Selenium-dependent Glutathione Peroxidase Activity Research Articles

Selenium Content and Glutathione Peroxidase Activity in Tissues of the Dairy Cow After Short-Term Feeding1,2

The effect of selenium and vitamin E on concentrations of selenium and selenium-dependent glutathione peroxidase in tissues of dairy cows was studied. Selenium (5 mg/day) and vitamin E (2 g/day) were supplemented for 10 days in a 2×2 factorial arrangement. Selenium supplementation increased content of selenium in whole blood, plasma, ovary, and liver and increased activity of selenium-dependent glutathione peroxidase in liver. Activity of selenium-dependent glutathione peroxidase in follicular fluid was closely correlated with selenium-dependent glutathione peroxidase in plasma. Significant activity of selenium-dependent glutathione peroxidase was detected in luteal tissue of the ovary. Relationships were linear between content of selenium and selenium-dependent glutathione peroxidase in ovary, uterus, and adrenal tissues.

Differential changes of glutathione S-transferase activity by dietary selenium

Dietary selenium deficiency produced increased activity of the glutathione S-transferases in the liver, kidney and duodenal mucosa. In these tissues, the residual activity of total glutathione peroxidase that included selenium-independent activity was considerably higher than that of selenium-dependent glutathione peroxidase. The enhanced activity of glutathione S-transferases was restored to control level 48 hr after an injection of selenite equivalent to the amount of daily selenium intake. Under the same conditions, selenium-dependent glutathione peroxidase activity increased with time and reached 11.9, 11.6 and 46.2% of the activity in the liver, kidney and duodenal mucosa of selenium-supplemented rats, respectively, 48 hr after selenite injection, whereas total glutathione peroxidase activity was not altered except in the kidney. These differential changes of glutathione S-transferase activity were intimately related to those of selenium-dependent glutathione peroxidase activity produced by selenium depletion and repletion, suggesting that the glutathione S-transferase activity was regulated by dietary selenium. Present findings support the idea that glutathione S-transferases having selenium-independent glutathione peroxidase activity function as a substitute for selenium-dependent glutathione peroxidase in selenium-deficient rats.

Differential distribution of glutathione and glutathione-related enzymes in rabbit kidney: Possible implications in analgesic nephropathy

Whole tissue reduced glutathione (GSH) concentration was found to be lowest in rabbit renal inner medulla and progressively higher in outer medulla and cortex. Activities of cytosolic glutathione reductase in inner medulla and outer medulla were similar, and each was only approximately 50% of that of cortex. Whole tissue and microsomal γ-glutamyl transpeptidase activities were high in cortex and outer medulla but were low in inner medulla. Cytosolic activity of selenium-dependent glutathione peroxidase (GPx-I) was similar in both outer medulla and inner medulla but was only 50% of that of cortex. Activity of cytosolic selenium-independent glutathione peroxidase (GPx-II) was highest in cortex and lowest in inner medulla (∼15% of cortex and ∼50% of outer medulla). Cytosolic glutathione S-transferase activity with l-chloro-2,4-dinitrobenzene as substrate was high in all three regions of kidney. With 1,2-dichloro-4-nitrobenzene and 1,2-epoxy-(4-nitrophenoxy)propane as substrates, cytosolic glutathione S-transferase activities were very low in cortex, outer medulla, and inner medulla. Microsomal activities of glutathione reductase, GPx-I, GPx-II and glutathione S-transferases were much lower than activities of corresponding cytosolic enzymes. Activities of the glutathione peroxidases in renal inner medulla would hence be expected to cause little interference to prostaglandin endoperoxide synthetase mediated cooxidative activation of paracetamol. It has been demonstrated that the paracetamol metabolite can react rapidly with GSH, forming not only glutathione conjugate but also paracetamol itself and oxidized glutathione. Low GSH concentrations, as well as low activities of glutathione reductase, GPx-I, GPx-II, and γ-glutamyl transpeptidase, may therefore render the inner medullary region of kidney particularly vulnerable to paracetamol-related analgesic nephropathy.

Dietary selenium deficiency causes decreased N-oxygenation of N, N-dimethylaniline and increased mutagenicity of diemthylnitrosamine in the isolated rat liver/cell culture system

Male Wistar rats were fed diets of varying selenium content in order to obtaine selenium-deficient and selenium-supplemented rats. After 5–6 weeks on the respective diet, the rats were used to investigate how selenium influences the effects of dimethylnitrosamine (DMN) on some lvier enzymes and related reactions. The selenium-dependent glutathione peroxidase activity in postmicrosomal supernatant from liver was about 1% in selenium-deficient rats as compared to selenium-supplemented rats or rats fed a standard diet. The highest DMN-demethylase activity was observed in postmitochondrial supernatant from selenium-deficient rat lvier, and the lowest in selenium-supplemented rats. No dietary effect was observed on hepatic microsomal cytochrome P450 levels. C-oxygenation of N, N-dimethylaniline (DMA) was not affected by the selenium level. On the other hand, selenium deficiency seemed to reduce N-oxygenation of DMA. The mutagenicity of DMN in Chinease hamster V79 cells after metabolic activation by the isolated perfused rat liver, was approximately doubled when selenium-deficient livers were used as compared to selenium-supplemented livers and livers from rats fed a standard diet. A negative correlation between DMA- N-oxygenation and mutagenicity from DMN was observed, whereas no correlation between DMA- C-oxygenation and mutagenicity from DMN was found.

Dietary Selenium Requirement of Fingerling Channel Catfish

Two experiments were conducted in aquaria to determine the minimum dietary selenium requirement of fingerling channel catfish (Ictalurus punctatus). Casein-gelatin diets containing graded levels of supplemental selenium (as Na2SeO3) ranging from 0 to 15 mg/kg were fed to catfish for 15 weeks in experiment 1 to broadly define their selenium requirement and toxicity levels. Although growth of catfish was affected by dietary selenium level, significant differences in weight gain were not easily discernible due to variability among the groups of fish. Weight gain data generally indicated that the basal diet containing 0.06 mg Se/kg diet caused growth depression, and a supplemental selenium level of 15 mg/kg also caused a reduced growth response, which indicated selenium toxicity. Selenium concentrations in edible muscle tissue increased almost linearly with increasing dietary selenium levels. Liver and plasma selenium-dependent glutathione peroxidase (Se GSH-Px) activities indicated the selenium requirement of fingerling channel catfish was between 0.1 and 0.5 mg Se/kg diet. In experiment 2, casein-gelatin diets containing incremental levels of supplemental selenium were fed to catfish for 14 weeks to more precisely determine their minimum dietary selenium requirement. Growth data and liver and plasma Se GSH-Px activities indicated that the minimum selenium requirement of fingerling channel catfish fed adequate vitamin E was 0.25 mg Se/kg dry diet. Based on these data, it appears that selenium supplementation of commercial catfish feeds is warranted.

Sex Differences in the Glutathione Peroxidase Activity of Various Tissues of the Rat

The glutathione peroxidase activity of blood, liver and brain of sexually mature male and female rats has been determined using both hydrogen peroxide and cumene hydroperoxide as substrates. No significant differences were found on comparison of the glutathione peroxidase activity of the brain tissues of male and female animals. Observed mathematically significant differences between the enzyme activity in the blood of male and females were probably of little physiological significance. Both selenium and non-selenium dependent glutathione peroxidase activities were significantly higher in the livers of female rats but this was much more apparent when cumene hydroperoxide was reduced. Castrated male rats had significantly higher hepatic enzyme activities, approaching those of the female animals. The lipoperoxide concentration was significantly higher in the livers of female rats.

Selenium and Mitochondrial Integrity in the Pancreas of the Chick

The activity of selenium-dependent glutathione peroxidase (SeGSHPx) in the pancreas was found to decline in response to Se depletion in the chick at approximately the same time as the appearance of histopathological changes in the acinar cells of the pancreas, indicative of the onset of nutritional pancreatic atrophy. Oral administration of 10 µg of Se as Na2SeO3 to Se-deficient chicks increased SeGSHPx activity in liver and plasma within 12 hours but not in pancreas until 36 hours. Histological sings of regeneration were not observed within 48 hours of Se repletion; however, acinar nuclei were significantly larger within 12 hours of Se repletion. The apparent rates of synthesis of DNA, RNA, and protein in liver, and of DNA in pancreas, were not affected by Se deficiency. The apparent rates of synthesis of pancreatic RNA and protein were depressed in Se-deficient chicks and were increased within 24 and 36 hours, respectively, upon Se repletion. Electron microscopic observation of pancreatic acinar cells from Se-deficient chicks revealed disruption of inner and outer mitochondrial membranes. However, no impairment was noted in rate of oxygen uptake, respiratory control index, or ADP:oxygen ratio of pancreatic mitochondria due to Se deficiency. The mitochondrial protein yield of the Se-deficient chick pancreases was 27% less than that of Se-adequate chicks. These data suggest that a decreased number of functional acinar mitochondria may result in nutritional pancreatic atrophy in the Se-deficient chick.

Selenium-dependent and non-selenium-dependent glutathione peroxidases in human tissue extracts

A method for the assessment of both selenium-dependent and non-selenium-dependent glutathione peroxidases on crude tissue extracts in human is described. The enzyme activity is measured by the coupled assay system in which oxidation of reduced glutathione (GSH) is coupled to NADPH oxidation catalyzed by glutathione reductase. Total glutathione peroxidase activity is measure with cumene hydroperoxide as substrate. Selenium-dependent glutathione peroxidase is measured with tert-butyl hydroperoxide. This substrate is preferable to H 2O 2 which gives too high blank values compared to the assay values. The difference between total glutathione peroxidase and selenium-dependent glutathione peroxidase activities represents the non-selenium-dependent glutathione peroxidase activity. Studies of substrate specificity of the two glutathione peroxidases separated by gel filtration as well as linearity and recovery studies are presented. For a given tissue, the relative amounts of the two glutathione peroxidases given by our assay are identical to those estimated by quantifying the elution peaks after gel filtration. Based on the percentages of the two glutathione peroxidases, human tissues can be classified in four groups: (1) the non-selenium-dependent glutathione peroxidase is predominant in liver, in renal cortex and skeletal muscle; (2) non-selenium-dependent and selenium-dependent glutathione peroxidases are in equal amounts in renal medulla; (3) the selenium-dependent glutathione peroxidase is predominant in adrenal glands and platelets: (4) the selenium-dependent glutathione peroxidase represents 100% of the glutathione peroxidase activity in the other organs. The heart and the brain are of special interest in this group because of the physiological role and the regulation of the selenoenzyme.

Effect of Selenium on the Subcellular Distribution of Glutathione Peroxidase in Rat Liver, Epididymal Fat Pad and Seminal Vesicle

The effect of dietary selenium on the subcellular distribution of selenium-dependent glutathione peroxidase (SeGSH-Px) activity in rat liver, epididymal fat pad, and seminal vesicle was determined. Tissues were fractionated by differential centrifugation, and the subcellular distribution of SeGSH-Px activity was determined by comparison with the distribution of biochemical marker enzymes. Liver SeGSH-Px activity was located in both the cytosol and mitochondria. In epididymal fat pad and seminal vesicle, SeGSH-Px activity was located primarily in the cytosol; association with another subcellular organelle, however, was indicated. In liver and epididymal fat pad, SeGSH-Px activity increased linearly, and, in the seminal vesicle, increased linearly and quadratically, with increasing dietary selenium concentration. Distribution of SeGSH-Px activity among the cellular fractions from the tissues, however, was not affected by dietary selenium supplementation.

Effects of riboflavin and selenium deficiencies on glutathione and its relating enzyme activities with respect to lipid peroxide content of rat livers.

Weanling male rats were fed a riboflavin- or selenium-deficient diet for 5 weeks, and the glutathione content and its relating enzyme activities in the livers were examined. The glutathione content and glutathione reductase activity were decreased by deficiency of riboflavin, but not by that of selenium. Glutathione peroxidase activity was increased by addition of selenium to the diet, but without its addition, the activity was higher in the riboflavin-deficient rats than in the riboflavin-sufficient rats, in spite of the increase of lipid peroxides in the former rats. Glucose 6-phosphate dehydrogenase activity was decreased significantly by riboflavin deficiency. In another experiment, riboflavin was intraperitoneally injected into rats fed the diet to which neither riboflavin nor selenium had been added. The glutathione content and activities of glutathione reductase and glucose 6-phosphate dehydrogenase returned to the control level of riboflavin-sufficient rats in 24 h, the lipid peroxide level in 48 h, and the glutathione peroxidase activity, being higher than that in the control rats, in 72 h after the injection, respectively. These findings indicate that the increase of lipid peroxides in the livers of riboflavin-deficient rats is caused by the decrease in the glutathione content as well as glutathione reductase activity rather than by that in the selenium-dependent glutathione peroxidase activity.

Selenium and glutathione peroxidase activity in blood of the nutria ( Myocastor coypus): Comparison with guinea-pig, rat, rabbit and some non-mammalian vertebrates

1. 1. Plasma selenium and selenium-dependent glutathione peroxidase (GSH-Px) activity in plasma and erythrocytes were measured in the nutria ( Myocastor coypus) and the alligator ( Alligator mississippiensis). Plasma and erythrocyte GSH-Px activity only were measured in the rat, rabbit, chicken, quail, turtle ( Chrysemys picta), snake ( Nerodia fasciata), bullfrog ( Rana catesbiana) and catfish ( Ictalurus punctatus). 2. 2. Plasma GSH-Px activity ranged from a low of 0.021 EU/ml in the bullfrog to a high of 1.97 EU/ml in the alligator. 3. 3. Erythrocyte GSH-Px activity was lower in the nutria (2.8 EU/g Hb) than all other species studied. Whereas the rat had the highest activity (539 EU/g Hb). The variation among the mammalian species studied was greater than between the different orders of vertebrates. There was no relationship between plasma and erythrocyte GSH-Px activity and no phylogenetic pattern in the differences in activity between species. 4. 4. There was a relationship between plasma selenium concentration and enzyme activity: animals with the lowest enzyme activity had lowest selenium levels.

Glutathione S-transferases of the bovine retina. Evidence that glutathione peroxidase activity is the result of glutathione S-transferase.

We have purified two isoenzymes of glutathione S-transferase from bovine retina to apparent homogeneity through a combination of gel-filtration chromatography, affinity chromatography and isoelectric focusing. The more anionic (pI = 6.34) and less anionic (pI = 6.87) isoenzymes were comparable with respect to kinetic and structural parameters. The Km for both substrates, reduced glutathione and 1-chloro-2,4-dinitrobenzene, bilirubin inhibition of glutathione conjugation to 1-chloro-2,4-dinitrobenzene, 1-chloro-2,4-dinitrobenzene inactivation of enzyme activity and molecular weight were similar. However, pH optimum and energy of activation were found to differ considerably. Retina was found to have no selenium-dependent glutathione peroxidase activity. The total glutathione peroxidase activity fractionated with the transferases in the gel-filtration range of mol.wt. 49000 and expressed activity with only organic hydroperoxides as substrate. Only the more anionic isoenzyme expressed both transferase and peroxidase activity.

Glutathione peroxidase activities of animal tissues

1. 1. Selenium-dependent, selenium-independent and total glutathione peroxidase activities were measured in liver, kidney, heart, lung, intestine and stomach of a variety of animals, mostly mammals. 2. 2. Total glutathione peroxidase activity of liver was highest in the order: hamster > gerbil ≈ rabbit > mouse > rat. Total glutathione peroxidase activity was adequate in the livers of other mammals and in the kidney, heart, lung, intestine and stomach of all mammals. 3. 3. The total glutathione peroxidase activities in the following tissues correlated significantly: liver and kidney; heart and lung; and intestine and stomach. Selenium-dependent glutathione peroxidase activities in rat tissues correlated with reported activities of copper-superoxide dismutase in the same tissues. 4. 4. Considering the phylogenetic distribution of total glutathione peroxidase activities, the rodent limb of the phylogenetic tree has the highest activities.

Relationship of Selenium-Dependent Glutathione Peroxidase Activity and Nutritional Pancreatic Atrophy in Selenium-Deficient Chicks

Experiments were conducted to determine the relationship of selenium-dependent glutathione peroxidase (SeGSHpx) activity to the onset of nutritional pancreatic atrophy (NPA) in chicks. Chicks produced from hens fed a low Se, low vitamin E practical diet had low activities of SeGSHpx in plasma and pancreas at hatching. The SeGSHpx activity remained at a low level before and during all stages of the onset of NPA. Selenium supplementation of the diet prevented NPA and resulted in significant elevations in SeGSHpx activity. Although the early stages of NPA are believed to involve mitochondrial swelling, no significant differences were found in rate of oxygen uptake, respiratory control index or adenosine diphosphate to oxygen ratio between pancreatic mitochondria isolated from Se-deficient or Se-adequate chicks. Improvements in maternal Se status significantly increased chick pancreatic SeGSHpx activity at hatching and delayed the onset of NPA in chicks fed a Se-deficient diet. A significant proportion of the second-generation Se- and vitamin E-depleted chicks used in these studies was found to grow at nearly normal rates when fed the Se-deficient diet. These chicks, designated as refractory to the growth-depressing effect of severe Se-deficiency, were biochemically deficient (i.e., had very low SeGSHpx activities) and showed NPA. It is concluded that the Se-responsive lesion which results in NPA in the chick is different from that which results in depressed growth.selenium glutathione pancreatic atrophy

Drug-induced lipid peroxidation in mice—I Modulation by monooxegenase activity, glutathione and selenium status

Feeding male mice for 2 days with sucrose leads to a decrease of total liver glutathione by more than 50 per cent when compared to controls. Such animals were intoxicated with 300 mg/kg paracetamol and upon administration of inducers of the drug-metabolizing system, in vivo and vitro lipid peroxidation in these animals was largely increased as well as the susceptibility to the drug. Pretreatment of the mice with methylcholanthrene led to a 28-fold, with benzo(α)pyrene to a 22-fold, and with phenobarbital to a tenfold increase in ethane exhalation. In vivo administration of various monooxygenase inhibitors showed that all agents effectively inhibit paracetamol-induced lipid peroxidation. It is concluded that phase I metabolism of paracetamol is a prerequisite for the manifestation of drug-induced lipid peroxidation. Selenium deficiency in mice neither affected hepatic levels of glutathione nor its decrease following sucrose feeding, nor glutathione transferase, superoxide dismutase, catalase and glutathione reductase activity. Selenium-dependent glutathione peroxidase activity of selenium-deficient mice, reactive with H 2O 2 as well as with t-butylhydroperoxide, decreased to 5 per cent of the supplemented controls. A glutathione peroxidase activity, which utilized cumenehydroperoxide as a substrate but insensitive to selenium deficiency, was found. Selenium-deficient diethylmaleate-pretreated animals were much more susceptible to paracetamol-induced lipid peroxidation than controls. Supplemented diethylmaleate-pretreated animals showed no signs of lipid peroxidation if treated with 100 mg/kg aminopyrine or ethylmorphine. However, deficient animals exhibited high ethane exhalation rates, drastically increased serum transaminases, loss of hepatic glutathione and mortality upon administration of these drugs. Qualitatively similar results with lower ethane exhalation rates were observed when 125 mg/kg furosemide was administered to diethylmaleate-pretreated selenium-deficient or -adequate mice. Even administration of 200 mg/kg ethoxycoumarin in combination with diethylmaleate lead to significant lipid peroxidation in phenobarbital-induced mice. The results demonstrate that in vivo selenium-dependent glutathione peroxidase plays a predominant role within the glutathione redox couple system. The enzyme protects the liver from peroxidative damage evoked by phase I metabolism of various drug types, as long as sufficient glutathione is available. It is suggested that activated oxygen released from the microsomal monooxygenase is the species responsible for the observed lipid peroxidation accompanied by severe acute liver lesions.

Evidence for the occurrence of selenium-independent glutathione peroxidase activity in rat liver microsomes

Rat liver microsomes exhibit selenium-independent glutathione peroxidase activity which is associated with glutathione S-transferase activity. The peroxidase activity is not due to contamination with either soluble selenium-dependent or selenium-independent glutathione peroxidase activities of the cytosol. N-Ethylmaleimide treatment which stimulates rat liver microsomal glutathione transferase activity concomitantly stimulates the glutathione peroxidase activity. In contrast, N-ethylmaleimide depresses both enzyme activities of the cytosol. A protein exhibiting both glutathione peroxidase and glutathione transferase activity was isolated from the microsomes and purified to homogeneity by DEAE cellulose ion-exchange and S-hexylglutathione Sepharose 6B affinity chromatography.

Effect of Selenium on Appetite in the Selenium-Deficient Chick

Experiments were conducted to determine the initial effects of oral selenium administration on selenium-deficient chicks. Administration of 5 microgram selenium as seleno-DL-methionine increased voluntary feed consumption within 2-3 hours, whereas selenite did not have a significant effect until 3-4 hours. Spontaneous activity, body weight gain and plasma glucose concentration increased 6-8 hours after selenium administration. The earliest response in the specific activity of selenium-dependent glutathione peroxidase occurred in plasma at 8 hours and in liver at 24 hours after selenium administration. The onset of pancreatic atrophy, however, was not affected by the level of feed intake suggesting that the effect of selenium upon appetite may be distinct from the involvement of selenium in nutritional pancreatic atrophy and fibrosis.

Rat hepatic cytosolic glutathione-dependent enzyme protection against lipid peroxidation in the nadph-microsomal lipid peroxidation system

Dialyzed rat liver cytosol (105 000 × g supernatant), when added along with 2.5 mM glutathione, blocked malonaldehyde formation in the NADPH-microsomal lipid peroxidation system, thus protecting against lipid peroxidation. Preheating the cytosol for 10 min at 60° C destroyed its protective ability. Ammonium sulfate fractionation and Sephadex G-100 gel filtration of cytosol indicated that more than one glutathione-dependent protective enzyme was present. Fractions from the G-100 column containing the selenoenzyme glutathione peroxidase failed to protect, but fractions containing the glutathione S-transferases, which have non selenium-dependent glutathione peroxidase activity, did protect. The glutathione S-transferases were purified further with ion exchange chromatography and shown to have protective activity. Thus the rat hepatic cytosolic glutathione-dependent enzyme protection against lipid peroxidation in the NADPH-microsomal lipid peroxidation system is in part due to some of the glutathione S-transferases. The selenium-dependent glutathione peroxidase appears to have no protective effect in this system.

Influence of Selenium, Vitamin E, and Ethoxyquin on Lipid Peroxidation in Muscle Tissues from Fowl during Low Temperature Storage

The influences of various factors which affect the selenium-vitamin E status of laying hens on lipid peroxidation in muscle tissues during low temperature storage were studied. Laying hens from 32 to 56 weeks of age were fed low selenium and low vitamin E practical diets supplemented with different levels of Na2SeO3, d1-α-tocopherol acetate, ethoxyquin, and/or peroxidized corn oil. Vitamin E status as indicated by plasma vitamin E activity was improved by supplements of vitamin E, selenium, or ethoxyquin. Selenium status as indicated by plasma selenium-dependent glutathione peroxidase activity was improved by selenium supplementation. Incorporation of peroxidized corn oil into diets did not depress plasma vitamin E but increased plasma glutathione peroxidase when those diets contained supplemental selenium. Lipid peroxidation as indicated by the 2-thiobarbituric acid (TBA) method in muscle samples held at −20 C for up to 270 days was reduced by dietary selenium in M. pectoralis when corn oil was fed. Supplemental vitamin E or ethoxyquin reduced TBA values developed in M. gastrocnemius. Results indicate that dietary selenium and other factors affecting selenium status may be useful in retarding the development of oxidative rancidity in frozen poultry products.

Sex differences in biochemical manifestations of selenium deficiency in rat liver with special reference to heme metabolism

In addition to its well-known effect on glutathione peroxidase, selenium deficiency causes: (1) a defect in hepatic heme metabolism characterized by a phenobarbital-mediated increase in microsomal heme oxygenase activity, and (2) an increase in hepatic glutathione S-transferase activity. Since these effects were reported in selenium-deficient male rats, and since female rats have a lower selenium requirement than males, we examined whether these effects were sex-dependent. Weanling male rats, female rats, castrated male rats, and testosterone-treated female rats were fed either a selenium-deficient or a control diet. After 8 weeks, selenium-dependent hepatic glutathione peroxidase activity was 1 per cent of respective control values in each of the selenium-deficient groups. Hepatic glutathione, S-transferase activity was doubled by selenium deficiency in normal, unoperated males but was unaffected in the other groups. In control diet fed rats phenobarbital given as a single injection caused either no significant change or a decrease in the activity of hepatic microsomal heme oxygenase, the rate-limiting enzyme in heme degradation. In contrast, microsomal heme oxygenase activity was stimulated by phenobarbital in all selenium-deficient rats. The stimulation was greatest in males and least in females with intermediate values in castrated males and testosterone-treated females. These results demonstrate a marked effect of sex, castration of males, and testosterone treatment of females on the response of hepatic heme metabolism to phenobarbital and on glutathione S-transferase activity in selenium deficiency even though glutathione peroxidase was reduced to the same extent by selenium deficiency in all groups.