Orville A. Levander
U.S. Department of Agriculture, ARS
Beltsville Human Nutrition Research Center
Nutrient Requirements and Functions Laboratory
Beltsville, MD 20705
In 1957, Klaus Schwarz started the ultra trace element nutrition era with the discovery that trace amounts of selenium in the feed would protect vitamin E-deficient rats from dietary liver necrosis (Schwarz and Foltz, 1957). Soon thereafter, selenium deficiency was found to be involved in the etiology of several economically-important nutritional diseases of farm animals, and in the early 1960's, the usefulness of selenium supplements in counteracting these conditions became apparent (NRC, 1983). In 1973, the biochemical significance of selenium was established by its role in the peroxide-destroying enzyme glutathione peroxidase (Rotruck et al., 1973); more recent research has revealed other roles for selenium in thyroid hormone metabolism (Arthur et al., 1997; St. Germain and Galton, 1997) and in the reduction of thioredoxin (Gladyshev et al., 1998; Berggren et al., 1999).
Conclusive evidence of a role for selenium in human nutrition, however, did not materialize until 1979 when scientists in the People's Republic of China published their results in English showing the protective effect of selenium against a juvenile cardiomyopathy known as Keshan disease (Keshan Disease Research Group, 1979).
The first dietary standard for selenium in human nutrition in the U.S.A. was the so-called Estimated Safe and Adequate Daily Dietary Intake (ESADDI), which was published by the Food and Nutrition Board (FNB) of the National Academy of Sciences (FNB, 1980). Because of the time needed for the extensive internal and external review process of FNB reports, the 1980 edition of the Recommended Dietary Allowances (RDAs) did not contain any reference to Keshan disease even though the initial paper on this topic in English appeared in the open literature the year before.
As a result, the ESADDI was based on extrapolation of animal data to humans. Numerous studies carried out with both laboratory and farm animals suggested that the dietary level needed to prevent deficiency disease was similar across a wide range of species, and a consensus requirement of .1 to .2 ppm emerged. If one assumed that adults consume a half kilo of food daily on a dry weight basis, then a nutritional requirement of 50 µg per day could be calculated (500 g x 0.1 µg/g). This figure, in fact, became the lower limit of the 1980 ESADDI for adults (Table 1). Because of the lack of quantitative data on human requirements and the absence of any information about the nutritional bioavailability of selenium from different human diets, the ESADDI was given in the form of a range, 50 to 200 µg/day for adults, to reflect this uncertainty. The upper limit of the range was chosen to take into account any possible variation in human selenium requirements and also to prevent abuse of selenium supplements. The ESADDI for younger age groups was extrapolated from adult values. No specific recommendations were made for lactating or pregnant women.
| Table 1. Estimated safe and adequate daily dietary selenium intake. | ||
|---|---|---|
| Group | Age (years) | ESADDI (µg) |
| Infants | 0.0 to 0.5 | 10 to 40 |
| 0.5 to 1.0 | 20 to 60 | |
| Children | 1.0 to 3.0 | 20 to 80 |
| 4.0 to 6.0 | 30 to 120 | |
| 7.0 to 10.0 | 50 to 200 | |
| 11.0+ | 50 to 200 | |
| Adults | 50 to 200 | |
|
Source: FNB (1980). |
||
Once the linkage between Keshan disease and selenium had been found, interest in the role of selenium in human nutrition increased rapidly. As of February 10, 1999, PubMed, the National Library of Medicine's on-line research service to Medline, listed 1,003 "hits" (publications) indexed under the descriptor "selenium in human nutrition" for the period 1966 to 1998 (more 1998 "hits" may eventually become available as they are catalogued into the system). The period 1966 up to the release of the ESADDI in 1980 accounted for approximately 10% of the total "hits." The publication rate for papers about "selenium in human nutrition" started out slowly at about two per year during the late 1960's, but by the late 1970's, about 14 papers were being published annually. However, from 1980 to 1989, the publication rate averaged 38 per year over the decade of the eighties. All this new information suggested to trace element nutritionists that it might be time to start considering selenium as a candidate for the Recommended Dietary Allowances (RDAs), and an RDA for selenium was proposed in 1989 (FNB, 1989).
For establishing an RDA for selenium, the RDA committee initially considered one of the traditional methods for estimating mineral requirements, the metabolic balance study (Mertz, 1987). In this approach, the amount ingested is compared to the amount excreted. At the level at which intake equals excretion, the subject is said to be "in balance" and that intake is considered to meet the requirement of the individual. However, in the case of selenium, the body adapts so well to fluctuations in intake that a zero balance point can be obtained over a wide range of intakes (Levander and Morris, 1985). Thus, the metabolic balance technique was deemed not helpful in estimating selenium requirements.
A better approach was based on the dietary selenium intakes observed in the presence and absence of selenium deficiency (i.e., in different areas of China with or without Keshan disease). After carefully mapping the incidence of Keshan disease, one could then compare dietary selenium intakes in those areas with and without the disease. In such a way, it was possible to estimate a kind of minimum dietary requirement for selenium. The Chinese scientists found that Keshan disease was prevented in those areas in which the intake of selenium by adult males was at least 19.1 µg per day (Yang et al., 1988).
Another time-honored procedure for estimating requirements for nutrients is to measure an enzyme that is dependent for its activity on the nutrient in question. By comparing the enzyme activity vs. the dietary intake of the nutrient, one can determine the point at which the enzymatic activity plateaus off. The inflection point at which the enzymatic activity is maximized is taken as the level of dietary intake of the nutrient that satisfies the requirement.
Chinese scientists applied the enzyme maximization approach to adult males who were long-term residents of a Keshan disease area (Yang et al., 1987). Because of their prolonged residence in a selenium-poor region, all the subjects were of extremely low initial selenium status. The men were fed graded doses of selenomethionine. After several months, plasma glutathione peroxidase activity maximized in those persons whose total selenium intake was 40 µg per day (30 µg from supplement and 10 µg from the native diet) or more. Calculation of the RDA for North Americans based on this study required the use of a body weight adjustment factor and a safety factor (Levander, 1991). This calculation yielded a figure of 70 and 55 µg/day for the typical North American adult male and female, respectively (Table 2). RDAs for other age and sex groups were extrapolated on the basis of body weight and provision for growth. Incremental requirements for females during pregnancy and lactation were established by taking into account fetal accretion and lacteal losses of selenium, respectively.
| Table 2. Recommended dietary allowances for selenium. | ||||||
|---|---|---|---|---|---|---|
| Category | Age (years) or Condition | Weight | Height | Selenium |
||
| (kg) | (lb) | (cm) | (in) | (µg) | ||
| Infants | .5 to .5 | 6 | 13 | 60 | 24 | 10 |
| .5 to 1.0 | 9 | 20 | 71 | 28 | 15 | |
| Children | 1.0 to 3.0 | 13 | 29 | 90 | 35 | 20 |
| 4.0 to 6.0 | 20 | 44 | 112 | 44 | 20 | |
| 7.0 to 10.0 | 28 | 62 | 132 | 52 | 30 | |
| Males | 11.0 to 14.0 | 45 | 99 | 157 | 62 | 40 |
| 15.0 to 18.0 | 66 | 145 | 176 | 69 | 50 | |
| 19.0 to 24.0 | 72 | 160 | 177 | 70 | 70 | |
| 25.0 to 50.0 | 79 | 174 | 176 | 70 | 70 | |
| 51.0 + | 77 | 170 | 173 | 68 | 70 | |
| Females | 11.0 to 14.0 | 46 | 101 | 157 | 62 | 45 |
| 15.0 to 18.0 | 55 | 120 | 163 | 64 | 50 | |
| 19.0 to 24.0 | 58 | 128 | 164 | 65 | 55 | |
| 25.0 to 50.0 | 63 | 138 | 163 | 64 | 55 | |
| 51.0 + | 65 | 143 | 160 | 63 | 55 | |
| Pregnant | 65 | |||||
| Lactating | 1st 6 months | 75 | ||||
| 2nd 6 months | 75 | |||||
| Source: FNB (1989). | ||||||
Suggestions on how possibly to improve the next version of the selenium RDA have been offered (Combs, 1994) and, as this review is being written, the FNB Panel on Antioxidants and Related Nutrients is deliberating the latest dietary standard for selenium as part of the newly established RDI process (see below).
In 1990, the Food and Agricultural Organization of the United Nations (FAO), the International Atomic Energy Agency (IAEA) and the World Health Organization (WHO) convened an Expert Consultation on Trace Elements in Human Nutrition to update the 1973 report issued by the WHO Expert Committee on Trace Elements in Human Nutrition (WHO, 1973). The 1990 Consultation took a two-tiered approach to requirements (FAO/IAEA/WHO, 1996). The basal requirement was defined as "the intake needed to prevent pathologically relevant and clinically detectable signs of impaired function attributable to inadequacy of the nutrient," whereas the normative requirement was defined as the "intake that serves to maintain a level of tissue storage or other reserve that is judged by the Expert Consultation to be desirable."
The 1990 Consultation also utilized the term "safe range of population mean intakes", which refers to "the range of population group mean intakes deemed adequate to sustain existing nutriture in healthy population groups." It was felt that defining the nutritional standard in terms of population mean intakes would be more useful to public health nutritionists for both planning and assessment purposes. To calculate the population minimum mean selenium intake likely to meet basal requirements, the Consultation relied on analytical data of Chinese diets collected in areas with and without Keshan disease. Since WHO weight standards differ somewhat from the weights in the Chinese study, a correction had to be applied, and the final value was 19.1 x 65/60 = 21 µg/day for adult males.
The 1990 Consultation used plasma glutathione peroxidase activity as the basis for its normative selenium requirement just as the FNB committee had earlier used that approach for the basis of its RDA for selenium. However, unlike the RDA committee, the Consultation did not feel that full maximization of plasma glutathione peroxidase activity was needed for health. Rather, the Consultation chose 67% of the maximum attainable activity as a reasonable goal based on the fact that abnormalities in the ability of red blood cells to metabolize hydrogen peroxide only became apparent when the activity of the enzyme declined to 25% or less of normal. The normative requirement for the standard 65-kg reference man was calculated on this basis to be 26 µg/day. From this figure, it was possible to derive the lower limit of the safe range of population mean selenium intake to meet the normative requirement of most adult males of approximately 40 µg/day, assuming an interindividual variability of dietary selenium intake of 16% (26/168). Values for other sex and age groups were extrapolated on the basis of the basal metabolic rate and a full listing of the lower limit of the safe ranges of population mean intakes of dietary selenium to meet basal and normative requirements is presented in Table 3.
| Table 3. Lower limits of the safe range of population mean intakes of dietary selenium (in µg/day). | ||||
|---|---|---|---|---|
|
Age range (years) of duration of pregnancy or lactation |
Sex |
Weight (kg) |
Basal SePImin |
Normative SePImin |
| 0.00 to .25 | M&F | 5 | 3 | 6 |
| .25 to .50 | M&F | 7 | 5 | 9 |
| .50 to 1.00 | M&F | 9 | 6 | 12 |
| 1.00 to 3.00 | M&F | 12 | 10 | 20 |
| 3.00 to 6.00 | M&F | 17 | 12 | 24 |
| 6.00 to 10.00 | M&F | 25 | 14 | 25 |
| 10.00 to 12.00 | F | 37 | 16 | 30 |
| 12.00 to 15.00 | F | 48 | 16 | 30 |
| 15.00 to 18.00 | F | 55 | 16 | 30 |
| 18.00 to <60.00 | F | 55 | 16 | 30 |
| 10.00 to 12.00 | M | 35 | 16 | 30 |
| 12.00 to 15.00 | M | 48 | 19 | 36 |
| 15.00 to 18.00 | M | 64 | 21 | 40 |
| 18.00 to <60.00 | M | 65 | 21 | 40 |
| Pregnancy | ||||
| First trimester | 18 | 39 | ||
| Second trimester | 18 | 39 | ||
| Third trimester | 18 | 39 | ||
| Lactation | ||||
| 0 to 3 months | 21 | 42 | ||
| 3 to 6 months | 25 | 46 | ||
| 6 to 12 months | 26 | 52 | ||
| Source: FAO/IAEA/WHO (1996). | ||||
The Reference Daily Intake suggested by the Nordic countries is given as a range, 30 to 60 µg/d for adult males and females (Standing Nordic Committee on Food, 1989). The German Estimated Value for Adequate Supply is also stated as a range, though somewhat broader, 20 to 100 µg/d for adults (Committee on Nutrient Requirements, 1991). The British Reference Nutrient Intake distinguishes between gender and is 75 and 60 µg/d for adult males and females, respectively (Department of Health, 1991). The Australians also distinguish between the sexes and suggest 85 and 70 µg/d for men and women, respectively (Truswell et al., 1990). Thus, there is reasonable agreement among international nutrition authorities regarding the need for selenium, but there still remain some differences in dietary recommendations.
Committees of nutritionists convened to establish dietary standards generally are reluctant to consider toxicological aspects of nutrients, probably because of a lack of experience and/or expertise. The toxicity of essential trace elements, however, has begun to attract some attention, if for no reason other than the fact that, in some cases, the nutritional and toxicological standards may overlap (Mertz, 1998).
In the case of selenium, the upper limit of the ESADDI has sometimes been used as a quasi-toxicological standard, although it was never meant to be interpreted that way. Excessive selenium intakes and toxicity were discussed in the text of the 1989 RDAs, but no attempt was made to set any safe upper limit of selenium intake. The FAO/IAEA/WHO Consultation committed itself to a maximal daily safe dietary selenium intake of 400 µg, but a fair amount of arbitrariness was admittedly built into its derivation. Perhaps the best standard for overexposure at the present time is the so-called Reference Dose (RfD), although the RDI panel on Antioxidants will consider an upper level for selenium (see below).
The RfD is a toxicological standard designed to protect individuals from excessive environmental exposure to various substances. The RfD is defined as an "estimate (with uncertainty spanning perhaps an order of magnitude) of a daily exposure to the human population (including sensitive subgroups) that is likely to be without an appreciable risk of deleterious effects during a lifetime" (Dourson, 1994). The RfD for selenium established by the U.S. Environmental Protection Agency was based on studies of human selenosis in China (Poirier, 1994). The Chinese human selenium toxicity studies found a no-observed-adverse-effect level (NOAEL) of 850 µg/day for selenium or 15 µg Se/kg/day for their 55 kg subjects. Then the RfD was calculated as follows:
RfD = 15 µg Se/kg/day = 5 µg Se/kg day 3
In this equation, 15 µg Se/kg/day is the NOAEL from a chronic human toxicity study and 3 is the uncertainty factor (UF) to account for sensitive individuals. Thus, for the 70 kg reference male, the RfD would be 350 µg/day, substantially higher than the RDA of 70 µg/day for the same age and sex group.
During the past few years the FNB has embarked on an ambitious project involving a new approach to establishing dietary standards (IOM, 1998a). The new standards are called the Dietary Reference Intakes (DRI) and consist of four reference values: the Recommended Dietary Allowance (RDA), the Estimated Average Requirement (EAR), the Adequate Intake (AI), and the Tolerable Upper Intake Level (UL). The RDA is defined as "the average daily dietary intake level that is sufficient to meet the nutrient requirements of nearly all (97 to 98 percent) healthy individuals in a particular life-stage and gender group." Thus, the definition resembles that put forward in earlier editions of the RDA.
The EAR is defined as "the daily intake value that is estimated to meet the requirements, as defined by the specified indicator of adequacy, in half of the individuals in a life-stage or gender group." The EAR is used to set the RDA according to the following relationship:
RDA = EAR + 2 SDEAR
If insufficient data exist to determine an EAR (and hence an RDA), then an Adequate Intake (AI) is set. Therefore, there is greater uncertainty regarding an AI than an RDA.
The UL brings the dietary standard setting process into the realm of toxicology and is defined as "the highest level of nutrient intake that is likely to pose no risks of adverse health effects to almost all individuals in the general population." A risk assessment model has been developed for the purpose of establishing UL for nutrients, which consists of hazard identification followed by a dose-response assessment (IOM, 1998b).
The Panel on Dietary Antioxidants and Related Compounds, of which selenium is a part, is already well into its deliberations. One of the first acts of the panel was to define a dietary antioxidant: "a substance in foods that significantly decreases the adverse effects of reactive oxygen species, reactive nitrogen species or both on normal physiological functions in humans" (IOM, 1998c). Further communications from the Panel are awaited with interest.
From this brief review, the reader can easily discern that each group of experts that has undertaken the task to establish dietary standards for selenium has taken a somewhat different approach to the problem. This can, in part, be explained by the different scientific facts available to different committees as progress was made in our understanding and appreciation of selenium's role in human nutrition. However, the differences in approach also probably reflect differences in attitude and background of different committee or panel members.
In the case of selenium, at least two areas of research in particular need additional attention. First of all, the promising effect of selenium as a cancer chemopreventive agent in humans needs further investigation. The impressive data of Clark et al. (1996) need corroboration and clarification. Second, better indicators of selenium overexposure are needed, especially if increased intakes of the element are to be recommended for cancer prevention. On a basic biochemical level, there are also many exciting new leads for research developing with the role of selenium in thioredoxin reductase. If the next forty years of selenium research are as rewarding as the past forty, the new generation of selenium scientists should have very fulfilling careers indeed.
Arthur, J.R., F. Nicol, J.H. Mitchell and G.J. Beckett. 1997. Selenium and iodine deficiencies and selenoprotein function. Biomed. Environ. Sci. 10:129.
Berggren, M.M., J.F. Mangin, J.R. Gasdaka and G. Powis. 1999. Effect of selenium on rat thioredoxin reductase activity: increase by supranutritional selenium and decrease by selenium deficiency. Biochem. Pharmacol. 57:187.
Clark, L.C., G.F. Combs, Jr., B.W. Turnbull, E.H. Slate, D.K. Chalker, J. Chow, L.S. Davis, R.A. Glover, G.F. Graham, E.G. Gross, A. Krongrad, J.L. Lesher, Jr., H.K. Park, B.B. Sanders, Jr., C.L. Smith and J.R. Taylor. 1996. Effects of selenium supplementation for cancer prevention in patients with carcinoma of the skin. A randomized controlled trial. Nutritional Prevention of Cancer Study Group. JAMA 276:1957.
Combs, J.F., Jr. 1994. Essentiality and toxicity of selenium: a critique of the Recommended Dietary Allowance and the Reference Dose. In: Risk Assessment of Essential Elements. Mertz, W., C.O. Abernathy and S.S. Olin. ILSI Press, Washington, D.C.
Committee on Nutrient Requirements. 1991. Recommendations on Nutrient Intake. 5th edition. Frankfurt, Germany.
Department of Health. 1991. Dietary Reference Values for Food Energy and Nutrients for the United Kingdom (Report on Health and Social Subjects, No. 41). London, UK.
Dourson, M.L. 1994. Methods for establishing oral reference doses. In: Risk Assessment of Essential Elements. Mertz, W., C.O. Abernathy and S.S. Olin. ILSI Press, Washington, D.C.
FAO/IAEA/WHO. 1996. Trace Elements in Human Nutrition and Health. World Health Organization, Geneva.
FNB (Food and Nutrition Board). 1980. Recommended Dietary Allowances, 9th edition. National Academy of Sciences, Washington, D.C.
FNB (Food and Nutrition Board). 1989. Recommended Dietary Allowances, 10th edition. National Academy Press, Washington, D.C.
Gladyshev, V.N., V.M. Factor, F. Housseau and D.L. Hatfield. 1998. Contrasting patterns of regulation of the antioxidant selenoproteins, thioredoxin reductase, and glutathione peroxidase in cancer cells. Biochem. Biophys. Res. Commun. 251:488.
IOM (Institute of Medicine). 1998a. Dietary Reference Intakes. Thiamin, Riboflavin, Niacin, Vitamin B6, Folate, Vitamin B12, Pantothenic Acid, Biotin, and Choline. National Academy Press, Washington, D.C. (Prepublication copy).
IOM (Institute of Medicine). 1998b. Dietary Reference Intakes: A Risk Assessment Model for Establishing Upper Intake Levels for Nutrients. National Academy Press, Washington, D.C. (Prepublication copy).
IOM (Institute of Medicine). 1998c. Dietary Reference Intakes. Proposed Definition and Plan for Review of Dietary Antioxidants and Related Compounds. National Academy Press, Washington, D.C.
Keshan Disease Research Group. 1979. Observations on effect of sodium selenite in prevention of Keshan disease. Chin. Med. J. 92:471.
Levander, O.A. 1991. Scientific rationale for the 1989 Recommended Dietary Allowances for selenium. J. Am. Diet. Assoc. 91:1572.
Levander, O.A. and V.C. Morris. 1985. What can balance studies tell us about human dietary selenium requirements? In: Proc. 5th Intl. Symposium on Trace Elements in Man and Animals. Mills, C.F., I. Bremner and J.K. Chesters, Eds. Commonwealth Agricultural Bureaux.
Mertz, W. 1998. A perspective on mineral standards. J. Nutr. 128:375S.
Mertz, W. 1987. Use and misuse of balance studies. J. Nutr. 117:1811.
NRC (National Research Council). 1983. Selenium in Nutrition. National Academy Press, Washington, D.C.
Poirier, K.A. 1994. Summary of the derivation of the reference dose for selenium. In: Risk Assessment of Essential Elements. Mertz, W., C.O. Abernathy and S.S. Olin. ILSI Press, Washington, D.C.
Rotruck, J.T., A.L. Pope, H.E. Ganther, A.B. Swanson, D.G. Hafeman and W.G. Hoekstra. 1973. Selenium: biochemical role as a component of glutathione peroxidase. Science 179:588.
Schwarz, K. and C.M. Foltz. 1957. Selenium as an integral part of factor 3 against dietary necrotic liver degeneration. J. Am. Chem. Soc. 79:3292.
Standing Nordic Committee on Food. 1989. Nordic Nutrition Recommendations. Copenhagen, Denmark.
St. Germain, D.L. and V.A. Galton. 1997. The deiodinase family of selenoproteins. Thyroid 7:655.
Truswell, A.S., I.E. Dreosti, R.M. English, I.H.E. Rutishauser, N. Palmer. 1990. Recommended Nutrient Intakes. Australian Papers. Sydney: Australian Professional Publications.
WHO (World Health Organization). 1973. Trace Elements in Human Nutrition. Report of a WHO Expert Committee. World Health Organization, Geneva.
Yang, G.Q., K. Ge, J. Chen and X. Chen. 1988. Selenium-related endemic diseases and the daily selenium requirements of humans. World Rev. Nutr. Diet. 55:98.
Yang, G.Q., L.Z. Zhu, S.J. Liu, L.Z. Gu, P.C. Qian, J.H. Huang and M.D. Lu. 1987. Human selenium requirements in China. In: Selenium in Biology and Medicine. G.F. Combs, Jr., J.E. Spallholz, O.A. Levander and J.E. Oldfield, Eds. Van Nostrand Reinhold Company, New York, NY.