Low- and high-Km transport of dinitrophenyl glutathione in inside out vesicles from human erythrocytes.

Kinetic studies on the low- and high-Km transport systems for S-2,4-dinitrophenyl glutathione (DNP-SG) present in erythrocyte membranes were performed using inside-out plasma membrane vesicles. The high-affinity system showed a Km of 3.9 microM a Vmax of 6.3 nmol/mg protein per h, and the low-affinity system a Km of 1.6 mM and a Vmax of 131 nmol/mg protein per h. Both uptake components were inhibited by fluoride, vanadate, p-chloromercuribenzoate (pCMB) and bis(4-nitrophenyl)dithio-3,3'-dicarboxylate (DTNB). The low-Km uptake process was less sensitive to the inhibitory action of DTNB as compared to the high-Km process. N-Ethylmaleimide (1 mM) inhibited the high-Km process only. The high-affinity uptake of DNP-SG was competitively inhibited by GSSG (Ki = 88 microM). Vice versa, DNP-SG inhibited competitively the low-Km component of GSSG uptake (Ki = 3.3 microM). The high-Km DNP-SG uptake system was not inhibited by GSSG. The existence of a common high-affinity transporter for DNP-SG and GSSG in erythrocytes is suggested.

S-(4-azidophenacyl)[35S]glutathione photoaffinity labeling of rat liver plasma membrane-associated proteins.

A method for the synthesis of the glutathione conjugate S-(4-azidophenacyl)[35S]glutathione is described. The compound was used for photoaffinity labeling of proteins present in canalicular membrane vesicles (CMV), sinusoidal membrane vesicles (SMV), mitochondria and microsomes from rat liver. Most of the radioactivity introduced by photoaffinity labeling of CMV appeared in the 25-29 kDa range. Further labeled proteins were observed in bands at 37, 105 and about 120 kDa. 79% of the 25-29 kDa associated radioactivity was recovered in the supernatant after extensive revesiculation (washing) of the vesicles, together with the 37 kDa protein. CMV and SMV contained glutathione S-transferase (GST) activity which in CMV was decreased by 75% by washing. Photolabeling of a mixture of purified basic GST subunits from rat liver resulted in a band pattern at 25-29 kDa similar to that in the membrane preparations. Isoelectric focusing of the CMV indicated the presence of basic soluble GST subunits. S-Hexylglutathione-Sepharose affinity chromatography showed reversible binding of photolabeled proteins at 25-29 kDa. Difference photoaffinity labeling with GSSG, S-hexylglutathione, taurocholate and phenylmethylsulfonyl fluoride decreased the radioactivity bound by GST, but not that introduced into the 105 kDa protein band present in CMV. It is concluded that membrane-associated basic GST isoenzymes are present in standard membrane vesicle preparations. In the cell, the function may be transport of GST-bound compounds across the membrane and protection of the membranes against electrophiles.

ATP-stimulated uptake of S-(2,4-dinitrophenyl)glutathione by plasma membrane vesicles from rat liver.

ATP-stimulated uptake of S-(2,4-dinitrophenyl)glutathione with a high activity of 0.35 nmol/min per mg protein is found in a rat liver plasma membrane vesicle preparation enriched in sinusoidal marker enzymes. Transport takes place into an osmotically active space. Vanadate and S-(azidophenacyl)glutathione inhibit transport, whereas Ca2+, EGTA and ouabain are without effect.

Identification of a hepatic plasma membrane glutathione S-transferase activated by N-ethylmaleimide.

Rat liver plasma membranes exhibit membrane-bound glutathione S-transferase activity. The specific activity in isolated canalicular membranes was 83 +/- 8 mU/mg protein and 50 +/- 3 mU/mg protein in the sinusoidal membranes. Whereas microsomal and outer mitochondrial glutathione S-transferases were stimulated seven and four-fold with N-ethylmaleimide, respectively, the plasma membrane activity was activated two-fold. Western blot analysis, using an antibody against the microsomal glutathione S-transferase, shows the presence of a 17 kDa protein in canalicular and sinusoidal membrane fractions. The antibody reaction was about three-fold higher in the canalicular compared to the sinusoidal membrane fraction. These data support the conclusion that the plasma membrane glutathione S-transferase is closely related to the microsomal and outer mitochondrial membrane enzyme.

Inhibition of taurocholate efflux from rat hepatic canalicular membrane vesicles by glutathione disulfide.

In right-side out rat hepatic canalicular membrane vesicles glutathione disulfide (GSSG) inhibited the efflux of taurocholate approx. 70% in the presence or approx. 55% in the absence of a valinomycin-mediated K+ diffusion potential; maximal inhibition occurred at 5 mM GSSG. The inhibition by GSSG was abolished by dithioerythritol. Neither dithioerythritol alone nor GSH inhibited taurocholate efflux. S-(2,4-Dinitrophenyl)glutathione and N-ethylmaleimide showed intermediate inhibitory effects.

Hepatic mercapturic acid formation: involvement of cytosolic cysteinylglycine S-conjugate dipeptidase activity.

The role of cysteinylglycine S-conjugate dipeptidases in the intrahepatic mercapturic acid pathway was investigated in rat liver. Subcellular compartmentation studies and liver perfusions were performed using monochlorobimane and bimane S-conjugates as model compounds. The major part (over 95%) of total hepatic cysteinylglycine S-conjugate dipeptidase activity was located in the cytosol. Lower specific activity appeared in the canalicular plasma membrane fraction. Similar hepatic localization of dipeptidase activity was seen in the guinea pig. In intact rat liver perfused with monochlorobimane, the major products were the glutathione S-conjugate (mBSG) and the cysteinylglycine S-conjugate (mBCG) in bile. Minor amounts of the cysteine S-conjugate (mBCys) and the mercapturic acid (mBNAc) were formed, indicating a limitation in further metabolism of the dipeptide S-conjugate in the biliary space. However, when the dipeptide S-conjugate was offered to the sinusoidal space in liver perfusions, substantial uptake and conversion to mBNAc was observed, and only trace amounts of the infused dipeptide appeared in bile. The data suggest that cytosolic cysteinylglycine S-conjugate dipeptidase as identified here is involved in hepatic mercapturic acid formation from sinusoidal cysteinylglycine S-conjugates. This is especially of significance for species such as guinea pig and human, in which dipeptide S-conjugates are generated in the sinusoidal domain of the liver due to the presence of high gamma-glutamyltranspeptidase activity.

Identification and quantitation of glutathione in hepatic protein mixed disulfides and its relationship to glutathione disulfide.

The amount of glutathione present in hepatic protein mixed disulfides was determined to be 20-30 nmole/g liver. This was established using two specific enzymatic methods: (a) the coupled assay with DTNB and glutathione (GSSG) reductase and (b) a newly developed test using GSH transferase and 1-chloro-2,4-dinitrobenzene for the estimation of GSH released from proteins after borohydride treatment; further, these results were confirmed by HPLC analysis. Thus, authentic glutathione makes up only 2-6% of the value for total protein mixed disulfides. The latter were determined with the generally employed o-phthalaldehyde assay, which is not necessarily specific for GSH. The amount of glutathione mixed disulfides depends linearly on the content of glutathione disulfide in the liver cell in the range studied. By increasing the GSSG levels from 20 to about 60 nmole/g liver with paraquat, nitrofurantoin or t-butyl hydroperoxide, glutathione protein mixed disulfides are increased by a similar amount.

Subunit specificity and organ distribution of glutathione transferase-catalysed S-nitrosoglutathione formation from alkyl nitrites in the rat.

Glutathione transferase (GST)-catalysed S-nitrosoglutathione (GSNO) formation from alkyl nitrites was determined with the homodimers 1-1, 2-2, 3-3, and 4-4 isolated from rat liver. The 4-4 isoform showed a high specificity for the alkyl nitrites. Total GST activities were studied in homogenates from different organs. The liver showed highest GST activity both with amyl nitrite and with 1-chloro-2,4-dinitrobenzene (CDNB) as substrate, the activity ratio of amyl nitrite over CDNB being 3.8. In lung and heart, these ratios were 6.2 and 5.7, respectively, indicating a selectivity of these organs for alkyl nitrite metabolism and GSNO formation.

Transport of ebselen in plasma and its transfer to binding sites in the hepatocyte.

In vivo transport in plasma and in vitro transfer of ebselen to binding sites in the hepatocyte were studied. More than 90% of intravenously administered ebselen in mouse plasma is bound by selenium-sulfur bonds to reactive thiols in serum albumin. In in vitro experiments the uptake of [14C]-ebselen from a complex prepared with bovine serum albumin (BSA) was determined in isolated perfused rat liver. Radioactive ebselen metabolites were excreted into bile. In isolated hepatocytes, radioactivity was bound to all subcellular organelles. Ebselen is transferred from the BSA complex to membrane-associated proteins after reductive cleavage of the Se-S bond effected by endogenous protein thiols. In contrast, when proteins were separated by dialysis membranes, ebselen transfer from its BSA complex occurred only in the presence of externally added reductants. Among the physiological reductants tested, ebselen release from the BSA complex was highest with glutathione (75%) and lowest with ascorbic acid (less than 10%). Quantitative release of ebselen from its BSA complex was only achieved by the combined action of reductant, notably 2-mercaptoethanol, and guanidine thiocyanate, suggesting that ebselen interacts with proteins by covalent Se-S bonds as well as by ionic charge interactions.

Protein S-thiolation and redox regulation of membrane-bound glutathione transferase.

Membrane-bound GST transferase (GSTm) occurs in hepatic microsomal and plasma membranes as well as in the outer mitochondrial membrane, and it is known to be activated by N-ethylmaleimide. We recently analysed the activation by GSSG in some detail. The approximately 5-fold stimulation is reversed upon reduction of GSSG by GSSG reductase. In steady-state experiments, the Kox value was determined to be 0.05, i.e. 20 times more GSSG than GSH produces half-maximal activation. Kox is independent of the total glutathione concentration, indicating that S-thiolation by mixed disulfide formation, rather than interchain or intrachain disulfide bridge formation, is responsible for activation. In Western blots, a 17.7 kDa band, in addition to the 17.3 kDa band, was detected upon treatment with GSSG or with GSH plus t-butyl hydroperoxide. We suggest that under oxidative stress, GSTm is activated through direct S-thiolation of the enzyme. Dethiolation occurs via thiol disulfide exchange governed by the cellular glutathione redox state.

Relation between glutathione redox changes and biliary excretion of taurocholate in perfused rat liver.

The influence of the intracellular glutathione status on bile acid excretion was studied in the perfused rat liver. Perturbation of the thiol redox state by short term additions of diamide (100 microM) or hydrogen peroxide (250 microM) or t-butyl hydroperoxide (250 microM) led to a reversible inhibition of biliary taurocholate release without affecting hepatic uptake; inhibition amounted to 45% for diamide and 90% for the hydroperoxides. Concomitantly, the bile acid accumulated intracellularly. Bile flow increased from 1.3 to 2.0 microliters X min-1 X g liver-1 upon infusion of taurocholate (10 microM); the latter value was suppressed to 1.2 microliters X min-1 X g liver-1 by the addition of t-butyl hydroperoxide (250 microM). Similarly, the hepatic disposition of another bile constituent, bilirubin, was suppressed by 70% upon addition of hydrogen peroxide. While the addition of hydrogen peroxide inhibited also the endogenous release of bile acids almost completely, endogenous bile flow was much less affected, decreasing from 1.3 to 1.0 microliters X min-1 X g liver-1. Measurement of [14C]erythritol clearance showed bile/perfusate ratios of about unity both in the absence and presence of hydrogen peroxide, suggesting canalicular origin of the bile under both conditions. In livers from Se-deficient rats low in Se-GSH peroxidase (less than 5% of controls), hydrogen peroxide inhibited taurocholate transport substantially less, providing evidence for the involvement of glutathione in mediating the inhibition observed in normal livers. The percentage inhibition of taurocholate release and intracellular glutathione disulfide (GSSG) content were closely correlated. The addition of t-butyl hydroperoxide caused a several-fold increase of biliary GSSG release, whereas biliary GSH release was even decreased. The results establish a role of glutathione in canalicular taurocholate disposition.