Immunochemical evidence for multiple steroid-inducible hepatic cytochromes P-450 in the rat

Expression of the pregnane X receptor in mice antagonizes the cholic acid-mediated changes in plasma lipoprotein profile

OBJECTIVE

Modification of lipoprotein metabolism by bile acids has been mainly explained by activation of the farnesyl X receptor (FXR). The aim of the present study was to determine the relative contribution of the pregnane X receptor (PXR), another bile acid-activated nuclear receptor to changes in plasma lipoprotein profile.

METHODS AND RESULTS

Wild-type mice, Pxr-deficient mice, and Pxr-null mice expressing human PXR (Pxr-null SXR-Tg mice) were fed a cholic acid-containing diet, and consequences on plasma lipoprotein profiles and target gene expression were assessed. Cholic acid produced significant decreases in high-density lipoprotein (HDL) cholesterol, plasma apolipoprotein (apo)A-I and hepatic apoA-I mRNA in wild-type mice. Interestingly, the effect of cholic acid was significantly more pronounced in Pxr-deficient mice, indicating that PXR contributes to the weakening of the effect of bile acids on lipoprotein metabolism. Reciprocally, changes in HDL/apoA-I profiles were abolished in Pxr-null SXR-Tg mice in which PXR-responsive genes, particularly those involved in bile acid detoxification were readily activated after cholic acid treatment.

CONCLUSIONS

PXR expression in mice antagonizes the cholic acid-mediated downregulation of plasma HDL cholesterol and apoA-I, and magnification of PXR/SXR-mediated changes may constitute a new mean to counteract the effects of bile acids on plasma lipoproteins.

Disrupted bile acid homeostasis reveals an unexpected interaction among nuclear hormone receptors, transporters, and cytochrome P450

Sister of P-glycoprotein (SPGP) is the major hepatic bile salt export pump (BSEP). BSEP/SPGP expression varies dramatically among human livers. The potency and hierarchy of bile acids as ligands for the farnesyl/bile acid receptor (FXR/BAR) paralleled their ability to induce BSEP in human hepatocyte cultures. FXR:RXR heterodimers bound to IR1 elements and enhanced bile acid transcriptional activation of the mouse and human BSEP/SPGP promoters. In FXR/BAR nullizygous mice, which have dramatically reduced BSEP/SPGP levels, hepatic CYP3A11 and CYP2B10 were strongly but unexpectedly induced. Notably, the rank order of bile acids as CYP3A4 inducers and activators of pregnane X receptor/steroid and xenobiotic receptor (PXR/SXR) closely paralleled each other but was markedly different from their hierarchy and potency as inducers of BSEP in human hepatocytes. Moreover, the hepatoprotective bile acid ursodeoxycholic acid, which reverses hydrophobic bile acid hepatotoxicity, activates PXR and efficaciously induces CYP3A4 (a bile-metabolizing enzyme) in primary human hepatocytes thus providing one mechanism for its hepatoprotection. Because serum and urinary bile acids increased in FXR/BAR -/- mice, we evaluated hepatic transporters for compensatory changes that might circumvent the profound decrease in BSEP/SPGP. We found weak MRP3 up-regulation. In contrast, MRP4 was substantially increased in the FXR/BAR nullizygous mice and was further elevated by cholic acid. Thus, enhanced hepatocellular concentrations of bile acids, due to the down-regulation of BSEP/SPGP-mediated efflux in FXR nullizygous mice, result in an alternate but apparent compensatory up-regulation of CYP3A, CYP2B, and some ABC transporters that is consistent with activation of PXR/SXR by bile acids.

Growth hormone regulation and developmental expression of rat hepatic CYP3A18, CYP3A9, and CYP3A2

The present study investigated the role of growth hormone (GH) in hepatic CYP3A18 and CYP3A9 expression in prepubertal and adult male rats. For comparison, the effects of GH on CYP3A2 expression were also measured. Initial experiments demonstrated that CYP3A18 mRNA levels were greater during puberty and adulthood than during the prepubertal period, CYP3A9 mRNA was not expressed until puberty and its expression increased in adulthood, and CYP3A2 mRNA levels were relatively constant from prepuberty to adult life. Hypophysectomy, which results in the loss of multiple pituitary factors including GH, increased CYP3A2 and CYP3A18 mRNA expression 3- to 4-fold, but it did not affect CYP3A9 mRNA levels or CYP3A-mediated testosterone 2beta- or 6beta-hydroxylase activity in adult rats. GH administered as twice daily s.c. injections (0.12 microg/g body weight) to hypophysectomized or intact adult rats did not affect CYP3A18 or CYP3A9 mRNA expression. The same treatment decreased CYP3A2 mRNA and protein and testosterone 2beta- and 6beta-hydroxylase activity levels in intact but not hypophysectomized rats. However, in intact prepubertal rats, intermittent GH administration decreased CYP3A18 and CYP3A2 mRNA levels, but a higher dosage (3.6 microg/g) was required to suppress CYP3A2. Overall, the present study demonstrated that: (a) the constitutive expression of CYP3A18, CYP3A9, and CYP3A2 does not require the presence of GH, (b) CYP3A18 is more sensitive than CYP3A9 to GH modulation in adult rats; and (c) CYP3A2 is less sensitive to the suppressive influence of GH during the prepubertal period than during adult life.

Cytochrome P450 metabolism of estradiol in hamster liver and kidney

Estradiol induces kidney tumors in Syrian hamsters. The elevated conversion of estradiol to 4-hydroxylated metabolites in kidney compared to the predominant 2-hydroxylation in liver and other organs, where tumors are not induced by this treatment, has been proposed to be the basis of estrogen-induced carcinogenesis. In this study, we examined the hepatic and renal enzymes catalyzing the formation of catecholestrogens to understand the differences in estrogen metabolism in these organs. In liver, 2-hydroxylation of estradiol is the major metabolic pathway with 4-hydroxylation a minor by-product and with the formation of both catechols responding coordinately to the same inhibitors. Western blot analysis and inhibition studies suggest that the major form catalyzing hepatic estrogen 2-hydroxylation is a member of the CYP3A family, as previously observed with rat liver microsomes, and that 4-hydroxylation is a by-product of this metabolism. In the kidney, 4-hydroxylation of estradiol appears to be catalyzed by more than one enzyme according to the Eadie-Hofstee analysis. Both 2- and 4-hydroxylation in the kidney are affected differentially by inhibitors and are induced by beta-napthoflavone. Western blots of renal microsomes reveal that CYP1A2 is induced whereas CYP1A1 is detectable in kidney, but not induced by this treatment. Finally, a part of the 2-hydroxylation and a small part of the 4-hydroxylation by kidney microsomes may be catalyzed by a member of the CYP3A family, since these reactions are partially inhibited by CYP3A inhibitors such as progesterone and other progestins, although renal enzyme levels are much lower than those in the liver as revealed by Western blot. Our data suggest that estrogen 2-hydroxylation in the hamster kidney is catalyzed by members of the CYP1A and CYP3A families, which also contribute to 4-hydroxylation. The majority of 4-hydroxyestradiol formation in the hamster kidney may be catalyzed by a form(s) of the newly discovered CYP1B family that has yet to be characterized.

Differential interaction of erythromycin with cytochromes P450 3A1/2 in the endoplasmic reticulum: a CO flash photolysis study

The kinetics of CO binding to cytochromes P450, measured by the flash photolysis technique, were used to probe the interaction of erythromycin with cytochromes P450 in rat liver microsomes. Addition of erythromycin generates substrate difference spectra using microsomes from rats treated with phenobarbital or dexamethasone but not from untreated rats, showing that it binds to P450s induced by these agents. In contrast, erythromycin and/or a monoclonal antibody to P450 3A1/2 accelerated CO binding to microsomes from rats treated with phenobarbital but had no effect on microsomes from untreated or dexamethasone-treated rats. Based on the differential amounts and inducibilities of the P450 3A1 and 3A2 forms in these microsomal samples, these results indicate that erythromycin increased the rate for P450 3A2 but not P450 3A1. The divergent effects of erythromycin on these P450s, which exhibit 89% sequence similarity, were consistent with a model of the P450 substrate binding site in which erythromycin forms a more rigid complex with P450 3A1 than P450 3A2. These results demonstrate the sensitivity of P450 conformation/dynamics to substrate binding, and show that CO binding kinetics can distinguish among closely related P450s in a microsomal environment.

P-glycoprotein: a major determinant of rifampicin-inducible expression of cytochrome P4503A in mice and humans

The P-glycoprotein (Pgp) efflux pump can influence the hepatocellular concentration of xenobiotics that are modulators and substrates of cytochrome P4503A (CYP3A). We tested the hypothesis that Pgp is a determinant of drug-inducible expression of CYP3A. The magnitude of CYP3A induction by rifampicin was compared in the human parental colon carcinoma cell line LS 180/WT (wild type) and in two derivative clones overexpressing the human multidrug resistance gene MDR1 (also designated PGY1) because of either drug selection (LS 180/ADR) or transfection with MDRI cDNA (LS 180/MDR). In both MDR1 cDNA-overexpressing clones, rifampicin induction of CYP3A mRNA and protein was decreased and required greater rifampicin concentrations compared with parental cells. The role of Pgp in regulation of CYP3A expression in vivo was analyzed in mice carrying a targeted disruption of the mdr1a mouse gene. Oral treatment with increasing doses of rifampicin resulted in elevated drug levels in the livers of mdr1a (-/-) mice compared with mdr1a (+/+) mice at all doses. Consistent with the enhanced accumulation of rifampicin in mdr1a (-/-) mice, lower doses of rifampicin were required for induction of CYP3A proteins, and the magnitude of CYP3A induction was greater at all doses of rifampicin in mdr1a (-/-) mice compared with mdr1a (+/+) mice. We conclude that Pgp-mediated transport is a critical element influencing the CYP3A inductive response.

Distribution and induction of CYP3A1 and CYP3A2 in rat liver and extrahepatic tissues

Previously, we have shown that highly specific antibodies against cytochrome P450 enzymes can be produced by targeting a 5-amino acid sequence at the C-terminus. Although rat CYP3A1 and CYP3A2 share 89% amino acid sequence similarity, they differ by 3 out of 5 of their C-terminal residues. In an effort to produce antibodies specific to each form, rabbits were immunised with the peptides IITGS and VINGA, corresponding to the C-termini of CYP3A1 and CYP3A2, respectively. Both antibodies bound strongly to hepatic microsomal fraction from rats treated with pregnenolone 16 alpha-carbonitrile (PCN) in enzyme-linked immunosorbent assay. Binding of the anti-IITGS antibody was strongly inhibited by incubation with IITGS, but VINGA was 60 times less effective. Conversely, binding of the anti-VINGA antibody was inhibited by VINGA 100 times more effectively than IITGS. Similar inhibition of antibody binding was also found using immunoblotting. Immunoadsorption using the anti-IITGS antibody yielded a single protein from solubilised hepatic microsomal fraction from PCN-treated rats, which was recognised only by the anti-IITGS antibody. Both antibodies bound to single proteins in the liver which were increased following treatment with PCN, but only the anti-IITGS antibody recognised protein in the lung, small intestine, and kidney of untreated and PCN-treated rats. Also, the binding of the two antibodies to hepatic and extrahepatic microsomal fractions from uninduced and induced rats showed differences in the expression of proteins recognised by the two antibodies, providing further evidence of antibody specificity. Thus, the binding of anti-IITGS and anti-VINGA antibodies is mutually exclusive and consistent with specific binding to their target antigens, CYP3A1 and CYP3A2, respectively. Immunocytochemistry was used to determine the distribution of CYP3A1 and CYP3A2. In the liver of untreated animals, both CYP3A1 and CYP3A2 were found to be expressed in the centrilobular region. However, some CYP3A1 immunoreactivity was also detected in many, but not all, hepatocytes throughout the lobule. However, following treatment of rats with PCN, both CYP3A1 and CYP3A2 were found to be strongly expressed in hepatocytes throughout the lobule, although CYP3A2 showed greater expression in the centrilobular region. PCN treatment was also found to result in induction of CYP3A1 in specific regions of the small intestine, lung, and kidney.

Human cytochrome P450 3A4: enzymatic properties of a purified recombinant fusion protein containing NADPH-P450 reductase

Human cytochrome P450 3A4 is recognized as the catalyst for the oxygen-dependent metabolism of a diverse group of medically important chemicals, including the immunosuppressive agent cyclosporin; macrolide antibiotics, such as erythromycin; drugs such as benzphetamine, nifedipine, and cocaine; and steroids; such as cortisol and testosterone to name but a few. We have engineered the cDNA for human cytochrome P450 3A4 by linkage to the cDNA for the rat or human flavoprotein, NADPH-P450 reductase (NADPH:ferrihemoprotein oxidoreductase, EC 1.6.2.4). An enzymatically active fusion protein (rF450[mHum3A4/mRatOR]L1) has been expressed at high levels in Escherichia coli and purified to homogeneity. Enzymatic studies show a requirement for lipid, detergent, and cytochrome b5 for the 6 beta-hydroxylation of steroids and the N-oxidation of nifedipine. In contrast, these additions are not required for the N-demethylation of erythromycin or benzphetamine. A spectrophotometrically detectable metabolite complex of P450 3A4 is formed during the metabolism of triacetyloleandomycin, and this has a pronounced inhibitory effect on the metabolism of both testosterone and erythromycin. These results relate to the interpretation of current methods used to assess the in vivo activity of P450 3A4.

Effects of caloric restriction on rodent drug and carcinogen metabolizing enzymes: implications for mutagenesis and cancer

Caloric restriction in rodents results in increased longevity and a decreased rate of spontaneous and chemically induced neoplasia. The low rates of spontaneous neoplasia and other pathologies have made calorically restricted rodents attractive for use in chronic bioassays. However, caloric restriction also alters hepatic drug metabolizing enzyme (DME) expression and so may also alter the biotransformation rates of test chemicals. These alterations in DME expression may be divided into two types: (1) those that are the direct result of caloric restriction itself and are detectable from shortly after the restriction is initiated; (2) those which are the result of pathological conditions that are delayed by caloric restriction. These latter alterations do not usually become apparent until late in the life of the organism. In rats, the largest direct effect of caloric restriction on liver DMEs is an apparent de-differentiation of sex-specific enzyme expression. This includes a 40-70% decrease in cytochrome P450 2C11 (CYP2C11) expression in males and a 20-30% reduction of corticosterone sulfotransferase activity in females. Changes in DME activities that occur late in life in calorically restricted rats include a stimulation of CYP2E1-dependent 4-nitrophenol hydroxylase activity and a delay in the disappearance of male-specific enzyme activities in senescent males. It is probable that altered DME expression is associated with altered metabolic activation of chemical carcinogens. For example the relative expression of hepatic CYP2C11 in ad libitum-fed or calorically restricted rats of different ages is closely correlated with the amount of genetic damage in 2-acetylaminofluorene- or aflatoxin B1-pretreated hepatocytes isolated from rats of the same age and caloric intake. This suggests that altered hepatic drug and carcinogen metabolism in calorically restricted rats can influence the carcinogenicity of test chemicals.

Cultured adult rat jejunal explants as a model for studying regulation of CYP3A

Enzymes within the CYP3A subfamily are major Phase I drug-metabolizing enzymes present in hepatocytes and small bowel enterocytes. These enzymes are highly inducible in the liver by many structurally diverse compounds, including a number of commonly used medications. Studies indicate that CYP3A enzymes present in small bowel enterocytes are also inducible. However, the regulation of CYP3A enzymes in this tissue has not been well characterized, in part because in vivo studies are difficult, especially in humans. Our goals was to develop an in vitro model to study the regulation of CYP3A in enterocytes. To this end, we defined culture conditions under which adult rat jejunal explants maintained viable appearing villi for 21 hr. When dexamethasone, the prototypical inducer of CYP3A1 in rat hepatocytes, was added to the culture medium, there was a time-dependent induction of CYP3A1 mRNA and CYP3A protein in explant enterocytes which was essentially indistinguishable from the time course of induction of CYP3A1 mRNA and protein in enterocytes in vivo. This effect of dexamethasone appeared to be specific since dexamethasone had no consistent effect on the explant concentration of another enterocyte specific mRNA, intestinal fatty acid binding protein. Using this explant culture model, we found that CYP3A1 mRNA was also inducible by clotrimazole but we were unable to detect induction by rifampicin or troleandomycin. Our observations suggest that jejunal explants may provide an appropriate model for the study of the regulation of CYP3A and other drug-metabolizing enzymes.

Differential responses of rat hepatic microsomal carboxylesterase isozymes to glucocorticoids and pregnenolone 16 alpha-carbonitrile

Differences in the responses to glucocorticoids and pregnenolone 16 alpha-carbonitrile (PCN) of three isozymes of hepatic microsomal carboxylesterase, namely RL1, RL2 and RH1, in male rats were studied. The administration of dexamethasone dose-dependently increased isocarboxazid hydrolase activity, whereas p-nitrophenyl acetate-hydrolyzing activity was decreased dose-dependently. Betamethasone, methylprednisolone and PCN also markedly increased isocarboxazid hydrolase activity. A radial immunodiffusion assay indicated that carboxylesterase reactive with antibodies was induced by these steroids. Carboxylesterase isozyme RL2 was strongly induced by dexamethasone, methylprednisolone and PCN. In contrast, RL1 and RH1 were decreased by dexamethasone, but not by the other steroids. Estradiol benzoate had a synergic effect on the PCN-induced increase of isocarboxazid hydrolase, but the actions of the glucocorticoids were not affected. It is concluded that hepatic microsomal carboxylesterase isozymes in rats differ considerably from each other in their response to various steroids. These data are also indicative of the importance of glucocorticoids in hepatic xenobiotic metabolism.

pH effects on the N-demethylation and formation of the cytochrome P-450 iron II nitrosoalkane complex for erythromycin derivatives

The effects of pH on access to the cytochrome P-450 active site, N-demethylation and formation of the cytochrome P-450 Fe(II)-RNO metabolite complex for a series of erythromycin derivatives were examined. Studies were performed with dexamethasone-treated rat liver microsomes containing large amounts of cytochrome P-450 3A isozymes. In addition to factors such as hydrophobicity or hindrance around the dimethyl-amino function, the ionisation state of the N(CH3)2 group played an important role in the recognition and metabolism of the substrate by cytochrome P-450. Esterification of the desosamine in the beta position of the N(CH3)2 group leads to lower pKa values for the R--N+ H(CH3)2 <--> [R--N (CH3)2] + H+ equilibrium. At physiological pH, the amine group is mainly in the unprotonated form. Consequently, easier access to the protein active site and significant formation of cytochrome P-450 Fe(II)-RNO metabolite complex are observed for these derivatives. These results led us to interpret the formation of cytochrome P-450 Fe(II)-RNO metabolite complex as a series of multiple steps equilibria depending on the ionisation state of the N(CH3)2 group, the partition coefficient of the substrate between the microsomal layer and the aqueous media and a series of metabolic reactions leading partially to the final inhibitory nitrosoalkane-cytochrome P-450 Fe(II) complex.

Evidence for the involvement of a distinct form of cytochrome P450 3A in the oxidation of digitoxin by rat liver microsomes

The preceding paper (B. Gemzik, D. Greenway, C. Nevins, and A. Parkinson (1992). Regulation of two electrophoretically distinct proteins recognized by antibody against rat liver cytochrome P450 3A1. J. Biochem. Toxicol., 7 (43-52).) described the regulation of two rat liver microsomal proteins (50- and 51-kDa) recognized by antibody against P450 3A1. It was also shown that changes in the levels of the 51-kDa 3A protein were usually paralleled by changes in the rate of testosterone 2 beta-, 6 beta-, and 15 beta-hydroxylation. The present study demonstrates that age- and sex-dependent changes in the 50-kDa protein were paralleled by changes in the rate of digitoxin oxidation to digitoxigenin bisdigitoxoside. Induction or suppression of the 50-kDa protein by treatment of rats with various xenobiotics were also paralleled by changes in the rate of digitoxin oxidation. These results suggest that, contrary to previous assumptions, the conversion of digitoxin to digitoxigenin bisdigitoxoside and the conversion of testosterone to 2 beta-, 6 beta-, and 15 beta-hydroxytestosterone are primarily catalyzed by different forms of P450 3A. Further evidence for this conclusion was obtained from studies in which the suicide inhibitor, chloramphenicol, was administered to mature female rats previously treated with pregnenolone-16 alpha-carbonitrile (PCN), which induces both the 50-kDa and the 51-kDa protein. Treatment of mature female rats with PCN alone caused a marked increase (16- to 18-fold) in the 6 beta-hydroxylation of testosterone and the rate of digitoxin oxidation. Treatment of PCN-induced rats with chloramphenicol caused a approximately 70% decrease in liver microsomal testosterone 6 beta-hydroxylation, but had no effect on the rate of conversion of digitoxin to digitoxigenin bisdigitoxoside. The oxidation of testosterone by purified 3A1 (a 51-kDa protein) was also inhibited by chloramphenicol in a time- and reduced nicotinamide adenine dinucleotide phosphate (NADPH)-dependent manner. In addition to testosterone and chloramphenicol, purified 3A1 also metabolized troleandomycin, but it was unable to convert digitoxin to digitoxigenin bisdigitoxoside. Testosterone inhibited the microsomal oxidation of digitoxin, but digitoxin did not inhibit testosterone oxidation. This suggests that testosterone is a substrate for the 3A enzyme that metabolizes digitoxin, but that this form of P450 3A does not contribute significantly to testosterone oxidation by rat liver microsomes.(ABSTRACT TRUNCATED AT 400 WORDS)

The human hepatic cytochromes P450 involved in drug metabolism

The cytochromes P450 are a superfamily of hemoproteins that catalyze the metabolism of a large number of xenobiotics and endobiotics. The type and amount (i.e., the animal's phenotype) of the P450s expressed by the animal, primarily in the liver, thus determine the metabolic response of the animal to a chemical challenge. A majority of the characterized P450s involved in hepatic drug metabolism have been identified in experimental animals. However, recently at least 12 human drug-metabolizing P450s have been characterized at the molecular and/or enzyme level. The characterization of these P450s has made it possible to "phenotype" microsomal samples with respect to their relative levels of the various P450s and their metabolic capabilities. The purpose of this review is to compare and contrast the human P450s involved in drug metabolism with their related forms in the rat and other experimental species.

Regulation of two electrophoretically distinct proteins recognized by antibody against rat liver cytochrome P450 3A1

We recently reported that antibody against purified P450 3A1 (P450p) recognizes two electrophoretically distinct proteins (50 and 51 kDa) in liver microsomes from male and female rats, as determined by Western immunoblotting. Depending on the source of the liver microsomes, the 51-kDa protein corresponded to 3A1 and/or 3A2 which could not be resolved by sodium dodecyl sulfate (SDS)polyacrylamide gel electrophoresis. The other protein (50 kDa) appears to be another member of the P450 IIIA gene family. Both proteins were markedly intensified in liver microsomes from male or female rats treated with pregnenolone-16 alpha-carbonitrile, dexamethasone, troleandomycin, or chlordane. In contrast, treatment of male or female rats with phenobarbital intensified only the 51-kDa protein. Treatment of male rats with Aroclor 1254 induced the 51-kDa protein, but suppressed the 50-kDa form. In addition to their changes in response to inducers, the 50- and 51-kDa proteins also differed in their developmental expression. For example, the 50-kDa protein was not expressed until weaning (3 weeks), whereas the 51-kDa protein was expressed even in 1-week-old rats. At puberty (between weeks 5 and 6), the levels of the 50-kDa and 51-kDa proteins markedly declined in female but not in male rats, which introduced a large sex difference (male greater than female) in the levels of both proteins. Changes in the level of the 51-kDa protein were paralleled by changes in the rate of testosterone 2 beta-, 6 beta-, and 15 beta-hydroxylation. In male rats, the marked increase in the levels of the 50-kDa protein between weeks 2 and 3 coincided with a three- to four fold increase in the rate of testosterone 2 beta-, 6 beta-, and 15 beta-hydroxylation, which suggests that the 50-kDa protein catalyzes the same pathways of testosterone oxidation as the 51-kDa protein. However, this developmental increase in testosterone oxidation may have resulted from an activation of the 51-kDa 3A protein. These results indicate that the two electrophoretically distinct proteins recognized by antibody against P450 3A1 are regulated in a similar but not identical manner, and suggest that the 51-kDa 3A protein is the major microsomal enzyme responsible for catalyzing the 2 beta-, 6 beta-, and 15 beta-hydroxylation of testosterone.

Cytochrome P450 IIIA1 (P450p) requires cytochrome b5 and phospholipid with unsaturated fatty acids

In contrast to other P450 enzymes purified from rat liver microsomes, purified P450 IIIA1 (P450p) is catalytically inactive when reconstituted with NADPH-cytochrome P450 reductase and the synthetic lipid, dilauroylphosphatidylcholine. However, purified P450 IIIA1 catalyzes the oxidation of testosterone when reconstituted with NADPH-cytochrome P450 reductase, cytochrome b5, an extract of microsomal lipid, and detergent (Emulgen 911). The present study demonstrates that the microsomal lipid extract can be replaced with one of several naturally occurring phospholipids, but not with cholesterol, sphingosine, sphingomyelin, ceramide, cerebroside, or cardiolipin. The ratio of the testosterone metabolites formed by purified P450 IIIA1 (i.e., 2 beta-, 6 beta-, and 15 beta-hydroxytestosterone) was influenced by the type of phospholipid added to the reconstitution system. The ability to replace microsomal lipid extract with several different phospholipids suggests that the nature of the polar group (i.e., choline, serine, ethanolamine, or inositol) is not critical for P450 IIIA1 activity, which implies that P450 IIIA1 activity is highly dependent on the fatty acid component of these lipids. To test this possibility, P450 IIIA1 was reconstituted with a series of synthetic phosphatidylcholines. Those phosphatidylcholines containing saturated fatty acids were unable to support testosterone oxidation by purified P450 IIIA1, regardless of the acyl chain length (C6 to C18). In contrast, several unsaturated phosphatidylcholines supported testosterone oxidation by purified P450 IIIA1, and in this regard dioleoylphosphatidylcholine (PC(18:1)2) was as effective as microsomal lipid extract and naturally occurring phosphatidylcholine or phosphatidylserine. These results confirmed that P450 IIIA1 activity is highly dependent on the fatty acid component of phospholipids. A second series of experiments was undertaken to determine whether microsomal P450 IIIA1, like the purified enzyme, is dependent on cytochrome b5. A polyclonal antibody against purified cytochrome b5 was raised in rabbits and was purified by affinity chromatography. Anti-cytochrome b5 caused a approximately 60% inhibition of testosterone 2 beta-, 6 beta-, and 15 beta-hydroxylation by purified P450 IIIA1 and inhibited these same reactions by approximately 70% when added to liver microsomes from dexamethasone-induced female rats. Overall, these results suggest that testosterone oxidation by microsomal cytochrome P450 IIIA1 requires cytochrome b5 and phospholipid containing unsaturated fatty acids.

Differential inhibition of individual human liver cytochromes P-450 by cimetidine

Cimetidine binds to cytochrome P-450 and inhibits hepatic metabolism of various drugs in humans. However, cytochrome P-450 is a family of enzymes rather than a single protein, and effects of cimetidine on individual human liver cytochromes P-450 have not been previously characterized. Metabolism of selected substrates and cimetidine-binding assays have been performed using human liver microsomes, purified human liver cytochromes P-450, and cytochrome P-450 complementary DNA-expressed yeast proteins to probe interaction of cimetidine with these individual enzymes. Cimetidine (3.0 mmol/L) in incubations reduced bufuralol hydroxylase activity by 80% and strongly inhibited microsomal nifedipine oxidation (23% +/- 13% of control activity). The same concentration of cimetidine produced intermediate inhibition of cytochrome enzymes responsible for ethoxyresorufin deethylation and aniline hydroxylation (77% +/- 6% and 68% +/- 17% of activity in control microsomal incubations, respectively), but little effect on tolbutamide hydroxylation was observed. Concordantly, the calculated binding constant for the binding of cimetidine to a purified cytochrome P-450 with high tolbutamide hydroxylase activity was 4.4 mmol/L, whereas the calculated binding concentration constant for a purified cytochrome P-450-metabolizing nifedipine was 0.7 mmol/L. These studies show a high variability in the effect of cimetidine on drug metabolism by individual human liver cytochromes P-450. In vitro studies using human liver microsomes and genetically engineered human cytochromes P-450 can be very useful in exploring important clinical questions of hepatic drug metabolism.

The inhibition of drug oxidation by anhydroerythromycin, an acid degradation product of erythromycin

The inhibition of steroid 6 beta-hydroxylase activity by anhydroerythromycin, an acid breakdown product of erythromycin, has been studied and compared to the effects of erythromycin using liver microsomes from control and dexamethasone pretreated rats and human liver microsomes. Both anhydroerythromycin and erythromycin were found to be demethylated, thus both fulfil the prerequisites for possible metabolite-cytochrome P450 complex information. The formation of a metabolite-cytochrome P450 complex was demonstrated for anhydroerythromycin by preincubating NADPH fortified microsomes with anhydroerythromycin. This complex formation could be reversed by incubating the microsomes in 50 microM potassium ferricyanide. Anhydroerythromycin was a more potent inhibitor of androst-4-ene-3,17-dione (androstenedione) 6 beta-hydroxylation than erythromycin. Kinetic analysis shows that there are probably two cytochromes P450 involved in androstenedione 6 beta-hydroxylation in control rat microsomes both of which are inhibited by anhydroerythromycin. There are at least two forms of cytochrome P450 responsible for androstenedione 6 beta-hydroxylation in microsomes from dexamethasone pretreated rats but only the high affinity form is inhibited by anhydroerythromycin. "Atypical" kinetics were observed in human microsomes but inhibition of androstenedione 6 beta-hydroxylation was observed with 5 microM anhydroerythromycin at all androstenedione concentrations used. Inconsistencies have been observed in the literature with respect to clinical interactions observed with erythromycin. Since anhydroerythromycin appears to be a more potent inhibitor of androstenedione 6 beta-hydroxylation than erythromycin, we speculate that the variable blood levels of anhydroerythromycin found after dosing with erythromycin may explain these discrepancies.

Effects of ethinylestradiol and testosterone implants on hepatic microsomal cytochrome P450 monooxygenases of birth gonadectomized male and female Dark Agouti rats

Monooxygenases in the cytochrome P450 IIIA subfamily are induced by a number of their xenobiotic substrates and by testosterone, an endobiotic substrate of importance in their regulation. 17 alpha-Ethinylestradiol (EE) is also metabolized by these enzymes and in this study Dark Agouti rats were used to examine the effects of subcutaneous implantation of controlled release silastic capsules containing EE to determine if this steroid also induces these enzymes. Data were compared with results obtained from equivalent groups of animals implanted with capsules containing testosterone propionate (TP). Liver microsomes prepared from male and female rats were used to identify intrinsic gender differences in the monooxygenases studied and gender differences in the responses to the implanted steroids were also determined. Effects due to imprinting of growth hormone secretion patterns were controlled by using male and female birth gonadectomized animals. Results obtained from groups with blank implants showed there were no effects due to the silastic implant material itself on the monooxygenases studied. The specific activities of erythromycin N-demethylation in liver microsomes of both EE and TP implanted male and female birth gonadectomized animals were enhanced relative to corresponding blank implanted controls consistent with both steroids having an effect to induce activity attributable to cytochrome P450 IIIA isoforms. Immunoinhibition studies using microsomes from EE treated female rats with erythromycin as substrate provided further evidence for this steroid having this induction effect. The specific activity of ethylmorphine N-demethylation was however not increased in microsomes prepared from the EE implanted female animals and was decreased in the corresponding male preparations. These findings distinguished the response to this steroid from that to TP and suggested induction by this estrogen of an isoform(s) having a more limited range of substrates than has characteristically been found in this subfamily. EE treatment also caused an increase in diazepam C3 hydroxylase consistent with an effect to induce P450 IIIA activity but this was found only in microsomes from birth gonadectomized female animals. This was in contrast to the effect of TP treatment which produced increases in this monooxygenase in both male and female animals. Another gender specific effect of EE was a striking decrease in morphine N-demethylase activity seen only in birth gonadectomized male rats. This again contrasted with the effect of TP which caused a marked increase in this activity in liver microsomes of both male and female birth gonadectomized animals consistent with the proposal that testosterone is important in the regulation of this activity.(ABSTRACT TRUNCATED AT 400 WORDS)

Differential stability of drug-metabolizing enzyme activities in primary rat hepatocytes, cultured in the absence or presence of dexamethasone

The effects of primary hepatocyte culture on the rat cytochrome P450-dependent monooxygenase system and several conjugating enzyme activities were examined using a culture system similar to those used for evaluation of chemicals as potential genotoxins. Cytochrome P450 and cytochrome b5 contents progressively decreased throughout the 72-h culture period to less than 25% of initial values, whereas cytochrome P450 reductase rapidly decreased by 50% during attachment, but then remained stable. Cytochrome P450-dependent testosterone hydroxylase activities decreased more rapidly in culture than did cytochrome P450 content reaching less than 50% of attachment levels by 24 h. Cytochrome P450IIIA immunoreactive protein decreased at a similar rate to testosterone-6 beta-hydroxylase. Activated UDP-glucuronyltransferase activities towards 1-naphthol and testosterone declined more slowly over the 72 h than cytochrome P450 and remained at 50-60% of initial values at 72 h. UDP-glucuronyltransferase activity towards digitoxigenin monodigitoxoside (DIG) did not decrease during culture. Glutathione-S-transferase and sulfotransferase activities also declined during the 72 h at rates which appeared to be isozyme-dependent. Addition of 1 microM dexamethasone (DEX) to the culture medium increased UDP-glucuronyltransferase activity towards DIG, cytochrome P450 reductase and testosterone-6 beta-hydroxylase activities up to 2.5-, 2.0- and 7-fold, respectively and induced cytochrome P450IIIA immunoreactive protein(s) in the hepatocytes after 24 and 48 h of culture; DEX was less effective at the 72 h time-point. DEX treatment also significantly accelerated the decreases in glutathione-S-transferase activities and in sulfotransferase activities towards 1-naphthol and estrone. Thus, it appears that primary rat hepatocytes cultured under standard conditions, not only rapidly lose their monooxygenase capabilities, but also some of their capacity for conjugation. Furthermore, the use of DEX in cell culture medium to enhance cell survival does not maintain total drug-metabolizing enzyme capability, but appears to transiently and selectively increase expression of certain isozymes at the expense of others.

Differential regulation of liver P-450III cytochromes in choline-deficient rats: implications for the erythromycin breath test as a parameter of liver function

Progressive liver fibrosis in rats develops when they are fed a diet deficient in choline. This diet also results in a pronounced and selective decrease in the liver microsomal content of a phase I drug-metabolizing enzyme belonging to the cytochrome P-450III gene family. Because P-450III cytochromes characteristically catalyze the N-demethylation of erythromycin, we believed that the production of breath CO2 from erythromycin would be dramatically reduced in choline-deficient rats. However, when 12 choline-deficient rats were compared with 9 control rats, the reduction in CO2 production from erythromycin (mean decrease 71%) was essentially identical to that from aminopyrine (mean decrease 69%), a substrate believed to be metabolized normally by the hepatocyte in fibrotic liver disease. Furthermore, we found that the relative erythromycin and aminopyrine demethylase activities were comparable when measured in vitro in liver microsomes prepared from the choline-deficient rats. To determine the molecular basis for the erythromycin demethylase activity in the choline-deficient rats, the liver microsomes were subjected to immunoblot analysis using a variety of polyclonal and monoclonal antibodies capable of distinguishing individual P-450III-related proteins. Our studies confirm that a major erythromycin demethylase belonging to the P-450III family, termed P-450p, was greatly reduced in the choline-deficient rat liver. However, the specific concentration of a second P-450p-related protein was essentially normal and that of a third P-450p-related protein was actually increased in the choline-deficient rat liver.(ABSTRACT TRUNCATED AT 250 WORDS)

Pronounced and differential effects of ionic strength and pH on testosterone oxidation by membrane-bound and purified forms of rat liver microsomal cytochrome P-450

The aim of this study was to determine the effects of ionic strength and pH on the different pathways of testosterone oxidation catalyzed by rat liver microsomes. The catalytic activity of cytochromes P-450a (IIA1), P-450b (IIB1), P-450h (IIC11) and P-450p (IIIA1) was measured in liver microsomes from mature male rats and phenobarbital-treated rats as testosterone 7 alpha-, 16 beta-, 2 alpha- and 6 beta-hydroxylase activity, respectively. An increase in the concentration of potassium phosphate (from 25 to 250 mM) caused a marked decrease in the catalytic activity of cytochromes P-450a (to 8%), P-450b (to 22%) and P-450h (to 23%), but caused a pronounced increase in the catalytic activity of cytochrome P-450p (up to 4.2-fold). These effects were attributed to changes in ionic strength, because similar but less pronounced effects were observed with Tris-HCl (which has approximately 1/3 the ionic strength of phosphate buffer at pH 7.4). Testosterone oxidation by microsomal cytochromes P-450a, P-450b, P-450h and P-450p was also differentially affected by pH (over the range 6.8-8.0). The pH optima ranged from 7.1 (for P-450a and P-450h) to 8.0 (for P-450p), with an intermediate value of 7.4 for cytochrome P-450b. Increasing the pH from 6.8 to 8.0 unexpectedly altered the relative amounts of the 3 major metabolites produced by cytochrome P-450h. The decline in testosterone oxidation by cytochromes P-450a, P-450b and P-450h that accompanied an increase in ionic strength or pH could be duplicated in reconstitution systems containing purified P-450a, P-450b or P-450h, equimolar amounts of NADPH-cytochrome P-450 reductase and optimal amounts of dilauroylphosphatidylcholine. This result indicated that the decline in testosterone oxidation by cytochromes P-450a, P-450b and P-450h was a direct effect of ionic strength and pH on these enzymes, rather than a secondary effect related to the increase in testosterone oxidation by cytochrome P-450p. Similar studies with purified cytochrome P-450p were complicated by the atypical conditions needed to reconstitute this enzyme. However, studies on the conversion of digitoxin to digitoxigenin bisdigitoxoside by liver microsomes, which is catalyzed specifically by cytochrome P-450p, provided indirect evidence that the increase in catalytic activity of cytochrome P-450p was also a direct effect of ionic strength and pH on this enzyme.(ABSTRACT TRUNCATED AT 400 WORDS)

Reconstitution of testosterone oxidation by purified rat cytochrome P450p (IIIA1)

Cytochrome P450p (IIIA1) has been purified from rat liver microsomes by several investigators, but in all cases the purified protein, in contrast to other P450 enzymes, has not been catalytically active when reconstituted with NADPH-cytochrome P450 reductase and dilauroylphosphatidylcholine. We now report the successful reconstitution of testosterone oxidation by cytochrome P450p, which was purified from liver microsomes from troleandomycin-treated rats. The rate of testosterone oxidation was greatest when purified cytochrome P450p (50 pmol/ml) was reconstituted with a fivefold molar excess of NADPH-cytochrome P450 reductase, an equimolar amount of cytochrome b5, 200 micrograms/ml of a chloroform/methanol extract of microsomal lipid (which could not be substituted with dilauroylphosphatidylcholine), and the nonionic detergent, Emulgen 911 (50 micrograms/ml). Testosterone oxidation by cytochrome P450p was optimal at 200 mM potassium phosphate, pH 7.25. In addition to their final concentration, the order of addition of these components was found to influence the catalytic activity of cytochrome P450p. Under these experimental conditions, purified cytochrome P450p converted testosterone to four major and four minor metabolites at an overall rate of 18 nmol/nmol P450p/min (which is comparable to the rate of testosterone oxidation catalyzed by other purified forms of rat liver cytochrome P450). The four major metabolites were 6 beta-hydroxytestosterone (51%), 2 beta-hydroxytestosterone (18%), 15 beta-hydroxytestosterone (11%) and 6-dehydrotestosterone (10%). The four minor metabolites were 18-hydroxytestosterone (3%), 1 beta-hydroxytestosterone (3%), 16 beta-hydroxytestosterone (2%), and androstenedione (2%). With the exception of 16 beta-hydroxytestosterone and androstenedione, the conversion of testosterone to each of these metabolites was inhibited greater than 85% when liver microsomes from various sources were incubated with rabbit polyclonal antibody against cytochrome P450p. This antibody, which recognized two electrophoretically distinct proteins in liver microsomes from troleandomycin-treated rats, did not inhibit testosterone oxidation by cytochromes P450a, P450b, P450h, or P450m. The catalytic turnover of microsomal cytochrome P450p was estimated from the increase in testosterone oxidation and the apparent increase in cytochrome P450 concentration following treatment of liver microsomes from troleandomycin- or erythromycin-induced rats with potassium ferricyanide (which dissociates the cytochrome P450p-inducer complex). Based on this estimate, the catalytic turnover values for purified, reconstituted cytochrome P450p were 4.2 to 4.6 times greater than the rate catalyzed by microsomal cytochrome P450p.

Specific drug binding to rat liver cytochrome P-450 isozymes induced by pregnenolone-16 alpha-carbonitrile and macrolide antibiotics. Implications for drug interactions

Clinical interactions of macrolides with various drugs lead to elimination impairment, increase of plasma concentration and overdose-like effects, resulting from modifications of their metabolism. Previous studies have shown that treatment of rats by the macrolide antibiotics of the oleandomycin and erythromycin series lead to the induction of an hepatic cytochrome P-450 which is implicated into their own metabolism. We have characterized PCN or macrolides induced cytochromes P-450 by their specific ability to interact with macrolide derivatives and, using the cytochrome P-450 spectral binding assays, we have shown that some compounds, implicated in drug interaction with macrolides, interact preferentially with the same cytochromes. This strongly suggests that specific blockage of cytochrome P-450 IIIA1 family by macrolides, is responsible for these drug interactions and that these interactions can be predicted easily by simple in vitro tests such as those described herein.

Characterization of a cDNA encoding a new member of the glucocorticoid-responsive cytochromes P450 in human liver

Adult human liver contains a form of cytochrome P450, termed HLp, that resembles the glucocorticoid-inducible cytochrome P450p in rat liver in its structure, function, and regulation and catalyzes the oxidation of such clinically important substrates as cyclosporin, nifedipine, erythromycin, and midazolam. Recent evidence, however, suggests that HLp may represent two or more closely related forms of cytochromes P450, one of which is termed P450nf. To search for additional members of the Class III human subfamily of HLp related genes, we screened a human liver cDNA library cloned in phage vector lambda gt11 with oligonucleotides and with a cDNA fragment related to HLp. We isolated a full-length cDNA (1709 nucleotides) encoding a new form of human cytochrome P450 termed HLp2. Analysis of HLp2 cDNA predicted a protein of 502 amino acids, weighing 57,294 Da 83% similar to HLp. HLp2 appears to represent a distinct gene as judged by partial sequence analysis of a cloned human gene and by hybridizations of Southern blots, under conditions of varying stringency, with a 3'-portion of HLp cDNA and with an oligonucleotide specific for HLp2. Northern blot analysis revealed that HLp/P450nf was present in all samples of liver mRNA from adult patients not treated with inducers of HLp, whereas HLp2 mRNA was undetectable in more than two-thirds. Human fetal liver RNA contained mRNA species 2.1 and 1.9 kb which hybridized with an HLp2 oligonucleotide. We conclude that HLp2 represents a third member of the Class III glucocorticoid-responsive gene family that is expressed in both fetal and adult human liver and may account for polymorphism in metabolism of clinically important drugs.

Metabolism of dihydroergotamine by a cytochrome P-450 similar to that involved in the metabolism of macrolide antibiotics

1. Previous studies have shown that the macrolide antibiotics, such as oleandomycin and erythromycin, enhance their own transformation into a stable metabolite-cytochrome P-450 complex, thus impairing monooxygenase activity. This cytochrome P-450 induced by macrolides is similar to the major form induced in rats by pregnenolone-16 alpha-carbonitrile (PCN) (III A1 isozyme). 2. The cytochrome P-450 isozyme induced in rats by PCN or macrolide antibiotics bound dihydroergotamine (DHE) with high affinity and was also capable of metabolizing the drug. However, phenobarbital administration enhanced the metabolism of DHE to a greater extent than would be expected from the levels of the PB-PCNE isoenzyme, indicating that other cytochrome P-450 proteins may also be involved in DHE metabolism. 3. DHE metabolism was inhibited by macrolide antibiotics both ex vivo and in vitro. The metabolite-cytochrome P-450 complex formed by the antibiotics impairs the metabolism of DHE, so that when the complex is dissociated the metabolic activity is restored. These findings explain the observed clinical interactions between macrolides and other drugs, and such an approach may prove useful in their prediction.

Modulation of the induction of rat hepatic cytochromes P-450 by selenium deficiency

The induction by phenobarbital of liver microsomal cytochrome P-450 has been demonstrated to be impaired in rats fed a selenium-deficient diet. Cytochrome P-450 isozyme specific immunologic and molecular techniques were used in the present study to better define the role of selenium in the induction of cytochrome P-450 by phenobarbital. Phenobarbital treatment of the selenium-deficient rats resulted in an increase in the level of total cytochrome P-450 50% of that observed with control rats and in a 10-fold increase in microsomal heme oxygenase. Quantitative immunoblot analyses demonstrated that the levels of cytochromes P-450b + e and P-450p in the phenobarbital-treated selenium-deficient rats were approximately 50% of those found in the phenobarbital-treated control rats. Finally, RNA hybridization studies using cDNA probes to cytochromes P-450b + e or P-450p demonstrated that the accumulations of the RNAs encoding these cytochromes P-450 were unaffected by the selenium status of the rats. These studies suggest that the impaired phenobarbital induction of the cytochromes P-450 in the selenium-deficient rats is the result of an increase in the degradation of the cytochromes P-450 or a decrease in the translation of the mRNAs coding for them.

Induction of cytochrome P-450 in cultured rat hepatocytes. The heterogeneous localization of specific isoenzymes using immunocytochemistry

Primary cultures of rat hepatocytes were exposed to phenobarbitone, clofibric acid, beta-naphthoflavone, isosafrole or dexamethasone for 3 days, and the induction of several cytochrome P-450 isoenzymes was demonstrated by increased catalytic activity, by Western blotting and by immunocytochemistry. The profiles of isoenzymes induced in vitro were compared with those induced in liver microsomes of rats dosed with the same agents. Clofibric acid, an agent which has not been thoroughly investigated previously, was shown to induce both in vivo and in vitro several P-450 isoenzymes normally inducible by phenobarbitone (PB1a, PB3a and PB3b) or steroids (PB2c). Immunocytochemical studies demonstrated that the inducible isoenzymes of cytochrome P-450 are not distributed evenly throughout the hepatocyte population, and increasing concentrations of phenobarbitone or beta-naphthoflavone in the medium results in an increasing proportion of 'induced' cells. However, whereas maximal concentrations of beta-naphthoflavone resulted in virtually all cells containing induced levels of MC1b, a maximal concentration of phenobarbitone resulted in only 30% of the cells containing induced levels of PB3a/PB3b. These results are discussed in relation to the heterogeneous distribution and induction of cytochrome P-450 in the intact liver.

In vitro interaction of rat liver cytochromes P-450 with erythromycin, oleandomycin and erythralosamine derivatives. Importance of structural factors

Several derivatives of the erythromycin, erythralosamine and oleandomycin series have been prepared. Their abilities to bind to rat liver microsomal cytochrome P-450 and to lead to the formation of stable 456 nm absorbing cytochrome P-450-metabolite complexes after their oxidative microsomal metabolism in vitro have been compared. The obtained data confirmed that cytochrome P-450 induced in rats either by macrolides or by 16 alpha-pregnenolone carbonitrile were the major isozymes involved in the binding of macrolides to liver microsomes and in metabolite-complex formation. They showed that (i) hydrophobicity was in general a beneficial factor for these two properties, (ii) the presence of a bulky substituent in position 3 of erythromycin dramatically decreased their affinity for these isozymes, and (iii) the simultaneous presence of bulky substituents in position 2' and 3 prevented iron-metabolite complex formation. These results led to the selection of two compounds, erythralosamine-2'-benzoate and erythralosamine-2',3-diacetate, which exhibited a particularly high affinity for macrolide inducible cytochrome P-450 and were very good precursors of cytochrome P-450-iron-metabolite complex formation.

Characterization of a phenobarbital-inducible dog liver cytochrome P450 structurally related to rat and human enzymes of the P450IIIA (steroid-inducible) gene subfamily

A cytochrome P450 called PBD-1 isolated from liver microsomes of an adult male Beagle dog treated with phenobarbital (PB) is structurally and functionally similar to members of the P450IIIA gene subfamily in rat and human liver microsomes. The sequence of the first 28 amino-terminal residues of PBD-1 is identical in 15 and 20 positions, respectively, to the P450IIIA forms P450p from rat and P450NF (and HLp) from human. Upon immunoblot analysis, anti-PBD-1 IgG recognizes PCNa (P450p) and PCNb (PB/PCN-E) from rat, P450NF from human, and two proteins in liver microsomes from both untreated and PB-treated dogs. Similarly, anti-PCNb IgG cross-reacts with PBD-1 and with at least one protein in microsomes from untreated dogs and two proteins in microsomes from PB-treated dogs. P450IIIA-form marker steroid 6 beta-hydroxylase activities increase 2.5-fold upon PB-treatment of dogs and are selectively inhibited by anti-PBD-1 IgG. NADPH-dependent triacetyloleandomycin (TAO) complex formation and erythromycin demethylase, also marker activities for P450IIIA forms from rats and humans, increase 4- and 5-fold in dog liver microsomes upon PB treatment, whereas immunochemically reactive PBD-1 is induced 3-fold. In microsomes from PB-treated dogs, 5 mg anti-PBD-1 IgG/nmol P450 inhibits greater than 75 and 50% of TAO complex formation and erythromycin demethylase activity, respectively. TAO complex formation is not inhibited by chloramphenicol, a selective inhibitor of the major PB-inducible dog liver cytochrome P450, PBD-2. These data suggest that PBD-1 or another immunochemically related form is responsible for a major portion of macrolide antibiotic metabolism by microsomes from PB-treated dogs and for steroid 6 beta-hydroxylation by microsomes from both untreated and PB-treated dogs. Major species differences were noted, however, in the apparent Km for 6 beta-hydroxylation of androstenedione by liver microsomes from untreated rats (24 microM), humans (380 microM), and untreated dogs (4700 microM).

Identification of cytochrome P450a (P450IIA1) as the principal testosterone 7 alpha-hydroxylase in rat liver microsomes and its regulation by thyroid hormones

In the preceding paper, evidence was presented that rat liver microsomes contain two structurally related isozymes of cytochrome P450, namely cytochromes P450a and P450m, that can both catalyze the 7 alpha-hydroxylation of testosterone. The aim of the present study was to determine the extent to which these two P450 isozymes are responsible for the 7 alpha-hydroxylation of testosterone catalyzed by rat liver microsomes. Four monoclonal antibodies against cytochrome P450a, designated A2, A4, A5, and A7, were prepared in BALB/c mice. Monoclonal antibodies A2 (an IgM), A4 (an IgG2b), and A5 (an IgG1) were determined to be distinct immunoglobulins, whereas A7 could not be distinguished from A5. All of the antibodies were highly specific for cytochrome P450a; none cross-reacted with cytochrome P450m or with 10 other P450 isozymes purified from rat liver microsomes. Competition experiments between unlabeled and horseradish peroxidase-conjugated antibodies revealed that each of the monoclonal antibodies recognized the same epitope on cytochrome P450a. None of the monoclonal antibodies bound to denatured cytochrome P450a, suggesting that they each bound to a spatial epitope. A monospecific, polyclonal antibody against cytochrome P450a was also prepared, as described in the preceding paper. The levels of cytochrome P450a in liver microsomes were determined by single radial immunodiffusion, Western immunoblot (with polyclonal antibody), and enzyme-linked immunosorbent assay with monoclonal antibody. The levels of cytochrome P450a declined with age in male but not female rats, and were inducible up to 10-fold by treatment of rats with various xenobiotics. The levels of cytochrome P450a (but not cytochrome P450m) were also elevated (approximately 3-fold) by thyroidectomy of mature male rats. Near normal levels of cytochrome P450a were restored by treatment of athyroid rats with triiodothyronine, whereas treatment with thyroxine was less effective in this regard. These changes in the levels of cytochrome P450a were highly correlated (r = 0.995) with changes in testosterone 7 alpha-hydroxylase activity. None of the monoclonal antibodies inhibited the catalytic activity of cytochrome P450a when reconstituted with NADPH-cytochrome P450 reductase and lipid. In contrast, the polyclonal antibody not only inhibited the catalytic activity of purified cytochrome P450a, but also completely inhibited (greater than 96%) the 7 alpha-hydroxylation of testosterone catalyzed by liver microsomes from immature and mature rats of both sexes and by liver microsomes from male rats treated with a variety of cytochrome P450 inducers.(ABSTRACT TRUNCATED AT 400 WORDS)

Purification of two isozymes of rat liver microsomal cytochrome P450 with testosterone 7 alpha-hydroxylase activity

Cytochrome P450a was purified to electrophoretic homogeneity from liver microsomes from immature male Long-Evans rats treated with Aroclor 1254. Rabbit polyclonal antibody raised against cytochrome P450a cross-reacted with cytochromes P450b, P450e, and P450f (which are structurally related to cytochrome P450a). The cross-reacting antibodies were removed by passing anti-P450a over an N-octylamino-Sepharose column containing these heterologous antigens. The immunoabsorbed antibody recognized only a single protein (i.e., cytochrome P450a) in liver microsomes from immature male rats treated with Aroclor 1254 (i.e., the microsomes from which cytochrome P450a was purified). However, the immunoabsorbed antibody recognized three proteins in liver microsomes from mature male rats, as determined by Western immunoblot. As expected, one of these proteins (Mr 48,000) corresponded to cytochrome P450a. The other two proteins did not correspond to cytochromes P450b, P450e, or P450f (as might be expected if the antibody were incompletely immunoabsorbed), nor did they correspond to cytochromes P450c, P450d, P450g, P450h, P450i, P450j, P450k, or P450p. One of these proteins was designated cytochrome P450m (Mr approximately 49,000), the other cytochrome P450n (Mr approximately 50,000). Like cytochrome P450a, cytochrome P450n was present in liver microsomes from both male and female rats. However, whereas cytochrome P450a was detectable in liver microsomes from 1-week-old rats, cytochrome P450n was barely detectable until the rats were at least 3 weeks old. Furthermore, in contrast to cytochrome P450a, the levels of cytochrome P450n did not decline appreciably with age in postpubertal male rats. Cytochrome P450m was detectable only in liver microsomes from postpubertal (greater than 4 week-old) male rats. Cytochromes P450m and P450n were isolated from liver microsomes from mature male rats and purified to remove cytochrome P450a. When reconstituted with NADPH-cytochrome P450 reductase and lipid, cytochrome P450n exhibited little testosterone hydroxylase activity, whereas cytochrome P450m catalyzed the 15 alpha-, 18-, 6 beta-, and 7 alpha-hydroxylations of testosterone at 10.8, 4.6, 2.0, and 1.9 nmol/nmol P450/min, respectively. The ability of cytochrome P450m to catalyze the 7 alpha-hydroxylation of testosterone was not due to contamination with cytochrome P450a, which catalyzed this reaction at approximately 25 nmol/nmol P450a/min. Cytochrome P450m also converted testosterone to several minor metabolites, including androstenedione and 15 beta-, 14 alpha-, and 16 alpha-hydroxytestosterone.(ABSTRACT TRUNCATED AT 400 WORDS)

Antibodies targeted against hypervariable and constant regions of cytochromes P450IIB1 and P450IIB2

Fusion proteins constructed between beta-galactosidase and six different segments of either cytochrome P450IIB1 or cytochrome P450IIB2 (ranging from 18 to 33 amino acids in length) were expressed in Escherichia coli. Rabbit antibodies raised against these fusion proteins were first adsorbed through a beta-galactosidase column and then immunopurified on a second column containing the corresponding fusion protein. With the exception of the antibodies directed against the hydrophobic amino-terminal segment of cytochrome P450IIB1, all the antipeptide antibodies recognized the major phenobarbital-inducible cytochromes P450IIB1 and -IIB2 on immunoblots of liver microsomal proteins. Two of the antibodies were raised against regions where cytochromes P450IIB1 and -IIB2 differ in primary structure, and were differentially reactive toward these two highly homologous cytochromes. Several of the antipeptide antibodies were also reactive with a third phenobarbital-inducible microsomal protein expressed in livers of some individual Sprague-Dawley rats which was shown to be more highly related to P450IIB1 than P450IIB2. This P450IIB1-related P450, designated P450IIB1*, was purified to apparent homogeneity and shown to hydroxylate the steroid hormones testosterone and androstenedione with the well-defined regiospecificity and high catalytic activity characteristic of P450IIB1. A fourth microsomal protein detected using the antipeptide antibodies appeared to be more highly related to P450IIB2. Because the segments on the P450 molecules recognized by these antipeptide antibodies are known, it is possible to predict where P450IIB1* and the P450IIB2-related protein differ from cytochromes P450IIB2 and -IIB1, respectively. These studies demonstrate the utility of site-specific anti-P450 antibodies raised to fusion peptides for studies on the expression of structurally related P450s and polymorphic variants within the cytochrome P450 gene superfamily.

Erythromycin breath test as an assay of glucocorticoid-inducible liver cytochromes P-450. Studies in rats and patients

The major P-450IIIA gene family member present in human liver is HLp which, like its rat liver orthologue P-450p, is inducible by glucocorticoids and catalyzes erythromycin N-demethylation. To develop a practical method to estimate the amounts of HLp in patients [14C]N-methyl erythromycin was injected into rats that had been pretreated with dexamethasone or with inducers of other forms of cytochrome P-450. The rate of demethylation of this substrate, measured simply as 14CO2 in the breath, correlated well with the concentrations of immunoreactive P-450p protein (r = 0.70), holocytochrome P-450p (r = 0.70), or with erythromycin N-demethylase activity (r = 0.90) determined in the liver microsomes prepared from each rat. Next, [14C]N-methyl erythromycin was administered to 30 patients and there was a sixfold interindividual variation in breath 14CO2 production seemingly unrelated to medications, smoking status or age. However, the average breath test values were twofold greater in female as compared to male patients (P less than 0.01). Breath 14CO2 production rose in patients retested after treatment with the P-450IIIA inducers dexamethasone (P less than 0.05) or rifampicin (P less than 0.05) and was decreased after treatment with the HLp inhibitor triacetyloleandomycin (P less than 0.05). We conclude that the erythromycin breath test provides a convenient assay of P-450IIIA cytochromes in rats and in some patients.

Interrelationship between RU38486 and the P450 activities in rat liver

Microsomal P450 monooxygenases contribute actively to the biotransformation of the antiglucocorticoid RU38486, an 11 beta-substituted nor-steroid. Pretreatment of adult rats by inducers of specific forms, belonging to different P450 subfamilies, affects the ability of liver microsomes to metabolize RU38486. Phenobarbital and pregnenolone 16 alpha-carbonitrile increase the metabolic activity of liver microsomes whereas methylcholanthrene decreases their capacity to oxidize the steroid. Thus P450 forms IIIA, IIB1,2 and IIC7 are good candidates to be involved in the degradation of this peculiar molecule. Our study has been completed by investigating whether RU38486 would influence the P450 spectrum. Whereas the treatment of rats with either a glucocorticoid (cortisol, dexamethasone) or an antiglucocorticoid (pregnenolone 16 alpha-carbonitrile) has been shown to induce the P450 activity by increasing the hepatic concentration of form IIIA, we observed a slight decrease of the P450 activity by treating the animals with RU38486. Moreover RU38486 was able to antagonize the P450 induction by the other steroids as well as it inhibits the synthesis of various liver enzymes induced by glucocorticoids (for instance tyrosine aminotransferase). These findings may be important for the therapeutic use of RU38486 since its inhibitory effect on P450 activity may be at the origin of drug interactions by modifying the endogenous hormonal status.

Developmental aspects of P450IIIA: prenatal activity and inducibility

We have used antiserum of defined specificity as well as a specific inducers and inhibitors of P450IIIA1(2) to determine the fetal occurrence and inducibility of this enzyme in rats. Apparently absent from uninduced fetal rat liver (or present in extremely low amounts) cytochrome P450IIIA1(2) becomes increasingly inducible as a function of gestational age. In adult rats, it is now apparent that there are at least two inducible members and one male-specific constitutive member of the IIIA subfamily. The ontogenesis of these enzymes from 2 weeks post partum to puberty has also been determined. The male-specific occurrence of P450IIIA2 subject to testosterone imprinting and maintenance has been proposed. Inconsistencies persist, however. Waxman et al. have proposed the perinatal occurrence in male and female rats with subsequent suppression in females, whereas others have not detected P450IIIA1(2) in uninduced perinatal rat liver. These differences remain unresolved and reflect the difficulties in defining the individual enzyme specificities for various substrates and of antiserum reactivity. Approaches recently applied to investigations of the IIB subfamily of cytochromes P-450 should contribute greatly to the elucidation of factors governing the ontogenesis of IIIA in rats and humans. Recently, cDNA probes capable of discriminating P450IIB1 and P450IIB2 (commonly referred to as P450s b and e, respectively) were utilized to discriminate the developmental regulation of these immune cross-reactive enzymes. cDNA probes specific for the constitutive and inducible P450IIIA enzymes should clarify the P450IIIA ontogeny in rats. However, in light of regulatory differences among the human and rat members of P450IIIA, it is apparent that the extrapolation of human biotransforming potential from results of animal models must be approached with great caution.

Inhibition of steroid 5 alpha-reductase and its effects on testosterone hydroxylation by rat liver microsomal cytochrome P-450

It has been shown previously that liver microsomal steroid 5 alpha-reductase activity increases with age in female but not male rats, which coincides with a female-specific, age-dependent decline in the cytochrome P-450-dependent oxidation of testosterone to 1 beta-, 2 alpha-, 2 beta-, 6 alpha-, 6 beta-, 7 alpha-, 15 beta-, 16 alpha-, 16 beta-, and 18-hydroxytestosterone and androstenedione. To determine whether the increase in steroid 5 alpha-reductase activity is responsible for the decrease in testosterone oxidation, we have examined the effects of the steroid 5 alpha-reductase inhibitor, 4-MA (17 beta-N,N-diethylcarbamoyl-4-methyl-4-aza-5 alpha-androstan-3-one), on the pathways of testosterone oxidation catalyzed by rat liver microsomes. We have also determined which hydroxytestosterone metabolites are substrates for steroid 5 alpha-reductase. At concentrations of 0.1 to 10 microM, 4-MA completely inhibited steroid 5 alpha-reductase activity without inhibiting the pathways of testosterone oxidation catalyzed by liver microsomes from rats of different age and sex, and from rats induced with phenobarbital or pregnenolone-16 alpha-carbonitrile. 4-MA (10 microM) had little or no effect on the oxidation of testosterone catalyzed by liver microsomes from mature male rats (which have low steroid 5 alpha-reductase activity). In contrast, the hydroxylated testosterone metabolites formed by liver microsomes from mature female rats (which have high steroid 5 alpha-reductase activity) accumulated to a much greater extent in the presence of 4-MA. Evidence is presented that 4-MA increases the accumulation of hydroxytestosterones by two mechanisms. First, 4-MA inhibited the 5 alpha-reduction of those metabolites (such as 6 beta-hydroxytestosterone) that were found to be excellent substrates for steroid 5 alpha-reductase. In the absence of 4-MA, these metabolites eventually disappeared from incubations containing liver microsomes from mature female rats. Second, 4-MA inhibited the formation of 5 alpha-dihydrotestosterone, which otherwise competed with testosterone for oxidation by cytochrome P-450. This second mechanism explains why 4-MA increased the accumulation of metabolites (such as 7 alpha-hydroxytestosterone) that were found to be poor substrates for steroid 5 alpha-reductase. Despite its marked effect on the accumulation of hydroxylated testosterone metabolites, 4-MA had no effect on their initial rate of formation by liver microsomes from either male or female rats.(ABSTRACT TRUNCATED AT 400 WORDS)

Human liver microsomal steroid metabolism: identification of the major microsomal steroid hormone 6 beta-hydroxylase cytochrome P-450 enzyme

Cytochrome P-450-dependent steroid hormone metabolism was studied in isolated human liver microsomal fractions. 6 beta hydroxylation was shown to be the major route of NADPH-dependent oxidative metabolism (greater than or equal to 75% of total hydroxylated metabolites) with each of three steroid substrates, testosterone, androstenedione, and progesterone. With testosterone, 2 beta and 15 beta hydroxylation also occurred, proceeding at approximately 10% and 3-4% the rate of microsomal 6 beta hydroxylation, respectively, in each of the liver samples examined. Rates for the three steroid 6 beta-hydroxylase activities were highly correlated with each other (r = 0.95-0.97 for 25 individual microsomal preparations), suggesting that a single human liver P-450 enzyme is the principal microsomal 6 beta-hydroxylase catalyst with all three steroid substrates. Steroid 6 beta-hydroxylase rates correlated well with the specific content of human P-450NF (r = 0.69-0.83) and with its associated nifedipine oxidase activity (r = 0.80), but not with the rates for debrisoquine 4-hydroxylase, phenacetin O-deethylase, or S-mephenytoin 4-hydroxylase activities or the specific contents of their respective associated P-450 forms in these same liver microsomes (r less than 0.2). These correlative observations were supported by the selective inhibition of human liver microsomal 6 beta hydroxylation by antibody raised to either human P-450NF or a rat homolog, P-450 PB-2a. Anti-P-450NF also inhibited human microsomal testosterone 2 beta and 15 beta hydroxylation in parallel to the 6 beta-hydroxylation reaction. This antibody also inhibited rat P-450 2a-dependent steroid hormone 6 beta hydroxylation in uninduced adult male rat liver microsomes but not the steroid 2 alpha, 16 alpha, or 7 alpha hydroxylation reactions catalyzed by other rat P-450 forms. Finally, steroid 6 beta hydroxylation catalyzed by either human or rat liver microsomes was selectively inhibited by NADPH-dependent complexation of the macrolide antibiotic triacetyloleandomycin, a reaction that is characteristic of members of the P-450NF gene subfamily (P-450 IIIA subfamily). These observations establish that P-450NF or a closely related enzyme is the major catalyst of steroid hormone 6 beta hydroxylation in human liver microsomes, and furthermore suggest that steroid 6 beta hydroxylation may provide a useful, noninvasive monitor for the monooxygenase activity of this hepatic P-450 form.

Multiplicity of steroid-inducible cytochromes P-450 in rat liver microsomes

Highly purified preparations of steroid-inducible cytochromes P-450 have been isolated from liver microsomes of adult male Sprague-Dawley rats treated with pregnenolone-16 alpha-carbonitrile (PCN), phenobarbital (PB), or triacetyloleandomycin (TAO). The interrelationships among these preparations as well as their relationship to the major forms of cytochrome P-450 of this gene family identified in other laboratories have been evaluated by amino-terminal sequence analysis, sodium dodecyl sulfate-polyacrylamide gel electrophoresis, chromatography on DEAE-Sephacel, and Ouchterlony double-immunodiffusion. The results indicate that a cytochrome P-450 previously isolated in this laboratory and referred to as PCNa (P.E. Graves, L.S. Kaminsky, and J. Halpert (1987) Biochemistry 26, 3887-3894) is the major cytochrome P-450 induced in rats by TAO, and corresponds to P-450p (K.A. Hostetler, S.A. Wrighton, P. Kremers, and P.S. Guzelian (1987) Biochem. J. 245, 27-33) and to P450PCN1 (F.J. Gonzalez, B.J. Song, and J.P. Hardwick (1986) Mol. Cell. Biol. 6, 2969-2976). A second previously isolated cytochrome P-450 termed PCNb in this laboratory appears to be identical to PB/PCN-E (F.P. Guengerich, G.A. Dannan, S.T. Wright, M.V. Martin, and L.S. Kaminsky (1982) Biochemistry 21, 6019-6030). PCNb as well as a third cytochrome P-450 termed PCNc isolated from PB-treated rats both correspond in amino-terminal sequence to the putative protein product of the pP450PCN2/cDNA clone of Gonzalez et al. These results document at the protein level the multiplicity of steroid-inducible rat liver cytochromes P-450.