BIOCHEMICAL
MEDICINE
AND
METABOLIC
BIOLOGY
45, 65-73 (191)
Human Jejunal Glutathione Reductase: Purification and Evaluation of the NADPH- and Glutathione-Induced Changes in Redox State HAMD~ 66tis
AND NAZMI
OZER
Department of Biochemistry, Faculty of Medicine, Hacettepe University, Ankara, Turkey Received June 21, 1990 Human proximal jejunal glutathione reductase (EC 1.6.4.2) was purified to homogeneity by affinity chromatography on 2’, 5’-ADP-Sepharose 4B. In most of its molecular and kinetic properties, the enzyme resembled glutathione reductase from other sources: The subunit mass was 56 kDa; the isolectric point and pH optimum were 6.75 and 7.25, respectively; Michaelis constants, determined at pH 7.4, 37°C. fell within the range of previously reported values [K,(NADPH) = 20 PM, K,(GSSG) = 80 PM]. The response of the enzyme to reducing conditions, on the other hand, had unique features: Preincubation with 1 mM NADPH resulted in 90% loss of activity which could be partially reversed by 2 mM GSSG, but not GSH. (Treatment with GSSG regenerated 68% of the original activity.) Reduction by GSH also caused inactivation which potentially amounted to > 80%. This inactivation could not be reversed by GSSG. The protective effect of GSSG against inactivation by GSH was studied. Except where [GSSG] far exceeded [GSH], the presence of GSSG in the preincubation medium decreased the extent of inhibition without affecting the rate constant for approach to equilibrium activity. At [GSSG] > [GSH] a decrease in the rate constant for inactivation was also observed. The results were interpreted in terms of a three-step mechanism: (I) preequilibrium reduction of E,, to Ered; (2) rate-limiting change in conformation from Ered to Elredr and (3) irreversible conversion to catalytically inferior products. 0 1991 Academic Press. Inc.
Glutathione reductase (EC 1.6.4.2) is a flavoprotein which catalyzes the NADPH-dependent reduction of oxidized glutathione. The GSSG/GSH redox system has been implicated in the regulation of metabolic activity (1). Reduced glutathione contributes to the maintenance of the integrity of cellular membranes and plays an important role in detoxification [2,3]. Hence glutathione reductase has been widely studied with respect to structure and kinetic mechanism (4-15). Studies on enzyme isolated from various mammalain sources and from yeast have revealed common properties such as a homodimeric structure (M, = lOO,OOO-125,000) with 1 FAD/subunit, a branching reaction mechanism involving ping-pong and sequential pathways, and an inhibitory effect of NADPH in the absence of GSSG. However, differences exist in detail (4,7,13) which might be due to differences in the species or tissue source of the enzyme used. The following is a report on human small intestinal glutathione reductase. The 65 0885-4505/91 $3.00 Copyright 0 1991 by Academic Press. Inc. All rights of reproduction in any form reserved.
66
HUMAN
JEJUNAL
GLUTATHIONE
REDUCTASE
small intestine is an important route of entry for xenobiotics, and characterization of glutathione reductase, together with information on glutathione S-transferases (16), should allow assessment of the glutathione-related detoxifying potential of this tissue. This study also provides an opportunity for comparing the properties of glutathione reductases from different tissues of the same organism. The behavior of human erythrocyte glutathione reductase has been described previously (57). MATERIALS AND METHODS Chromatographic media were obtained from Pharmacia-LKB, Sweden. GSH and GSSG were from Boehringer-Mannheim, FRG. All other chemicals were standard commercial products obtained from Sigma or Aldrich, USA. The tissue source of glutathione reductase was a surgical specimen obtained from the proximal jejunum of a 30-year-old female patient undergoing gastrojejunostomy indicated by chronic duodenal ulcer and pyloric stenosis. Enzyme pur$cation. Except where specified, all purification steps were carried out at OXC. The jejunal specimen was washed with ice-cold physiological saline and the mucosal cells were scraped off carefully with a scalpel. The cells (35g) were homogenized in 75 ml 0.25 M sucrose containing 1 mM EDTA and 0.2 mM dithioerythritol (DTE), using a Virtis homogenizer at 23,000 rpm for 2 min. The homogenate was centrifuged at 20,OOOg for 30 min. The supernatant obtained was recentrifuged at 105,OOOg for 1 hr. The cytosolic fraction was applied to a 5.5 x 17-cm column of Sephadex G-25 equilibrated with 10 mM Tris-HCl (pH 7.8) containing 1 mM EDTA and 0.2 mM DTE. Effluent fractions containing glutathione reductase and glutathione S-transferase were processed through hexylglutathione-Sepharose 4B to remove the latter enzyme. Unretained material was applied to a 2 x lo-cm column of 2’,5’-ADP-Sepharose 4B preequilibrated with 40 mM potassium phosphate buffer, pH 7.4. The column was washed successively with 400 and 40 mM phosphate buffer, pH 7.4. Glutathione reductase was eluted with the 40 mM buffer supplemented with 0.5 mM NADPH. Active fractions were pooled and applied to DEAE-Sepharose CL-6B (0.9 x 5 cm) preequilibrated with 40 mM potassium phosphate, pH 7.4. Elution was carried out at room temperature, using the same buffer. This procedure resolved glutathione reductase (eluting at 40 mM potassium phosphate) from NADP+ and NADPH, which eluted at 110 and 190 mM buffer, respectively. The DEAESepharose CLdB step was repeated to remove residual cofactor and to concentrate the enzyme. Chromatofocusing. Chromatofocusing of GSSGR (21 pg) was performed on PBE 94 (0.6 x 3 cm) preequilibrated with 25 mM imidazole-HCl, pH 7.4. Elution was carried out using Polybuffer 74 (12 ml) prepared by 1:9.6 dilution of the commercial stock and adjustment of pH to 4.3 with HCl. SDS-polyactylumide gel electrophoresis. The subunit molecular weight and purity of the enzyme were tested by SDS-PAGE according to Laemmli (17). Slab gels of 12.5% acrylamide were used. Staining was done with Coomassie brilliant blue R-250.
ocliis
67
AND GzER
TABLE 1 Purification of Human Jejunal Glutathione Reductase Step 105,000 g supematant, Sephadex G-25 eluate 2’,5’-ADP-Sepharose 4B eluate First DEAE-Sepharose CLdB Second DEAESepharose CLdB a An underestimation,
Volume (ml)
Protein (mg)
Activity (units)
Sp. Act. W/w3
280
1192
107
0.059
1
100
50
0.55
54”
98
1661
50
56
0.52
82
158
2678
II
0.36
81
225
3814
76
3.6
Enrichment (fold)
Yield %
due to the inhibitory effect of NADPH in the eluate.
Assays. Activity was determined at 37°C in 100 mMpotassium phosphate buffer (pH 7.4) containing 4 mM EDTA, 1 mM GSSG, and 0.1 mM NADPH. The reaction was initiated by the addition of enzyme and followed by monitoring the absorbance at 340 nm. One unit of enzyme was defined as that amount catalyzing the consumption of 1 pmole NADPH/min under the assay conditions described. Specific activities were calculated on the basis of protein concentration determined by the method of Bradford (18). Long-term Effects of NADPH, GSSG, and GSH. The enzyme (4.5-7 pg/ml) was preincubated at 37°C in 100 mM potassium phosphate buffer (pH 7.4) containing appropriate concentrations of the ligand(s) under study. Residual activity was determined by 170-fold dilution of aliquots of the preincubation mixture into the assay medium. RESULTS The purification procedure described achieved about 4000-fold enrichment of human jejunal glutathione reductase with 76% yield (Table 1). The enzyme appeared homogeneous in SDS-PAGE and had a subunit mass of 56 kDa (Fig. 1) and a specific activity of 225 U/mg. FAD content was found to be 1 per subunit, by taking the absorptivity of enzyme-bound FAD as 11.3 mM- ’ cm- ’ at 462 nm (8). The enzyme was effectively free of NADPH, as verified by its response to added GSSG (see below). Isoelectric properties. The p1 of the enzyme was estimated by chromatofocusing, which gave an activity peak centered at pH 6.75 (Fig. 2). The pH optimum. The effect of pH on activity was determined at 1 mM GSSG/O.l mM NADPH. At each pH the observed activity was corrected for specific buffer effects by performing assays at three different buffer concentrations and extrapolating to zero buffer concentration. Optimal activity was obtained at pH 7.25. Michaelis parameters. In the [GSSG] range = 0.02-2.4 mM and the [NADPH] range 0.04-0.2 mM, Lineweaver-Burk plots for GSSG at different fixed levels of
68
HUMAN
JEJUNAL
GLUTATHIONE
REDUCTASE
-
f
f f + 1. SDS-polyacrylamide gel electrophoresis of affinity-purified human jejunal glutathione reductase. The arrows correspond to the points of migration of the standards (starting from the cathode: bovine serum albumin, 66,000; ovalbumin, 45,000; rat liver GST l-l, 25,000; lysozyme, 14,300; cyt c, 12,400). FIG.
NADPH and for NADPH at different fixed levels of GSSG were parallel, indicative of a ping-pong mechanism (19). Secondary double-reciprocal plots of V,,, versus the concentration of the fixed substrate (7) gave K,(NADPH) = 20 PM and KJGSSG) = 80 PM (Fig. 3). Inactivation by NADPH. Preincubation of purified glutathione reductase with 1 mM NADPH resulted in progressive loss of activity (Fig. 4). Approximately 65% of lost activity was recovered upon addition of 2 mM GSSG, while 2 mM GSH was without effect.
DH
FIG.
2.
Chromatofocusing
profile of glutathione reductase: (0) activity, (0) pH.
ocltis
0
69
AND GZER
IO
30
40
1 I s $-I plots for NADPH (0) and for GSSG (0) at different FIG. 3. Intercept replots of Lineweaver-Burk fixed concentrations of the second substrate.
Effect of GSH. Reduced glutathione also caused rapid inactivation of glutathione reductase (Fig. 5). The highest degree of inhibition (73%) was obtained at 0.4 mM GSH. Further increase in [GSH] had a protective effect, which appeared to arise from the presence of endogenous GSSG in the GSH stocks, amounting to 11 ? 2% of th GSH equivalents; i.e., a solution nominally 1 mM
0-l 0
10
20 Time
30 (min)
40
180
inactivation of glutathione reductase. (A) Decay of activity in the presFIG. 4. NADPH-induced ence of 1 mM NADPH. (B) Activity following addition of 2 mM GSSG. (0) Activity following addition of 2 rnM GSH. GSSG and GSH were added to separate aliquots of the inactivation mixture at the times indicated by the arrows). (A) Stability of the enzyme in the absence of ligands; (0) stability of the enzyme in the presence of 2 mM GSSG. In preincubation. [El = 4.5 pg/ml.
70
HUMAN
JEJUNAL
GLUTATHIONE
REDUCTASE
FIG. 5. (A) GSH-induced inactivation of glutathione reductase. Nominal concentration of GSH: a, 0.1 mM; b, 0.4 mM; c, 5mM; d, 10 mM. (See text for comments on actual GSH concentrations.) (B) Semilogarithmic plots for the approach to equilibrium activity at 10 mM GSH. In preincubation [El = 7 pg/ml.
in GSH contained 0.89 mM GSH and 0.055 mM GSSG.’ The possible effect of GSSG on GSH-induced inactivation was put to test (Fig. 6). While no significant reversal of inactivation was observed (Fig. 6, curve c), addition of GSSG at the onset of inactivation by GSH altered both the rate and the extent of inhibition (Fig. 6, curves a and b).
b
oc 0
IO
20
30 TimeCmifl)
40
50
-
60
FIG. 6. Effect of oxidized glutathione on GSH-induced inactivation. Decay of activity in the presence of (0) 0.4 mM GSH and (0) 0.4 mM GSH + 2 mM GSSG. (A) Response of inactivated enzyme to added GSSG (2 mM). In preincubation, [El = 7 kg/ml.
’ These values were estimated (a) by treating freshly prepared GSH solutions, nominally 0.5 mM in GSH, with 0.1 mM NADPH and excess yeast glutathione reductase (Boehringer) and noting the change in absorbance at 340 nm, and (b) by titrating the stock solutions with DTNB to deduce the actual concentration of GSH.
71
8cxYs AND OZER 2GSH
GSSG E;ed
2GSH Preequilibrium
+
Products
GSSG SIOW SCHEME
Fast 1.
DISCUSSION In properties such as molecular mass, isoelectric point, pH optimum, and affinity for the substrates GSSG and NADPH, human jejunal glutathione reductase was found to resemble its counterpart in the human erythrocyte as well as those in other sources. The response of the enzyme to preincubation with NADPH and the protective and reactivating effects of GSH and GSSG, on the other hand, differed significantly from what was reported earlier for human erythrocyte and yeast glutathione reductases (7,13). These properties were therefore studied in further detail. Effect ofNADPH. Treatment of jejunal glutathione reductase with mM NADPH resulted in 91% inhibition of activity with a half-life of 1.5 min (Fig. 4). Both the rate and the extent of inhibition were similar to those for yeast glutathione reductase (13), but not for the human erythrocyte enzyme which is only 50% inhibited by 1 mM NADPH (7). GSSG (2 mM) caused reactivation to 68% of the original activity, as observed with mouse liver and yeast glutathione reductase (4,13) and qualitatively reported for erythrocyte glutathione reductase (7). The most significant departure in the inhibition pattern of jejunal glutathione reductase by NADPH was that, contrary to earlier reports (7,13), GSH neither protected the enzyme against inactivation nor caused reversal of inhibition. The marginal increase in activity observed upon addition of 2 mM GSH to NADPH-inactivated enzyme (Fig. 4) is more likely to be due to GSSG contamination in the GSH stocks (see above) than GSH per se. Effect of GSH. Reduced glutathione caused inactivation of jejunal glutathione reductase (Fig. 5) with a rate constant that was independent of reagent concentration in the range 0.1-10 mM (nominal [GSH]): The decay to equilibrium activity, which conformed to first-order kinetics (Fig. 5b), had a half-life of 0.8 f 0.1 min. This constancy of the kinetic parameters suggests that the rate-limiting step in GSH-induced inactivation may be a conformational change which follows the initial reductive step(s). The extent of inhibition, on the other hand, was found to vary with glutathione concentration as implied by Fig. 5 and supported by the data in Fig. 6. This observation compares favorably with results obtained with yeast glutathione reductase (13) and is consistent with a mechanism (Scheme 1) in which a rate-limiting change in conformation (from Ered to Efred) is succeeded by further steps leading to at least two products, a low-specific-activity form and a high-specific-activity form, the latter being favored by the presence of GSSG. Although endogenous (or added) GSSG would also be expected to affect the level of Ered relative to Etotal, and hence the rate constant for inactivation
72
HUMAN
JEJUNAL
GLUTATHIONE
REDUCTASE
(cf. Eqs. [l]-[3]), this is observed only when additional GSSG is introduced the system (Fig. 6, curve b; half-life for inactivation, 8.5 min). [Et - E,,,J[GSH]2 - d(activity) dt
(1)
= K
= &%d1 K[GSH]’ = kf[GSSG] + K[GSH]’
into
(2)
Et
(3)
Under the conditions in Fig. 5, the [GSSG] term must be negligible relative to K[GSH]‘. Scheme 1 would also account for the inability of GSSG to reverse GSH-induced inactivation (Fig. 6, curve c). The potentially reversible Eox-E’,d conversion (see below) becomes effectively unidirectional by the fast processes leading to products. The model also fits the data on NADPH-induced inactivation with the following restrictions: (1) The different conformational and covalent states need not be identical to those in the GSH-induced inactivation system. (2) Since reactivation by GSSG was observed, conversion of Efred to products must be slow, such that at the time of addition GSSG (Fig. 4), a significant fraction (ca. 60%) of the enzyme population exhibiting low activity is still in the E’ state which retains its capacity to revert to E,,. The increased stability of E’ in the NADPH system may be due to the formation of an E’ * NADPH complex (20). In the preceding argument, the primary change leading to inactivation has been taken to be a conformational change at the level of the reduced, dimeric enzyme. Alternative interpretations could be based on the reported tendency of NADPHreduced human erythrocyte glutathione reductase to aggregate (7) or of NADPHreduced Escherichia coli enzyme to dissociate into monomers (21). The first possibility may be ruled out, since, if rate-limiting, it would necessarily yield higher than first-order kinetics of inactivation. This was observed neither in the NADPH-induced changes nor in the GSH-induced changes in the activity of the jejunal glutathione reductase. Hence, in the present case, aggregation (if present) could only be the result rather than the cause of inactivation. The possibility of inactivation by dissociation into monomers remains to be tested. However, the present data point to finite residual activity in the GSH-induced inactivation system which conflicts with the known structure of the active site of human erythrocyte glutathione reductase, which is composed of domains contributed by different subunits (22) and should therefore lose its catalytic potential upon dissociation of the dimer. Human jejunal glutathione reductase, while a typical representative of its class with respect to basic molecular and kinetic properties, undergoes NADPH- and GSH-induced changes in redox state and conformation which are at variance with the qualitative information available on human erythrocyte glutathione reductase (7). A detailed comparative study is necessary to evaluate any tissuespecific differences in the behavior of the enzyme and their relevance in the regulation of GSH/GSSG levels in different cells.
iieus
AND OZER
73
REFERENCES 1. Gilbert, H. F., in “Methods in Enzymology” (F. Wold and K. Moldave, Eds.), Vol. 107, pp. 330-351, Academic Press, Orlando, FL, 1984. 2. Williamson, J. M., Boettcher, B.. and Meister, A., Proc. Nat/. Acad. Sci. USA 79, 6246-6249) (1982). 3. Carlberg, I.. and Mannervik. B., in “Methods in Enzymology” (A. Meister, Ed.), Vol. 113, pp. 484-490, Academic Press, Orlando, FL, 1985. 4. Lopez-Barea, J., and Lee, C. Y., Eur. 1. Biochem. 98, 487-499 (1979). 5. Scott, E. M., Duncan, 1. W., and Ekstrand, V., J. Biol. Chem. 238, 3928-3933 (1963). 6. Staal, G. E. J., and Veeger, C., Biochim. Biophys. Acta 185, 49-62. (1969). 7. Worthington, D. J.. and Rosemeyer, M. A.. Eur. J. Biochem. 67, 231-238 (1976). 8. Massey, V., and Williams, C. H., Jr. J. Biol. Chem. 240, 4470-4480 (1965). 9. Icen, A. L.. FEBS Lerr. 16, 29-32 (1971). 10. Mannervik, B., Biochem. Biophys. Res. Commun. 53, 1151-1158 (1973). 11. Moroff. G., and Brandt. K. G., Biochim. Biophys. Acta 410, 21-31 (1975). 12. Thieme. R., Pai, E. F., Schirmer, R. H., and Schulz, G. E., J. Mol. Biol. 152, 763-782 (1981). 13. Pinto, M. C., Mata, A. M., and Lopez-Barea, J.. Arch. Biochem. Biophys. 228, I-12 (1984). 14. Acan, N. L.. and Tezcan, E. F., FEBS Left. 250, 72-74 (1989). 15. Ghisla. S.. and Massey, V.. Eur. J. Biochem. 181, 1-17 (1989). 16. ozer, N., Erdemli, 6, Sayek, I., and iizer, I., Biochem. Med. Metab. Biol., 44, 142-150 (1990). 17. Laemmli, U. K., Nature (London) 227, 680-685 (1970). 18. Bradford, M. M., Anal. Biochem. 72, 248-254 (1976) 19. Segel, I. H., “Enzyme Kinetics,” p. 610. Wiley-Interscience, New York (1975). 20. Williams, C. H., Jr., Arscott, L. D., and Jones, E. T.. in “Flavins and Flavoproteins” (T. P. Singer, Ed.) pp. 455463. Elsevier, Amsterdam (1976). 21. Arscott, L. D., Drake, D. M., and Williams, C. H., Jr., Biochemistry 28, 3591-3598 (1989). 22. Schulz, G. E.. Schirmer. R. H., Sachsenheimer, W., and Pal, E. F., Nature (London) 373, 120124 (1978).
MEDICINE
AND
METABOLIC
BIOLOGY
45, 65-73 (191)
Human Jejunal Glutathione Reductase: Purification and Evaluation of the NADPH- and Glutathione-Induced Changes in Redox State HAMD~ 66tis
AND NAZMI
OZER
Department of Biochemistry, Faculty of Medicine, Hacettepe University, Ankara, Turkey Received June 21, 1990 Human proximal jejunal glutathione reductase (EC 1.6.4.2) was purified to homogeneity by affinity chromatography on 2’, 5’-ADP-Sepharose 4B. In most of its molecular and kinetic properties, the enzyme resembled glutathione reductase from other sources: The subunit mass was 56 kDa; the isolectric point and pH optimum were 6.75 and 7.25, respectively; Michaelis constants, determined at pH 7.4, 37°C. fell within the range of previously reported values [K,(NADPH) = 20 PM, K,(GSSG) = 80 PM]. The response of the enzyme to reducing conditions, on the other hand, had unique features: Preincubation with 1 mM NADPH resulted in 90% loss of activity which could be partially reversed by 2 mM GSSG, but not GSH. (Treatment with GSSG regenerated 68% of the original activity.) Reduction by GSH also caused inactivation which potentially amounted to > 80%. This inactivation could not be reversed by GSSG. The protective effect of GSSG against inactivation by GSH was studied. Except where [GSSG] far exceeded [GSH], the presence of GSSG in the preincubation medium decreased the extent of inhibition without affecting the rate constant for approach to equilibrium activity. At [GSSG] > [GSH] a decrease in the rate constant for inactivation was also observed. The results were interpreted in terms of a three-step mechanism: (I) preequilibrium reduction of E,, to Ered; (2) rate-limiting change in conformation from Ered to Elredr and (3) irreversible conversion to catalytically inferior products. 0 1991 Academic Press. Inc.
Glutathione reductase (EC 1.6.4.2) is a flavoprotein which catalyzes the NADPH-dependent reduction of oxidized glutathione. The GSSG/GSH redox system has been implicated in the regulation of metabolic activity (1). Reduced glutathione contributes to the maintenance of the integrity of cellular membranes and plays an important role in detoxification [2,3]. Hence glutathione reductase has been widely studied with respect to structure and kinetic mechanism (4-15). Studies on enzyme isolated from various mammalain sources and from yeast have revealed common properties such as a homodimeric structure (M, = lOO,OOO-125,000) with 1 FAD/subunit, a branching reaction mechanism involving ping-pong and sequential pathways, and an inhibitory effect of NADPH in the absence of GSSG. However, differences exist in detail (4,7,13) which might be due to differences in the species or tissue source of the enzyme used. The following is a report on human small intestinal glutathione reductase. The 65 0885-4505/91 $3.00 Copyright 0 1991 by Academic Press. Inc. All rights of reproduction in any form reserved.
66
HUMAN
JEJUNAL
GLUTATHIONE
REDUCTASE
small intestine is an important route of entry for xenobiotics, and characterization of glutathione reductase, together with information on glutathione S-transferases (16), should allow assessment of the glutathione-related detoxifying potential of this tissue. This study also provides an opportunity for comparing the properties of glutathione reductases from different tissues of the same organism. The behavior of human erythrocyte glutathione reductase has been described previously (57). MATERIALS AND METHODS Chromatographic media were obtained from Pharmacia-LKB, Sweden. GSH and GSSG were from Boehringer-Mannheim, FRG. All other chemicals were standard commercial products obtained from Sigma or Aldrich, USA. The tissue source of glutathione reductase was a surgical specimen obtained from the proximal jejunum of a 30-year-old female patient undergoing gastrojejunostomy indicated by chronic duodenal ulcer and pyloric stenosis. Enzyme pur$cation. Except where specified, all purification steps were carried out at OXC. The jejunal specimen was washed with ice-cold physiological saline and the mucosal cells were scraped off carefully with a scalpel. The cells (35g) were homogenized in 75 ml 0.25 M sucrose containing 1 mM EDTA and 0.2 mM dithioerythritol (DTE), using a Virtis homogenizer at 23,000 rpm for 2 min. The homogenate was centrifuged at 20,OOOg for 30 min. The supernatant obtained was recentrifuged at 105,OOOg for 1 hr. The cytosolic fraction was applied to a 5.5 x 17-cm column of Sephadex G-25 equilibrated with 10 mM Tris-HCl (pH 7.8) containing 1 mM EDTA and 0.2 mM DTE. Effluent fractions containing glutathione reductase and glutathione S-transferase were processed through hexylglutathione-Sepharose 4B to remove the latter enzyme. Unretained material was applied to a 2 x lo-cm column of 2’,5’-ADP-Sepharose 4B preequilibrated with 40 mM potassium phosphate buffer, pH 7.4. The column was washed successively with 400 and 40 mM phosphate buffer, pH 7.4. Glutathione reductase was eluted with the 40 mM buffer supplemented with 0.5 mM NADPH. Active fractions were pooled and applied to DEAE-Sepharose CL-6B (0.9 x 5 cm) preequilibrated with 40 mM potassium phosphate, pH 7.4. Elution was carried out at room temperature, using the same buffer. This procedure resolved glutathione reductase (eluting at 40 mM potassium phosphate) from NADP+ and NADPH, which eluted at 110 and 190 mM buffer, respectively. The DEAESepharose CLdB step was repeated to remove residual cofactor and to concentrate the enzyme. Chromatofocusing. Chromatofocusing of GSSGR (21 pg) was performed on PBE 94 (0.6 x 3 cm) preequilibrated with 25 mM imidazole-HCl, pH 7.4. Elution was carried out using Polybuffer 74 (12 ml) prepared by 1:9.6 dilution of the commercial stock and adjustment of pH to 4.3 with HCl. SDS-polyactylumide gel electrophoresis. The subunit molecular weight and purity of the enzyme were tested by SDS-PAGE according to Laemmli (17). Slab gels of 12.5% acrylamide were used. Staining was done with Coomassie brilliant blue R-250.
ocliis
67
AND GzER
TABLE 1 Purification of Human Jejunal Glutathione Reductase Step 105,000 g supematant, Sephadex G-25 eluate 2’,5’-ADP-Sepharose 4B eluate First DEAE-Sepharose CLdB Second DEAESepharose CLdB a An underestimation,
Volume (ml)
Protein (mg)
Activity (units)
Sp. Act. W/w3
280
1192
107
0.059
1
100
50
0.55
54”
98
1661
50
56
0.52
82
158
2678
II
0.36
81
225
3814
76
3.6
Enrichment (fold)
Yield %
due to the inhibitory effect of NADPH in the eluate.
Assays. Activity was determined at 37°C in 100 mMpotassium phosphate buffer (pH 7.4) containing 4 mM EDTA, 1 mM GSSG, and 0.1 mM NADPH. The reaction was initiated by the addition of enzyme and followed by monitoring the absorbance at 340 nm. One unit of enzyme was defined as that amount catalyzing the consumption of 1 pmole NADPH/min under the assay conditions described. Specific activities were calculated on the basis of protein concentration determined by the method of Bradford (18). Long-term Effects of NADPH, GSSG, and GSH. The enzyme (4.5-7 pg/ml) was preincubated at 37°C in 100 mM potassium phosphate buffer (pH 7.4) containing appropriate concentrations of the ligand(s) under study. Residual activity was determined by 170-fold dilution of aliquots of the preincubation mixture into the assay medium. RESULTS The purification procedure described achieved about 4000-fold enrichment of human jejunal glutathione reductase with 76% yield (Table 1). The enzyme appeared homogeneous in SDS-PAGE and had a subunit mass of 56 kDa (Fig. 1) and a specific activity of 225 U/mg. FAD content was found to be 1 per subunit, by taking the absorptivity of enzyme-bound FAD as 11.3 mM- ’ cm- ’ at 462 nm (8). The enzyme was effectively free of NADPH, as verified by its response to added GSSG (see below). Isoelectric properties. The p1 of the enzyme was estimated by chromatofocusing, which gave an activity peak centered at pH 6.75 (Fig. 2). The pH optimum. The effect of pH on activity was determined at 1 mM GSSG/O.l mM NADPH. At each pH the observed activity was corrected for specific buffer effects by performing assays at three different buffer concentrations and extrapolating to zero buffer concentration. Optimal activity was obtained at pH 7.25. Michaelis parameters. In the [GSSG] range = 0.02-2.4 mM and the [NADPH] range 0.04-0.2 mM, Lineweaver-Burk plots for GSSG at different fixed levels of
68
HUMAN
JEJUNAL
GLUTATHIONE
REDUCTASE
-
f
f f + 1. SDS-polyacrylamide gel electrophoresis of affinity-purified human jejunal glutathione reductase. The arrows correspond to the points of migration of the standards (starting from the cathode: bovine serum albumin, 66,000; ovalbumin, 45,000; rat liver GST l-l, 25,000; lysozyme, 14,300; cyt c, 12,400). FIG.
NADPH and for NADPH at different fixed levels of GSSG were parallel, indicative of a ping-pong mechanism (19). Secondary double-reciprocal plots of V,,, versus the concentration of the fixed substrate (7) gave K,(NADPH) = 20 PM and KJGSSG) = 80 PM (Fig. 3). Inactivation by NADPH. Preincubation of purified glutathione reductase with 1 mM NADPH resulted in progressive loss of activity (Fig. 4). Approximately 65% of lost activity was recovered upon addition of 2 mM GSSG, while 2 mM GSH was without effect.
DH
FIG.
2.
Chromatofocusing
profile of glutathione reductase: (0) activity, (0) pH.
ocltis
0
69
AND GZER
IO
30
40
1 I s $-I plots for NADPH (0) and for GSSG (0) at different FIG. 3. Intercept replots of Lineweaver-Burk fixed concentrations of the second substrate.
Effect of GSH. Reduced glutathione also caused rapid inactivation of glutathione reductase (Fig. 5). The highest degree of inhibition (73%) was obtained at 0.4 mM GSH. Further increase in [GSH] had a protective effect, which appeared to arise from the presence of endogenous GSSG in the GSH stocks, amounting to 11 ? 2% of th GSH equivalents; i.e., a solution nominally 1 mM
0-l 0
10
20 Time
30 (min)
40
180
inactivation of glutathione reductase. (A) Decay of activity in the presFIG. 4. NADPH-induced ence of 1 mM NADPH. (B) Activity following addition of 2 mM GSSG. (0) Activity following addition of 2 rnM GSH. GSSG and GSH were added to separate aliquots of the inactivation mixture at the times indicated by the arrows). (A) Stability of the enzyme in the absence of ligands; (0) stability of the enzyme in the presence of 2 mM GSSG. In preincubation. [El = 4.5 pg/ml.
70
HUMAN
JEJUNAL
GLUTATHIONE
REDUCTASE
FIG. 5. (A) GSH-induced inactivation of glutathione reductase. Nominal concentration of GSH: a, 0.1 mM; b, 0.4 mM; c, 5mM; d, 10 mM. (See text for comments on actual GSH concentrations.) (B) Semilogarithmic plots for the approach to equilibrium activity at 10 mM GSH. In preincubation [El = 7 pg/ml.
in GSH contained 0.89 mM GSH and 0.055 mM GSSG.’ The possible effect of GSSG on GSH-induced inactivation was put to test (Fig. 6). While no significant reversal of inactivation was observed (Fig. 6, curve c), addition of GSSG at the onset of inactivation by GSH altered both the rate and the extent of inhibition (Fig. 6, curves a and b).
b
oc 0
IO
20
30 TimeCmifl)
40
50
-
60
FIG. 6. Effect of oxidized glutathione on GSH-induced inactivation. Decay of activity in the presence of (0) 0.4 mM GSH and (0) 0.4 mM GSH + 2 mM GSSG. (A) Response of inactivated enzyme to added GSSG (2 mM). In preincubation, [El = 7 kg/ml.
’ These values were estimated (a) by treating freshly prepared GSH solutions, nominally 0.5 mM in GSH, with 0.1 mM NADPH and excess yeast glutathione reductase (Boehringer) and noting the change in absorbance at 340 nm, and (b) by titrating the stock solutions with DTNB to deduce the actual concentration of GSH.
71
8cxYs AND OZER 2GSH
GSSG E;ed
2GSH Preequilibrium
+
Products
GSSG SIOW SCHEME
Fast 1.
DISCUSSION In properties such as molecular mass, isoelectric point, pH optimum, and affinity for the substrates GSSG and NADPH, human jejunal glutathione reductase was found to resemble its counterpart in the human erythrocyte as well as those in other sources. The response of the enzyme to preincubation with NADPH and the protective and reactivating effects of GSH and GSSG, on the other hand, differed significantly from what was reported earlier for human erythrocyte and yeast glutathione reductases (7,13). These properties were therefore studied in further detail. Effect ofNADPH. Treatment of jejunal glutathione reductase with mM NADPH resulted in 91% inhibition of activity with a half-life of 1.5 min (Fig. 4). Both the rate and the extent of inhibition were similar to those for yeast glutathione reductase (13), but not for the human erythrocyte enzyme which is only 50% inhibited by 1 mM NADPH (7). GSSG (2 mM) caused reactivation to 68% of the original activity, as observed with mouse liver and yeast glutathione reductase (4,13) and qualitatively reported for erythrocyte glutathione reductase (7). The most significant departure in the inhibition pattern of jejunal glutathione reductase by NADPH was that, contrary to earlier reports (7,13), GSH neither protected the enzyme against inactivation nor caused reversal of inhibition. The marginal increase in activity observed upon addition of 2 mM GSH to NADPH-inactivated enzyme (Fig. 4) is more likely to be due to GSSG contamination in the GSH stocks (see above) than GSH per se. Effect of GSH. Reduced glutathione caused inactivation of jejunal glutathione reductase (Fig. 5) with a rate constant that was independent of reagent concentration in the range 0.1-10 mM (nominal [GSH]): The decay to equilibrium activity, which conformed to first-order kinetics (Fig. 5b), had a half-life of 0.8 f 0.1 min. This constancy of the kinetic parameters suggests that the rate-limiting step in GSH-induced inactivation may be a conformational change which follows the initial reductive step(s). The extent of inhibition, on the other hand, was found to vary with glutathione concentration as implied by Fig. 5 and supported by the data in Fig. 6. This observation compares favorably with results obtained with yeast glutathione reductase (13) and is consistent with a mechanism (Scheme 1) in which a rate-limiting change in conformation (from Ered to Efred) is succeeded by further steps leading to at least two products, a low-specific-activity form and a high-specific-activity form, the latter being favored by the presence of GSSG. Although endogenous (or added) GSSG would also be expected to affect the level of Ered relative to Etotal, and hence the rate constant for inactivation
72
HUMAN
JEJUNAL
GLUTATHIONE
REDUCTASE
(cf. Eqs. [l]-[3]), this is observed only when additional GSSG is introduced the system (Fig. 6, curve b; half-life for inactivation, 8.5 min). [Et - E,,,J[GSH]2 - d(activity) dt
(1)
= K
= &%d1 K[GSH]’ = kf[GSSG] + K[GSH]’
into
(2)
Et
(3)
Under the conditions in Fig. 5, the [GSSG] term must be negligible relative to K[GSH]‘. Scheme 1 would also account for the inability of GSSG to reverse GSH-induced inactivation (Fig. 6, curve c). The potentially reversible Eox-E’,d conversion (see below) becomes effectively unidirectional by the fast processes leading to products. The model also fits the data on NADPH-induced inactivation with the following restrictions: (1) The different conformational and covalent states need not be identical to those in the GSH-induced inactivation system. (2) Since reactivation by GSSG was observed, conversion of Efred to products must be slow, such that at the time of addition GSSG (Fig. 4), a significant fraction (ca. 60%) of the enzyme population exhibiting low activity is still in the E’ state which retains its capacity to revert to E,,. The increased stability of E’ in the NADPH system may be due to the formation of an E’ * NADPH complex (20). In the preceding argument, the primary change leading to inactivation has been taken to be a conformational change at the level of the reduced, dimeric enzyme. Alternative interpretations could be based on the reported tendency of NADPHreduced human erythrocyte glutathione reductase to aggregate (7) or of NADPHreduced Escherichia coli enzyme to dissociate into monomers (21). The first possibility may be ruled out, since, if rate-limiting, it would necessarily yield higher than first-order kinetics of inactivation. This was observed neither in the NADPH-induced changes nor in the GSH-induced changes in the activity of the jejunal glutathione reductase. Hence, in the present case, aggregation (if present) could only be the result rather than the cause of inactivation. The possibility of inactivation by dissociation into monomers remains to be tested. However, the present data point to finite residual activity in the GSH-induced inactivation system which conflicts with the known structure of the active site of human erythrocyte glutathione reductase, which is composed of domains contributed by different subunits (22) and should therefore lose its catalytic potential upon dissociation of the dimer. Human jejunal glutathione reductase, while a typical representative of its class with respect to basic molecular and kinetic properties, undergoes NADPH- and GSH-induced changes in redox state and conformation which are at variance with the qualitative information available on human erythrocyte glutathione reductase (7). A detailed comparative study is necessary to evaluate any tissuespecific differences in the behavior of the enzyme and their relevance in the regulation of GSH/GSSG levels in different cells.
iieus
AND OZER
73
REFERENCES 1. Gilbert, H. F., in “Methods in Enzymology” (F. Wold and K. Moldave, Eds.), Vol. 107, pp. 330-351, Academic Press, Orlando, FL, 1984. 2. Williamson, J. M., Boettcher, B.. and Meister, A., Proc. Nat/. Acad. Sci. USA 79, 6246-6249) (1982). 3. Carlberg, I.. and Mannervik. B., in “Methods in Enzymology” (A. Meister, Ed.), Vol. 113, pp. 484-490, Academic Press, Orlando, FL, 1985. 4. Lopez-Barea, J., and Lee, C. Y., Eur. 1. Biochem. 98, 487-499 (1979). 5. Scott, E. M., Duncan, 1. W., and Ekstrand, V., J. Biol. Chem. 238, 3928-3933 (1963). 6. Staal, G. E. J., and Veeger, C., Biochim. Biophys. Acta 185, 49-62. (1969). 7. Worthington, D. J.. and Rosemeyer, M. A.. Eur. J. Biochem. 67, 231-238 (1976). 8. Massey, V., and Williams, C. H., Jr. J. Biol. Chem. 240, 4470-4480 (1965). 9. Icen, A. L.. FEBS Lerr. 16, 29-32 (1971). 10. Mannervik, B., Biochem. Biophys. Res. Commun. 53, 1151-1158 (1973). 11. Moroff. G., and Brandt. K. G., Biochim. Biophys. Acta 410, 21-31 (1975). 12. Thieme. R., Pai, E. F., Schirmer, R. H., and Schulz, G. E., J. Mol. Biol. 152, 763-782 (1981). 13. Pinto, M. C., Mata, A. M., and Lopez-Barea, J.. Arch. Biochem. Biophys. 228, I-12 (1984). 14. Acan, N. L.. and Tezcan, E. F., FEBS Left. 250, 72-74 (1989). 15. Ghisla. S.. and Massey, V.. Eur. J. Biochem. 181, 1-17 (1989). 16. ozer, N., Erdemli, 6, Sayek, I., and iizer, I., Biochem. Med. Metab. Biol., 44, 142-150 (1990). 17. Laemmli, U. K., Nature (London) 227, 680-685 (1970). 18. Bradford, M. M., Anal. Biochem. 72, 248-254 (1976) 19. Segel, I. H., “Enzyme Kinetics,” p. 610. Wiley-Interscience, New York (1975). 20. Williams, C. H., Jr., Arscott, L. D., and Jones, E. T.. in “Flavins and Flavoproteins” (T. P. Singer, Ed.) pp. 455463. Elsevier, Amsterdam (1976). 21. Arscott, L. D., Drake, D. M., and Williams, C. H., Jr., Biochemistry 28, 3591-3598 (1989). 22. Schulz, G. E.. Schirmer. R. H., Sachsenheimer, W., and Pal, E. F., Nature (London) 373, 120124 (1978).
