Human CYP2D6 and mouse CYP2Ds: organ distribution in a humanized mouse model.

Polymorphic cytochrome P450 (P450) 2D6 (CYP2D6) metabolizes several classes of therapeutic drugs, endogenous neurochemicals, and toxins. A CYP2D6-humanized transgenic mouse line was previously developed to model CYP2D6-poor and -extensive metabolizer phenotypes. Human CYP2D6 was detected in the liver, kidney, and intestine of these animals. In this study, we investigated further the cellular expression and relative tissue levels of human CYP2D6 in these transgenic mice in liver, intestine, kidney, and brain. In addition, we compared this with the expression of mouse CYP2D enzymes in these organs. In humans, these organs are of interest with respect to P450-mediated drug metabolism, toxicity, and disease. The expression of human CYP2D6 and mouse CYP2D enzymes in humanized and wild-type mice was quantified by immunoblotting and detected at the cellular level by immunocytochemistry. The cell-specific expression of human CYP2D6 in liver, kidney, and intestine in humanized mice was similar to that reported in humans. The expression patterns of mouse CYP2D proteins were similar to those in humans in liver and kidney but substantially different in intestine. Human CYP2D6 was not detected in brain of transgenic mice. Mouse CYP2D proteins were detected in brain, allowing, for the first time, a direct comparison of CYP2D expression among mouse, rat, and human brain. This transgenic mouse model is useful for investigating CYP2D6-mediated metabolism in liver, kidney, and especially the intestine, where expression patterns demonstrated substantial species differences.     

A transgenic mouse expressing human CYP1A2 in the pancreas

A transgenic mouse line expressing the human cytochrome P450 CYP1A2 in the pancreas under the control of the mouse elastase promoter was established. The expression of CYP1A2 was specific to the transgenic pancreas and was not found in the control wild-type mouse pancreas. The level of CYP1A2 expressed in pancreatic microsomes from transgenic mice was comparable to that of the endogenously expressed CYP1A2 protein in the liver, as judged by western blotting analyses. Estrone metabolism was used to determine the activity of CYP1A2 expressed in the pancreas of the transgenic mouse. The transgenic pancreas exhibited almost one-third to one-half of the activity of wild-type or CYP1A2 transgenic mouse liver, whereas the wild-type pancreas demonstrated no activity. The addition of NADPH-cytochrome P450 oxidoreductase to the reaction mixture containing pancreatic microsomes from the transgenic mice did not increase the estrone metabolism activity significantly. This transgenic mouse line provides another useful tool to study human CYP1A2 and its relation to chemical toxicity and carcinogenesis.     

A CYP2B6-humanized mouse model and its potential applications

CYP2B6 is a human microsomal cytochrome P450 enzyme with broad substrate selectivity. CYP2B6 is the only functional member of the human CYP2B gene subfamily, which differs from the situation in rodents, such as mouse, where multiple functional Cyp2b genes are expressed. Recent studies with Cyp2b knockout or knockdown mouse models have yielded insights into the in vivo roles of mouse CYP2B enzymes in drug disposition and xenobiotic toxicity. A CYP2B6-humanized mouse model (CYP2A13/2B6/2F1-transgenic/Cyp2abfgs-null), which expresses human CYP2B6 in the liver, and human CYP2A13 and CYP2F1 in the respiratory tract, but not any of the mouse Cyp2b genes, has also been established. In the CYP2B6-humanized mouse, the CYP2B6 transgene is expressed primarily in the liver, where it was found to be active toward prototype CYP2B6 substrate drugs. The regulatory elements of the CYP2B6 transgene appear to be compatible with mouse nuclear receptors that mediate CYP2B induction. Therefore, the CYP2B6-humanized mouse is a valuable animal model for studying the impact of CYP2B6 expression or induction on drug metabolism, drug efficacy, drug-drug interaction, and drug/xenobiotic toxicity. In this mini-review, we provide a brief background on CYP2B6 and the Cyp2b-knockout and CYP2B6-humanized mice, and discuss the potential applications and limitations of the current models.     

CYP1A1 and CYP1A2 expression: Comparing ‘humanized’ mouse lines and wild-type mice; comparing human and mouse hepatoma-derived cell lines

Human and rodent cytochrome P450 (CYP) enzymes sometimes exhibit striking species-specific differences in substrate preference and rate of metabolism. Human risk assessment of CYP substrates might therefore best be evaluated in the intact mouse by replacing mouse Cyp genes with human CYP orthologs; however, how “human-like” can human gene expression be expected in mouse tissues? Previously a bacterial-artificial-chromosome-transgenic mouse, carrying the human CYP1A1_CYP1A2 locus and lacking the mouse Cyp1a1 and Cyp1a2 orthologs, was shown to express robustly human dioxin-inducible CYP1A1 and basal versus inducible CYP1A2 (mRNAs, proteins, enzyme activities) in each of nine mouse tissues examined. Chimeric mice carrying humanized liver have also been generated, by transplanting human hepatocytes into a urokinase-type plasminogen activator(+/+)_severe-combined-immunodeficiency (uPA/SCID) line with most of its mouse hepatocytes ablated. Herein we compare basal and dioxin-induced CYP1A mRNA copy numbers, protein levels, and four enzymes (benzo[a]pyrene hydroxylase, ethoxyresorufin O-deethylase, acetanilide 4-hydroxylase, methoxyresorufin O-demethylase) in liver of these two humanized mouse lines versus wild-type mice; we also compare these same parameters in mouse Hepa-1c1c7 and human HepG2 hepatoma-derived established cell lines. Most strikingly, mouse liver CYP1A1-specific enzyme activities are between 38- and 170-fold higher than human CYP1A1-specific enzyme activities (per unit of mRNA), whereas mouse versus human CYP1A2 enzyme activities (per unit of mRNA) are within 2.5-fold of one another. Moreover, both the mouse and human hepatoma cell lines exhibit striking differences in CYP1A mRNA levels and enzyme activities. These findings are relevant to risk assessment involving human CYP1A1 and CYP1A2 substrates, when administered to mice as environmental toxicants or drugs.     

Inhibition of CYP2E1 catalytic activity in vitro by S-adenosyl-L-methionine

The objective of this work was to evaluate the possible in vitro interactions of S-adenosyl-l-methionine (SAM) and its metabolites S-(5'-Adenosyl)-l-homocysteine (SAH), 5'-Deoxy-5'-(methylthio)adenosine (MTA) and methionine with cytochrome P450 enzymes, in particular CYP2E1. SAM (but not SAH, MTA or methionine) produced a type II binding spectrum with liver microsomal cytochrome P450 from rats treated with acetone or isoniazid to induce CYP2E1. Binding was less effective for control microsomes. SAM did not alter the carbon monoxide binding spectrum of P450, nor denature P450 to P420, nor inhibit the activity of NADPH-P450 reductase. However, SAM inhibited the catalytic activity of CYP2E1 with typical substrates such as p-nitrophenol, ethanol, and dimethylnitrosamine, with an IC(50) around 1.5-5mM. SAM was a non-competitive inhibitor of CYP2E1 catalytic activity and its inhibitory actions could not be mimicked by methionine, SAH or MTA. However, SAM did not inhibit the oxidation of ethanol to alpha-hydroxyethyl radical, an assay for hydroxyl radical generation. In microsomes engineered to express individual human P450s, SAM produced a type II binding spectrum with CYP2E1-, but not with CYP3A4-expressing microsomes, and SAM was a weaker inhibitor against the metabolism of a specific CYP3A4 substrate than a specific CYP2E1 substrate. SAM also inhibited CYP2E1 catalytic activity in intact HepG2 cells engineered to express CYP2E1. These results suggest that SAM interacts with cytochrome P450s, especially CYP2E1, and inhibits the catalytic activity of CYP2E1 in a reversible and non competitive manner. However, SAM is a weak inhibitor of CYP2E1. Since the K(i) for SAM inhibition of CYP2E1 activity is relatively high, inhibition of CYP2E1 activity is not likely to play a major role in the ability of SAM to protect against the hepatotoxicity produced by toxins requiring metabolic activation by CYP2E1 such as acetaminophen, ethanol, carbon tetrachloride, thioacetamide and carcinogens.     

Inhibition of urethane-induced carcinogenicity in cyp2e1-/- in comparison to cyp2e1+/+ mice.

Urethane is an established animal carcinogen and has been classified as "reasonably anticipated to be a human carcinogen." Until recently, urethane metabolism via esterase was considered the main metabolic pathway of this chemical. However, recent studies in this laboratory showed that CYP2E1, and not esterase, is the primary enzyme responsible for urethane oxidation. Subsequent studies demonstrated significant inhibition of urethane-induced genotoxicity and cell proliferation in Cyp2e1-/- compared to Cyp2e1+/+ mice. Using Cyp2e1-/- mice, current studies were undertaken to assess the relationships between urethane metabolism and carcinogenicity. Urethane was administered via gavage at 1, 10, or 100 mg/kg/day, 5 days/week, for 6 weeks. Animals were kept without chemical administration for 7 months after which they were euthanized, and urethane carcinogenicity was assessed. Microscopic examination showed a significant reduction in the incidences of liver hemangiomas and hemangiosarcomas in Cyp2e1-/- compared to Cyp2e+/+ mice. Lung nodules increased in a dose-dependent manner and were less prevalent in Cyp2e1-/- compared to Cyp2e+/+ mice. Microscopic alterations included bronchoalveolar adenomas, and in one Cyp2e1+/+ mouse treated with 100 mg/kg urethane, a bronchoalveolar carcinoma was diagnosed. Significant reduction in the incidence of adenomas and the number of adenomas/lung were observed in Cyp2e1-/- compared to Cyp2e1+/+ mice. In the Harderian gland, the incidences of hyperplasia and adenomas were significantly lower in Cyp2e1-/- compared to Cyp2e+/+ mice at the 10 mg/kg dose, with no significant differences observed at the high or low doses. In conclusion, this work demonstrated a significant reduction of urethane-induced carcinogenicity in Cyp2e1-/- compared to Cyp2e1+/+ mice and proved that CYP2E1-mediated oxidation plays an essential role in urethane-induced carcinogenicity.     

In vivo induction of human cytochrome P450 enzymes expressed in chimeric mice with humanized liver.

The induction and inhibition of human cytochrome P450 (P450) enzymes are clinically responsible for drug interactions. Although the induction of P450s is investigated using human hepatocytes in the drug development process, there are some disadvantages, such as the decline of the enzyme activity during culture. In the present study, we examined the in vivo induction potency in chimeric mice with humanized liver, which was recently established in Japan to clarify whether this chimeric mouse model would be more suitable for human induction studies. Rifampicin and 3-methylcholanthrene (3-MC) were used in vivo as typical P450 inducers in the chimeric mice. The expression levels of human CYP3A4 mRNA and CYP3A4 protein and dexamethasone 6-hydroxylase activity, specific for human CYP3A4, were increased 8- to 22-, 3- to 10-, and 5- to 12-fold, respectively, by treatment with rifampicin. In addition, the expression levels of human CYP1A2 mRNA and CYP1A2 protein were also increased 2- to 9- and 5-fold, respectively, by treatment with 3-MC. Although other human P450s are expressed in the chimeric mice, there were few effects by the treatment of rifampicin and 3-MC on the mRNA, protein, and enzyme activity of those P450s. It was demonstrated that human P450s expressed in the chimeric mice with humanized liver were induced by rifampicin and 3-MC. This chimeric mouse model may be a useful animal model to estimate and predict the in vivo induction of P450s in humans.     

Metabolic oxidation and toxification of N-methylformamide catalyzed by the cytochrome P450 isoenzyme CYP2E1.

Alkylformamides, for example N-methylformamide, are hepatotoxic in rodents and humans. The mechanism by which N-methylformamide exerts its hepatotoxicity involves metabolic oxidation at the formyl moiety to yield a short-lived intermediate, perhaps methyl isocyanate, which reacts with glutathione to afford S-(N-methylcarbamoyl)glutathione. The hypothesis that the cytochrome P450 isozyme CYP2E1 catalyzes the metabolic toxification of N-methylformamide was tested. Hepatocytes obtained from mice that had received acetone, an inducer of CYP2E1, were incubated for up to 4 hr with N-methylformamide (5 and 10 mM). Whereas N-methylformamide caused cytotoxicity in these cells, as measured by release from the cells of lactate dehydrogenase, it was barely toxic, under these conditions, to cells from untreated mice. Coincubation of N-methylformamide with dimethylsulfoxide (10 mM), a CYP2E1 inhibitor, for 4 or 6 hr abolished the hepatocytotoxicity of N-methylformamide. Metabolism of N-methylformamide to S-(N-methylcarbamoyl) glutathione was measured in incubates with liver microsomes from rats, mice, or humans in the presence of glutathione. Pretreatment of rodents with acetone or ethanol induced the rate of metabolism of N-methylformamide and of p-nitrophenol, a known CYP2E1 substrate, but it did not increase aminopyrine N-demethylation. Metabolism of N-methylformamide and p-nitrophenol was elevated in microsomes from animals that had received acetone (1%) in their drinking water for 1 week to 230% and 200%, respectively, of control values in mouse microsomes and to 310% and 240%, respectively, of control values in rat microsomes. Pretreatment of animals with 4-methylpyrazole (200 mg/kg intraperitoneally, once daily for 3 days) increased metabolism of N-methylformamide to 410% of control values in rat liver microsomes but was without effect on murine microsomal metabolism of N-methylformamide. The metabolism of this compound was strongly inhibited by the CYP2E1 substrates or inhibitors dimethylsulfoxide (1-100 mM), p-nitrophenol (100 microM), and diethyldithiocarbamate (100 microM), which did not affect aminopyrine N-demethylation. A polyclonal antibody against rat CYP2E1 (10 mg of IgG/nmol of cytochrome P450) inhibited N-methylformamide metabolism in liver microsomes from rats and from a human by 75% and 80%, respectively. The rate of metabolism of N-methylformamide to S-(N-methylcarbamoyl) glutathione was determined in liver microsomes from six humans and correlated with extent of metabolic hydroxylation of chlorzoxazone, a CYP2E1 probe, and with amount of immunodetectable enzyme using an anti-rat CYP2E1 antibody (r = 0.81 and 0.80, respectively). The results suggest that CYP2E1 is the predominant, if not sole, cytochrome P450 isozyme responsible for the metabolic toxification of hepatotoxic N-alkylformamides.     

An improved TK-NOG mouse as a novel platform for humanized liver that overcomes limitations in both male and female animals

We developed a novel immunodeficient NOG mouse expressing HSVtk mutant clone 30 cDNA under the control of mouse transthyretin gene enhancer/promoter (NOG-TKm30) to acquire fertility in males and high inducibility of liver injury in females. Maximum human albumin levels (approx. 15 mg/mL plasma) in both male and female NOG-TKm30 mice engrafted with human hepatocytes (humanized liver mice) were observed 8–12 weeks after transplantation. Immunohistochemical analyses revealed abundant expression of major human cytochrome P450 (CYP) enzymes (CYP1A2, CYP2C9, CYP2D6, CYP2E1, and CYP3A4) in reconstituted liver with original zonal distribution. In vivo drug–drug interactions were observed in humanized liver mice as decreased area under the curve of midazolam (CYP3A4/5 substrate) and omeprazole (CYP3A4/5 and CYP2C19 substrate) after oral administration of rifampicin. Furthermore, we developed a pregnant model for evaluating prenatal exposure to drugs. The detection of thalidomide metabolites in the fetuses of pregnant humanized liver mice indicates that the novel TK model can be used for developmental toxicity studies requiring the assessment of human drug metabolism. These results suggest that the limitations of traditional TK-NOG mice can be addressed using NOG-TKm30 mice, which constitute a novel platform for humanized liver for both in vivo and in vitro studies.     

Differential metabolism of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) in mice humanized for CYP1A1 and CYP1A2

The procarcinogen 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) is the most abundant heterocyclic amine formed during the cooking of foods. Metabolism of PhIP by CYP1A2 differs substantially between humans and rodents, with more N2-hydroxylation (activation) and less 4'-hydroxylation (detoxication) in humans. Therefore, the human response to PhIP and other heterocyclic amine exposure may not be accurately reflected in the laboratory rodent. By generating mouse models expressing the human genes, species differences in heterocyclic amine metabolism can be addressed. Two transgenic mouse lines were developed, one expressing the human CYP1A1 CYP1A2 transgene in a mouse Cyp1a1-null background (hCYP1A1) and another expressing human CYP1A1 CYP1A2 in a mouse Cyp1a2-null background (hCYP1A2). Expression of human CYP1A2 protein was detected in the liver and also at considerably lower levels in extrahepatic tissues such as lung, kidney, colon, and heart. In the hCYP1A1 and hCYP1A2 mice, 3-methylcholanthrene (3-MC) induced both human CYP1A1 and CYP1A2 protein in the liver. Differences in the metabolism of the heterocyclic amine PhIP were observed between wild-type and hCYP1A2 mice. PhIP was preferentially metabolized by N2-hydroxylation in hCYP1A2 mice, whereas in wild-type mice, 4'-hydroxylation was the predominant pathway. Since the N2-hydroxylation pathway for PhIP metabolism has been reported to be predominant in humans, these results illustrate the potential effectiveness of using these transgenic, humanized mice as models for determining human health risks to PhIP and other heterocyclic amines instead of wild-type mice.     

P450 and carcinogenesis

Multiple forms of cytochrome P450 play important roles in metabolic activation of a variety of environmental procarcinogens. Large species differences in substrate specificities between experimental animals and humans are critical factors in evaluation of chemical safety. To study the role of human P450s in genotoxic activation of environmental chemicals, transgenic bacteria expressing both human P450s and P450 reductase have been developed for the mutagenicity test. Mice lacking CYP1A2, and CYP1B1, and CYP2E1 were prepared to investigate the mechanism of procarcinogen activation in vivo. The first human transgenic animals were mice carrying human fetus-specific CYP3A7. Using these transgenic mice, mutagenic activation of a natural mycotoxin, aflatoxin B1, catalyzed by CYP3A7 in vivo was demonstrated. This observation was clear in extrahepatic tissues that did not express mouse CYP3A enzymes. In conclusion, P450s are key factors involved in metabolic activation of environmental procarcinogens for their biological actions.     

Characterization of transgenic mouse strains using six human hepatic cytochrome P450 probe substrates

1. Transgenic mice were evaluated with six human cytochrome P450 (CYP) selective probe substrates, as little is known about their metabolism in the mouse. Mouse strains characterized include C57BL/SJL, FVB/N, mdr 1a/1b (-/-), ob/ob and ACCA. 2. Human CYP probe substrates used for characterization of mouse CYP activities included bufuralol, testosterone, dextromethorphan, phenacetin, diclofenac and S-mephenytoin. Activities were compared with those obtained in human liver microsomes and in human recombinant enzyme preparations. All transgenic mouse strains showed similar apparent K(m) with bufuralol, testosterone and dextromethorphan which compared favourably with those observed in human liver microsomes. 3. K(m) for phenacetin O-deethylase and S-mephenytoin 4'-hydroxylation were more variable across strains and in some cases demonstrated biphasic kinetics. Phenacetin O-deethylase activity was low in all mouse strains except FVB/N and mdr 1a/1b (-/-). Diclofenac 4-hydroxylation did not occur to any significant extent in the five strains of mouse evaluated here. 4. The findings suggest the validity of using five of the probes for transgenic mouse hepatic CYP characterization and gross comparison with data generated with human CYP.     

Expression of human cytochromes P450 in chimeric mice with humanized liver.

Recently, a chimeric mouse line in which the liver could be replaced by more than 80% with human hepatocytes was established in Japan. Because the chimeric mouse produces human albumin (hAlb), replacement by human hepatocytes could be estimated by the hAlb concentration in the blood of chimeric mice. In this study, we investigated human major cytochrome P450 (P450) in the livers of chimeric mice by mRNA, protein, and enzyme activity using real-time polymerase chain reaction, Western blot analysis, and high-performance liquid chromatography, respectively. Chimeric mice with humanized liver generated using hepatocytes from a Japanese and white donor were used. Human P450 mRNAs were expressed in the liver of chimeric mice, and major human P450 proteins such as CYP1A2, CYP2C9, and CYP3A4 were detected. The expression of P450 mRNA and protein was correlated with the hAlb concentration in the blood. The enzyme activities such as diclofenac 4'-hydroxylase activity, dexamethasone 6-hydroxylase activity, and coumarin 7-hydroxylase activity, activities that are specific to human P450 but not to murine P450, were increased in a hAlb concentration-dependent manner. The chimeric mice with nearly 90% replacement by human hepatocytes demonstrated almost the same protein contents of human P450s and drug-metabolizing enzyme activity as those of the donor. It was confirmed that genomic DNA from the livers of the chimeric mice and that from the liver of the donor exhibited the same genotype. In conclusion, the chimeric mice exhibited a similarly efficient capacity of drug metabolism as humans, suggesting that they could be a useful animal model for drug development.     

Cytochrome P450-dependent drug oxidation activities and their expression levels in liver microsomes of chimeric TK-NOG mice with humanized livers

Hepatic cytochrome P450 (P450)-dependent drug oxidation activity has not been completely characterized in chimeric TK-NOG mice with humanized livers (humanized liver mice). In this study, we examined several drug oxidation activities catalyzed by liver microsomes from humans, humanized liver mice, and TK-NOG mice using 9 P450 substrates. The catalytic activities of liver microsomes from humans and humanized liver mice showed relatively similar rates of oxidation of 7-ethoxyresorufin, coumarin, 7-pentoxyresorufin, flurbiprofen, S-mephenytoin, chlorzoxazone, and midazolam, whereas bufuralol 1′-hydroxylation and propafenone 4′-hydroxylation (rodent-specific propafenone oxidation activity) were higher in humanized liver mice than in humans. In addition, P450 protein expression levels in the humanized mouse liver were quantified using a liquid chromatography–tandem mass spectrometry-based protein quantification method. Quantification of P450 enzymes showed a 3-fold difference between human and humanized liver mouse livers, except for CYP2B6, which showed an approximately 6-fold difference. Overall, most P450-dependent drug oxidation activities were comparable between liver microsomes from human and humanized liver mice based on the similar expression levels of human P450 enzymes. However, some differences were observed between both species, including considerable differences in bufuralol 1′-hydroxylation and propafenone 4′-hydroxylation activities.     

Acetone catabolism by cytochrome P450 2E1: studies with CYP2E1-null mice.

Previous experiments in vitro have suggested that cytochrome P450 2E1 (CYP2E1) is involved in acetone catabolism by converting acetone to acetol and then to methylglyoxal, both intermediates in the gluconeogenic pathway. In the present study, CYP2E1-null mice were used to demonstrate the role of CYP2E1 in acetone catabolism in vivo. The blood acetone level in male CYP2E1-null mice was 3.3 +/- 0.9 microg/mL, which was similar to levels of their sex- and age-matched parental lineage strains C57BL/6N (2.3 +/- 0.2 microg/mL) and 129/Sv (3.5 +/- 0.3 microg/mL) mice (both are CYP2E1 wild-type). After fasting for 48 hr, the blood acetone levels in the CYP2E1 wild-type mice were increased by 2.5- to 4.4-fold, but that in the CYP2E1-null mice increased 28-fold. These results clearly demonstrate that CYP2E1 plays a vital role in the catabolism of acetone under fasting conditions.     

Role of CYP3A and CYP2E1 in alcohol-mediated increases in acetaminophen hepatotoxicity: comparison of wild-type and Cyp2e1(-/-) mice.

CYP2E1 is widely accepted as the sole form of cytochrome P450 responsible for alcohol-mediated increases in acetaminophen (APAP) hepatotoxicity. However, we previously found that alcohol [ethanol and isopentanol (EIP)] causes increases in APAP hepatotoxicity in Cyp2e1(-/-) mice, indicating that CYP2E1 is not essential. Here, using wild-type and Cyp2e1(-/-) mice, we investigated the relative roles of CYP2E1 and CYP3A in EIP-mediated increases in APAP hepatotoxicity. We found that EIP-mediated increases in APAP hepatotoxicity occurred at lower APAP doses in wild-type mice (300 mg/kg) than in Cyp2e1(-/-) mice (600 mg/kg). Although this result suggests that CYP2E1 has a role in the different susceptibilities of these mouse lines, our findings that EIP-mediated increases in CYP3A activities were greater in wild-type mice compared with Cyp2e1(-/-) mice raises the possibility that differential increases in CYP3A may also contribute to the greater APAP sensitivity in EIP-pretreated wild-type mice. At the time of APAP administration, which followed an 11 h withdrawal from the alcohols, alcohol-induced levels of CYP3A were sustained in both mouse lines, whereas CYP2E1 was decreased to constitutive levels in wild-type mice. The CYP3A inhibitor triacetyloleandomycin (TAO) decreased APAP hepatotoxicity in EIP-pretreated wild-type and Cyp2e1(-/-) mice. TAO treatment in vivo resulted in inhibition of microsomal CYP3A-catalyzed activity, measured in vitro, with no inhibition of CYP1A2 and CYP2E1 activities. In conclusion, these findings suggest that both CYP3A and CYP2E1 contribute to APAP hepatotoxicity in alcohol-treated mice.