The preparation of ligandin with glutathione-S-transferase activity from porcine liver cytosol affinity chromatography on bromosulphophthalein-Sepharose.

A simple and rapid method for the purification of porcine ligandin with glutathione-S-transferase activity is presented. After ion-exchange chromatography on DEAE-Sephadex, ligandin is isolated from porcine liver cytosol by affinity chromatography on bromosulphophthalein-Sepharose and gel filtration of Sephadex G-100. Evidence is presented that the purified ligandin is homogeneous with respect to polyacrylamide-gel electrophoresis (7.5%) and sodium dodecylsulphate-gel electrophoresis. Physico-chemical investigations show that the purified ligandin has properties similar to those of ligandin isolated from other species with respect to molecular weight, amino-acid composition, secondary structure and catalytic activity. As is the case for human and rat ligandin, porcine ligandin binds bilirubin. Evidence is also presented that porcine liver cytosol contains several bromosulphophthalein-binding proteins with basic isoelectric points lacking catalytic activity.     

Triamcinolone activation of renal ammonia production.

The effects of scillaren and dinitrophenol on bilirubin excretion by the perfused rat liver were studied. Both compounds inhibited bile flow, scillaren by 20 to 40%, and dinitrophenol by 60 to 80%. Bilirubin excretion was also impaired. However, the effect of scillaren on bilirubin excretion was less than that on bile flow, as indicated by an increase in the bile bilirubin concentration, whereas dinitrophenol had a greater effect on bilirubin excretion than on bile flow. Dinitrophenol also inhibited the hepatic removal of unconjugated bilirubin from the perfusate, probably because it impaired the initial uptake and/or storage of unconjugated bilirubin by the perfused liver.     

Cytokeratin 18 is an M-cell marker in porcine Peyer's patches

The intermediate filaments of the dome epithelium of porcine Peyer's patches were studied by immunohistochemistry. The labelling patterns of monospecific antibodies directed against cytokeratins 8, 18 and 19 differed considerably. About 40% of the dome epithelial cells were intensely labelled by three different anti-cytokeratin 18 antibodies, indicating that large amounts of cytokeratin 18 are present in these cells. In order to verify that these cytokeratin-18-immunoreactive cells were M-cells, uptake studies using fluorescein-labelled yeast particles were performed. Numerous yeast particles were found exclusively in dome epithelial cells that were highly positive for cytokeratin 18, thus representing M-cells. In contrast, the content of cytokeratin 19 in M-cells was lower than that in neighbouring enterocytes. The labelling intensity of cytokeratin 8 did not differ between M-cells and enterocytes. In addition, the absence of vimentin and desmin from the dome epithelium of porcine Peyer's patches was demonstrated. The results show (1) that porcine M-cells differ from enterocytes in the composition of their cytoskeleton, (2) that cytokeratin 18 is a useful marker for detecting porcine M-cells and (3) that this marker directly correlas with M-cell function.     

Cell of origin of distinct cultured rat liver epithelial cells, as typed by cytokeratin and surface component selective expression

The cell of origin of the nonparenchymal epithelioid cells that emerge in liver cell cultures is unknown. Cultures of rat hepatocytes and several types of nonparenchymal cells obtained by selective tissue dispersion procedures were typed with monoclonal antibodies to rat liver cytokeratin and vimentin, polyvalent antibodies to cow hoof cytokeratins and porcine lens vimentin, and monoclonal antibodies to surface membrane components of ductular oval cells and hepatocytes. Immunoblot analysis revealed that, in cultured rat liver nonparenchymal epithelial cells, the anti-rat hepatocyte cytokeratin antibody recognized a cytokeratin of relative mass (Mr) 55,000 and the anti-cow hoof cytokeratin antibody reacted with a cytokeratin of Mr 52,000, while the anti-vimentin antibodies detected vimentin in both cultured rat fibroblasts and nonparenchymal epithelial cells. Analyses on the specificity of anti-cytokeratin and anti-vimentin antibodies toward the various cellular structures of liver by double immunofluorescence staining of frozen tissue sections revealed unique reactivity patterns. For example, hepatocytes were only stained with anti-Mr 55,000 cytokeratin antibody, while the sinusoidal cells reacted only with the anti-vimentin antibodies. In contrast, epithelial cells of the bile ductular structures and mesothelial cells of the Glisson capsula reacted with all the anti-cytokeratin and anti-vimentin antibodies. It should be stressed, however, that the reaction of the anti-vimentin antibodies on bile ductular cells was weak. The same analysis on tissue sections using the anti-ductular oval cell antibody revealed that it reacted with bile duct structures but not with the Glisson capsula. The anti-hepatocyte antibody reacted only with the parenchymal cells. The differential reactivity of the anti-cytokeratin and anti-vimentin antibodies with the various liver cell compartments was confirmed in primary cultures of hepatocytes, sinusoidal cells, and bile ductular cells, indicating that the present panel of antibodies to intermediate filament constituants allowed a clear-cut distinction between cultured nonparenchymal epithelial cells, hepatocytes, and sinusoidal cells. Indirect immunofluorescence microscopy on nonfixed and paraformaldehyde-fixed cultured hepatocytes and bile ductular cells further confirmed that both anti-hepatocyte and anti-ductular oval cell antibodies recognized surface-exposed components on the respective cell types.(ABSTRACT TRUNCATED AT 400 WORDS)     

Characteristics of the cell surface antigen, p72, associated with a variety of human tumours and mitogen-stimulated T-lymphoblasts

We recently purified two closely related 33 kDa proteins from rat hepatic cytosol, designated bile acid binder I and II, which selectively bind bile acids with comparable affinity as glutathione S-transferase B. This work has now been extended to human liver in which we have identified a similar cytosolic binding activity in the 30-40 kDa fraction from gel filtration. Subsequent chromatofocusing and hydroxyapatite chromatography resulted in the isolation of a homogeneous monomeric protein of 36 kDa. The binding affinity of this protein for lithocholate using the displacement of 1-anilino-8-naphthalenesulfonate (ANS) was 0.1 microM, whereas human hepatic glutathione S-transferases purified from glutathione affinity chromatography demonstrated no competitive displacement of ANS.     

Isolation and characterization of hyodeoxycholic-acid: UDP-glucuronosyltransferase from human liver

The enzyme hyodeoxycholic-acid: UDP-glucuronosyltransferase was purified about 230-fold from a solubilized human liver microsomal preparation utilizing anion-exchange chromatography, ampholyte-displacement chromatography and UDP-hexanolamine--Sepharose affinity chromatography. The homogeneity of the final enzyme preparation was judged by two criteria: the appearance of a single band of Mr 52000 in SDS/PAGE; the elution of a single peak in reversed-phase FPLC. The isolated enzyme catalyzed the glucuronidation of the 6 alpha-hydroxy bile acids hyodeoxycholic and hyocholic acids, and of the steroid hormone estriol, with a ratio of relative reaction rates of 13:1:2.7. UDP-glucuronosyltransferase activities toward the 3 alpha-hydroxy bile acid lithocholic acid, androsterone, testosterone, bilirubin and p-nitrophenol were not detectable in the pure enzyme preparation and were shown to be separated from enzyme activity toward hyodeoxycholic acid during ampholyte-displacement chromatography and/or UDP-hexanolamine--Sepharose affinity chromatography. Two-substrate kinetic analysis of hyodeoxycholic-acid-conjugating activity gave a sequential mechanism with apparent Km values of 12 microM and 4 microM for hyodeoxycholic acid and UDP-glucuronic acid, respectively. Phospholipids were required for reconstitution of maximal activity toward hyodeoxycholic acid. Phosphatidylcholine was the most effective activator of enzyme activity.     

Enzymatic conversion of bilirubin monoglucuronide to diglucuronide by rat liver plasma membranes.

Formation of bilirubin monoglucuronide from unconjugated bilirubin requires a microsomal enzyme, UDP-glucuronate glucuronyltransferase (EC 2.4.1.17). Conversion of bilirubin monoglucuronide to bilirubin diglucuronide, the major bilirubin conjugate in bile, was studied in subcellular fractions of rat liver. The highest specific activity for bilirubin diglucuronide formation occurred in a fraction highly enriched in plasma membranes. Studies of reaction stoichiometry and utilization of UDP-D-[14C]glucuronic acid revealed that conversion of bilirubin monoglucuronide to bilirubin diglucuronide is not catalyzed by UDP-glucuronyltransferase, and results from transglucuronidation of bilirubin monoglucuronide, with formation of bilirubin diglucuronide and unconjugated bilirubin. When unconjugated bilirubin was infused intravenously into rats at rates exceeding the maximal hepatic excretory capacity, bilirubin monoglucuronide accumulated in serum and bilirubin diglucuronide was found exclusively in bile as the predominant bilirubin metabolite. These results suggest that formation of bilirubin diglucuronide occurs at the surface membrane of the liver cell. Conversion of bilirubin monoglucuronide to bilirubin diglucuronide may play a role in the transport of bilirubin glucuronides from liver to bile.     

Purification, redox sensitivity, and RNA binding properties of SECIS-binding protein 2, a protein involved in selenoprotein biosynthesis.

In mammalian selenoprotein mRNAs, the highly structured 3' UTR contains selenocysteine insertion sequence (SECIS) elements that are required for the recognition of UGA as the selenocysteine codon. Our previous work demonstrated a tight correlation between codon-specific translational read-through and the activity of a 120-kDa RNA-binding protein that interacted specifically with the SECIS element in the phospholipid hydroperoxide glutathione peroxidase mRNA. This study reports the RNA binding and biochemical properties of this protein, SECIS-binding protein 2 (SBP2). We detected SBP2 binding activity in liver, hepatoma cell, and testis extracts from which SBP2 has been purified by anion exchange and RNA affinity chromatography. This scheme has allowed us to identify a 120-kDa polypeptide that co-elutes with SBP2 binding activity from wild-type but not mutant RNA affinity columns. A characterization of SBP2 biochemical properties reveals that SBP2 binding is sensitive to oxidation and the presence of heparin, rRNA, and poly(G). SBP2 activity elutes with a molecular mass of approximately 500 kDa during gel filtration chromatography, suggesting the existence of a large functional complex. Direct cross-linking and competition experiments demonstrate that the minimal phospholipid hydroperoxide glutathione peroxidase 3' UTR binding site is between 82 and 102 nucleotides, which correlates with the minimal sequence necessary for translational read-through. SBP2 also interacts specifically with the minimally functional 3' UTR of another selenoprotein mRNA, deiodinase 1.     

Expression of neural cell adhesion molecule in human liver development and in congenital and acquired liver diseases

In the liver, neural cell adhesion molecule (NCAM) is a marker of immature cells committed to the biliary lineage and is expressed by reactive bile ductules in human liver diseases. We investigated the possible role of NCAM in the development of intrahepatic bile ducts and aimed at determining whether immature biliary cells can contribute to the repair of damaged bile ducts in chronic liver diseases. Therefore, we performed immunohistochemistry for NCAM and bile duct cell markers cytokeratin 7 and cytokeratin 19 on frozen sections of 85 liver specimens taken from 14 fetuses, 10 donor livers, 18 patients with congenital liver diseases characterized by ductal plate malformations (DPMs), and 43 cirrhotic explant livers. Duplicated ductal plates and incorporating bile ducts during development showed a patchy immunoreactivity for NCAM, while DPMs were continuously positive for NCAM. Bile ducts showing complete or patchy immunoreactivity for NCAM were found in cirrhotic livers, with higher frequency in biliary than in posthepatitic cirrhosis. Our results suggest that NCAM may have a function in the development of the intrahepatic bile ducts and that NCAM-positive immature biliary cells can contribute to the repair of damaged bile ducts in chronic liver diseases.     

Human hepatic beta-glucuronidase: an enzyme kinetic study

beta-Glucuronidase (EC 3.2.1.31) was purified from human liver and its activity was determined by enzyme kinetic method employing phenolphthalein glucuronic acid (PGA) and conjugated bilirubin, primarily bilirubin diglucuronide purified from human bile, as substrates in the absence or presence of D-glucaro-1,4-lactone. The enzyme was capable of acting on both PGA and conjugated bilirubin with Michaelis constants of 0.435 mmol/l at 56 degrees C and 1.02 mmol/l at 37 degrees C, respectively. Both reactions were beta-glucuronidase-specific because both were inhibited by D-glucaro-1,4-lactone in a competitive fashion. Conjugated bilirubin acted as a noncompetitive inhibitor of the enzyme for PGA in a two-substrate system. The study indicates that these two substrates bind at different catalytic sites of the enzyme and, on molar base, conjugated bilirubin had higher affinity for the enzyme and less degree of inhibition by D-glucaro-1,4-lactone than PGA. Whether such catalytic sites are also common for other beta-D-glucuronid ethers and esters remains to be proven.     

A comparative study of bile acid CoA:amino acid:N-acyltransferase (BAT) from four mammalian species.

1. Bile acid CoA:amino acid:N-acyltransferase (BAT) was partially purified from dog, human, pig and rat livers. The interspecies variation in substrate specificity and kinetics were determined for glycine and taurine. 2. BAT activity from dog liver formed bile acid conjugates with taurine exclusively, whereas BAT activity from each of the other species formed conjugates with both taurine and glycine. 3. Biliary composition of glycine and taurine bile acid conjugates could partly be accounted for by substrate affinity (Km) and turnover number (Vmax) of BAT activity. 4. A monospecific anti-human BAT polyclonal antibody reacted on Western blot analysis with a 40 kDa band in a 100,000 g supernatant fraction from rat liver. 5. Immunoabsorption chromatography using an anti-human BAT antibody-Sepharose affinity column showed that both the immunoreactive protein band and BAT activity were removed from the 100,000 g supernatant fraction from human and rat livers.     

Phosphatidylinositol 3-kinase-dependent signaling modulates taurochenodeoxycholic acid-induced liver injury and cholestasis in perfused rat livers

Taurochenodeoxycholic acid (TCDCA), but not glycochenodeoxycholic acid (GCDCA), activates a phosphatidylinositol 3-kinase (PI3-K)-mediated survival pathway in vitro. Here, the effects of PI3-K inhibition on TCDCA- and GCDCA-induced hepatocellular injury, apoptosis, and bile secretion were examined in the intact liver. In isolated perfused rat livers, bile flow was determined gravimetrically. Hepatovenous lactate dehydrogenase and alanine aminotransferase efflux as markers of liver integrity and biliary secretion of 2,4-dinitrophenyl-S-glutathione (DNP-GS) were determined photometrically. Apoptosis was assessed by immunohistochemistry of active caspase-3 and cytokeratin 18 in liver tissue. Phosphorylation of protein kinase B (PKB/Akt) as a readout of PI3-K activity was determined by immunoblot analysis. Bile acid concentrations were determined by gas chromatography. TCDCA (25 muM) induced moderate liver injury by hepatocellular apoptosis and distinctly reduced bile flow and DNP-GS secretion. In contrast, GCDCA (25 muM) induced severe liver injury by extensive hepatocyte apoptosis. TCDCA strongly activated PI3-K, whereas GCDCA did not markedly affect PI3-K activity. Inhibition of PI3-K by 100 nM wortmannin enhanced TCDCA-induced liver injury and apoptosis and tended to aggravate the cholestatic effect of TCDCA. In contrast, wortmannin reduced GCDCA-induced liver injury and apoptosis. Bile acid uptake tended to be reduced by wortmannin. The cholestatic effect of GCDCA was aggravated by wortmannin. Inhibition of PI3-K markedly aggravated TCDCA-induced but not GCDCA-induced liver damage and hepatocyte apoptosis. Thus TCDCA appears to block its inherent toxicity by a PI3-K-dependent survival pathway in the intact liver.     

Purification and characterization of bile acid-CoA:amino acid N-acyltransferase from human liver

The bile acid-conjugating enzyme, bile acid-CoA: amino acid N-acyltransferase, was purified 480-fold from the soluble fraction of homogenized frozen human liver. Purification was accomplished by a combination of anion exchange chromatography, chromatofocusing, glycocholate-AH-Sepharose affinity chromatography, and high performance liquid chromatography (HPLC) gel filtration. Following purification, the reduced, denatured enzyme migrated as a single 50-kDa protein band by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. A similar molecular mass was obtained for the native enzyme by HPLC gel filtration. Elution from the chromatofocusing column suggested an apparent isoelectric point of 6.0 (+/- 0.2). Using a rabbit polyclonal antibody raised against the purified enzyme, Western blot analysis using 100,000 x g human liver supernatant confirmed that the affinity-purified polyclonal antibody was specific for human liver bile acid-CoA:amino acid N-acyltransferase. The purified enzyme utilized glycine, taurine, and 2-fluoro-beta-alanine (a 5-fluorouracil catabolite), but not beta-alanine, as substrates. Kinetic studies revealed apparent Km values for taurine, 2-fluoro-beta-alanine, and glycine of 1.1, 2.2, and 5.8 mM, respectively, with corresponding Vmax values of 0.33, 0.19, and 0.77 mumol/min/mg protein. These data demonstrate that a single monomeric enzyme is responsible for the conjugation of bile acids with glycine or taurine in human liver.     

Binding of bile acids, oleic acid, and organic anions by rat and human hepatic Z protein

Binding affinities of purified Z proteins from rat and human liver for bile acids, oleic acid, and organic anions were studied. Purification of Z protein from both rat and human hepatic cytosol was performed by gel filtration, chromatofocusing, and hydroxyapatite chromatography. Both purified proteins showed the same molecular weight (Mr = 14,000) and isoelectric points were 6.9 and 6.5 for rat and human proteins, respectively. Binding studies were performed by the competitive displacement of 1-anilino-8-naphthalene sulfonate. Rat and human Z proteins exhibited similar binding affinities for bile acids, oleic acid, and organic anions. Among various bile acids, both proteins bound monohydroxy bile acids with high affinity and trihydroxy bile acids with low affinity; sulfates were bound with higher and glucuronides with lower affinity than their parent bile acids. In comparison with GSH S-transferases, rat Z protein had lower affinity for bile acids than rat GSH S-transferase B and human Z protein had higher affinity for bile acids than human cationic GSH S-transferase. The role for Z protein in the intracellular binding of bile acids may be particularly important in human liver.     

Antigenic interrelationship between the 40-kilodalton cytokeratin polypeptide and desmoplakins

We describe here antigenic cross-reactivity between the human 40-kilodalton cytokeratin polypeptide [Moll et al] and components of bovine desmosomal plaque, namely desmoplakins I and II. This relationship was revealed by an antibody (KM 4.62), raised against cytoskeletal preparation of cultured human breast adenocarcinoma cells (MCF-7) and selected by immunoblotting and immunofluorescent labeling. In cultured human cells that contain the 40-kD cytokeratin, antibody KM 4.62 labeled arrays of filaments throughout the cytoplasm. This antibody labeled the basal layer of nonkeratinizing squamous epithelia as well as various simple (normal and malignant) epithelia and epithelial elements of the thymus. In liver tissue, labeling was obtained only in bile ducts and canaliculi but not in the hepatocytes. In bovine cells and tissues, on the other hand, immunofluorescent labeling with antibody KM 4.62 was strictly confined to desmosomes. This was verified by double immunolabeling with both antibody KM 4.62 and specific cytokeratin or desmosomal antibodies. Immunoblotting analysis indicated that the former antibody reacts specifically with the high molecular weight components of the bovine desmosomal plaque, namely desmoplakins I and II. These immunochemical results suggest that bovine desmoplakins share same structural relationship with the human acidic, 40-kD cytokeratin.     

Cloning,sequencing and expression of porcine CD40 ligand in Escherichia coli and human and porcine cells

The CD40L ligand (CD40L) plays an important role in the interaction between antigen-specific T lymphocytes and antigen-presenting cells. The porcine CD40L encoding gene was isolated from porcine peripheral blood mononuclear cells (PBMC) using RT-PCR. Sequence analysis of the cloned CD40L gene showed an open reading frame of 786 base pairs encoding a 262 amino acid protein with a predicted molecular mass of 29 kD. The deduced amino acid sequence of the porcine CD40L shared 82%, 88% and 93% similarity with the CD40L protein of mouse, human and cattle. The isolated CD40L sequence was expressed as a hexahistidine fusion protein in Escherichia coli and purified by affinity chromatography. The analysis of the CD40L-expression in human 293 and porcine MAX cells by immunofluorescence showed its location on the cell surface.     

Immunoexpression of intermediate filaments and morphological changes in the liver and bile duct of rats infected with Fasciola hepatica

We investigated the immunoexpression of the intermediate filament proteins, cytokeratin and desmin, and the morphological changes in the liver of rats during experimental fasciolosis at 4, 7 and 10 weeks post-infection. Rats were infected with 30 Fasciola hepatica metacercariae. Paraffin sections of the liver were stained using H & E, PAS and azan stains. Immunohistochemical reactions were performed using antibodies against cytokeratin and desmin. The experimental F. hepatica infection led to fibrosis and cirrhosis of the liver, and to inflammation of the common bile ducts. The expression of cytokeratin was increased in the epithelial cells of both the liver bile ductules at 4, 7 and 10 weeks post-infection and in the common bile ducts at 7 and 10 weeks post-infection compared to uninfected rats; expression in the common bile ducts was more intense. The myofibroblasts of the liver and smooth myocytes of the interlobular bile ducts and common bile ducts, showed a slight increase in desmin expression compared to the uninfected rats. The increased expression of cytokeratins in the hyperplastic rat common bile duct epithelium during the biliary phase of fasciolosis at 7 and 10 weeks post-infection may be explained by mechanical irritation by the parasite and an inflammatory reaction in the bile duct epithelium and in periductal fibrous tissue.     

UDP-glucuronosyltransferase activity and bilirubin conjugation in the bullfrog.

Bile pigments of bile and serum of Rana catesbeiana were investigated by means of high-pressure liquid chromatography. The major pigment in both bile and serum was bilirubin IX alpha. Bilirubin UDP-glucuronosyltransferase activity was found in the livers of all animals examined, but no conjugated bilirubin was detectable in the bile. Frog bile was found to contain large amounts of beta-glucuronidase. When the beta-glucuronidase inhibitor saccharo-1,4-lactone was introduced into the gall bladder followed by an exogenous bilirubin load, bilirubin glucuronide appeared in the bile.     

Pathophysiological basis of liver disease in cystic fibrosis employing a DeltaF508 mouse model

The molecular pathogenesis of cystic fibrosis (CF) liver disease is unknown. This study investigates its earliest pathophysiological manifestations employing a mouse model carrying DeltaF508, the commonest human CF mutation. We hypothesized that, if increased bile salt spillage into the colon occurs as in the human disease, then this should lead to a hydrophobic bile salt profile and to "hyperbilirubinbilia" because of induced enterohepatic cycling of unconjugated bilirubin. Hyperbilirubinbilia may then lead to an increased bile salt-to-phospholipid ratio in bile and, following hydrolysis, precipitation of divalent metal salts of unconjugated bilirubin. We document in CF mice elevated fecal bile acid excretion and biliary secretion of more hydrophobic bile salts compared with control wild-type mice. Biliary secretion rates of bilirubin monoglucuronosides, bile salts, phospholipids, and cholesterol are increased significantly with an augmented bile salt-to-phospholipid ratio. Quantitative histopathology of CF livers displays mild early cholangiopathy in approximately 53% of mice and multifocal divalent metal salt deposition in cholangiocytes. We conclude that increased fecal bile acid loss leads to more hydrophobic bile salts in hepatic bile and to hyperbilirubinbilia, a major contributor in augmenting the bile salt-to-phospholipid ratio and endogenous beta-glucuronidase hydrolysis of bilirubin glucuronosides. The confluence of these perturbations damages intrahepatic bile ducts and facilitates entrance of unconjugated bilirubin into cholangiocytes. This study of the earliest stages of CF liver disease provides a framework for investigating the molecular pathophysiology of more advanced disease in murine models and in humans with CF.     

Comparison of the binding sites of GSH S-transferases of the Ya- and Yb-subunit classes: effect of glutathione on the binding of bile acids

We have previously observed that the Ya subunit-containing glutathione (GSH) S-transferases from rat liver exhibit a common high affinity binding site for lithocholic acid, bilirubin, and sulfobromophthalein (BSP) (1984. J. Lipid Res. 25: 1177-1183). Subsequently we found that cholic acid and its amidates bound to a site on the Ya subunit separate for the lithocholic acid/bilirubin site (1986. J. Lipid Res. 27: 955-966). We now have extended this work by showing that amidates of lithocholic acid as well as chenodeoxycholic acid and its amidates competitively displace [14C]lithocholic acid from the Ya subunit. GSH did not inhibit binding of any of the ligands to the high affinity Ya site, but did inhibit binding to the cholic acid site on the Ya subunit. We have also defined the binding sites and effects of GSH on the Yb class of subunits. Lithocholic, chenodeoxycholic, and cholic acids (and amidates) shared a common site on the Yb or Y'b subunit, whereas BSP and bilirubin were bound at a different site. Both the bile acid and organic anion sites on the Yb subunit were inhibited by GSH. The inhibition by GSH in all cases (Ya cholic acid site or Yb bile acid or bilirubin sites) was saturable, of the competitive type, and incomplete at maximal GSH concentrations, suggesting that when GSH binds to its distinct substrate site, it induces a conformational change in the proteins affecting the other binding sites.