Binding of bile acids by glutathione S-transferases from rat liver

Binding of bile acids and their sulfates and glucuronides by purified GSH S-transferases from rat liver was studied by 1-anilino-8-naphthalenesulfonate fluorescence inhibition, flow dialysis, and equilibrium dialysis. In addition, corticosterone and sulfobromophthalein (BSP) binding were studied by equilibrium and flow dialysis. Transferases YaYa and YaYc had comparable affinity for lithocholic (Kd approximately 0.2 microM), glycochenodeoxycholic (Kd approximately to 60 microM), and cholic acid (Kd approximately equal 60 microM), and BSP (Kd approximately 0.09 microM). YaYc had one and YaYa had two high affinity binding sites for these ligands. Transferases containing the Yb subunit had two binding sites for these bile acids, although binding affinity for lithocholic acid (Kd approximately 4 microM) was lower than that of transferases with Ya subunit, and binding affinities for the other bile acids were comparable to the Ya family. Sulfated bile acids were bound with higher affinity and glucuronidated bile acids with lower affinity by YaYa and YaYc than the respective parent bile acids. In the presence of GSH, binding of lithocholate by YaYc was unchanged and binding by YbYb' was inhibited. Conversely, GSH inhibited the binding of cholic acid by YaYc but had less effect on binding by YbYb'. Cholic acid did not inhibit the binding of lithocholic acid by YaYa.     

Identification of S-acyl glutathione conjugates of bile acids in human bile by means of LC/ESI-MS

Previous work from this laboratory has reported the biotransformation of bile acids (BA) into the thioester-linked glutathione (GSH) conjugates via the intermediary metabolites formed by BA:CoA ligase and shown that such GSH conjugates are excreted into the bile in healthy rats as well as rats dosed with lithocholic acid or ursodeoxycholic acid. To examine whether such novel BA-GSH conjugates are present in human bile, we determined the concentration of the GSH conjugates of the five BA that predominate in human bile. Bile was obtained from three infants (age 4, 10, and 13 months) and the BA-GSH conjugates quantified by means of liquid chromatography (LC)/electrospray ionization (ESI)-linear ion trap mass spectrometry (MS) in negative-ion scan mode, monitoring characteristic transitions of the analytes. By LC/ESI-MS, only primary BA were present in biliary BA, indicating that the dehydroxylating flora had not yet developed. GSH conjugates of chenodeoxycholic and lithocholic acid were present in concentrations ranging from 27 to 1120 pmol/ml, several orders of magnitude less than those of natural BA N-acylamidates. GSH conjugates were not present, however, in the ductal bile obtained from 10 adults (nine choledocholithiasis, one bile duct cancer). Our results indicate that BA-GSH conjugates are formed and excreted in human bile, at least in infants, although this novel mode of conjugation is a very minor pathway.     

Inhibition of cytosolic glutathione S-transferase activity from rat liver by copper

H(2)O(2) inactivation of particular GST isoforms has been reported, with no information regarding the overall effect of other ROS on cytosolic GST activity. The present work describes the inactivation of total cytosolic GST activity from liver rats by the oxygen radical-generating system Cu(2+)/ascorbate. We have previously shown that this system may change some enzymatic activities of thiol proteins through two mechanisms: ROS-induced oxidation and non-specific Cu(2+) binding to protein thiol groups. In the present study, we show that nanomolar Cu(2+) in the absence of ascorbate did not modify total cytosolic GST activity; the same concentrations of Cu(2+) in the presence of ascorbate, however, inhibited this activity. Micromolar Cu(2+) in either the absence or presence of ascorbate inhibited cytosolic GST activity. Kinetic studies show that GSH but no 1-chloro-2,4-dinitrobenzene prevent the inhibition on cytosolic GST induced by micromolar Cu(2+) either in the absence or presence of ascorbate. On the other hand, NEM and mersalyl acid, both thiol-alkylating agents, inhibited GST activity with differential reactivity in a dose-dependent manner. Taken together, these results suggest that an inhibitory Cu(2+)-binding effect is likely to be negligible on the overall inhibition of cytosolic GST activity observed by the Cu(2+)/ascorbate system. We discuss how modification of GST-thiol groups is related to the inhibition of cytosolic GST activity.     

Formation of delta 2- and delta 3-cholenoic acids from bile acid 3-sulfates by a human intestinal Fusobacterium strain

We isolated two strains of an unnamed Fusobacterium species from human intestinal microflora, which stereospecifically transformed bile acid 3-sulfates into C-3-unsubstituted, ring A-unsaturated bile acids. Both 3 alpha- and 3 beta-sulfates of 5 beta-bile acids were metabolized to delta 3-5 beta-cholenoic acids; 3 beta-sulfates of 5 alpha-bile acids were converted into a mixture of delta 2-5 alpha-bile acids and 3 alpha-hydroxy-5 alpha-bile acids, whereas 3 alpha-sulfates of 5 alpha-bile acids were left intact. Unsulfated bile acids were not transformed into unsaturated derivatives. These strains differ from previously isolated intestinal bacteria, which desulfated bile acid sulfates without further transformation.     

Long-chain fatty acids and their acyl-CoA esters cause the translocation of phosphatidate phosphohydrolase from the cytosolic to the microsomal fraction of rat liver

A translocation of phosphatidate phosphohydrolase from the cytosolic to the microsomal fraction was promoted in cell-free extracts of rat liver by oleate and palmitate and their CoA esters. Oleate was more potent in this respect than palmitate and the CoA esters were more effective than the unesterified acids. Octanoate, octanoyl-CoA and CoA did not cause the translocation. It is proposed that the interaction of phosphatidate phosphohydrolase with the membranes that synthesize glycerolipids causes it to become metabolically active. This enables the liver to increase its capacity for triacylglycerol synthesis in response to an increased supply of fatty acids.     

Synthesis of N-acetylglucosaminides of unconjugated and conjugated bile acids.

3-N-Acetylglucosaminides of unconjugated, glycine- and taurine-conjugated bile acids have been synthesized. Bile acids appropriately protected were condensed with acetochloroglucosamine through the 3 alpha-hydroxyl group by means of the Koenigs-Knorr reaction using cadmium carbonate as a catalyst. Subsequent borohydride reduction and/or alkaline hydrolysis provided desired 3-N-acetylglucosaminides of unconjugated bile acids. Glycine-conjugates were obtained from N-acetylglucosaminides of unconjugated bile acids and ethyl glycinate by the carbodiimide method. The preparation of N-acetylglucosaminides of taurine-conjugates was attained by the Koenigs-Knorr reaction of bile acid p-nitrophenyl esters followed by condensation with taurine. 7-N-Acetylglucosaminides of ursodeoxycholates were prepared in a similar fashion. The convenient synthesis of 3-N-acetylglucosaminides of unconjugated bile acids is also described.     

Formation of delta 2- and delta 3-cholenoic acids from bile acid 3-sulfates by a human intestinal Fusobacterium strain.

We isolated two strains of an unnamed Fusobacterium species from human intestinal microflora, which stereospecifically transformed bile acid 3-sulfates into C-3-unsubstituted, ring A-unsaturated bile acids. Both 3 alpha- and 3 beta-sulfates of 5 beta-bile acids were metabolized to delta 3-5 beta-cholenoic acids; 3 beta-sulfates of 5 alpha-bile acids were converted into a mixture of delta 2-5 alpha-bile acids and 3 alpha-hydroxy-5 alpha-bile acids, whereas 3 alpha-sulfates of 5 alpha-bile acids were left intact. Unsulfated bile acids were not transformed into unsaturated derivatives. These strains differ from previously isolated intestinal bacteria, which desulfated bile acid sulfates without further transformation.     

Chemical synthesis of bile acid acyl-adenylates and formation by a rat liver microsomal fraction

In mammals, unconjugated bile acids formed in the intestine by bacterial deconjugation are reconjugated (N-acylamidated) with taurine or glycine during hepatocyte transport. Activation of the carboxyl group of bile acids to form acyl-adenylates is a likely key intermediate step in bile acid N-acylamidation. To gain more insight into the process of bile acid adenylate formation, we first synthesized the adenylates of five common, natural bile acids (cholic, deoxycholic, chenodeoxycholic, ursodeoxycholic, and lithocholic acid), and confirmed their structure by proton NMR. We then investigated adenylate formation by subcellular fractions of rat liver (microsomes, mitochondria, cytosol) using a newly developed LC method for quantifying adenylate formation. The highest activity was observed in the microsomal fraction. The reaction required Mg²⁺ and its optimum pH was about pH 7.0. In term of maximum velocity (V_max) and the Michaelis constant (K_m), the catalytic efficiency of the enzyme under the conditions used was highest with cholic acid of the bile acids tested. The formation of cholyl-adenylate was strongly inhibited by lithocholic and deoxycholic acid, as well as by palmitic acid; ibuprofen and valproic acid were weak inhibitors. In cholestatic disease, such adenylate formation might lead to subsequent bile acid conjugation with glutathione or proteins.     

Similarities between rat liver mitochondrial and cytosolic glutathione reductases and their apoenzyme accumulation in riboflavin deficiency

Glutathione reductase has been purified 5,500-fold from rat liver mitochondrial matrix in a yield of 30%. The mitochondrial enzyme was immunochemically indistinguishable from that of the cytosol and the subunit molecular weight was apparently similar to that of the cytosolic enzyme, that is, 50,000 daltons. The optimum pH and kinetic properties investigated were not significantly different from those of the cytosolic enzyme. When rats were fed a riboflavin-deficient diet, the enzyme activity in the mitochondria decreased to a greater extent than that in the cytosol, and greater accumulation of apo-enzyme in the former than that in the latter was confirmed by the amount of immunoprecipitable protein, activation by FAD addition in vitro, and the enzyme activity recovery after injection of riboflavin, into riboflavin-deficient rats.     

Biosynthesis of hydroxyl-linked glucuronides of short-chain bile acids by rat liver 3-hydroxysteroid UDP-glucuronosyltransferase

Microsomal preparations from livers of Sprague-Dawley rats catalyze the glucuronidation of 3 alpha-hydroxy-5 beta-H (3 alpha, 5 beta) short-chain bile acids (C20-C23), predominantly at the hydroxyl group, while the glucuronidation of 3 beta, 5 beta short-chain bile acids occurs exclusively at the carboxyl group. A similar pattern of conjugation was also observed in Wistar rats having normal levels of 3-hydroxysteroid UDP-glucuronosyltransferase. Significant reductions of formation rates for hydroxyl-linked, but not carboxyl-linked, short-chain bile acid glucuronides were observed in hepatic microsomes from Wistar rats with low 3-hydroxysteroid UDP-glucuronosyltransferase activity. 3-Hydroxysteroid UDP-glucuronosyltransferase, purified to homogeneity from Sprague-Dawley liver microsomes, catalyzed the 3-O-glucuronidation of 3 alpha, 5 beta C20-23 bile acids, as well as of lithocholic and isolithocholic acids (C24). The apparent Michaelis constants (KM) for short-chain bile acids were similar to the value obtained for androsterone. 3 alpha, 5 beta-C20 and 3 beta, 5 beta-C20 competitively inhibited glucuronidation of androsterone by the purified 3-hydroxysteroid UDP-glucuronosyltransferase. Purified 17 beta-hydroxysteroid and p-nitrophenol UDP-glucuronosyltransferases did not catalyze the glucuronidation of bile acids. In addition, none of the purified transferases catalyzed the formation of carboxyl-linked bile acid glucuronides. The results show that 3-hydroxysteroid UDP-glucuronosyltransferase, an enzyme specific for 3-hydroxyl groups of androgenic steroids and some conventional bile acids, also catalyzes the glucuronidation of 3 alpha-hydroxyl (but not carboxyl) groups of 3 alpha, 5 beta short-chain bile acids.     

Acid-catalyzed formation of isoprene from a mevalonate-derived product using a rat liver cytosolic fraction

Isoprene formation in a rat liver cytosolic fraction is shown to be increased 146-fold by acid treatment. This acid catalysis is dependent upon prior incubation of the cytosolic fraction with DL-mevalonate and is stimulated when the incubation also contains ATP. Formation of isoprene proceeds linearly through 5 h of acid treatment and is nearly complete at 10 h. These results suggest that the acid-catalyzed isoprene formation arises from the decomposition of dimethylallyl pyrophosphate via a carbonium ion mechanism. Chemical model studies using 3-methyl-2-buten-1-ol and 3-methyl-3-buten-1-ol (the alcohols corresponding to dimethylallyl pyrophosphate and isopentenyl pyrophosphate, respectively) confirm this hypothesis. At a pH less than or equal to 1, an 85% decomposition of 3-methyl-2-buten-1-ol to isoprene occurred after 24 h, while 3% of 3-methyl-3-buten-1-ol was converted to isoprene under identical conditions and time. It is concluded that the predominant immediate precursor of isoprene is dimethylallyl pyrophosphate and at low pH the ultimate fate of dimethylallyl pyrophosphate is complete conversion to isoprene. These conclusions have important biochemical and methodological implications.     

Inhibition of glutathione S-transferase by bile acids.

The effects of bile acids on the detoxification of compounds by glutathione conjugation have been investigated. Bile acids were found to inhibit the total soluble-fraction glutathione S-transferase activity from rat liver, as assayed with four different acceptor substrates. Dihydroxy bile acids were more inhibitory than trihydroxy bile acids, and conjugated bile acids were generally less inhibitory than the parent bile acid. At physiological concentrations of bile acid, the glutathione S-transferase activity in the soluble fraction was inhibited by nearly 50%. This indicates that the size of the hepatic pool of bile acids can influence the ability of the liver to detoxify electrophilic compounds. The A, B and C isoenzymes of glutathione S-transferase were isolated separately. Each was found to be inhibited by bile acids. Kinetic analysis of the inhibition revealed that the bile acids were not competitive inhibitors of either glutathione or acceptor substrate binding. The microsomal glutathione S-transferase from guinea-pig liver was also shown to be inhibited by bile acids. This inhibition, however, showed characteristics of a non-specific detergent-type inhibition.     

Regulation of linoleic acid delta 6-desaturation by a cytosolic lipoprotein-like fraction in isolated rat liver microsomes

A peripheral component of the delta 6-fatty acid-desaturase system of rat liver microsomes has been isolated from the cytosol by ultracentrifugation at a saline density of 1.26 g/ml. It exhibited lipoprotein characteristics with an approximate protein/lipid ratio of 1.22 and free fatty acids and phosphatidylcholine as its main lipid components. Linoleic acid desaturation activity diminished in washed microsomes, since they lost the adsorbed cytosolic fraction. Addition of the factor reactivated the reaction and the recovery was dependent on the concentration of the factor in the medium. Linoleic acid and linoleyl-CoA were bound by the cytosolic fraction. However, the transport of substrate to the desaturase was not apparently a main function of the cytosolic fraction, since transport occurred equally in the absence of the factor. Moreover, the solubilization of linoleyl-CoA was not enhanced and the free monomeric concentration was not altered by the presence of the cytosolic fraction. In addition, the factor did not divert delta 6-desaturase substrate to or from other metabolic pathways such as esterification to phospholipids. gamma-Linolenic acid produced by delta 6-desaturation of linoleic acid in the microsomes inhibited the desaturase, but it was removed by the factor from the membrane towards the cytosol, preventing the inhibition. The anti-inhibitory effect of the cytosolic factor was blockaded by addition of columbinic acid or gamma-linolenic acid to the factor. Moreover, the inhibitory effect of arachidonic acid was not prevented by addition of the cytosolic fraction. These results suggest that the cytosolic fraction studied would optimize the delta 6-desaturation of linoleic acid in vitro in rat liver microsomes by removal of the product, gamma-linolenic acid, as it is formed.     

Presence of protein-bound unconjugated bile acids in the cytoplasmic fraction of rat brain

Using liquid chromatography/electrospray ionization mass spectrometry, we have found three unconjugated bile acids [cholic acid (CA), chenodeoxycholic acid (CDCA), and deoxycholic acid (DCA)] in the rat brain cytoplasmic fraction. CDCA was detected only upon extraction with high concentrations of guanidine, indicating that it is bound noncovalently to protein in the brain. The most abundant of the three, it was present at a concentration of 1.6 nmol/g wet weight (approximately 15 mg of protein) of brain, corresponding to almost 30 times its serum concentration. CA and DCA were present at 1/30th the concentration of CDCA. Bile acids conjugated with amino acids, sulfuric acid, and glucuronic acid were not detected. These data clearly demonstrate that unconjugated CDCA and, to a lesser extent, CA and DCA, exists in the rat brain.