Metabolism of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine by human cytochrome P4501A1, P4501A2 and P4501B1

While the metabolic activation of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) by N-hydroxylation has been well documented, the relative roles of the human cytochrome P450 (CYP) enzymes that catalyze this reaction have not been established. Previous studies indicated that the mutagenic activation product, 2-hydroxyamino-PhIP (N2-OH-PhIP), is produced primarily by CYP1A2, and to a lesser extent by CYP1A1. We recently reported that human CYP1B1 also produces N2-OH-PhIP (Carcinogenesis, 18, 1793-1798, 1997). In the present study, we examined PhIP metabolism by microsomes containing recombinant human CYP1A1, 1A2 or 1B1 expressed in Sf9 insect cells and compared the kinetic values for PhIP metabolite formation. PhIP metabolites were analyzed by high pressure liquid chromatography with fluorescence and absorbance detection. Vmax values for N2-OH-PhIP formation were 90, 16 and 0.2 nmol/min/nmol P450, and the apparent Km values were 79, 5.1 and 4.5 microM for human CYP1A2, 1A1 and 1B1, respectively. The non-mutagenic metabolite, 4'-hydroxy-PhIP, was also formed by all three CYP enzymes with Vmax values of 1.5, 7.8 and 0.3 nmol/ min/nmol P450 and apparent Km values of 43, 8.2 and 2.2 microM for human CYP1A2, 1A1 and 1B1, respectively. Although the Vmax for N2-OH-PhIP production was highest for CYP1A2, the catalytic efficiency (Vmax/Km) of CYP1A1 was greater than that of CYP1A2. These results suggest that, for humans, extrahepatic CYP1A1 may be more important than previously thought for the metabolic activation of the dietary carcinogen PhIP.     

CYP4T1--a cytochrome P450 expressed in rainbow trout (Oncorhynchus mykiss) liver

Total RNA isolated from a rainbow trout (Oncorhynchus mykiss) liver was subjected to RT/PCR using degenerate primers designed from homologous regions amongst cytochrome P450 CYP4 proteins. PCR amplification resulted in a single electrophoretic band which was excised, purified and sequenced directly, using cycle sequencing. The deduced protein sequence demonstrated the closest amino acid identity to rabbit CYP4B1 (54.6%) and rat CYP4B2 (55.4%). Phylogenic analysis of this sequence was found to be significantly different to any other CYP4 sequence and has been named CYP4T1. This represents the first CYP4 family member to be identified in an aquatic vertebrate.     

Major involvement of rabbit liver cytochrome P4501A in thiabendazole 5-hydroxylation

1. Thiabendazole is a widely used food preservative and anthelmintic drug for breeding animal species. In order to characterize precisely the cytochrome P450 isozyme(s) involved in its major route of metabolism, a rapid and sensitive spectrofluorimetric method was developed for the simultaneous determination of thiabendazole and its main hepatic metabolite 5-hydroxythiabendazole. 2. The kinetics of thiabendazole 5-hydroxylation were determined in microsomal preparations from control rabbits or animals previously treated with either beta-naphthoflavone, isosafrole, phenobarbital, rifampicin or clofibrate. These treatments led to specific induction of CYP1A1, 1A2, 2B4, 3A6 and 4A1 respectively. 3. By considering this panel of characterised microsomal preparations, only those obtained from BNF-treated rabbits exhibited an increase in thiabendazole 5-hydroxylase activity Ethoxyresorufin O-deethylation in these microsomes was solely inhibited by thiabendazole. These argue for a specific involvement of the CYP1A subfamily. 4. In the CYP1A subfamily, CYP1A2 appears to be responsible for basal 5-hydroxylation and further unidentified metabolism of thiabendazole in control livers. However, the major involvement of CYP1A1 is supported by the following characteristics of 5-hydroxylation of thiabendazole: (1) the correlation with CYP1A1 expression and (2) the inhibition by ellipticine and not by furafylline, inhibitors of CYP1A1 and CYP1A2 respectively. 5. All these data demonstrated that the rabbit cytochrome P4501A is predominantly involved in thiabendazole 5-hydroxylation which has been suspected to be critical in terms of safety of the parent drug.     

Effects of estradiol and progesterone on cytochrome P4501A1 expression in Hepa 1c1c7 cells

The effects of estradiol and progesterone on the expression of cytochrome P4501A1 were investigated in Hepa 1c1c7 cells. Both steroids, at 10 microM concentration, increased P4501A1-mediated 7-ethoxyresorufin O-deethyalase activity and amounts of its immunoreactive protein and CYP1A1 mRNA. Gel shift assay revealed that the steroids could induce both AhR transformation and binding of the ligand-AhR complex to its specific DNA recognition site. Transient transfection demonstrated that 5'flanking region of CYP1A1 could respond to the steroid action. The competitive binding assay showed that the steroids bound to AhR with moderate affinity. These results suggested that steroidal structure can be AhR ligands and induce CYP1A1 expression in AhR-dependent manner.     

Identification of c-fos related genes and their induction by neural activation in rainbow trout brain

The anterior portion of the neonatal telencephalic subventricular zone (SVZa) contains proliferating cells that generate an immense number of neurons destined to become the granule and periglomerular cells of the olfactory bulb. In contrast to other immature neurons in the central nervous system, cells arising in the SVZa maintain the ability to divide as they traverse the rostral migratory stream to their final destinations despite expressing an antigenic marker of differentiated neurons (Menezes et al. [1995] Molec. Cell. Neurosci. 6:496-508). Because of their considerable proliferative capacities and unusual mitotic behavior, we decided to determine the cell cycle length of proliferating cells within the SVZa and within the migratory pathway used by SVZa-derived cells. Following the methodology of Nowakowski et al. [1989](J. Neurocytol. 18:311-318), postnatal day 2 rat pups were exposed to 5'-bromo-2'deoxyuridine (BrdU) for increasing periods of time before perfusion. By plotting the percentage of nuclei undergoing DNA synthesis in the SVZa at each time versus the BrdU labeling interval, we determined that approximately 15% of the SVZa population is actively dividing and that these cells have a cycle length of approximately 14 hr, significantly less than the 18.6 hr determined to be the cycle length of dividing cells in more posterior, glia-generating regions of the subventricular zone (Thomaidou et al. [1997] J. Neurosci. 17:1075-1085). The cycle length of cells dividing in the mid portion of the rostral migratory stream, however, is considerably longer: 17.3 hr. This may reflect the need for these cells to coordinate the processes of migration and division. Our studies also suggest that there may be regional differences in the types of descendants produced by the proliferating cells. Retroviral lineage tracing studies showed that those cells that divide within the rostral migratory stream, like proliferating cells within the SVZa, make cells destined for the olfactory bulb. Unlike the progenitors that divide within the SVZa and generate more granule cells than periglomerular cells, the proliferating cells within the migratory pathway generate more periglomerular cells than granule cells. Collectively the proliferating cells of the SVZa and migratory pathway produce a large number of olfactory bulb interneurons. Our work suggests that this may be achieved in part by the relatively rapid divisions of progenitor cells within the SVZa and in part by the ongoing division of migrating cells en route to the olfactory bulb.     

Metabolism of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) by human cytochrome P4501B1

Cytochrome P4501B1 (CYP1B1) is the most recently identified member of the dioxin-inducible CYP1 family. CYP1B1 is constitutively expressed in most human tissues, including colon and breast, and can activate numerous chemically diverse carcinogens. We evaluated the metabolism of the dietary heterocyclic amine carcinogen 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) by microsomes from yeast expressing the human CYP1B1 protein. PhIP metabolites were analysed by HPLC with fluorescence and absorbance detection. We found that human CYP1B1 metabolizes PhIP to three products: N2-OH-PhIP, a mutagenic activation product; 4'-OH-PhIP, a detoxification product; and 2-OH-PhIP, the mutagenic potential of which is unknown. Metabolite identity was confirmed by co-elution with authentic standards and synchronous fluorescence spectroscopy. The identity of the 2-OH-PhIP standard was additionally confirmed by mass spectrometry. Kinetic studies of the formation of N2-OH-PhIP, 4'-OH-PhIP and 2-OH-PhIP by CYP1B1 indicated apparent Km values of 5.7 +/- 1.3, 2.2 +/- 0.5 and 1.3 +/- 0.2 microM, respectively. Apparent turnover rates were 0.40 +/- 0.03, 0.93 +/- 0.02 and 0.04 +/- 0.00 nmol product/min nmol P450, respectively. At saturating levels of substrate, CYP1B1-mediated formation of the non-mutagenic metabolite 4'-OH-PhIP was favored two-fold over that of the mutagenic metabolite, N2-OH-PhIP and >10-fold over that of 2-OH-PhIP. The formation of N2-OH-PhIP, a potent mutagen implicated in the etiology of human colon and breast cancer, indicates that CYP1B1 may play an important role in PhIP-mediated carcinogenesis.     

Determination of human hepatic cytochrome P4501A2 activity in vitro use of tacrine as an isoenzyme-specific probe

Oxidative metabolism of the cognition activator tacrine (1,2,3,4-tetrahydro-9-aminoacridine) is thought to be catalyzed by cytochrome P4501A2 (CYP1A2). In this study, the use of tacrine as a specific substrate to measure CYP1A2 activity in vitro was investigated. Tacrine metabolism was assessed in 16 human liver microsomal samples. Initially, the percentage conversion of tacrine to stable metabolites (i.e. 1-, 2-, 4-, and 7-hydroxytacrine) at a single time point was correlated with levels of CYP1A2 apoprotein. Apoprotein was detected by immunoquantification using a monospecific CYP1A2 antipeptide antibody. Significant correlations were seen between CYP1A2 content and the degree of 1-hydroxylation (r = 0.81, p < 0.001), 7-hydroxylation (r = 0.70, p < 0.001), and metabolism to all stable products (r = 0.82, p < 0.001). The major metabolite formed in all livers was 1-hydroxytacrine. The conversion of tacrine to this metabolite was examined in more detail. The rate of formation varied from 19.2 pmol min-1 mg-1 to 101.0 pmol min-1 mg-1. There was a significant correlation (r = 0.84, p < 0.001) between the rate of formation and CYP1A2 levels. Tacrine metabolism was also compared with the rate of formation of 3-methylxanthine, from theophylline, a reaction previously shown to be catalyzed by CYP1A2. Significant correlations were found between 3-methylxanthine formation and all quantified tacrine metabolites. The rate of 3-methylxanthine generation also correlated with CYP1A2 apoprotein levels. It is concluded, therefore, that tacrine is a valuable probe for the determination of human hepatic CYP1A2 activity in vitro.     

Glutathione S-transferase M1 and T1 and cytochrome P4501A1 polymorphisms in relation to the risk for benign and malignant head and neck lesions

BACKGROUND: Susceptibility to head and neck cancer in a particular individual may depend in part on the metabolic balance between Phase 1 enzymes, such as cytochromes P450 (CYPs), and Phase II enzymes, such as glutathione S-transferases (GSTs). Genetic variability in CYP and GST isoenzymes may contribute to individual differences in susceptibility to chemical carcinogenesis. GSTM1 and GSTT1 null genotypes as well as polymorphic variants in the CYP1A1 gene that may help determine the risk for head and neck cancer have been described in previous reports. METHODS: Polymorphisms of GSTM1, GSTT1, and CYP1A1 in whole blood were detected by polymerase chain reaction (PCR) in 185 patients with head and neck squamous cell carcinoma (HNSCC), 78 patients with benign head and neck lesions (BHNL), and 207 blood donors. RESULTS: GSTM1 null genotype was demonstrated to be equally frequent in patients with HNSCC (50.8%), patients with BHNL (47.4%), and blood donors (51.7%). GSTT1 null genotype occurred significantly more often in patients with BHNL (33.3%) as compared with blood donors (20.3%), demonstrating that lack of GSTT1 may be a risk factor for BHNL. Presence of the rare valine in the CYP1A1/Nco1 site was found in 33 patients with HNSCC (17.8%), in 20 patients with BHNL (25.6%), and in 34 blood donors (16.4%). The frequencies with which the presence of the rare cytosine nucleotide in the CYP1A1/Msp1 site was detected were 17.8%, 15.4%, and 15.9%, respectively. CONCLUSIONS: The occurrence of polymorphic variants in the GSTM1, GSTT1, and CYP1A1 genes did not differ between the groups investigated, therefore indicating no significant contribution to the development of head and neck cancer.     

Differential inhibition of epidermal growth factor signaling pathways in rat hepatocytes by long-term ethanol treatment

BACKGROUND & AIMS: Long-term ethanol intake suppresses liver regeneration in vivo and ethanol interferes with epidermal growth factor (EGF)-induced DNA synthesis in vitro. Therefore, the effects of long-term ethanol treatment on EGF-activated signaling reactions in rat hepatocytes were investigated. METHODS: Hepatocytes from long-term ethanol-fed rats and pair-fed controls were stimulated with EGF (0.5-20 nmol/L) for 15-120 seconds. Tyrosine phosphorylation of EGF receptor (EGFR), Shc, and phospholipase-C gamma1 (PLC gamma), and growth factor receptor binding protein 2 (Grb2) coprecipitation with EGFR and Shc were analyzed by Western blotting. RESULTS: EGFR autophosphorylation was suppressed at all EGF concentrations in ethanol-fed cells compared with pair-fed cells, without significant differences in total EGFR protein or EGFR tyrosine kinase activity detected in cell lysates, suggesting that intracellular factors suppressed EGFR function. EGF-induced PLC gamma tyrosine phosphorylation and inositol 1,4,5-trisphosphate (InsP3) formation were suppressed, but cytosolic [Ca2+]c elevation was little affected, indicating enhanced InsP3-mediated intracellular Ca2+ release in ethanol-fed cells. Grb2 binding to EGFR was suppressed, but EGF-induced Shc tyrosine phosphorylation and Grb2 association with Shc were not significantly decreased. CONCLUSIONS: Long-term ethanol feeding suppressed EGF-induced receptor autophosphorylation in rat hepatocytes with differential inhibition of downstream signaling processes mediated by PLC gamma, Shc, and Grb2. Altered patterns of downstream signals emanating from EGFR may contribute to deficient liver regeneration in chronic alcoholism.     

Glyphosate is an inhibitor of plant cytochrome P450: functional expression of Thlaspi arvensae cytochrome P45071B1/reductase fusion protein in Escherichia coli

Radiofrequency (RF) energy has many advantages in thermal tumor ablation protocols. With the recent development of open MRI systems, interventional MRI procedures, including thermal ablation, have become the focus of great research interest. However, the significant interference between RF generators and MR imagers has prevented simultaneous imaging and RF ablation and, until now, has limited the role of RF-based thermal therapy in interventional MRI. Here, a simple switching circuit designed with consideration of patient safety provides compatibility between open MRI systems and RF thermal lesion generators. The experimental results show that the switching circuit allows imaging during RF ablation and opens new opportunity for MR-guided thermal therapy.     

2-Methoxy-4-nitroaniline and its isomers induce cytochrome P4501A (CYP1A) enzymes with different selectivities in the rat liver

We reported previously that 2-methoxy-4-nitroaniline (2-MeO-4-NA) is a selective inducer of cytochrome P4501A2 (CYP1A2) in the rat liver, and its molecular size is the smallest among known CYP1A2-selective inducers. In the present study, a structure-activity relationship on the CYP1A2-selective induction has been investigated using isomeric nitroanisidines and their related chemicals. Western blot analyses revealed that the chemicals removed a substituent (amino, methoxyl or nitro group) from a 2-MeO-4-NA molecule had no capacity for inducing CYP1A enzymes in rat livers. On the other hand, isomeric nitroanisidines such as 2-MeO-4-NA, 2-MeO-5-NA and 4-MeO-2-NA induced both CYP1A2 and CYP1A1 enzymes with different selectivities. As judged from the induced levels of CYP1A proteins, 2-MeO-4-NA (CYP1A2/CYP1A1 ratio; 9.5) and 4-MeO-2-NA (0.3) were the most selective inducers of CYP1A2 and CYP1A1, respectively, among the isomeric nitroanisidines (0.44 mmol/kg) used. The induced level of CYP1A2 protein was in the order 2-MeO-4-NA > 2-MeO-5-NA > 4-MeO-2-NA, although no significant difference was observed on their CYP1A2 mRNA level. On the contrary, increases in the levels of CYP1A1 mRNA and protein were in the order 4-MeO-2-NA > 2-MeO-5-NA > 2-MeO-4-NA. The present findings indicate that all three substituents (amino, methoxyl and nitro groups) are necessary components of nitroanisidines for induction of CYP1A enzymes, and also show that regio-isomeric positions of these substituents determine the selectivity in the induction of CYP1A enzymes.     

Absence of the glutathione S-transferase M1 gene increases cytochrome P4501A2 activity among frequent consumers of cruciferous vegetables in a Caucasian population

The cancer protective effect of cruciferous vegetables has been attributed to induction of phase II enzymes. But cruciferous vegetables also induce cytochrome P4501A2 (CYP1A2), which catalyzes the metabolic activation of various procarcinogens, including aromatic amines in tobacco. Thus, frequent intake of cruciferous vegetables could also result in cancer-enhancing effects. GSTM1 is involved in the detoxification of various carcinogens, but it also enhances the excretion of isothiocyanates and possibly other enzyme inducers in cruciferous vegetables. We, therefore, hypothesized that GSTM1 null genotype might be associated with increased CYP1A2 activity among frequent consumers of cruciferous vegetables because the excretion of CYP1A2 inducers contained in these vegetables may be partially inhibited in the absence of the GSTM1 enzyme. Three hundred twenty-eight non-Hispanic white residents of Los Angeles County (265 males and 63 females) were genotyped for the presence or absence of GSTM1 alleles and phenotyped for CYP1A2 activity. Information on usual dietary habits was obtained from these subjects through in-person interviews. Among frequent (at least once a week) consumers of broccoli, GSTM1 null individuals exhibited a 21% higher geometric mean level of CYP1A2 activity relative to GSTM1 non-null individuals (5.24 versus 4.32, two-sided P = 0.01). No such difference was observed in subjects who consumed broccoli less frequently (two-sided P = 0.39). This interactive effect of GSTM1 genotype and vegetable intake on CYP1A2 activity also was observed when overall intake of the five cruciferous vegetables under study (broccoli, cabbage, cauliflower, Brussels sprouts, and mustard greens) was examined. Among weekly consumers of cruciferous vegetables, GSTM1 null individuals showed a 16% higher geometric mean level of CYP1A2 activity relative to GSTM1 non-null individuals (5.03 versus 4.33, two-sided P = 0.02), whereas no difference was evident among those who consumed cruciferous vegetables less frequently (two-sided P = 0.35). Our results suggest that cruciferous vegetables contain CYP1A2 inducers, which are deactivated in the presence of GSTM1.     

Characterization of cytochrome P450 (CYP3A12) induction by rifampicin in dog liver

We investigated the mechanisms of renal vascular wall thickening in a rat model of N-nitro L-arginine methyl ester (L-NAME)-induced hypertension. To separate the effects of L-NAME-induced hypertension from other effects of nitric oxide (NO) inhibition, we created two models of L-NAME-induced hypertension: both had the same blood pressure level but NO inhibition was moderate in one group (group M) and severe in the other (group S). Urinary excretion of nitrates and nitrites was lower in group S than in group M. Wall thickening and lipid deposition in renal vessels were significantly greater in group S than in groups M. Simple and multiple regression analyses indicated that renal vascular wall thickening was more strongly correlated with lipid deposition than with blood pressure. The number of vessels positive for staining with Sudan black B was negatively correlated with urinary NO excretion. Expression of fibronectin and transforming growth factor-beta was greater in the Sudan black B-positive than in the Sudan black B-negative vessels, suggesting that extracellular matrix production was increased in vessels with lipid deposition. Lipid deposition and increased production of extracellular matrix may contribute to renal vascular wall thickening in L-NAME-induced hypertension. Some mechanisms independent of hypertension play important roles in vascular wall thickening induced by NO inhibition.     

Metabolism of 2-amino-alpha-carboline. A food-borne heterocyclic amine mutagen and carcinogen by human and rodent liver microsomes and by human cytochrome P4501A2

In the present study, the responsive properties of afferent units in caudal nerve of streptozotocin (STZ) induced diabetic rats were investigated. It was found that some of the C and A delta units spontaneously discharged and their mechanical threshold was decreased significantly in diabetic rats. The response of the C units of diabetic and control rats to a 1 min sustained mechanical stimulus of threshold strength was similar, whereas, discharge frequencies during sustained suprathreshold mechanical stimulus were significantly greater in the C units of diabetic rats. After removal of threshold or suprathreshold mechanical stimulus, after discharges were also greater in C units of diabetic rats. Conduction velocity of C units, A delta units and A beta mechanical receptive units were also decreased in diabetic rats. However, there was no significant difference in the subunits of all kinds of receptive units between diabetic and control rats. The present data suggested that the C and A delta units with lowered mechanical threshold contribute to the decrease of the behavioral nociceptive threshold of diabetic rats and abnormal discharge of the C and A delta units may be a peripheral factor in hypersensitivity to painful stimuli and paresthesia of diabetic rats.     

Hepatic cellular distribution of cytochrome P-450 IA1 in rainbow trout (Oncorhynchus mykiss): an immunohisto- and cytochemical study

Monoclonal antibody 1-12-3 reactive against scup (Stenotomus chrysops) cytochrome P450 E (a teleost CYP IA1) has been used to immunohistochemically localize CYP IA1 within hepatocytes and presumably sinusoidal endothelial and biliary epithelial cells of scup and trout. The goal of the present study was to extend immunohistochemical studies to the ultrastructural level determining intracellular locations of CYP IA1 in fish liver. Juvenile trout (5-10 g) were given i.p. injections once (50 micrograms/g b beta-naphthoflavone in cod liver oil; 0.5-ml injectate volume). After 5 days, livers were fixed (0.25% glutaraldehyde) via vascular in situ perfusion, removed, cut in 100-microns slices, infiltrated, and embedded in LR White monomer. Ultrathin sections of exposed livers were incubated in monoclonal antibody 1-12-3, rabbit anti-mouse IgG, and protein G colloidal gold. Membranes of granular endoplasmic reticulum in perinuclear regions of hepatocytes were consistently labeled. In addition, hepatocyte plasma membrane, particularly microvilli at bile canaliculi, was labeled. Biliary epithelial cells were labeled on luminal plasma membrane surrounding biliary passageway. Plasma membrane facing sinusoid and immediately subjacent cytoplasm was labeled in endothelial cells. Presence of CYP IA1 in sinusoidal endothelium could contribute to detoxication and/or bioactivation of blood borne chemicals. Granular endoplasmic reticulum was not uniformly labeled in hepatocytes. Rather, distribution seemed sequestered within highly specific regions and not dispersed along all membrane surfaces. Localization within biliary epithelial cells could signify potential of this cell type to bioactivate polycyclic aromatic hydrocarbons and may explain the common finding of biliary as well as hepatocytic tumors of trout liver.     

Regulation of cytochrome P450 expression by inhibitors of hydroxymethylglutaryl-coenzyme A reductase in primary cultured rat hepatocytes and in rat liver

It was previously demonstrated that treatment of primary cultured rat hepatocytes with lovastatin, an inhibitor of 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) reductase, induced the mRNAs for several cytochromes P450 (P450s), including CYP2B1/2, CYP3A1/2, and CYP4A. In this study, we have compared the effects of lovastatin with those of three additional HMG-CoA reductase inhibitors (simvastatin, pravastatin, and the structurally dissimilar drug fluvastatin) on P450 expression in primary cultured rat hepatocytes, and we have also characterized the effects of in vivo treatment with fluvastatin on P450 expression in rat liver. Treatment of cultured hepatocytes with lovastatin, simvastatin, or fluvastatin increased CYP2B1/2, CYP3A1/2, and CYP4A mRNA and immunoreactive protein levels over the dose range (3 x 10(-6) to 3 x 10(-5) M) required to increase the amount of HMG-CoA reductase mRNA. The increases in CYP2B1/2 levels produced by 3 x 10(-5) M fluvastatin treatment were larger than those produced by lovastatin or simvastatin treatment or by treatment with 10(-4) M phenobarbital. In contrast, treatment of cultured hepatocytes with 3 x 10(-5) M lovastatin, simvastatin, or fluvastatin increased CYP3A1/2 and CYP4A mRNA and immunoreactive protein to lower levels than those produced by treatment with 10(-5) M dexamethasone or 10(-4) M ciprofibrate. Treatment of cultured hepatocytes with pravastatin had little or no effect on the amount of any of the P450s examined, although this drug induced HMG-CoA reductase mRNA as effectively as did fluvastatin. Incubation of hepatocytes with 10(-4) M fluvastatin increased CYP1A1 mRNA to 67% of the level induced by treatment with 10(-5) M beta-naphthoflavone. Doses of 50 or 100 mg/ kg/day fluvastatin administered for 3 days to rats increased the hepatic levels of CYP2B1/2 and CYP4A mRNA and immunoreactive protein, although to much lower levels than those produced by treatment with phenobarbital or ciprofibrate, respectively. Treatment of rats with fluvastatin had no effect on hepatic levels of CYP3A1/2 mRNA or immunoreactive protein. However, treatment with 50 mg/kg/day fluvastatin induced CYP1A1 mRNA and protein. The effects of fluvastatin treatment on P450 expression seen in primary cultured rat hepatocytes thus largely recapitulated the effects seen in vivo. The differences in effects among the HMG-CoA reductase inhibitors suggest that simple inhibition of HMG-CoA reductase cannot explain all of the effects of these drugs on P450 expression.