Two pathways of iron-catalyzed oxidation of bilirubin: effect of desferrioxamine and trolox, and comparison with microsomal oxidation.

The bilirubin-degrading activity of liver microsomes from rats induced with 3-methylcholanthrene has been shown to be markedly stimulated by addition of 3,3',4,4',5,5'-hexabromobiphenyl, a polyhalogenated chemical which resembles in size and shape the most effective inducers of cytochrome P450IA1, but lacks the structural features necessary for it to be metabolised. The degradation of bilirubin by this microsomal system has been compared to oxidation by a chemical model system involving H2O2 and Fe-EDTA (ethylenediaminetetraacetic acid). In both systems bilirubin disappearance was accompanied by bleaching. However, when either desferrioxamine or Trolox were present in the chemical model system, the rate of bilirubin oxidation was greatly enhanced and, at the same time, bilirubin was largely or entirely converted to biliverdin, a pathway of oxidation which proceeds by dehydrogenation. In the presence of desferrioxamine, biliverdin was also further oxidised to an unidentified red pigment.

Formation of cobalt protoporphyrin by chicken hepatocytes in culture. Relationship to decrease of 5-aminolaevulinate synthase caused by cobalt.

Cobalt protoporphyrin generated from 5-amino[4-14C]laevulinate by homogenates or primary cultures of chick embryo liver exposed to CoCl2 was found to be radioactivity unextractable by acid/acetone, when extra protein was added. The activity of ferrochelatase was required for formation of cobalt protoporphyrin since inhibition of ferrochelatase with 3,5-diethoxycarbonyl-1,4-dihydrocollidine (in the presence of cycloheximide) inhibited formation of cobalt protoporphyrin and resulted in accumulation of protoporphyrin. Cobalt protoporphyrin was detected spectrophotometrically in hepatocyte cultures exposed to the combination of 2-allyl-2-isopropylacetamide and CoCl2: (1) as the pyridine haemochrome of the protein pellet remaining after acid-acetone extraction of the cells, or (2) as the material extracted from the protein pellet with acetic acid-pyridine-chloroform. The amount of cobalt protoporphyrin increased with increasing CoCl2 concentration as cellular haem declined. The decrease in haem was about equal to the amount of cobalt protoporphyrin that accumulated. 2-Allyl-2-isopropylacetamide and polychlorinated biphenyls were both powerful inducers of 5-aminolaevulinate synthase. The former led to protoporphyrin accumulation, whereas with the latter, uroporphyrin accumulated, probably due to a concomitant decrease in activity of uroporphyrinogen decarboxylase. The decrease in activity of 5-aminolaevulinate synthase produced by administration of CoCl2 was greater after treatment with 2-allyl-2-isopropylacetamide than after treatment with allylisopropylacetamide and 3,4,3',4'-tetrachlorobiphenyl. We conclude: (a) that cobalt protoporphyrin is readily formed in cultured hepatocytes, and (b) that its formation accounts for the action of cobalt on 5-aminolaevulinate synthase.

Species differences in the covalent binding of [14C]tamoxifen to liver microsomes and the forms of cytochrome P450 involved.

Species differences in the NADPH-dependent covalent binding of [14C]tamoxifen to liver microsomes have been studied using preparations from humans, female F344 rats and DBA/2 mice. Protein binding has been used as an index of metabolic activation and as a surrogate for DNA binding in order to establish which forms of cytochrome P450 are responsible for genotoxicity. A panel of 12 human liver microsomes has been characterized and immunoquantified for nine cytochrome P450 isoenzymes. Binding of tamoxifen (45 microM) (25 +/- 2.5 pmol/15 min/mg protein, mean +/- SE) correlated (P < 0.05) with CYP3A4 and CYP2B6 content. Covalent binding of [14C]tamoxifen to microsomal preparations from human breast tumour tissue could also be detected but at levels 7-fold lower than in liver. The covalent binding of tamoxifen to mice, rat or human liver microsomal preparations increased with increasing substrate concentration. Covalent binding of [14C]tamoxifen (45 microM) in rats was 3.8-fold and mice 17-fold higher than in human liver microsomal preparations. In mice, the apparent Km (9.6 +/- 1.9 microM) was very much lower than for rats (119 +/- 41 microM). Pretreatment of female rats with phenobarbitone or dexamethasone resulted in a 4- to 5-fold increase in [14C]tamoxifen binding, relative to controls, consistent with the involvement of CYP2B1 and CYP3A1 in the metabolic activation. It cannot be distinguished at present if the same reactive metabolites are involved in protein and DNA binding. The greater potential of mouse liver microsomes to activate tamoxifen, relative to rats, does not reflect DNA damage or hepatocarcinogenicity seen following dosing with tamoxifen in vivo. It is concluded that covalent binding of tamoxifen to protein in vitro cannot be directly related to the carcinogenic potential of this compound. However, in the three species investigated, results suggest that the rat is a better model than the mouse for human liver microsomal activation of tamoxifen both with respect to kinetic parameters and the pattern of metabolic products.

Metabolic activation of halothane to neoantigens in C57Bl/10 mice: immunochemical studies.

C57Bl/10 mice were given halothane (10 mmol/kg, intraperitoneally) and microsomal proteins were analysed for the presence of trifluoroacetylated (TFA) neoantigens by SDS-gel electrophoresis followed by immunoblotting using a polyclonal anti-TFA antibody. In microsomal preparations from liver, lung and olfactory tissues, a 54 kDa TFA adduct was detectable 1 h after dosing. After 3-48 h, multiple bands were detected in liver (45-100 kDa) and in the lung (26-57 kDa) and in one experiment in which [14C]halothane was given, several immunoreactive bands from liver microsomes were shown to contain a covalently bound metabolite of the drug. In olfactory tissue, initially (1 h), a major band of 54 kDa and a less prominent component of about 50 kDa were seen. The number of bands increased at later times but the additional bands were far fewer than in liver. The rate of decay of the 54 kDa adduct was also measured in both liver and olfactory microsomes and found to be compatible with the reported turnover of total liver cytochrome P-450. 24 h after treating mice with halothane (10 mmol/kg), no TFA neoantigens could be detected on the outer cell surface of isolated viable hepatocytes when analysed by fluorescence activated flow cytometry. In contrast, non-viable cells, or those fixed in acetone were all positive. Using immunohistochemistry, TFA neoantigens were demonstrated in the centrilobular area of the liver, the non-ciliated bronchiolar epithelial (Clara) cells of the lung, proximal tubular cells of the kidney and the respiratory and olfactory epithelium of nasal tissues.

Neoantigen formation and clastogenic action of HCFC-123 and perchloroethylene in human MCL-5 cells.

In this study, the metabolic activation of 2,2-dichloro-1,1,1-trifluoroethane (hydrochlorofluorocarbons-123, HCFC-123), halothane or 1,1-dichloro-1-fluoroethane (HCFC-141b) was compared to that of perchloroethylene, using lymphoblastoma derived cell lines expressing human CYP1A1, CYP1A2, CYP2E1, CYP2A6 and CYP3A4 (MCL-5 cells). A dose dependent increase in micronucleus formation was detected over a nominal concentration range of 0.05-2 mM for HCFC-123 and halothane, but this was not seen with HCFC-141b. No dose response for HCFC-123 was seen in a control cHo1 cell line not expressing this cytochrome P450's. Cell lines expressing individual human cytochrome P-450 (CYP) forms were also used to define the enzymes responsible for the clastogenic events and to investigate the formation of immunoreactive protein by microsomal fractions. It was shown that CYP2E1 or CYP2B6 catalysed the clastogenic response, but CYP2D6, CYP3A4, CYP1A2 or CYP1A1 all appeared to be inactive. The formation of neoantigenic trifluoroacetylated protein adducts by microsomal mixtures incubated with HCFC-123 and NADPH was catalysed primarily by CYP2E1 and to a lesser extent by CYP2C19, whereas, only trace levels of immunoreactive protein were seen with microsomes expressing CYP2B6 or CYP2C8. With perchloroethylene as a substrate, the extent of activation was low in comparison with HCFC-123, as judged by the absence of micronuclei formation in the MCL-5 cell line and the weak immunoreactivity of proteins following Western blotting. CYP1A2, CYP2B6 and CYP2C8 appeared to be responsible for perchloroethylene immunoreactivity and in contrast to the findings with the HCFC's, no activation of perchloroethylene by CYP2E1 could be detected. These results show that even though both saturated and unsaturated halocarbons can result in neoantigen formation, there is a marked difference in the specificity of the CYP enzymes involved in their metabolic activation.

Drug-induced conversion of liver haem into modified porphyrins. Evidence for two classes of products.

A porphyrin with inhibitory activity towards protohaem ferro-lyase (EC 4.99.1.1) was isolated from the liver of mice given either griseofulvin or isogriseofulvin. This porphyrin resembles closely in chromatographic and spectral properties the inhibitory pigment isolated after treatment with 3,5-diethoxycarbonyl-1,4-dihydrocollidine, but differs from other green pigments that do not inhibit protohaem ferro-lyase. A hypothesis is proposed to account for the differences in properties between the two groups of pigments and for the mechanism of inhibition of the enzyme.

High-performance liquid chromatography of N-alkylprotoporphyrins: an investigation of their formation and loss under chemical and enzymatic conditions in vitro.

A new technique for resolving N-alkylprotoporphyrins into each structural isomer is described. The technique has been used to investigate the rate of formation and loss of N-alkylporphyrins during reaction of the parent porphyrin with alkyl iodides and to establish the conditions required for optimal yields of the various isomers. Preferential loss of the isomers bearing the N-alkyl group on one of the vinyl-substituted pyrrole rings is observed on prolonged incubation and HI generated during the reaction has been shown to be responsible. A method for detection and partial resolution by HPLC of N-alkylprotoporphyrins produced by liver microsomes in vitro is also described. Microsomes from rats induced with 3-methylcholanthrene produce significantly more N-ethylprotoporphyrin from either 4-ethyl-3,5-diethoxycarbonyl-1,4-dihydro- 2,6-dimethyl-pyridine or ethylhydrazine than do microsomes from control animals, but the isomeric composition of the isolated N-alkylporphyrin differs from that reported in vivo. Evidence that authentic N-alkylporphyrins are lost during incubation with microsomes has been obtained, and here again, the isomers bearing the N-alkyl group on vinyl-substituted pyrrole rings are preferentially lost. Experiments with 14C-labeled N-methylprotoporphyrin show that approximately 40% of the lost porphyrin could be recovered bound covalently to the microsomal pellet.

Stimulation of the pathway of porphyrin synthesis in the liver of rats and mice by griseofulvin, 3,5-Diethoxycarbonyl-1,4-dihydrocollidine and related drugs: evidence for two basically different mechanisms.

Griseofulvin and isogriseofulvin cause, like 3,5-diethoxycarbonyl-1,4-dihydrocollidine, a fall in the activity of the hepatic enzyme porphyrin-metal chelatase and accumulation of protoporphyrin in the liver. Analogues of either griseofulvin or 3,5-diethoxycarbonyl-1,4-dihydrocollidine which do not decrease the chelatase activity are not porphyrogenic on their own, but can potentiate the porphyria caused by 3,5-diethoxycarbonyl-1,4-dihydrocollidine. This suggests the existence of two basically different mechanisms by which drugs stimulate the pathway of porphyrin synthesis in the liver.

Inhibition of haem synthesis caused by cobalt in rat liver. Evidence for two different sites of action.

Cobalt inhibits liver haem synthesis in vivo by acting at least two different sites in the biosynthetic pathway: (1) synthesis of 5-aminolaevulinate and (2) conversion of 5-amino-laevulinate into haem. The first effect is largely, if not entirely, due to inhibition of the activity of 5-aminolaevulinate synthase, rather than to inhibition of the formation of the enzyme. The second effect results from diversion of 5-aminolaevulinate into an unidentified liver pool with solubility properties similar to those of cobalt protoporphyrin.

The effect of cobaltous chloride on liver haem metabolism in the rat. Evidence for inhibition of haem synthesis and for increased haem degradation.

After a single dose of cobaltous chloride there was a marked inhibition of liver 5-aminolaevulinate (5-ALA) synthetase (at 1 h) and this was followed in turn by a stimulation of haem oxygenase (at 3 h) and by a return of the synthetase activity to normal or above normal (at 17 h). Bile cannulation experiments were performed 1 and 17 h after administration of CoCl2. At 1 h there was a marked decrease in bile porphyrin content, no change in bile concentration of bilirubin, but a decrease in the conversion of [14C]-5-ALA to bilirubin and to liver haem. At 17 h, an the other hand, the bile excretion of both porphyrins and bilirubin was significantly greater than in controls and more radioactivity (from [14C]-5-ALA) appeared in the bile as bilirubin. It is concluded that the effects of cobalt on liver haem metabolism are complex and time-dependent. There is first inhibition of liver haem synthesis at two different steps of the pathway (synthesis of 5-ALA and conversion of 5-ALA to haem), with diversion of [14C]-5-ALA into a relatively stable liver pool different from haem; and at a later stage there is also an increase in the rate of liver haem degradation.

Studies on the mechanism of experimental porphyria produced by 3,5-diethoxycarbonyl-1,4-dihydrocollidine. Role of a porphyrin-like inhibitor of protohaem ferro-lyase.

A green porphyrin-like pigment with inhibitory properties towards protohaem ferro-lyase activity was isolated from the livers of mice and rats after treatment with 3,5-diethoxycarbonyl-1,4-dihydrocollidine. Mice, which are more sensitive to the porphyrogenic properties of the drug, produce more inhibitor. The non-porphyrogenic analogue 3,5-diethoxycarbonylcollidine does not cause accumulation of the pigment in the liver. The inhibitory substance is present in control liver at low but measurable concentrations.

Inhibition of protohaem ferro-lyase in experimental porphyria. Isolation and partial characterization of a modified porphyrin inhibitor.

1. A modified porphyrin with inhibitory activity towards protohaem ferro-lyase was purified from the livers of mice treated with 3,5-diethoxycarbonyl-1,4-dihydrocollidine, and partially characterized. 2. The inhibitor can be labelled by 5-amino[4-14C]-laevulinate, suggesting that it originates from pre-labelled liver haem. No radioactivity from 3,5-diethoxycarbonyl-1,4-dihydro[2,6-14C]collidine can be recovered bound to the purified abnormal porphyrin when the radioactive drug is used to induce its formation. 3. Similar modified porphyrins isolated from the livers of animals treated with 2-allyl-2-isopropylacetamide, secobarbitone or 1-ethynylcyclohexanol did not exhibit inhibitory activity toward protohaem ferro-lyase. 4. The inhibition of protohaem ferro-lyase was progressive, could be slowed down by cooling and partially prevented by preincubating the enzyme with the porphyrin substrate. Once established, inhibition could not be reversed by addition of the substrate 5. These results suggest that the modified porphyrin irreversibly inhibits protohaem ferro-lyase and may be used as a sensitive and selective reagent to titrate the number of active centres of the enzyme.

Inhibition of protohaem ferro-lyase by N-substituted porphyrins. Structural requirements for the inhibitory effect.

N-Methyl mesoporphyrin was a powerful inhibitor of protohaem ferro-lyase in vitro, whereas N-ethyl mesoporphyrin and N-methyl coproporphyrin were not and neither was the newly described green pigment produced by giving rats ethylene. This suggests that the size of the substituent at a pyrrole nitrogen and also the number of carboxylic acid side chains of the substituted porphyrin are important for the inhibitory effect. Evidence that N-methyl mesoporphyrin inhibited the enzyme, whereas the ethylene-derived pigment did not, was also obtained in vivo.