denekamp, J.; Rojas, A.; Stewart, F. A. Is radioprotection by WR2721 restricted to normal tissues? In: Nygaard, 0. F., Simic, M. G., eds. Radioprotectors and anticarcinogens. New York, NY: Academic Press; 1983:655-679. Grdina, D. J.; Guilford, W. H.; Sigdestad, C. P.; Giometti, C. S. Effects of radioprotectors on DNA damage and repair, proteins, and cell-cycle progression. Pharmac. Ther. 39: 133137;1988. Grdina, D. J.; Nagy, B. The effect of 2-[(aminopropy)amino]ethanithiol (WR1065) on radiation-induced DNA damage and repair and cell progression in V79 cells. Br. J. Cancer 54:933-94 1; 1986. Grdina, D. J.; Nagy, B.; Hill, C. K.; Wells, R. L.; Peraino, C. The radioprotector WR 1065 reduces radiation-induced mutations at the hypoxanthine-guanine phosphoribosyl transferase locus in V79 cells. Carcinogenesis 6:929931;1985. Harris, J. W.; Phillips, T. L. Radiobiological and biochemical studies of thiophosphate radioprotective compounds related to cysteamine. Radiat. Res. 46:362-379; 197 1. Hill, C. K.; Nagy, B.; Peraino, C.; Grdina, D. J. 2-[(aminopropyl)amino]ethanethiol (WR1065) is anti-neoplastic and anti-mutagenic when given during 6oCo y-ray irradiation. Carcinogenesis 7:665-668; 1986. Meechan, P. J.; Vaughan, A. T. M.; Giometti, C. S., Grdina, D. J. Association of WR-1065 with CHO AA8 cells, nuclei, and nucleoids. Radiat, Res. 125: 152- 157; 199 1. Milas, L.; Murray, D.; Brock, W. A.; Meyn, R. E. Radioprotectors in tumor radiotherapy: Factors and settings determining therapeutic ratio. Pharmac. Ther. 39: 179187;1988. Millar, J. L.; McElwain, T. J.; Clutterbuck, R. D.; Wist, E. A. The modification of melphalan toxicity in tumor

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

bearing mice by S-2-(3-aminopropylamino)-ethylphosphorothioic acid (WR2721). Am. J. Clin. Oncol (CCT) 5:321328;1982. Murley, J. S.; Grdina, D. J.; Meechan, P. J. Accumulation of CHO cells in G2 phase following exposure to WR-1065. Radiat. Res. 126:223-228; 199 1. Murray, D.; Prager, A.; Milas, L. Radioprotection ofcultured mammalian cells by the aminothiols WR-1065 and WR255591: Correlation between protection against DNA double-strand breaks and cell killing after y radiation, Radiat. Res. 120:154-163;1989. Murray, D.; VanAnkeren, S. C.; Milas, L.; Meyn, R. Radioprotective action of WR- 1065 on radiation-induced DNA strand breaks in cultured Chinese hamster ovary cells. Radiat. Res. 113:155-170;1988. Phillips, T. L. Rationale for initial clinical trials and future development of radioprotectors. Cancer Clin. Trials 3: 165173;1980. Roti Roti, J. L.; Wright, W. D. Visualization of DNA loops in nucleoids from Hela cells: Assays for DNA damage and repair. Cytometry 8:461-467;1987. Sigdestad, C. P.; Guilford, W.; Perrin, J.; Grdina, D. J. Cell cycle redistribution of cultured cells after treatment with chemical radiation protectors. Cell Tissue Kinet. 2 1: 193200;1988. Sigdestad, C. P.; Treaty, S. H.; Knapp, L. A.; Grdina, D. J. The effect of 2-[(aminopropyl)ethanethiol (WR-1065) on radiation induced DNA double strand damage and repair in V79 cells. Br. J. Cancer 55:477-482;1987. Smoluk, G. D.; Fahey, R. C.; Calabro-Jones, P. M.; Aguilera, J. A.; Ward, J. F. Radioprotection of cells in culture by WR272 1 and derivatives: Form of the drug responsible for protection. Cancer Res. 48:3641-3647;1988. Smoluk, G. D.; Fahey, R. C.; Ward, J. F. Equilibrium dialysis studies of the binding of radioprotector compounds to DNA. Radiat. Res. 107: 194-204; 1986. Smoluk, G. D.; Fahey, R. C.; Ward, J. F. Interaction of glutathione and other low-molecular-weight thiols with DNA: Evidence for counterion condensation and coion depletion near DNA. Radiat. Res. 114:3-10;1988.

Selective radioprotection by WR1065 0 X. ZHANGel al.

22. Taylor, Y. C.; Duncan, P. G.; Zhang, X.; Wright, W. D. Differences in the DNA supercoiling response of irradiated cell lines from ataxia-telangiectasia versus unaffected individuals. Int. J. Radiat. Biol. 59:359-37 1; 199 I. 23. Taylor, Y. C.; Zhang, X.; Parsian, A. J.; Duncan, P. G. Image analysis-based measurement of DNA supercoiling changes in transformed and nontransformed human cell lines. Exp. Cell Res. 197:222-228; 1991. 24. Vaughan, A. T. M.; Grdina, D. J.; Meechan, P. J.; Milner, A. E.; Gordon, D. J. Conformational changes in chromatin structure induced by the radioprotective aminothiol, WR 1065. Br. J. Cancer 60:893-896;1989. 25. Washburn, L. C.; Carlton, J. E.; Hayes, R. L. Distribution of WR-272 1 in normal and malignant tissues of mice and

719

rats bearing solid tumors: Dependence on tumor type, drug dose and species. Radiat. Res. 59:475-483;1974. 26. Yuhas, J. M. Active versus passive absorption kinetics as the basis for selective protection of normal tissues by S-2(3-aminopropylaminoj-ethylphosphorothioic acid. Cancer Res. 40: 15 19-1524;1980. 21. Yuhas, J. M.; Spellman, J. M.; Culo, F. The role of WR2721 in radiotherapy and/or chemotherapy. Cancer Clin. Trials 3:21 l-216;1980. 28. Zheng, S.; Newton, G. L.; Gonick, G.; Fahey, R. C.; Ward, J. F. Radioprotection of DNA by thiols: Relationship between the net charge on a thiol and its ability to protect DNA. Radiat. Res. 114:l l-27:1988.