Blood

selenium

and glutathione in New Dakota1

C. D. THOMSON,1

peroxidase Zealand,
AND M. HOWEt
of

activity Oregon,

of populations and South

M. A. BEILSTEIN,
57702, USA

P. D. WHANGER,

M. F. ROBINSON,1

Department of Agricultural Chemistry, Oregon State University, Human Nutrition, University of Otago, Dunedin, New Zealand South Dakota

Corvallis, OR 97330, USA; tDepartment and tRoute 4, Box 1660, Rapid City,

ABSTRACT The relationshipof whole blood selenium glutathioneperoxidase(GPX) activity was

for individuals who in New Zealand, Oregon, Dakota represented, respectively,

(Se)
examined and

to

South

populations

with exposure to low, medium, and high amounts of Se. The mean (respective) blood Se levels were 60, 200, and 400 ng/ml. Intergroup differences in blood Se levels were highly significant (P < 0.001). GPX assays
were coupled performed procedure using two variations the of an of enzyme-

the two methods. Despite a fourfolddifference n absoluteaci tivities measured by thesemethods, the GPX activities were highly correlated (r .86) between procedures.
to assess equivalence

Average
(P <

blood GPX

activity was significantlylower

0.001) for the New Zealand group compared with the othertwo groups,but therewas no difference in GPX activitiesetween the Oregon and South b Dakota groups. Linear regression of GPX vs. Se values
within these each group indicated only data from in a significant correlation Zealand also group showed a parameters of parameters

the

New groups
=

(r
for

.46, P < 0.01).

combined

Comparison all three

of these

significant positive orrelation c (r .60, P < 0.001). A saturation model (ln GPX k,+k2 (Se)1] fits the combined data better(r .80, P < 0.01)than does
= =

direct

comparison

of the two parameters.

These

results

suggestthat GPX activity is an appropriate indicator ofhuman Se status only in populations with below normal exposure to Se, as activity of this enzyme is saturated at relatively low levels. WHANGER, P. D.; BElLSTEIN, M. A; ThOMsON, C. D.; ROBINSON, M. F.; HOWE, M. Blood selenium and glutathione peroxidase activity of populations in New Zealand, Oregon, and South Dakota. FASEB]. 2: 2996-3002; 1988.
Key Wonis: glutathione peroxidase selenium bloodS Wendel procedure hemoglobin concentration . P and V method sample size

SELENIUM (SE) AS AN INTEGRAL PART of the enzyme glutathione peroxidase (GPX) (EC 1.11.1.9) is the only identified function of Se in animals (2). Therefore blood GPX activity has been widely used as an indicator of Se status in both animals and humans. Although blood GPX activity has been shown in many studies of animals to closelyreflectSe status (as indicated by its levels in the blood) (3-8), the relationship of blood Sc and GPX activity in humans is unclear. Several studies of human blood GPX activityhave demonstrated a positivecorrelation with blood Se levels(Table 1). A greater number of other studies,however, have either failed to detect such a relationship or have demonstrated a negative correlation (Table 2). Several studies that have demonstrated a positive relationship of Se and GPX in humans were conducted either with individuals in geographic areas of low Se (9-11) or with patients who were depleted of Se because of health or nutritional problems (10, 14, 15). The failureto detect such a positiverelationshipin humans with higher Se status may result from a lack of correlation, or the range of values may be too small to detect such a relationship. In this study, Se levels and GPX activities were determined from blood collected from New Zealanders, Oregonians, and South Dakotans who represented populations with exposure to low, medium, and high amounts of Se, respectively. Blood samples from these three regions provided a range of Se concentrations of greater than sixfold between high and low groups. Another objective of this study was to compare two GPX assay procedures over a wide range of activities. Since GPX activity cannot be assayed by saturating substrate concentrations (21), measured activity is dependent on assay conditions. Although most current GPX assays are based on the enzyme-linked procedure of Paglia and Valentine (22), a standard assay for GPX does not exist, and substrate concentrations usually vary between laboratories. Comparisons of data among

Published

with

the approval

of Oregon

State

Agricultural

Ex-

periment Station as Technical Paper No. 8490. A preliminary report of these results has been presented (see ref 1).

2996

0892-6638/88/0002-2996/$0l.50.

FASEB

TABLE

1. Studies

supporting

a positive

relation of Se and GPX in human blood Regression analysis P < < < < < < < < < 0.001 0.001 0.01 0.001 0.001 0.001 0.01 0.001 0.001 References 9 9 9 10 11 12 13 14 15

Study New New

site

Sample
264b

size

WE Se range, 20-220 20-100 80-220

ppb

r .71 .74 .27

Zealand Zealand

New Zealand
New Zealand New Zealand New York

222k 1326 19

107a 66
100

30-70 20-100 130-440

120-600 16-120

.98 .62 .86

.67 .51

Texas Texas
Boston

io
186

20-190

.76

If reported differently, converted to whole blood (WB) basis, assuming: 60% WB Se in RECs, hematocrit = 45%, hemoglobin = 34% RBC wt, WB dry wt - 21% wet wt. #{176}WB vs. WE Se, primarily normal controls. GPX RBC GPX vs. RBC Sc, multiple samples from 4 total parenteral nutrition (TPN) patients during Se depletion. dWB GPX vs. WB Se, infants and children. RBC GPX vs. RBC Se, 23 maternal samples at time of delivery, 23 paired-cord blood samples, and 20 nonpregnant adult females. 1RBC GPX vs. RBC Se, university students. RBC GPX vs. plasma Se, multiple samples from 7 TPN patients during Se depletion. !, RBC GPX vs. plasma Se, 8 unsupplemented TPN patients, and 10 healthy
controls.

laboratories are therefore difficult and sometimes impossible. All samples in this study were assayed either by a method published by Wendel (23) or by the method regularly used in the Oregon State University (OSU) laboratory (24), which is a modification of the Paglia and Valentine method (22) and hereafter will be referred to as the modified P and V method. Its purpose was to determine whether the activities measured by the two procedures were linearly related over a wide range of activities. PROCEDURES Subject selection

mineral-vitamin supplementation. All were in good health and exhibited no deficiency or toxicity. Sample preparation

participants
signs of

Se

Because of the historically established differences in Se exposure that the regions represent,participantsin the study were recruited from three geographic areas: the country of New Zealand and the states of Oregon and South Dakota. Sampling was not controlled for age, sex, duration of residence in the study area, diet, or TABLE
Study

Blood samples taken from individuals in New Zealand and South Dakota were drawn into heparinized tubes and immediately frozen in liquid nitrogen for shipment on dry ice. On arrival at OSU, samples were transferred to a - 60#{176}C freezer. Blood samples drawn from Oregon residents were also frozen in liquid nitrogen and assayed shortly afterward on thawing. When thawed, blood samples were diluted 1:6(wt/vol) in water at 4#{176}C. 0.10-ml aliquot of each dilution A was further diluted with 0.90 ml water and 1.0 ml doublestrength Drabkins diluent (23) for determination of the hemoglobin concentration by the absorbance at 540 nm (25). Initial dilutions were maintained at 4#{176}C until enzyme assays were performed within a few hours of thawing.

2. Studies supporting no relation or an inverse relation of Se and GPX in human blood site Sample 106
42d

size

WB Se range, 40-105 100-220

Ppb

Regression

analysis,

References
16 9

Germany New Zealand

.11, ns .17, ns

San Diego
Germany New Mexico

10
541 335

143-162
90-130 140-250 No

.11,ns
relationship - .19, ns - .49, ns

17
18 19 13 14 20 15 15

Texas

356

0-100

Texas
Oregon Boston

70
110

16-120
70-120

Negative correlation
- .53,

8
86 16 10 differently, converted as in Table 1.
controls.

20-90
30-170 20-170 130-190
bRBC GPX vs. REC Se, normal controls.

Boston Boston Boston 1f reported

P < 0.05 .05, ns - .05, ns .29, ns .40, ns

ns, slope not significantly different RBC

15 15
from 0.
GPX

dWE GPX vs. WE Se, normal

WB GPX vs. WB Se, includes two subjects takingdailysupplements of 200 and 450 ,g Se.

vs. RBC Se, 37 expectant mothers and 17 controls. wB GPX vs. WE Se, high Sc exposure via drinking water. 6RBC GPX vs. RBC Se, multiple samples from 7 TPN patients during Se depletion. RBC GPX vs. REC Se, monthly samples from 13 expectant mothers through pregnancy and delivery. RBC GPX vs. plasma Se, 8 unsupplemented TPN patients. 6RBC GPX vs. plasma Se, 8 TPN patientsafter31 to 117 days supplementation at 100 sg Se per day as seleniousacid. REC GPX vs.plasma Se, pre- and postsupplementationTPN patients. RBC GPX vs. plasma Se, controls for TPN patient study.

BLOOD

SELENIUM

AND

GLUTATHIONE

PEROXIDASE

2997

Selenium

assays

GPX

assays:

modified

P and

V method

(24)

Aliquots of 0.20 ml of the initial dilutions were combined with 0.20 ml double-strength Drabkins diluent (23) plus 2.40 ml water. Aliquots of 0.1 ml of these dilutions were combined with 0.8 ml of reaction mixture in 1.5-ml glass microcuvettes preequiibrated at 37#{176}C. The reaction mixture contained 0.84 EU/mi of glutathione reductase, 6.25 mM reduced glutathione, and Glutathione peroxidase assays: Wendel procedure 0.31 mM f3-NADPH in pH 7.0, 0.125-M phosphate buffer. After 6 mm preincubation at 37#{176}C, 0.10 ml of Aliquots of the initial 1:6 whole blood dilutions were 25-mM t-butylhydroperoxide in water was added to the diluted in water to give final hemoglobin concentracuvettes. The assay mixtures were mixed in the cuvettes tions of 3.0 mg/mI. Aliquots of 0.50 ml of the 3.0 mg and immediately placed in the 37#{176}C water-jacketed hemoglobin/mi dilutions were combined with 0.2 ml of automatic recording spectrophotometer to measure the Wendels transformation solution at room temperature. rate of change of 340-nm absorbance. Absorbance The transformation solution consisted of 0.45 mM changes were measured over 2 mm. Blank assays were K3Fe(CN)s and 4.5 mM KCN in pH 7.0, 0.125 M phosperformed with water containing Drabkins diluent at phate buffer. After 5 mm incubation at room temperathe same dilutionas the samples. After subtracting the ture, 0.50-ml aliquots of sample in transformation solublank rate, the sample rates were multiplied by 1608 tion were transferred to 1.0-mi glass microcuvettes (24) to yield enzyme activity as nmol fi-NADPH containing 0.40 ml of reaction mixture preequiibrated oxidized per min/ml of dilute sample. These values at 37#{176}C. The reaction mixture contained 1.5 enzyme by the hemoglobin concentrations of the units (EU)/ml glutathione reductase (EC 1.6.4.2)solated were divided i dilutions to yield enzyme activity as nmol j3-NADPH from yeast, 2.5 mM reduced glutathione, and 0.63 mM oxidized per mm/mg hemoglobin. f3-NADPH in pH 7.0, 0.125 M phosphate buffer. After Reduced glutathione, glutathione reductase, and j310 mm preincubation at 37#{176}C, ml of 12-mM t-butyl0.10 NADPH were obtained from Sigma Chemical Co. (St. hydroperoxide in water was added to the mixtures in the Louis, Mo.) and t-butylhydroperoxide was obtained cuvettes. The assay mixtures were mixed in the cuvettes from Pfaltz and Bauer, Inc. (Waterbury, Conn.). by covering with paraflim (American Can Co., Greenwich, Conn.) and inverting several times. The cuvettes were placed in an automatic recording spectrophotomRESULTS eter with 37#{176}C water-jacketed cuvette holders (Hitachi 100-80A, Nissei Sanyo America, Ltd., Mountain View, The average Se concentrations and GPX activities (Table 3) were significantly lower (P < 0.001) for the Calif.), and the rate of change in 340-nm absorbance New Zealand samples compared with the Oregon and was measured over 2 mm. Blank assays were performed South Dakota samples. Despite an overlap in the ranges with water diluted in transformation solution and reaction mixture at the same concentration as the samples. of Se values, the Se concentrations from the Oregon After subtracting the blank rate,the sample rates(opti- samples were significantlyower (P < 0.001) than those l of the South Dakota group. These differences were cal density units/mm) were multiplied by 160.8 to give found regardilessof normalization of Sc values (ng enzyme activities as nmol 13-NADPH oxidized per Se/mg hemoglobin or ng Se/ml whole blood). The sigmm/mg hemoglobin. Duplicate 5.0-mi aliquots were taken from the initial 1:6 whole blood dilutions for fluorometric Se analysis after wet digestion with nitric and perchloric acids (26). Se concentrations of samples were normalized to both ml of whole blood and mg hemoglobin.

TABLE 3. Concentrations of whole blood selenium and glulathione

peroxidase

by geographic region
Glutathione peroxidase

Selenium Population New Zealand Sample size 34 ng/mg hemoglobin ng/ml whole blood 59 11 (41=101) 202 25 (155-278) 397 128 (221=640) 192 154 41-640 Wendel assay, units/mg hemoglobin 6.2 1.2 (3.9-8.8) 7.7 1.7 (4.9-10.7) 7.6 1.8 (4.6-11.3) 7.1 1.7 Modified P&V assay, units/mg hemoglobin 26.9 4.9 (17.5-35.2) 40.0 6.1 (27.2-49.0)

0.45 0.8 (0.32_0.66)b 1.46 0.20 (1.15-1.86)

Oregon

23

South

Dakota

21
78

2.78

0.97 (1.50-5.47)

40.8

(28.7-51.3)

6.6

All
Values

1.37 1.09 0.32-5.47 P < 0.001.

34.5 8.8

3.9-11.3

17.5-51.3

are

standard

deviation,

6Values in parenthesesare ranges.

2998

Vol. 2

Nov. 1988

The FASEBJournal

WHANGER

ET AL.

nificantly higher Se concentration of the South Dakota blood samples compared with the Oregon samples did not result in higher GPX activity by either assay method. The results from the two GPX assay procedures were highly correlated. The correlation coefficients were, respectively,0.94, 0.86, 0.96, and 0.86 for the New Zealand, Oregon, and South Dakota blood samples, and when all were combined. For all of the samples together, a slope of 4.54 was obtained for the regression line. As can be seen from Table 3, the use of the modified P and V method resulted in four- to fivefoldgreater GPX activitythan did the Wendel procedure. One of the greatest differences in the assay components between the two procedures is the concentration of reduced glutathione: 5 vs. 1 mM. Therefore, the effect of reduced glutathione concentration on activity was determined. A curvilinear response was obtained with increased glutathione concentrations up to the highest concentration used, which was 6.0 mM with the Wendel procedure (Fig. 1, top). Although not readily apparent, the response to concentrations of up to 1 mM appears to be linear. Using the modified P and V method, a linear response in GPX activitywas ob-

.5O

djo #{149}
#{149}0

0 0 0

4 Z

#{149}
00
r,Qis

0 0

E C30.
o

#0

#{149}

00

oNew Zealand #{149}Oregon

South Dakota
I I

8
0
2
C,

Selenium content

(ng/mg Hb)

2. Glutathioneperoxidaseactivitys. selenium v blood from people living in the country of New Zealand states of Oregon and South Dakota.
Figure

content

of

and

in the

60

40

20

tamed with concentrations of up to 10 mM (Fig. 1, bottom). Use of the modified P and V procedure as an example indicates an activity of 38 units with 5 mM as compared with only 13 units for 1 mM reduced glutathione. The t-butylhydroperoxide concentration in the Wendel procedure is about twice (2.5 vs. 1.2 mM) that in the modified P and V method, but this would result in greater GPX activity. Therefore, the major reason for such a difference between the two assay procedures appears to be the concentration of reduced glutathione used. The plot of GPX activities (modified P and V) vs. Se concentrations (ng Se/mg hemoglobin; Fig. 2) indicated a linear relationship between these variables only at very low Se concentrations, i.e., in the New Zealand group. Linear regression of GPX vs. Se by geographic group gave a significantpositivecorrelation (r .46, P < 0.01) only for the New Zealand group. Correlation coefficients for GPX vs. Se in the Oregon and South Dakota groups were 0.11 and -0.02, respectively. The correlationcoefficient, however, for allsamples was 0.60 (P < 0.001). The lack of an increase in GPX activityat higher Se concentrations suggests saturation of the ability to synthesize GPX. The data from this study were fitted to the following saturation model: GPX k1 x antiln (k2/[Se]), with GPX activities determined by the modified P and V procedure and Se concentrations expressed as ng Se/mg hemoglobin. Transformation of thisequation to:
= =

ln (GPX)

in (k1)

k2 x [Se]

9
allowed

Glutathione
Figure 1.Effectf reduced o one peroxidase (GPX) activity Concentrations of 0.25, 0.5, 1.0,

regression analysis of ln (GPX) vs. [Se] (Fig. 3) (Table 4). In this model, k1 represents the average glutathione concentration glutathi- blood GPX activity of a population on with theoretically inas assayed by two different
2.0, 3.0, and 6.0

content (mM)

methods.

mM-reduced

glutathione were used in the Wendel assay (top),and concentrations of 1.0, 4.0, 8.0, and 10.0-mM reduced glutathione were used in the modified Paglia and Valentine assay (bottom).

for all sam(P < 0.001) and a value fork1 of 45.2. When the regression was performed with only the New Zealand samples, the corplcs gave a correlation coefficient of -0.80 2999

finite blood

Sc levels. The regression

analysis

BLOOD

SELENIUM

AND

GLUTATHIONE

PEROXIDASE

(ng Se/mg
3. Natural log per mg hemoglobin.
Figure of

Hb)
vs. the inverse of Se content

GPX

activity

relation coefficient was -0.51 (P < 0.01) and k1 was 46.4. The similarity of the maximum average GPX values estimated from a population with low Se status and a population with a broad range of blood Se levels indicates that the model is a good predictor. DISCUSSION The results described here are in agreement with a previous study of blood from New Zealanders (9) that suggested that GPX activity was correlated with blood Se levelsonly in populations with very low Se status. However, this is in disagreement with two studies (12, 13) that found a strong correlation of blood Se and GPX in populations with relativelyhigh Se status. There is no apparent explanation for these conflicting results, but the selenomethionine content may be a contributing factor, as there is not a corresponding increase of GPX activity with Se levels due to this selenoamino acid (27-29). Also, the percentage of erythrocyte Se associated with GPX has been suggested as one of the factors affecting the correlation of GPX activity with Se levels (30, 31). These results, which show a correlation of GPX activity with blood Se levels in blood from New Zealanders

but not in blood from people of higher status, are similar to results obtained from animals. With increasing levels of dietary Se, GPX activity has been shown to reach a plateau before the concentrations of Sc in tissues do (3, 8, 32). Figures 2 and 3 suggest that after biological saturation is reached, GPX does not increase any more even though Se levels are further elevated. The percentage of erythrocyte Se associated with GPX has been shown to be greater in blood samples from New Zealanders than in the samples from Oregonians (31). Therefore, the greater percentage of Sc associated with GPX appears to be at low Se status, which suggeststhat this enzyme has priority for Se over other systems. These results, which show a highly significant correlation between the results obtained with the two assays, suggest that the disagreement in the literature is not due solely to different assay procedures. Both methods demonstrated a correlation between GPX activity and blood Se in the New Zealanders, but not in the Oregon or South Dakota groups. Because the blood from all sources was assayed under identical conditions, the differences in the blood from the three sources are real. The inability to assay GPX at saturating substrate concentrations (21, 33) introduces greater variability into the assay because a slight variation in assay components will be reflected in assay activity (Fig. 1). The kinetics and mechanism of GPX action have been extensively studied (34, 35), and it was not our purpose to repeat these studies here. We wanted only to emphasize the influence of reduced glutathione concentration on the activity of GPX. It is well known that GPX cannot be saturated with both glutathione and hydrogen peroxide simultaneously, but at a given concentration of one substrate, saturation can be determined for the other. The peroxide concentrations in the study shown in Fig. 1 were held constant with the various levels of reduced glutathione. Comparison of the two GPX methods demonstrates that the assays are highly correlated. Therefore, it appears likely that all GPX assays based on the enzymelinked method of Paglia and Valentine (22) will produce results similar to those obtained by the Wendel procedure (23). Thus the differences in the relationship of blood GPX and Se observed in various human populations cannot be attributed to any major extent to differences in GPX assay conditions.
and gluiat/zione peroxid.ase

TABLE

4. Regression analysis for concentrations of whole blood selenium

Modified P and V vs. selenium, ng/mg hemoglobin

In (Modified P and V) vs. [Ser, mg hemoglobin/ng

Population
New Zealand

Sample size 34 23 21
78

I
14.3 35.1

Slope
28.3 3.37

r
.46 .11
- .02

Slope -0.243

r - .51

<

0.01 ns ns

3.84

< 0.01

Oregon South Dakota

All I, 45.2.

41.2
27.8 Values

-0.16
4.85 are
standard

.60 deviation,

<

0.001 in, 46.4.

3.8l

-0.230

- .80

< 0.001

intercept.

significance.

ns, slope not significantly

different from 0.

1ln,

3000

Vol. 2

Nov. 1988

The FASEB Journal

WHANGER

ET AL.

has been shown in several a more precise indication of human blood Se levels;however, platelet preparation is much more troublesome than whole blood or red blood cell (RBC) preparation, and it may not be feasible for epidemiological studies where samples must be transported before processing. Furthermore, platelet GPX may primarily reflect recent Sc intake, whereas RBC and whole blood Se indicate Se status over a longer period of time and, therefore, they may be more appropriate to the study of chronic disease conditions and diseases of long incubation. In conclusion, the results from two assay methods used most commonly forGPX activity were highly correlated, and both indicated a correlation of GPX activity with Se in blood from people with low Se status but not for those with higher Se status. Thus the data from many studies that show no correlation of GPX activity to blood Se reflect factors that are mostly unidentified, and apparently are not a result of the techniques used.
The author ideas spent for this research were generated a 2-month study leave (Deutscher when the senior Akademischer

Platelet GPX activity studies (36, 38) to provide

REA, H. H.; DOESBURG, V. M.; M. F. Selenium concentrations and glutathione peroxidase activitiesin whole blood of New Zealand residents. Br.]. Nutr. 37: 457-460; 1977. 10. VAN Rij, A. M.; THOMSON, C. D.; MCKENZIE, J. M.; ROBINSON, M. F. Selenium deficiency in total parenteral nutrition. Am.]. Clin. Nut,. 32: 2076-2085; 1979.

9. THOMSON,

C. D.;

ROBINSON,

11.

MCKENZIE, RoBINsoN,

R. L.;

REA,

H.

M.;

THOMSON,

C. D.;

M. F. tathione peroxidase infants and children. 1978.

Selenium concentration and gluactivity in blood of New Zealand Am.]. Gun. Nutr. 31: 1413-1418;

12.

13.

RUDOLPH, peroxidase red cells. LANE, H.

N.;

WONG, S. L. Selenium and glutathione activity in maternal and cord plasma and DUDRICK,

Pediatr.

W.;
levels

selenium

12: 789-792; 1978. S.; WARREN, D. C. Blood and glutathione peroxidase activities in

Res.

university and chronic intravenous hyperalimentation subjects. Proc. Soc. Exp. Biol. Med. 167: 383-390; 1981. 14. LANE, H. W.; BARROSO, A. D.; DUDRICK, S. ENGLERT, D. A.; FACFADYEN, B. V. Selenium status of

J.;

J.

seven IVH patients. McHowell, J. M.; Gawthorne, M.; White, C. L., eds. Trace element metab. man

Proc. Symp. 4th. Canberra, Australia: Australian Academy Science; 1982: 30-33. 15. BAPTISTA, R. BISTRIAN, B. R.; BLACKBURN, G. L.; MILLER, D. G.; CHAMPAGNE, C. D.; BUCHANAN, L.

anim.,

mt. J.;

Austauschdienst) in the laboratory of Dr. A. Wendel, Universit#{227}t Tubingen, Physiologisch-Chemisches Instit#{252}t, Tubingen 1, D-7400 FederalRepublic of Germany. The research was supported by U.S. Public Health ServiceResearch Grant DK 38306 from the National
Instituteof Diabetes and Digestive and Kidney Diseases, and by the

Utilizing selenious acid to reverse selenium in totalparenteral nutritionpatients.Am.]. 39: 816-820; 1984. 16.

deficiency Clin. Nutr.

SCHMIDT, K.; HELLER, W. Selenkonzentration und Activit#{227}t Glutathion-peroxydase in Lysat der Menschlischer Erythrozyten. Blut 33: 247-251; 1976 (Eng.

Medical

Research

Council

of New Zealand. 17.

abstr.).

REFERENCES P. D. The relationship of glutathione peroxidase in populations of low, medium and high selenium status. FASEB ]. 2: A1088; 1988 (abstr.). 2. OH, S. H.; GANTHER, H. E.; HOEKSTRA, W. G. Selenium as a component of glutathioneperoxidase isolated from ovine erythrocytes.Biochemistry 13:
WHANGER,

1.

BEILSTEIN, M. blood selenium

A.;

18.

and

G. N.; WHITE, D. A. Selenium in human nutrition: dietary intakes and effects of supplementation. Bioinorg. Chem. 8: 303-318; 1978. BEHNE, D.; WOLTERS, W. Selenium content and glutathione peroxidase activity in the plasma and erythrocytes of non-pregnant and pregnant women. J. Clin. Chem. Clin. Biochem. 17: 133-135; 1979.
SCHRAUZER,

19. VALENTINE, SCHLUCHTER,

J.

L.;

KANG,

H.

K.;

DANG,

P.-M.;

1825-1829;

1974. 20.

3. OH, S. H.; POPE, A. L.; HOEKSTRA, W. G. Dietary selenium requirement of sheep fed a practical-type diet as assessed by tissue glutathione peroxidase and other criteria. J. Anim. Sci. 42: 984-992; 1976. 4. WHANGER, P. D.; WEswIc, P. H.; SCHMITZ, J. A.; OLDFIELD, J. E. Effects of selenium and vitamin E on blood selenium levels, tissue glutathione peroxidase
activities purified and white or hay diets.

M. Selenium concentrations and glutathione peroxidase activities in a population exposed to selenium via drinking water. ]. Toxicol. Environ. Health 6: 731-736; 1980. BUTLER, J. A.; WHANGER, P. D.; TRIPP, M. J. Blood selenium and glutathione peroxidase activity in pregnant women: comparative assays in primates and other animals. Am.]. Clin. Nutr. 36: 15-32; 1982.
FLOHE,

21.

L.;

GUNZLER,

W. A. Glutathione

peroxidase.

J.

muscle Nutr.

disease in sheep fed 107: 1298-1307; 1977. 22.

5.

THOMPSON, FLowER,

W.

peroxidase
Res.

H.; MCMURRAY, C. H.; BLANCHThe levels of selenium and glutathione activity in blood of sheep, cows and pigs.

R.

J.

Benohr, H. C.; Sies, H.; Waller, H. D.; Wendel, A., eds.Glutathione: Proc. 16th Conj German Soc. Biol. Chem.; March 1973; Tubingen, FRG; Stuttgart, FRG: Georg Thieme Verlag; 1974: 132-145. PAGLIA, D. E.; VALENTINE, W. N. Studies on the quantitative and qualitative characterization of erythrocyte

Floh#{233}, L.;

Vet. Sci. 20: 229-231; 1976. 6. HOFFMAN, C.; RIvINuS, B.; SWANSON,

L. Effect of intramuscular administration of selenium and vitamin E in dairy heifers on erythrocyte glutathione peroxidase activity and blood selenium levels. ]. Anim. Sci. 47: 192-197; 1978. 7. CHAVEZ, E. R. Effect of dietary selenium on glutathione
peroxidase activity in piglets.

glutathione 158-169; 23. 24.

WENDEL,

peroxidase. 1967. A. Glutathione

J.

Lab.

Clin. Methods

Med.

70:

peroxidase.

Enzymol.

77: 325-333; 1981. BLACK, R. S.; TRIPP, WESWIG, P. H. Selenium

WHANGER, P. proteins in ovine tissue: Distribution of selenium and glutathione peroxidase

M.

J.;

D.; III.
in

Can.

].

Anim.

Sd.

59: 25.

8.

HAFEMAN, Effect of glutathione

67-75; 1979. D. G.; 580-587; 1974.

SUNDE, R. A.; HOEKSTRA, W. G. dietary selenium on erythrocyte and liver peroxidase in the rat. ]. Nutr. 104:

26.

tissuecytosols.Bioinorg.Chem. 8: 161-172; 1978. ANTONINI, E.; BRUNORI, M. Hemoglobin and methemoglobin. Surgenor, D. M., ed. The red blood cell. New York, N.Y.: Academic; 1975: 753-797. BROWN, M. W.; WATKINSON, J. H. An automated fluorometric method for the determination of nano-

BLOOD

SELENIUM

AND GLUTATHIONE

PEROXIDASE

3001

gram quantitiesof selenium.

29-35; 27.
WHANGER,

Analyt.

Chim.

Acta

89:

1977.

eds.

dietary
nine
dase LANE,

P. D.; BUTLER, J. A. Effects of various levels of selenium as selenite or selenomethioselenium

in rats.].
SHIRLEY,

Benohr, H. C.; Sies, H.; Waller, H. D.; Wendel, A., Glut athione: Proc. 16th Conf German Soc. Biol. Ghe,n.; March, 1973: Tubingen, FRG; George

on tissue
activity

levels NuIr.

and glutathione peroxi118: 846-852; 1988.

CERDA,J.J.

34.

28.

H. W.; various

R. L.;

Glutathione
tissue of rats 35.

proxidase

activity

in intestinal

and

liver

fed
29.

levels
Nutr.

of selenium,
109: 444-452; of
SUNDE,

sulfur
1979.

and

alpha36.

tocopherol.].
WASCHULEwSKI,

I. H.;
on

R. A. Effect
tissue

methionine

utilization

of dietary selenium from

peroxidase in

30.

Deposition of dietaryorganicand inorganicselenium in raterythrocyte proteins.].NuIr. 116: 1701-1710; 1986. 37. 31. WHANGER, P. D.; ROBINSON, M. F.; FELDMAN, E. B.; BEILSTEIN, M. A.; BUTLER, J. A. Variations in the distribution of selenium between erythrocyte glutathione peroxidase and hemoglobin in different human populations. Federation Proc. 45: 474; 1986 (abstr.). 38. 32. MEYER, W. R.; MAHAN, D. C.; MoxoN, A. L. Value
of dietary selenium and vitamin E for weanling swine as measured by performance and tissue selenium and glutathione peroxidase activities. ]. Anim. Sci. 52:

dietary selenomethionine for glutathione the rat.]. Nutr. 118: 367-374; 1988. BEILSTEIN, M. A.; WHANGER, P. D.

180-184. W. A.; EICHELE, E. peroxidase, V. The kinetic mechanism. Hoppe-Seylers Z. Physiol.C/tern. 353: 987-999; 1972. GUNZLER, W. A.; VERGIN, H.; MULLER, I.; FLOH, L. Glutathion-Peroxidase, VI. Die Reaktion der Glutathion-Peroxidase mit verschiedenen Hydroperoxiden. Hoppe-Seylers Z. Physiol. Chem. 353: 1001-1004; 1972. LEVANDER, 0. A.; ALFTHAN, G.; ARVILOMMI, H.; GREF, C. G.; HUTTUNEN, J. K.; KATAJA, M.; KolvIsTOINEN, P.; PIKKARAINEN,J. Bioavailability of selenium to Finnish men as assessed by platelet glutathione peroxidase activity and other blood parameters. Am.].
Theime FLOH, 1974:

L.; Glutathione

Verlag; Stuttgart, FRG: LOSCHEN, G.; GiNZLER,

Gun.

Nutr.

37: 887-897;

1983.

MENZEL,

STEINER, G.; LOMBECK, F. K. Glutathione peroxidase and transferase activity of platelets. Eur.

H.;

I.;
].

OHNESORGE, glutathione S-

Pediatr.140:

244-247;
LANE,

1983.

H. W.; LOTSPEICH, LARD, DUDRICK, S. of selenium supplementation tients receiving chronic

C. A.;

MOORE,

J.;

J.;

WARREN, D. on selenium total parenteral

C. E.; C. The

BALeffect status of panutrition.

].

Parenter.

Enteral

Nutr.

11: 177-182;

1987.

302-311;

1981.
critical remarks on

33.

GiJNZLER,

nation

W. A. Some of glutathione

peroxidase

in blood.

the determiFloh#{233}, L.;

Received for publication March 28, 1988. Accepted for publication June 22, 1988.

3002

Vol. 2

Nov. 1988

The FASEBJournal

WHANGER

ET AL.