The redox status of glutathione (L-gamma-glutamyl-L-cysteinylglycine, GSH) plays an important role in a number of different cellular reactions including cellular oxidative stress. Using A549 human small cell Lung carcinoma fibroblasts, toe investigated the role of exogenous GSH on the intracellular GSH/glutathione disulfide (oxidized glutathione, GSSG) redox ratio in GSH-depleted cells by treating with L-buthionine-(S,R)-sulfoximine (BSO). GSH levels decreased after BSO treatment. Although BSO is a well-recognized inhibitor of GSH biosynthesis and has no known effect on GSSG reductase, surprisingly, the levels of GSSG also decreased. Incubation of control or GSH-depleted cells with exogenous GSH did not alter intracellular GSH or GSSG levels to any significant extent. Therefore, the ratio of GSH/GSSG also was not altered significantly either in the controls or the BSO-treated cells. It appears that there may be cellular homeostatic mechanisms that would maintain a constant GSH/GSSG ratio, irrespective of the changes in intracellular GSH concentration.

Alterations in intracellular glutathione (GSH) content are known to affect intrinsic responses to ionizing radiation. More recently, it became apparent that radiation responses may depend also on the expression of specific oncogenes, including ras. These findings, suggesting a possible link between GSH and ras, led us to examine the effect of various GSH modulators on ras expression. Treatment of c-Ha-ras-transformed NIH 3T3 cells with L-buthionine S'R'-sulfoximine, dimethylfumarate, or N',N'-1,3-bis(trans-4-hydroxycyclohexyl)-N'-nitrosourea resulted in dose- and time-dependent reduction in ras mRNA steady-state levels followed by a decrease in ras-encoded p21 protein production. The effect on ras correlated with the extent of GSH decline, was common to different members of the ras family, and was independent of the mode of oncogene activation or cell phenotype. Indeed, similar drug effects were observed with murine cells in which overexpression of the c-Ha-ras proto-oncogene was due to transcriptional activation (PR4, nontumorigenic) or gene amplification (NIH 136, tumorigenic) and with malignant cells expressing a mutated Ha-ras (RS504). Moreover, N-ras, EJras, and Ki-ras in human tumor cells were similarly affected. Molecular analysis revealed a significant decrease in ras mRNA half-life in cells subjected to GSH inhibition, an effect that required de novo protein synthesis, but there was no change in the rate of gene transcription. These results indicate that pharmacological manipulation of cellular GSH content can down-regulate ras expression at the posttranscriptional level by destabilizing ras transcripts. The potential clinical implications are discussed.

Critical cellular biomolecules are damaged by ionizing radiation either directly, or indirectly by radicals generated from the radiolysis of cellular water (1,2). The free radicals thus produced can further interact among themselves or with cellular oxygen to perpetuate the radiation effects. Depending on the linear energy transfer (LET), dose rate, and duration of exposure these effects may vary. At the molecular level it is generally believed that damage to DNA is the critical event. Damage to DNA can be detected at the cellular level by chromosomal aberrations (3) and micronuclei formation (4). It should be possible to do this semi-invasively using lymphocytes from peripheral blood. Radiation damage may vary from transformation and mutagenesis in in vitro systems to carcinogenesis and cataractogenesis in the long term and lethality due to acute exposures. Lethality may be due to bone marrow depletion (hematopoietic syndrome), destruction of gastrointestinal mucosa (gastrointestinal syndrome), and neurovascular destruction (central nervous system syndrome) (5). Exposure to a nuclear accident may also cause injuries from shock, burns, or wounds that may accentuate lethal effects. Protection from these effects and mechanisms of protection are of importance in determining suitable regimens for different radiation exposure scenarios, such as radiation therapy, nuclear accidents, and manned space missions. The purpose of this paper is to review some of the new data obtained on radioprotection with agents which show a small margin of radioprotection and to show a possible correlation between antioxidant mechanisms and radioprotection based on these data.

Cells cope with radiation damage through several mechanisms: (1) increased DNA repair activity, (2) scavenging and inactivation of radiation-induced radical molecules, and (3) entry into a G0-like quiescent state. We have investigated a chromosomal rearrangement to elucidate further the molecular and genetic mechanisms underlying these phenomena. A mutant of Escherichia coli JM83 (phi 80dlacZ delta M15) was isolated that demonstrated significantly increased resistance to both ionizing and ultraviolet radiation. Surviving fractions of mutant and wild-type cells were measured following exposure to standardized doses of radiation. Increased radioresistance was directly related to a chromosomal alteration near the bacteriophage phi 80 attachment site (attB), as initially detected by the LacZ- phenotype of the isolate. Southern hybridization of chromosomal DNA from the mutant and wild-type E. coli JM83 strains indicated that a deletion had occurred. We propose that the deletion near the attB locus produces the radioresistant phenotype of the E. coli JM83 LacZ- mutant, perhaps through the alteration or inactivation of a gene or its controlling element(s).

In this study, we present data which suggests that there is a correlation between EJras mRNA expression and increased radioresistance. The experimental system c onsisted of two human osteof5arcoma (HOS subclones transfected differential levels of tad. Northern blot analysis showed that the differential expreaaion of raa waa concomitant with an increase in cellular radioresistance as measured by a marked increase in the Do and D. The increased radioresistance was not ex%lained by variations in cell cycle nor via increaaes in antioxidant enzymes. The close correlation between an increased ras mRNA level and radioresistance suggests that the EJraa protein (p21) may participate in the response of human tumor cells to ionizing radiation. with the mutate h ras, which express indicated that there may be a threshold level of a expression necessary for cells to exhibit the radioresistant phenotype (41. While other oncogenes have been implicated in an increase in radioresistance, the ~(LB family appears to play a key role in intrinsic cellular radioresistance.

The cellular Ha-ras oncogene, activated by missense mutations, has been implicated in intrinsic resistance to ionizing radiation. This study shows that the overexpression of the unmutated gene (proto-oncogene) may also be involved in how the cells respond to radiation. The experimental system consisted of mouse NIH 3T3-derived cell lines which carry multiple copies of a transcriptionally activated human c-Ha-ras proto-oncogene. Both tumorigenic (RS485) and revertant nontumorigenic subclones (PR4 and 4C3) which have high levels of ras expression exhibited a marked increase in radioresistance as measured by D0 compared to control NIH 3T3 cells. Other nontransformed cells with elevated levels of ras (phenotypically revertant line 4C8-A10) also had a significantly increased resistance to radiation, further indicating an association between ras and radioresistance. The increased radioresistance of the RS485 and phenotypic revertants could not be explained by a differential expression of the myc or metallothionein I genes or by variations in cell cycle. The correlation between increased ras proto-oncogene expression and radioresistance suggests that the ras encoded p21, a plasma membrane protein, may participate in the cellular responses to ionizing radiation.

The most effective radioprotective agents exhibit toxicities that can limit their usefulness. It may be possible to use combinations of agents with different radioprotective mechanisms of action at less toxic doses, or to reduce the toxicity of the major protective compound by adding another agent. With regard to the latter possibility, improved radioprotection and reduced lethal toxicity of the phosphorothioate WR-2721 was observed when it was administered in combination with metals (selenium, zinc or copper). The known mechanisms of action of potential radioprotective agents and varying effects of different doses and times of administration in relation to radiation exposure must be considered when using combined-agent regimens. A number of receptor-mediated protectors and other biological compounds, including endotoxin, eicosanoids and cytokines, have at least an additive effect when administered with thiol protectors. Eicosanoids and other bioactive lipids must be administered before radiation exposure, whereas some immunomodulators have activity when administered either before or after radiation exposure. For example, the cytokine interleukin-1 administered simultaneously with WR-2721 before irradiation or after irradiation enhances the radioprotective efficacy of WR-2721. The most effective single agents or combinations of protectors result in a decrement in locomotor activity, an index of behavioral toxicity. Recent evidence indicates that administration of the CNS stimulant caffeine mitigates the behavioral toxicity of an effective radioprotective dose of the phosphorothioate WR-3689 without altering its radioprotective efficacy. These examples indicate that the use of combinations of agents is a promising approach for maximizing radioprotection with minimal adverse effects.

We will review the relationships between glutathione (GSH), protein thiols, and cellular responses to radiation, peroxides, and peroxide-producing drugs. Our primary interest involves the behavior of sulfhydryls as electron and hydrogen carriers, and their capacity to protect various target molecules against radiation and peroxidative damage. We used reagents such as L-buthionine sulfoximine (LBSO), alone and in combination with N-ethyl maleimide (NEM), diamide, and dimethylfumarate, to decrease GSH so that it could no longer participate in the electron transfer reactions. Our results indicate that aerobic sensitization produced by GSH depletion can be further enhanced if electron-accepting agents, such as tertiary butyl hydroperoxide (t-BOOH), are present during irradiation. Hydroperoxide is a substrate for glutathione peroxidase and diverts electrons and hydrogen away from target molecules during its reduction. Sensitivity to radiation seems to be due to the inhibition of the mitochondria's capacity to reduce hydroperoxide. We will also report the mitochondria's ability to reduce the oxygen radicals produced by radiation and drugs. Data also indicate that t-BOOH oxidizes protein thiols which are enzymatically involved in repair of DNA damage.

In summary, there are theoretically several ways by which a cell's content of DNA repair genes or other radioprotective proteins can be increased. At present the most feasible way to do so seems to be to search for repair systems, such as the adaptive repair observed in human lymphocytes, whose activity can be increased by exogenous inducing agents. Now that this effect has been established in human lymphocytes, a search for similar systems in other cells should begin. Transfection of DNA repair genes into bone marrow cells that could be used to make the blood-forming organs more radioresistant is actually very close to being possible; the problem is more a matter of morality than of feasibility. The practicability of amplification of DNA repair genes or genes coding for other radioprotective proteins in the cells of a whole organism is at present close to zero. It is clear, however, that continued research on the basic mechanisms of cellular radiosensitivity could eventually lead to genetically engineered radioresistance in humans, but only if society approves.

The effect of changes in both the intracellular glutathione (GSH) concentration and the concentration of extracellular reducing equivalents on the aerobic radiosensitization was studied in three cell lines: CHO-10B4, V79, and A549. Intracellular GSH was metabolically depleted after the inhibition of GSH synthesis by buthionine sulfoximine (BSO), while the extracellular environment was controlled through the replacement of growth medium with a thiol-free salt solution and in some experiments by the exogenous addition of either GSH or GSSG. Each of the cell lines examined exhibited an enhanced aerobic radioresponse when the intracellular GSH was extensively depleted (GSH less than 1 nmol GSH/10(6) cells after 1.0 mM BSO/24 h treatment) and the complexity of the extracellular milieu decreased. Although the addition of oxidized glutathione (5 mM GSSG/30 min) to cells prior to irradiation was without effect, much or all of the induced radiosensitivity was overcome by the addition of reduced glutathione (5 mM GSH/15 min). However, the observation that the exogenous GSH addition restores the control radioresponse without increasing the intracellular GSH concentration was entirely unexpected. These results suggest that a number of factors exert an influence on the extent of GSH depletion and determine the extent of aerobic radiosensitization. Furthermore, the interaction of exogenous GSH with--but without penetrating--the cell membrane is sufficient to result in radiorecovery.

Our data show that A549 cells are increasingly radiosensitive with prolonged exposure to L-BSO. The resulting glutathione and protein thiol depleted cells show both loss of shoulder and slope modification. Furthermore, there is an increase in single strand DNA breaks and irrepairable cross-linking. The aerobic radiation damage in the thiol depleted state appears to be different from that obtained with hypoxic cells. Any postulated role for GSH in reducing or preventing peroxidative radiation damage must also include protection against single strand DNA breaks as well as involvement in repairing DNA-protein cross-links. The latter effect may be related to decreased protein thiol content as reflected in a decreased enzyme capacity to repair DNA damage.

There is considerable interest in the role of thiols and redox enzymes as modulators of chemo- and radioresponses from both therapeutic and theoretical perspectives. The metabolism of GSH and the functioning of the GSH redox cycle can have a significant impact on the response of cells to both chemical and radiation stresses. From this rapidly developing research area, attention has been focused on recent progress in several areas: (a) the use of GSH-depletion to reverse chemo-induced resistance to further chemical treatment; (b) the significance of GSH transferase activity in chemoresistant cells; (c) the effectiveness of GSH-depletion in sensitizing radioresistant cells; and (d) the evidence for an alternative GSH-linked recovery pathway. From an examination of these areas, it seems warranted to conclude the following: First, manipulation of the redox state is effective in enhancing the cytotoxic effect of radiation and some chemotherapeutic agents in radio- and chemoresistant cells; second, additional possibilities for a therapeutic advantage may be gained as new redox modifiers are developed; and last, membrane domains may provide important sites for redox alterations by naturally occurring substrates or by drugs specifically designed to alter this interaction.

In a study of the relation between chronic inflammation and carcinogenesis, C3H mouse fibroblasts of the 10T 1/2 clone 8 line (10T 1/2 cells) were exposed to human neutrophils stimulated to synthesize reactive oxygen intermediates or to a cell-free enzymatic system generating superoxide (xanthine oxidase plus hypoxanthine). After exposure, the 10T 1/2 cells were either placed in tissue culture or immediately injected into athymic nude mice. Both malignant and benign tumors developed in the mice injected with treated cells, but not in those injected with control cells; in one instance cells grown from one of the benign tumors subsequently developed a malignant phenotype. Malignant transformation was also observed in treated cells in the experiments in vitro.

Stimulated human phagocytes produce sister chromatid exchanges in cultured mammalian cells by a mechanism involving oxygen metabolites. Experiments were designed to determine whether antioxidants inhibit this process. Superoxide dismutase, catalase, and hydroxyl radical scavengers (benzoate, mannitol) protected target Chinese hamster ovary cells from phagocyte-induced sister chromatid exchanges, implicating the involvement of hydroxyl radicals in this chromosomal damage. N-acetylcysteine and beta-carotene were also protective. alpha-Tocopherol (greater than 5 microM) protected target cells exposed to phagocytes but not to enzymatically generated oxidants when the vitamin was added just before the source of oxygen radicals, suggesting, as reported by others, that the principal action of tocopherol in this setting was to inhibit the release of oxidants from phagocytes. On the other hand, cultivation of target cells with supplemental tocopherol protected them from the toxic effects of the enzymatic oxidant-producing system, indicating a role for membrane-associated free radicals in the mechanism of sister chromatid exchange induction. Low concentrations of sodium selenite (0.1-1.0 microM) protected the target cells. However, higher concentrations (10 microM) of selenite had no effect on oxidant-induced sister chromatid exchange formation, and 0.1 mM selenite increased the number of exchanges. Sodium selenite concentrations of 0.1 mM also decreased the intracellular glutathione concentration of target cells during an oxidant stress, and reducing target cell glutathione concentrations with buthionine sulfoximine increased their sensitivity to oxygen-related chromosomal damage. Therefore, the potentiation of oxygen radical-induced chromosomal damage observed with high concentrations of selenite may result from a decrease in the thiol antioxidant defense systems within the cell. The findings suggest that the hydroxyl radical has an important role in the production of phagocyte-induced cytogenetic injury, membrane-derived intermediates may be involved, depletion of intracellular glutathione renders cells more susceptible to this injury, and supplementation of target cells with antioxidants can protect them from oxygen radical-generated chromosomal injury.

We have measured the rate of GSH resynthesis in plateau phase cultures of A549 human lung carcinoma cells subjected to a fresh medium change. Buthionine sulfoximine (BSO) blocks this resynthesis. Diethyl maleate (DEM) causes a decrease in accumulation of GSH. If DEM is added concurrently with BSO there is a rapid decline in GSH that is maximal in the presence of 0.5 mM DEM. GSH depletion rapidly occurs when BSO is added to log phase cultures which initially are higher in GSH content. Twenty-four hr treatment of A549 cells with BSO results in cells that are more radiosensitive in air and show a slight hypoxic radiation response. A 2 hr treatment with either 0.25 mM or 0.5 mM DEM results in some hypoxic sensitization and little increase in the aerobic radiation response. The 24 hr BSO + 2 hr DEM treatment sensitizes hypoxic cells to a greater degree than either agent alone but does not increase the aerobic response more than is seen with BSO alone. Cells treated simultaneously with BSO + DEM show little increase in the hypoxic radiation response, compared to DEM alone, but are more sensitive under aerobic conditions. Decreased cell survival for aerobically irradiated log phase A549 cells occurs within minutes after addition of a mixture of BSO + DEM. The decreased cell survival following aerobic irradiation, after prolonged treatment with BSO or acute exposure to BSO + DEM, may be in part due to inhibition of glutathione peroxidases. For example, glutathione-S-transferase, known to have glutathione peroxidase activity (non-selenium), is nearly completely inhibited by the BSO treatments. In addition, cellular capacity to react with peroxide (glutathione peroxidase, selenium containing) was also inhibited. We suggest that the enhanced aerobic radiation response is related to an inability of GSH depleted cells to inactivate either peroxy radicals or hydroperoxides that may be produced during irradiation of BSO treated cells. Furthermore, enhancement of the aerobic radiation response may be useful In vivo if normal tissue responses are not also increased.