Facile production of minor metabolites for drug development using a CYP3A shuffled library.

Metabolic profiling of new drugs is limited by the difficulty in obtaining sufficient quantities of minor metabolites for definitive structural identification. Biocatalytic methods offer the potential to produce metabolites that are difficult to synthesize by traditional medicinal chemistry. We hypothesized that the regioselectivity of the drug metabolizing cytochrome P450s could be altered by directed evolution to produce minor metabolites of drugs in development. A biocatalyst library was constructed by DNA shuffling of four CYP3A forms. The library contained 11 ± 4 (mean ± SD) recombinations and 1 ± 1 spontaneous mutations per mutant. On expression in Escherichia coli, 96% of mutants showed detectable activity to at least one probe substrate. Using testosterone as a model drug-like substrate, mutants were found that preferentially formed metabolites produced in only trace amounts by parental forms. A single 1.6L batch culture of one such mutant enabled the facile isolation of 0.3mg of the minor metabolite 1ß-hydroxytestosterone and its ab initio structural determination by 1D- and 2D-NMR spectroscopy.

Metabolism of the major Echinacea alkylamide N-isobutyldodeca-2E,4E,8Z,10Z-tetraenamide by human recombinant cytochrome P450 enzymes and human liver microsomes.

Echinacea preparations are used for the treatment and prevention of upper respiratory tract infections. The phytochemicals believed responsible for the immunomodulatory properties are the alkylamides found in ethanolic extracts, with one of the most abundant being the N-isobutyldodeca-2E,4E,8Z,10Z-tetraenamide (1). In this study, we evaluated the human cytochrome P450 enzymes involved in the metabolism of this alkylamide using recombinant P450s, human liver microsomes and pure synthetic compound. Epoxidation, N-dealkylation and hydroxylation products were detected, with different relative amounts produced by recombinant P450s and microsomes. The major forms showing activity toward the metabolism of 1 were CYP1A1, CYP1A2 (both producing the same epoxide and N-dealkylation product), CYP2A13 (producing two epoxides), and CYP2D6 (producing two epoxides and an hydroxylated metabolite). Several other forms showed less activity. In incubations with human liver microsomes and selective inhibitors, CYP2E1 was found to be principally responsible for producing the dominant, hydroxylation product, whereas CYP2C9 was the principal source of the epoxides and CYP1A2 was responsible for the dealkylation product. In summary, in this study the relative impacts of the main human xenobiotic-metabolizing cytochrome P450s on the metabolism of a major Echinacea alkylamide have been established and the metabolites formed have been identified.

Membrane integration of recombinant human P450 forms.

Amino terminal sequence modification of cytochrome P450 enzymes is often necessary to achieve expression in bacteria. The aim of this study was to examine the effect of such modifications on membrane integration and P450 activity. Forms that retained substantial N-terminal hydrophobic sequences remained unaffected by treatments to remove peripheral membrane proteins and were released only by detergent. Truncated P450s 2A13, 2C9 (delta 3-20), 2C19 (delta 3-20), 2D6 (DB11) and 2E1 remained principally membrane-bound, but some P450 was found in the soluble fraction and could be partially extracted by alkaline and high salt treatments. The subcellular localization of P450s 2C9 and 2C19 assessed by fluorescence microscopy mirrored the distribution between subcellular fractions. The MALLLAVFL modified forms of P450 2C9 YFP, P450 2C18 YFP and P450 2C19 YFP were found primarily at the periphery of the cells, whereas the truncated forms of P450 2C9 (delta 3-20) YFP and 2C19 (delta 3-20) YFP were observed at the periphery as well as inside the cells. N-terminal variants of P450s 2C9 and 2C19 showed altered kinetics towards form-selective substrates. Rates of diclofenac 4 -hydroxylation by P450 2C9 and luciferin H-EGE metabolism by P450 2C19 were higher for the MALLLAVFL-modified forms compared with the (delta 3-20) truncated forms despite supplementation of truncated form incubations with additional reductase. Thus, N-terminal sequence modifications changed the degree of membrane integration, potentially affecting subcellular localization, interactions with redox partners, and hence enzymatic activity.

Cytochrome P450 enzyme-mediated degradation of Echinacea alkylamides in human liver microsomes.

Echinacea preparations are widely used herbal remedies for the prevention and treatment of colds. In this study we have investigated the metabolism by human liver microsomes of the alkylamide components from an Echinacea preparation as well as that of pure synthetic alkylamides. No significant degradation of alkylamides was evident in cytosolic fractions. Time- and NADPH-dependent degradation of alkylamides was observed in microsomal fractions suggesting they are metabolised by cytochrome P450 (P450) enzymes in human liver. There was a difference in the susceptibility of 2-ene and 2,4-diene pure synthetic alkylamides to microsomal degradation with (2E)-N-isobutylundeca-2-ene-8,10-diynamide (1) metabolised to only a tenth the extent of (2E,4E,8Z,10Z)-N-isobutyldodeca-2,4,8,10-tetraenamide (3) under identical incubation conditions. Markedly less degradation of 3 was evident in the mixture of alkylamides present in an ethanolic Echinacea extract, suggesting that metabolism by liver P450s was dependent both on their chemistry and the combination present in the incubation. Co-incubation of 1 with 3 at equimolar concentrations resulted in a significant decrease in the metabolism of 3 by liver microsomes. This inhibition by 1, which has a terminal alkyne moiety, was found to be time- and concentration-dependent, and due to a mechanism-based inactivation of the P450s. Alkylamide metabolites were detected and found to be the predicted epoxidation, hydroxylation and dealkylation products. These findings suggest that Echinacea may effect the P450-mediated metabolism of other concurrently ingested pharmaceuticals.

Characterization of the human cytochrome P450 forms involved in metabolism of tamoxifen to its alpha-hydroxy and alpha,4-dihydroxy derivatives.

Tamoxifen is a known hepatocarcinogen in rats and is associated with an increased incidence of endometrial cancer in patients. One mechanism for these actions is via bioactivation, where reactive metabolites are generated that are capable of binding to DNA or protein. Several metabolites of tamoxifen have been identified that appear to predispose to adduct formation. These include alpha-hydroxytamoxifen, alpha,4-dihydroxytamoxifen, and alpha-hydroxy-N-desmethyltamoxifen. Previous studies have shown that cytochrome P450 (P450) enzymes play an important role in the biotransformation of tamoxifen. The aim of our work was to determine which P450 enzymes were capable of producing alpha-hydroxylated metabolites from tamoxifen. When tamoxifen (18 or 250 microM) was used as the substrate, P450 3A4, and to a lesser extent, P450 2D6, P450 2B6, P450 3A5, P450 2C9, and P450 2C19 all produced a metabolite with the same HPLC retention time as alpha-hydroxytamoxifen at either substrate concentration tested. This peak was well-separated from 4-hydroxy-N-desmethyltamoxifen, which eluted substantially later under the chromatographic conditions used. No alpha,4-dihydroxytamoxifen was detected in incubations with any of the forms with tamoxifen as substrate. However, when 4-hydroxytamoxifen (100 microM) was used as the substrate, P450 2B6, P450 3A4, P450 3A5, P450 1B1, P450 1A1, and P450 2D6 all produced detectable concentrations of alpha,4-dihydroxytamoxifen. These studies demonstrate that multiple human P450s, including forms found in the endometrium, may generate reactive metabolites in women undergoing tamoxifen therapy, which could subsequently play a role in the development of endometrial cancer.

Metabolic activation of polycyclic aromatic hydrocarbons and other procarcinogens by cytochromes P450 1A1 and P450 1B1 allelic variants and other human cytochromes P450 in Salmonella typhimurium NM2009.

A variety of polycyclic aromatic hydrocarbons and their dihydrodiol derivatives, arylamines, heterocyclic amines, and nitroarenes, were incubated with cDNA-based recombinant (Escherichia coli or Trichoplusia ni) systems expressing different forms of human cytochrome P450 (P450 or CYP) and NADPH-P450 reductase using Salmonella typhimurium tester strain NM2009, and the resultant DNA damage caused by the reactive metabolites was detected by measuring expression of umu gene in the cells. Recombinant (bacterial) CYP1A1 was slightly more active than any of four CYP1B1 allelic variants, CYP1B1*1, CYP1B1*2, CYP1B1*3, and CYP1B1*6, in catalyzing activation of chrysene-1,2-diol, benz[a]anthracene-trans-1,2-, 3,4-, 5,6-, and 8,9-diol, fluoranthene-2,3-diol, dibenzo[a,l]pyrene, benzo[c]phenanthrene, and dibenz[a,h]anthracene and several arylamines and heterocyclic amines, whereas CYP1A1 and CYP1B1 enzymes had essentially similar catalytic specificities toward other procarcinogens, such as (+)-, (-)-, and (+/-)-benzo[a]pyrene-7,8-diol, 5-methylchrysene-1,2-diol, 7,12-dimethylbenz[a]anthracene-3,4-diol, dibenzo[a,l]pyrene-11,12-diol, benzo[b]fluoranthene-9,10-diol, benzo[c]chrysene, 5,6-dimethylchrysene-1,2-diol, benzo[c]phenanthrene-3,4-diol, 7,12-dimethylbenz[a]anthracene, benzo[a]pyrene, 5-methylchrysene, and benz[a]anthracene. We also determined activation of these procarcinogens by recombinant (T. ni) human P450 enzymes in S. typhimurium NM2009. There were good correlations between activities of procarcinogen activation by CYP1A1 preparations expressed in E. coli and T. ni cells, although basal activities with three lots of CYP1B1 in T. ni cells were very high without substrates and NADPH in our assay system. Using 14 forms of human P450s (but not CYP1B1) (in T. ni cells), we found that CYP1A2, 2C9, 3A4, and 2C19 catalyzed activation of several of polycyclic aromatic hydrocarbons at much slower rates than those catalyzed by CYP1A1 and that other enzymes, including CYP2A6, 2B6, 2C8, 2C18, 2D6, 2E1, 3A5, 3A7, and 4A11, were almost inactive in the activation of polycyclic aromatic hydrocarbons examined here.

Specificity of 17beta-oestradiol and benzo[a]pyrene oxidation by polymorphic human cytochrome P4501B1 variants substituted at residues 48, 119 and 432.

1. Eight human cytochrome P4501B1 (CYP1B1) allelic variants, namely Arg48 Ala119 Leu432, Arg48 Ala119 Val432 Gly48 Ala119 Leu432, Gly48 Ala119 Val432, Arg48 Ser119 Leu432, Arg48 Ser119 Val432, Gly48 Ser119 Leu432 and Gly48 Ser119 Va1432 (all with Asn453), were expressed in Escherichia coli together with human NADPH-P450 reductase and their catalytic specificities towards oxidation of 17beta-oestradiol and benzo[a]pyrene were determined. 2. All of the CYP1B1 variants expressed in bacterial membranes showed Fe2+.CO versus Fe2+ difference spectra with wavelength maxima at 446 nm and they reacted with antibodies raised against recombinant human CYP1B1 in immunoblots. The ratio of expression of the reductase to CYP1B1 in these eight preparations ranged from 0.2 to 0.5. 3. CYP1B1 Arg48 variants tended to have higher activities for 17beta-oestradiol 4-hydroxylation than Gly48 variants, although there were no significant variations in 17beta-oestradiol 2-hydroxylation activity in these eight CYP1B1 variants. Interestingly, ratios of formation of 17beta-oestradiol 4-hydroxylation to 2-hydroxylation by these CYP1B1 variants were higher in all of the Val432 forms than the corresponding Leu432 forms. 4. In contrast, Leu432 forms of CYP1B1 showed higher rates of oxidation of benzo[a]pyrene (to the 7,8-dihydoxy-7,8-dihydrodiol in the presence of epoxide hydrolase) than did the Val432 forms. 5. These results suggest that polymorphic human CYP1B1 variants may cause some altered catalytic specificity with 17beta-oestradiol and benzo[a]pyrene and may influence susceptibilities of individuals towards endogenous and exogenous carcinogens.

Metabolic activation of heterocyclic amines and other procarcinogens in Salmonella typhimurium umu tester strains expressing human cytochrome P4501A1, 1A2, 1B1, 2C9, 2D6, 2E1, and 3A4 and human NADPH-P450 reductase and bacterial O-acetyltransferase.

We investigated roles of different forms of cytochrome P450 (P450 or CYP) in the metabolic activation of heterocyclic amines (HCAs) and other procarcinogens to genotoxic metabolite(s) in the newly developed umu tester strains Salmonella typhimurium (S. typhimurium) OY1002/1A1, OY1002/1A2, OY1002/1B1, OY1002/2C9, OY1002/2D6, OY1002/2E1 and OY1002/3A4, which express respective human P450 enzymes and NADPH-cytochrome P450 reductase (reductase) and bacterial O-acetyltransferase (O-AT). These strains were established by introducing two plasmids into S. typhimurium TA1535, one carrying both P450 and the reductase cDNA in a bicistronic construct under control of an IPTG-inducible double tac promoter and the other, pOA102, carrying O-AT and umuC"lacZ fusion genes. Expression levels of CYP were found to range between 35 to 550 nmol/l cell culture in the strains tested. O-AT activities in different strains ranged from 52 to 125 nmol isoniazid acetylated/min/mg protein. All HCAs tested, and 2-aminoanthracene and 2-aminofluorene exhibited high genotoxicity in the OY1002/1A2 strain, and genotoxicity of 2-amino-3-methylimidazo [4,5-f]quinoline was detected in both the OY1002/1A1 and OY1002/1A2 strains. 3-Amino-1,4-dimethyl-5H-pyrido[4,3-b]-indole and 3-amino-1-methyl-5H-pyrido[4,3-b]-indole were activated in the OY1002/1A1, OY1002/1B1, OY1002/1A2, and OY1002/3A4 strains. Aflatoxin B(1) exhibited genotoxicity in the OY1002/1A2, OY1002/1A1, and OY1002/3A4 strains. beta-Naphthylamine and benzo[a]pyrene did not exhibit genotoxicity in any of the strains. These results suggest that CYP1A2 is the major cytochrome P450 enzyme involved in bioactivation of HCAs.

Exploiting the versatility of human cytochrome P450 enzymes: the promise of blue roses from biotechnology.

The cytochrome P450 (P450) enzymes involved in drug metabolism are among the most versatile biological catalysts known. A small number of discrete forms of human P450 are capable of catalyzing the monooxygenation of a practically unlimited variety of xenobiotic substrates, with each enzyme showing a more or less wide and overlapping substrate range. This versatility makes P450s ideally suited as starting materials for engineering designer catalysts for industrial applications. In the course of heterologous expression of P450s in bacteria, we observed the unexpected formation of blue pigments. Although this was initially assumed to be an artifact, subsequent work led to the discovery of a new function of P450s in intermediary metabolism and toxicology, new screens for protein engineering, and potential applications in the dye and horticulture industries.

Oxidation of indole by cytochrome P450 enzymes.

Indole is a product of tryptophan catabolism by gut bacteria and is absorbed into the body in substantial amounts. The compound is known to be oxidized to indoxyl and excreted in urine as indoxyl (3-hydroxyindole) sulfate. Further oxidation and dimerization of indoxyl leads to the formation of indigoid pigments. We report the definitive identification of the pigments indigo and indirubin as products of human cytochrome P450 (P450)-catalyzed metabolism of indole by visible, (1)H NMR, and mass spectrometry. P450 2A6 was most active in the formation of these two pigments, followed by P450s 2C19 and 2E1. Additional products of indole metabolism were characterized by HPLC/UV and mass spectrometry. Indoxyl (3-hydroxyindole) was observed as a transient product of P450 2A6-mediated metabolism; isatin, 6-hydroxyindole, and dioxindole accumulated at low levels. Oxindole was the predominant product formed by P450s 2A6, 2E1, and 2C19 and was not transformed further. A stable end product was assigned the structure 6H-oxazolo[3,2-a:4, 5-b']diindole by UV, (1)H NMR, and mass spectrometry, and we conclude that P450s can catalyze the oxidative coupling of indoles to form this dimeric conjugate. On the basis of these results, we propose that the P450/NADPH-P450 reductase system can catalyze oxidation of indole to a variety of products.

Formation of a dihydroxy metabolite of phenytoin in human liver microsomes/cytosol: roles of cytochromes P450 2C9, 2C19, and 3A4.

Formation of four oxidative metabolites from the anticonvulsant drug phenytoin (DPH) catalyzed by human liver microsomal cytochrome P450 (P450) enzymes was determined simultaneously. Under the conditions in which linearity for formation of 4'-hydroxylated DPH (4'-HPPH; main metabolite) was observed, human liver cytosol increased microsome-mediated DPH oxidation. 3',4'-Dihydroxylated product (3', 4'-diHPPH) formation was 10 to 40% of total DPH oxidation in the presence of liver cytosol. 3'-Hydroxy DPH formation was catalyzed by only one of the human liver microsomal samples examined and 3', 4'-dihydrodiol formation could not be detected in all samples. In the presence of liver cytosol, 3',4'-diHPPH formation activity from 100 microM 4'-HPPH was correlated with testosterone 6beta-hydroxylation activity and CYP3A4 content. However, 3', 4'-diHPPH formation using 1 or 10 microM 4'-HPPH as a substrate was not correlated with contents of any P450s or marker activities. Of 10 cDNA-expressed human P450 enzymes examined, CYP2C19, CYP2C9, and CYP3A4 catalyzed 3',4'-diHPPH formation from the primary hydroxylated metabolites (3'-hydroxy-DPH and 4'-HPPH). Fluvoxamine and anti-CYP2C antibody inhibited 3',4'-diHPPH formation from 10 microM 4'-HPPH in a human liver sample that contained relatively high levels of CYP2C, whereas ketoconazole and anti-CYP3A antibody showed inhibitory effects on the activities in liver microsomal samples in which CYP3A4 levels were relatively high. These results suggest that CYP2C9, CYP2C19, and CYP3A4 all have catalytic activities in 3',4'-diHPPH formation from primary hydroxylated metabolites in human liver and that the hepatic contents of these three P450 forms determine which P450 enzymes play major roles of DPH oxidation in individual humans.

Roles of NADPH-P450 reductase in the O-deethylation of 7-ethoxycoumarin by recombinant human cytochrome P450 1B1 variants in Escherichia coli.

Four human cytochrome P450 1B1 (CYP1B1) allelic variants were purified from membranes of Escherichia coli in which respective CYP1B1 cDNAs and human NADPH-P450 reductase cDNA have been introduced. Purified CYP1B1 variants were used to reconstitute 7-ethoxycoumarin O-deethylation activities with purified rabbit liver or recombinant (rat) NADPH-P450 reductase in the phospholipid vesicles and compared with those catalyzed by CYP1B1 enzymes in the membranes of E. coli in monocistronic (by adding the reductase) and bicistronic (without addition of extra reductase) systems. In the bicistronic system, the ratio of expression of NADPH-P450 reductase to CYP1B1 proteins was found to range from 0.2 to 0.5. Purified CYP1B1 enzymes (under optimal reconstitution conditions) catalyzed 7-ethoxycoumarin O-deethylation at rates one-third to one-fourth of those catalyzed by membranes of E. coli coexpressing CYP1B1 and the reductase proteins. Full catalytic activities in reconstituted systems were achieved with a twofold molar excess of NADPH-P450 reductase to CYP1B1; in membranes of E. coli with the monocistronic CYP1B1 construct, an eightfold molar excess of reductase to CYP1B1 was required. However, in membranes of bicistronic constructs, there was no additional stimulation of 7-ethoxycoumarin O-deethylation by extra NADPH-P450 reductase, despite the fact that the molar ratio of expression levels of reductase to CYP1B1 was <0.5. These results suggest that NADPH-P450 reductase produced in the bacterial membranes is more active in interacting with CYP1B1 proteins in the bicistronic system than the reductase added to artificial phospholipid vesicles or bacterial membranes.

Phenytoin metabolism by human cytochrome P450: involvement of P450 3A and 2C forms in secondary metabolism and drug-protein adduct formation.

The anticonvulsant phenytoin (5,5-diphenylhydantoin) provokes a skin rash in 5 to 10% of patients, which heralds the start of an idiosyncratic reaction that may result from covalent modification of normal self proteins by reactive drug metabolites. Phenytoin is metabolized by cytochrome P450 (P450) enzymes primarily to 5-(p-hydroxyphenyl-),5-phenylhydantoin (HPPH), which may be further metabolized to a catechol that spontaneously oxidizes to semiquinone and quinone species that covalently modify proteins. The aim of this study was to determine which P450s catalyze HPPH metabolism to the catechol, proposed to be the final enzymatic step in phenytoin bioactivation. Recombinant human P450s were coexpressed with NADPH-cytochrome P450 reductase in Escherichia coli. Novel bicistronic expression vectors were constructed for P450 2C19 and the three major variants of P450 2C9, i.e., 2C9*1, 2C9*2, and 2C9*3. HPPH metabolism and covalent adduct formation were assessed in parallel. P450 2C19 was the most effective catalyst of HPPH oxidation to the catechol metabolite and was also associated with the highest levels of covalent adduct formation. P450 3A4, 3A5, 3A7, 2C9*1, and 2C9*2 also catalyzed bioactivation of HPPH, but to a lesser extent. Fluorographic analysis showed that the major targets of adduct formation in bacterial membranes were the catalytic P450 forms, as suggested from experiments with human liver microsomes. These results suggest that P450 2C19 and other forms from the 2C and 3A subfamilies may be targets as well as catalysts of drug-protein adduct formation from phenytoin.

Association of CYP1B1 genetic polymorphism with incidence to breast and lung cancer.

Cytochrome P450 1B1 (CYP1B1) participates in the metabolic activation of a number of procarcinogens including benzo[a]pyrene and the hydroxylation of 17beta-estradiol at the C-4 position. In this study, we investigated the association between CYP1B1 genetic polymorphism and breast or lung cancer incidence. The Ala-Ser polymorphism at codon 119 in presumed substrate recognition site 1 was significantly associated with the incidence of breast or squamous cell carcinoma of the lung. On the other hand, Leu-Val polymorphism at codon 432 did not show any association to the cancers. An allele containing both Ala and Leu simultaneously, comprised 75% of alleles among 315 Japanese healthy controls, was significantly inversely associated with breast cancer incidence. When expressed in a recombinant system, this CYP1B1 cDNA showed the lowest 17beta-estradiol 4-hydroxylase activity among four different variant forms of CYP1B1. Thus, inter-individual differences in activation of procarcinogens or metabolism of oestrogen originating from genetic polymorphisms of the human CYP1B1 gene may contribute to the susceptibility of human cancers.

What makes P450s work? Searches for answers with known and new P450s.

Random mutagenesis has been developed as an approach for the study of human cytochrome P450 (P450) enzymes and their structure and function. Sensitive screening methods are critical for the success of this approach. We have developed one system that takes advantage of the ability of human P450 1A2 to activate heterocyclic amines to mutagenic products [A. Parikh, P. D. Josephy, and F. P. Guengerich, Biochemistry, 38, 5283-5289 (1999)]. Mutants with both attenuated and enhanced activity have been recovered and subjected to further kinetic analysis. For phenacetin O-deethylation, the E225I mutant had kcat 6x > wild type; D320A had kcat 1/10x < wild type (and Km 15 x > wild type). With all three P450s, the rate of first electron reduction was similar, and all had similar binding constants for phenacetin (approximately 15 microM). All three forms yielded intermolecular, noncompetitive kinetic deuterium isotope effects of 1.5-2 [DV and D(V/K)] for O-deethylation of [OCD2CH3]-phenacetin. All three forms of P450 1A2 also formed a minor product, the acetol (C-hydroxylation of the acetyl group). This reaction had a deuterium isotope effect of approximately 14 with all three forms of the enzyme, and C-H bond breaking is the rate-determining step. Another approach to P450 2A6 involves the recent observation that this P450 can accumulate indigo [E. M. J. Gillam, A. M. A. Aguinaldo, L. M. Notley, D. Kim, R. G. Mundkowski, A. A. Volkov, F. H. Arnold, P. Soucek, J. T. DeVoss, and F. P. Guengerich, Biochem. Biophys. Res. Commun 265, 469-472 (1999)]. Current results indicate that this process involves the conversion of endogenous indole to indoxyl by the P450. The reaction may be used in assays of random mutants and has some potential applications in industry.

Formation of indigo by recombinant mammalian cytochrome P450.

The development of bicistronic systems for coexpression of recombinant human cytochrome P450 enzymes (P450s) with their redox partner, NADPH-cytochrome P450 reductase (NPR), has enabled P450 activity to be reconstituted within bacterial cells. During expression of recombinant P450 2E1 and some other forms, we observed the formation of a blue pigment in bacterial cultures. The pigment was extracted from cultures and shown to comigrate with standard indigo on TLC. UV-visible spectroscopy and mass spectrometric analysis provided further support for identification of the pigment as indigo. Indigo is known to form following the spontaneous oxidation of 3-hydroxyindole. Accordingly, we speculated that indole, formed as a breakdown product of tryptophan in bacteria, was hydroxylated by the P450 system, leading to indigo formation. Bacterial membranes containing recombinant P450 2E1 and human NPR were incubated in vitro with indole and shown to catalyze formation of a blue pigment in a time- and cofactor-dependent manner. These studies suggest potential applications of mammalian P450 enzymes in industrial indigo production or in the development of novel colorimetric assays based on indole hydroxylation.

Enhancement of cytochrome P-450 3A4 catalytic activities by cytochrome b(5) in bacterial membranes.

Activities of testosterone, nifedipine, and midazolam oxidation by recombinant cytochrome P-450 (P-450) 3A4 coexpressed with human NADPH-P-450 reductase (NPR) in bacterial membranes (CYP3A4/NPR membranes) were determined in comparison with those of other recombinant systems and of human liver microsomes with high contents of CYP3A4. Growth conditions for Escherichia coli transformed with the bicistronic construct affected expression levels of CYP3A4 and NPR; an excess of NPR over P-450 in membrane preparations enhanced CYP3A4-dependent testosterone 6beta-hydroxylation activities of the CYP3A4/NPR membranes. Cytochrome b(5) (b(5)) and apolipoprotein b(5) further enhanced the testosterone 6beta-hydroxylation activities of CYP3A4/NPR membranes after addition to either bacterial membranes or purified enzymes. NPR was observed to enhance catalytic activity when added to the CYP3A4/NPR membranes, either in the form of bacterial membranes or as purified NPR (in combination with cholate and b(5)). Apparent maximal activities of testosterone 6beta-hydroxylation in CYP3A4/NPR membranes were obtained when the molar ratio of CYP3A4/NPR/b(5) was adjusted to 1:2:1 by mixing membranes containing each protein. Testosterone 6beta-hydroxylation, nifedipine oxidation, and midazolam 4- and 1'-hydroxylation activities in CYP3A4/NPR membranes plus b(5) systems were similar to those measured with microsomes of insect cells coexpressing CYP3A4 with NPR and/or of human liver microsomes, based on equivalent CYP3A4 contents. These results suggest that CYP3A4/NPR membrane systems containing b(5) are very useful models for prediction of the rates for liver microsomal CYP3A4-dependent drug oxidations.

Catalytic properties of polymorphic human cytochrome P450 1B1 variants.

Four polymorphic human cytochrome P450 (CYP) 1B1 allelic variants, namely Arg48,Ala119,Leu432,Asn453, Arg48,Ser119,Leu432,Asn453, Arg48, Ala119,Val432,Asn-453 and Arg48,Ser119,Val432,Asn453, were expressed in Escherichia coli together with human NADPH-P450 reductase and the recombinant proteins (in bacterial membranes) were used to assess whether CYP1B1 polymorphisms affect catalytic activities towards a variety of P450 substrates, including diverse procarcinogens and steroid hormones. Activities for activation of 19 procarcinogens to DNA-damaging products by these four CYP1B1 variants in a Salmonella typhimurium NM2009 umu response system were found to be essentially similar, except that a Arg48, Ser119,Leu432,Asn453 variant was slightly more active (1.2- to 1.5-fold) than the other three CYP1B1 enzymes in catalyzing activation of (+)- and (-)-benzo[a]pyrene-7, 8-diols, 7,12-dimethylbenz[a]anthracene-3,4-diol, benzo[g]chrysene-11,12-diol, benzo[b]fluoranthene-9,10-diol, 2-amino-3,5-dimethylimidazo[4,5-f]quinoline, 2-amino-3-methylimidazo[4,5-f]quinoline and 2-aminofluorene. Kinetic analysis of 17beta-estradiol hydroxylation showed that V(max) values for 4-hydroxylation ranged between 0.9 and 1.5 nmol/min/nmol P450 for 4-hydroxylation and 0.3 and 0.6 nmol/min/nmol P450 for 2-hydroxylation in these CYP1B1 variants, with K(m) values ranging from 1 to 9 microM. Interestingly, the ratio of product formation of 4-hydroxyestradiol to 2-hydroxyestradiol was higher for the Val432 variants of CYP1B1 variants than the Leu432 variants of the enzyme. The same trend was noted in the ratio of estrone 4-hydroxylation to estrone 2-hydroxylation catalyzed by CYP1B1 variants. Mutation in the CYP1B1 genes also affected the K(m) and V(max) values in the 6beta-hydroxylation of testosterone and 6beta- and 16alpha-hydroxylation of progesterone. These results indicate that the polymorphisms in the human CYP1B1 gene cause some alterations in catalytic function towards procarcinogens and steroid hormones and thus may make some contribution to susceptibilities of individuals towards mammary and lung cancers in humans.