- Email: [email protected]
Effect of selenium deficiency on the disposition of plasma glutathione
ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 240, No. 1, July, pp. 166-171, 1985
Effect of Selenium Deficiency KRISTINA Department
of Medicine, Health Science Received
on the Disposition
E. HILL’
AND
RAYMOND
Division of Gastroenterology Center, 7703 Floyd Curl Drive,
December
26, 1984, and in revised
of Plasma Glutathione’ F. BURK
and Nutrition, San Antonio, form
February
University of Texas Texas 7828h 8, 1985
Selenium deficiency causes increased hepatic synthesis and release of GSH into the blood. The purpose of this study was to examine the effect of selenium deficiency on the disposition of plasma glutathione. Plasma glutathione concentration was 40 * 3.4 nmol GSH equivalents/ml in selenium-deficient rats and 17 + 5.4 nmol GSH equivalents/ml in control rats. The half-life and systemic clearance of plasma glutathione were found to be the same in selenium-deficient and control rats (& = 3.4 f 0.7 min). Because selenium-deficient plasma glutathione concentration was twice that of control, the determination that selenium deficiency did not affect glutathione plasma systemic clearance indicated that the flux of glutathione through the plasma was doubled by selenium deficiency. It has been proposed that the kidney is responsible for the removal of a major fraction of plasma glutathione. In these studies, renal clearance accounted for 24% of plasma systemic glutathione clearance in controls and 44% in selenium-deficient rats. This indicates that a significant amount of glutathione is metabolized at extrarenal sites, especially in control animals. More than half of the increased plasma glutathione produced in selenium deficiency was removed by the kidney. Thus, selenium deficiency results in a doubling of cysteine transport in the form of glutathione from the liver to the periphery as well as a doubling of plasma glutathione concentration. Q 1985 Academic PXSS, h.
Glutathione (y-glutamylcysteinylglytine) turnover in rat liver is increased twofold by selenium deficiency. Hepatic glutathione synthesis and hepatic glutathione release into the plasma are doubled (1). As a result of this increased glutathione release, plasma glutathione concentration is increased in selenium deficiency. Glutathione in plasma and interstitial fluid is catabolized to its constituent amino acids by tissues which contain the extracellular enzymes y-glutamyltranspeptidase (T-GTP)~ and dipeptidase (2). It has
been reported that the kidney is a major site of plasma glutathione removal and degradation (3-5). The present studies were undertaken to examine the effect of selenium deficiency on the fate of plasma glutathione, with special attention to renal handling of plasma glutathione. EXPERIMENTAL Animals. Male Sprague-Dawley rats (250-450 g) which had been fed the experimental diet from weaning for at least 8 weeks were used in all experiments. The diet was prepared as described previously (6). The control diet contained 0.5 mg/kg selenium as NaaSeOa and 100 IU/kg vitamin E as DL-a-tocopheryl acetate. The selenium-deficient diet omitted the selenium. The rats had free access to food and water up to the time of the experiment. Selenium deficiency was verified by measurement of glutathione peroxidase in the livers of some of the rats fed the selenium-deficient diet (7). Liver weights
i This work was supported by NIH Grant ES 02497 and Grant AQ-870 from the Robert A. Welch Foundation of Houston, Tex. ‘To whom correspondence should be addressed. 3 Abbreviations used: y-GTP, y-glutamyltranspeptidase; DTNB, 5,5’-dithiobis(2-nitrobenzoic acid). 0003-9861185
$3.00
Copyright 0 1985 by Academic Press, Inc. All rights of reproduction in any form reserved.
166
PLASMA
GLUTATHIONE
IN
THE
and body weights were measured in these rats over a 12-month period. Selenium-deficient livers weighed 3.58 t 0.34 g liver/100 g body wt (n = 28) and control livers weighed 3.32 f 0.27 g liver/I00 g body wt (n = 29). Chemicals. GSH, GSSG, glutathione reductase, DTNB, 5-sulfosalicylic acid, and trichloroacetic acid were purchased from Sigma Chemical Company, St. Louis, Missouri. NADPH was obtained from Boehringer Mannheim, Indianapolis, Indiana. [%]GSH and Protosol were obtained from New England Nuclear, Boston, Massachusetts. Ready Solv EP liquid scintillant was obtained from Beckman Instruments, Inc., Houston, Texas. All other chemicals were of reagent grade and obtained from local sources. Methods. Total glutathione concentration was measured with the glutathione reductase-DTNB recirculating assay of Tietze (8). Plasma was prepared by three methods for glutathione determination to allow comparison of the methods. Blood was removed from the aorta of an anesthetized rat (sodium pentobarbital, 50 mg/kg body wt) and anticoagulated with EDTA (1 mg/ml). An aliquot of whole blood was immediately mixed with an equal volume of 10 mM DTNB. DTNB-treated plasma was separated by centrifugation and assayed for glutathione as described by Adams et al (9). The remainder of the blood was centrifuged for 1.5 min in a Beckman microfuge. An aliquot of the resultant plasma was mixed with an equal volume of 10 mM DTNB and assayed for glutathione. A O.&ml aliquot of plasma was deproteinized with 0.25 ml of 10% 5-sulfosalicylic acid and assayed for glutathione as described by Anderson and Meister (10). Plasma glutathione concentration was determined in arterial and renal venous blood. Renal vein blood was obtained from animals in which the vena cava had been clamped above and below the renal veins to prevent mixing of other venous blood when renal venous blood was removed. Plasma samples with greater than 0.2% hemolysis (11) were not used for glutathione determinations. Tissue samples (1 g) were homogenized in 10% trichloroaeetic acid with a Brinkmann polytron. The homogenates were centrifuged at 15009 for 10 min and an aliquot of the supernatant was neutralized wit11 0.3 M NaH2P04 before it was assayed for glutathione. The rate of removal of [%]GSH from the plasma was measured after injecting [%]GSH (3 pCi, 60 pmol) into the inferior vena cava above the renal veins. At the desired times, blood was removed using a heparinized syringe (25 ~1 heparin solution/300 ~1 blood). Blood was taken from the vena cava below the renal veins with the vena cava clamped so that the blood sampled had its origin distal to the kidneys. Plasma radioactivity was determined by adding a loo-p1 aliquot of plasma directly to 10 ml of scintil-
SELENIUM-DEFICIENT
RAT
167
lant. At each time point it was necessary to determine the amount of a5S remaining in the plasma as GSH. The GSH was separated from other S-containing compounds in a manner similar to that described by Griffith (12). Blood was withdrawn from the aorta and immediately mixed with EDTA (1 mg/ml) and carrier GSH (20 rmol/ml). The blood was centrifuged for 1.5 min in a Beckman microfuge and the plasma was immediately deproteinized with 10% trichloroacetic acid. A 500-~1 aliquot of acid supernatant was diluted with 5.0 ml Hz0 and the acidic sample was applied to a Dowex 1-X (acetate form) column (0.5 X 7.0 cm). The column was eluted with two 5.0-m] aliquots of Hz0 and 0.3, 0.6, 1.0, 1.4, and 1.8 M acetic acid. Cystine eluted in the column flow through and the Hz0 eluates; GSH eluted with 1.0 M acetic acid. The percentage of counts injected remaining in the plasma at a given time was corrected to reflect the percentage present as GSH. The amount of % radioactivity retained in the kidney was determined after injection of [%]GSH (3 &i, 60 pmol) into the vena cava. At the desired time the blood was removed from the aorta and the right kidney was removed and weighed. The kidney was homogenized in saline and a 250-p] aliquot was digested in 1.0 ml of Protosol. The tissue homogenate was digested by incubating the sample in a scintillation vial at 55°C for approximately 2 h. After the sample was digested and cooled, 10 ml of scintillant was added to the vial and the radioactivity was determined. The radioactivity of tissue and plasma samples was determined using a Model 3320 Packard liquid scintillation counter. Measurement of renal blood flow in pentobarbital anesthetized rats was performed using a 2-mm blood flow transducer on a square-wave electromagnetic flowmeter (Carolina Medical Electronics, Inc.) (13). Renal plasma flow was calculated from the relationship renal
plasma = renal
flow blood
flow * (1 - corrected
hematocrit),
where corrected hematocrit = 0.95 X hematocrit. r-GTP activity was measured by determining the rate of formation of p-nitroaniline (Sigma Kit No. 415, Sigma Chemical Co., St. Louis, MO.). Systemic plasma clearance values (Clr) were calculated from the relationship (14) Clr = dose/AUC, where AUC is the area under the curve. AUC was determined from the graph of [%]GSH remaining in the plasma with time (Fig. 2); dose was the initial radioactivity of [%]GSH injected intravenously. The fractional rate of disappearance, k, was determined from the terminal phase of the [%]GSH disappearance curve (t = 2.3, 4, 5 min). Volume of distribution ( V,) was calculated from the relationship Clr = V, * k. Clr, k, and V, were determined individually for each rat and then average values were calculated. Cl,,,,
168
HILL
was calculated as the product and renal extraction ratio.
of renal
plasma
AND
BURK
flow
RESULTS
There has been considerable discussion about the methodology employed to determine plasma glutathione concentrations (9,lO). Figure 1 compares three techniques we used to prepare plasma for determination of glutathione in control and selenium-deficient rats. In confirmation of the results of Anderson and Meister (lo), glutathione measurable in acid-deproteinized plasma decreased rapidly with time (Fig. lb). In contrast, when plasma or whole blood was treated with DTNB immediately, the measurable plasma glutathione increased slightly, but not significantly, over the subsequent 2 h (Fig. la). A potential explanation of this is the oxidation of GSH to GSSG by DTNB which would decrease its reactivity with other plasma constituents such as cystine (15,16). When DTNB-treated whole blood was allowed to stand for 1 h, the measurable glutathione increased significantly (data not shown). DTNB may thus cause release of glutathione from cells. As a result of this comparison, subsequent plasma glutathione determinations were performed after treating rapidly separated plasma with DTNB. 0 50-s-
oJ-
0
T
T b
I
2
0
I
2
hr
FIG. 1. Comparison of glutathione concentration in control (0) and selenium-deficient (A) plasma. Plasma was prepared for assay by DTNB treatment of whole blood (--, a), DTNB treatment of plasma (---, a), and acid deproteinization of plasma (b); see Experimental. Each point represents the mean of at least five animals. One SD is represented by the bracket.
0
I
2
3
4
5
6
min FIG. 2. Rate of removal of glutathione from selenium-deficient (a) and control (0) plasma. A tracer dose of [%]GSH was injected into the vena cava. The amount of % remaining in the plasma as GSH at each time point is expressed as a percentage of the [%]GSH injected; see ExperimentaL Each point represents the mean of at least six animals. One SD is represented by the bracket. The lines were determined by linear regression analysis using control and selenium-deficient values.
Selenium deficiency caused a twofold increase in plasma glutathione concentration (Fig. I). Selenium-deficient arterial plasma contained 40 + 3.4 nmol glutathione/ml plasma and control arterial plasma contained 17 -+ 5.4 nmol glutathione/ml plasma. The disappearance of plasma glutathione was measured to determine whether it was affected by selenium deficiency. Figure 2 shows that plasma [35S]GSH disappeared at the same rate in seleniumdeficient and control rats. The removal of glutathione from the plasma appeared to be biphasic, suggesting the presence of a distribution phase and an elimination phase (14). The distribution phase could represent equilibration of the label between the plasma and interstitial fluid compartments. This equilibration would account for the decrease of radioactive material to approximately 30% of the total injected dose within 30 s. GSH halflife values of 3.4 4 0.7 min were calculated using the points in the elimination portion of the curve. From these results we calculated the fractional rate of disappearance (k), systemic plasma clearance (Clr), and volume of distribution (V,) of plasma GSH. As Table Ia shows, these pharmacokinetic values were not affected by selenium deficiency. Because the glutathione concentration was two times greater in
PLASMA
GLUTATHIONE
IN
THE
SELENIUM-DEFICIENT
TABLE
169
RAT
I
PLASMAGLUTATHIONEINSELENIUM-DEFICIENT AND CONTROLRATS" ?l a. Kinetics of plasma glutathione Fractional rate of disappearance V, (ml/100 g body wt) CIT (ml plasma/min a 100 g body b. Plasma glutathione Aorta Renal vein
(min-‘) wt)
Selenium
deficient
4 4 4
0.20 + 0.04 22.5 + 3.9 4.42 t 0.91
0.22 * 0.04 21.4 +- 4.1 4.59 * 1.05
5 6
17 * 5.46 5.2 k 0.9’
40 +- 3.4b 9.5 f 1.6b
-
0.69
0.76
6 -
1.52 + 0.27’ 1.05 0.24
2.64 31 0.62” 2.01 0.44
levels
c. Kinetics of glutathione removal by the kidneys Renal extraction ratio Rena1 plasma flow (ml plasma/min * 100 g body ,wt) (ml plasma/min . 100 g body wt) CL, ~L”JC1T (LValues *.“Values test.
Control
are means + SD. with the same superscript
are statistically
selenium-deficient plasma than in control plasma and because the clearance did not change, the:se results indicate that selenium deficiency doubles the rate of production of plasma glutathione. The kidney is thought to be the major organ responsible for the removal of plasma glutathione (3-5). The increased turnover of plasma glutathione in selenium-deficient rats prompted an examination of its removal by the kidneys. y-GTP is the enzyme responsible for the initial step of plasma glutathione degradation. The enzymatic activity of renal r-GTP was 1.91 t 0.12 pmol p-nitroaniline formed/mg protein. min in control kidney homogenate and 1.94 + 0.20 /*mol p-nitroaniline formed/mg protein - min in selenium-deficient kidney homogenate. Thus, selenium deficiency does not affect the total activity of y-GTP in the kidney. Arterial and renal vein plasma glutathione were measured in seleniumdeficient and control rats (Table Ib). The selenium-deficient kidney removed a greater amount of glutathione per milliliter plasma than did the control kidney (30 nmol vs 1.2 nmol) due apparently to a slight increase in extraction ratio and to
different
from
each other
(P < 0.05)
by Student’s
t
the higher plasma glutathione concentration. In addition, renal plasma flow was higher in selenium-deficient rats than in control rats (Table Ic). The increase in renal plasma flow and the slightly higher extraction ratio in selenium deficiency resulted in an increased renal plasma clearance of glutathione (Clrenal) in comparison to control. Renal clearance accounted for 44% of the systemic clearance in seleniumdeficient rats but only for 24% in controls. Renal removal of plasma glutathione was also examined by determining the amount of radioactivity retained by the kidney after intravenous administration of [?S]GSH. The selenium-deficient kidney retained 1.3-1.7 times more injected radioactivity than did the control kidney (Fig. 3). This is consistent with the increased removal of plasma glutathione in the selenium-deficient rat. The relatively constant amount of radioactivity retained by selenium-deficient or control kidney is supportive of the rapid turnover of glutathione in the kidney (4). After intravenous administration of [35S]GSH, 24-h urine collection showed that identical percentages of the total injected [35S] were present in control and
170
HILL
AND
FIG. 3. Amount of ‘?S retained by control (0) and selenium-deficient (A) kidneys expressed as a percentage of the [Y$JGSH injected. The radioactivity remaining in the kidneys following the intravenous injection of a tracer dose of [?S]GSH (see Experimental) was determined at the time points shown. Each point is the mean of at least three animals. One SD is represented by the bracket.
selenium-deficient urine (15 + 3% (n = 5) and 15 + 2% (n = 5), respectively). Because the initial glutathione pool is twice as great in selenium-deficient plasma as in control, these urinary data are consistent with a twofold increase in the metabolism of plasma GSH by the selenium-deficient kidney. Even though there was increased removal of plasma glutathione by the selenium-deficient kidney, there was no difference between selenium-deficient and control kidney glutathione concentrations. The control kidney contained 2.32 + 0.30 pmol GSH equivalents/g (n = 8) and the selenium-deficient kidney contained 2.48 +- 0.29 pmol GSH equivalents/g (n = 9). Thus glutathione metabolism (removal from the plasma and degradation) in the kidney appears to be affected by selenium deficiency. DISCUSSION
Glutathione metabolism is markedly affected by selenium deficiency. Glutathione synthesis and release by the liver are doubled in selenium deficiency (1). Plasma giutathione concentration is doubled as a
BURK
result. The plasma glutathione half-life is the same in selenium-deficient and control rats. Since the glutathione concentration is two times greater in selenium-deficient plasma than in control plasma, finding the same plasma clearance in both nutritional states indicates that the amount of plasma glutathione turned over in selenium deficiency is two times greater than in control. The total amount of glutathione released into the systemic circulation can be estimated from arterial glutathione concentration and Clr. From this calculation, it is estimated that the seleniumdeficient rat releases 2.4 times more glutathione into the extracellular fluid than does the control rat (182 nmol GSH equivalents/min -100 g body wt in selenium deficiency vs 75 nmol GSH equivalents/ min * 100 g body wt in control). The value of 75 nmol/min -100 g body wt calculated for the input of glutathione into (or its removal from) the circulation of control rats agrees reasonably well with that of 59 nmol GSH equivalents/min - 100 g body wt reported by Lauterburg et al. (17). Hepatic glutathione release has been proposed as the primary source of plasma glutathione (17). Using perfused liver data from Ref. (18) and liver wt/lOO g body wt for control and selenium-deficient livers (see Experimental), we estimate liver glutathione release to be 35 and 72 nmol GSH equivalents/min. 100 g body wt in control and selenium-deficient rats, respectively. These values are less than half the glutathione calculated to be released into the extracellular fluid (see above). They suggest that there may be other sites of glutathione release into the circulation. This assumes that release of glutathione by the perfused liver is the same as in vivo release and clearly this may not be the case. The kidney is known to be a major site of plasma glutathione removal. Hahn et al. (3) demonstrated in mice that a large amount of radioactivity was present in the kidney 5 min after intravenous injection of [14C]glycine-labeled glutathione. Griffith and Meister (4) measured plasma glutathione levels in rats which had had
PLASMA
GLUTATHIONE
IN
THE
a bilateral nephrectomy and compared them with the levels in sham-operated rats. They concluded that 67% of plasma glutathione removal occurred via the kidneys. Haberle et aL (5) measured arterial and renal vein plasma glutathione concentrations to assess renal removal of plasma glutathione. They concluded that the kidney accounted for approximately 50% of the glutathione removed from the plasma. The results presented in Table Ic indicate that renal plasma glutathione clearance accounts for 24% of the systemic plasma glutathione clearance in control rats. These results suggest that while the kidney removes a large amount of plasma glutathione, other sites are also responsible for the :removal of most of the plasma glutathione. The amount of glutathione removed by the kidneys can be calculated from the product of the arteriovenous difference in glutathione concentration and the renal plasma flow. The selenium-deficient kidney removes 81 nmol GSH equivalents/ min. 100 g body wt from the plasma compared with 18 nmol GSH equivalents removed/mine 100 g body wt by the control kidney. The increased removal of 63 nmol GSH equivalents/min 0100 g body wt by the selenium-deficient kidney accounts for 60% of the increased extracellular glutathione turnover in the selenium-deficient rat (182 nmol GSH equivalents/min * 100 g body wt for selenium-deficient vs 75 nmol GSH equivalents/min -100 g body wt for control). Thus, a major portion of the increased extracellular glutathione turned over in selenium deficiency enters the kidney. Possible reasons for this increased extracellular glutathione turnover in selenium (deficiency are (a) increased kidney requirement for glutathione or cysteine and (b) need for increased GSH concentration in extracellular fluid. In summary, selenium deficiency affects glutathione metabolism in rat liver, plasma, and kidney. The increased release of glutathione from selenium-deficient rat liver results in increased plasma glutathione. There is a corresponding increase in the removal of plasma glutathione by
SELENIUM-DEFICIENT
RAT
171
the selenium-deficient kidney. Removal of other sites may also be increased but over half the increase appears to be accounted for by the kidney. This indicates that selenium deficiency affects the interorgan metabolism of glutathione. ACKNOWLEDGMENTS The authors thank Mr. J. M. Lane for technical assistance, Dr. H. J. Reineck, Mrs. T. Craig, and Mrs. M. Parma for renal blood flow measurements, Mrs. S. R. B. Allerheiligen for helpful discussions concerning pharmacokinetics, Dr. T. M. Ludden for helpful suggestions and critical review of the manuscript, and Mrs. R. E. Ortiz for secretarial assistance. REFERENCES 1. HILL, K. E., AND BURK, R. F. (1982) J. Viol. Chem. 257, 10668-10672. 2. MCINTYRE, T. M., AND CURTHOYS, N. P. (1980) Int. J Biochem. 12, 545551. 3. HAHN, R., WENDEL, A., AND FLOHC, L. (1978) B&him. Biophys. Acta 539, 324-337. 4. GRIFFITH, 0. W., AND MEISTER, A. (1979) Proc. Natl. Acaa! Sci. USA 76, 5606-5610. 5. HABERLE, D., WAHLL~NDER, A., AND SIES, H. (1979) FEBS Lett. 108, 335-340. 6. LAWRENCE, R. A., AND BURK, R. F. (1978) J. N&r. 108, 211-215. 7. LAWRENCE, R. A., AND BURK, R. F. (1976) Biochem. Biophys. Res. Commun. 71, 952-958. 8. TIETZE, F. (1969) Anal Biochem. 27, 502-522. 9. ADAMS, J. D., LAUTERBURG, B. H., AND MITCHELL, J. R. (1983) J. Pharmacol Exp. Ther. 227,749754. 10. ANDERSON, M. E., AND MEISTER, A. (1980) J. Biol. Chem. 255, 9530-9533. 11. WENDEL, A., AND CIKRYT, P. (1980) FEBS Lett. 120, 209-211. 12. GRIFFITH, 0. W. (1981) J. Biol. Chem. 256, 49004904. 13. REINECK, H. J., O’CONNOR, G. J., LIFSCHITZ, M. D., AND STEIN, J. H. (1980) J. Lab. Clin. Med. 96, 356-362. 14. GIBALDI, M., AND PERRIER, D. (1982) Pharmacokinetics, pp. l-55, Dekker, New York. 15. BEATTY, P., AND REED, D. J. (1981) Biochem. Pharmacol. 30, 1227-1230. 16. BANNAI, S. (1984) B&him. Biophys. Acta 779, 289-306. 17. LAUTERBURG, B. H., AND MITCHELL, J. R. (1981) J. CZin. Invest. 67, 1415-1424. 18. KRIETER, P. A., ZIEGLER, D. M., HILL, K. E., AND BURK, R. F. (1985) B&hem. PharmacoZ. 34, 955-960.
Effect of Selenium Deficiency KRISTINA Department
of Medicine, Health Science Received
on the Disposition
E. HILL’
AND
RAYMOND
Division of Gastroenterology Center, 7703 Floyd Curl Drive,
December
26, 1984, and in revised
of Plasma Glutathione’ F. BURK
and Nutrition, San Antonio, form
February
University of Texas Texas 7828h 8, 1985
Selenium deficiency causes increased hepatic synthesis and release of GSH into the blood. The purpose of this study was to examine the effect of selenium deficiency on the disposition of plasma glutathione. Plasma glutathione concentration was 40 * 3.4 nmol GSH equivalents/ml in selenium-deficient rats and 17 + 5.4 nmol GSH equivalents/ml in control rats. The half-life and systemic clearance of plasma glutathione were found to be the same in selenium-deficient and control rats (& = 3.4 f 0.7 min). Because selenium-deficient plasma glutathione concentration was twice that of control, the determination that selenium deficiency did not affect glutathione plasma systemic clearance indicated that the flux of glutathione through the plasma was doubled by selenium deficiency. It has been proposed that the kidney is responsible for the removal of a major fraction of plasma glutathione. In these studies, renal clearance accounted for 24% of plasma systemic glutathione clearance in controls and 44% in selenium-deficient rats. This indicates that a significant amount of glutathione is metabolized at extrarenal sites, especially in control animals. More than half of the increased plasma glutathione produced in selenium deficiency was removed by the kidney. Thus, selenium deficiency results in a doubling of cysteine transport in the form of glutathione from the liver to the periphery as well as a doubling of plasma glutathione concentration. Q 1985 Academic PXSS, h.
Glutathione (y-glutamylcysteinylglytine) turnover in rat liver is increased twofold by selenium deficiency. Hepatic glutathione synthesis and hepatic glutathione release into the plasma are doubled (1). As a result of this increased glutathione release, plasma glutathione concentration is increased in selenium deficiency. Glutathione in plasma and interstitial fluid is catabolized to its constituent amino acids by tissues which contain the extracellular enzymes y-glutamyltranspeptidase (T-GTP)~ and dipeptidase (2). It has
been reported that the kidney is a major site of plasma glutathione removal and degradation (3-5). The present studies were undertaken to examine the effect of selenium deficiency on the fate of plasma glutathione, with special attention to renal handling of plasma glutathione. EXPERIMENTAL Animals. Male Sprague-Dawley rats (250-450 g) which had been fed the experimental diet from weaning for at least 8 weeks were used in all experiments. The diet was prepared as described previously (6). The control diet contained 0.5 mg/kg selenium as NaaSeOa and 100 IU/kg vitamin E as DL-a-tocopheryl acetate. The selenium-deficient diet omitted the selenium. The rats had free access to food and water up to the time of the experiment. Selenium deficiency was verified by measurement of glutathione peroxidase in the livers of some of the rats fed the selenium-deficient diet (7). Liver weights
i This work was supported by NIH Grant ES 02497 and Grant AQ-870 from the Robert A. Welch Foundation of Houston, Tex. ‘To whom correspondence should be addressed. 3 Abbreviations used: y-GTP, y-glutamyltranspeptidase; DTNB, 5,5’-dithiobis(2-nitrobenzoic acid). 0003-9861185
$3.00
Copyright 0 1985 by Academic Press, Inc. All rights of reproduction in any form reserved.
166
PLASMA
GLUTATHIONE
IN
THE
and body weights were measured in these rats over a 12-month period. Selenium-deficient livers weighed 3.58 t 0.34 g liver/100 g body wt (n = 28) and control livers weighed 3.32 f 0.27 g liver/I00 g body wt (n = 29). Chemicals. GSH, GSSG, glutathione reductase, DTNB, 5-sulfosalicylic acid, and trichloroacetic acid were purchased from Sigma Chemical Company, St. Louis, Missouri. NADPH was obtained from Boehringer Mannheim, Indianapolis, Indiana. [%]GSH and Protosol were obtained from New England Nuclear, Boston, Massachusetts. Ready Solv EP liquid scintillant was obtained from Beckman Instruments, Inc., Houston, Texas. All other chemicals were of reagent grade and obtained from local sources. Methods. Total glutathione concentration was measured with the glutathione reductase-DTNB recirculating assay of Tietze (8). Plasma was prepared by three methods for glutathione determination to allow comparison of the methods. Blood was removed from the aorta of an anesthetized rat (sodium pentobarbital, 50 mg/kg body wt) and anticoagulated with EDTA (1 mg/ml). An aliquot of whole blood was immediately mixed with an equal volume of 10 mM DTNB. DTNB-treated plasma was separated by centrifugation and assayed for glutathione as described by Adams et al (9). The remainder of the blood was centrifuged for 1.5 min in a Beckman microfuge. An aliquot of the resultant plasma was mixed with an equal volume of 10 mM DTNB and assayed for glutathione. A O.&ml aliquot of plasma was deproteinized with 0.25 ml of 10% 5-sulfosalicylic acid and assayed for glutathione as described by Anderson and Meister (10). Plasma glutathione concentration was determined in arterial and renal venous blood. Renal vein blood was obtained from animals in which the vena cava had been clamped above and below the renal veins to prevent mixing of other venous blood when renal venous blood was removed. Plasma samples with greater than 0.2% hemolysis (11) were not used for glutathione determinations. Tissue samples (1 g) were homogenized in 10% trichloroaeetic acid with a Brinkmann polytron. The homogenates were centrifuged at 15009 for 10 min and an aliquot of the supernatant was neutralized wit11 0.3 M NaH2P04 before it was assayed for glutathione. The rate of removal of [%]GSH from the plasma was measured after injecting [%]GSH (3 pCi, 60 pmol) into the inferior vena cava above the renal veins. At the desired times, blood was removed using a heparinized syringe (25 ~1 heparin solution/300 ~1 blood). Blood was taken from the vena cava below the renal veins with the vena cava clamped so that the blood sampled had its origin distal to the kidneys. Plasma radioactivity was determined by adding a loo-p1 aliquot of plasma directly to 10 ml of scintil-
SELENIUM-DEFICIENT
RAT
167
lant. At each time point it was necessary to determine the amount of a5S remaining in the plasma as GSH. The GSH was separated from other S-containing compounds in a manner similar to that described by Griffith (12). Blood was withdrawn from the aorta and immediately mixed with EDTA (1 mg/ml) and carrier GSH (20 rmol/ml). The blood was centrifuged for 1.5 min in a Beckman microfuge and the plasma was immediately deproteinized with 10% trichloroacetic acid. A 500-~1 aliquot of acid supernatant was diluted with 5.0 ml Hz0 and the acidic sample was applied to a Dowex 1-X (acetate form) column (0.5 X 7.0 cm). The column was eluted with two 5.0-m] aliquots of Hz0 and 0.3, 0.6, 1.0, 1.4, and 1.8 M acetic acid. Cystine eluted in the column flow through and the Hz0 eluates; GSH eluted with 1.0 M acetic acid. The percentage of counts injected remaining in the plasma at a given time was corrected to reflect the percentage present as GSH. The amount of % radioactivity retained in the kidney was determined after injection of [%]GSH (3 &i, 60 pmol) into the vena cava. At the desired time the blood was removed from the aorta and the right kidney was removed and weighed. The kidney was homogenized in saline and a 250-p] aliquot was digested in 1.0 ml of Protosol. The tissue homogenate was digested by incubating the sample in a scintillation vial at 55°C for approximately 2 h. After the sample was digested and cooled, 10 ml of scintillant was added to the vial and the radioactivity was determined. The radioactivity of tissue and plasma samples was determined using a Model 3320 Packard liquid scintillation counter. Measurement of renal blood flow in pentobarbital anesthetized rats was performed using a 2-mm blood flow transducer on a square-wave electromagnetic flowmeter (Carolina Medical Electronics, Inc.) (13). Renal plasma flow was calculated from the relationship renal
plasma = renal
flow blood
flow * (1 - corrected
hematocrit),
where corrected hematocrit = 0.95 X hematocrit. r-GTP activity was measured by determining the rate of formation of p-nitroaniline (Sigma Kit No. 415, Sigma Chemical Co., St. Louis, MO.). Systemic plasma clearance values (Clr) were calculated from the relationship (14) Clr = dose/AUC, where AUC is the area under the curve. AUC was determined from the graph of [%]GSH remaining in the plasma with time (Fig. 2); dose was the initial radioactivity of [%]GSH injected intravenously. The fractional rate of disappearance, k, was determined from the terminal phase of the [%]GSH disappearance curve (t = 2.3, 4, 5 min). Volume of distribution ( V,) was calculated from the relationship Clr = V, * k. Clr, k, and V, were determined individually for each rat and then average values were calculated. Cl,,,,
168
HILL
was calculated as the product and renal extraction ratio.
of renal
plasma
AND
BURK
flow
RESULTS
There has been considerable discussion about the methodology employed to determine plasma glutathione concentrations (9,lO). Figure 1 compares three techniques we used to prepare plasma for determination of glutathione in control and selenium-deficient rats. In confirmation of the results of Anderson and Meister (lo), glutathione measurable in acid-deproteinized plasma decreased rapidly with time (Fig. lb). In contrast, when plasma or whole blood was treated with DTNB immediately, the measurable plasma glutathione increased slightly, but not significantly, over the subsequent 2 h (Fig. la). A potential explanation of this is the oxidation of GSH to GSSG by DTNB which would decrease its reactivity with other plasma constituents such as cystine (15,16). When DTNB-treated whole blood was allowed to stand for 1 h, the measurable glutathione increased significantly (data not shown). DTNB may thus cause release of glutathione from cells. As a result of this comparison, subsequent plasma glutathione determinations were performed after treating rapidly separated plasma with DTNB. 0 50-s-
oJ-
0
T
T b
I
2
0
I
2
hr
FIG. 1. Comparison of glutathione concentration in control (0) and selenium-deficient (A) plasma. Plasma was prepared for assay by DTNB treatment of whole blood (--, a), DTNB treatment of plasma (---, a), and acid deproteinization of plasma (b); see Experimental. Each point represents the mean of at least five animals. One SD is represented by the bracket.
0
I
2
3
4
5
6
min FIG. 2. Rate of removal of glutathione from selenium-deficient (a) and control (0) plasma. A tracer dose of [%]GSH was injected into the vena cava. The amount of % remaining in the plasma as GSH at each time point is expressed as a percentage of the [%]GSH injected; see ExperimentaL Each point represents the mean of at least six animals. One SD is represented by the bracket. The lines were determined by linear regression analysis using control and selenium-deficient values.
Selenium deficiency caused a twofold increase in plasma glutathione concentration (Fig. I). Selenium-deficient arterial plasma contained 40 + 3.4 nmol glutathione/ml plasma and control arterial plasma contained 17 -+ 5.4 nmol glutathione/ml plasma. The disappearance of plasma glutathione was measured to determine whether it was affected by selenium deficiency. Figure 2 shows that plasma [35S]GSH disappeared at the same rate in seleniumdeficient and control rats. The removal of glutathione from the plasma appeared to be biphasic, suggesting the presence of a distribution phase and an elimination phase (14). The distribution phase could represent equilibration of the label between the plasma and interstitial fluid compartments. This equilibration would account for the decrease of radioactive material to approximately 30% of the total injected dose within 30 s. GSH halflife values of 3.4 4 0.7 min were calculated using the points in the elimination portion of the curve. From these results we calculated the fractional rate of disappearance (k), systemic plasma clearance (Clr), and volume of distribution (V,) of plasma GSH. As Table Ia shows, these pharmacokinetic values were not affected by selenium deficiency. Because the glutathione concentration was two times greater in
PLASMA
GLUTATHIONE
IN
THE
SELENIUM-DEFICIENT
TABLE
169
RAT
I
PLASMAGLUTATHIONEINSELENIUM-DEFICIENT AND CONTROLRATS" ?l a. Kinetics of plasma glutathione Fractional rate of disappearance V, (ml/100 g body wt) CIT (ml plasma/min a 100 g body b. Plasma glutathione Aorta Renal vein
(min-‘) wt)
Selenium
deficient
4 4 4
0.20 + 0.04 22.5 + 3.9 4.42 t 0.91
0.22 * 0.04 21.4 +- 4.1 4.59 * 1.05
5 6
17 * 5.46 5.2 k 0.9’
40 +- 3.4b 9.5 f 1.6b
-
0.69
0.76
6 -
1.52 + 0.27’ 1.05 0.24
2.64 31 0.62” 2.01 0.44
levels
c. Kinetics of glutathione removal by the kidneys Renal extraction ratio Rena1 plasma flow (ml plasma/min * 100 g body ,wt) (ml plasma/min . 100 g body wt) CL, ~L”JC1T (LValues *.“Values test.
Control
are means + SD. with the same superscript
are statistically
selenium-deficient plasma than in control plasma and because the clearance did not change, the:se results indicate that selenium deficiency doubles the rate of production of plasma glutathione. The kidney is thought to be the major organ responsible for the removal of plasma glutathione (3-5). The increased turnover of plasma glutathione in selenium-deficient rats prompted an examination of its removal by the kidneys. y-GTP is the enzyme responsible for the initial step of plasma glutathione degradation. The enzymatic activity of renal r-GTP was 1.91 t 0.12 pmol p-nitroaniline formed/mg protein. min in control kidney homogenate and 1.94 + 0.20 /*mol p-nitroaniline formed/mg protein - min in selenium-deficient kidney homogenate. Thus, selenium deficiency does not affect the total activity of y-GTP in the kidney. Arterial and renal vein plasma glutathione were measured in seleniumdeficient and control rats (Table Ib). The selenium-deficient kidney removed a greater amount of glutathione per milliliter plasma than did the control kidney (30 nmol vs 1.2 nmol) due apparently to a slight increase in extraction ratio and to
different
from
each other
(P < 0.05)
by Student’s
t
the higher plasma glutathione concentration. In addition, renal plasma flow was higher in selenium-deficient rats than in control rats (Table Ic). The increase in renal plasma flow and the slightly higher extraction ratio in selenium deficiency resulted in an increased renal plasma clearance of glutathione (Clrenal) in comparison to control. Renal clearance accounted for 44% of the systemic clearance in seleniumdeficient rats but only for 24% in controls. Renal removal of plasma glutathione was also examined by determining the amount of radioactivity retained by the kidney after intravenous administration of [?S]GSH. The selenium-deficient kidney retained 1.3-1.7 times more injected radioactivity than did the control kidney (Fig. 3). This is consistent with the increased removal of plasma glutathione in the selenium-deficient rat. The relatively constant amount of radioactivity retained by selenium-deficient or control kidney is supportive of the rapid turnover of glutathione in the kidney (4). After intravenous administration of [35S]GSH, 24-h urine collection showed that identical percentages of the total injected [35S] were present in control and
170
HILL
AND
FIG. 3. Amount of ‘?S retained by control (0) and selenium-deficient (A) kidneys expressed as a percentage of the [Y$JGSH injected. The radioactivity remaining in the kidneys following the intravenous injection of a tracer dose of [?S]GSH (see Experimental) was determined at the time points shown. Each point is the mean of at least three animals. One SD is represented by the bracket.
selenium-deficient urine (15 + 3% (n = 5) and 15 + 2% (n = 5), respectively). Because the initial glutathione pool is twice as great in selenium-deficient plasma as in control, these urinary data are consistent with a twofold increase in the metabolism of plasma GSH by the selenium-deficient kidney. Even though there was increased removal of plasma glutathione by the selenium-deficient kidney, there was no difference between selenium-deficient and control kidney glutathione concentrations. The control kidney contained 2.32 + 0.30 pmol GSH equivalents/g (n = 8) and the selenium-deficient kidney contained 2.48 +- 0.29 pmol GSH equivalents/g (n = 9). Thus glutathione metabolism (removal from the plasma and degradation) in the kidney appears to be affected by selenium deficiency. DISCUSSION
Glutathione metabolism is markedly affected by selenium deficiency. Glutathione synthesis and release by the liver are doubled in selenium deficiency (1). Plasma giutathione concentration is doubled as a
BURK
result. The plasma glutathione half-life is the same in selenium-deficient and control rats. Since the glutathione concentration is two times greater in selenium-deficient plasma than in control plasma, finding the same plasma clearance in both nutritional states indicates that the amount of plasma glutathione turned over in selenium deficiency is two times greater than in control. The total amount of glutathione released into the systemic circulation can be estimated from arterial glutathione concentration and Clr. From this calculation, it is estimated that the seleniumdeficient rat releases 2.4 times more glutathione into the extracellular fluid than does the control rat (182 nmol GSH equivalents/min -100 g body wt in selenium deficiency vs 75 nmol GSH equivalents/ min * 100 g body wt in control). The value of 75 nmol/min -100 g body wt calculated for the input of glutathione into (or its removal from) the circulation of control rats agrees reasonably well with that of 59 nmol GSH equivalents/min - 100 g body wt reported by Lauterburg et al. (17). Hepatic glutathione release has been proposed as the primary source of plasma glutathione (17). Using perfused liver data from Ref. (18) and liver wt/lOO g body wt for control and selenium-deficient livers (see Experimental), we estimate liver glutathione release to be 35 and 72 nmol GSH equivalents/min. 100 g body wt in control and selenium-deficient rats, respectively. These values are less than half the glutathione calculated to be released into the extracellular fluid (see above). They suggest that there may be other sites of glutathione release into the circulation. This assumes that release of glutathione by the perfused liver is the same as in vivo release and clearly this may not be the case. The kidney is known to be a major site of plasma glutathione removal. Hahn et al. (3) demonstrated in mice that a large amount of radioactivity was present in the kidney 5 min after intravenous injection of [14C]glycine-labeled glutathione. Griffith and Meister (4) measured plasma glutathione levels in rats which had had
PLASMA
GLUTATHIONE
IN
THE
a bilateral nephrectomy and compared them with the levels in sham-operated rats. They concluded that 67% of plasma glutathione removal occurred via the kidneys. Haberle et aL (5) measured arterial and renal vein plasma glutathione concentrations to assess renal removal of plasma glutathione. They concluded that the kidney accounted for approximately 50% of the glutathione removed from the plasma. The results presented in Table Ic indicate that renal plasma glutathione clearance accounts for 24% of the systemic plasma glutathione clearance in control rats. These results suggest that while the kidney removes a large amount of plasma glutathione, other sites are also responsible for the :removal of most of the plasma glutathione. The amount of glutathione removed by the kidneys can be calculated from the product of the arteriovenous difference in glutathione concentration and the renal plasma flow. The selenium-deficient kidney removes 81 nmol GSH equivalents/ min. 100 g body wt from the plasma compared with 18 nmol GSH equivalents removed/mine 100 g body wt by the control kidney. The increased removal of 63 nmol GSH equivalents/min 0100 g body wt by the selenium-deficient kidney accounts for 60% of the increased extracellular glutathione turnover in the selenium-deficient rat (182 nmol GSH equivalents/min * 100 g body wt for selenium-deficient vs 75 nmol GSH equivalents/min -100 g body wt for control). Thus, a major portion of the increased extracellular glutathione turned over in selenium deficiency enters the kidney. Possible reasons for this increased extracellular glutathione turnover in selenium (deficiency are (a) increased kidney requirement for glutathione or cysteine and (b) need for increased GSH concentration in extracellular fluid. In summary, selenium deficiency affects glutathione metabolism in rat liver, plasma, and kidney. The increased release of glutathione from selenium-deficient rat liver results in increased plasma glutathione. There is a corresponding increase in the removal of plasma glutathione by
SELENIUM-DEFICIENT
RAT
171
the selenium-deficient kidney. Removal of other sites may also be increased but over half the increase appears to be accounted for by the kidney. This indicates that selenium deficiency affects the interorgan metabolism of glutathione. ACKNOWLEDGMENTS The authors thank Mr. J. M. Lane for technical assistance, Dr. H. J. Reineck, Mrs. T. Craig, and Mrs. M. Parma for renal blood flow measurements, Mrs. S. R. B. Allerheiligen for helpful discussions concerning pharmacokinetics, Dr. T. M. Ludden for helpful suggestions and critical review of the manuscript, and Mrs. R. E. Ortiz for secretarial assistance. REFERENCES 1. HILL, K. E., AND BURK, R. F. (1982) J. Viol. Chem. 257, 10668-10672. 2. MCINTYRE, T. M., AND CURTHOYS, N. P. (1980) Int. J Biochem. 12, 545551. 3. HAHN, R., WENDEL, A., AND FLOHC, L. (1978) B&him. Biophys. Acta 539, 324-337. 4. GRIFFITH, 0. W., AND MEISTER, A. (1979) Proc. Natl. Acaa! Sci. USA 76, 5606-5610. 5. HABERLE, D., WAHLL~NDER, A., AND SIES, H. (1979) FEBS Lett. 108, 335-340. 6. LAWRENCE, R. A., AND BURK, R. F. (1978) J. N&r. 108, 211-215. 7. LAWRENCE, R. A., AND BURK, R. F. (1976) Biochem. Biophys. Res. Commun. 71, 952-958. 8. TIETZE, F. (1969) Anal Biochem. 27, 502-522. 9. ADAMS, J. D., LAUTERBURG, B. H., AND MITCHELL, J. R. (1983) J. Pharmacol Exp. Ther. 227,749754. 10. ANDERSON, M. E., AND MEISTER, A. (1980) J. Biol. Chem. 255, 9530-9533. 11. WENDEL, A., AND CIKRYT, P. (1980) FEBS Lett. 120, 209-211. 12. GRIFFITH, 0. W. (1981) J. Biol. Chem. 256, 49004904. 13. REINECK, H. J., O’CONNOR, G. J., LIFSCHITZ, M. D., AND STEIN, J. H. (1980) J. Lab. Clin. Med. 96, 356-362. 14. GIBALDI, M., AND PERRIER, D. (1982) Pharmacokinetics, pp. l-55, Dekker, New York. 15. BEATTY, P., AND REED, D. J. (1981) Biochem. Pharmacol. 30, 1227-1230. 16. BANNAI, S. (1984) B&him. Biophys. Acta 779, 289-306. 17. LAUTERBURG, B. H., AND MITCHELL, J. R. (1981) J. CZin. Invest. 67, 1415-1424. 18. KRIETER, P. A., ZIEGLER, D. M., HILL, K. E., AND BURK, R. F. (1985) B&hem. PharmacoZ. 34, 955-960.