Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Cytochromes P450 (CYP) are mixed function oxidases that are predominantly expressed in the liver and biotransform drugs, endogenous compounds, dietary constituents, and environmental toxins (Lieber, 1997; Woodcroft and Novak, 1998; Zhukov and Ingelman-Sundberg, 1999). Although alcohol dehydrogenase (ADH) is the major contributor to the metabolism of ethanol (Crabb, 1995), the CYP2E1 enzyme has been proposed to account for approximately 20% of ethanol metabolism at physiologically relevant blood alcohol concentrations (Lieber, 1994; Matsumoto et al., 1996). Ethanol consumption leads to increased CYP2E1 mRNA transcript levels in alcohol abusers (Lieber, 1999), but ethanol can also increase CYP2E1 levels by stabilizing the enzyme and preventing its degradation (Zhukov and Ingelman-Sundberg, 1999). ADH is not induced by ethanol (Lieber, 1994), suggesting that the induction of CYP2E1 by chronic ethanol is responsible for the 2-fold increase in ethanol metabolism observed in heavy drinkers (Lieber, 1994, 1999).

CYP2E1 activity is also induced by endogenous substances (e.g., acetone) as well as by pathophysiological states such as diabetes, obesity, and hypertriglycerinemia (Lieber, 1997, 1999). The underlying mechanisms involve transcriptional, post-transcriptional, and post-translational regulation (Lieber, 1997, 1999). CYP2E1 has been proposed to contribute to gluconeogenesis because it metabolizes acetone to an intermediate of this energy-producing pathway (Lieber, 1997). This physiological role may contribute to its high functional conservation across species; human CYP2E1 exhibits catalytic properties that are similar to its rat ortholog and they appear to share many regulatory features (Lieber, 1999).

There is mounting evidence that CYP2E1 is a key factor in the pathogenesis of alcoholic liver disease (ALD), a risk factor for hepatocellular carcinoma (Jarvelainen et al., 2000). Due to its high NADPH oxidase activity (Zhukov and Ingelman-Sundberg, 1999), CYP2E1 generates ethanol- and oxygen-derived free radicals that can initiate lipid peroxidation, oxidative stress, and Kupffer cell activation, thereby propagating cellular injury and DNA strand breaks (Jarvelainen et al., 2000). Furthermore, CYP2E1 induction correlates with lipid peroxidation and pathological severity during chronic ethanol exposure (Jarvelainen et al., 2000).

This enzyme also bioactivates cytotoxins, including carbon tetrachloride, many organic solvents, and acetaminophen, and its activity strongly correlates with degree of tissue injury induced by these toxins (Lieber, 1997; Woodcroft and Novak, 1998). The enhanced toxicity associated with the 5- to 10-fold induction of CYP2E1 during chronic ethanol intake may explain the increased vulnerability of alcohol abusers to therapeutically and industrially used xenobiotics, such as acetaminophen and bromobenzene (Lieber, 1999).

Epidemiological findings have established an interactive influence of ethanol drinking and tobacco smoking on cancer development; the synergistic risk for certain cancers is about 50% higher than the sum of the risks from cigarette smoking or ethanol drinking alone. CYP2E1 bioactivates tobacco smoke and other procarcinogens (Lieber, 1997); the enzyme is implicated in benzene-induced hematological malignancies in humans (Rothman et al., 1997) and nitrosodimethylamine-induced tumorigenesis in animals (Lieber, 1997). Tobacco smoke has been shown to induce CYP2E1 activity in animal models and in humans (Villard et al., 1998; Benowitz et al., 1999). Therefore, the induction of CYP2E1 by both tobacco smoke and ethanol may contribute to their synergistic increase in risk of liver disease development.

A higher percentage of alcoholics smoke (80-95%) compared with nonalcoholics (25-30%) (Batel et al., 1995), and twin studies indicate that, for equal ethanol consumption, heavy smokers had higher ethanol elimination rates compared with nonsmokers (Kopun and Propping, 1977). The incidence of alcoholism is 10 times more likely in smokers than nonsmokers (Batel et al., 1995). Among nonalcoholics, smokers report drinking alcohol at levels that are about twice that of nonsmokers (Batel et al., 1995). Consistent with the idea that smoking increases ethanol metabolism leading to increased consumption, long-term nicotine exposure was shown to enhance ethanol consumption in rats (Blomqvist et al., 1996). Ethanol effects on central nicotinic receptors or other central or peripheral receptor systems have been postulated to contribute to this interaction between ethanol and nicotine (Blomqvist et al., 1996; Ericson et al., 2000). However, cigarette smoke has been shown to increase CYP2E1 activity (Villard et al., 1998; Benowitz et al., 1999). Therefore, we postulated that in addition to pharmacodynamic alterations, the higher ethanol consumption among smokers may be partly due to tobacco smoke constituent(s) (i.e., nicotine) increasing the metabolism of ethanol (metabolic tolerance), requiring dependent individuals to drink more ethanol for the same effect. The purpose of this study is to examine whether of behaviorally and pharmacologically relevant doses of ethanol and nicotine increase CYP2E1 protein and activity in rat liver.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. Protease inhibitor cocktail tablets and chemiluminescence blotting substrate were purchased from Roche Diagnostics (Laval, QC, Canada). Recombinant viral-expressed rat CYP2E1 in lymphoblastoid cells; baculoviral-expressed rat CYP1A1, CYP2B1, CYP3A4, CYP2C11, and CYP2A2 in insect cells; and expressed rat CYP2A1 Supersomes were purchased from GENTEST (Woburn, MA) and pyridine-induced rat liver microsomes were purchased from Oxford Biomedical Research, Inc. (Oxford, MI). Protran nitrocellulose membranes were purchased from Schleicher & Schuell (Keene, NH). Biotinylated anti-rabbit IgG secondary antibody, ABC Elite kit, and DAB/hydrogen peroxide kit were purchased from Vector Laboratories (Burlington, ON, Canada), and Neutravidin-conjugated horseradish peroxidase was purchased from Pierce Chemical (Rockford, IL). Protein assay kits, prestained molecular markers, and Bio-Dot microfiltration apparatus were purchased from Bio-Rad (Hercules, CA), StrataPrep Total RNA kit from Stratagene (San Diego, CA), and Millipore Ultrafree-DA Centrifugal Filter for DNA extraction (Millipore Corporation, Bedford, MA). Ethanol, nicotine bitartrate, chlorzoxazone (CZN), and 2-benzoxazolinone were purchased from Sigma-Aldrich Canada Ltd. (Oakville, ON, Canada).

Animals. Four groups (n = 4/group) of adult male Wistar rats (250-300 g; Charles River, St-Constant, QC, Canada) were injected subcutaneously, once per day, for 7 days with either 0, 0.1, 0.3, or 1.0 mg of nicotine base per kilogram of body weight, in the form of nicotine bitartrate in sterile saline (pH adjusted to 7.4). Saline and ethanol (in saline), at doses of 0.3, 1.0, and 3.0 g/kg of body weight (n = 4/group), were administered by gavage to food-deprived (2-4 h, to facilitate constant absorption) male Wistar rats once daily for 7 days. Animals were sacrificed by decapitation 4 h after the last drug injection and livers were removed, frozen immediately in liquid nitrogen, and stored at 80°C until processed for protein, activity, and mRNA studies. All experimental procedures described in this study were carried out in accordance with the guidelines for the care and use of laboratory animals and approved by the Animal Care Committee of the University of Toronto.

Membrane Preparation. Portions of livers were homogenized manually in 100 mM Tris, pH 7.4, with 0.1 mM EDTA, 0.32 M sucrose, 0.1 mM dithiothreitol, and protease inhibitor cocktail (1 tablet/50 ml of homogenizing buffer) on ice. Homogenates were centrifuged twice at 3000g for 3 min, and the membrane fractions were prepared by 110,000g centrifugation (Sorvall RC2-B Combi Plus Ultraspeed centrifuge; Sorvall, Newton, CT) of the supernatant fraction. The resulting membrane pellets were resuspended in 100 mM Tris, pH 7.4, 0.1 mM EDTA, 0.1 mM dithiothreitol, 1.15% w/v KCl, 20% v/v glycerol, and stored at 80°C. Microsomes from pyridine-induced rat liver, as well as from rat CYP2E1-expressed lymphoblastoid cells, were used as positive controls. For kinetic studies, liver microsomes were prepared in cold 1.15% KCl by centrifugation at 9,000g followed by 100,000g.

Immunoblotting. Untreated rat liver membranes were serially diluted and used to construct standard curves to establish conditions of linearity. Membrane proteins (2.5 µg of hepatic membranes and 0.18 pmol of expressed CYP2E1 were used as a standard) were separated by SDS-polyacrylamide gel electrophoresis (4% stacking and 8% separating gels), transferred onto nitrocellulose membrane, and probed with a rabbit polyclonal anti-rat CYP2E1 antibody at a 1:4000 dilution for 1.5 h; this antibody has been shown to be selective for CYP2E1 at concentrations used in this study (Hansson et al., 1990). Membranes were incubated with a secondary biotinylated goat anti-rabbit IgG (1:6000 dilution) for 1 h, followed by Neutravidin-conjugated horseradish peroxidase (1:40,000 dilution) for 20 min before detection by chemiluminescence and analyzed with an imaging system (Imaging Research Inc., St. Catherines, ON, Canada). Control blots were processed without primary antibody.

Immunohistochemistry. Liver samples were fixed in 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4, cryoprotected in 20% sucrose in phosphate buffer, and rapidly frozen in isopentane cooled on dry ice. Sections (16 µm thick) were cut on a freezing microtome, collected in well plates in PBS (10 mM sodium phosphate buffer, 0.9% sodium chloride) (pH 7.4). Sections were washed and blocked with PBS containing 4% Triton, 5% normal goat serum, 2% bovine serum albumin, and 1% skimmed milk for 2 h at room temperature. Sections were incubated at 4°C for 4 days in rabbit anti-rat CYP2E1 antibody (used in immunoblotting assays) diluted 1:1000 in PBS containing 2.5% Triton, 1% normal goat serum, and 1% bovine serum albumin. Sections were washed with PBS, incubated for 1 h at room temperature with biotinylated goat anti-rabbit IgG diluted 1:2000, and rewashed in PBS. Endogenous peroxidase activity in the sections was quenched with 0.3% hydrogen peroxide in PBS for 5 min. The antigen-antibody complex was visualized using the avidin-biotin complex technique followed by reaction with 3,3'-diaminobenzidine. Sections were dehydrated and mounted using xylene in Permount; control sections were incubated without primary antibody.

Chlorzoxazone Metabolism and Chemical Inhibition Assays. CZN 6-hydroxylation was assayed according to the method of Leclercq et al. (1998) for rat liver microsomes with the following modifications. For kinetic parameter studies, rat hepatic microsomes (0.1 mg) were mixed with 25 µl of 10 mM -NADPH, 25 µl of 50 mM MgCl₂ in Tris-HCl buffer, pH 7.4, to a final volume of 250 µl. Incubations were carried out at 37°C. Initial studies indicated that this reaction was linear up to 1.6 mg of microsomal protein and to an incubation time of 40 min; final assays used incubations of 0.1 mg of microsomal protein for 20 min. CZN stock solutions were prepared in methanol and all other stock solutions were made in Tris-HCl. CZN solutions were dried under N₂ and reconstituted in Tris-HCl prior to incubations. The reaction was stopped by the addition 2 ml of ethyl acetate and 100 µl of 400 µM 2-benzoxazolinone (internal standard) and subsequently extracted. The organic phase was evaporated to dryness and reconstituted into 150 µl of mobile phase consisting of 19% acetonitrile in 10 mM sodium acetate buffer, pH 4.5. Formation of 6-hydroxy-CZN (6-OH-CZN) was measured by high-performance liquid chromatography with UV detection at 295 nm. A Waters Spherisorb S5 ODS2 column (4.6 × 150 mm; Waters, Bedford, MA) was used to separate CZN, 6-OH-CZN, and internal standard by using a mobile phase. The separation was performed with isocratic elution at a flow rate of 1 ml/min. The retention times for CZN, 6-OH-CZN, and 2-benzoxazolinone were 19.6, 4.7, and 6.5 min, respectively. CZN metabolic assays (5-400 µM) were performed to obtain kinetic parameters (K_m and V_max) in rat liver microsomes. Chemical inhibition studies were performed using 150 µM CZN, 0.1 mg of hepatic microsomal protein, or 11.5 pmol of expressed CYP2E1, incubated for 20 min at 37°C with either 10 or 100 µM nicotine bitartrate, or 100 or 500 µM aniline. Because aniline was dissolved in methanol and low methanol concentrations strongly inhibit CYP2E1 activity, solutions for aniline experiments were dried under N₂ and reconstituted prior to incubation. Because diethyldithiocarbamate is a mechanism-based inhibitor of CYP2E1, microsomes were preincubated for 20 min with 31.5 or 315 µM DDC prior to the addition of CZN. Substrate and inhibitor concentrations were chosen based on our current experimental data and published data (Jayyosi et al., 1995; Bourrie et al., 1996; Eagling et al., 1998). Negative controls consisted of incubations without NADPH or with heat-denatured microsomal protein.

RNA Slot Blot Analysis. Total RNA from rat livers was isolated using StrataPrep Total RNA kit and its quality assessed by electrophoresis in 1.2% agarose gel. Yeast tRNA was added to all samples (9.5 µg) and cDNA standards (10 µg) to reduce background. Total liver RNA (2.5 µg), serially diluted full-length CYP2E1 cDNA (0.63-20 pg, to ensure linearity of the assay and as a positive control), cDNAs for CYP2B6, CYP2A6, CYP2D6, and CYP3A4 (1.0 and 10.0 pg each, as negative controls) were applied directly to nylon membranes under vacuum and denaturing conditions by using a Bio-Dot microfiltration apparatus according to manufacturer's instructions. Membranes were prehybridized for at least 60 min at 43°C in buffer [50% formamide, 120 mM Na₂HPO₄ (pH 7.2), 7% SDS, and 250 mM NaCl] then hybridized for 16 h with either 1.4 × 10⁷ cpm/ml [-³²P]dCTP random-primed full-length human CYP2E1 cDNA or with two end-labeled rat oligonucleotide hybridization probes (each 2.0 × 10⁷ cpm/ml [-³²P]dATP) in the same buffer. Specifically, the rat oligonucleotides that are the reverse complementary sequences to 5'-TTTATTTCAGACACATTTTTC-3' and 5'-GAAGTTTTCATTGAACAAACT-3' that correspond to positions 721 to 740 and 613 to 633 of CYP2E1 mRNA, respectively, were used as probes; these sequences were specific for CYP2E1 mRNA. In addition, blots were probed with a 1.3 × 10⁷ cpm/ml [-³²P]dCTP random-labeled 517-base pair PCR product of -actin (a loading control), which was made using the forward primer 5'-CACCACAGCTGAGAGGGAAATCGTGCGTGA-3', the reverse primer 5'-ATTTGCGGTGCACGATGGAGGGGCCGGACT-3', and rat brain -actin cDNA template followed by gel extraction of the PCR product. A serial dilution of the -actin PCR product (0.008-1000 pg) was also probed to verify the linearity of our detection system. Blots were washed at room temperature sequentially in 2× then 0.5× saline sodium citrate with 0.1% SDS for 15 min each and exposed to Kodak OMAT-XR film for 1 to 8 days at 80°C. All films were analyzed using a digital imaging system (Imaging Research Inc.).

Determination of Plasma Nicotine Levels. Plasma concentrations of nicotine in frozen rat plasma samples were determined by a well established gas-liquid chromatography technique with 5-methylnicotine as the internal standard (Pacifici et al., 1993). The limit of detection of the assay was 0.5 ng/ml and there was a linear relationship between detected chromatographic peak and nicotine concentration (data not shown).

Analysis. The kinetic parameters (K_m and V_max) for CZN metabolism by liver microsomes from rats treated with saline, nicotine, or ethanol were analyzed using Enzfitter computer software, version 1.05 (Elsevier Biosoft Inc., Cambridge, UK). Treatment groups were considered to be significantly different from control if p < 0.05, using unpaired student t-tests.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

CYP2E1 Is Induced by Low Doses of Ethanol and Nicotine. An immunoblotting assay was developed to measure hepatic CYP2E1. Detection of serially diluted CYP2E1 indicated that the immunoblotting signal was linear up to 10 µg of protein from untreated liver membranes (Fig. 1A); no band was detected in the absence of primary antibody (Fig. 1A). The specificity of the rabbit antibody for rat CYP2E1 has been demonstrated (Hansson et al., 1990); we have confirmed this finding because no cross-reactivity was observed with expressed rat CYP2C11, 2B1, 3A2, 1A1, 2A1, or 2A2 isozymes under immunoblotting conditions used in this study (Fig. 1B). The immunoreactive band from expressed rat CYP2E1 microsomes comigrated with liver microsomes from rats treated with the CYP2E1 inducer pyridine in vivo and untreated liver membranes (Fig. 1B). A 7-day ethanol treatment dose dependently induced CYP2E1 protein in rat hepatic membranes (Fig. 2, A and B). Ethanol doses of 0.3, 1.0, and 3.0 g/kg dose dependently induced CYP2E1 by 1.6-, 1.9-, and 2.4-fold, respectively, compared with saline-treated animals. In addition, 0.1-, 0.3-, and 1.0-mg/kg doses of nicotine for 7 days significantly increased CYP2E1 in rat liver by 1.4-, 1.8-, and 1.5-fold compared with saline controls. The apparent saturation of the level of CYP2E1 induction at the low-to-middle nicotine doses has been observed with other CYP2E1 inducers (Woodcroft and Novak, 1998).

View larger version (30K): [in this window] [in a new window]   Fig. 1.   Western blotting of hepatic CYP2E1. Representative immunoblots incubated with (+1°) and without (1°) primary antibody are shown with standard curves for A, a dilution curve of untreated rat hepatic CYP2E1 (±S.E.M., four experiments). B, immunodetectable bands were observed with 0.184 and 0.092 pmol of CYP2E1 from human lymphoblastoid-expressed rat CYP2E1 (2E1) microsomes that comigrated with 2.0 µg of pyridine-treated microsomes (pyridine) and 2.0 µg of untreated rat liver (liver) membrane preparations. Cross-reactivity of the rabbit anti-rat polyclonal antibody with 0.184 pmol of rat expressed CYP2C11 (2C11), CYP2B1 (2B1), CYP3A2 (3A2), CYP1A1 (1A1), CYP2A1 (2A1), and CYP2A2 (2A2) isozymes was assessed. Quantitative measurements in relative optical density units multiplied by pixel area (density units).

View larger version (34K): [in this window] [in a new window]   Fig. 2.   Chronic ethanol or nicotine induce CYP2E1 in rat liver. A, ethanol dose dependently induces CYP2E1 in rat liver relative to saline-treated rats; mean of four animals/group (±S.D.). B, representative immunoblot. C, low-dose nicotine induces hepatic CYP2E1 in rat liver relative to saline-treated animals; mean of four animals/group (±S.D.). D, representative immunoblot. Significant differences are indicated by *p < 0.05 and **p < 0.001.

Region-Specific CYP2E1 Induction by Behaviorally Relevant Doses of Ethanol and Nicotine. Immunohistochemical analyses clearly demonstrate that a short 7-day treatment with a 3.0-g/kg dose of ethanol markedly increase CYP2E1 immunoreactivity in the region surrounding the central vein of each liver lobule compared with saline treatment (Fig. 3, A and B). Nicotine also enhanced CYP2E1 immunostaining in the same zonal pattern (Fig. 3, C and D). Staining was undetectable when sections were processed in the absence of primary antibody (Fig. 3E).

View larger version (188K): [in this window] [in a new window]   Fig. 3.   Behaviorally relevant doses of ethanol and nicotine increased CYP2E1 immunoreactivity in the centrilobular region of rat liver. Immunohistochemical analysis demonstrates that a 7-day treatment with 3.0 g/kg ethanol (B), relative to its saline control (A), and 1.0 mg/kg nicotine (D), relative to its saline control (C), increased hepatic CYP2E1 immunostaining in the region surrounding the central vein (v) of the liver lobule. E, no staining was observed in sections processed without primary antibody. Scale bar, 500 µm.

In Vitro Kinetics of Chlorzoxazone Hydroxylation in Rat Liver. We determined protein concentrations and incubation times that resulted in the linear formation of 6-OH-CZN with less than 15% CZN disappearance. Microsomes incubated without NADPH or heat denatured rat microsomes did not produce detectable product (data not shown). Ten substrate concentrations ranging from 5 to 400 µM and Eadie-Hofstee analyses were used to determine the kinetic parameters (K_m and V_max) of CZN hydroxylation in rat liver microsomes. Kinetic analyses of CZN hydroxylation by hepatic microsomes from saline and ethanol (3 g/kg)-treated animals yielded similar K_m values of 150 ± 21 and 221 ± 65 µM, respectively (mean ± S.D., p = 0.100) (Table 1). Liver microsomes from saline and nicotine-treated (1.0 mg/kg) rats produced similar K_m affinity constants of 109 ± 22 and 131 ± 18 µM (p = 0.250) (Table 1).

Inhibition CYP2E1 Hydroxylation of Chlorzoxazone. At our estimated K_m value of 150 µM CZN, expressed rat CYP2E1 catalyzed 6-hydroxylation of CZN with a velocity of 51.7 (pmol/pmol CYP2E1/min), which was completely abolished by 100 µM aniline, a CYP2E1 competitive inhibitor, and was significantly inhibited by 31.5 µM DDC, a mechanism-based inhibitor (Fig. 4B; p < 0.01). Inhibitors were tested at concentrations equal to reported IC₅₀ values (Bourrie et al., 1996; Eagling et al., 1998). Nicotine, at 100 µM, did not inhibit expressed CYP2E1-mediated CZN metabolism.

View larger version (19K): [in this window] [in a new window]   Fig. 4.   A, representative CZN metabolism by hepatic microsomes from an individual saline-treated animal. CZN was metabolized to 6-hydroxy-CZN (3 experiments, mean ± S.E.M.) with a Km of 181 ± 25 and Vmax of 1.08 ± 0.07 nmol/mg of protein/min. Eadie-Hofstee plot is shown in the inset. B, inhibition of CZN 6-hydroxylation by expressed rat CYP2E1 by 100 µM aniline (AN) and 31.5 µM DDC, but not by 100 µM nicotine. Significant inactivation by aniline and DDC [compared with uninhibited CYP2E1 control (CN)] is indicated by **p < 0.001.

View larger version (32K): [in this window] [in a new window]   Fig. 5.   Effects of CYP2E1 inhibitors on 6-OH-CZN formation. Liver microsomes from rats treated with saline, ethanol (A), and nicotine (B) were incubated with 150 µM CZN alone (CN) or with 100 (hatched) and 500 µM (solid) aniline (AN), 31.5 (hatched) and 315 µM (solid) DDC or 10 (hatched) and 100 µM (solid) nicotine bitartrate. Each column represents mean velocities calculated for four individual animals for each treatment group (+S.D.). Significant inactivation is indicated by *p < 0.05 and **p < 0.001.

Effects of Ethanol and Nicotine on CYP2E1 mRNA Levels. Total RNA was extracted from rat livers and slot blot analysis was used to compare CYP2E1 mRNA between different treatment groups. A serial dilution of CYP2E1 cDNA was used to determine the linear range of the assay, and all samples fell within this range (Fig. 6A). No signal was detected from 1 and 10 pg of cDNAs of CYP2B6, CYP2A6, CYP2D6, and CYP3A4, indicating the specificity of the CYP2E1 probes (Fig. 6B). No significant difference in CYP2E1 mRNA levels was observed between livers from ethanol and vehicle-treated (2.24 ± 0.16 and 2.15 ± 0.22) (p = 0.56) or nicotine-treated animals compared with their saline controls (2.28 ± 0.30 and 2.19 ± 0.14) (p = 0.59) (Fig. 6, C and D); similar results were observed with blots hybridized with rat CYP2E1 oligonucleotide probes (data not shown). The -actin levels between livers from ethanol and vehicle-dosed animals (0.78 ± 0.11 and 1.01 ± 0.26) (p = 0.18) or between livers from nicotine and vehicle-dosed animals (1.21 ± 0.14 and 1.32 ± 0.30) (p = 0.55) were also not significantly different; likewise, no significant differences were observed in the levels of CYP2E1/-actin between ethanol and nicotine and their respective control groups.

View larger version (47K): [in this window] [in a new window]   Fig. 6.   Slot blotting of total RNA from livers of treated animals. A, detection of a serial dilution of CYP2E1 cDNA was linear from 0 to 9.8 pg; all samples fell in this range (inset: representative dilution curve blot). B, no cross-reactive signal from cDNA of CYP2B6 (2B6), CYP2A6 (2A6), CYP2D6 (2D6), and CYP3A4 (3A4) when loaded at 1 pg (first lane) and 10 pg (second lane). C, representative blots of CYP2E1 and -actin mRNA levels in ethanol-control (E-C), ethanol (E), nicotine-control (N-C), and nicotine (N) treatment groups. D, no significant differences in CYP2E1 mRNA levels observed between ethanol- (E) and nicotine (N)-treated animals compared with their saline vehicle controls (E-C and N-C, respectively); mean of four animals/group (+S.D.).

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

In this study, behaviorally relevant doses of ethanol and nicotine were chosen (see Experimental Procedures) to model the effects of ethanol drinking and tobacco consumption on the levels and activity of hepatic CYP2E1. We showed that low-level ethanol treatments caused a dose-dependent increase in CYP2E1 protein. We also found that short-term exposure to nicotine at low doses induces CYP2E1. Our kinetic data are the first to show an increase in hepatic CZN metabolism after low-level ethanol or nicotine treatment specifically related to induction of CYP2E1 protein. Because the enzyme's affinity for CZN was not altered, the increased V_max is consistent with an increased amount of active CYP2E1 enzyme after both ethanol and nicotine treatments (Table 1), and the increase in V_max is similar in magnitude to the observed CYP2E1 protein elevation detected by immunoblotting (Fig. 2; Table 1).

These findings further elucidate, and supported by the results from previous studies, that suggest cigarette smoke can increase CYP2E1 activity. CYP2E1 protein, activity, and mRNA levels were induced by in vivo cigarette smoke exposure in mice (Villard et al., 1998). Previous animal studies with higher doses and longer durations of nicotine administration and studies with cigarette smoke showed increased enzymatic activities in liver attributed to CYP2E1 (Bhagwat et al., 1998; Villard et al., 1998). Another animal study showed that either orally administered tobacco extract or nicotine at concentrations that model oral tobacco consumption were shown to up-regulate aniline hydroxylation (Kaur and Ali, 1982), which is predominantly mediated by CYP2E1. Cigarette smoking also significantly enhanced the metabolism of CZN in humans (Benowitz et al., 1999), reflecting increased CYP2E1 activity (Lucas et al., 1999). Among a group of nonalcoholics, female smokers exhibited significantly higher 6-OH-CZN production compared with nonsmoking females; however, not all studies examining the effects of cigarette smoke on CYP2E1 activity agree (Lucas et al., 1999). Therefore, the effects of long-term cigarette smoke or nicotine exposure in humans need to be assessed in a well designed pharmacokinetic study that minimizes the effects of any potential confounding variables (e.g., dietary constituents that modulate CYP2E1 activity).

CYP2E1 can bioactivate cigarette nitrosamines and its induction by nicotine may contribute to the development of tobacco-related cancers (Lieber, 1997; Woodcroft and Novak, 1998). Moreover, a 25 to 40% concomitant increase in reactive oxygen species generation and lipid peroxidation was observed in tissues from nicotine-treated rats (Wetscher et al., 1995), possibly mediated by nicotine-induced CYP2E1. Interestingly, the nicotine nitrosation product 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (a tobacco-smoke procarcinogen) also induced CYP2E1 activity (Bhagwat et al., 1998), and CYP2E1 has been proposed as a mediator induced by 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone-induced chromosomal aberrations (Abdel-Rahman et al., 2000). Our findings suggest that nicotine, which is present in tobacco smoke at severalfold higher concentrations than other known CYP2E1 inducers, may contribute to the enhanced levels of CYP2E1 found in smokers, which has clinical, toxicological, and possible physiological implications.

We confirmed that the enhancement in CZN hydroxylation was mediated by CYP2E1 by chemical inhibitory studies with aniline, a competitive CYP2E1 inhibitor and DDC, a mechanism-based CYP2E1 inhibitor. From the observed inhibition of CZN hydroxylation by these fairly selective CYP2E1 inhibitors, we estimate that CYP2E1 contributes to 80% or more of the CZN metabolism in microsomes from saline and drug-treated animals. Minor contributions from CYP3A and CYP1A1 that can metabolize CZN in rats may also occur (Jayyosi et al., 1995).

Constitutively expressed and ethanol-induced CYP2E1 is almost exclusively localized to the centrilobular region of each liver lobule in rats and humans (Lieber, 1999). Our findings suggest that low doses of nicotine or ethanol cause an induction of hepatic CYP2E1 that is specific to the centrilobular region, which may exacerbate the hepatotoxicity of the CYP2E1 substrates ethanol, acetaminophen, carbon tetrachloride, and N-nitrosodimethylamine in this hepatic region (Tsutsumi et al., 1989). Our immunohistochemical analyses also suggest that CYP2E1 induction, as estimated by liver homogenates used in immunoblotting and kinetic studies, may underestimate the levels of CYP2E1 attained centrilobularly in rat liver.

High-dose ethanol is a well established inducer of CYP2E1 in animal models and in human alcohol abusers (Lieber, 1999). Chronic high-dose ethanol treatment (12-13 g/kg/day) for 20 to 30 days increases CYP2E1 protein 6- to 9-fold via ligand stabilization (Eliasson et al., 1988). When a blood alcohol concentration of 300 mg/dl is attained, there is a 5- to 6-fold elevation in mRNA, which is accompanied by 12- to 13-fold induction of CYP2E1 protein in rats (Ronis et al., 1993). Our findings demonstrate that considerably lower doses of ethanol for 7 days dose dependently induce CYP2E1 protein and CYP2E1-mediated 6-OH-CZN activity in rat.

Intermittent and chronic doses of nicotine (similar to those used in this study) have been shown to increase ethanol consumption (Blomqvist et al., 1996). Some studies have suggested that central nicotinic receptors are responsible for this behavioral interaction (Blomqvist et al., 1996), whereas other findings dispute this proposal (Dyr et al., 1999) or implicate other receptor systems as the underlying mediators of this interplay (Ericson et al., 2000). Interestingly, nicotine was observed to decrease hepatic ADH activity (Bhagwat et al., 1998). We have shown that nicotine increased hepatic CYP2E1 and may account for one component of the cross-tolerance observed between ethanol and nicotine (Ericson et al., 2000). Of interest, we have also found that CYP2B1/2, the nicotine-metabolizing CYPs in rats, were induced by ethanol in the livers of the same animals used in this study and that the magnitude of induction of CYP2B1/2 was similar to that observed for CYP2E1 (2.0-, 3.0-, and 2.7-fold for 0.3-, 1.0-, and 3.0-g/kg doses, respectively) (Schoedel et al., 2001). These findings provide strong evidence that the induction of ethanol and nicotine metabolic pathways may synergistically contribute to the development of cross-tolerance.

One mechanism by which nicotine could increase CYP2E1 is by transcriptional regulation. However, we found that there were no differences in mRNA levels between livers from saline- or nicotine-treated animals, similar to the results following ethanol treatments (Fig. 6). An alternative mechanism is protein stabilization, a process whereby many CYP2E1 substrates, including ethanol and low-molecular weight ligands, increase CYP2E1 levels by slowing its high NADPH oxidase activity and its generation of reactive oxygen species (Zhukov and Ingelman-Sundberg, 1999). These reactive oxygen metabolites are thought to oxidize and modify the enzyme, labeling CYP2E1 for auto-degradation by the 26 proteosomal complex (Zhukov and Ingelman-Sundberg, 1999). Therefore, we assessed if nicotine induction involved an interaction of nicotine with CYP2E1's active site. During the short incubation time with nicotine (10 µM or 100 µM) there was no indication of an interaction between nicotine and CYP2E1 (i.e., nicotine did not inhibit CYP2E1-hydroxylation of CZN) suggesting it does not interact with the catalytic site. However nicotine may still act as a heme ligand (as does imidazole, a structurally similar compound) to stabilize the enzyme (Eliasson et al., 1988; Lieber, 1999; Zhukov and Ingelman-Sundberg, 1999). Alternatively, nicotine also bears structural similarity to pyridine, which translationally activates CYP2E1 (Lieber, 1999), and may be the mechanism underlying nicotine's induction of the enzyme. Current experiments are underway to assess 1) whether and how nicotine acts to stabilize CYP2E1, and 2) whether nicotine can increase CYP2E1 levels at lower doses than tested here. These mechanistic studies, in combination with the latter dose-response experiments that will identify lower nonsaturating doses of nicotine that "induce" the enzyme, will enable us to optimize the design of combination experiments of nicotine and ethanol to test whether their inductive effects are addictive or synergistic.

The generation of hydroxyethyl radicals, reactive oxygen species, and acetaldehyde by CYP2E1 in heavy drinkers is implicated in the pathogenesis of ALD (Lieber, 1999). In addition, alcoholics without clinical symptoms of liver disease exhibited lower CZN metabolism compared with alcoholic patients with ALD, further implicating CYP2E1 in the development of ALD (Albano et al., 1999). CYP2E1 has also been linked to the development of nonalcoholic steatohepatitis via its generation of oxidative stress (Leclercq et al., 2000). Therefore, nicotine-induced CYP2E1 may not only synergistically increase ethanol-mediated liver damage, but also increase the susceptibility to a number of nonethanol-associated liver diseases.

In vivo cigarette smoke exposure induced CYP2E1 in mice, which was associated with a concomitant increase in cigarette smoke-induced DNA strand breaks, implicating CYP2E1 in this tobacco-related toxicity. Of the tobacco constituents implicated in toxicity, chronic treatment with 0.6 mg/kg nicotine was shown to increase lipoperoxides, hydroperoxides, conjugated dienes, and free fatty acids in liver, lungs, and heart compared with control rats (Ashakumary and Vijayammal, 1996). Furthermore, ethanol and nicotine produced an additive enhancement of lipid peroxidation and depletion of antioxidants (Ashakumary and Vijayammal, 1996). Our data suggest that nicotine-induced CYP2E1 may contribute to nicotine-induced oxidative stress and free radical generation.

In summary, we showed that like chronic ethanol, low, behaviorally relevant doses of nicotine induced CYP2E1 protein and activity in the centrilobular region of rat liver. In addition to smokers, millions of nonsmokers may be exposed therapeutically to nicotine during smoking cessation therapy and, potentially, during treatment for ulcerative colitis (Guslandi, 1999), Alzheimer's disease (White and Levin, 1999), and Parkinson's disease and other neuropsychiatric movement disorders (Erdmann, 1996; Kelton et al., 2000). Chronic nicotine may contribute to the observed cross-tolerance between ethanol and nicotine by increasing the clearance of ethanol. Considering the magnitude of induction and CYP2E1's role in oxidative stress generation and procarcinogen bioactivation, the enhanced levels of CYP2E1 by nicotine exposure may also contribute to susceptibility to alcoholic and nonalcoholic liver diseases. It remains to be determined whether nicotine and ethanol synergistically induce CYP2E1; whether nicotine alone, or with ethanol, induces CYP2E1 in humans; and whether this induction of CYP2E1 is associated with any advantageous, or detrimental, effects.