9 1985 by The Humana Press Inc. All rights of an)" nature whatsoever reserved. 0163--4984/85/8008-0019503.40
Selenium Retention and Inhibition of Cell Growth in Mouse Mammary Epithelial Cell Lines In Vitro DANIEL MEDINA,* DAVID MORRISON, AND CAROLJ. OBORN Department of Cell Biology, Baylor College of ~ledicine, Houston, TX 77030 Received July 24, 1984; Accepted February 25, 1985
ABSTRACT The steady state levels of growth inhibitory doses of inorganic selenium were examined in five different mammary epithelial cell lines: MOD, COMMA-D, COMMA-F, COMMA-T, and YN-4. The retention of selenium was monitored using a radioactive isotope, ;SSe. Growth inhibition correlated with high levels of selenium in the cell. Generally, the retention of intracellular selenium was not dependent upon cell density,, cell number, net growth rate, or tumorigenicity of the mammary cell lines. One cell line, COMMA-D, exhibited an unique response wherein the amount of selenium retained was low and the growth inhibitory effects of selenium were negligible when the cells were exposed to selenium at low density. However, at high cell densities, the COMMA-D cells responded like the other four cell lines. The growth inhibitory effect of selenium was reversible; upon removal of selenium from the medium, cells start synthesizing DNA within 24 h. The retention of selenium was influenced by constituents in the growth medium. In particular, cysteine, but not methionine, purines, or pyrimidines altered selenium retention and counteracted the growth inhibitory effects of selenium. These results indicated that the mammary cell lines, particula~ COMMA-D and MOD are good model systems to examine the uptake, retention, localization, and function of inorganic selenium under conditions where it acts as a growth inhibitory agent. * Author to whom all correspondence and reprint requests should be addressed. Biological Trace Element Research
19
Vol. 8, 1985
?4edina, Morrison, and Oborn
20
Index Entries: Selenium retention, and growth inhibition in mammary cells; reversibility, of Se-inhibited growfl~ in mammary cells; cell culture, Se retention and growth inhibition in mammaD'; mouse mammary cells, Se retention and growth of; mammary cells, Se retention and growth in mouse; epithelial ceils, Se retention and growth in mouse mammary.
INTRODUCTION The inhibition of m a m m a D' tumorigenesis by inorganic selenium a d d e d to the diet or water is a well-documented fact in both the m o u s e and rat m a m m a r y tumor models (1-9). The m e c h a n i s m s of the chemopreventive effects of selenium supplementation have not been resolved; however, it appears that regulation of glutathione peroxidase and inhibition of lipid peroxidation may not be involved (10-15). In order to examine at a cellular and biochemical level the possible m e c h a n i s m s of action of selenium an in vitro model system has been developed over the past several years. Previous results have demonstrated that m a m m a r y epithelial cell lines as well as cell lines of other organs, are sensitive to the growth inhibitory effects of selenium (16-23). The results so far have indicated that low doses (5 x 10-SM Se) stimulate cell growth, whereas high doses (5 x 10-6M Se) inhibit cell growth, but in a reversible manner. The inhibition of m a m m a r y cell growth correlated with a decrease in the uptake of 3H-thymidine into DNA, a decreased DNA-labeling index, and a blockage of cells in the S / G 2 phases of the cell cycle (20,21,24). In vitro cell lines afford a simple and quantifiable model system to examine several aspects of selenium function. Previous experiments examined the retention and localization of low doses of selenium using 75Se as a tracer for selenium (25). Those results indicated that there was a progressive increase in the a m o u n t of selenium retained by the m a m m a r y cells over 24 h and the majority of selenium localized in the cytoplasm. In the experiments reported herein, the intraceUular retention of selenium was examined in five different m a m m a r y epithelial lines u n d e r conditions in which the growth of the cells was inhibited by selenium. The reversibility of selenium inhibition of growth as well as some factors that influenced selenium retention were also examined in specific cell lines.
MATERIAL AND METHODS Ceil Lines The origin and cellular characteristics of mouse m a m m a r y epithelial cell lines YN-4 and COMIVIA-D have been detailed in (20,21,26,27). The COMMA-T and COMMA-F lines were derived from early passage COMMA-D by partial trypsinization (COMMA-T) and by plating spontaneous floaters (COMMA-F) (unpublished data). The latter two cell lines Biological Trace Element Research
VoL 8. l g85
Selenium Retention in Mammary Cells
21
exhibit properties of epithelial cells, but do not produce normal mamma W gland upon transplantation into syngeneic mice. The MOD cell line was derived from a D2 mammary tumor and produces mammary adenocarcinomas upon transplantation into syngeneic mice.
Cell Culture The cell lines were routinely grown and maintained in Falcon plastic flasks in Dulbecco's Modified Eagles Medium (DMEM) with 10% fetal bovine serum (FBS) (Hy-clone Inc., Logan, Utah), insulin (5 >g/mL), gentamicin (50 I,g/mL), and N-2-hydroxyethylpiperazine-N'-2-ethane sulfonic acid buffer (HEPES). Cultures were maintained at 37~ in air: CO2 (92.5:7.5) and subcultured weekly. All cell lines were epithelial, as judged by the presence of keratin intermediate filaments, the absence of vimentin intermediate filaments, and the ability to form "domes" at high saturation densities. For experiments, cells were collected from the parent flasks by trypsinization, washed twice in DMEM with serum and insulin, and plated at 3 x 10 4 cells/cm2 into 35 mm Falcon Petri dishes. For the experiments shown in Tables 1, 2, and 3, and Fig. 1, the cells were plated into dishes containing serum-flee DMEM with chemically defined supplements of fibronectin (1 I~g/mL), transferrin (10 Dg/mL), epidermal growth factor (10 ng/mL), insulin (5 #.g/mL), HEPES buffer (15 mM), and gentamicin (50 ~g/mL). For the experiment shown in Table 4, the MOD cells were plated at a higher cell density (5-7 x 104 cells/cm3) into DMEM containing 5% FBS and insulin (5 Dg/mL). This variation in experimental plan was adopted in order to insure rapid growth of the MOD cells because this cell line has recently developed a preference for serum in the growth medium.
General
Experimental Protocol
The basic experimental approach was to plate the cells on day 0. On d I (low density) or d 3 (high density), the medium was changed to selenium-containing medium. Selenium as Na2SeO3 (Pflatz and Bauer, Stanford, CT) was added to the medium at either 5 x 10-7 or 5 X 10-6M. Extensive experiments had demonstrated previously that 5 x 10-;M Se had no effect on cell growth and 5 x 10-6M Se significantly inhibited cell growth (20,21). The level of selenium present in control medium was not measured since, for the majority of the experiments, the medium did not contain serum, which is the major source of selenium in cell culture medium. In serum-free medium, the primary source of selenium would be the deionized water. The level of selenium in the water supply of the Texas Medical Center has been measured by us and is below the level of sensitivity of the assay (199% of the cells in each of the three g r o u p s were viable. The u n e x p e c t e d ratio of s e l e n i u m retained for cells e x p o s e d to 5 x 10 -6 VS 5 • 10-7M Se was e x a m i n e d with the idea that a factor or set of factors in the g r o w t h m e d i u m were altering the steady-state levels of intracellul~ir selenium. The cells w e r e g r o w n for 3 d in regular g r o w t h med i u m . O n d 3, the m e d i u m was r e m o v e d , the cells rinsed with PBS buffer, a n d t h e n the cells e x p o s e d to two different media (DMEM or PBS-glucose-growth factors) with the two concentrations of s e l e n i u m for 24 h. The results are s h o w n in Table 3. C o l u m n 3 s h o w s that little g r o w t h occurred over the 24 h. C o l u m n 4 s h o w s that the level of 75Se iI1 cells g r o w n in the salt-glucose m e d i u m followed the expected ratio (i.e., 13), whereas the level of 75Se in cells g r o w n in DMEM exhibited the u n u s u a l high ratio (i.e., 118). T h e data in c o l u m n 4 s h o w that the u n u s u a l h i g h ratio of Se retention was a result of t h e altered retention of 5 x 10-7M Se w h e n cells were g r o w n in DMEM c o m p a r e d to salt m e d i u m (216 vs 1551 CPM, respectively). These results indicated a factor or set of factors in DMEM m o d u l a t e d the r e t e n t i o n of s e l e n i u m . This m o d u l a t i o n was o v e r c o m e by high selenium concentra9 tions since the cells e x p o s e d to 5 x 10-6M Se exhibited the s a m e selen i u m retention in both media. O n e likely set of candidates responsible for m o d u l a t i n g the r e t e n t i o n of s e l e n i u m were j u d g e d to be a m i n o acids. The effects of t h e a m i n o acids, cysteine a n d m e t h i o n i n e , a n d of nucleotides on s e l e n i u m m e d i a t e d inhibition of the g r o w t h of the M O D cell line are illustrated in Table 4. M O D cells were g r o w n f r o m 3 d, then the cells were e x p o s e d to s e l e n i u m a n d a high dose of a m i n o acids or nucleotides singly or in combination. The net cell n u m b e r was m e a s u r e d 48 h later. A pilot experim e n t h a d s h o w n that m e t h i o n i n e alone (5 x 10-2M) did n o t inhibit cell Biological Trace Element Research
VoL 8, 1985
Selenium Retention in Mammary Cells
31
growth and did not counteract the effect of selenium; however, cysteine alone at 5 x 10-2M inhibited cell growth (data not shown). H o w e v e r , the cells in the presence of lower doses of cvsteine (either 2.5 or 5.0 x 10-3M) and cysteine plus selenium (5 • 10-~M Se) grew to the same levels as control cells. In contrast, the growth of cells in the presence of 5 x 10-6M Se alone was markedly depressed. The purine and pyrimidine mixes h a d no effect on cell growth nor counteracted the effect of selenium. The ability of cysteine to counteract the growth inhibitory effects of selenium was correlated with the ability of cysteine to alter the retention of selenium. The level of 75Se (5 p,Ci/mL; 5 x 10-6M cold Se) into M O D cells 30 h after the addition of selenium was 14,198 +- 1200 CPM/104 cells, whereas in the presence of cvsteine (2.5 x 10-3M) and selenium, the level was 374 +- 16 CPM/104 cells.
DISCUSSION A previous study examined the localization of low concentrations of selenium in three m a m m a r y epithelial cell lines with different degrees of responsiveness to selenium (25). In that study, the vast majority of selen i u m was intracellularly localized; however, neither the absolute level of selenium retained nor its distribution into subcellular components could explain the stimulatory effect of selenium on the growth of the YN-4 cell lines. Since, in erythrocytes, the metabolism of selenium is dosed e p e n d e n t (29,31), the experiments reported herein concentrated on the retention of high doses of selenium in a series of m a m m a r y epithelial ceil lines. The results differ from the earlier experiments on m a m m a r y cells since the cellular response to selenium was an inhibition of growth. Several interesting results were noteworthy. First, the data strongly indicate that intracellular selenium retention and inhibition of cell growth were linked events. The growth of four of five cell lines (MOD, COMMA-F, COMMA-T, YN-4) was significantly inhibited by 5 x 10-6M selenium regardless w h e t h e r selenium was administered to low or high density cultures. Since the four cell lines encompassed tumorigenic (MOD) and nontumorigenic populations (COMMA lines), it was evident that neither cell density nor the tumorigenic potential of the individual cell populations significantly influenced the retention of selenium and the response of the cells to selenium. The subcellular localization of high doses of selenium in these cells was not examined. Second, the response of the COMMA-D cell line to selenium differed from the other four cell lines. Selenium had no effect on the growth of COMMA-D cells w h e n administered to low density cultures; however, it exhibited a moderate degree of growth inhibition w h e n administered to high density cultures. This response was paradoxical since the refractoriness of low density COMMA-D cell cultures could not be attribBiological Trace Element Research
VoL 8. 1985
32
34edina, /Horrison, and Oborn
uted to a different cell density or to a different growth rate when compared to the other cell lines (Table 1). Furthermore, the qualitative nature of the growth inhibition at high density cultures seemed to be similar. The growth inhibitory effect was not a generalized toxic event since both COMMA-D and MOD (data not shown) cells were greater than 99% viable after selenium treatment. Similarly, the growth inhibitory effect of selenium on both COMMA-D cells (Table 2) and YN-4 cells (21) was rapidly reversible once selenium was withdrawn from the medium. An inspection of the data in Table 1 and Fig. 1 suggests that the lack of a growth inhibitory effect of selenium in low density cultures might be correlated with a low level of selenium in the cells. The preliminary experiment in Fig. 1 indicates that MOD cells exposed to high doses of selenium retained a high level of selenium starting at 1 h after exposure. In contrast, COMMA-D cells retained low levels of selenium over 48 h. The level of 75Se retention (CPM/104 cells) in COMMA-D cells was 40x less than the uptake in comparable cultures (Table 1: MOD, Experiment 1; YN-4, Experiment 3). However in high density cultures, the difference in retention between COMMA-D and other cell lines was only 2 x. The data suggest that COMMA-D cells metabolize and/or compartmentalize high doses of selenium differently than MOD cells. The basis for the low level of retention in COMMA-D cells is not known, but it would be interesting to determine whether this was caused by the unique cell subpopulations that comprise COMMA-D or to the presence/absence of a protein that might regulate uptake, efflux, and sequestration of intracellular selenium. Third, the experiment that examined the loss of selenium from COMI~LA-D cells indicated that the recovery from selenium-mediated cytostasis was relatively rapid. The recovery was relatively rapid with respect to the cellular events recorded, i.e., DNA synthesis and cell growth. By 24 h after withdrawal of selenium from the medium, the level of intracellular selenium was decreased 52% and the increase in thymidine labeling index was 2.5 x. These data provide one explanation for the reversible inhibition of mammary tumorigenesis observed in animal experiments (32,33). The rapid appearance of tumors after removal of selenium supplementation to the diet might be attributed to the depletion of intracellular selenium with a consequent increase in DNA synthesis and tumor growth. These data also lend themselves to several interpretations with respect to the mechanism of selenium action. The metabolism of selenium has been carefully documented in erythrocytes and hepatocytes, where it has been demonstrated that selenium is rapidly (within minutes) taken up by the cell and then released. The reduction of selenite to selenide involves glutathione and glutathione reductase in an energy dependent process (28,29,31,3"4-36). However, the role of hydrogen selenide and other metabolic products as intermediates in the synthesis of selenoproteins or their interactions with cellular macromolecules at the steady state conditions as occurred in the experiments reported herein Biological Trace Element Research
Vol. 8, 1985
Selenium Retention in Mammary Cells
33
are not well understood. One interpretation of the results reported herein would suggest that selenium modulates a series of intracellular events in a rapid fashion. For instance, there may be a set of selenoproteins or selenium-modulated proteins that are rapidly synthesized or degraded. Selenoproteins other than GSH-PX have been reported in m a m m a l i a n cells, although their functions and biochemical characteristics are unresolved at this time (37-44). If these proteins are involved in DNA synthesis, then the rapid increase of 3H-thymidine labeling w o u l d suggest these proteins have a rapid turnover rate. A careful d o c u m e n t a t i o n of selenium uptake, efflux, and localization combined with a sensitive analysis of protein changes occurring during these events might unravel a critical function for selenium. A second interpretation w o u l d suggest that selenium modulates glutathione metabolism and this modulation, in turn, could affect DNA synthesis. LeBoeuf and Hoekstra (45) have reported that oxidized glutathione as well as nonprotein sulfhydrals (i.e., GSH) in liver are increased after continuous exposure ~o high levels of selenium. Other investigators have d e m o n s t r a t e d similarly a role of cellular sulfhydrals in selenium metabolism (28,29,31,36). Although a modification of gluthathione metabolism could affect D N A synthesis (46); as yet, the selenium modulation of gluththione metabolism and subsequent inhibition of DNA synthesis have not been linked together causally in one system. Fourth,-the experiments shed light on factors that regulate retention of inorganic selenium. It is k n o w n that selenium diffuses across the cell membrane; however, the disproportionate ratio of selenium retention d e m o n s t r a t e d in these experiments suggested some factors influenced the overall process of uptake and efflux. The experiments p r e s e n t e d herein did not examine uptake separately from efflux. It has been s h o w n that these are two separate events in erythrocytes and are regulated inde-. p e n d e n t l y (28,29,31,36). Whether the same process occurs in epithelial cells is not clear. However, the results presented in Tables 3 and 4 indicate some factors that may influence retention. The results in Table 3 d e m o n s t r a t e d that factors in the growth m e d i u m strongly influenced the retention of selenium w h e n selenium was in relatively low concentrations in the m e d i u m . A similar result was s h o w n by Gasiewicz and Smith (29) w h o noted that selenium efflux was d e p e n d e n t upon the nature of the extracellular m e d i u m , i.e., plasma versus a salt buffer. The data provided by Experiment 4 demonstrated that the amino acid cysteine, but not m e t h i o n i n e or nucleotides, altered the retention of selenium and thus counteracted the growth inhibitory effect of selenium even w h e n selenium was at high concentrations. Thus, factors that regulate overall intracellular selenium retention can have significant effects on the biological response of the cell. In s u m m a r y , these experiments reinforce the importance of intracellular accumulation of selenium for its growth inhibitory effect; however, they do not define the nature of the intracellular localization. The current emphasis of on-going experiments is to elucidate the macromolecules Biological Trace Element Research
Vol. 8, 1985
Medina, ~1orrison, and Oborn
34
that interact w i t h or are m o d u l a t e d by s e l e n i u m a n d h o w t h e s e interactions affect D N A synthesis.
ACKNOWLEDGMENT S u p p o r t e d by N I H research grants CA-11944 (DM) a n d GM-07330 (DGM).
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32.
H. J. Thompson and P. J. Becci, ]. ,Natl. Cancer Inst. 65, 1299 (1980). C. Ip, Cancer Res. 41, 2683 (1981). C. Ip, Cancer Res. 41, 4386 (1981). C. W. Welsch, M. Gt~odrich-Smith, C. K. Brown, H. D. Greene, and E. I. Hamel, Carcinogenesis 2, 519 (1981). G. N. Schrauzer and D. Ishmael, Ann. Clin. Lab. Sci. 4, 441 (1974). G. N. Schrauzer, D. A. White, and C. J. Schneider, Bioinorg. Chem. 6, 265 (1976). D. Medina and F. Shepherd, Cancer Lett. 8, 241 (1980). D. Medina and F. Shepherd, Carcinogenesis 2, 451 (1981). D. Medina, H. W. Lane, and F. Shepherd, Carcinogenesis 4, 1159 (1983). C. Ip and D. Sinha, Carcinogenesis 2, 435 (1981). H. W. Lane and D. Medina, Cancer Res. 43, 1558 (1983). H. W. Lane, C. K. Tracey, and D. Medina, [. Nutr. 114, 323 (1984). P. M. Horvath and C. [p, Cancer Res. 43, 5335 (1983). H. W. Lane and D. Medina, J. Natl. Cancer Inst. (in press). D. Medina, in Diet, Nutrition, and Cancer, B. Reddy and L. Cohen, eds., CRC Press, Boca Raton, F1. (in press). W. L. McKeehan, W. G. Hamilton, and R. G. Ham, Proc. Natl, Acad. Sci. USA 73, 2023 (1976). G. A. Greeder and J. A. Milner, Science 209, 825 (1980). J. A. Milner and C. Y. Hsu, Cancer Res. 41, 1652 (1981). A. M. Watrach, J. A. Milner, and M. A. Watrach, Cancer Lett. 15, 137 (1982). D. Medina and C. J. Oborn, Cancer Lett. 13, 333 (1981). D. Medina and C. J. Oborn, Cancer Res. 44, 4361 (1984). D. W. Gruenwedel, and M. K. Cruickshank, Toxicol. Appl. Pharmacol. 50, 1 (1979). M. Kasuya, Toxicol. Appl. Pharmacol. 35, 11 (1976). D. Medina, H. W. Lane, and C. M. Tracey, Cancer Res. 43, 2460 (1983). D. Medina, H. W. Lane, and C. J. Oborn, Cancer Lett. 15, 301 (1982). D. Medina, Fo Miller, C. J. Oborn, and B. B. Asch, Cancer Res. 43, 2100 (1983). K. G. Danietson, C. J. Oborn, E. M. Durban, J. S. Butel, and D. Medina, Proc. Natl. Acad. Sci, USA 81, 3756 (1984). A. L. Lee, A. Dong, and J. Yano, Can ]. Biochem. 47, 791 (1969). T. A. Gasiewicz and J. C. Smith, Chem. Biol. Interactions. 21, 299 (1978). D. Medina and C. J. Oborn, Cancer Res. 40, 3982 (1980). M. Sandholm, Acta Pharmacol. Toxicol. 33, 6 (1973). G. N. Schrauzer, J. E. McGinness, and K. Kuehn, Carcinogenesis 1, 199 (1980).
Biological Trace Element Research
Vo/. 8, ! 985
Selenium Retention in A4ammary Cells
35
33. C. Ip and M. M. lp, Carcinogenesis 2, 915 (1981). 34. P. L. Wright and M. D. Bell, Proc. Soc. Exptl. Biol. Med. 114, 379 (1963). 35. R. A. Sunde and W. G. Hoekstra, Biochem. Biophys. Res. Commun. 93, 1181, (1980). 36. H. E. Ganther, in Advances in Nutritional Research, H. H. Draper, ed., Plenum, NY, 2, 107 (1979). 37. H. I. Calvin, J. Exp. Zool. 204, 445 (1978). 38. R. S. Black, M. J. Tripp, P. D. Whanger, and P. H. Weswig, Bioinorganic Chem. 8, 161, (1978). 39. V. Pallini and E. Bacci, ]. Submicr. Cytol. 11, 165 (1979). 40. K. P. McConneU, R. M. Burton, T. Kute, and P. J. Higgins, Biodfim. Biophys. Acta 558, 113 (1979). 41. R. F. Burk and P. E. GregoD', Arch. Biochem. Biophys. 213, 73 (1982). 42. M. A. Motsenbocker and A. L. Tappel Biochim. Biophys. Acta 709, 160 (1982). 43. M. A. Motsenbocker and A. L. Tappel Biochim. Biophys. Acta 719, 147 (1982). 44. K. D. Danielson, C. J. Oborn, and D. Medina, J. Cell Biol. 97, 2 (1983). 45. R. A. LeBoeuf and W. G. Hoekstra, J. Nutr. 113, 845 (1983). 46. A. Meister, Science 220, 472 (1983).
Biological Trace Element Research
Vol. 8. 1985
Selenium Retention and Inhibition of Cell Growth in Mouse Mammary Epithelial Cell Lines In Vitro DANIEL MEDINA,* DAVID MORRISON, AND CAROLJ. OBORN Department of Cell Biology, Baylor College of ~ledicine, Houston, TX 77030 Received July 24, 1984; Accepted February 25, 1985
ABSTRACT The steady state levels of growth inhibitory doses of inorganic selenium were examined in five different mammary epithelial cell lines: MOD, COMMA-D, COMMA-F, COMMA-T, and YN-4. The retention of selenium was monitored using a radioactive isotope, ;SSe. Growth inhibition correlated with high levels of selenium in the cell. Generally, the retention of intracellular selenium was not dependent upon cell density,, cell number, net growth rate, or tumorigenicity of the mammary cell lines. One cell line, COMMA-D, exhibited an unique response wherein the amount of selenium retained was low and the growth inhibitory effects of selenium were negligible when the cells were exposed to selenium at low density. However, at high cell densities, the COMMA-D cells responded like the other four cell lines. The growth inhibitory effect of selenium was reversible; upon removal of selenium from the medium, cells start synthesizing DNA within 24 h. The retention of selenium was influenced by constituents in the growth medium. In particular, cysteine, but not methionine, purines, or pyrimidines altered selenium retention and counteracted the growth inhibitory effects of selenium. These results indicated that the mammary cell lines, particula~ COMMA-D and MOD are good model systems to examine the uptake, retention, localization, and function of inorganic selenium under conditions where it acts as a growth inhibitory agent. * Author to whom all correspondence and reprint requests should be addressed. Biological Trace Element Research
19
Vol. 8, 1985
?4edina, Morrison, and Oborn
20
Index Entries: Selenium retention, and growth inhibition in mammary cells; reversibility, of Se-inhibited growfl~ in mammary cells; cell culture, Se retention and growth inhibition in mammaD'; mouse mammary cells, Se retention and growth of; mammary cells, Se retention and growth in mouse; epithelial ceils, Se retention and growth in mouse mammary.
INTRODUCTION The inhibition of m a m m a D' tumorigenesis by inorganic selenium a d d e d to the diet or water is a well-documented fact in both the m o u s e and rat m a m m a r y tumor models (1-9). The m e c h a n i s m s of the chemopreventive effects of selenium supplementation have not been resolved; however, it appears that regulation of glutathione peroxidase and inhibition of lipid peroxidation may not be involved (10-15). In order to examine at a cellular and biochemical level the possible m e c h a n i s m s of action of selenium an in vitro model system has been developed over the past several years. Previous results have demonstrated that m a m m a r y epithelial cell lines as well as cell lines of other organs, are sensitive to the growth inhibitory effects of selenium (16-23). The results so far have indicated that low doses (5 x 10-SM Se) stimulate cell growth, whereas high doses (5 x 10-6M Se) inhibit cell growth, but in a reversible manner. The inhibition of m a m m a r y cell growth correlated with a decrease in the uptake of 3H-thymidine into DNA, a decreased DNA-labeling index, and a blockage of cells in the S / G 2 phases of the cell cycle (20,21,24). In vitro cell lines afford a simple and quantifiable model system to examine several aspects of selenium function. Previous experiments examined the retention and localization of low doses of selenium using 75Se as a tracer for selenium (25). Those results indicated that there was a progressive increase in the a m o u n t of selenium retained by the m a m m a r y cells over 24 h and the majority of selenium localized in the cytoplasm. In the experiments reported herein, the intraceUular retention of selenium was examined in five different m a m m a r y epithelial lines u n d e r conditions in which the growth of the cells was inhibited by selenium. The reversibility of selenium inhibition of growth as well as some factors that influenced selenium retention were also examined in specific cell lines.
MATERIAL AND METHODS Ceil Lines The origin and cellular characteristics of mouse m a m m a r y epithelial cell lines YN-4 and COMIVIA-D have been detailed in (20,21,26,27). The COMMA-T and COMMA-F lines were derived from early passage COMMA-D by partial trypsinization (COMMA-T) and by plating spontaneous floaters (COMMA-F) (unpublished data). The latter two cell lines Biological Trace Element Research
VoL 8. l g85
Selenium Retention in Mammary Cells
21
exhibit properties of epithelial cells, but do not produce normal mamma W gland upon transplantation into syngeneic mice. The MOD cell line was derived from a D2 mammary tumor and produces mammary adenocarcinomas upon transplantation into syngeneic mice.
Cell Culture The cell lines were routinely grown and maintained in Falcon plastic flasks in Dulbecco's Modified Eagles Medium (DMEM) with 10% fetal bovine serum (FBS) (Hy-clone Inc., Logan, Utah), insulin (5 >g/mL), gentamicin (50 I,g/mL), and N-2-hydroxyethylpiperazine-N'-2-ethane sulfonic acid buffer (HEPES). Cultures were maintained at 37~ in air: CO2 (92.5:7.5) and subcultured weekly. All cell lines were epithelial, as judged by the presence of keratin intermediate filaments, the absence of vimentin intermediate filaments, and the ability to form "domes" at high saturation densities. For experiments, cells were collected from the parent flasks by trypsinization, washed twice in DMEM with serum and insulin, and plated at 3 x 10 4 cells/cm2 into 35 mm Falcon Petri dishes. For the experiments shown in Tables 1, 2, and 3, and Fig. 1, the cells were plated into dishes containing serum-flee DMEM with chemically defined supplements of fibronectin (1 I~g/mL), transferrin (10 Dg/mL), epidermal growth factor (10 ng/mL), insulin (5 #.g/mL), HEPES buffer (15 mM), and gentamicin (50 ~g/mL). For the experiment shown in Table 4, the MOD cells were plated at a higher cell density (5-7 x 104 cells/cm3) into DMEM containing 5% FBS and insulin (5 Dg/mL). This variation in experimental plan was adopted in order to insure rapid growth of the MOD cells because this cell line has recently developed a preference for serum in the growth medium.
General
Experimental Protocol
The basic experimental approach was to plate the cells on day 0. On d I (low density) or d 3 (high density), the medium was changed to selenium-containing medium. Selenium as Na2SeO3 (Pflatz and Bauer, Stanford, CT) was added to the medium at either 5 x 10-7 or 5 X 10-6M. Extensive experiments had demonstrated previously that 5 x 10-;M Se had no effect on cell growth and 5 x 10-6M Se significantly inhibited cell growth (20,21). The level of selenium present in control medium was not measured since, for the majority of the experiments, the medium did not contain serum, which is the major source of selenium in cell culture medium. In serum-free medium, the primary source of selenium would be the deionized water. The level of selenium in the water supply of the Texas Medical Center has been measured by us and is below the level of sensitivity of the assay (199% of the cells in each of the three g r o u p s were viable. The u n e x p e c t e d ratio of s e l e n i u m retained for cells e x p o s e d to 5 x 10 -6 VS 5 • 10-7M Se was e x a m i n e d with the idea that a factor or set of factors in the g r o w t h m e d i u m were altering the steady-state levels of intracellul~ir selenium. The cells w e r e g r o w n for 3 d in regular g r o w t h med i u m . O n d 3, the m e d i u m was r e m o v e d , the cells rinsed with PBS buffer, a n d t h e n the cells e x p o s e d to two different media (DMEM or PBS-glucose-growth factors) with the two concentrations of s e l e n i u m for 24 h. The results are s h o w n in Table 3. C o l u m n 3 s h o w s that little g r o w t h occurred over the 24 h. C o l u m n 4 s h o w s that the level of 75Se iI1 cells g r o w n in the salt-glucose m e d i u m followed the expected ratio (i.e., 13), whereas the level of 75Se in cells g r o w n in DMEM exhibited the u n u s u a l high ratio (i.e., 118). T h e data in c o l u m n 4 s h o w that the u n u s u a l h i g h ratio of Se retention was a result of t h e altered retention of 5 x 10-7M Se w h e n cells were g r o w n in DMEM c o m p a r e d to salt m e d i u m (216 vs 1551 CPM, respectively). These results indicated a factor or set of factors in DMEM m o d u l a t e d the r e t e n t i o n of s e l e n i u m . This m o d u l a t i o n was o v e r c o m e by high selenium concentra9 tions since the cells e x p o s e d to 5 x 10-6M Se exhibited the s a m e selen i u m retention in both media. O n e likely set of candidates responsible for m o d u l a t i n g the r e t e n t i o n of s e l e n i u m were j u d g e d to be a m i n o acids. The effects of t h e a m i n o acids, cysteine a n d m e t h i o n i n e , a n d of nucleotides on s e l e n i u m m e d i a t e d inhibition of the g r o w t h of the M O D cell line are illustrated in Table 4. M O D cells were g r o w n f r o m 3 d, then the cells were e x p o s e d to s e l e n i u m a n d a high dose of a m i n o acids or nucleotides singly or in combination. The net cell n u m b e r was m e a s u r e d 48 h later. A pilot experim e n t h a d s h o w n that m e t h i o n i n e alone (5 x 10-2M) did n o t inhibit cell Biological Trace Element Research
VoL 8, 1985
Selenium Retention in Mammary Cells
31
growth and did not counteract the effect of selenium; however, cysteine alone at 5 x 10-2M inhibited cell growth (data not shown). H o w e v e r , the cells in the presence of lower doses of cvsteine (either 2.5 or 5.0 x 10-3M) and cysteine plus selenium (5 • 10-~M Se) grew to the same levels as control cells. In contrast, the growth of cells in the presence of 5 x 10-6M Se alone was markedly depressed. The purine and pyrimidine mixes h a d no effect on cell growth nor counteracted the effect of selenium. The ability of cysteine to counteract the growth inhibitory effects of selenium was correlated with the ability of cysteine to alter the retention of selenium. The level of 75Se (5 p,Ci/mL; 5 x 10-6M cold Se) into M O D cells 30 h after the addition of selenium was 14,198 +- 1200 CPM/104 cells, whereas in the presence of cvsteine (2.5 x 10-3M) and selenium, the level was 374 +- 16 CPM/104 cells.
DISCUSSION A previous study examined the localization of low concentrations of selenium in three m a m m a r y epithelial cell lines with different degrees of responsiveness to selenium (25). In that study, the vast majority of selen i u m was intracellularly localized; however, neither the absolute level of selenium retained nor its distribution into subcellular components could explain the stimulatory effect of selenium on the growth of the YN-4 cell lines. Since, in erythrocytes, the metabolism of selenium is dosed e p e n d e n t (29,31), the experiments reported herein concentrated on the retention of high doses of selenium in a series of m a m m a r y epithelial ceil lines. The results differ from the earlier experiments on m a m m a r y cells since the cellular response to selenium was an inhibition of growth. Several interesting results were noteworthy. First, the data strongly indicate that intracellular selenium retention and inhibition of cell growth were linked events. The growth of four of five cell lines (MOD, COMMA-F, COMMA-T, YN-4) was significantly inhibited by 5 x 10-6M selenium regardless w h e t h e r selenium was administered to low or high density cultures. Since the four cell lines encompassed tumorigenic (MOD) and nontumorigenic populations (COMMA lines), it was evident that neither cell density nor the tumorigenic potential of the individual cell populations significantly influenced the retention of selenium and the response of the cells to selenium. The subcellular localization of high doses of selenium in these cells was not examined. Second, the response of the COMMA-D cell line to selenium differed from the other four cell lines. Selenium had no effect on the growth of COMMA-D cells w h e n administered to low density cultures; however, it exhibited a moderate degree of growth inhibition w h e n administered to high density cultures. This response was paradoxical since the refractoriness of low density COMMA-D cell cultures could not be attribBiological Trace Element Research
VoL 8. 1985
32
34edina, /Horrison, and Oborn
uted to a different cell density or to a different growth rate when compared to the other cell lines (Table 1). Furthermore, the qualitative nature of the growth inhibition at high density cultures seemed to be similar. The growth inhibitory effect was not a generalized toxic event since both COMMA-D and MOD (data not shown) cells were greater than 99% viable after selenium treatment. Similarly, the growth inhibitory effect of selenium on both COMMA-D cells (Table 2) and YN-4 cells (21) was rapidly reversible once selenium was withdrawn from the medium. An inspection of the data in Table 1 and Fig. 1 suggests that the lack of a growth inhibitory effect of selenium in low density cultures might be correlated with a low level of selenium in the cells. The preliminary experiment in Fig. 1 indicates that MOD cells exposed to high doses of selenium retained a high level of selenium starting at 1 h after exposure. In contrast, COMMA-D cells retained low levels of selenium over 48 h. The level of 75Se retention (CPM/104 cells) in COMMA-D cells was 40x less than the uptake in comparable cultures (Table 1: MOD, Experiment 1; YN-4, Experiment 3). However in high density cultures, the difference in retention between COMMA-D and other cell lines was only 2 x. The data suggest that COMMA-D cells metabolize and/or compartmentalize high doses of selenium differently than MOD cells. The basis for the low level of retention in COMMA-D cells is not known, but it would be interesting to determine whether this was caused by the unique cell subpopulations that comprise COMMA-D or to the presence/absence of a protein that might regulate uptake, efflux, and sequestration of intracellular selenium. Third, the experiment that examined the loss of selenium from COMI~LA-D cells indicated that the recovery from selenium-mediated cytostasis was relatively rapid. The recovery was relatively rapid with respect to the cellular events recorded, i.e., DNA synthesis and cell growth. By 24 h after withdrawal of selenium from the medium, the level of intracellular selenium was decreased 52% and the increase in thymidine labeling index was 2.5 x. These data provide one explanation for the reversible inhibition of mammary tumorigenesis observed in animal experiments (32,33). The rapid appearance of tumors after removal of selenium supplementation to the diet might be attributed to the depletion of intracellular selenium with a consequent increase in DNA synthesis and tumor growth. These data also lend themselves to several interpretations with respect to the mechanism of selenium action. The metabolism of selenium has been carefully documented in erythrocytes and hepatocytes, where it has been demonstrated that selenium is rapidly (within minutes) taken up by the cell and then released. The reduction of selenite to selenide involves glutathione and glutathione reductase in an energy dependent process (28,29,31,3"4-36). However, the role of hydrogen selenide and other metabolic products as intermediates in the synthesis of selenoproteins or their interactions with cellular macromolecules at the steady state conditions as occurred in the experiments reported herein Biological Trace Element Research
Vol. 8, 1985
Selenium Retention in Mammary Cells
33
are not well understood. One interpretation of the results reported herein would suggest that selenium modulates a series of intracellular events in a rapid fashion. For instance, there may be a set of selenoproteins or selenium-modulated proteins that are rapidly synthesized or degraded. Selenoproteins other than GSH-PX have been reported in m a m m a l i a n cells, although their functions and biochemical characteristics are unresolved at this time (37-44). If these proteins are involved in DNA synthesis, then the rapid increase of 3H-thymidine labeling w o u l d suggest these proteins have a rapid turnover rate. A careful d o c u m e n t a t i o n of selenium uptake, efflux, and localization combined with a sensitive analysis of protein changes occurring during these events might unravel a critical function for selenium. A second interpretation w o u l d suggest that selenium modulates glutathione metabolism and this modulation, in turn, could affect DNA synthesis. LeBoeuf and Hoekstra (45) have reported that oxidized glutathione as well as nonprotein sulfhydrals (i.e., GSH) in liver are increased after continuous exposure ~o high levels of selenium. Other investigators have d e m o n s t r a t e d similarly a role of cellular sulfhydrals in selenium metabolism (28,29,31,36). Although a modification of gluthathione metabolism could affect D N A synthesis (46); as yet, the selenium modulation of gluththione metabolism and subsequent inhibition of DNA synthesis have not been linked together causally in one system. Fourth,-the experiments shed light on factors that regulate retention of inorganic selenium. It is k n o w n that selenium diffuses across the cell membrane; however, the disproportionate ratio of selenium retention d e m o n s t r a t e d in these experiments suggested some factors influenced the overall process of uptake and efflux. The experiments p r e s e n t e d herein did not examine uptake separately from efflux. It has been s h o w n that these are two separate events in erythrocytes and are regulated inde-. p e n d e n t l y (28,29,31,36). Whether the same process occurs in epithelial cells is not clear. However, the results presented in Tables 3 and 4 indicate some factors that may influence retention. The results in Table 3 d e m o n s t r a t e d that factors in the growth m e d i u m strongly influenced the retention of selenium w h e n selenium was in relatively low concentrations in the m e d i u m . A similar result was s h o w n by Gasiewicz and Smith (29) w h o noted that selenium efflux was d e p e n d e n t upon the nature of the extracellular m e d i u m , i.e., plasma versus a salt buffer. The data provided by Experiment 4 demonstrated that the amino acid cysteine, but not m e t h i o n i n e or nucleotides, altered the retention of selenium and thus counteracted the growth inhibitory effect of selenium even w h e n selenium was at high concentrations. Thus, factors that regulate overall intracellular selenium retention can have significant effects on the biological response of the cell. In s u m m a r y , these experiments reinforce the importance of intracellular accumulation of selenium for its growth inhibitory effect; however, they do not define the nature of the intracellular localization. The current emphasis of on-going experiments is to elucidate the macromolecules Biological Trace Element Research
Vol. 8, 1985
Medina, ~1orrison, and Oborn
34
that interact w i t h or are m o d u l a t e d by s e l e n i u m a n d h o w t h e s e interactions affect D N A synthesis.
ACKNOWLEDGMENT S u p p o r t e d by N I H research grants CA-11944 (DM) a n d GM-07330 (DGM).
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32.
H. J. Thompson and P. J. Becci, ]. ,Natl. Cancer Inst. 65, 1299 (1980). C. Ip, Cancer Res. 41, 2683 (1981). C. Ip, Cancer Res. 41, 4386 (1981). C. W. Welsch, M. Gt~odrich-Smith, C. K. Brown, H. D. Greene, and E. I. Hamel, Carcinogenesis 2, 519 (1981). G. N. Schrauzer and D. Ishmael, Ann. Clin. Lab. Sci. 4, 441 (1974). G. N. Schrauzer, D. A. White, and C. J. Schneider, Bioinorg. Chem. 6, 265 (1976). D. Medina and F. Shepherd, Cancer Lett. 8, 241 (1980). D. Medina and F. Shepherd, Carcinogenesis 2, 451 (1981). D. Medina, H. W. Lane, and F. Shepherd, Carcinogenesis 4, 1159 (1983). C. Ip and D. Sinha, Carcinogenesis 2, 435 (1981). H. W. Lane and D. Medina, Cancer Res. 43, 1558 (1983). H. W. Lane, C. K. Tracey, and D. Medina, [. Nutr. 114, 323 (1984). P. M. Horvath and C. [p, Cancer Res. 43, 5335 (1983). H. W. Lane and D. Medina, J. Natl. Cancer Inst. (in press). D. Medina, in Diet, Nutrition, and Cancer, B. Reddy and L. Cohen, eds., CRC Press, Boca Raton, F1. (in press). W. L. McKeehan, W. G. Hamilton, and R. G. Ham, Proc. Natl, Acad. Sci. USA 73, 2023 (1976). G. A. Greeder and J. A. Milner, Science 209, 825 (1980). J. A. Milner and C. Y. Hsu, Cancer Res. 41, 1652 (1981). A. M. Watrach, J. A. Milner, and M. A. Watrach, Cancer Lett. 15, 137 (1982). D. Medina and C. J. Oborn, Cancer Lett. 13, 333 (1981). D. Medina and C. J. Oborn, Cancer Res. 44, 4361 (1984). D. W. Gruenwedel, and M. K. Cruickshank, Toxicol. Appl. Pharmacol. 50, 1 (1979). M. Kasuya, Toxicol. Appl. Pharmacol. 35, 11 (1976). D. Medina, H. W. Lane, and C. M. Tracey, Cancer Res. 43, 2460 (1983). D. Medina, H. W. Lane, and C. J. Oborn, Cancer Lett. 15, 301 (1982). D. Medina, Fo Miller, C. J. Oborn, and B. B. Asch, Cancer Res. 43, 2100 (1983). K. G. Danietson, C. J. Oborn, E. M. Durban, J. S. Butel, and D. Medina, Proc. Natl. Acad. Sci, USA 81, 3756 (1984). A. L. Lee, A. Dong, and J. Yano, Can ]. Biochem. 47, 791 (1969). T. A. Gasiewicz and J. C. Smith, Chem. Biol. Interactions. 21, 299 (1978). D. Medina and C. J. Oborn, Cancer Res. 40, 3982 (1980). M. Sandholm, Acta Pharmacol. Toxicol. 33, 6 (1973). G. N. Schrauzer, J. E. McGinness, and K. Kuehn, Carcinogenesis 1, 199 (1980).
Biological Trace Element Research
Vo/. 8, ! 985
Selenium Retention in A4ammary Cells
35
33. C. Ip and M. M. lp, Carcinogenesis 2, 915 (1981). 34. P. L. Wright and M. D. Bell, Proc. Soc. Exptl. Biol. Med. 114, 379 (1963). 35. R. A. Sunde and W. G. Hoekstra, Biochem. Biophys. Res. Commun. 93, 1181, (1980). 36. H. E. Ganther, in Advances in Nutritional Research, H. H. Draper, ed., Plenum, NY, 2, 107 (1979). 37. H. I. Calvin, J. Exp. Zool. 204, 445 (1978). 38. R. S. Black, M. J. Tripp, P. D. Whanger, and P. H. Weswig, Bioinorganic Chem. 8, 161, (1978). 39. V. Pallini and E. Bacci, ]. Submicr. Cytol. 11, 165 (1979). 40. K. P. McConneU, R. M. Burton, T. Kute, and P. J. Higgins, Biodfim. Biophys. Acta 558, 113 (1979). 41. R. F. Burk and P. E. GregoD', Arch. Biochem. Biophys. 213, 73 (1982). 42. M. A. Motsenbocker and A. L. Tappel Biochim. Biophys. Acta 709, 160 (1982). 43. M. A. Motsenbocker and A. L. Tappel Biochim. Biophys. Acta 719, 147 (1982). 44. K. D. Danielson, C. J. Oborn, and D. Medina, J. Cell Biol. 97, 2 (1983). 45. R. A. LeBoeuf and W. G. Hoekstra, J. Nutr. 113, 845 (1983). 46. A. Meister, Science 220, 472 (1983).
Biological Trace Element Research
Vol. 8. 1985
