Changes in Ornithine Decarboxylase Activity and Polyamine Levels in Response to Eight Different Forms of Selenium H. J. Thompson, C. Ip, and H. E. Ganther HJT.
Laboratory of Nutrition Research, AMC Cancer Research Center, Denver, Colorado.-CI. Department of Breast Surgery, Roswell Park Cancer Institute, Buflalo, New York.-HEG. Department of Nutritional Sciences, University of Wisconsin, Madison, Wisconsin
ABSTRACT The biological activity of selenium is known to depend on its chemical form. In this study, eight forms of selenium that differed in oxidation state or degree of methylation were studied for their acute effects on the activities of ornithine decarboxylase (ODC) and S-adenosylmethionine decarboxylase (AdoMet DC) and on the concentrations of the polyamines putrescine, spermidine, and spermine in the liver. The polyamine pathway was studied because it is involved in the control of cell growth and in the cell’s response to trophic, carcinogenic, and toxic stimuli, activities that selenium has been reported to affect. Female Sprague Dawley rats were administered 12 pmol Se/kg body weight via intraperitoneal injection and were sacrificed six hours later. Injection of sodium selenate, sodium selenite, selenomethionine, Se-methylselenocysteine, selenobetaine, and selenobetaine methyl ester resulted in significant increases in liver selenium, whereas injection of dimethylselenoxide and trimethylselenonium chloride did not. ODC activity and AdoMet DC activity were induced by those selenium compounds that also increased liver selenium content, but the magnitude of enzyme induction by those compounds was not correlated with the hepatic concentration of total selenium determined fluorometrically. Furthermore, the induction of ODC activity by the various forms of selenium did not result in concomitant increases in putrescine, spermidine, and spermine except in the case of selenite. Given that alterations in the metabolism of selenium are induced when the level of tissue selenium is elevated and that the relative abundance of various selenometabolites can be affected by the point of entry of selenium into intermediary metabolism, these data suggest that the changes that were observed in enzyme activities and polyamine levels are likely to be associated with the accumulation of a specific metabolite of selenium. The relevance of these findings to elucidation of the biological activities attributable to various forms of selenium is under investigation.
Address reprint requests and correspondence to: Dr. Thompson, Laboratory AMC Cancer Research Center, 1600 Pierce Street, Denver, CO 80214.
of Nutrition
Research,
283 Journal of Inorganic Biochemistry, b&283-292 (1991) 0 1991 Elsevier Science Publishing Co., Inc., 655 Avenue of the Americas, NY, NY 10010 0162-0134/91/$3.50
284
H. J. Thompson et al.
Abbreviations Omithine
decarboxylase,
ODC; S-adenosylmethionine
decarboxylase.
AdoMet DC.
INTRODUCTION In a recent commentary entitled “The two faces of selenium.” Oldfield [l] briefly reviewed the dichotomous effects of selenium, including its nutritional, anticancer. and toxic activities. Too little or too much selenium is known to be associated with pathological lesions in both animals and humans. One fascinating aspect of the role of selenium in biological systems is that its activity is manifested by a wide variety of chemical forms and not by the element per se 121. As shown in Figure I. reduction and methylation reactions are prominent features of selenium metabolism in animals. Metabolism via this pathway alters the chemical form of selenium and thereby confers or eliminates biological activity. For example, it has recently been reported that demethylation is a significant feature of selenium metabolism and influences the biological activity of methylated selenium compounds 131. Consequently, the point of entry of selenium into metabolism and its effect on the relative abundance of the various species of selenium in a tissue are likely to affect the biological activities associated with a particular tissue concentration of Se. For example, inorganic selenium. such as hydrogen selenide, is generally believed to be the selenium precursor for incorporation into glutathione peroxidase [4]. On the other hand, preliminary studies seem to suggest that methylated selenides may be the active species in cancer chemoprevention [5, 61. Thus convertibility to glutathione peroxidase may correlate well with nutritional potency but may not necessarily be a criterion for anticarcinogenic activity. The forms of selenium that account for its toxicity have yet to be elucidated. In this study. we investigated: 1) whether specific
GS-Se,-SG ! $
SELENOCYSTEINE
GS-SeH I ------+>
H,Se
t---__-_-____--i
cH,seH +---’ METHYLSELENOL
mM _______. ____+ SELENOBETAINE
L__._.__-.--.-
METHYL
ESTER -7
.__J
i cH3SecHJ
.+__-
DIMETHYLSELENIDE
1 w&se+ TRIMETHYLSELENONIUM
FIGURE 1.
Intermediary metabolism of selenium
~~~,-sw~H,-~H(N~~~)-cooH L:;zzJ
CH3-Se(O)-CH3 DIMETHYL
SELENOXIDE
SELENIUM
AND POLYAMINE
METABOLISM
285
forms of selenium could be implicated in the induction of ODC activity and the modulation of polyamine biosynthesis by controlling the point of entry of Se into its intermediary metabolism and 2) how the resulting observations correlated with the biological activities known to be associated with these selenium compounds. We chose to evaluate the effects of selenium on acute changes in ODC activity and polyamine levels since these parameters have been shown to change rapidly in response to trophic, carcinogenic, and toxic stimuli which are types of biological responses selenium has been reported to affect [7 - 111.
MATERIALS Selenium
AND METHODS
Treatment Female Sprague Dawley rats were used. They were housed in an environmentally controlled animal facility maintained at 25°C and 50% relative humidity with a 12-hour light, 12-hour dark cycle. All rats were fed an AIN-76A diet formulation [ 12, 131 before receiving any treatment. This diet provides 0.1 ppm Se as selenite from the mineral mix. At approximately 55 days of age, 72 rats were randomized into one of nine treatment groups shown in Table 1. Each rat was injected intraperitoneally at a dose of 12 pmol Se/kg body weight with one of the eight different selenium compounds. Rats were killed six hours after the injection and livers were excised and rapidly frozen in liquid nitrogen. Liver samples were subsequently analyzed for ODC and S-adenosylmethionine decarboxylase (AdoMet DC) activities as well as for polyamine and selenium content.
ODC and AdoMet DC Assays. Tissue homogenates were prepared and the enzyme activities in the 105,000 g supematant fraction were determined by the standard procedure of base entrapment of released 14C0, from D,L-[l-‘4C]omithine and S-adenosyl-L-[carboxyl14C] methionine, respectively [ 14, 151. Enzyme activity was calculated on the basis of the amount of i4C0, liberated in 60 min/mg protein. The
TABLE 1. Liver Selenium Concentrations Following a Single Injection of Various Selenium Compounds” Selenium Compound None Sodium selenate Sodium selenite Selenomethionine Selenobetaine Se-methylselenocysteine Selenobetaine methyl ester Dimethylselenoxide Trimethylselenonium chloride
Selenium Pg/gb 0.93 4.81 5.20 6.27 3.41 5.03 4.58 0.91 1.08
f + f jz f + + f +
o.02c 0.22d*” 0.25d,” O.lle 0.37d 0.22d,c 0.26d,e 0.38’ 0.06’
a Rats were given an i.p. injection of 12 pmol Se/kg body weight and were killed 6 hours later. b Each value is a mean f SEM. c,d*e Values with no superscript in common are statistically different, p < 0.05.
286
H. J. Thompson et al.
protein content of the supernatant fraction was measured 1161 using the Bio-Rad Protein Assay Kit.
by the method of Bradford
Polyamine Determination. The analysis of polyamines was performed using a modification of the method of Seiler and Wiechmann as described previously [ 171. Tissue homogenates were incubated with 0.5 N perchloric acid on ice for 30 min and then centrifuged to yield a protein free supernatant. This fraction was derivatized with dansyl chloride and polyamines were separated by one-dimensional thin layer chromatography. Putrescine, spermidine, and spermine content were quantified via fluorescence scanning using a Shimadzu Model CS 930 high speed scanner. Selenium Assay.
Duplicate samples of liver from each rat were wet ashed in a concentrated nitricjperchloric acid mixture (3: 1, v/v). Selenium content of the tissue digest was determined fluorometrically with the use of 2.3.diaminonapthalene [ 181.
Selenium Compounds.
Sodium selenate, sodium selenite and DL-selenomethionine were purchased from Sigma. The other compounds were synthesized as described elsewhere: selenobetaine and selenobetaine methyl ester [S]; Se-methylselenocysteine 161; dimethylselenoxide 161: and trimethylselcnonium chloride [ 191~
Statistica/ Analysis. All data were evaluated for their compliance to assumptions of normality. Tissue selenium, ODC, and AdoMet DC data were subjected to umvariate analysis of variance [20]. Pair-wise comparisons among treatment groups were made as recommended by Tukey [21]. Since tissue polyamine pools are metaboiitally interrelated, they do not necessarily vary independently. Therefore polyamine data were also analyzed by multivariate analysis of variance (MANOVA) 1221 in order to decipher statistically significant effects of treatment on individual components of polyamine biosynthesis (putrescine, spermidine. and spermine) as well as the overall effect of treatment on the biosynthetic pathway. The relationship between hepatic concentrations of selenium resulting from the injection of those compounds that increased liver ODC activity and the magnitude of ODC induction was further evaluated by polynomial regression analyses [20]. ODC and pofyamine data were also analyzed by the Kruskal Wallis procedure 1231 ttr establish a rank order 01‘ the effects of various selenium compounds on these biochemical parameters. RESULTS The effect of injecting an equimolar amount of selenium in one of eight different chemical forms on the concentration of selenium in the liver six hours after administration is shown in Table 1. With the exception of dimethylselenoxide and trimethylselenonium chloride each compound caused a significant increase in liver selenium. The magnitude of increase ranged from about 6.Sfold for selenomethionine to about 3.7-fold for selenobetaine. The activities of ODC and AdoMet DC found in the liver of rats following an acute injection of eight different forms of selenium are reported in Table 2. All except dimethylselenoxide and trimethylselenonium caused a significant induction of ODC activity. The magnitude of induction ranged from about 36fold for Se-methylselenocysteine to about s-fold for selenobetaine. The efficacy of these selenium compounds for ODC induction was roughly in the order: Se-methylselenocysteine > (selenomethionine, selenobetaine methyl ester, selenite) > (selenate, selenobetaine). The failure of dimethylselenoxide and trimethylselenonium to produce a significant
SELENIUM
AND POLYAMINE
METABOLISM
287
TABLE 2. Effect of Acute Selenium Injection on Liver ODC and AdoMet DC Activities
Selenium Compounda None Sodium selenate Sodium selenite Selenomethionine Selenobetaine Se-methylselenocysteine Selenobetaine methyl ester Dimethylselenoxide Trimethylselenonium chloride
ODCbxc Units/hr 15 137 218 247 199 540 232 25 22
+ + + + + f + + +
2d 30” 32e 40e 22e 73’ 25e 5d 6*
AdoMet DCbsC Units/hr 429 465 600 582 722 608 646 336 328
f f + f + + + + +
44d 72d 61e 41e 87e 99e 70e 30* 2gd
‘Rats were given an i.p. injection of 12 pmol Se/kg body weight and were killed 6 hours later. b Each value represents mean f SEM of eight observations. The ANOVA for effect of treatment with selenium was highly significant, p < 0.001. ‘A unit of enzyme activity is defined as a pmol CO, liberated mg protein. / d,e. Values within a column not sharing common superscripts are statistically different, p < 0.05.
increase in ODC activity may be related to their metabolic fates. Metabolism studies have indicated that trimetbylselenonium and dimethylselenoxide (S. Vadhanavikit and H. Ganther, unpublished data) are rapidly and almost quantitatively excreted; this is in agreement with our observation that their administration resulted in little selenium retention in the liver (Table 1). Comparison of the selenium concentration data (Table 1) with the ODC data (Table 2) suggests that elevated tissue selenium concentration is associated with ODC induction. However, when only those compounds that induced ODC activity were considered, the magnitude of ODC induction was not correlated with the concentration of liver selenium (r2 = 0.29, p > 0.09). AdoMet DC is the second most rate limiting step in polyamine biosynthesis. We observed (H. Thompson, unpublished data) that injection of 12 and 25 pmol selenite caused a decrease in liver S-adenosylmethionine: no treatment, 58.8 f 3.8 nmol/g versus selenite, 48.8 k 2.5 and 32.5 f 4.8 nmol/g, respectively. This suggests that the activity of AdoMet DC might also be affected by selenium and that effects of selenium on polyamine levels might be mediated through modulation of more than one enzyme. Data for AdoMet DC are presented in Table 2. Injection of selenite, selenomethionine, selenobetaine, and selenobetaine methyl ester resulted in significant increases in AdoMet DC activity. However, the stimulation of AdoMet DC was not as consistent and was much smaller in magnitude than that of ODC. Again, the injection of dimethylselenoxide or trimethylselenonium chloride failed to elicit any significant change in AdoMet DC activity. Table 3 shows the effect of treatment with the various selenium compounds on the liver concentration of putrescine, spermidine, and spermine. Statistical analyses of these data using MANOVA indicate highly significant effects of selenium treatment on each amine as well as on the total polyamine level in the liver. The magnitude of the effect of selenium was greatest for putrescine followed by spermidine and then
288
H. J. Thompson et al.
TABLE 3. Effect of Acute Selenium Injection on Changes in Liver Putrescine. Spermidine, and Spermine Total
Selenium Compound” -.
Putrescineh
Spermidineh
pmollg
pmol!g --__.---
‘44 + 9’ 103 f 15’.d
“None Sodium selenate Sodium
/Imol /g
1222 +_ 120d
93 +_27’.d 84 + 24’,d
795 888 755 701 75s 1036
71 t7’
Selenobetaine methyl ester Dimethylselenoxide Trlmethylselenonium chloride Overall MANOVA for treatment p value ----. -_.-
bLrnol,ig _~..-__-.--.___-._----
Polyamines’
624 +- 48’ 94s _+ i(Hc , regulated through a compensatory mechanism involving the interconversion oi‘ putrescine, spermidine. and spermine [34], and selenium has been reported to mducc one of the enzymes involved in the catabolism of polyamines via the acetylation pathway, nameI> spermidine acetyl transferase [35]. Enhanced catabolism of polyamines could oft‘ser increases in polyamine levels due to ODC induction. It 1s possible that various selenium compounds might differentially affect polyamine catabolism ~a the acetylation reaction. The implication of such effects in accounting for the biological acti+ of various forms of selenium requires further investigation. In summary, the results of this study. while confirming that selenite induces ODC activity and polyamine biosynthesis, indicate that not ail forms of selenium generate the same responses even at equivalent tissue concentrations of selenium. A most striking observation is that only those selenium compounds that have cancer inhibitory activity cause a significant induction of ODC levels. Sodium selenate. sodium selenite, selenomethionine, Se-.methylselenocysteine. seienobetaine, and selenobetaine methyl ester have all been shown to suppress tumorigenehts 15, 6. 361, Dimethylselenoxide, on the other hand, is much less active [C,], while trimethylselenonium does not inhibit tumor development even at a ievel of 40 ppm Se in the diet [ 191. The latter two compounds, which have inherently lower toxicity, also failed to induce ODC. Given these observations, it becomes especially important to determine if the induction of QDC activity is associated with the accumulation of particular species of selenium and whether the induction of ODC by selenium is related to a particular set of biological responses. At least three possible explanations merit investigation: ODC induction by selenium is required for its anticancer activity: it is a component of a toxic response to selenium: or it is a coincidental biochemical signal dissociated from either the anticarcinogenic or toxic actions of selenium. In addition, further studies of the modulation by various forms of selenium of both the
SELENIUM
AND POLYAMINE
METABOLISM
291
anabolic and catabolic pathways of polyamine metabolism may be useful in unraveling the mechanisms responsible for the various biological activities attributed to selenium.
This work was supported by Grant No. CA 45164 from the National Cancer Institute, NIH, and by the College of Agricultural and Life Sciences, University of Wisconsin, Madison, Wisconsin. The authors would like to thank Rose Marie Budnick, JoAnn Juzefyk, Tom Keow, Rita Pawlak, and Janet Treichel for their technical assistance and Joyce Manley and Connie Ryan for their assistance in preparing this manuscript.
REFERENCES 1. J. E. Oldfield, J. Nutr. 117, 2002 (1987). H. E. Ganther, J. Am. CON. Toxicol. 5, 1 (1986). H. E. Ganther, S. Vadhanavikit, and C. Ip, FASEB J. 4, A372 (1990). R. A. Sunde, Annu. Rev. Nutr. 10, 451 (1990). C. Ip and H. E. Ganther, Cancer Res. 50, 1206 (1990). C. Ip, C. Hayes, R. M. Budnick, and H. E. Ganther, Cancer Res. 51, 595 (1991). D. R. Morris and R. H. Fillingame, Annu. Rev. Biochem. 43, 303 (1974). J. Janne, H. Poso, and A. Raina, Biochim. Biophys. Acta 473, 241 (1978). E. S. Canellakis, D. Viceps-Madore, D. A. Kyriakidis, and J. S. Heller, Curr. Top. Cell. Regul. 15, 155 (1979). 10. D. H. Russell, Adv. Enzyme ReguI. 21, 201 (1983). 11. C. W. Tabor and H. Tabor, Annu. Rev. Biochem. 53, 749 (1984). 12. Report of the American Institute of Nutrition Ad Hoc Committee on Standards for Nutritional Studies, J. Nutr. 107, 1340 (1977). 13. Second Report of the American Institute of Nutrition Ad Hoc Committee on Standards for Nutritional Studies, J. Nutr. 110, 1726(1980). 14. J. E. Seely and A. E. Pegg, Methods Enzymol. 94, 158 (1983). 15. A. E. Pegg and H. Poso, Methods Enzymol. 94, 234 (1983). 16. M. M. Bradford, Anal. Biochem. 72, 248 (1976). 17. R. Minocha, H. J. Thompson, L. D. Meeker, and E. J. Herbst, Biochem. J. 223, 929
2. 3. 4. 5. 6. 7. 8. 9.
(1984). 18. R. Hasunuma, T. Ogawa, and Y. Kawanishi, Anal. Biochem. 126, 242 (1982). 19. C. Ip and H. E. Ganther, Carcinogenesis 9, 1481 (1988). 20. G. W. Snedecor and W. G. Cochran, Statistical Methods, Ed. 6, Iowa State University Press, Ames, Iowa, 1967, p. 258. 21. R. G. Steel and I. H. Torrie, Principles and Procedures of Statistics, Ed. 1, McGraw-Hill, New York, 1960, p. 109. 22. D. F. Morrison, Multivariate Statistical Methods, McGraw-Hill, New York, 1976. 23. J. B. Kruskal, Psychometrika 29, 115 (1964). 24. C. Ip, J. Am. Golf. Toxicol. 5, 7 (1986). 25. C. Ip, Ann. Clin. Res. 18, 22 (1986). 26. D. Medina and D. G. Morrison, Pathol. Zmmunopathol. Res. 7, 187 (1988). 27. T. G. O’Brien, Cancer Res. 36, 2644 (1976). 28. T. G. O’Brien, R. C. Simsiman, and R. K. Boutwell, Cancer Res. 35, 1662 (1975). 29. A. K. Verma, E. Erturk, and G. T. Bryan, Cancer Res. 43, 5964 (1983).
292
H. J. Thompson et al.
30. L. A. Van Rooijen. H. J. Derks. R. Van Wijk, and A. Bisschop. Curcinogenesis 31. 32. 33. 34. 35. 36.
(1984). S. Colombatto and M. A. Grille, Med. Sri. Res. 15, 987 (1087) C. Ip and H. J. Thompson. J. Nat/. Cancer Inst. 81. 839 i1989r. S. J. Foster. R. J. Kraus. and H. E. Ganther. Arch Aiochem. (1986). A. E. Pegg, Cancer Res. 48, 759 (1988). S. Colombatto, D. C. Pezzali, and M. A. Grillo. Inr J. Biochern. C. Ip and C. Hayes. C’mcinogenesis 10, 921 (1989).
Received May 7, 1991; accepted June 5. 1991
Biophys.
5, 225
251, 17
19. 725 (1987).
Laboratory of Nutrition Research, AMC Cancer Research Center, Denver, Colorado.-CI. Department of Breast Surgery, Roswell Park Cancer Institute, Buflalo, New York.-HEG. Department of Nutritional Sciences, University of Wisconsin, Madison, Wisconsin
ABSTRACT The biological activity of selenium is known to depend on its chemical form. In this study, eight forms of selenium that differed in oxidation state or degree of methylation were studied for their acute effects on the activities of ornithine decarboxylase (ODC) and S-adenosylmethionine decarboxylase (AdoMet DC) and on the concentrations of the polyamines putrescine, spermidine, and spermine in the liver. The polyamine pathway was studied because it is involved in the control of cell growth and in the cell’s response to trophic, carcinogenic, and toxic stimuli, activities that selenium has been reported to affect. Female Sprague Dawley rats were administered 12 pmol Se/kg body weight via intraperitoneal injection and were sacrificed six hours later. Injection of sodium selenate, sodium selenite, selenomethionine, Se-methylselenocysteine, selenobetaine, and selenobetaine methyl ester resulted in significant increases in liver selenium, whereas injection of dimethylselenoxide and trimethylselenonium chloride did not. ODC activity and AdoMet DC activity were induced by those selenium compounds that also increased liver selenium content, but the magnitude of enzyme induction by those compounds was not correlated with the hepatic concentration of total selenium determined fluorometrically. Furthermore, the induction of ODC activity by the various forms of selenium did not result in concomitant increases in putrescine, spermidine, and spermine except in the case of selenite. Given that alterations in the metabolism of selenium are induced when the level of tissue selenium is elevated and that the relative abundance of various selenometabolites can be affected by the point of entry of selenium into intermediary metabolism, these data suggest that the changes that were observed in enzyme activities and polyamine levels are likely to be associated with the accumulation of a specific metabolite of selenium. The relevance of these findings to elucidation of the biological activities attributable to various forms of selenium is under investigation.
Address reprint requests and correspondence to: Dr. Thompson, Laboratory AMC Cancer Research Center, 1600 Pierce Street, Denver, CO 80214.
of Nutrition
Research,
283 Journal of Inorganic Biochemistry, b&283-292 (1991) 0 1991 Elsevier Science Publishing Co., Inc., 655 Avenue of the Americas, NY, NY 10010 0162-0134/91/$3.50
284
H. J. Thompson et al.
Abbreviations Omithine
decarboxylase,
ODC; S-adenosylmethionine
decarboxylase.
AdoMet DC.
INTRODUCTION In a recent commentary entitled “The two faces of selenium.” Oldfield [l] briefly reviewed the dichotomous effects of selenium, including its nutritional, anticancer. and toxic activities. Too little or too much selenium is known to be associated with pathological lesions in both animals and humans. One fascinating aspect of the role of selenium in biological systems is that its activity is manifested by a wide variety of chemical forms and not by the element per se 121. As shown in Figure I. reduction and methylation reactions are prominent features of selenium metabolism in animals. Metabolism via this pathway alters the chemical form of selenium and thereby confers or eliminates biological activity. For example, it has recently been reported that demethylation is a significant feature of selenium metabolism and influences the biological activity of methylated selenium compounds 131. Consequently, the point of entry of selenium into metabolism and its effect on the relative abundance of the various species of selenium in a tissue are likely to affect the biological activities associated with a particular tissue concentration of Se. For example, inorganic selenium. such as hydrogen selenide, is generally believed to be the selenium precursor for incorporation into glutathione peroxidase [4]. On the other hand, preliminary studies seem to suggest that methylated selenides may be the active species in cancer chemoprevention [5, 61. Thus convertibility to glutathione peroxidase may correlate well with nutritional potency but may not necessarily be a criterion for anticarcinogenic activity. The forms of selenium that account for its toxicity have yet to be elucidated. In this study. we investigated: 1) whether specific
GS-Se,-SG ! $
SELENOCYSTEINE
GS-SeH I ------+>
H,Se
t---__-_-____--i
cH,seH +---’ METHYLSELENOL
mM _______. ____+ SELENOBETAINE
L__._.__-.--.-
METHYL
ESTER -7
.__J
i cH3SecHJ
.+__-
DIMETHYLSELENIDE
1 w&se+ TRIMETHYLSELENONIUM
FIGURE 1.
Intermediary metabolism of selenium
~~~,-sw~H,-~H(N~~~)-cooH L:;zzJ
CH3-Se(O)-CH3 DIMETHYL
SELENOXIDE
SELENIUM
AND POLYAMINE
METABOLISM
285
forms of selenium could be implicated in the induction of ODC activity and the modulation of polyamine biosynthesis by controlling the point of entry of Se into its intermediary metabolism and 2) how the resulting observations correlated with the biological activities known to be associated with these selenium compounds. We chose to evaluate the effects of selenium on acute changes in ODC activity and polyamine levels since these parameters have been shown to change rapidly in response to trophic, carcinogenic, and toxic stimuli which are types of biological responses selenium has been reported to affect [7 - 111.
MATERIALS Selenium
AND METHODS
Treatment Female Sprague Dawley rats were used. They were housed in an environmentally controlled animal facility maintained at 25°C and 50% relative humidity with a 12-hour light, 12-hour dark cycle. All rats were fed an AIN-76A diet formulation [ 12, 131 before receiving any treatment. This diet provides 0.1 ppm Se as selenite from the mineral mix. At approximately 55 days of age, 72 rats were randomized into one of nine treatment groups shown in Table 1. Each rat was injected intraperitoneally at a dose of 12 pmol Se/kg body weight with one of the eight different selenium compounds. Rats were killed six hours after the injection and livers were excised and rapidly frozen in liquid nitrogen. Liver samples were subsequently analyzed for ODC and S-adenosylmethionine decarboxylase (AdoMet DC) activities as well as for polyamine and selenium content.
ODC and AdoMet DC Assays. Tissue homogenates were prepared and the enzyme activities in the 105,000 g supematant fraction were determined by the standard procedure of base entrapment of released 14C0, from D,L-[l-‘4C]omithine and S-adenosyl-L-[carboxyl14C] methionine, respectively [ 14, 151. Enzyme activity was calculated on the basis of the amount of i4C0, liberated in 60 min/mg protein. The
TABLE 1. Liver Selenium Concentrations Following a Single Injection of Various Selenium Compounds” Selenium Compound None Sodium selenate Sodium selenite Selenomethionine Selenobetaine Se-methylselenocysteine Selenobetaine methyl ester Dimethylselenoxide Trimethylselenonium chloride
Selenium Pg/gb 0.93 4.81 5.20 6.27 3.41 5.03 4.58 0.91 1.08
f + f jz f + + f +
o.02c 0.22d*” 0.25d,” O.lle 0.37d 0.22d,c 0.26d,e 0.38’ 0.06’
a Rats were given an i.p. injection of 12 pmol Se/kg body weight and were killed 6 hours later. b Each value is a mean f SEM. c,d*e Values with no superscript in common are statistically different, p < 0.05.
286
H. J. Thompson et al.
protein content of the supernatant fraction was measured 1161 using the Bio-Rad Protein Assay Kit.
by the method of Bradford
Polyamine Determination. The analysis of polyamines was performed using a modification of the method of Seiler and Wiechmann as described previously [ 171. Tissue homogenates were incubated with 0.5 N perchloric acid on ice for 30 min and then centrifuged to yield a protein free supernatant. This fraction was derivatized with dansyl chloride and polyamines were separated by one-dimensional thin layer chromatography. Putrescine, spermidine, and spermine content were quantified via fluorescence scanning using a Shimadzu Model CS 930 high speed scanner. Selenium Assay.
Duplicate samples of liver from each rat were wet ashed in a concentrated nitricjperchloric acid mixture (3: 1, v/v). Selenium content of the tissue digest was determined fluorometrically with the use of 2.3.diaminonapthalene [ 181.
Selenium Compounds.
Sodium selenate, sodium selenite and DL-selenomethionine were purchased from Sigma. The other compounds were synthesized as described elsewhere: selenobetaine and selenobetaine methyl ester [S]; Se-methylselenocysteine 161; dimethylselenoxide 161: and trimethylselcnonium chloride [ 191~
Statistica/ Analysis. All data were evaluated for their compliance to assumptions of normality. Tissue selenium, ODC, and AdoMet DC data were subjected to umvariate analysis of variance [20]. Pair-wise comparisons among treatment groups were made as recommended by Tukey [21]. Since tissue polyamine pools are metaboiitally interrelated, they do not necessarily vary independently. Therefore polyamine data were also analyzed by multivariate analysis of variance (MANOVA) 1221 in order to decipher statistically significant effects of treatment on individual components of polyamine biosynthesis (putrescine, spermidine. and spermine) as well as the overall effect of treatment on the biosynthetic pathway. The relationship between hepatic concentrations of selenium resulting from the injection of those compounds that increased liver ODC activity and the magnitude of ODC induction was further evaluated by polynomial regression analyses [20]. ODC and pofyamine data were also analyzed by the Kruskal Wallis procedure 1231 ttr establish a rank order 01‘ the effects of various selenium compounds on these biochemical parameters. RESULTS The effect of injecting an equimolar amount of selenium in one of eight different chemical forms on the concentration of selenium in the liver six hours after administration is shown in Table 1. With the exception of dimethylselenoxide and trimethylselenonium chloride each compound caused a significant increase in liver selenium. The magnitude of increase ranged from about 6.Sfold for selenomethionine to about 3.7-fold for selenobetaine. The activities of ODC and AdoMet DC found in the liver of rats following an acute injection of eight different forms of selenium are reported in Table 2. All except dimethylselenoxide and trimethylselenonium caused a significant induction of ODC activity. The magnitude of induction ranged from about 36fold for Se-methylselenocysteine to about s-fold for selenobetaine. The efficacy of these selenium compounds for ODC induction was roughly in the order: Se-methylselenocysteine > (selenomethionine, selenobetaine methyl ester, selenite) > (selenate, selenobetaine). The failure of dimethylselenoxide and trimethylselenonium to produce a significant
SELENIUM
AND POLYAMINE
METABOLISM
287
TABLE 2. Effect of Acute Selenium Injection on Liver ODC and AdoMet DC Activities
Selenium Compounda None Sodium selenate Sodium selenite Selenomethionine Selenobetaine Se-methylselenocysteine Selenobetaine methyl ester Dimethylselenoxide Trimethylselenonium chloride
ODCbxc Units/hr 15 137 218 247 199 540 232 25 22
+ + + + + f + + +
2d 30” 32e 40e 22e 73’ 25e 5d 6*
AdoMet DCbsC Units/hr 429 465 600 582 722 608 646 336 328
f f + f + + + + +
44d 72d 61e 41e 87e 99e 70e 30* 2gd
‘Rats were given an i.p. injection of 12 pmol Se/kg body weight and were killed 6 hours later. b Each value represents mean f SEM of eight observations. The ANOVA for effect of treatment with selenium was highly significant, p < 0.001. ‘A unit of enzyme activity is defined as a pmol CO, liberated mg protein. / d,e. Values within a column not sharing common superscripts are statistically different, p < 0.05.
increase in ODC activity may be related to their metabolic fates. Metabolism studies have indicated that trimetbylselenonium and dimethylselenoxide (S. Vadhanavikit and H. Ganther, unpublished data) are rapidly and almost quantitatively excreted; this is in agreement with our observation that their administration resulted in little selenium retention in the liver (Table 1). Comparison of the selenium concentration data (Table 1) with the ODC data (Table 2) suggests that elevated tissue selenium concentration is associated with ODC induction. However, when only those compounds that induced ODC activity were considered, the magnitude of ODC induction was not correlated with the concentration of liver selenium (r2 = 0.29, p > 0.09). AdoMet DC is the second most rate limiting step in polyamine biosynthesis. We observed (H. Thompson, unpublished data) that injection of 12 and 25 pmol selenite caused a decrease in liver S-adenosylmethionine: no treatment, 58.8 f 3.8 nmol/g versus selenite, 48.8 k 2.5 and 32.5 f 4.8 nmol/g, respectively. This suggests that the activity of AdoMet DC might also be affected by selenium and that effects of selenium on polyamine levels might be mediated through modulation of more than one enzyme. Data for AdoMet DC are presented in Table 2. Injection of selenite, selenomethionine, selenobetaine, and selenobetaine methyl ester resulted in significant increases in AdoMet DC activity. However, the stimulation of AdoMet DC was not as consistent and was much smaller in magnitude than that of ODC. Again, the injection of dimethylselenoxide or trimethylselenonium chloride failed to elicit any significant change in AdoMet DC activity. Table 3 shows the effect of treatment with the various selenium compounds on the liver concentration of putrescine, spermidine, and spermine. Statistical analyses of these data using MANOVA indicate highly significant effects of selenium treatment on each amine as well as on the total polyamine level in the liver. The magnitude of the effect of selenium was greatest for putrescine followed by spermidine and then
288
H. J. Thompson et al.
TABLE 3. Effect of Acute Selenium Injection on Changes in Liver Putrescine. Spermidine, and Spermine Total
Selenium Compound” -.
Putrescineh
Spermidineh
pmollg
pmol!g --__.---
‘44 + 9’ 103 f 15’.d
“None Sodium selenate Sodium
/Imol /g
1222 +_ 120d
93 +_27’.d 84 + 24’,d
795 888 755 701 75s 1036
71 t7’
Selenobetaine methyl ester Dimethylselenoxide Trlmethylselenonium chloride Overall MANOVA for treatment p value ----. -_.-
bLrnol,ig _~..-__-.--.___-._----
Polyamines’
624 +- 48’ 94s _+ i(Hc , regulated through a compensatory mechanism involving the interconversion oi‘ putrescine, spermidine. and spermine [34], and selenium has been reported to mducc one of the enzymes involved in the catabolism of polyamines via the acetylation pathway, nameI> spermidine acetyl transferase [35]. Enhanced catabolism of polyamines could oft‘ser increases in polyamine levels due to ODC induction. It 1s possible that various selenium compounds might differentially affect polyamine catabolism ~a the acetylation reaction. The implication of such effects in accounting for the biological acti+ of various forms of selenium requires further investigation. In summary, the results of this study. while confirming that selenite induces ODC activity and polyamine biosynthesis, indicate that not ail forms of selenium generate the same responses even at equivalent tissue concentrations of selenium. A most striking observation is that only those selenium compounds that have cancer inhibitory activity cause a significant induction of ODC levels. Sodium selenate. sodium selenite, selenomethionine, Se-.methylselenocysteine. seienobetaine, and selenobetaine methyl ester have all been shown to suppress tumorigenehts 15, 6. 361, Dimethylselenoxide, on the other hand, is much less active [C,], while trimethylselenonium does not inhibit tumor development even at a ievel of 40 ppm Se in the diet [ 191. The latter two compounds, which have inherently lower toxicity, also failed to induce ODC. Given these observations, it becomes especially important to determine if the induction of QDC activity is associated with the accumulation of particular species of selenium and whether the induction of ODC by selenium is related to a particular set of biological responses. At least three possible explanations merit investigation: ODC induction by selenium is required for its anticancer activity: it is a component of a toxic response to selenium: or it is a coincidental biochemical signal dissociated from either the anticarcinogenic or toxic actions of selenium. In addition, further studies of the modulation by various forms of selenium of both the
SELENIUM
AND POLYAMINE
METABOLISM
291
anabolic and catabolic pathways of polyamine metabolism may be useful in unraveling the mechanisms responsible for the various biological activities attributed to selenium.
This work was supported by Grant No. CA 45164 from the National Cancer Institute, NIH, and by the College of Agricultural and Life Sciences, University of Wisconsin, Madison, Wisconsin. The authors would like to thank Rose Marie Budnick, JoAnn Juzefyk, Tom Keow, Rita Pawlak, and Janet Treichel for their technical assistance and Joyce Manley and Connie Ryan for their assistance in preparing this manuscript.
REFERENCES 1. J. E. Oldfield, J. Nutr. 117, 2002 (1987). H. E. Ganther, J. Am. CON. Toxicol. 5, 1 (1986). H. E. Ganther, S. Vadhanavikit, and C. Ip, FASEB J. 4, A372 (1990). R. A. Sunde, Annu. Rev. Nutr. 10, 451 (1990). C. Ip and H. E. Ganther, Cancer Res. 50, 1206 (1990). C. Ip, C. Hayes, R. M. Budnick, and H. E. Ganther, Cancer Res. 51, 595 (1991). D. R. Morris and R. H. Fillingame, Annu. Rev. Biochem. 43, 303 (1974). J. Janne, H. Poso, and A. Raina, Biochim. Biophys. Acta 473, 241 (1978). E. S. Canellakis, D. Viceps-Madore, D. A. Kyriakidis, and J. S. Heller, Curr. Top. Cell. Regul. 15, 155 (1979). 10. D. H. Russell, Adv. Enzyme ReguI. 21, 201 (1983). 11. C. W. Tabor and H. Tabor, Annu. Rev. Biochem. 53, 749 (1984). 12. Report of the American Institute of Nutrition Ad Hoc Committee on Standards for Nutritional Studies, J. Nutr. 107, 1340 (1977). 13. Second Report of the American Institute of Nutrition Ad Hoc Committee on Standards for Nutritional Studies, J. Nutr. 110, 1726(1980). 14. J. E. Seely and A. E. Pegg, Methods Enzymol. 94, 158 (1983). 15. A. E. Pegg and H. Poso, Methods Enzymol. 94, 234 (1983). 16. M. M. Bradford, Anal. Biochem. 72, 248 (1976). 17. R. Minocha, H. J. Thompson, L. D. Meeker, and E. J. Herbst, Biochem. J. 223, 929
2. 3. 4. 5. 6. 7. 8. 9.
(1984). 18. R. Hasunuma, T. Ogawa, and Y. Kawanishi, Anal. Biochem. 126, 242 (1982). 19. C. Ip and H. E. Ganther, Carcinogenesis 9, 1481 (1988). 20. G. W. Snedecor and W. G. Cochran, Statistical Methods, Ed. 6, Iowa State University Press, Ames, Iowa, 1967, p. 258. 21. R. G. Steel and I. H. Torrie, Principles and Procedures of Statistics, Ed. 1, McGraw-Hill, New York, 1960, p. 109. 22. D. F. Morrison, Multivariate Statistical Methods, McGraw-Hill, New York, 1976. 23. J. B. Kruskal, Psychometrika 29, 115 (1964). 24. C. Ip, J. Am. Golf. Toxicol. 5, 7 (1986). 25. C. Ip, Ann. Clin. Res. 18, 22 (1986). 26. D. Medina and D. G. Morrison, Pathol. Zmmunopathol. Res. 7, 187 (1988). 27. T. G. O’Brien, Cancer Res. 36, 2644 (1976). 28. T. G. O’Brien, R. C. Simsiman, and R. K. Boutwell, Cancer Res. 35, 1662 (1975). 29. A. K. Verma, E. Erturk, and G. T. Bryan, Cancer Res. 43, 5964 (1983).
292
H. J. Thompson et al.
30. L. A. Van Rooijen. H. J. Derks. R. Van Wijk, and A. Bisschop. Curcinogenesis 31. 32. 33. 34. 35. 36.
(1984). S. Colombatto and M. A. Grille, Med. Sri. Res. 15, 987 (1087) C. Ip and H. J. Thompson. J. Nat/. Cancer Inst. 81. 839 i1989r. S. J. Foster. R. J. Kraus. and H. E. Ganther. Arch Aiochem. (1986). A. E. Pegg, Cancer Res. 48, 759 (1988). S. Colombatto, D. C. Pezzali, and M. A. Grillo. Inr J. Biochern. C. Ip and C. Hayes. C’mcinogenesis 10, 921 (1989).
Received May 7, 1991; accepted June 5. 1991
Biophys.
5, 225
251, 17
19. 725 (1987).
