Cancer Letters, 65 (1992) 43-49

Elsevier Scientific Publishers Ireland Ltd.

43

Inhibition by selenium of DNA and RNA synthesis in normal and malignant human cells in vitro

Fikrat I. Abdullaev, Christina MacVicar and Gerald D. Frenkel

Department of Biological Sciences, Rutgers Uniuersity, Newark, NJ 07102 (USA)

(Received 23 March 1992) (Revision received 29 April 1992) (Accepted 30 April 1992)

Smnmary

Seoeral studies haoe demonstrated dij- jerences between normal and malignant cells in their sensitiuity to uarious effects of selenite. We have compared the effect of selenite on

DNA and RNA synthesis in two pairs of normal and malignant human cell lines. One pair of cells, CCL-210 (normal lung jibroblasts) and A549 (lung adenocarcinoma cells), exhibited a large difference in their sensitiuity to selenite but no significant difference in their sensitivity to selenodiglutathione. They also had a large difference in the leuel of intracellular suljhydryl (SH) compounds. In contrast the other pair of cells, WI-38 (normal fetal lung jibroblasts) and WL38VA (SV-40 transformed WI-38 cells) both had low leuels of intracellular SH com-

pounds and exhibited similar (low) sensitiuity to selenite. Our results indicate that differences between normal and malignant cells in their sensitiuity to selenite could be due to a dij- jerence in the reaction of selenite with in- tracellular suljhydryl compounds to form selenotrisuljides.

Correspondence to: Gerald D. Frenkel, Department of

Biological Sciences, Rutgers University, Newark, NJ 07102,

USA.

Keywords: DNA synthesis; RNA synthesis; selenium; sulfhydryl compounds

Introduction

Selenium compounds such as selenite have been known for some time to inhibit the development of tumors in animals. Never- theless it has proven extremely difficult to establish the precise mechanism of this anticar- cinogenic effect [4,22,23]. One mechanism which has been proposed is a cytotoxic effect on tumor cells. Support for this idea has come from the demonstration that selenite induces a variety of cytotoxic effects in tumor cells in vivo and in vitro [1,3,8,15,16,18,21,24,28].

The observation that selenium compounds are anticarcinogenic in animals has suggested their potential utilization as chemotherapeutic and/or chemopreventive agents in human be- ings [2,17,29]. An important factor in realizing this potential is the selectivity of these com- pounds for tumor cells. Several previous studies have supported the generalization that normal and malignant cells differ in their sensi- tivity to the effects of selenium compounds on cellular functions [1,6,7,18,19,21,27]. Although the possible reasons for this differen- tial sensitivity have been discussed, its molecular basis is not yet understood. In order

0304-3835/92/$05.00 0 1992 Elsevier Scientific Publishers Ireland Ltd

Printed and Published in lreland

44

to investigate the relative sensitivity of normal and malignant cells to selenite we have been studying its inhibitory effects on cellular DNA and RNA synthesis [ 1,8,10,1 l] which are im- portant measures of cytotoxicity [22]. In this report we describe differences between cells which could account for the greater sensitivity of the tumor cells to selenite.

Experimental Procedures

Materials

Cell culture media and serum were obtained from GIBCO. [3H]Thymidine and [3H]uridine were purchased from NEN, diethylmaleate (DEM) from Sigma. Non-radioactive sodium selenite was obtained from Gallard Schlesinger and [ 75Se]selenite from Amersham. Selenodi- glutathione was formed by the reaction of selenite (333 PM) with glutathione at a 1:4- molar ratio in culture medium (without serum) at room temperature for 30 min. Under these conditions essentially all of the selenite is con- verted to selenodiglutathione, as determined by spectroscopic and thin-layer chromato- graphic analysis [9,12].

Cells The cell lines utilized were: CCL-210 (nor-

mal lung fibroblast cells), A549 (lung adenocarcinoma cells), WI-38 (normal fetal lung fibroblast cells) and WI-38VA (SV40- transformed WI-38 cells). All of the cell lines are of human origin and were obtained from the American Type Culture Collection. All cells were grown in a 1: 1 mixture of Dulbecco’s Modified Eagle’s Medium and Ham’s F12 with 10% fetal calf serum at 37OC in a 5% CO2 at- mosphere. Under these conditions the growth rates (doubling times) were 36, 22, 38 and 20 h for the CCL-210, A549, WI-38 and WI- 38VA cells, respectively.

DNA and RNA synthesis Cells were grown in monolayer cultures in

lOO-mm dishes. Sodium selenite was added to the cultures as indicated in the individual

experiments and incubation was continued at 37OC for 1 h. r3H]Thymidine (86 Ci/mmol) or [3H]uridine (42 Ci/mmol) was then added to a final concentration of 4 &i/ml and in- cubation continued for 15 min at 37OC. The incorporation of radioactivity into the acid precipitable form was then measured as described previously [8]. All determinations were carried out in duplicate.

Quantitation of cellular non-protein sulfhydryl

(SH) compounds The level of cellular non-protein SH com-

pounds was determined with exponentially growing cultures with Ellman’s reagent [5], utilizing the procedure of Sedlak and Lindsay [26] as described previously [lo].

Measurement of selenium uptake Cells were exposed to 25 FM [75Se]se1enite

(0.5 Ci/mmol) for -1 h and cell-associated selenium was determined as described previously [3].

Thin layer chromatography The selenium-containing species were

analyzed by TLC (elution solvent:isobutyric acid:H20/NH40H = 66:33:1) and auto- radiography, as described previously [ 121.

Results

As part of a comparative study of the relative sensitivities of cells to selenium compounds we examined the effect of selenite on DNA and RNA synthesis in two human cell lines: normal lung fibroblasts (CCL-210) and lung tumor cells (A549). We observed that the potency of selenite was higher with the tumor cells than the normal cells (Fig. 1). Thus, with both DNA and RNA synthesis, 50% inhibition occurred with 50-60 PM selenite for the normal cells, but with less than 10 PM selenite for the tumor cells.

A number of previous studies have demonstrated the involvement of cellular sulfhydryl compounds in the cellular effects of

0 20 40 60 60 100 Table 1. Intracellular SH levels.

B

20 -

00 0 20 40 60 a0 100

[SelenlteJ(~M)

Fb. 1. Inhibition of DNA and RNA synthesis by selenite. Cells were exposed to the indicated concentra- tion of selenite for 1 h, after which the incorporation of [H3]thymidine or [H3]uridine into DNA or RNA, respec- tively, was measured as described in Experimental Pro- cedures. The results are presented as a percentage of the incorporation in the absence of added selenite. A: DNA synthesis; B: RNA synthesis. (0): 210 cells; (0): A549 cells.

selenite [ lO,ll, 19,20,25,28]. This stems from the fact that the first step in the metabolism of selenite [14] is its reaction with glutathione or other cellular SH compounds-to form corresponding selenotrisulfide [ 131:

HsSeOs + 4 RSH - RSSeSR + RSSR

+ 3 H20

the

45

Cells SH compounds nmoles/106 cells

A549 32.9 zt 2.6 210 5.3 l 0.3 WI-38 2.9 l 0.4 WL38VA 2.8 zt 0.5

One possible explanation for the differing sensitivity of these two cell types to selenite is a difference in their intracellular level of SH compounds. To examine this we measured these levels and found that the level of SH compounds was in fact approximately 6-fold higher in the tumor cells (Table I). If the higher potency of selenite in the tumor cells does in fact result from the higher intracellular level of SH compounds, then the potency of the selenotrisulfide product of the above reaction would be expected to be the same in the two cell types. To examine this, selenite was reacted with glutathione, the predominant cellular SH compound, to form selenodigluta- thione (see Experimental Procedures) and the effect of this compound on nucleic acid syn- thesis in the two cell types was examined. As shown in Fig. 2, selenodiglutathione inhibited DNA and RNA synthesis in both cell types with approximately the same potency. This in- dicates that the differing levels of SH com- pounds of the two cell types could explain their differing sensitivities to selenite.

Further support for this idea comes from studies on another pair of normal and malig- nant cells, which do not exhibit a difference in their sensitivities to selenite. As shown in Fig. 3, both WI-38 cells and their SV40- transformed derivative WI-38VA cells were relatively insensitive to the inhibitory effects of selenite on nucleic acid synthesis. As shown in Table I, both cell types have approximately the same relatively low intracellular SH level. Fur- thermore, both cell types were sensitive to selenodiglutathione (Fig. 3)) indicating that

46

80

60

60

0

0 20 40 60 80 100

1. ’ * '.'*'. ’

1 3 I ’ I ’ I ’ I

0 20 40 60 80 100

[Selenodiglutathione] (PM)

Fig. 2. Inhibition of DNA and RNA synthesis by

selenodiglutathione. The experiment was carried out as in Fig. 1, except that prior to addition to cells the selenite was reacted with glutathione to form selenodiglutathione, as described in Experimental Procedures. The results are

presented as a percentage of the incorporation in the absence of added selenodiglutathione. A: DNA syn-

thesis; B: RNA synthesis. The symbols are the same as in Fig. 1.

0 20 40 60 80 100 I* '.I- '*'='.

160 A.

A

120

100

80

60

40

20 -

B

01 0 20 40 60 80 100

[Selenlum]pM

Fig. 3. Inhibition of DNA and RNA synthesis in WI-38

and WI-38VA cells by selenite. Cells were incubated for 1 h with the indicated concentration of selenium com-

pound and DNA and RNA synthesis was measured as in Fig. 1. A: DNA synthesis; B: RNA synthesis. (0): WI-38

cells, selenite; @): WI-38 cells, selenodiglutathione; (01: WI-38VA cells, selenite; (@: WI-38VA cells, selenodiglutathione.

Table II. Uptake of selenite by A549 and CCL-210 cells.

Cells Cell-associated selenium (pmoles/106 cells)

Total Acid soluble

A549 583’ zt 78 19

CCL-210 18.5’ f 0.5 6

A549/CCL-210 31 3

‘Mean of two experiments.

Acid precipitable

502 12 42

their insensitivity to selenite can be overcome by by-passing the requirement for the reaction to form the selenotrisulfide.

We have also detected another difference between the A549 and CCL-210 cells which could account for their diiering sensitivities to selenite. Table II shows that after a l-h ex- posure to selenite there was more cell- associated selenium, both acid precipitable and acid soluble, with the A549 cells than with the CCL-210 cells. Thus, the greater inhibitory effect of selenite in the A549 cells could be due to greater uptake and/or retention of selenium compounds.

We have also observed that there is a qualitative difference in the selenium- containing compounds which result upon incubation of selenite with the acid soluble frac- tion of extracts of the A549 and CCL-210 cells. Analysis of these compounds by thin- layer chromatography and autoradiography is shown in Fig. 4. Based upon our earlier work [12], we can conclude that with the A549 extract the major Se-containing species is selenodiglutathione (spot no. 1) , whereas with the CCL-210 extract there are significant amounts of other species, including selenodimercaptoethylamine (spot no. 4) and the mixed selenotrisulfides of glutathione and mercaptoethylamine (spot no. 3) and gluta- thione and cysteine (spot no. 2 ) , respectively. This suggests that there may be a difference in the composition of the intracellular SH pools; the qualitative difference in the selenium- containing species which are formed could ex- plain the difference in potency of selenite in the two types of cells.

Dkussion

Our comparative experiments with the A549 tumor cells and the CCL-210 normal cells have demonstrated three differences possible reasons for their differential sensitivity to selenite. One is a difference in their in- tracellular levels of SH compounds, a second is a difference in the amount of intracellular Se

l-

-4

-3

Ai 1 CCL-210

F& 4. Analysis of the Se-containing products of the reaction of selenite with the acid-soluble fraction of extracts of A549 and CCL-210 cells. The acid soluble

fraction of the cell extracts was prepared as described previously 1121 and incubated with 75Se]selentte for 30 min at room temperature. The resulting mixture was analyzed by thin layer chromatography on PEI-cellulose and autoradiography as described in Experimental Pro- cedures.

48

after exposure to selenite and the third is a qualitative difference in their potential for selenotrisulfide formation. The fact that the cells show no difference in sensitivity to selenodiglutathione strongly suggests the im- portance of the reaction in which it is formed. All three differences between the cells could be related to this reaction. Our previous results have also indicated that this reaction is impor- tant in determining the extent of inhibition of DNA and RHA synthesis by selenite [lo, 111. They have also provided a biochemical ex- planation for the requirement for this reaction: purified DNA and RHA polymerases, the en- zymes which catalyze the synthesis, are in- hibited by the selenotrisulfides but not by selenite. Furthermore, there are differences in the potency of various selenotrisulfides in this inhibitory effect 191. Thus our earlier observa- tions together with the present findings lead us to suggest that differences between normal and tumor cells in their sensitivity to selenite can result from quantitative and/or qualitative dif- ferences in the intracellular reaction of selenite to form selenotrisulfides.

Although the present results provide ad- ditional evidence of the importance of selenotrisulfide formation for the inhibition by selenite of nucleic acid synthesis, it is important to realize that the formation of selenotrisulfides is only the first step in the intracellular metabolism of selenite and numerous other species are formed [14]. Our results do not rule out the involvement of other selenium- containing metabolites in the inhibitory effects of selenite .

In most cases malignant cells are more sen- sitive to selenite than normal cells [1,6,7,18,19,21,27]. The question arises as to whether this difference may reflect a property of the transformed state. Our studies with WI- 38 and WI-38VA cells bear upon this question since the latter were derived by in vitro transformation of the former. The absence of any significant difference between these cells in their sensitivity to selenite suggests that the transformed phenotype itself is not sufficient to generate a differential sensitivity to selenite.

Nevertheless, our demonstration that cellular SH compounds and selenotrisulfide formation are important suggests an approach to assess- ing which tumors are likely to be prove most sensitive to selenium compounds as chemotherapeutic agents.

Acknowledgmenta

We thank Dr. Paula Caffrey for helpful discussions. This work was supported by Grant ES-04087 from the National Institutes of Health and by a grant from the American Insti- tute for Cancer Research. This is Publication No. 98 from the Department of Biological Sciences, Rutgers-Newark.

References

1

2

3

4

5

6

7

8

9

10

11

Abdullaev, F.I., Allakhverdiev, LA. and Mamedova, G.R. (1989) Inhibition of RNA synthesis in animal cells by sodium selenite. Biokhimiya, 54, 145- 148. Batist, G. (1988) Selenium. Preclinical studies of antican- cer therapeutic potential. Biol. Trace Elem. Res., 15, 223 - 229. Caffrey, P.B. and Frenkel, G.D. (1991) Inhibition of cell colony formation by selenite: involvement of glutathione. Mol. Pharmacol., 39, 281-284. Combs, G.F. and Combs, S.B. (1986) The Role of Selenium in Nutrition. Academic Press, Orlando, Florida. Ellman, G.L. (1959) Tissue sulfhydryl groups. Arch. Biochem. Biophys., 82, 70- 77. Fico, M.E., Watrach, A., Watrach, M. and Milner, J.A. (1985) Differences in sensitivity of neoplastic and non- neoplastic canine mammary cells to selenium. Fed. Proc., 44, 1673 - 1678. Fico, M.E., Poirier, K.A., Watrach, A., Watrach, M. and Milner, J.A. (1986) Differential effects of selenium on nor- mal and nonneoplastic mammary cells. Cancer Res., 46, 384-388. Frenkel, G.D. (1985) Effects of sodium selenite and selenate on DNA and RNA synthesis in vitro. Toxicol. Lett., 25, 219-223. Frenkel, G.D., Walcott, A. and Middleton, C. (1987) In- hibition of RNA and DNA polymerases by product of the reaction of selenite with sulfhydryl compounds. Mol. Phar- macol., 31, 112- 116. Frenkel, G.D. and Falvey, D. (1988) Evidence for the in- volvement of cellular sulfhydryl compounds in the inhibi- tion of DNA synthesis by selenite. Mol. Pharmacol., 34, 573-577. Frenkel, G.D. and Falvey, D. (1989) Involvement of sulfhydryl compounds in the inhibition of cellular RNA syn- thesis by selenite. Biochem. Pharmacol., 38,2849 - 2852.

49

12

13

14

15

16

17

18

19

20

Frenkel, G.D., Falvey, D. and MacVicar, C. (1991) Prod-

uct of the reaction of selenite with intracellular sulhydryl

compounds. Biol. Trace Elem. Res., 30, 9- 18.

Ganther, H.E. (1968) Selenotrisulfides. Formation by

reaction of thiols with selenious acid. Biochemistry., 7,

2898 - 2905.

Ganther, H.E. (1986) Pathways of selenium metabolism

including respiratory products. J. Am. Coil. Toxicol., 5,

1-5.

Golczewski, J. and Frenkel, G.D. (1989) Cellular

selenoproteins and the effects of selenite on cell prolifera-

tion Biol. Trace Elem. Res., 20, 115- 126.

Gruenwedel, D.W. and Cruikshank, M.K. (1979) The in-

fluence of sodium selenite on the viability and intracellular

synthetic activity (DNA, RNA and protein synthesis) of

Hela S - 3 cells. Toxicol. Appl. Pharmacol., 50, 1 - 7.

Ip, C. and Thompson, H.J. (1989) New approaches to

cancer chemoprevention with difluoromethylornithine and

selenite. J. Natl. Cancer Inst., 81, 839-843.

Kuchan, N.J., Fico-Santoro, M.A. and Milner, J.A.

(1990) Consequences of selenite supplementation on the

growth and metabolism of cultures of canine mammary

cells. J. Biochem. Nutr., 1, 478-483.

Kuchan, N.J. and Milner, J.A. (1991) Influence of in-

tracellular glutathione on selenite-mediated growth inhibi-

tion of canine mammary tumor cells. Cancer Lett., 57,

181-186.

LeBoeuf, R., Laishes, B. and Hoekstra, W. (1985) Effects

of selenium on cell proliferation in rat liver and mammalian

cells as indicated by cytokinetic and biochemical analysis.

Cancer Res., 45, 5496- 5504.

21

22

23

24

25

26

27

28

29

Medina, D. and Oborn, C. (1981) Differential effects of

selenium on the growth of mouse mammary cells in vitro.

Cancer Lett., 13, 333-344.

Medina, D. (1986) Mechanism of selenium inhibition of

tumorigenesis. J. Am. Coil. Toxicol., 5, 21-26.

Medina, D. and Morrison, D. (1988) Current ideas on

selenium as a chemopreventative agent. Pathol. Im-

munopathol. Res., 7, 187 - 199.

Milner, J.A. and Hsu, C.R. (1981) Inhibitory effect of

selenium on the growth of L1210 leukemic cells. Cancer

Res., 41, 1652- 1656.

Poirier, K.A. and Milner, J.A. (1983) Factors influencing

the antitumorigenic properties of selenium. J. Nutr., 113,

2147 -2154.

Sedlak, J. and Lindsay, R.H. (1968) Estimation of total,

protein-bound and non-protein sulfhydryl groups in tissue

with Ellman’s Reagent. Anal. Biochem., 25, 192 - 205.

Thompson, H.J. and Anne, M.R. (1990) Differences in

selenium concentrations in target tissiues and their

relevance to its anticarcinogenicity. Nutr. Res., 10,

81-89.

Vernie, L.N., DeVries, M., Karreman, L., Topp, R. and

Bont, W.S. (1983) Inhibition of amino acid incorporation

in a cell-free system and inhibition of protein synthesis in

cultured cells by reaction products of selenite and thiols.

B&hem. Biophys. Acta, 739, l- 7.

Yu, S.Y.,Ao,P., Wang,L.M., Huang, S.L., Chen, H.C.,

Lu, X.P. and Liu, Q.Y. (1988) Biochemical and cellular

aspects of the anticancer activity of selenium. Biol. Trace.

Elem. Res., 15, 243-255.