Click for next page ( 58


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 57
- Metabolism DIETARY FORMS A number of organic selenium compounds have been identified in plants or plant products. These compounds include selenocystine, selenocysteine, Se-methylselenocysteine, selenohomocystine, selenomethionine, Se-methyl- selenomethionine, selenomethionine selenoxide, selenocystathionine, and di- methyl diselenide (Shrift, 1969~. There is some evidence for the presence of selenite and selenate in plants (Butler and Peterson, 1967; Olson et al., 19701. Selenocystine, selenocysteine, selenomethionine, and Se-methylsele- nomethionine, however, appear to be the major selenium compounds in seeds or forages commonly consumed by livestock (Peterson and Butler, 1962; Shrift, 1969; and Olson et al., 1970~. Thus, organic selenium is the major form for animals consuming natural feeds. Feeds and forages grown in certain areas of North America, however, do not contain enough selenium to meet livestock requirements (see the sec- tion, "Regional Distribution in Crops". One way to correct this deficiency has been to mix grains grown in high-selenium areas with selenium-defi- cient feeds (Ullrey et al., 19771. Approval has now been given to add inor- ganic selenium to feeds deficient in this element, and the most common forms used are sodium selenite or selenate. Selenized yeast tablets, con- taining primarily organic selenium, are available as human supplements and have been shown to increase blood selenium levels (Schrauzer and White, 19781. 57

OCR for page 57
58 ABSORPTION S ELK NIUM IN NUTRITION The absorption of selenium is significantly lower in ruminants than in monogastric animals. The retention of selenium taken orally as selenite was found to be 77 percent in swine as compared to only 29 percent for sheep (Wright and Bell, 19661. Essentially, no absorption of selenium oc- curred in the rumen and abomasum, and the greatest absorption of sele- nium occurred in the small intestine and the cecum of sheep. In swine, no absorption occurred in the stomach and the first part of the small intes- tine, but the greatest absorption occurred in the last part of the small intes- tine, the cecum, and the colon. Similar observations have been noted for rats. The greatest absorption of selenomethionine and selenite occurred in the duodenum, followed in decreasing amounts by the jejunum and ileum, with essentially no absorption in the stomach (Whanger et al., 1976~. By use of the everted intestinal sacs of hamsters, McConnell and Cho (1965) found that selenomethionine was transported against a concentration gra- dient, whereas selenite and selenocystine were not. The transport of sele- nomethionine was inhibited by methionine, but the transport of selenite and selenocystine was not inhibited by their respective sulfur analogues. Some differences in the absorption of selenium incorporated into pro- tein, as compared to the unbound form, have been found. The intestinal absorption of 75Se from a kidney homogenate of rabbits dosed with 75Se- selenomethionine was 87 percent, as compared to 91 percent for seleno- methionine mixed with unlabeled rabbit kidney homogenates (Thomson et al., 19751. 75Se-selenite or 75Se-selenomethionine was injected into the coe- lomic cavity of fish, and their muscle removed, homogenized, and fed in diets to rats. The intestinal absorption of 75Se given as labeled fish was less complete than that of 75 Se mixed with unlabeled fish (Richold et al., 19771. In a short-term human study, selenite was almost as well absorbed as selenomethionine in young women (Thomson et al., 1978a). In a pro- longed study, three individuals were given different treatments. One was given 100 ,ug selenium as selenomethionine, the second was given 100 ,ug selenium as selenite, and the third was given 65 ,ug selenium in mackerel daily for 4 to 10 weeks (Robinson et al., 1978a). Selenite-Se was not ab- sorbed (45 percent) as well as selenomethionine (75 percent) or fish-Se (66 percent). Thus, different dietary forms of selenium may have an influence on absorption in humans. VASCULAR TRANSPORT From 75 to 85 percent of the selenium in ovine erythrocytes is associated with glutathione peroxidase (GSH-Px)(Oh et al., 1974). Essentially all of

OCR for page 57
Metabolism 59 the selenium in erythrocytes of rats is associated with GSH-Px, but only about 10 percent of the selenium is associated with this enzyme in human erythrocytes (Behne and Wolters, 1979~. Even less of the selenium (1.5 per- cent) in human plasma is associated with this enzyme. Patterns similar to those in humans have been noted for the rhesus monkey (Whanger, un- published observations). Thus, the major proportion of selenium is associ- ated with GSH-Px in rat or sheep erythrocytes, but not in primate erythro- cytes. Even though there are differences in the amount of selenium associated with GSH-Px between primates and other animals, the metabolism of sele- nite by blood in vitro is similar. The uptake and release of selenite by bovine (Jenkins and Hidiroglou, 1972) or human erythrocytes (Lee et al., 1969) involve sulfydryl groups, and the binding of selenite to plasma proteins is dependent upon the presence of erythrocytes (Lee et al., 1969; Sandholm, 1974, 1975~. This binding of selenium to plasma proteins is not energy dependent and does not require protein synthesis (Porter et al., 1979~. These relationships have been clarified further with studies on rat erythro- cytes (Gasiewicz and Smith, 1978~. The uptake and subsequent metabo- lism of selenite are dependent upon the reduced glutathione concentrations in the erythrocytes. The results indicate that H2Se or a similar product of GSSeSG reduction by glutathione reductase was the final product of sele- nite metabolism by rat erythrocytes. In mice plasma the protein-bound selenium is mainly located in al- bumin, with smaller amounts possibly situated with the or- and 3-globulins (Sandholm, 1974~. In contrast, the important selenium-binding proteins in plasma of humans appear to be the lipoproteins. When human blood was incubated with 75Se-selenite in vitro, the most heavily labeled proteins in the plasma were the o`-lipoproteins and an unidentified fraction located electrophoretically between the oil- and 32-globulin fractions (Sandholm, 19751. This is consistent with the findings of Burk (1974), who showed that up to 16 percent of the plasma 75 Se was found in very low density lipopro- teins after the administration of 75Se-selenite to patients. The 75Se bound to human plasma proteins is absorbed by lymphocytes in preference to sel- enite, suggesting that plasma proteins function as carriers of selenium to these lymphocytes (Porter et al., 1979~. BODY RETENTION AND TISSUE DISTRIBUTION The tissue distribution of selenium in various animals using either 75Se or stable selenium has been reported by several investigators. With required dietary selenium intake, the kidney contains the highest concentration of selenium, followed by the liver and other glandular tissues such as the

OCR for page 57
60 SELENIUM IN NUTRITION spleen and pancreas. Intestinal and lung tissues can have relatively high concentrations. Cardiac muscle contains appreciably more than skeletal muscle. Wool and hair may have relatively high concentrations, but ner- vous tissue has low concentrations. Selenium concentrations in muscle, liver, and kidney for sheep, cattle, and swine and in muscle, liver, and eggs for poultry fed various levels of dietary selenium are shown in Table 6. Some typical values for selenium in tissues in addition to those shown in Table 6 are (in ppm wet weight) cardiac muscle, 0.15 to 0.20; pancreas, 0.34 to 0.44; ovary, 0.19 to 0.25; cerebrum, 0.07 to 0.09; and wool, 0.21 to 0.49. Not only is the tissue content of selenium dependent upon the level in the diet, but also upon the chemical form. In general, selenium is deposited in tissues at higher concentrations when present in diets as organic rather than as inorganic selenium. A possible exception to this involves using as a selenium source fish meal, in which heavy metals are thought to complex the selenium (Mahan and Moxon, 1978~. Ullrey et al. (1977) fed sheep and cattle diets composed of low-selenium ingredients from Michigan and added selenite selenium to make the diets contain 0.20 and 0.30 ppm sele- nium, respectively. These workers then used grains from South Dakota to formulate diets containing similar levels of selenium. In both sheep and cattle the selenium content in muscle, liver, kidney, and plasma was higher when the South Dakota diets were fed than when the same amount of sele- nium as selenite was fed in the Michigan diets. Likewise in pigs the sele- nium content in muscle, liver, and kidney was higher when the diet con- taining 0.45 ppm selenium was composed of ingredients from South Dakota than when the same level of selenium, predominantly as selenite, was fed in diets composed of ingredients from Michigan (Ku et al., 1973~. Also, the selenium in various grain products for swine must be considered. Inorganic and organic selenium contributed from brewers' grains or dis- tillers' grains and solubles were shown to result in similar weekly serum selenium response curves within each selenium level with 4-week-old weanling swine, while selenium contributed by fish meal had lower levels (Mahan and Moxon, 1978~. At 0.4 ppm dietary selenium intake, muscular tissue selenium levels were greater, compared to the selenite source, when fish meal or brewers' grains were fed, whereas nonmuscular tissue (liver, kidney, and testes) had similar selenium concentrations for all groups. Fish meal provided the poorest selenium retention of all test products eval- uated. This difference in response to various organic selenium sources has also been demonstrated with poultry. Selenium given as selenomethionine results in higher content in muscle than when given as selenite or seleno- cystine (Osman and Latshaw, 19761. Latshaw and Biggert (1981) also showed that dietary selenomethionine results in higher egg selenium con

OCR for page 57
Metabolism 61 centrations than does dietary selenite. Selenium given as selenomethionine or as seleniferous wheat resulted in more selenium in eggs than when sele- nium was given as selenocystine, Astragalus selenium, or fish meal (Lat- shaw and Osman, 19751. Consistent with tissue data on sheep, cattle, and swine, when selenium is supplied as a natural feed ingredient (0.42 ppm) for hens, it causes higher muscle, liver, and egg selenium content than when the same amount is supplied predominantly as selenite (Latshaw, 1975~. Furthermore, the availability of selenium must be taken into con- sideration. For example, using chicks as the test animals, the availability of selenium from plant sources was found to be higher (60 to 210 percent) than from animal sources (8 to 25 percent), as compared to sodium selenite (Cantor et al., 1975b). In lambs suffering from white muscle disease (WMD), levels of selenium less than 0.05 ppm in muscle and less than 0.1 ppm in liver (dry basis) have been observed (Allaway et al., 1966~. When the dietary intake of selenium is very low, the kidney of lambs has a higher concentration of selenium than does the liver, but when the dietary intake of selenium is increased, the liver usually has a higher concentration of selenium than does the kid- ney (Ewan et al., 1968c; Oh et al., 1976b). These workers fed lambs an artificial milk composed of torula yeast, glucose, stripped lard, vitamins, and minerals. Tissue selenium levels were very low in unsupplemented lambs (less than 0.02 in muscle, less than 0.08 in liver, and less than 0.2 ppm in kidney). Supplementation with selenium as selenite up to 1.0 ppm caused the greatest increase of selenium content in the liver. In work with practical diets (Paulson et al., 1968a; Oh et al., 1976a) various levels of selenium as selenite (up to 0.52 ppm) were added to deficient diets, and muscle, liver, and kidney levels increased respectively from 0.02, 0.05, and 0.5 to 0.1, 0.9, and 3.8 ppm. Paulson et al. (1968a) determined the sele- nium content of lambs fed "commercial" diets at several state agricultural experiment stations. These diets contained 0.16 to 0.52 ppm selenium; the selenium in skeletal muscle ranged from 0.2 to 1.6 ppm, in the livers from 0.61 to 2.5 ppm, in the kidneys from 3.8 to 7.5 ppm (dry basis), and in blood from 0.23 to 0.61 ppm. These workers also reported that the blood reached 0.71 ppm in sheep grazing native grasses containing 1.15 ppm selenium. Ullrey et al. (1978) and Paulson et al. (1968a) determined the tissue selenium residue of lambs fed up to 264 ppm selenium in salt. At this highest level of selenium intake in salt the muscle, liver, blood, and kidney contained, respectively, 0.57, 6.5, 0.36, and 4.8 ppm selenium (dry-weight basis). At high levels of selenium intake the livers and kidneys of cattle and sheep have been found to contain 5 to 25 ppm of selenium (fresh-weight basis). Moxon et al. (1944) reported 5.6 ppm selenium in liver and 3.0 ppm

OCR for page 57
62 ._ - V2 so ._ o - :> o .~ Ct - ._ Cal A V, ._ - Ct ._ ._ ._ V2 o U. o ._ Ct sly C) o V, ._ V) V) C) ~ V7 ~ . - Ct CQ _ 't ,,, V) 5- To A - . - ~ ,... Cal ~ ;2 ._ 'a 3 US ~ ~ O ~ )_ ~ ~ IN X _ ~Cd ~_ Cd 4_ _ ~_ ~ ~ b1) _ a, _ ~g ~ ~g ~ 3 it' ~ '36 ~3 c's 0 3 c': 4) ~'- 3 o o o o ~ ~ r~ r~ oO oo C~ ~00 0 00 0 0 0 0 0 0 ~ 0 oo ~ ~ 0 oo ~ oo ~ 1~ ~ ~ oo ~o ~ ~ ~ oo ~ oo ~ ~ . . . . . . . . . . . . . . . . . . O ~O ~ ro ~ r~ ~ ~ ~ ~ u~ 1 1 1 1 1 1 1 1 1 ~ ~oo 0 0 0 0 ~ U: ._ U ~ C~ ~oo ~ 0 ~ ~ ~ ~ ~ 0 0 ~0 0 0 ~ 0 0 0 ~0 0 0 0 ~0 C~ ~ ~ oo ~ ~ ~ ~ ~ ~ ~ ~ ~ 0 . . . . . . . . . . .. . . .. . . O ~O O ~ ~O O ~ O ~ ~ ~ ~D~ ~ ~ CN 1 1 1 1 1 1 1 1 1 1 1 1 0 U ~0 ~0 0 ~00 ~0 0 u, ~O ~O O r ~0 0 0 0 0 0 ~ _ E~ ~ ~o 0 ~ ~ ~0 ~ 0 ~o ~ u~ oo ~ ~ oo .= ~ O ~ ~ ~ ~ ~ ~ ~ u ~ ~ ~ ~ ~ ~D 0 r ~0 0 0 0 C) 1 1 1 1 1 1 1 1 1 1 1 1 ~ ~ ~ ~o ~ ~ o ct ~ ~ O 00 0 0 ~O ~ ~ ~ r~ V) ~ o o o o o o o o o o o o 0 .- ~ C) C) C} C) C) C) C) L~ ~ .o .o .o ~ ~ ~.o ~ ~ .o .C) .o ~ ~ s:: .o S~ O ~ ~Ct Ct ~ ~ ~Ct ~ ~ Ct ~Ct O O O ~ ~ ~ O ~ ~ O O O ~ ~ ~ O ~'_ ._ ~ Ct e _ e _ ~ ~1 ~ ~' ~D V? E ~/ _e x O ~ ~ ,,3 V) ~o Uro) O Cd _ U) ~ ~ =) ~ I ~ ~,o I V) U) - 0 O O ~ ,,,, ,'s 0 0 0 ~ ~ r~ ~ r~ ~ ~ r~ ~ ~ ~ ., a~ _ ~ _ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ct~= ~ ,= 3 ~ 3 3 .= ~ ~ ~ - ~ ~ o r~ ~ u~ r~ ~ ~ ~4 ~

OCR for page 57
63 o CN U~o c~ c~ r~ ro~ ... . . .. o__ _ _ ~o ~oo r~ r~ _ ~ ~ U~ ... . . .. ooo o o oo C) ~ 3 3 o _` - - ~: - ~ .= c~ 3 ~ ~v ce 3 - - CO c C~ C~ = o - . ~_ ~ 3 ~ 3 . ~ 3 s~ o oo o ~ ~ ~ _ 3 _` - - Ct ~ ._ s~ _ ;^ a~ - - .= c~ 3 ~ Q.) ce 3 0 0 0 oO ~ - ~ a~ ~ ~ Oo ~- ~ c~ ~ ~ ~ 0 r~ r~ u ~r-, . . .. . . . .. . . . . . r~ ~ _ _ __ ~ _ 1 1 1 o o ~C~ _ o o o ~ oo o ~ ~ oo ~ o _ . . . . . . . . . . a~ 0 oo ~ O O ~ O O O c~ ~ r~ o U~ - o o o oo ~ ~ . . o o C o U~ o o o o - o C) C) ~C) C) C) C) C) ~C ~C) ~ 'c ~,,4-C ~,O-C C C C C C C C C C C C C C C C .Q .Q C O ~ O ~ O O ~ O ~ ~ O ~ ~ O ~ ~ O ~ O O ~ O ~ V) ~ O - ~O V~ - - (~S CT ~t~] C (1) C 8. 0 =, O - C V7 Cd ~ O C ~ C ~C e. ~ ~3 't' 3 s, C ~ ~ ='- ~ ~ ~ ~ ::i ~, ~ o ~ o 't ~ =.= ~ 'd .= .= .= ~ ~ ~ ~ ~ o ^ ^ ^ ^ ~ ^ ~ o v~ u~ u~ ~c ~ ~ ~ ~ ~ o ~ ~ o o ~ ~ o ._ ~ O V~ ~ ~ r~ ~ (~ U~ ~D V:} ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ O O O O ~ - O O ~ ~ C) _ ~ = p~ (~s i~ ~ 0 ~ ~ 0 ~ ~ 0 ~ ~ ~ ~ ~ ~ V V ~ ~ ~ ~ ~ ~ ~ _` DE Y tts 3 3 ,, ~ 3 ,4 E ~ ~ ~

OCR for page 57
64 .s s 1 Cal U. Cal - ._ _ ._ - o o ._ Ct sit C) o v ._ V) Ct Ct ~ ~ ~4 V, _ ~ c., au ~ ~ lo, ~ _ Ed ~ ._ ~ V V, :^ ._ . C) Cal [.L1 = so o I ~ o sol o v Ct - Ct ._ 3 ~3 hi.S ~ ~ _ A) ~ 3 ~ 3 cat ~_ ~ 3 As ~ ~cd ~Ct s:- ~' 3 ~ 3 3 ~ 3 o o ~ ~ Lr) ~o o ~C~ ~ ~ U ~U~ C~ _ _ _ _ _ .~:3 ,s ~- - ro r~ O u ~ O ~ oo r~ 0 ~r~ ~ oo oO oO . . . . . . o oo oo ~O r~ . . . ~o ~ r~ ~ u: u~ O ~ r~ ~ O ~ a~ oo ~ . .. . . . . ... r ~0 ~0 0 0 0 0 ~ o ~ ~ ~ ~ ~ ~ o ooooo r~ ~ ~D oO a~ 0 r~ ~ ~ ~ . . .. . . . . . . ... o o oo o o o o o o o ~ U) ~ ~ r~ O~ ~ ~ ~r~ n ~ ~ ~ ~ ~00 oo ~o o o o~ ~ ~ o o o o o ~ .. .. . . .. . . . . . . ... o oo o o oo o o o o o o o 11 oo ~C~ C) C) CJ C) C) _C) _ _ _ ~CJCJ _C)C)C) C) ._ ._._ ._._._._._._ ._ Ct C~ ~ ~ ~ C~ Ct e~ ~ Ct Ct C ~ o ~o o o ~ ~ ~ o O O ~O ~ ~ ~ O O O ~ O O O O a., 0 o u' ~o o ~c.} ~ ~ ~ c) ~ ^ ~ o ~ ) ~ ao ~ ~ ~ ~ - _ ~ O ~~ ~3 ~ s~ Q) o ~ ~ ~ ~, 0 ~ o 0., ~ ~ ~ c = ~ (t ~.o ~ ~ o ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ,o ~ ~ =^ ~o o ~ 0 ~ 0 0 0 0 ~ 0 0 ~ 0 0^ 0 ~ ^~ ~0 ~ . ~ o ~ ~ ~ o ~ ~ ~ ~ o ~ ~ ~ ~ ~ ~ ~ ~ ~ o - ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ V V ~^ ~ <~ ~c~ ~ 3 ~,~ ~ 3 ~ ~Y ~ ~ vS) ~0 _ ~ O ~ O ~ _ V) ~c4 ~P~ _ ~ ~ _ ~1

OCR for page 57
65 U) _ 3 ~ - , 3 ~ O ~c~3 (~\ _ ~ _ s a~ ~O ~ ~ ~ _` ~ ~ o o o oo oo o o oo ~ ~ ~ r~ ~ _ 1 1 1 _ ~ r~ oo oo _ ~ _ oo ~ oo U~ O ~ ro ~ ~ r~ . . . . . . o ~o ~ ~ ~ o . r~ . . ~ ~:: =: 0 ~ oO ~ oo oo r~ ~r ~D oO 0- __ ~ ~ ~o _ ~ ~ ~ ~oo ~r~ . . . . . . . . . . . . . . . . . o o o o o o o o o - o o o o o 1 1 1 ~D . . . o o o ~ "5 ~ ~ r~ ~ . . . o o o 06 oo a to c o O4 o to 0 0 ~ ~ ~ u: ~ cr ~ ~c~oo Lr' ~ u~ ~ O ~ ~ ~ r~ ~0 r~ ~ c~ 0 ~ _ _ ~c~ ro .. . . ...... . .......... O Ol Olo0000 0 ~00000000- oo ~ ~ oo .. . . . ~ o o o C)C ~CJC) c-} c) - c) c) c ~- c) c) ~c) - - c ~c) c, c) . - . - . - . - . - . - . - . - . - . - . - . - . - s: ~ o ~ ~ ~o ~ ~ ~ o o ~ ~ ~ ~ o o ~ o o o ~o o o o ~ ~ o o o o O a ~_ O c 0 _ 0 9 0 0 0 ~ 0 c 0 ~ c c c : ~= ~ ~ ~ E c c u ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~, o 0) O ~ V, V, au V ~) ~ ~ ~ ~ ~ ~ ~ a~ v~ 3 v~ 3 ~ v~ c~ >, c~ c~ ~ ~v ~s: c~ u~ v, u~ 04 ~ ~ ~ o 0 0 0 .G) o O ~ ~ ~ ~ ~r ~ 0 0 0 0 ~ ~ ,,= ~ _ ~ _ ~ ~ _ ~ ~ 0 0 0 ,5:) ~ ~ ~ ~ ~ ~ ~ ~ P 0 ~ 0 ~ c13 0 0 sO O O 0 0^ 0 0^ 0^ 0 0 0 0 ~ v, ~ 0^ ~ ~ . _ _ c) __~___ ~_ ~_______~ 0 ~ ~ ~ 0 ~0 "o ~ ~ ~ ~ ~ ~ ~ 0 v v v v v ~: ~ 3 ~ c~ S~ 1 ~ u, c; ._ v ~ v' . _ ~ 3 _~ ~ ~s c' c~ u, ~ ~ s~ v~ ct s~ ~ O ._ s" ct ~o ~ s~ Z 0 cd 5- . o o - . cd cd t a, c) c) ~ ~ o o c) c) ~ ~ - - ~ ~ o d 50 cd :^ s~ o -- s~ c) c)~ ~o - o ~o ~s . -~ ~'v v'~ 'e. - . -. - ~D~ 'e - o v' ~ - - s-d e s~ ~ (~o - e ~ ~ o o s~ c~ ~ t 'e 'e ~ a' . ~ ~y . - ~ - ~ s:: . - . - o o . - . - cd cd : ) o o ) c~ v, - - a ~ ~: . .- ~ o s~ - o ~ :~ cd e ~ o ) - cd o ~ o - cd cd ~ o ~ . - . c) 3 ~ c; a~ ~ cO~ _ cd u,

OCR for page 57
66 SELENIUM IN NUTRITION in muscle (both fresh-weight basis) of steers that had been pastured on seleniferous rangeland for 3 years. Maag and Glenn (1967) fed selenite at levels up to 12 to 24 mg selenium/kg body weight to steers until 6 out of 8 animals died from selenium poisoning. The selenium content of the mus- cles of these animals ranged from 0.10 to 0.73 ppm and the livers con- tained 5.0 to 12.3 ppm (fresh-weight basis). When levels of up to 0.3 ppm selenium in diets were fed to cattle, muscle, liver, and kidneys contained, respectively, 0.14, 0.50, and 1.58 ppm selenium (wet-weight basis; Ullrey et al., 1977~. Addition of up to 0.2 ppm selenium as selenite to diets for pigs contain- ing high levels of natural selenium does not significantly increase the tissue selenium content (Groce et al., 1973b; Jenkins and Winter, 1973; and Ku et al., 19731. Concentrations of up to 2.2, 3.38, and 11.0 ppm (wet-weight basis) were obtained, respectively, for muscle, liver, and kidney when nat- ural dietary selenium levels were 0.8 to 0.9 ppm. Whole-blood values ranged up to 0.35 ppm (Jenkins and Winter, 19731. The concentrations of selenium in muscle, liver, and kidney in deficient pigs appear to be 0.07, 0.08, and 1.2 ppm, respectively (dry-matter basis; Ewan, 19711. Feeding practical diets (0.13 to 0.2 ppm Se) to turkeys resulted in 0.17 to 0.22 ppm and 0.62 to 0.70 ppm selenium (fresh weight), respectively, in muscle and liver (Cantor and Scott, 1975~. The addition of up to 0.2 ppm selenium as selenite to this diet did not result in significant increases of selenium in these tissues. The addition of 0.06 ppm selenium in various forms to torula yeast-based diets (0.03 ppm Se) for chicks resulted in up to 0.9, 0.78, and 1.4 ppm selenium (lyophilized tissues), respectively, in mus- cle, liver, and kidney (Osman and Latshaw, 19761. Addition of 0.32 ppm selenium as selenite to a corn-soybean basal diet (0.10 ppm Se) resulted in 0.42, 0.82, and 0.74 ppm selenium (dry-weight basis), respectively, in muscle, liver, and eggs of laying hens (Latshaw, 19751. The selenium con- tent of eggs can also be influenced by dietary selenium. Various levels of selenium and different chemical forms of this element were added to a corn-soybean meal basal diet (0.05 ppm Se) for laying hens, and concen- trations of up to 1.5 and 0.63 ppm selenium (dry weight) were obtained for egg white and egg yolk, respectively (Latshaw and Osman, 1975~. How- ever, the relative selenium concentrations found in egg whites and egg yolks may sometimes be reversed (Arthur, 19721. The selenium content of human tissues appears to show patterns similar to tissues from animals fed diets with required levels of selenium. The mean selenium content (wet-weight basis) was highest in kidney (1.09 ppm) followed by the liver (0.54 ppm), spleen (0.34 ppm), and testes and pancreas (0.30 ppm), with the lowest content in brain (0.13 ppm) for hu- mans ranging from 9 months to 68 years of age (Schroeder et al., 1970~.

OCR for page 57
Metabolism 67 The mean selenium content of hair was found to be about 0.57 ppm, which is similar to that from rats fed control diets. This is significantly higher than the hair values (0.047 to 0.088 ppm) reported for people living in a selenium-deficient area of mainland China (Keshan Disease Research Group, 1979a). The calculated human body burden of selenium for Amer- icans was about 15 mg (Schroeder et al., 19701. This is two to three times more than that calculated for New Zealanders (4.7 to 10 mg), which is not surprising since the selenium status of the latter is known to be low. As with animals, the blood selenium levels in humans are apparently influenced by dietary intake. The blood selenium in people of New Zealand averaged about 0.07 ppm, which is low in comparison to the values re- ported for people (as summarized by Griffiths and Thomson, 1974) living in the United Kingdom (0.32 ppm), Sweden (0.12 ppm), Canada (0.18 ppm), or the United States (0.22 ppm). The blood selenium levels in peo- ple arriving in New Zealand from the United States have been shown to gradually decrease from 0.20 ppm to 0.07 ppm within about 200 days. The lowest blood values reported thus far are 0.01 to 0.03 ppm for humans suffering from Keshan disease in mainland China (Keshan Disease Re- search Group, 1979b). McKenzie et al. (1978) have summarized the blood selenium levels in children living in various countries; the levels range from a low of 0.04 ppm in New Zealand to a high of 0.81 ppm in Venezuela. The peak of selenium retention on a whole-body basis (Mush et al., 1967) and in tissues such as blood, liver, muscle, kidney, spleen, and lung (Brown and Burk, 1973) is reached within hours (up to 24) after the injec- tion of 75 Se as SeO32- . In tissues like brain and thymus, however, the max- imum is not reached until about 2 days after injection (Brown and Burk, 19731. The kidney retains the greatest amount of selenium (percent of dose) for short time periods (48 hours) followed by the liver, pancreas, heart, and muscle in decreasing order (Kincaid et al., 19771. The small intestine, however, may retain a large amount (up to 8.5 percent of dose) of the 75Se. The importance of selenium in male reproduction is indicated by the incorporation of 75Se as SeO32- in the reproductive organs. In contrast to other tissues, the maximum incorporation of selenium in testes and epidid- ymis of rats was reached 2 to 3 weeks after injection (Brown and Burk, 1973; and McConnell et al., 1979b). As an indication of the accumula- tion of selenium in male reproductive organs, about 40 percent of the total body 75Se was found in testes plus epididymis of rats 3 weeks after injection (Brown and Burk, 19731. Within the sperm, 75 Se became associated pri- marily with the midpiece of the sperm tail. Calvin (1978) extended these studies with rat sperm and found the selenium to be primarily in the tail keratin, a disulfide-stabilized fraction obtained by extracting isolated tails

OCR for page 57
68 SELENIUM IN NUTRITION with sodium dodecyl sulfate. The name selenoflagellin was proposed for this selenium-binding polypeptide of 17,000 molecular weight (MOO) in sperm. Calvin suggested that this molecule may be essential for proper as- sembly of the rat sperm tail. In bulls, the peak in accumulation of 75 Se in whole semen was reached 40 days after injection (Smith et al., 1979), which is similar to results with a ram in which the maximum incorporation of 75Se in sperm was reached 49 days after injection (Tripp et al., 1979~. In the bull the epididymis and testis retained the greatest amount of 75Se (CPM/g basis), except for the kidney, 23 days after injection. Among the accessory glands, the prostate and seminal vesicles contained the highest levels.of 75Se. Some studies on the binding of 75Se to proteins in testes have been done (McConnell et al., 1979b). 75Se is bound to three proteins with molecular weights of57,000, 45,000, and 15,000. Thetwohighermolecu- lar-weight compounds predominate at short labeling times (1 to 24 hours), whereas the 15,000 molecular-weight compound predominates after longer labeling periods (1 or more weeks). METABOLISM There are many factors that influence the metabolism of selenium. Among these are the chemical forms of selenium, sulfur, arsenic, metals, microor- ganisms, and vitamin E, and previous selenium intake. INORGANIC SELENIUM There is very little doubt concerning the ability of animal tissues to convert inorganic selenium to organic forms. This is demonstrated by the incorpo- ration of 75Se from selenite into dimethyl selenide (Hsieh and Ganther, 1975), into GSH-Px (Oh et al., 1974; Forstrom et al., 1978), and into sele- noamino acids (Godwin and Fuss, 1972; Olson and Palmer, 19761. The selenium residue in glutathione peroxidase has been indicated to be se- lenocysteine (Forstrom et al., 19781. A small amount of selenocystine was found in tissues of rabbits after administration of 75Se-selenite (Godwin and Fuss, 1972~. Evidence was also obtained for some other unidentified selenoamino compounds. Similarly, a selenocysteine derivative was found in p~ronase digests of acetone powders of liver and kidney tissues from rats given selenite (Olson and Palmer, 1976~. These workers found no seleno- methionine under the conditions of study. Thus, the conversion of selenite to selenomethionine in tissues has not as yet been demonstrated. Godwin et al. (1971) isolated 75Se-selenomethionine from ewe milk protein follow- ing the intraruminal administration of 75Se-selenite, but this could be due

OCR for page 57
Metabolism 69 to the actions of the rumen microorganisms as discussed elsewhere (see the section, "Rumen Microorganisms". Although the pathways for reduction of selenite to selenide have been fairly well established (Ganther, 1979), the pathways for conversion of sel- enide to selenoamino acids have not been fully delineated. Since seleno- cysteine is one of the forms of selenium identified in animal tissues, most of the work has been done on its incorporation into proteins. Evidence has been presented for a specific transfer RNA in rat liver for selenocysteine (Hawkes et al., 1979~. By this mechanism the selenocysteine could be in- corporated during translation via the action of this tRNA and its charging enzymes. An alternative mechanism is that the selenocysteine is formed in situ from some previously incorporated amino acid such as cysteine or serine, which would be susceptible to a posttranslational modification. This type of reaction could occur similarly to the cysteine synthase reac- tion, which can produce cysteine from serine and sulfide if selenide or its equivalent is substituted for the sulfide (Olson and Palmer, 1976~. ORGANIC SELENIUM In contrast to sulfur, selenium compounds tend to undergo reductive path- ways in tissues. However, reduced selenium compounds can be metabo- lized by animal tissues. Some evidence has been presented for formation of selenocystathionine, selenoglutathione, selenotaurine, and selenocysteic acid in tissue homogenates of chicks and mice injected with 75Se-seleno- methionine (Martin and Gerlach, 1972~. However, differences in either the metabolism of various selenium compounds or their biopotency have been observed. In both rats (Miller et al., 1973) and sheep (Jacobsson, 1966) greater retention of radioactivity was found in the pancreas after the injec- tion of 75Se-selenoamino acids than after the injection of selenite or sele- nate. Selenomethionine was found to be four times as effective as either selenite or selenocystine in prevention of pancreatic degeneration in chicks (Cantor et al., 1975a). In contrast, selenite, selenate, or selenocystine were more effective than selenomethionine in the prevention of exudative di- athesis in chicks, which was highly correlated with plasma GSH-Px activity (Cantor et al., 1975b). Other differences in selenium compounds have also been noted in chicks. Feeding selenomethionine to hens resulted in more selenium in egg white than in egg yolk, whereas feeding selenite or seleno- cystine resulted in more selenium in egg yolk than egg white (Latshaw and Osman, 1975~. A higher concentration of selenium was found in the pan- creas and breast muscle (Osman and Latshaw, 1976) of chicks fed seleno- methionine than when fed either selenite or selenocystine. These authors concluded that selenocystine is not incorporated into protein but is metab

OCR for page 57
70 SELENIUM IN NUTRITION olized like inorganic selenium compounds. In work by others, marked dif- ferences were found in the biopotency of various organic selenium com- pounds in the prevention of dietary liver necrosis in rats (Schwarz and Fredga, 19751. Thus the data of these different investigators suggest that the various organic selenium compounds are not metabolized to common intermediates. REDUCTION OF SELENIUM Evidence has been obtained for the formation of hydrogen selenide from selenite in tissues (Diplock et al., 1973; Hsieh and Ganther, 19751. There are some formidable problems, however, involved in studying the produc- tion of hydrogen selenide. Oxygen must be rigorously excluded in order to prevent oxidation of H2Se to See. The volatile selenium is readily deposited on glass, nylon, and polypropylene and plastic surfaces. Polystyrene and polyethylene are more satisfactory (Ganther, 1979~. Carrier gas, rate of gas delivery, and types of trapping agents used are other critical factors (Dip- lock et al., 19731. Identification of H2Se as the acid-volatile product is usu- ally based on its trapping behavior, in which silver nitrate or sodium arse- nite are very effective. Nitric acid, however, is a poor trapping agent for this gas. McConnell and Portman (1952a) identified dimethyl selenide in the exhaled gases of rats given 75Se-labeled inorganic selenium. Dimethyl sele- nide can be trapped in soluble form by using nitric acid trapping solutions. Ganther (1979) has provided the basic information on the mechanism for reduction of selenite to selenide. A specific requirement for GSH, an- aerobic conditions (for best activity), and NADPH are essential for this reduction. Reduction of selenite to selenide occurs by a series of reactions involving initially the nonenzymic reaction of selenite with GSH to form an intermediate in which selenium is joined to GSH in the S-Se-S linkage (re- action 1~. This is followed by NADPH-linked reduction of this intermedi- ate by GSH reductase to form H2Se (reactions 2 and 31: NADPH NADP NADPH NADP H2 SeO3 + GSH ~ GSSeSG~GSSe~H2 Se Both the microsomal and the cytoplasmic fractions catalyze the methyl- ation of selenium, apparently by methyltransferases acting upon the hy- drogen selenide (Hsieh and Ganther, 19771. The microsomal enzyme sys- tem is quite labile and is exceedingly sensitive to arsenite. The activity of the liver microsomal system can be increased four-fold by feeding rats a stock diet rather than a purified diet (Hsieh and Ganther, 1976~. The cyto

OCR for page 57
Metabolism 71 solic Se-methyltransferase can be readily purified from rat liver or kidney and is not nearly as sensitive to arsenite as the microsomal one. S-adeno- sylmethionine is an effective methyl donor in either system. McConnell and Portman (1952b) reported that rodents tolerated rather large amounts of dimethyl selenide (the LD50 at 24 hours was 1.3 g selenium or 1.8 g di- methyl selenide/kg of body weight for rats), suggesting that it is a detoxifi- cation product of selenium metabolism. While these data have not been confirmed by others, Obermeyer et al. (1971) found that the intraperito- neal injection of male albino rats (Sprague-Dawley) raised on a commer- cial rat chow with up to 0.8 g Se (1.1 g dimethyl selenide)/kg of body weight caused no deaths in 5 animals 24 hours after injection. Parizek et al. (1976) reported 90 percent mortality in 24 hours after intraperitoneal injection of adult male rats with only 2.2 mg dimethyl selenide/kg of body weight, which is about 400 times lower than the LD50 reported by the pre- vious workers. There is no obvious explanation for this large difference, but some of it could be due to a sex-linked difference in lethality of this compound (Parizek et al., 1976~. The selenium status of the animal could also have had an effect. No mortality was observed in dimethyl selenide- injected rats that had been drinking water containing 1.0 ppm selenium, whereas up to 90 percent mortality occurred in rats injected with this com- pound without prior treatment with selenium (Parizek et al., 19761. Thus, the reason rats used by McConnell and Portman (1952b) were more resis- tant to dimethyl selenide may be due to previous exposures to higher levels of selenium. PRE VI O U S S E LE NIUM I NTAKE The degree to which the tissues have been previously saturated by dietary selenium greatly influences the retention of a subsequent dose. Supple- mentation of diets for rats with various levels of selenium up to 5.0 ppm caused a progressive drop in retention of 75Se-selenite from more than 50 percent to less than 20 percent of the dose (Hopkins, 1962~. Marked in- creases in the urinary excretion of 75Se accounted for much of the de- creased retention. An inverse relationship between the selenium status of sheep and the retention of a dose of 75Se-selenite has been observed (Mush et al., 1967; Lopez et al., 19691. This is also reflected in the uptake of selenite by erythrocytes. Erythrocytes from selenium-deficient sheep will take up a larger amount of radioactive selenium during in vitro incubation than will those from animals with adequate selenium (Wright and Bell, 1963; Weswig et al., 19651. This test has been suggested as a possible use- ful diagnostic procedure in anticipating selenium-responsive diseases. Furthermore, the fixation of 75Se into rumen bacteria in vitro is inversely

OCR for page 57
72 SELENIUM IN NUTRITION proportional to the previous dietary intake of selenium by the host sheep (Hidiroglou et al., 1968~. The selenium status also appears to influence the number of selenium-binding proteins in plasma. Two peaks of protein- bound 75Se were seen on gel filtration chromatography of plasma from rats fed a selenium-adequate diet, whereas only one peak was present from rats fed a selenium-deficient diet (Burk, 19731. RUMEN MICROORGANISMS The rumen microorganisms are probably responsible for the lower absorption of selenium in ruminants than in nonruminants. Much of the dietary sele- nium is reduced to insoluble forms by rumen microbes (Cousins and Cairney, 1961; Peterson and Spedding, 1963; Whanger et al., 19681. A greater per- centage of inorganic than organic selenium is converted into these insoluble forms, and a high-carbohydrate diet is more favorable to selenium conversion into insoluble forms than a high-roughage diet (Whanger et al., 1968~. Even though rumen microbes convert a portion of the selenium into in- soluble forms, they also incorporate selenium into their proteins. Seleno- methionine was shown to be incorporated into bacterial protein when ru- men fluid was incubated in vitro with this compound (Paulson et al., 1968b). Characterization of the 75Se-containing compounds in rumen mi- crobes revealed the presence of 75Se-selenomethionine after incubation with 75Se-selenite in vitro (Hidiroglou et al., 1968~. Thus, rumen microbes appear to be able to convert inorganic selenium to organic selenium com- pounds, as well as to incorporate organic selenium compounds into bacte- rial proteins. INFLUENCE OF SULFUR The sulfur analogues of selenium compounds appear to have the greatest influence on selenium metabolism. This is demonstrated by nearly a three- fold increase in urinary excretion of selenium following a parenteral dose of sodium selenate when rats were given sulfate either parenterally or in the diet. Sulfate had only slight effects on the urinary excretion of selenium that was administered in the form of selenite (Ganther and Baumann, 1962a). Rather high levels of selenium and sulfur were used in these exper- iments. Although sulfur has been implicated in promoting selenium defi- ciency problems, there are no consistent reports on the influence of this element on selenium metabolism. Paulson et al. (1966) found little differ- ence in the selenium uptake by tissues when lactating ewes, fed diets con- taining either 0.28 or 0.62 percent sulfur, were dosed intraruminally with 75Se-selenate. When 75Se was injected into the rumen, Pope et al. (1979) found that wethers fed a low-sulfur diet (0.05 percent) maintained higher

OCR for page 57
Metabolism 73 blood selenium levels and excreted less selenium in urine than animals fed higher levels of sulfur (0.11 to 0.2 percent). This is consistent with research by others (White and Somers, 1977), which showed that sheep fed a low- sulfur diet (0.07 percent) had significantly higher plasma and wool sele- nium levels when given selenomethionine. Hence, the alteration of sele- nium metabolism appears to be the greatest with sulfur-deficient animals. INFLUENCE OF ARSENIC The beneficial effect of arsenic on selenium poisoning was noted in the 1930s when 5 ppm of sodium arsenite in the drinking water was found to give full protection against liver damage in rats fed diets containing 15 ppm selenium (Moxon and Rhian, 19431. Several chemical forms of arse- nic, including some organic arsenicals, are effective against selenium tox- icity (Levander, 1977~. Even though arsenic counteracts selenium toxicity, it was found to inhibit expiration of dimethyl selenide. This apparent para- dox was elucidated by data showing that arsenic increases the biliary excre- tion of selenium, resulting in decreased tissue levels (Levander, 1977~. In the reverse situation, selenium has also been shown to increase the biliary excretion of arsenic. Arsenic, however, cannot substitute for the metabolic functions of selenium; all attempts either to promote or prevent selenium deficiency diseases in animals by feeding arsenic have been unsuccessful. INFLUENCE OF OTHER ELEMENTS The toxicity of selenium can be reversed by such metals as copper, mer- cury, and cadmium when given at high dietary levels (Hill, 1974), suggest- ing that metals can influence the metabolism of selenium. Cadmium is equal to or superior to arsenite as an inhibitor of selenium volatilization in rats (Ganther and Bauman, 1962b), providing further evidence that metals can alter selenium metabolism. Some evidence is available to indi- cate that even tellurium will induce selenium and vitamin E deficiency in ducklings (Van Vleet, 1977b). Other elements, such as thallium, silver, lead, and cobalt, have been shown to affect selenium metabolism, as was discussed in a review by Diplock, (1976~. For a more detailed discussion on metal-selenium interaction, see the section, "Nutritional and Metabolic Interrelationships. " EFFECTS OF VITAMIN E Although a metabolic interrelationship between selenium and vitamin E has definitely been established (Hoekstra, 1975), moderate levels of vita- min E do not affect selenium metabolism,. The distribution and excretion

OCR for page 57
74 SELENIUM IN NUTRITION of selenium in young rats depleted of selenium and vitamin E were no dif- ferent from that in animals receiving protective amounts of vitamin E in the diet (Hopkins, 19621. Massive doses of vitamin E, however, apparently decrease selenium volatilization in rats and increase selenium retention in the liver and carcass. Since selenium requirements have been shown to be dependent upon the vitamin E status of animals (Thompson and Scott, 1969), moderate levels of this vitamin must be influencing the effectiveness of selenium. Furthermore, the vitamin E status of animals apparently af- fects the oxidation state of tissue selenium, since the proportion of sele- nium that was acid-volatile was significantly less in vitamin E-deficient rats than in vitamin E-supplemented ones (Diplock et al., 19731. MISCELLANEOUS FACTORS A number of other factors will influence the metabolism of selenium, as discussed in various reviews (Ganther, 1965; Levander, 1972; Diplock, 19761. These include the quantity and nature of protein in diets, previous adaptations of the animal to selenium, and the presence of BAL (2,3-di- mercaptopropanol), penicillamine, sodium dehydrocholate, linseed meal, bromobenzene, and phosphates. Some of the protective factors in linseed meal have been recently identified. These include two newly identified cya- nogenic glycosides, linustatin and neolinustatin, which have been shown to provide significant protection against growth depression due to selenium toxicity (Palmer et al., 19801. These respective compounds were found to be present in linseed meal at concentrations of 0.17 to 0.19 percent. Along similar lines, potassium cyanide dosing to ewes was reported to increase the incidence of white muscle disease in their lambs (Rudert and Lewis, 19781. Moreover, cyanide protects partially against selenium toxicity in rats (Palmer and Olson, 1979~. However, attempts to induce selenium defi- ciency by adding cyanide to water for rats had little effect (Palmer and Olson, 19811. Other examples of the possible influence of dietary compo- nents are comparisons of crude diets (commercial) to purified ones. Two to three times as much volatile selenium was exhaled when 75Se-selenite was injected into rats that had been fed certain crude diets than when it was injected into those that had been fed purified basal diets that permitted good growth rates. Not all crude diets, however, possessed this ability to promote exhalation of volatile selenium (Ganther, 1965~. EXCRETION The primary route of excretion of selenium is in the urine of monogastric animals, regardless of whether it is given orally or injected (Hopkins, 1962;

OCR for page 57
Metabolism 75 Burk et al., 19721; but the urinary route in ruminants is dependent upon the method of administration. When selenium is administered orally to ruminants, most is excreted with feces (Cousins and Cairney, 1961; Peter- son and Spedding, 1963; Paulson et al., 1966; Lopez et al., 19691. In con- trast, when selenium was injected either intravenously or subcutaneously in ruminants, more was excreted in urine than in feces (Wright and Bell, 1966~. There is also an age effect, as indicated by the work of Ewan et al. (1968c). 75 Se given orally to young lambs (8 to 10 weeks of age) fed a syn- thetic liquid diet was recovered to the extent of 66 to 75 percent in urine and 34 to 27 percent in feces. The rumen microbes no doubt contribute to this age effect, since they are present in greatest numbers in fully devel- oped ruminants. The peak of urinary excretion occurs 2 to 3 days after administration of the dose (Mush et al., 1967; Burk et al., 1972~. The size of dose, as well as the selenium status of the animal, influences the speed of excretion and the relative amounts that may be excreted via urine. In lambs (Lopez et al., 1969), calves (Kincaid et al., 1977), and rats (Burk et al., 1972), urinary excretion of selenium was found to be directly proportional to the selenium status of the animals. Essentially all of the selenium excreted in feces of ruminants is in insolu- ble forms, and very little is available for uptake by plants (Cousins and Cairney, 1961; Peterson and Spedding, 19631. The poor availability is evi- dent by data showing that less than 0.3 percent of 75Se was taken up by three common pasture species of plants growing for 75 days on feces from sheep dosed orally with Se-selenite. Trimethyl selenide is the major urinary metabolite of selenium (Byard, 1969; Palmer et al., 19691. This compound is essentially unavailable nutritionally as a source of selenium for the pre- vention of liver necrosis in rats (Tsay et al., 1970), suggesting it to be a detoxification product of selenium. Mortality and survival time were not significantly affected by 0.15 or 1.5 ppm selenium added as trimethyl sel- enide, but 0.15 ppm selenium added as sodium selenite gave complete protection. An injection of 75Se-labeled trimethyl selenite was rapidly ex- creted in unchanged form in the urine (70 percent in the first 12 hours). Excretion of selenium by humans is very similar to patterns for mono- gastric animals. From 43 to 86 percent of an oral dose of selenium appears in the urine of humans (Thomson, 1972; van Rij et al., 1979~. The daily urinary loss for patients was less for those with lower blood and plasma selenium concentrations (van Rij et al., 1979), indicating the body can con- serve this element. In patients given 75Se-selenite, there were only trace amounts of radioactivity expired in air, and no dermal losses were detected (Thomson and Stewart, 1974~. Peak excretion of 75 Se occurred 4 to 5 days after dosing. Subsequent studies from this same laboratory were con

OCR for page 57
76 SELENIUM IN NUTRITION ducted with 75Se-selenomethionine in women (Griffiths et al., 1976~. As with selenite, very few respiratory or dermal losses of selenium occurred after administration of selenomethionine, but the urinary excretion was about half that observed for selenite. Based on balance studies with New Zealand women, the minimum dietary requirement of selenium for main- tenance of normal human health was estimated to be probably not more than 20 ,ug/day (Stewart et al.,1978~.