9 1990 by the Humana Press, Inc. All rights of any nature, whatsoever, reserved. 0163-4984/90/2402-0147 $02.00
Selenium Concentrations in the Human Thyroid Gland JAN AASETH, 1 HARALD FREY, 2 EYSTEIN GLATTRE, *,3 GUNNAR NORHEIM,4 JETMUND RINGSTAD, 1 AND YNGVAR T H O M A S S E N 5
~Department of Occupational Medicine, N-9012 University Hospital, Tromso, Norway,. 21VledicalDepartment, Aker Hospital, Oslo; 3The Norwegian Cancer Registry, Montebello, Oslo; 4National Veterinary Institute, Oslo; 5National Institute of Occupat~bnal Health, Oslo, Norway Received July 7, 1989; Accepted September 8, 1989
ABSTRACT Recently, we found that prediagnostic serum selenium concentration was significantly lower for cases developing thyroid cancer (n = 43) than for controls. We assumed that redistribution of serum selenium into the affected tissue took place in the prediagnostic period. The present study was carried out to determine the physiological concentration of selenium in the thyroid, since very few data are available in the literature~ The concentrations of selenium in the thyroid (n = 45) and liver samples from Norwegians who had died because of acute illness or accidents were determined by hydride generation atomic absorption spectrometry. The mean selenium concentration was found to be 0.72 + 0.44 ~g/g in the thyroid and 0.45 --- 0.11 I~g/g in the liver tissue. The surprisingly high concentration of selenium in apparently normal thyroids indicates that selenium has important functions in this organ. The remarkably broad range, together with the observation that no significant correlation exists between thyroid and liver concentrations, suggest that factors other than the selenium status are important determinants for the selenium concentration in the thyroid gland. This observation is consistent with our hypothesis that in carcinogenesis, prediagnostic processes influence the serum-/thyroidratio of selenium. Index Entries: Selenium; thyroid gland; liver; carcinogenesis.
*Author to whom all correspondence and reprint requests should be addressed. Biological Trace Element Research
]. 47
Vol. 24, 1990
148
Aaseth et al.
INTRODUCTION A chemopreventive role of selenium in various malignancies is indicated by epidemiological and experimental studies (1-4). Recently, we have found that the mean prediagnostic serum concentration of selenium was significantly lower for cases developing thyroid cancer than for matched controls (5). In animals and humans only one functional seleno-protein, Sedependent glutathione peroxidase (Se-GSH-Px), has been characterized. This enzyme is involved in protecting cell constituents from peroxidative damage (6,7). A possible mechanism accounting for the chemopreventive effect of selenium against cancer is that Se-GSH-Px may reduce tissue concentration of toxic free radicals and hydroperoxides. This scavenging ability is of particular importance for the thyroid gland, since its crucial functions involve oxidation of trapped iodide (I-), which is incorporated into thyreoglobulin, the precursor of the thyroid hormones, thyroxin (T4), and triiodothyronin (Ts). The I- oxidation is brought about by a specific thyroid peroxidase, making use of physiologically produced hydrogenperoxide (H202) as a cofactor (8). It follows that the thyroid tissue must be continuously exposed to a physiological generation of H202 and probably also to a subsequent small amount of lipoperoxides (9). SeGSH-Px, which is capable of detoxifying H202 as well as lipoperoxides, is of particular importance in the thyroid gland, since it represents 100% of GSH-Px activity in this organ (9). It might be assumed that the thyroid gland should contain relatively high concentrations of selenium as an essential component of GSH-Px. The activity of this enzyme also depends on sufficiently high levels of glutathione (GSH). Remarkably high GSH levels have been demonstrated in the thyroid (10), whereas selenium thyroid concentrations are almost completely lacking in the literature (11). In the present paper, results show that the thyroid gland contains higher selenium concentrations than the liver and other human organs. MATERIALS AriD M E T H O D S Thyroid samples were collected from 45 persons aged 20-60 years, who suffered an acute death (accident or acute myocardial infarction). Before their death, all individuals included in the study were apparently healthy, using no medication. Patients with histologically detectable liver or thyroid pathology were not included in the material. Twenty-five persons had lived in Oslo, and 20 in the Tromso regions of Norway. Liver samples were also collected and used for selenium determinations. Liver tissue is considered to provide an index of whole body status of the trace element (12). Biological Trace Element Research
Vof. 24, I990
Thyroideal Selenium Levels
149
The thyroid as well as the liver samples were digested in a mixture of nitric and perchloric acid (13,14). Both selenium and mercury were determined by an automated hydride generator-atomic absorption spectroscopy system. All results are given on a wet weight basis. The quantitation limit for both elements was 0.02 p,g/g tissue. A standardized liver sample (National Bureau of Standards, US) was analyzed by the same method, and the results obtained were within 5% of the given value.
RESULTS The average concentration of selenium in the thyroid gland was 0.72 • 0.44 I~g/g (n = 45), the range being remarkably broad, viz 0.151.90 Ixg/g. The average value in the samples from Oslo, viz 0.73 • 0.36 Dg/g (n = 25), was not different from the average in Tromso, viz 0.71 • 0.54 Ixg/g (n = 20) (see Fig. 1). The average level in the liver was 0.45 • 0.11 I~g/g, the range being 0.15-0.76 t~g/g (n = 35), no differences being seen between Tromso and Oslo. No significant correlation was observed b e t w e e n the selenium concentrations in the thyroid gland and the liver (r = 0.1). Thus, the thyroid gland apparently is an exception with regard to the relatively constant pattern of organ distribution reported for selenium (13). This must implicate that whole body selenium status is not the most important factor determining the selenium m e a s u r e d in the thyroid gland. No significant differences could be found between m e n and w o m e n or between different age groups, so far as selenium values were concerned. Mercury exposure has previously been reported to increase the selenium uptake in the thyroid by codeposition of the two elements (11). However, in the present study, most of the patients had mercury levels in the gland lower than 0.02 I~g/g. The average liver level of mercury was 0.08 • 0.05 p,g/g. No significant correlations were found between selenium and mercury in these organs. The determination of other essential elements (sodium, potassium, magnesium, calcium, zinc, and copper) by inductively coupled plasma atomic emission spectrometry (15), indicated that the concentrations of these elements in the thyroid gland were much nearer to the extracellular fluid concentrations than those found in the liver. This might be explained from the "extracellular" composition of the follicular fluid of the gland. Our present observation of strikingly high selenium concentrations, and previous reports of high glutathione concentrations (10), are in contrast to such similarities with the extracellular mineral composition.
DISCUSSION The present results are a s s u m e d to be representative of the population of the southeastern and the northern parts of Norway with regard to Biological Trace Element Research
VoL 24, 1990
150
Aaseth et aL
Fig. 1. Selenium concentrations (~g/g) in thyroid glands. The bars give the number of thyroids (n) within various concentration intervals. selenium status. They are based on homogeneous autopsy material, which included no cases of thyroid or liver pathology. The selenium concentrations in the thyroid gland (mean 0.72 ~g/g wt) are high, compared to the liver concentrations. Nevertheless, our results are consistent with the very few values for thyroid selenium levels reported previously (11). Data concerning selenium concentrations in h u m a n liver are also relatively sparse. Our result (mean 0.45 tzg/g) is in relatively good agreement with most previous investigations (16-19). In animal studies there is very. good relationship between the selenium intake and concentrations found in the liver (20). A similar relationship is suggested to exist in humans (11). Serum concentrations are also used as an indicator of daily intake of selenium (21). Reference values for serum selenium levels in Norway, viz around 0.12 tJ,g/g, are higher than those in Sweden, but lower than North American levels (1t,2I). The lack of correlation between liver and thyroid selenium found in the present study, indicates that the whole body selenium status is not the main determinant of the concentration in the thyroid. This might be a result of increased prepathological selenium uptake in the thyroid gland or proteins with crucial functions in the thyroid gland that retains its selenium very tightly (see 22). Thus, in rats kept on a selenium-deficient diet, regulatory mechanisms strive to maintain the selenium in the thyroid (23). The protective role against thyroid cancer of raised tissue selenium previously suggested (5) may be linked to the retention of the element by thyroidal proteins. Biological Trace Element Research
Vol. 24, 1990
Thyroideal Selenium Levels
151
A protective action of selenium against heavy metal toxicity has been suggested previously (24). In the thyroid gland of rats, a specific protein with apparent molecular wt of 27,800 has been found (23). The amounts of selenium attached to this protein were higher than those attached to GSH-Px. High amounts of selenium linked to another protein with apparent molecular wt of 15,600 was also found in thyroid, liver, and other organs in rats. The functions of these " n e w " selenoproteins are unknown. It is of particular interest, however, that recent studies (25) have reported that the dehalogenase responsible for the T4-TB-Conversion in rat liver appears to be selenium-dependent. An attractive hypothesis is that selenium plays a role for thyroidaI dehalogenases necessary to detoxify atypically iodinated species resembling T 4 o r T 3. An interesting observation is that carcinogenic c o m p o u n d s capable of inducing thyroidal cancer in animal studies are aromatic- or polycyclic hydrocarbons that can be substrates for iodination (26). It might be speculated if the resulting iodo-compounds are particularly active carcinogenic promotors since halogenated hydrocarbons in general possess higher reactivity than the parent compounds. If so, the dehalogenases in the thyroid may serve as a chemopreventive action against cancer, and may even be a general mechanism of the anticarcinogenic effect of selenium. The possibility that the optimum selenium concentrations in the thyroid (and other tissues) are higher than those found in the present study cannot be excluded. The low liver-to-thyroid ratio of selenium concentrations seen in our m a t e r i a l ~ a v e r a g e 0.6--is never seen in healthy rats, only in selenium-deficient animals (23).
ACKNOWI EDGMENTS The authors thank the Norwegian Cancer Society for financial support.
REFERENCES 1. J. T. Salonen, G. Alfthan, J. K. Huttunen, and P. Puska, Amer. J. Epidemiol. 120, 342 (1984). 2. W. C. Willett and M. J. Stampfer, Acta Pharmacol. Toxicol. 59 (suppl. 7), 240 (1986). 3. G. N. Schrauzer, D. A. White, and C. J. Schneider, Bioinorg. Chem. 7, 23 (1977). 4. C. Ip, Ann. Clin. Res. 18, 22 (1986). 5. E. Glattre, Y. Thomassen, S. O. Thoresen, T. Haldorsen, P. G. LundLarsen, L. Theodorsen, and J. Aaseth. Int. J. Epidemiology 18, 45 (1989). 6. L. Flohe and I. Brand, Biochem. Biophys. Acta 191, 541 (1969). 7. National Research Council: Subcommittee on Selenium, Selenium in Nutrition, National Academy Press, Washington, DC, 1983. 8. L. J. De Groot and H. Niepomniszcze, Metabolism 26, 665 (1977). Biological Trace Element Research
VoL 24, I990
152
A a s e t h et aL
9. F. Carmignol, P. M. Sinet, and H. Jerome, Biochim. Biophys. Acta 759, 49 (1983). 10. P. C. Jocelyn, Biochemistry of the SH Group, Academic Press, London, New York, 1972, p. 269. 11. Y. Thomassen and J. Aaseth, Selenium, M. Ihnat, ed. CRC Press Inc., Boca Raton, FL, 1989. 12. E. J. Underwood, Trace Elements in Human and Animal Nutrition, Academic Press, New York, 1977, p. 545. 13. A. Haugen, Role, and G. Norheim, Abstracts, 10th Nordic Atomic Absorption Spectroscopy and Trace Element Conference, Turku, Finland Aug. 6-9, 1985, p. 50. 14. G. Norheim and R. lqaugen, Acta Pharmacol. ToxicoI. 59 (suppl. 7), 610 (1986). 15. Y. Thomassen, J. Aaseth, G. Norheim, and J. Ringstad, J. Trace Elem. Electrolytes Health Dis. (1989), submitted. 16. G. Norheim and J. Aaseth, J. Oslo City Hosp. 30, 105 (1980). 17. R. C. Dickson and R. H. Tomlinson, Clin. Chem. Acta 16, 311 (1967). 18. C. A. Johnson and j. F. Lewin, Anal. Chim. Acta 82, 79 (1976). 19. H. A. Schroeder, D. V. Frost, and J. J. Balassa, ]. Chron. Dis. 19, 1007 (1970). 20. G. f~vernes, K. Moksnes, E. Mo fbkland, and Arne Fr~slie, Acta Agric. Scand. 36, 1 (1986). 21. J. Aaseth and Y. Thomassen, Selenium in Medicine and Biology, Proceedings of the Second International Congress in Trace Elements in Medicine and Biology, France, N. N6ve, and A. Favier, eds., W. de Gruyter, Berlin, New York, 1988. 22. P. Goyens, J. Goldstein, B. Nsombola, H. Vis, and J. E. Dumont, Acta Endocrinologica 114, 497 (1987). 23. D. Behne, H. Hilmert, S. Scheid, H. Gessner, and W. Elger, Biochem. Biophys. Acta 966, 12 (1988). 24. L. Kosta, V. Zelenko, V. Ravnik, et al., IAEA-SM-175/27, Symposium on nuclear techniques in comparative studies on food and environmental contamination, International Atomic Energy Agency, Vienna, Austria, p. 541. 25. G. J. Becket, S. E. Beddows, P. C. Mortice, F. Nicol, and J R. Arthur, Biochem. I. 248, 443 (1987). 26. WHO, IARC Monographs on the evaluation of the carcinogenic risk of chemicals to humans, vol. 27, International Agency for Research on Cancer, Lyon, France 1982, pp. 63--154.
Biological Trace Element Research
VoL 24, 1990
Selenium Concentrations in the Human Thyroid Gland JAN AASETH, 1 HARALD FREY, 2 EYSTEIN GLATTRE, *,3 GUNNAR NORHEIM,4 JETMUND RINGSTAD, 1 AND YNGVAR T H O M A S S E N 5
~Department of Occupational Medicine, N-9012 University Hospital, Tromso, Norway,. 21VledicalDepartment, Aker Hospital, Oslo; 3The Norwegian Cancer Registry, Montebello, Oslo; 4National Veterinary Institute, Oslo; 5National Institute of Occupat~bnal Health, Oslo, Norway Received July 7, 1989; Accepted September 8, 1989
ABSTRACT Recently, we found that prediagnostic serum selenium concentration was significantly lower for cases developing thyroid cancer (n = 43) than for controls. We assumed that redistribution of serum selenium into the affected tissue took place in the prediagnostic period. The present study was carried out to determine the physiological concentration of selenium in the thyroid, since very few data are available in the literature~ The concentrations of selenium in the thyroid (n = 45) and liver samples from Norwegians who had died because of acute illness or accidents were determined by hydride generation atomic absorption spectrometry. The mean selenium concentration was found to be 0.72 + 0.44 ~g/g in the thyroid and 0.45 --- 0.11 I~g/g in the liver tissue. The surprisingly high concentration of selenium in apparently normal thyroids indicates that selenium has important functions in this organ. The remarkably broad range, together with the observation that no significant correlation exists between thyroid and liver concentrations, suggest that factors other than the selenium status are important determinants for the selenium concentration in the thyroid gland. This observation is consistent with our hypothesis that in carcinogenesis, prediagnostic processes influence the serum-/thyroidratio of selenium. Index Entries: Selenium; thyroid gland; liver; carcinogenesis.
*Author to whom all correspondence and reprint requests should be addressed. Biological Trace Element Research
]. 47
Vol. 24, 1990
148
Aaseth et al.
INTRODUCTION A chemopreventive role of selenium in various malignancies is indicated by epidemiological and experimental studies (1-4). Recently, we have found that the mean prediagnostic serum concentration of selenium was significantly lower for cases developing thyroid cancer than for matched controls (5). In animals and humans only one functional seleno-protein, Sedependent glutathione peroxidase (Se-GSH-Px), has been characterized. This enzyme is involved in protecting cell constituents from peroxidative damage (6,7). A possible mechanism accounting for the chemopreventive effect of selenium against cancer is that Se-GSH-Px may reduce tissue concentration of toxic free radicals and hydroperoxides. This scavenging ability is of particular importance for the thyroid gland, since its crucial functions involve oxidation of trapped iodide (I-), which is incorporated into thyreoglobulin, the precursor of the thyroid hormones, thyroxin (T4), and triiodothyronin (Ts). The I- oxidation is brought about by a specific thyroid peroxidase, making use of physiologically produced hydrogenperoxide (H202) as a cofactor (8). It follows that the thyroid tissue must be continuously exposed to a physiological generation of H202 and probably also to a subsequent small amount of lipoperoxides (9). SeGSH-Px, which is capable of detoxifying H202 as well as lipoperoxides, is of particular importance in the thyroid gland, since it represents 100% of GSH-Px activity in this organ (9). It might be assumed that the thyroid gland should contain relatively high concentrations of selenium as an essential component of GSH-Px. The activity of this enzyme also depends on sufficiently high levels of glutathione (GSH). Remarkably high GSH levels have been demonstrated in the thyroid (10), whereas selenium thyroid concentrations are almost completely lacking in the literature (11). In the present paper, results show that the thyroid gland contains higher selenium concentrations than the liver and other human organs. MATERIALS AriD M E T H O D S Thyroid samples were collected from 45 persons aged 20-60 years, who suffered an acute death (accident or acute myocardial infarction). Before their death, all individuals included in the study were apparently healthy, using no medication. Patients with histologically detectable liver or thyroid pathology were not included in the material. Twenty-five persons had lived in Oslo, and 20 in the Tromso regions of Norway. Liver samples were also collected and used for selenium determinations. Liver tissue is considered to provide an index of whole body status of the trace element (12). Biological Trace Element Research
Vof. 24, I990
Thyroideal Selenium Levels
149
The thyroid as well as the liver samples were digested in a mixture of nitric and perchloric acid (13,14). Both selenium and mercury were determined by an automated hydride generator-atomic absorption spectroscopy system. All results are given on a wet weight basis. The quantitation limit for both elements was 0.02 p,g/g tissue. A standardized liver sample (National Bureau of Standards, US) was analyzed by the same method, and the results obtained were within 5% of the given value.
RESULTS The average concentration of selenium in the thyroid gland was 0.72 • 0.44 I~g/g (n = 45), the range being remarkably broad, viz 0.151.90 Ixg/g. The average value in the samples from Oslo, viz 0.73 • 0.36 Dg/g (n = 25), was not different from the average in Tromso, viz 0.71 • 0.54 Ixg/g (n = 20) (see Fig. 1). The average level in the liver was 0.45 • 0.11 I~g/g, the range being 0.15-0.76 t~g/g (n = 35), no differences being seen between Tromso and Oslo. No significant correlation was observed b e t w e e n the selenium concentrations in the thyroid gland and the liver (r = 0.1). Thus, the thyroid gland apparently is an exception with regard to the relatively constant pattern of organ distribution reported for selenium (13). This must implicate that whole body selenium status is not the most important factor determining the selenium m e a s u r e d in the thyroid gland. No significant differences could be found between m e n and w o m e n or between different age groups, so far as selenium values were concerned. Mercury exposure has previously been reported to increase the selenium uptake in the thyroid by codeposition of the two elements (11). However, in the present study, most of the patients had mercury levels in the gland lower than 0.02 I~g/g. The average liver level of mercury was 0.08 • 0.05 p,g/g. No significant correlations were found between selenium and mercury in these organs. The determination of other essential elements (sodium, potassium, magnesium, calcium, zinc, and copper) by inductively coupled plasma atomic emission spectrometry (15), indicated that the concentrations of these elements in the thyroid gland were much nearer to the extracellular fluid concentrations than those found in the liver. This might be explained from the "extracellular" composition of the follicular fluid of the gland. Our present observation of strikingly high selenium concentrations, and previous reports of high glutathione concentrations (10), are in contrast to such similarities with the extracellular mineral composition.
DISCUSSION The present results are a s s u m e d to be representative of the population of the southeastern and the northern parts of Norway with regard to Biological Trace Element Research
VoL 24, 1990
150
Aaseth et aL
Fig. 1. Selenium concentrations (~g/g) in thyroid glands. The bars give the number of thyroids (n) within various concentration intervals. selenium status. They are based on homogeneous autopsy material, which included no cases of thyroid or liver pathology. The selenium concentrations in the thyroid gland (mean 0.72 ~g/g wt) are high, compared to the liver concentrations. Nevertheless, our results are consistent with the very few values for thyroid selenium levels reported previously (11). Data concerning selenium concentrations in h u m a n liver are also relatively sparse. Our result (mean 0.45 tzg/g) is in relatively good agreement with most previous investigations (16-19). In animal studies there is very. good relationship between the selenium intake and concentrations found in the liver (20). A similar relationship is suggested to exist in humans (11). Serum concentrations are also used as an indicator of daily intake of selenium (21). Reference values for serum selenium levels in Norway, viz around 0.12 tJ,g/g, are higher than those in Sweden, but lower than North American levels (1t,2I). The lack of correlation between liver and thyroid selenium found in the present study, indicates that the whole body selenium status is not the main determinant of the concentration in the thyroid. This might be a result of increased prepathological selenium uptake in the thyroid gland or proteins with crucial functions in the thyroid gland that retains its selenium very tightly (see 22). Thus, in rats kept on a selenium-deficient diet, regulatory mechanisms strive to maintain the selenium in the thyroid (23). The protective role against thyroid cancer of raised tissue selenium previously suggested (5) may be linked to the retention of the element by thyroidal proteins. Biological Trace Element Research
Vol. 24, 1990
Thyroideal Selenium Levels
151
A protective action of selenium against heavy metal toxicity has been suggested previously (24). In the thyroid gland of rats, a specific protein with apparent molecular wt of 27,800 has been found (23). The amounts of selenium attached to this protein were higher than those attached to GSH-Px. High amounts of selenium linked to another protein with apparent molecular wt of 15,600 was also found in thyroid, liver, and other organs in rats. The functions of these " n e w " selenoproteins are unknown. It is of particular interest, however, that recent studies (25) have reported that the dehalogenase responsible for the T4-TB-Conversion in rat liver appears to be selenium-dependent. An attractive hypothesis is that selenium plays a role for thyroidaI dehalogenases necessary to detoxify atypically iodinated species resembling T 4 o r T 3. An interesting observation is that carcinogenic c o m p o u n d s capable of inducing thyroidal cancer in animal studies are aromatic- or polycyclic hydrocarbons that can be substrates for iodination (26). It might be speculated if the resulting iodo-compounds are particularly active carcinogenic promotors since halogenated hydrocarbons in general possess higher reactivity than the parent compounds. If so, the dehalogenases in the thyroid may serve as a chemopreventive action against cancer, and may even be a general mechanism of the anticarcinogenic effect of selenium. The possibility that the optimum selenium concentrations in the thyroid (and other tissues) are higher than those found in the present study cannot be excluded. The low liver-to-thyroid ratio of selenium concentrations seen in our m a t e r i a l ~ a v e r a g e 0.6--is never seen in healthy rats, only in selenium-deficient animals (23).
ACKNOWI EDGMENTS The authors thank the Norwegian Cancer Society for financial support.
REFERENCES 1. J. T. Salonen, G. Alfthan, J. K. Huttunen, and P. Puska, Amer. J. Epidemiol. 120, 342 (1984). 2. W. C. Willett and M. J. Stampfer, Acta Pharmacol. Toxicol. 59 (suppl. 7), 240 (1986). 3. G. N. Schrauzer, D. A. White, and C. J. Schneider, Bioinorg. Chem. 7, 23 (1977). 4. C. Ip, Ann. Clin. Res. 18, 22 (1986). 5. E. Glattre, Y. Thomassen, S. O. Thoresen, T. Haldorsen, P. G. LundLarsen, L. Theodorsen, and J. Aaseth. Int. J. Epidemiology 18, 45 (1989). 6. L. Flohe and I. Brand, Biochem. Biophys. Acta 191, 541 (1969). 7. National Research Council: Subcommittee on Selenium, Selenium in Nutrition, National Academy Press, Washington, DC, 1983. 8. L. J. De Groot and H. Niepomniszcze, Metabolism 26, 665 (1977). Biological Trace Element Research
VoL 24, I990
152
A a s e t h et aL
9. F. Carmignol, P. M. Sinet, and H. Jerome, Biochim. Biophys. Acta 759, 49 (1983). 10. P. C. Jocelyn, Biochemistry of the SH Group, Academic Press, London, New York, 1972, p. 269. 11. Y. Thomassen and J. Aaseth, Selenium, M. Ihnat, ed. CRC Press Inc., Boca Raton, FL, 1989. 12. E. J. Underwood, Trace Elements in Human and Animal Nutrition, Academic Press, New York, 1977, p. 545. 13. A. Haugen, Role, and G. Norheim, Abstracts, 10th Nordic Atomic Absorption Spectroscopy and Trace Element Conference, Turku, Finland Aug. 6-9, 1985, p. 50. 14. G. Norheim and R. lqaugen, Acta Pharmacol. ToxicoI. 59 (suppl. 7), 610 (1986). 15. Y. Thomassen, J. Aaseth, G. Norheim, and J. Ringstad, J. Trace Elem. Electrolytes Health Dis. (1989), submitted. 16. G. Norheim and J. Aaseth, J. Oslo City Hosp. 30, 105 (1980). 17. R. C. Dickson and R. H. Tomlinson, Clin. Chem. Acta 16, 311 (1967). 18. C. A. Johnson and j. F. Lewin, Anal. Chim. Acta 82, 79 (1976). 19. H. A. Schroeder, D. V. Frost, and J. J. Balassa, ]. Chron. Dis. 19, 1007 (1970). 20. G. f~vernes, K. Moksnes, E. Mo fbkland, and Arne Fr~slie, Acta Agric. Scand. 36, 1 (1986). 21. J. Aaseth and Y. Thomassen, Selenium in Medicine and Biology, Proceedings of the Second International Congress in Trace Elements in Medicine and Biology, France, N. N6ve, and A. Favier, eds., W. de Gruyter, Berlin, New York, 1988. 22. P. Goyens, J. Goldstein, B. Nsombola, H. Vis, and J. E. Dumont, Acta Endocrinologica 114, 497 (1987). 23. D. Behne, H. Hilmert, S. Scheid, H. Gessner, and W. Elger, Biochem. Biophys. Acta 966, 12 (1988). 24. L. Kosta, V. Zelenko, V. Ravnik, et al., IAEA-SM-175/27, Symposium on nuclear techniques in comparative studies on food and environmental contamination, International Atomic Energy Agency, Vienna, Austria, p. 541. 25. G. J. Becket, S. E. Beddows, P. C. Mortice, F. Nicol, and J R. Arthur, Biochem. I. 248, 443 (1987). 26. WHO, IARC Monographs on the evaluation of the carcinogenic risk of chemicals to humans, vol. 27, International Agency for Research on Cancer, Lyon, France 1982, pp. 63--154.
Biological Trace Element Research
VoL 24, 1990
