A prevention strategy for circumventing drug resistance in cancer chemotherapy.

The development of drug resistance is considered to be a major cause for the failure of chemotherapy in a number of types of cancer, including ovarian, breast and lung. Most previous research has focused on approaches to reverse drug resistance once it has arisen, that is, on the use of agents which can make drug-resistant tumors more sensitive to chemotherapy. Unfortunately, this approach has thus far met with only limited clinical success. Because of the prevalence of drug resistance in cases of advanced cancer, there exists an urgent need to develop new approaches to dealing with this problem. We have hypothesized the feasibility of an alternative approach: the use of specific agents to prevent the development of resistance before it arises. Our initial studies to examine this hypothesis have focused on ovarian cancer. We have designed both in vitro and in vivo systems in which resistance develops rapidly after exposure of tumor cells or xenografts to melphalan or cisplatin. Using these systems we have shown that two selenium compounds, selenite and selenomethionine are able to prevent the induction of resistance. Furthermore, inclusion of selenite in a chemotherapeutic protocol can result in a significant enhancement of the efficacy of cisplatin in suppressing the growth of human ovarian tumor xenografts. These results have supported the idea that prevention may be a useful new approach to the problem of drug resistance in cancer chemotherapy.

Selenium compounds prevent the induction of drug resistance by cisplatin in human ovarian tumor xenografts in vivo.

PURPOSE: The development of drug resistance is a major cause for the failure of chemotherapy, particularly in ovarian cancer. Most previous research has focused on approaches to reverse drug resistance once it has arisen, that is, on the use of agents which can make drug-resistant tumors more sensitive to chemotherapy. We have suggested the feasibility of an alternative approach: the use of specific agents to prevent the development of resistance. METHODS: We designed an in vivo system to assay for the ability of compounds to prevent the induction of resistance by cisplatin. In this system, mice bearing tumors (which originated from A2780 human ovarian tumor cells) were treated with a low dose (2.6 mg/kg) of cisplatin and the tumors rapidly developed resistance to subsequent cisplatin treatment. Cell lines initiated from these tumors retained the resistant phenotype even after several months in culture. RESULTS: When either selenite or selenomethionine were administered (i.p., 1.5 mg/kg) close to the time of the initial cisplatin treatment, the induction of resistance was prevented. Similar treatments with sulfite or methionine had no effect on the induction of resistance by cisplatin. Studies in cells from treated tumors have indicated that the selenium compounds may prevent the induction of resistance by preventing a cisplatin-induced increase in glutathione level. CONCLUSIONS: Selenium compounds specifically prevent the induction by cisplatin of drug resistance in human ovarian tumors in vivo.

Rapid development of glutathione-S-transferase-dependent drug resistance in vitro and its prevention by ethacrynic acid.

Exposure of A2780 human ovarian tumor cells to a low concentration of melphalan in vitro for 7 days resulted in the development of melphalan resistance. This resistance was not a stable characteristic of the cells since it was lost after 2 weeks in culture in the absence of drug. The melphalan-resistant cells exhibited significant cross-resistance to cisplatin but only minor cross-resistance to doxorubicin. The resistant cells had elevated levels of glutathione-S-transferase activity and mRNA. Exposure of the cells to the ethacrynic acid resulted in a decrease in enzyme activity as well as a reversal of their drug-resistant phenotype, indicating that the enzyme is involved in the resistance. When ethacrynic acid was present during the 7-day exposure of the cells to melphalan, the development of drug resistance was prevented. This system may serve as a useful preliminary step in screening for agents which can prevent the development of chemotherapy-induced drug resistance in human cancer.

Treatment of human ovarian tumor xenografts with selenite prevents the melphalan-induced development of drug resistance.

Human ovarian tumors (derived from A2780 cells), growing as subcutaneous xenografts in immunodeficient mice, develop melphalan-resistant cells after a single treatment of the tumor-bearing animal with the drug [Caffrey, Zhang and Frenkel, submitted for publication]. Treatment of the animals with selenite by i.p. injection prevented the development of primary resistance to melphalan as well as cross-resistance to cisplatin. Selenite treatment also prevented the melphalan-induced increase in the cellular level of glutathione. In contrast, selenite administered s.c. or in drinking water had relatively little effect on the development of resistance. The results in this animal model suggest that selenite may be clinically useful in preventing the development of drug resistance during chemotherapy of cancer.

Rapid development of drug resistance in human ovarian tumor xenografts after a single treatment with melphalan in Vivo.

Human ovarian tumors from A2780 cells were grown as xenografts in immunodeficient mice, treated with a single i.p. dose of melphalan and tumor cells were removed and placed into tissue culture. The cells from the treated tumors exhibited an approximately 2-fold resistance to melphalan in vitro compared to cells taken from untreated tumors. This degree of resistance was similar to that of cells from tumors formed from melphalan-resistant A2780-ME cells. The cells from the treated tumors were also resistant to cisplatin but not to doxorubicin. They contained approximately 2-fold higher levels of glutathione than cells from the untreated tumors. Exposure of the cells to buthionine sulfoximine (a specific inhibitor of glutathione biosynthesis) eliminated the difference in glutathione levels as well as the difference in sensitivity to melphalan. When tumor-bearing animals were treated with buthionine sulfoximine in addition to melphalan the resulting tumor cells were not resistant to the drug. Resistance could also be demonstrated in the tumors themselves in vivo: the growth of previously untreated tumors was severely inhibited by a high dose of melphalan (11.7 mg/kg) administered i.p. to the animals, whereas the growth of tumors which had received prior treatment with melphalan was unaffected by the subsequent high dose. The rapid development of drug-resistant tumor cells after a single drug treatment in vivo makes this an excellent system for the investigation of the mechanisms by which resistance develops as well as for use in the screening for agents which can prevent it.

Prevention of the development of melphalan resistance in vitro by selenite.

Exposure of A2780 human ovarian tumor cells to a low concentration of melphalan in vitro for 7 d results in the development of melphalan resistance, which is dependent on elevated cellular levels of glutathione and glutathione S-transferase. The inclusion of selenite (at concentrations as low as 0.2 microM) during the exposure to melphalan completely prevented the development of resistance. Selenite did not prevent the melphalan-induced increase in glutathione, but it did prevent the increase in the activity of glutathione S-transferase. It also prevented the increase in the expression of the glutathione S-transferase gene, suggesting that this may be the mechanism by which it prevents the development of melphalan resistance. The results of this in vitro study suggest that selenite may prove to be useful in preventing the development of drug resistance in vivo.

Sensitivity of melphalan-resistant tumors to selenite in vivo.

Previous studies have demonstrated that melphalan-resistant human ovarian tumor cells exhibit a higher degree of sensitivity to the cytotoxic effects of selenite in vitro than comparable drug-sensitive cells (P.B. Caffrey, G.D. Frenkel, Selenite cytotoxicity in drug resistant and non-resistant human ovarian tumor cells, Cancer Res. 52 (1992) 4812-4816; P.B. Caffrey, G.D. Frenkel, The development of drug resistance by tumor cells in vitro is accompanied by the development of sensitivity to selenite, Cancer Lett. 81 (1994) 59-65). We have now examined the sensitivity of drug-resistant tumors to selenite in vivo. A2780 human ovarian tumor cells, or their melphalan-resistant derivative (A2780ME) cells were injected subcutaneously into nude mice and the resulting tumors were found to be melphalan-sensitive and -resistant, respectively, in vivo. Treatment with selenite (2 mg/kg Se s.c.), which had no overt toxic effect on the animals, resulted in a significant decrease in the rate of growth of the melphalan-resistant tumors, but not on the rate of growth of the drug-sensitive tumors. Thus, melphalan-resistant ovarian tumors are also more sensitive to selenite treatment in vivo.

The development of drug resistance by tumor cells in vitro is accompanied by the development of sensitivity to selenite.

The effects of selenite on cell viability and proliferation in a line of drug-sensitive human ovarian tumor (A2780) cells were compared with its effects on a melphalan-resistant derivative of these cells (A2780-ME) which had been developed in vitro (Hamilton et al. (1985) Biochemical Pharmacol., 34, 2583-2586). With the A2780-ME cells there was a 50% decrease in the number of viable cells (i.e. which exclude Trypan Blue dye) after exposure to less than 100 microM selenite for 6 h. In contrast, exposure to more than 300 microM selenite was required to achieve the same effect in the parent line. Similarly, exposure to 10 microM selenite resulted in a 50% decrease in A2780-ME cell proliferation, whereas this treatment had only a small inhibitory effect on proliferation of the parent cells. Thus, the development of melphalan resistance in vitro was accompanied by the development of selenite sensitivity. Pre-exposure of the two cell types to buthionine sulfoximine eliminated the difference in their intracellular glutathione levels, as well as most of their differential sensitivity to selenite. Furthermore, the two cell types did not exhibit a difference in sensitivity to selenodiglutathione, the product of the reaction of selenite with glutathione. Thus, the increase in intracellular glutathione, which has been shown to be responsible for the development of drug resistance in these cells is also responsible for the development of selenite sensitivity.

Selenite-induced inhibition of colony formation by buthionine sulfoximine-sensitive and resistant cell lines.

We previously demonstrated that treatment of HeLa cells with buthionine sulfoximine (BSO), which decreases the level of cellular glutathione, resulted in a decrease in the potency of selenite in inhibiting cell colony formation. We have now examined the effect of selenite on normal human lung fibroblast (CCL-210) cells, which resemble HeLa cells in their sensitivity to BSO, and on human lung adenocarcinoma (A549) cells, which are relatively insensitive to BSO. We have found that BSO treatment caused an approximately fourfold decrease in selenite potency in the CCL-210 cells, but had no significant effect on its potency in A549 cells. These results support the hypothesis that for selenite to exert its cytotoxic effect, it must undergo the reaction with an SH compound to form the selenotrisulfide. As a result of the lower sensitivity of the tumor cells to BSO, it was possible to achieve a large differential sensitivity to the cytotoxic effect of selenite.

Selenite cytotoxicity in drug resistant and nonresistant human ovarian tumor cells.

Our previous studies on selenite cytotoxicity led us to hypothesize that drug resistant tumor cells with high intracellular glutathione will exhibit a high degree of sensitivity to selenite. To examine this we studied the effects of selenite on drug resistant human ovarian tumor (NIH:OVCAR-3) cells in three assays of cytotoxicity: proliferation; cell viability (trypan blue exclusion); and attachment to a solid matrix. The cells were sensitive to low levels of selenite: concentrations as low as 5 microM inhibited cell proliferation and attachment; and viability was decreased by concentrations as low as 20 microM. In each of these assays the NIH:OVCAR-3 cells were more sensitive to selenite than were drug sensitive human ovarian tumor (A2780) cells. These results suggest the potential for the utilization of selenite in the treatment of some drug resistant tumors.

Inhibition of cell colony formation by selenite: involvement of glutathione.

Selenium is an essential trace element that has been shown to have anticarcinogenic activity. One mechanism that has been proposed for this activity is a cytotoxic effect of selenium on tumor cells. As a means of assessing its cytotoxicity, we have examined the effect of selenite on tumor cell viability, using as an assay the ability of the cells to form colonies. We have found that brief exposure of HeLa cells to micromolar concentrations of selenite resulted in significant inhibition of colony formation, indicating that this is an assay for selenite cytotoxicity that is more sensitive than those that have been employed previously. In order to investigate the involvement of cellular glutathione in selenite cytotoxicity, we treated cells with buthionine sulfoximine (BSO) before selenite exposure. This treatment, which resulted in a 7-fold reduction in the level of intracellular glutathione, also caused a significant decrease in the inhibitory effect of selenite on colony formation. However, when cells were exposed to selenite that had previously been reacted with glutathione, the BSO-induced decrease in cytotoxicity was eliminated. In contrast, reaction of selenite with other sulfhydryl compounds, such as cysteine and mercaptoethylamine, did not restore its potency in BSO-treated cells. The simplest explanation for these results is that, for selenite to exert its inhibitory effect, it must react with intracellular glutathione to form the selenodiglutathione derivative.

Selenium deficiency induced by zinc deprivation in a crustacean.

For intact daphnids reared in circumstances of controlled trace element exposure, one consequence of insufficient zinc (Zn) is an increased demand on the animal's pool of available selenium (Se). This demand is manifested by the type of cuticle deterioration associated with Se deficiency and by a depression of reproduction. In the presence of 25 parts per billion (ppb) Zn, 1 ppb Se eliminates these symptoms. In the absence of detectable Zn, 5 ppb Se eliminates overt cuticle damage and substantially increases reproduction. A shortening of life span resulting from Zn deprivation is not ameliorated by Se addition. We suggest that the interplay between Zn and Se concentrations reflects an underlying interplay between interdependent, but distinct, metabolic pathways; i.e., (for Se) glutathione peroxidase and (for Zn) Cu,Zn-superoxide dismutase-each offering protection against free radical damage. Because they are not necessarily localized in a given tissue, the key to recognition of such subtle, complex trace nutrient interactions has been use of intact animals in circumstances of control previously attainable only in tissue cultures.