Hepatic drug-metabolizing enzymes in primary and secondary tumors of human liver.

Significant increases in activities of epoxide hydrolase, UDP-glucuronosyltransferase, and glutathione S-transferase, and marked reductions in cytochrome P-450 mixed-function oxidase systems occur in hyperplastic nodules induced in rat liver by chemical mutagens. In contrast, activities of both oxidative (Phase I) and conjugative (Phase II) enzymes are decreased in hepatocellular carcinomas induced by peroxisome proliferators. The present work compares alterations induced by chemical mutagens or peroxisome proliferators with changes in enzyme activities that occur in primary and secondary hepatic tumors in man. The above activities, along with beta-glucuronidase and arylsulfatase, were measured in liver samples from 6 normal livers obtained at immediate autopsy, and liver specimens obtained by surgical biopsy from the following patients: 8 with hepatomas, 5 with nonmetastatic colorectal carcinomas, and 14 with metastatic colorectal carcinomas. Cytochromes P-450MP and P-450NF in addition to epoxide hydrolase were measured by immunoquantitation. Enzymes involved in conjugation reactions were either assayed fluorometrically (UDP-glucuronosyltransferase, beta-glucuronidase, sulfotransferase, and sulfatase) or spectrophotometrically (glutathione S-transferase) using umbelliferyl substrates or 1-chloro-2,4-dinitrobenzene. Secondary hepatic tumors showed no significant change in drug-metabolizing enzymes, in contrast to primary hepatomas, which displayed decreases in all of the measured drug metabolizing enzymes. Arylsulfatase was markedly depressed in primary hepatomas (14% of normal values). Thus, activities of drug-metabolizing enzymes in human primary tumors resemble those associated with altered hepatic foci induced by peroxisome proliferators such as ciprofibrate. The marked decreases in sulfatase that occurred in primary but not in secondary human tumors suggest that sulfation of endogenous compounds and xenobiotics may differ in patients with primary and secondary hepatic tumors.

Hepatic drug conjugation/deconjugation systems in hepatosplenic schistosomiasis.

The biotransformation capacity of the liver is altered in hepatosplenic schistosomiasis being more in favor of deconjugation pathways. These alterations should be considered when xenobiotics are administered and in studies of hepatic toxicity and carcinogenicity.

Stereoselective glucuronidation of (R)- and (S)-naproxen by recombinant rat phenol UDP-glucuronosyltransferase (UGT1A1) and its human orthologue.

Recombinant rat phenol UDP-glucuronosyltransferase (UGT1A1) conjugates (R)-naproxen at a much higher rate (> 17-fold) than its (S)-enantiomer, substantiating previous findings on stereoselective glucuronidation of racemic naproxen. In contrast, the recombinant human orthologue conjugated both enantiomers at equal rates. In line with high constitutive expression of UGT1A1 in extrahepatic tissues, a high R/S ratio of naproxen glucuronidation was found in rat testes, intestine, lung and kidney. The results demonstrate that (R)-naproxen represents a stereoselective substrate of rat UGT1A1, but not of the human orthologous UGT1A1.

Hepatic cytochrome P-450 system in experimental hepatosplenic schistosomiasis. Presence of an artifact in spectrophotometric analysis.

Several recent reports suggesting that the liver microsomal cytochrome P-450 system is impaired in Schistosoma mansoni infections in mice prompted a detailed investigation of the hepatic cytochrome P-450 system in murine schistosomiasis. Mice were prepared with three levels of Schistosoma mansoni infection and studied 8, 12 and 14 weeks post-exposure. Histological sections of the liver confirmed prominent granuloma formation and portal fibrosis which was dose and time dependent. Cytochrome P-450 levels appeared reduced grossly in microsomes from homogenates of infected livers, but accurate quantitation was complicated by the presence of a prominent peak at 422 nm. Ethylmorphine N-demethylase activity also appeared to be reduced in all infected animals, reaching a maximum decrease at 14 weeks of 44% of control values in the most heavily infected mice. NADPH-cytochrome c reductase activity and cytochrome b5 levels were similarly reduced. Utilization of techniques to separate hepatocytes from granulomatous material in infected livers eliminated the 422 nm peak and reversed several of the findings made with whole-liver derived microsomes. Cytochrome P-450 levels were reduced modestly (30%) at 8 weeks in the most heavily infected mice but returned to normal values by 14 weeks. Specific ethylmorphine N-demethylase activity of isolated hepatocytes was increased at 8 weeks with a return to normal by 14 weeks. Isolated granulomata were incriminated as one possible source of the 422 nm pigment in whole-liver derived microsomes but appeared unlikely to account fully for this finding. Thus, this investigation concluded that the cytochrome P-450 system is altered by experimental murine hepatosplenic schistosomiasis, but such alterations are subtle in nature and unlikely to contribute in major fashion to the observed changes in drug disposition.

Drug metabolizing enzyme activities in rat liver epithelial cell lines, hepatocytes and bile duct cells.

P450-dependent mono-oxygenase and conjugating enzyme activities were studied in rat liver epithelial cells (RLEs) and compared to those in hepatocytes and bile duct cells. Various RLE cell lines were investigated since (a) they are suspected to be derived from cells in the lineage from putative pluripotent stem cells to either hepatocytes or bile duct cells, and (b) they may represent targets of chemical carcinogens. Despite considerable variation between lines, common features were recognized. P450-dependent monooxygenase activities (7-ethoxyresorufin O-deethylase and 7-ethoxycoumarin O-deethylase) were undetectable in all RLEs and bile duct cells, and were uninducible by benz(a)anthracene. In contrast, glucuronosyltransferase (GT), sulfotransferase and GSH transferase activities were clearly detectable. Conjugating enzyme activities increased until confluency of the cell cultures was reached. Under the latter conditions, GT activities towards 4-methylumbelliferone or benzo(a)pyrene-3,6-quinol (substrates of a 3-methylcholanthrene-inducible phenol GT) were similar to those found in hepatocytes or bile duct cells. Using a selective cDNA probe, phenol GT mRNA was clearly detectable in RLE1. In contrast, GT activity towards 4-hydroxybiphenyl was much lower than in hepatocytes or bile duct cells (0.04- and 0.03-fold). Sulfotransferase and GSH transferase activities were also roughly comparable to those found in hepatocytes and in bile duct cells. The results suggest that RLEs and bile duct cells exhibit both high conjugating enzyme activities and a lack of P450-dependent mono-oxygenase activities, a pattern resembling the 'toxin-resistance phenotype' found in putative preneoplastic hepatocyte foci and nodules.

Application of solid-phase extraction to the isolation and determination of paracetamol and its metabolites.

A series of clean-up columns for solid-phase extraction (SPE) were packed with various C18 phases of different physico-chemical surface properties. These SPE columns were used for the isolation and determination of paracetamol and its metabolites from biological samples. The packing with a monomeric structure of chemically bonded phase showed the best recovery of tested substances in urine.

Sublobular distribution of transferases and hydrolases associated with glucuronide, sulfate and glutathione conjugation in human liver.

Activities of glucuronosyltransferase, sulfotransferase, glutathione S-transferase, beta-glucuronidase and sulfatase were determined in microdissected samples of periportal and pericentral sublobular regions from four human livers obtained at immediate autopsy. New methods are presented for the microdetermination of sulfotransferase and sulfatase activities in microdissected samples weighing 0.1 to 4 micrograms dry weight using umbelliferone and 4-methylumbelliferone sulfate as substrates. The three transferases were distributed heterogeneously across the liver lobule. Glucuronosyltransferase and glutathione S-transferase were localized predominantly in pericentral regions. In contrast, sulfotransferase activity was greater in periportal than pericentral regions. Average activities for glucuronosyltransferase and sulfotransferase were 23, and 50 mumoles X gm dry wt-1 X hr-1, respectively, in periportal regions, and 34 and 38 mumoles X gm dry st-1 X hr-1, respectively, in pericentral regions. Activities of glutathione S-transferase were considerably higher than those of the other transferases and were 8.3 mmoles X gm dry wt-1 X hr-1 in periportal areas and 12.2 mmoles X gm dry wt-1 hr-1 in pericentral areas. The two hydrolases studied, beta-glucuronidase and sulfatase, were evenly distributed across the liver lobule. The presence of significant hydrolase and transferase activities in both zones of the liver lobule supports the idea that net production of both sulfate and glucuronide conjugates may be influenced by futile cycling of conjugation-deconjugation reactions in both zones of the liver. Based on enhanced formation of sulfate but not glucuronide conjugates in homogenates of human liver treated with inhibitors of the hydrolases, it is suggested that futile cycling is more pertinent to the regulation of sulfation than glucuronidation.

Stereoselective (S)- and (R)-naproxen glucuronosyl transferases of rat liver.

Stereoselective glucuronidation of naproxen, one of the 2-arylpropionic acids that are widely used as anti-inflammatory drugs, was investigated in chromatofocusing fractions of solubilized liver microsomes from 3-methylcholanthrene- (MC) and phenobarbital- (PB) treated rats. On chromatofocusing of solubilized microsomes of PB-treated rats, two naproxen glucuronosyltransferase (GT) fractions were separated. The fraction eluting at pH 8.7 preferentially conjugated (S)-naproxen (S/R ratio = 1.6) and the fraction eluting at pH 7.8 mostly conjugated (R)-naproxen (S/R ratio = 0.7). Chromatofocusing of solubilized microsomes from MC-treated rats also resulted in the separation of two naproxen GT fractions, eluting at pH 9.4 (S/R ratio 0.2) and at pH 8.7 (S/R ratio 0.8). These two fractions coincided with the elution of known MC-inducible GT activities assigned to a GT isozyme variously termed 4-nitrophenol GT or GT-I. Interestingly, kidney microsomes, known to contain a high constitutive expression of GT-I, preferentially glucuronidated (R)-naproxen (S/R ratio 0.2). The S/R ratio of 0.8, observed with the pI 8.7 fraction of MC-treated rat liver, may be explained by the presence of a mixture of naproxen GTs, consisting of (R)-naproxen GT (S/R ratio 0.2) and of (S)-naproxen GT (S/R ratio 1.6). The results suggest that naproxen is conjugated by at least 3 GT isozymes in rat liver; these have been operationally designated (S)-naproxen GTPB (S/R ratio 1.6), (R)-naproxen GTPB (S/R ratio 0.7), and (R)-naproxen GTMC (S/R ratio 0.2). The latter isozyme is probably identical to the previously characterized MC-inducible GT-I. Thus, (S)- and (R)-naproxen represent useful substrates to distinguish different GT isozymes.

Hydrolysis of 4-methylumbelliferyl sulfate in periportal and pericentral areas of the liver lobule.

4-Methylumbelliferyl sulfate was used to characterize sulfatase activity in periportal and pericentral regions of the liver lobule in the perfused rat liver. Following infusion of 1.5 mM of this organic sulfatester, free 4-methylumbelliferone and 4-methylumbelliferyl glucuronide were formed at rates of 13 and 9 mumoles/g/h, respectively, in livers from fasted, phenobarbital-treated rats. 5-Pregnen-3 beta-ol, 20-one sulfate inhibited hydrolysis and metabolite production completely, whereas perfusion with nitrogen-saturated perfusate or FCCP decreased total metabolite formation by only 30%. 4-Methylumbelliferone formed from the hydrolysis of 4-methylumbelliferyl sulfate was monitored with micro-light guides placed on periportal and pericentral areas of the liver lobule. Detection of the desulfated product was always greater in the downstream region, i.e., infusion of 4-methylumbelliferyl sulfate produced a higher fluorescence signal in pericentral areas when perfusion was in the anterograde direction, while periportal areas demonstrated higher activity during perfusion in the retrograde direction. Perfusion with nitrogen-saturated perfusate abolished these differences. Taken together, these data suggest that uptake of organic sulfateesters is partially energy dependent, follows the hepatic oxygen gradient inversely, and is a major rate determinant for sulfatase activity in the liver.

Hydrolysis of organic sulfates in periportal and pericentral regions of the liver lobule: studies with 4-methylumbelliferyl sulfate in the perfused rat liver.

The hydrolysis of 4-methylumbelliferyl sulfate by liver sulfatases to free fluorescent 4-methylumbelliferone and the subsequent formation of the glucuronide conjugate were studied in the isolated perfused rat liver. In livers from fed, phenobarbital-treated rats, 4-methylumbelliferyl sulfate (0.25-1.5 mM) was hydrolyzed rapidly to free 4-methylumbelliferone at maximal rates of about 5 mumol/g/hr. A major fraction of the free 4-methylumbelliferone formed was converted to the glucuronide at maximal rates around 20 mumol/g/hr. Similar rates of hydrolysis were observed in livers from fasted, phenobarbital-treated or normal rats, although the ratio of glucuronide to free product was decreased markedly by fasting. In liver homogenates, however, rates of organic sulfate hydrolysis exceeded those observed in the perfused liver by at least 2-fold, suggesting that 4-methylumbelliferyl sulfate content is an important determinant of rates of hydrolysis in the perfused liver. There was a good correlation (r = 0.91) between rates of product formation and fluorescence of 4-methylumbelliferone detected from the liver surface with fiber optic light guides. Fluorescence of 4-methylumbelliferone produced from hydrolysis of 4-methylumbelliferyl sulfate was also monitored with micro-light guides placed on periportal and pericentral areas of the liver lobule for the estimation of local rates of product formation. When perfusions were in the anterograde direction, desulfation of 4-methylumbelliferyl sulfate was about 50% higher in pericentral (28.8 +/- 9.3 mumol/g/hr) than in periportal (18.2 +/- 2.7 mumol/g/hr) areas. Furthermore, 4-methylumbelliferyl sulfate content determined in microdissected samples was 1.5- to 2-fold higher in pericentral than in periportal regions of the liver lobule but the activity of 4-methylumbelliferyl sulfate sulfatase was identical in both zones of the liver lobule. We conclude, therefore, that the local substrate content is an important determinant of rates of 4-methylumbelliferyl sulfate hydrolysis in sublobular zones of the liver.

Species-dependent enantioselective glucuronidation of three 2-arylpropionic acids. Naproxen, ibuprofen, and benoxaprofen.

The enantioselective glucuronidation of several racemic 2-arylproprionic acids (naproxen, ibuprofen, and benoxaprofen) was investigated in vitro with immobilized microsomal protein from human, rhesus monkey, and rabbit liver as the source of UDP-glucuronyl-transferases. Human microsomes, solubilized microsomal protein, and immobilized protein all gave comparable enantioselectivity. The diastereomeric glucuronides were separated and quantitated by HPLC and characterized stereochemically by co-elution with glucuronides formed from authentic resolved enantiomers. Conjugation of the carboxylic acid moieties occurred stereoselectively with all three substrates. However, enantioselectivity varied qualitatively and quantitatively with substrate as well as with species. The glucuronidation of (S)-naproxen by human liver enzymes was inhibited in the presence of (R)-naproxen and vice versa. The ratio of the glucuronides of (S)-benoxaprofen to that of (R)-benoxaprofen in rhesus monkey urine varied between individual animals and appeared to change through time as dosing continued. Hydrolysis of the diasteriomeric glucuronides of (R)- and (S)-naproxen was differentially inhibited by addition of 1,4-saccharolactone.