Optimized human CYP4B1 in combination with the alkylator prodrug 4-ipomeanol serves as a novel suicide gene system for adoptive T-cell therapies.

Engineering autologous or allogeneic T cells to express a suicide gene can control potential toxicity in adoptive T-cell therapies. We recently reported the development of a novel human suicide gene system that is based on an orphan human cytochrome P450 enzyme, CYP4B1, and the naturally occurring alkylator prodrug 4-ipomeanol. The goal of this study was to systematically develop a clinically applicable self-inactivating lentiviral vector for efficient co-expression of CYP4B1 as an ER-located protein with two distinct types of cell surface proteins, either MACS selection genes for donor lymphocyte infusions after allogeneic stem cell transplantation or chimeric antigen receptors for retargeting primary T cells. The U3 region of the myeloproliferative sarcoma virus in combination with the T2A site was found to drive high-level expression of our CYP4B1 mutant with truncated CD34 or CD271 as MACS suitable selection markers. This lentiviral vector backbone was also well suited for co-expression of CYP4B1 with a codon-optimized CD19 chimeric antigen receptor (CAR) construct. Finally, 4-ipomeanol efficiently induced apoptosis in primary T cells that co-express mutant CYP4B1 and the divergently located MACS selection and CAR genes. In conclusion, we here developed a clinically suited lentiviral vector that supports high-level co-expression of cell surface proteins with a potent novel human suicide gene.

Interindividual variability of CYP2C19-catalyzed drug metabolism due to differences in gene diplotypes and cytochrome P450 oxidoreductase content.

Large interindividual variability has been observed in the metabolism of CYP2C19 substrates in vivo. The study aimed to evaluate sources of this variability in CYP2C19 activity, focusing on CYP2C19 diplotypes and the cytochrome P450 oxidoreductase (POR). CYP2C19 gene analysis was carried out on 347 human liver samples. CYP2C19 activity assayed using human liver microsomes confirmed a significant a priori predicted rank order for (S)-mephenytoin hydroxylase activity of CYP2C19*17/*17 > *1B/*17 > *1B/*1B > *2A/*17 > *1B/*2A > *2A/*2A diplotypes. In a multivariate analysis, the CYP2C19*2A allele and POR protein content were associated with CYP2C19 activity. Further analysis indicated a strong effect of the CYP2C19*2A, but not the *17, allele on both metabolic steps in the conversion of clopidogrel to its active metabolite. The present study demonstrates that interindividual variability in CYP2C19 activity is due to differences in both CYP2C19 protein content associated with gene diplotypes and the POR concentration.The Pharmacogenomics Journal advance online publication, 1 September 2015; doi:10.1038/tpj.2015.58.

Cytochrome P-450--catalyzed formation of delta 4-VPA, a toxic metabolite of valproic acid.

Liver damage induced by the antiepileptic drug valproic acid (VPA) is believed to be mediated by an unsaturated metabolite of the drug, delta 4-VPA. In studies of the biological origin of this hepatotoxic compound, it was found that liver microsomes from phenobarbital-treated rats catalyzed the desaturation of VPA to delta 4-VPA. Indirect evidence suggested that cytochrome P-450 was the responsible enzyme, a conclusion that was verified by studies with a purified and reconstituted form of the hemoprotein, which catalyzed the oxidation of VPA to 4- and 5-hydroxyvalproic acid and to delta 4-VPA. Desaturation of a nonactivated alkyl substituent represents a novel metabolic function of cytochrome P-450 and probably proceeds via the conversion of substrate to a transient free radical intermediate, which partitions between recombination (alcohol formation) and elimination (olefin production) pathways. These findings have broad implications with respect to the metabolic generation of olefins and may explain the increased hepatotoxic potential of VPA when it is administered in combination with potent enzyme-inducing anticonvulsants such as phenobarbital.

Gamma-glutamyl carboxylase (GGCX) tagSNPs have limited utility for predicting warfarin maintenance dose.

BACKGROUND: The pharmacogenetic factors contributing to warfarin dosing are of great interest to clinicians, and may have utility in the management of at-risk patients prescribed warfarin. Gamma-glutamyl carboxylase (GGCX), in its role as a key component of the vitamin K cycle, is a potential candidate gene associated with warfarin treatment. OBJECTIVE: To identify single nucleotide polymorphisms (SNPs) and correlated GGCX tagSNPs and test for association with warfarin maintenance dose. PATIENTS/METHODS: A small discovery population of European-descent individuals (n = 23) were resequenced for GGCX SNPs. Polymorphisms identified with > 5% minor allele frequency (MAF) were genotyped in a larger clinical population of 186 European patients. Univariate, multivariate and haplotype-based linear regression were used to assess the impact of GGCX SNPs on warfarin dose. RESULTS: We identified 37 SNPs in GGCX, of which 21 were present at > 5% MAF. These SNPs were binned, based on linkage disequilibrium, and six informative tagSNPs were identified. A single polymorphism at position 12970 (rs11676382; C/G-11%/89%) was associated with a warfarin maintenance dose across all analysis methods. GGCX-12970 explained 2% of the total variance in warfarin dose, in contrast to 21 and 8%, respectively, for VKORC1 and CYP2C9. CONCLUSIONS: The GGCX-12970 SNP had a small, but significant effect, on warfarin maintenance dose. Other polymorphisms in GGCX previously associated with warfarin dose were not confirmed in this study, suggesting that the effects of GGCX are potentially population/treatment-dependent and will not have broad utility for determining warfarin dosing.

Purification of macaque liver flavin-containing monooxygenase: a form of the enzyme related immunochemically to an isozyme expressed selectively in adult human liver.

A microsomal flavin-containing monooxygenase (FMO) was purified 77-fold from macacque liver microsomes on the basis of its methyl p-tolyl sulfoxidase activity. Sequential chromatography on anion- and cation-exchangers, lauryl-Sepharose and 2',5'-ADP-Sepharose provided a purified preparation which exhibited an apparent molecular mass of 59 kDa and a pI of 8.3. N-terminal amino-acid sequencing revealed the characteristic Gly-X-Gly-X-X-Gly consensus sequence for the putative FAD-binding domain of microsomal FMO. In marked contrast to the well-characterized hepatic and pulmonary forms present in experimental animals, the macacque liver enzyme displayed stereoselectivity for sulfoxidation of p-tolyl methyl sulfide on the pro-S rather than the pro-R face of the substrate. Polyclonal antibodies raised against the macacque liver form exhibited little or no cross-reactivity with major purified forms of the enzyme isolated from rabbit liver, guinea-pig liver or rabbit lung. Anti-macacque liver FMO did not cross-react with human fetal liver or adult kidney microsomes, but did recognize a 59 kDa constituent of human adult liver microsomes. The intensity of this immunoreactive 59 kDa band correlated well with human liver microsomal N,N-dimethylaniline N-oxygenase activity. We conclude that human adult liver selectively expresses a microsomal FMO which is functionally and immunochemically distinct from the FMO form(s) present in human fetal liver and adult kidney, and from the major hepatic and pulmonary forms present in common laboratory animals.

Human variability in hepatic glutathione S-transferase-mediated conjugation of aflatoxin B1-epoxide and other substrates.

Hepatic cytosolic fractions prepared from 14 human donors were analysed for glutathione S-transferase (GST) activity towards synthetic aflatoxin B1-8,9-epoxide (AFBO). In addition, GST-AFBO activity of pooled human liver cytosols was compared with rat, hamster, and mouse liver cytosol GST-AFBO activities. Consistent with previous studies, human liver cytosolic GSTs exhibited little activity towards AFBO. Hepatic GST-AFBO activities of rat, hamster, and mouse were 48-, 56-, and 312-fold greater, respectively, than observed for human liver using synthetic AFBO, and 70-, 465-, and 3545-fold greater, respectively, than observed for human liver using microsomally-generated AFBO. Furthermore, there was a 58-fold variation in hepatic GST-AFBO activities among the 14 human samples using synthetic AFBO as a substrate. Large interindividual variations were also observed with respect to GST activities towards bromosulfophthalein (BSP, 92-fold variation) and 3,4-dichloronitrobenzene (DCNB, 36-fold variation). Lesser interindividual variations were observed with respect to human liver GST activities towards benzo(a)pyrene-4,5-oxide (BaPO, 9-fold variation), 1-chloro-2,4-dinitrobenzene (CDNB, 8.5-fold variation), cumene hydroperoxide (CHP, 5-fold variation), and p-nitrophenyl acetate (NPA, 4-fold variation). No correlation was found among GST-AFBO activities and the presence of GST mu as determined by enzyme-linked immunosorbent assay (ELISA) or GST-trans-stilbene oxide (TSO) catalytic activity. Our observations support those of previous studies indicating that human liver cytosolic GSTs are relatively ineffective at conjugating AFBO. Furthermore, our data indicate that humans exhibit large inter-individual differences with respect to hepatic cytosolic GST conjugation of AFBO and certain other GST substrates.

Genetic association between sensitivity to warfarin and expression of CYP2C9*3.

Cytochrome P4502C9 (CYP2C9) is largely responsible for terminating the anticoagulant effect of racemic warfarin via hydroxylation of the pharmacologically more potent S-enantiomer to inactive metabolites. Mutations in the CYP2C9 gene result in the expression of three allelic variants, CYP2C9*1, CYP2C9*2 and CYP2C9*3. Both CYP2C9*2 and CYP2C9*3 exhibit altered catalytic properties in vitro relative to the wild-type enzyme. In the present study, a patient was genotyped who had proven unusually sensitive to warfarin therapy and could tolerate no more than 0.5 mg of the racemic drug/day. PCR-amplification of exons 3 and 7 of the CYP2C9 gene, followed by restriction digest or sequence analysis, showed that this individual was homozygous for CYP2C9*3. In addition, patient plasma warfarin enantiomer ratios and urinary 7-hydroxywarfarin enantiomer ratios were determined by chiral-phase high performance liquid chromotography in order to investigate whether either parameter might be of diagnostic value in place of a genotypic test. Control patients receiving 4-8 mg warfarin/day exhibited plasma S:R ratios of 0.50 +/- 0.25:1, whereas the patient on very low-dose warfarin exhibited an S:R ratio of 3.9:1. In contrast, the urinary 7-hydroxywarfarin S:R ratio of 4:1 showed the same stereoselectivity as that reported for control patients. Therefore, expression of CYP2C9*3 is associated with diminished clearance of S-warfarin and a dangerously exacerbated therapeutic response to normal doses of the racemic drug. Analysis of the plasma S:R warfarin ratio may serve as a useful alternative test to genotyping for this genetic defect.

Valproate hydroxylation by human fetal tissues and embryotoxicity of metabolites.

The oxidative biotransformation of sodium valproate was studied in liver, lung, brain, and adrenal homogenates from human conceptuses with gestational ages ranging from 50 to 77 days. Analyses of metabolites by GC/MS indicated the formation of 3-hydroxy-, 4-hydroxy-, and 5-hydroxyvalproic acid, with hydroxylation occurring preferentially at the 4- position. The adrenal homogenate was consistently the most active fetal tissue studied, with rates of hydroxylation similar to those in rat and macaque liver homogenates. Reaction rates in the fetal adrenal homogenate were approximately four times those in fetal liver and approximately 10 times the rates of the same reactions measured in fetal brain and lung. Although valproic acid itself (0.8 mmol/L) was highly embryotoxic to cultured whole rat embryos, none of the hydroxylated metabolites produced by human fetal tissues exhibited significant embryotoxicity at equimolar concentrations. This suggests that hydroxylation of valproic acid in human fetal tissues is a process of detoxification, and implies that valproic acid is a direct-acting teratogen.

Interaction of amiodarone with racemic warfarin and its separated enantiomorphs in humans.

To evaluate a stereoselective interaction for amiodarone and racemic warfarin, we performed a prospective study with its separated enantiomorphs. Single oral doses of racemic warfarin, 1.5 mg/kg, were administered to six normal subjects, with and without oral amiodarone, 400 mg daily, for the hypoprothrombinemic duration. Both the hypoprothrombinemia (P less than 0.001) and plasma warfarin concentrations (P less than 0.01) were significantly increased. The experiments were repeated separately with the R- and S-warfarin enantiomorphs. S-warfarin with amiodarone significantly increased both the hypoprothrombinemia (P less than 0.001) and plasma warfarin concentrations (P less than 0.01). R-warfarin with amiodarone significantly increased both the hypoprothrombinemia (P less than 0.001) and plasma warfarin concentrations (P less than 0.001). Thus amiodarone augmented the anticoagulant effect nonstereoselectively by reduced metabolic clearance of both warfarin enantiomorphs. Amiodarone and racemic warfarin can be a dangerous combination, particularly when either drug is added to a stabilized regimen of the other drug, unless the prothrombin times are monitored carefully.

Inhibition of human liver microsomal epoxide hydrolase by valproate and valpromide: in vitro/in vivo correlation.

On the basis of drug interactions with carbamazepine epoxide, it has been hypothesized that valproic acid and valpromide are inhibitors of epoxide hydrolase, but the role of epoxide hydrolase in these interactions has not been clearly established. In this study, therapeutic concentrations of valproic acid (less than 1 mmol/L) and valpromide (less than 10 mumol/L) inhibited hydrolysis of carbamazepine epoxide and styrene oxide in human liver microsomes and in preparations of purified human liver microsomal epoxide hydrolase. Valpromide (KI = 5 mumol/L) was 100 times more potent than valproic acid (KI = 550 mumol/L) as an inhibitor of carbamazepine epoxide hydrolysis in microsomes. After administration of carbamazepine epoxide to volunteers, the transdihydrodiol formation clearance was decreased 20% by valproic acid (blood concentration approximately 113 mumol/L) and 67% by valpromide (blood concentration less than 10 mumol/L). For both valproic acid and valpromide, a striking similarity exists between in vitro and in vivo inhibitory potencies. Valproic acid and valpromide are the first drugs known to inhibit microsomal epoxide hydrolase, an important detoxification enzyme, at therapeutic concentrations.

Intrinsic isotope effects suggest that the reaction coordinate symmetry for the cytochrome P-450 catalyzed hydroxylation of octane is isozyme independent.

The mechanism of the omega-hydroxylation of octane by three catalytically distinct, purified forms of cytochrome P-450, namely, P-450b, P-450c, and P-450LM2, was investigated by using deuterium isotope effects. The deuterium isotope effects associated with the omega-hydroxylation of octane-1,1,1-2H3, octane-1,8-2H2, and octane-1,1,8,8-2H4 by all three isozymes were determined. From these data the intrinsic isotope effects were calculated and separated into their primary and secondary components. The primary intrinsic isotope effect for the reaction ranged from 7.69 to 9.18 while the secondary intrinsic isotope effect ranged from 1.13 to 1.25. Neither the primary nor secondary isotope effect values were statistically different for any of the isozymes investigated. These data are consistent with a symmetrical transition state for a mechanism involving initial hydrogen atom abstraction followed by hydroxyl radical recombination which is essentially independent of the specific isozyme catalyzing the reaction. It is concluded that (1) in general the porphyrin-[FeO]3+ complex behaves as a source of a triplet-like oxygen atom, (2) the regioselectivity for the site of oxidation is dictated by the apoprotein of the specific isozyme of cytochrome P-450 catalyzing the reaction, and (3) the maximum primary intrinsic isotope effect for any cytochrome P-450 catalyzed oxidation of a carbon center is about 9, assuming no tunneling effects.

A refined 3-dimensional QSAR of cytochrome P450 2C9: computational predictions of drug interactions.

A ligand-based model is reported that predicts the Ki values for cytochrome P450 2C9 (CYP2C9) inhibitors. This CoMFA model was used to predict the affinity of 14 structurally diverse compounds not in the training set and appears to be robust. The mean error of the predictions is 6 microM. The experimentally measured Ki values of the 14 compounds range from 0.1 to 48 microM. Leave-one-out cross-validated partial least-squares gives a q2 value of between 0.6 and 0.8 for the various models which indicates internal consistency. Random assignment of biological data to structure leads to negative q2 values. These models are useful in that they establish a pharmacophore for binding to CYP2C9 that can be tested with site-directed mutagenesis. These models can also be used to screen for potential drug interactions and to design compounds that will not bind to this enzyme with high affinity.

Major differences between lung, skin and liver in the microsomal metabolism of homologous series of resorufin and coumarin ethers.

Phenoxazone and a homologous series of its ethers (methoxy to octoxy plus benzyloxy), and coumarin and a series of its ethers (methoxy to propoxy), were metabolized by liver, lung and skin microsomes of normal adult female BALB/c mice. For each series of substrates, and with each tissue, clear structure-activity relationships were seen, relating metabolic activity to the length of the ether side-chain. With the coumarin series of substrates the structure-activity relationships were almost identical in the three tissues, with liver more active than lung and lung more active than skin. Liver, lung and skin microsomes each showed very different structure-activity relationships, however, for metabolism of the phenoxazone series of substrates. Benzyloxyphenoxazone was metabolized almost twice as fast in lung as in liver, but for the other phenoxazone substrates the activities were much greater in liver than in lung or skin. Liver, lung and skin microsomal propoxy- and benzyloxyphenoxazone dealkylase activities differed in their sensitivities to inhibition by metyrapone and alpha-naphthoflavone. The structure-activity relationship and inhibitor data for the phenoxazone substrates are consistent with a view that mouse lung and sking cyt. P-450 are predominantly similar to phenobarbitone-induced and 3-methylcholanthrene-induced forms of hepatic cyt. P-450 respectively. The results also show that the pattern of microsomal metabolism of xenobiotics in lung and skin cannot be reliably predicted from that in liver.

Stereoselective sulfoxidation of sulindac sulfide by flavin-containing monooxygenases. Comparison of human liver and kidney microsomes and mammalian enzymes.

The stereoselective sulfoxidation of the pharmacologically active metabolite of sulindac, sulindac sulfide, was characterized in human liver, kidney, and cDNA-expressed enzymes. Kinetic parameter estimates (pH = 7.4) for sulindac sulfoxide formation in human liver microsomes (N = 4) for R- and S-sulindac sulfoxide were V(max) = 1.5 +/- 0.50 nmol/min/mg, K(m) = 15 +/- 5.1 microM; and V(max) = 1.1 +/- 0.36 nmol/min/mg, K(m) = 16 +/- 6.1 microM, respectively. Kidney microsomes (N = 3) produced parameter estimates (pH = 7.4) of V(max) = 0.9 +/- 0.29 nmol/min/mg, K(m) = 15 +/- 2.9 microM; V(max) = 0.5 +/- 0.21 nmol/min/mg, K(m) = 22 +/- 1.9 microM for R- and S-sulindac sulfoxide, respectively. In human liver and flavin-containing monooxygenase 3 (FMO3) the V(max) for R-sulindac sulfoxide increased 60-70% at pH = 8.5, but for S-sulindac sulfoxide was unchanged. In fourteen liver microsomal preparations, significant correlations occurred between R-sulindac sulfoxide formation and either immunoquantified FMO or nicotine N-oxidation (r = 0.88 and 0.83; P < 0.01). The R- and S-sulindac sulfoxide formation rate also correlated significantly (r = 0.85 and 0.75; P < 0.01) with immunoquantified FMO in thirteen kidney microsomal samples. Mild heat deactivation of microsomes reduced activity by 30-60%, and a loss in stereoselectivity was observed. Methimazole was a potent and nonstereoselective inhibitor of sulfoxidation in liver and kidney microsomes. n-Octylamine and membrane solubilization with lubrol were potent and selective inhibitors of S-sulindac sulfoxide formation. cDNA-expressed CYPs failed to appreciably sulfoxidate sulindac sulfide, and CYP inhibitors were ineffective in suppressing catalytic activity. Purified mini-pig liver FMO1, rabbit lung FMO2, and human cDNA-expressed FMO3 efficiently oxidized sulindac sulfide with a high degree of stereoselectivity towards the R-isomer, but FMO5 lacked catalytic activity. The biotransformation of the sulfide to the sulfoxide is catalyzed predominately by FMOs and may prove to be useful in characterizing FMO activity.

Isoform specificity of trimethylamine N-oxygenation by human flavin-containing monooxygenase (FMO) and P450 enzymes: selective catalysis by FMO3.

In the present study, we expressed human flavin-containing monooxygenase 1 (FMO1), FMO3, FMO4t (truncated), and FMO5 in the baculovirus expression vector system at levels of 0.6 to 2.4 nmol FMO/mg of membrane protein. These four isoforms, as well as purified rabbit FMO2, and eleven heterologously expressed human P450 isoforms were examined for their capacity to metabolize trimethylamine (TMA) to its N-oxide (TMAO), using a new, specific HPLC method with radiochemical detection. Human FMO3 was by far the most active isoform, exhibiting a turnover number of 30 nmol TMAO/nmol FMO3/min at pH 7.4 and 0.5 mM TMA. None of the other monooxygenases formed TMAO at rates greater than 1 nmol/nmol FMO/min under these conditions. Human fetal liver, adult liver, kidney and intestine microsomes were screened for TMA oxidation, and only human adult liver microsomes provided substantial TMAO-formation (range 2.9 to 9.1 nmol TMAO/mg protein/min, N = 5). Kinetic studies of TMAO formation by recombinant human FMO3, employing three different analytical methods, resulted in a Km of 28 +/- 1 microM and a Vmax of 36.3 +/- 5.7 nmol TMAO/nmol FMO3/min. The Km determined in human liver microsomes ranged from 13.0 to 54.8 microM. Therefore, at physiological pH, human FMO3 is a very specific and efficient TMA N-oxygenase, and is likely responsible for the metabolic clearance of TMA in vivo in humans. In addition, this specificity provides a good in vitro probe for the determination of FMO3-mediated activity in human tissues, by analyzing TMAO formation at pH 7.4 with TMA concentrations not higher than 0.5 mM.

Effects of oral vitamin K on S- and R-warfarin pharmacokinetics and pharmacodynamics: enhanced safety of warfarin as a CYP2C9 probe.

Evidence for the selectivity of S-warfarin metabolism by CYP2C9 is substantial, suggesting that warfarin may be a potential CYP2C9 phenotyping probe. It is, however, limited by its ability to elevate the international normalized ratio (INR) and potentially cause bleeding. The effect of vitamin K to attenuate the elevation of INR may enable the safe use of warfarin as a probe. The objective of this study was to investigate the pharmacokinetics and pharmacodynamics of S- and R-warfarin in plasma following the administration of warfarin alone versus warfarin and vitamin K in CYP2C9*1 homozygotes. Healthy adults received, in a randomized crossover fashion in a fasted state, warfarin 10 mg orally or warfarin 10 mg plus vitamin K 10 mg orally. Blood samples were obtained over 5 days during each phase. INR measurements were obtained at baseline and day 2 in each phase. INR, AUC0-infinity, and t1/2 of plasma S- and R-warfarin were examined. Eleven CYP2C9*1 homozygotes (3 men, 8 women) were enrolled. INR at day 2 following warfarin 10 mg was 1.18 +/- 0.19, which differed significantly from baseline (INR = 1.00 +/- 0.05) and warfarin with vitamin K (INR = 1.06 +/- 0.07). INR at baseline was not significantly different from warfarin with vitamin K. t1/2 and AUC0-infinity of both enantiomers did not significantly differ between the phases. It was concluded that INR is apparently attenuated by concomitant administration of a single dose of vitamin K without affecting the pharmacokinetics of either warfarin stereoisomer. Warfarin 10 mg may be safely used as a CYP2C9 probe in *1 homozygotes when given concomitantly with 10 mg of oral vitamin K.

A common genetic basis for idiosyncratic toxicity of warfarin and phenytoin.

CYP2C9 is mainly responsible for the metabolic clearance of phenytoin and (S)-warfarin. We have shown previously that mutations in the CYP2C9 gene are associated with diminished metabolism of (S)-warfarin, and so we have now studied the metabolism of phenytoin to its primary inactive metabolite, (S)-pHPPH, by these mutant enzymes. Kinetic parameters were determined for (S)-pHPPH formation using recombinant CYP2C9 variants purified from insect cells. The data demonstrate that the CYP2C9*3 gene product retains only 4-6% of the metabolic efficiency of the wild-type protein, CYP2C9*1, towards phenytoin and (S)-warfarin. Consequently, we suggest that homozygous expression of CYP2C9*3 may represent a common genetic basis for (apparently) idiosyncratic toxicities that have been reported for these two low therapeutic index drugs.

Prochiral sulfides as in vitro probes for multiple forms of the flavin-containing monooxygenase.

A homologous series of alkyl-substituted p-tolyl sulfides have been synthesized and evaluated as in vitro, isozyme-selective substrate probes for the microsomal flavin containing monooxygenases. Straight-chain and branched-chain alkyl homologs were metabolized to the corresponding (R)- and (S)-sulfoxides which were analyzed by chiral phase high-performance liquid chromatography. Initial studies demonstrated that the stereochemical composition of alkyl p-tolyl sulfoxides generated by FMO2, purified from rabbit lung, was a function of the degree of steric crowding about the prochiral center. In contrast, purified rabbit liver FMO1 formed the (R)-sulfoxide from the n-alkyl series of substrates in a highly stereoselective manner (> 90%). Similar results were obtained with these two rabbit cDNAs expressed in E. coli. In contrast to rabbit FMO1 and FMO2, a characteristic feature of catalysis by cDNA-expressed rabbit FMO3 was the lack of stereoselectivity observed for formation of methyl p-tolyl sulfoxide. Collectively, these data demonstrate that the stereochemical composition of sulfoxides generated from the n-alkyl series of sulfides is isozyme-dependent. Metabolism of methyl p-tolyl sulfide by detergent-solubilized hepatic microsomes from a wide variety of experimental animals yielded predominantly (R)- methyl p-tolyl sulfoxide, which, at least in rabbit liver, is indicative of catalysis dominated by FMO1. However, solubilized human and macaque liver preparations catalyzed this reaction in a relatively non-stereoselective manner. Macaque liver FMO was purified and the metabolite profile generated from the n-alkyl p-tolyl sulfides was found to be most similar to rabbit FMO3. Moreover, antibodies directed against macaque liver FMO selectively reacted with rabbit FMO3 and a microsomal protein expressed in adult human, but not fetal human liver, adult human kidney or adult human lung. Therefore, an FMO isoform expressed selectively in adult primate liver has catalytic and immunochemical properties consistent with its classification in the FMO3 family.

Cytosolic activation of hematin-dependent microsomal monooxygenase activity in the lung.

Additions of micromolar concentrations of hematin to washed rat pulmonary microsomal preparations resulted in marked (5-7-fold) increases in the NADPH-dependent generation of phenolic metabolites of benzo[a]pyrene (BaP). 9-Hydroxy-BaP was identified as the major reaction product. Additions of pulmonary cytosolic fractions to microsomes produced no measurable effect but cytosol and hematin added together elicited 25-30-fold increases in total phenolic products. Cytosolic fractions from other tissues, including rat kidneys and perfused rat livers, were also highly effective in enhancing the hematin-mediated increases in monooxygenase activity. However, cytosol from human placental tissues was only minimally effective when either pulmonary or placental microsomes were utilized as enzyme source. Superoxide dismutase and catalase (alone or in combination) had no measurable effect on hematin-mediated increases. Horseradish peroxidase effectively inhibited the hematin-dependent reactions but hematin-independent reactions were inhibited with equal effectiveness. Carbon monoxide profoundly inhibited all hematin-mediated increases in metabolite formation. The activating cytosolic component was non-dialyzable, inactivated by trypsin and heat, and eluted in the void volume from Sephadex G-150 columns. This suggested that the cytosolic factor(s) responsible for the increased hematin-dependent oxidation was a protein(s) with a high molecular weight or perhaps an aggregate or oligomer of proteinaceous material. HPLC profiles indicated a major effect on the generation of phenolics; quinones were also increased but only minimal increases in diols were observed. Results were consistent with the hypothesis that hematin-mediated increases in pulmonary monooxygenase activity result from an increased association of a small pool of pulmonary P-450-apoprotein(s) with the hematin prosthetic group to result in increased levels of an unidentified holocytochrome(s) with a relatively high substrate turnover number. The current data suggest a quaternary interaction among P-450 apoprotein(s), heme prosthetic group, reaction products (particularly 3-hydroxy-BaP) and a cytosolic protein(s). We postulate that the mechanism of action of the cytosolic factor is to facilitate the interaction of hematin with the apocytochrome.