Abstract
Bacillus Calmette-Guérin (BCG) and lipopolysaccharide (LPS) are well known potent activators of the cell-mediated immune system and thus lead to the decreases in cytochrome P450 (P450). In this study we used interleukin (IL)-1α/β, IL-6, or tumor necrosis factor α (TNFα) knockout (KO) mice to investigate how each cytokine is involved in P450 down-regulation, especially CYP3A11 and 2C29. BCG (40 mg/kg) was found to reduce both CYP3A11 and 2C29 mRNAs at 24 h after treatment in IL-1α/β KO mice in a manner similar to that seen in wild-type mice. CYP3A11 mRNA, but not CYP2C29 mRNA, was significantly decreased by BCG treatment in the TNFα KO mice, although the decrease was less than that of wild-type or IL-1α/β KO mice. In contrast, BCG showed no significant effect on CYP3A11 and 2C29 mRNAs in IL-6 KO mice. On the other hand, LPS was able to decrease CYP3A11 and 2C29 mRNA levels in all of the cytokine KO mice and markedly increased systemic levels of TNFα; BCG (40 mg/kg) lacked such activity. The present study has shown that IL-6 and TNFα are likely to be major factors involved in the down-regulation of CYP3A11 and 2C29 mRNAs in mice. In addition, there exist differences in the amount and/or kind of cytokines released by BCG or LPS, the latter being more potent than the former. This will be a possible reason for differential capability of P450 down-regulation between BCG and LPS.
Cytochromes P450 (P450s) are drug-metabolizing enzymes that oxidize various xenobiotics such as drugs, environmental pollutants, and chemical ingredients in food (Gonzalez, 1988). It is reported that direct damage of organs, inflammation, and infection result in down-regulation of P450 activities, which are due to activation of the immune system under diseased conditions (Morgan, 1997, 2001; Iber et al., 1999). Alterations of P450 expression can affect the pharmacokinetics of administered drugs, leading to changes in the duration and intensity of their intended efficacies. Therefore, elucidation of the underlying mechanism for P450 down-regulation following immune responses is expected to better predict pharmacokinetics of medicines, enabling better prediction of adverse effects of the medicines.
P450s produce superoxide-reactive oxygen when metabolizing their substrates (Morgan, 2001). For this reason, it is thought that P450 down-regulation defuses hyperoxidation in the case of activation in the immune system (Yasui et al., 2002). The immune system cell types are primarily leukocytes, such as lymphocytes, macrophages, and neutrophils, and upon activation, these cells release cytokines such as interleukin (IL)-1α/β, IL-6, and tumor necrosis factor α (TNFα). The release of these cytokines in turn has many effects upon other cells of the immune system. Numerous in vivo and in vitro studies show the reduction of both P450 mRNAs and proteins after treatment with the immune mediator lipopolysaccharide (LPS) or inflammatory cytokines, such as IL-1β, IL-6, and TNFα (Sewer and Morgan, 1997; Siewert et al., 2000). Since cytokines have many overlapping functions, it is very difficult to investigate the role of each cytokine exerting P450 down-regulation in whole animals. In addition, an immune stimulator, such as LPS, activates the immune system and promotes release of numerous cytokines. We previously demonstrated, using IL-1α/β and TNFα knockout (KO) mice, that LPS-mediated increase in serum cytokine concentrations and phosphorylation of mitogen-activated protein kinases (MAPKs) were different in KO mice as compared with wild-type mice (Oguro et al., 2002). In this regard, it is expected that the cytokine KO mouse would be a useful tool to examine a possible contribution of select cytokines involved in P450 down-regulation. Turpentine, a local tissue injury inducer, mimics aseptic inflammatory processes and down-regulates P450 mRNAs in mouse livers. Siewert et al. (2000) have shown that turpentine fails to down-regulate CYP1A2, 2A5, 2E1, and 3A11 mRNAs in IL-6 KO mice, suggesting an important role of IL-6 in turpentine-mediated P450 down-regulation. However, the mechanism of LPS-mediated P450 down-regulation remains to be determined.
Bacillus Calmette-Guérin (BCG) vaccine, used for tubercular prevention (McMurray, 2003) and/or medical treatment for bladder cancer (Nseyo and Lamm, 1997), is a potent nonspecific immune stimulant. BCG stimulates release of several kinds of cytokines, such as IL-1β, IL-2, IL-6, IL-12, TNFα, and interferon γ (IFNγ) (De Boer et al., 1992; O'Donnell et al., 1999). Matsuura et al. (1985) have reported suppressed microsomal P450 content and its enzymatic activities in rat liver for 15 days after BCG treatment. Therefore, it is possible that P450 content and its enzymatic activities are down-regulated by BCG administration during clinical treatment. However, the mechanism of P450 down-regulation by BCG has not been elucidated. The purpose of the present study was to investigate a possible role of cytokines in BCG-mediated P450 down-regulation. We examined the effects of BCG on P450 down-regulation in IL-1α/β, IL-6, and TNFα KO mice and compared these effects to those seen following LPS treatment in the same cytokine KO mice.
Materials and Methods
Reagents. Dried BCG vaccine was obtained from Japan BCG Laboratory (Tokyo, Japan). LPS from Escherichia coli serotype (O111:B4) was purchased from Sigma-Aldrich Japan (Tokyo, Japan). Adenosine 5′-[γ-32P]-triphosphate (5000 Ci/mmol) was obtained from Japan Isotope Association (Tokyo, Japan). Complete protease inhibitor cocktail was obtained from Roche Diagnostics (Mannheim, Germany). Anti-rat polyclonal CYP3A2 and 2C11 antibodies were kindly donated by Drs. S. Imaoka and Y. Funae, Laboratory of Chemical Biology, Osaka City University Medical School (Imaoka et al., 1987). Monoclonal antibody against signal transducer and activator of transcription (STAT) 1 (C terminus) and STAT3 was obtained from BD Biosciences Transduction Laboratories (Lexington, KY). Polyclonal antibodies against Phospho-STAT1 (Tyr701) and Phospho-STAT3 (Tyr705) were obtained from Cell Signaling Technology Inc. (Beverly, MA). All other reagents used were of the highest grade commercially available.
Animals and Treatments. All animal experiments were carried out under the control of the Regulation of the Committee of Animal Care and Welfare of Showa University. The IL-1α/β KO mice were established by Horai et al. (1998), the TNFα KO mice were established by Tagawa et al. (1997), and the IL-6 KO mice were established by Kopf et al. (1994). BALB/c mice (8 weeks old) were purchased from Japan SLC (Shizuoka, Japan) and mated with either IL-1α/β KO, IL-6 KO, or TNFα KO mice. Wild-type and KO mice progenies were selected by their respective gene expressions, and the lines for wild-type or KO mice were established. Dried BCG vaccine was suspended in saline and injected intraperitoneally at the doses indicated in the figures. LPS was dissolved in saline and injected intraperitoneally; a dose of 1 mg/kg is used for most of these P450 down-regulation studies. Control mice were injected with saline in volumes similar to that of treated mice.
RNA Extraction and Northern Blot. Total RNA was isolated from the liver of each mouse using the acid guanidine thiocyanate-phenol-chloroform extraction method as previously described (Chomczynski and Sacchi, 1987). Total RNA (20 μg) was fractionated by electrophoresis on 1.1% agarose gel, followed by transfer onto a nylon membrane. Northern blots were hybridized with 32P-labeled oligonucleotides for CYP3A11 (5′-TGTCCGATGTTCTTAGACACTGCCTTTCTG-3′, corresponding to the complement of nucleotides 1620–1649 of the published sequence in GenBank accession number X60452), CYP2C29 (5′-GGCCAGGCCCTCTCCAGCACAAATCCG TTT-3′, corresponding to the complement of nucleotides 1301–1330 of the published sequence in GenBank accession number D17674), and mouse 18S ribosomal RNA (5′-ACGGTATCTGATCGTCTTCGAACC-3′, corresponding to the complement of nucleotides 1044–1067 of the mouse gene for 18S rRNA). The hybridization levels were semiquantitated with a bio-imaging analyzer (BAS3000; Fuji Photo Film Co., Tokyo, Japan).
Preparation of Liver Microsomes. Livers were promptly removed from the body and were perfused with ice-cold 0.9% NaCl solution. The livers were homogenized immediately with 4 volumes of 1.15% (w/v) KCl solution. The homogenates were centrifuged at 9000g for 20 min, and the resulting supernatants were further ultracentrifuged at 105,000g for 90 min. The resulting microsomal pellet was suspended in 0.1 M sodium-potassium phosphate buffer (pH 7.4) containing 20% glycerol. Protein concentration was determined by the method of Lowry et al. (1951).
SDS-Polyacrylamide Gel Electrophoresis (PAGE) and Western Blotting. The relative levels of CYP3A and CYP2C proteins were detected by SDS-PAGE and Western blotting. Microsomal proteins (2.5 μg) were separated by SDS-PAGE (5% stacking gel, 10% separating gel) according to the method of Laemmli (1970) and transferred electrophoretically onto nitrocellulose membranes (Bio-Rad, Hercules, CA) at 80 mA for 50 min. Blots were immunolabeled with anti-rat polyclonal CYP3A2 and 2C11 antibodies. Peroxidase-conjugated proteins were detected with 4-chloro-1-naphthol and 0.006% H2O2 in 50 mM sodium-potassium phosphate buffer (pH 7.4).
Phosphorylation of STAT1 and STAT3. Liver was perfused with 10 mM Tris-HCl (pH 7.4) containing 1 mM EDTA, 250 mM sucrose, 1 mM p-amidinophenyl methanesulfonyl fluoride hydrochloride, 2 μg/ml aprotinine, 10 mM p-nitrophenyl phosphate, 20 mM β-glycerophosphate, 50 μM sodium orthovanadate, and 10 mM sodium molybdate and homogenized with 10 volumes of the homogenizing buffer (20 mM Tris-HCl, pH 7.4, containing 150 mM NaCl, 1% nonidet P-40, 100 μg/ml DNase I, 50 μg/ml RNase A, 1 mM p-amidinophenyl methanesulfonyl fluoride hydrochloride, 10 mM p-nitrophenyl phosphate, 20 mM β-glycerophosphate, 50 μM sodium orthovanadate, 10 mM sodium molybdate, and Complete protease inhibitor cocktail). The homogenized solution was centrifuged at 15,000 rpm for 20 min, and the supernatant was used for the immunoblot analysis. Protein concentration was determined by the method of Bradford (1976), using a commercially available protein reagent (Bio-Rad). The supernatant fraction (25 μg of protein) was solubilized in 2% SDS, and the proteins were separated by SDS-PAGE (5% stacking gel, 8% separating gel). After electrophoresis, the proteins were transferred to polyvinylidene difluoride membranes (Japan Genetics Co., Tokyo, Japan) at 80 mA for 50 min. Western blots were performed using monoclonal antibody against STAT1 (C terminus) or STAT3 and polyclonal antibodies against Phospho-STAT1 (Tyr701) or Phospho-STAT3 (Tyr705). Signal was detected using chemiluminescence (ECL; Amersham Biosciences UK, Ltd., Little Chalfont, Buckinghamshire, UK). Molecular weight was calculated with prestained SDS-PAGE standards (Bio-Rad) applied to the same gel run samples.
Hepatic Total P450 Content. Hepatic total P450 content was determined from a carbon monoxide difference spectrum of dithionite-treated microsomes dissolved in microsome solubilization buffer (0.1 M sodium-potassium phosphate buffer, pH 7.4, containing 20% glycerol) and recorded with a UV-300 spectrophotometer (Shimadzu, Tokyo, Japan) using a difference extinction coefficient (450–490 nm) of 91 mM -1 cm-1 as described by Omura and Sato (1964).
Testosterone 6β-Hydroxylase Activity. The method of Van der Hoeven (1984) was used to measure CYP3A activity. The reaction mixture contained 1 mg/ml liver microsomes, 1 mM NADPH, 3 mM MgCl2, 50 mM potassium phosphate buffer (pH 7.4), and 60 μM testosterone in a final volume of 1000 μl. The reaction mixture was incubated at 37° C for 10 min, and the reaction was terminated by cooling on ice. Methyltestosterone (100 μl; 5 nmol/ml in methanol) was added to the samples as an internal standard and extracted with 6 ml of dichloromethane. The mixture was centrifuged at 1600g for 10 min at 4° C. Organic solvent layer was isolated, and the solvent was evaporated under nitrogen gas stream at 45° C. The residue was dissolved in 100 μl of methanol and 100 μl of deionized water. A 50-μl aliquot of each sample was injected into the high-performance liquid chromatography (HPLC) apparatus and analyzed as described below. The determination of testosterone 6β-hydroxylase activity was carried out by using an HPLC-UV assay method. The HPLC system consisted of a model CCPM-α pump (TOSO Co., Tokyo, Japan), a model UV-8020 detector (TOSO), a model AS-8010 autosampler (TOSO), a model D-2500 integrator (Hitachi Co., Tokyo, Japan), and a 4.6× 250 mm CAPCELL PAK C18 SG120 column (Shiseido Co., Tokyo, Japan). The mobile phase consisted of methanol, H2O, and acetonitrile (39:60:1, v/; solvent A), and methanol, H2O, and acetonitrile (80:18:2, v/v; solvent B). A linear gradient from 90% solvent A to 10% solvent B was operated over 40 min at 1 ml/min. The eluate was monitored at 254 nm. Calibration curves were prepared from 1 to 10 μM by processing the authentic standard substances throughout the entire procedure. Under the present chromatographic conditions, 6β-hydroxytestosterone, testosterone, and methyltestosterone were eluted at 11.3, 25.3, and 28.9 min, respectively. 6β-Hydroxytestosterone was quantified by comparison with the standard curves, using the peak height ratio method.
Cytokine Concentrations. Serum IL-1β, IL-6, and TNFα concentrations were determined using the respective enzyme-linked immunosorbent assay kits (IL-1β, IL-6, and TNFα; Pierce Endogen, Rockford, IL).
Statistical Analysis. Data were subjected to statistical analysis by the Kruskal-Wallis nonparametric analysis of variance test, followed by Dunn's multiple comparison test. The accepted level of significance was set at p < 0.05.
Results
Down-Regulation of CYP3A11 and 2C29 mRNAs by BCG in Wild-Type Mice. We first examined time course- and dose-dependent effects of BCG on constitutive-type P450, CYP3A11, and 2C29 mRNAs, in wild-type mouse livers (Fig. 1). As shown in Fig. 1A, BCG (40 mg/kg) decreased CYP3A11 mRNA at 12 h after treatment, reached the bottom level at 24 h (26% of the controls), and then returned to control levels by 48 h. Although there was no significant difference, BCG administration also resulted in the decrease of CYP2C29 mRNA, but less than that seen in CYP3A11 mRNA under the present experimental conditions (Fig. 1). Dose-dependent effects of BCG on P450 are shown in Fig. 1B. BCG at a dose of 10 mg/kg did not produce any changes in either CYP3A11 or CYP2C29 mRNA when examined 24 h after administration. BCG dose dependently decreased CYP3A11 mRNA expression at doses of 20 mg/kg (25% of the controls) and 40 mg/kg (4.7% of the controls), and also significantly decreased CYP2C29 mRNA at a dose of 80 mg/kg (22% of the controls). BCG differentially altered CYP2E1 mRNA (increase and/or decrease) depending on the dose and time examined under the present experimental conditions (data not shown).
Change of CYP3A and 2C Proteins by BCG in Wild-Type Mice. Since the down-regulation of either CYP3A11 or 2C29 mRNA was observed after BCG administration, we examined changes in P450 protein levels in liver microsomes by Western blot analysis to examine time course effects (Fig. 2). Remarkable reductions of both CYP3A and CYP2C proteins were seen at 24 h after BCG administration. In addition, dose-dependent decreases of CYP3A and CYP2C protein levels produced by BCG were similar to those of their mRNAs, as shown in Fig. 1.
Suppression of 6β-Hydroxylase Activity after BCG Treatment of Wild-Type Mice. Matsuura et al. (1985) have reported that a single dose of BCG (40 mg/kg i.v.) decreases both total content and enzymatic activities of rat liver P450 up to 15 days after treatment. In contrast, we found that the BCG-mediated down-regulation of CYP3A11 mRNA and CYP3A protein in mouse livers returned to control levels within 48 h after administration. Therefore, we determined testosterone 6β-hydroxylase activity in mouse liver microsome (Fig. 3). Testosterone 6β-hydroxylase activity, a marker for CYP3A activity, was decreased to 25% of the control level at 24 h after BCG (40 mg/kg i.p.) treatment and sustained for 3 days. Since BCG was able to decrease CYP3A11 mRNA, CYP3A protein, and testosterone 6β-hydroxylase activity, we next examined the effect of BCG on CYP3A11 and 2C29 mRNAs in various cytokine KO mice. For comparison, we also investigated the effect of LPS on CYP3A11 and 2C29 mRNA in these mice.
Effect of BCG and LPS on CYP3A11 and 2C29 mRNAs in Various Cytokine KO Mice. LPS causes endotoxin shock and is a known down-regulator of various P450 mRNAs. LPS also stimulates release of numerous cytokines into the circulation from cells of the immune system such as macrophage. However, the role of each cytokine involved in P450 down-regulation mediated by LPS is not clear. Since we chose to focus on CYP3A rather than on CYP2C29, the dosage of 40 mg/kg was chosen for BCG. A 1 mg/kg dose of LPS was used because this dosage has been used to study P450 down-regulation. We increased the number of mice (12 each of wild-type and IL-6 KO mice; 5 each of IL-1α/β KO and TNFα KO mice) in each group and investigated the CYP3A11 and 2C29 mRNA expression by BCG administration. Individual differences for expression of CYP3A11 and 2C29 mRNAs under physiological conditions in both wild-type and KO mice tend to be generous; however, no statistical significance was seen (data not shown). As shown in Fig. 4, BCG markedly decreased CYP3A11 mRNA in IL-1α/β KO mice (20% of the IL-1α/β KO control mice) as well as that in wild-type mice (21% of the wild-type control mice). CYP3A11 mRNA in TNFα KO mice was also significantly decreased by BCG treatment (46% of the TNFα KO control mice), although the reduced level was less than that observed in wild-type or IL-1α/β KO mice. BCG administration resulted in the significant down-regulation of CYP2C29 mRNA in wild-type (52% of the controls) and IL-1α/β KO (35% of the controls) mice. BCG reduced CYP3A11 mRNA more than it does 2C29 mRNA in the livers of IL-1α/β KO and TNFα KO mice, an observation similar to that seen in wild-type mice. Interestingly, BCG failed to produce significant changes of CYP3A11 and CYP2C29 mRNAs in IL-6 KO mouse livers and CYP2C29 mRNA in TNFα KO mouse livers when compared with the respective controls. In contrast, LPS (1 mg/kg i.p.) markedly reduced CYP3A11 and 2C29 mRNAs in wild-type and all of the cytokine KO mice.
Effect of BCG and LPS on Hepatic P450 Content in Wild-Type and IL-6 KO Mice. BCG, but not LPS, failed to produce significant changes in CYP3A11 and 2C29 mRNAs in IL-6 KO mouse livers. We examined hepatic P450 content in wild-type and IL-6 KO mice after treatment with BCG or LPS (Fig. 5) and found that BCG significantly decreased hepatic P450 content in wild-type mice, but not in IL-6 KO mice. In contrast, LPS decreased P450 content at 24 h after treatment in both wild-type and IL-6 KO mouse livers.
Effect of BCG on Testosterone 6β-Hydroxylase Activity and CYP3A Protein in Wild-Type and IL-6 KO Mice. As shown in Fig. 4, BCG failed to produce significant changes in CYP3A11 mRNA in IL-6 KO mouse livers. We examined the effect of BCG on CYP3A protein levels and its reflected testosterone 6β-hydroxylase activity (Fig. 6). BCG decreased both testosterone 6β-hydroxylase activity and CYP3A protein in wild-type mouse livers. As expected, testosterone 6β-hydroxylase activity and CYP3A protein were not significantly decreased by BCG treatment in IL-6 KO mouse livers, just like that observed in CYP3A11 mRNA.
Effect of BCG and LPS on STAT1/3 Phosphorylation in Wild-Type and IL-6 KO Mice. STAT1 and STAT3 belong to the STAT family, which is considered to be involved in IL-6-mediated signal transduction. When a ligand combines with a receptor, Jaks associated with the receptor become activated and phosphorylation of the cytoplasmic area of gp130 occurs, thereby creating docking sites for STAT factors, such as STAT1 and STAT3. STATs then become phosphorylated and form homo- or heterodimers, and translocate into the nucleus, where they contribute to regulation of gene transcription. BCG failed to down-regulate CYP3A11 and 2C29 mRNAs in IL-6 KO livers, suggesting that IL-6 plays a key role in BCG-mediated effects on the P450 genes in mouse livers. Therefore, we examined phosphorylation of STAT1/3 in wild-type mouse livers treated with BCG or LPS. BCG administration resulted in the marked phosphorylation of both STAT1 and STAT3, 1 h later in wild-type mouse livers. Phosphorylated STAT3 slowly abated thereafter, and the band was still visible at 4 h after BCG administration (data not shown). On the other hand, phosphorylation of STAT1 was very transient and was undetectable at 2 h after BCG administration (data not shown). As shown in Fig. 7, BCG completely failed to phosphorylate STAT1/3 in IL-6 KO mice. As expected, LPS produced phosphorylation of STAT1/3 in wild-type animals. LPS also did not phosphorylate STAT3 in IL-6 KO mouse livers. However, LPS slightly increased phosphorylation of STAT1, although it was weaker than that in the wild-type mouse livers.
Serum IL-1β, IL-6, and TNFα Concentrations in Wild-Type and Various Cytokine KO Mice Treated with BCG or LPS. As shown in Fig. 7, there was a difference in STAT1 phosphorylation between BCG and LPS in wild-type and IL-6 KO mice. Although the effect of LPS on STAT1 phosphorylation in IL-6 KO mice was weak, CYP3A11 and 2C29 mRNAs were markedly down-regulated by LPS in IL-6 KO mice, similar to that observed in wild-type mice. Serum IL-1β, IL-6, and TNFα concentrations were measured in wild-type, IL-1α/β KO, IL-6 KO, and TNFα KO mice at 1 h after BCG or LPS administration (Fig. 8). Neither BCG nor LPS administration significantly elevated serum IL-1β concentration. Both BCG and LPS significantly increased serum IL-6 concentrations in wild-type, IL-1α/β KO, and TNFα KO mice, with LPS showing more potency than BCG. Additionally, serum TNFα concentration was strongly increased in wild-type, IL-1α/β KO, and IL-6 KO mice after LPS administration, but not after BCG at the dosage used in this study.
Discussion
In this study, we examined the possible roles of inflammatory cytokines in BCG-mediated CYP3A11 and 2C29 mRNA down-regulation and differences in P450 down-regulation caused by BCG and LPS. BCG is clinically used in the prevention of tuberculosis (McMurray, 2003) and in therapeutic regimens for treatment of bladder cancer (Nseyo and Lamm, 1997). In this study, since BCG was used at a high dose, as compared with typically lower clinical doses, there is a low probability that P450 activities of the liver are suppressed by BCG in a clinical setting. However, relatively infrequent systemic, disseminated BCG infection has been reported in clinical settings (Gonzalez et al., 2003), in which case there might be a decrease in the hepatic P450 activities of these patients.
We found that BCG down-regulated CYP3A11 mRNA expression in wild-type, IL-1α/β KO, and TNFα KO mouse livers and down-regulated CYP2C29 mRNA expression in wild-type and IL-1α/β KO mouse livers. Since BCG lacked the ability to decrease CYP3A11 and 2C29 mRNAs in IL-6 KO mouse livers, IL-6 appears to play a key role in down-regulation of these P450 mRNAs. Likewise, turpentine has been shown to lose its CYP3A11 down-regulatory potency in IL-6 KO mice (Siewert et al., 2000). The findings suggest that IL-6 seems to play a general role in CYP3A11 mRNA down-regulation produced by BCG, a bacterial immune activator, and turpentine, an aseptic inflammatory inducer, despite their differential action on mouse livers.
Signal transduction to the nucleus by IL-6 stimulation is carried out via activation of the gp130/Jak/STAT1/3 (Heinrich et al., 1998). In this study, we showed that BCG actually produced marked phosphorylation of STAT1/3 in wild-type mouse livers. As was expected, BCG lost its ability to phosphorylate both STAT1 and STAT3 in IL-6 KO mouse livers. Pan et al. (2003) reported that CYP3A11 and 2C29 mRNAs are markedly down-regulated by LPS in STAT1 KO mice and in wild-type mice. Thus, STAT1 is unlikely to be involved in LPS-induced CYP3A11 and 2C29 mRNA down-regulation. Accordingly, BCG-induced P450 mRNA down-regulation seems to be involved in STAT3, but not STAT1, activation. However, it was also reported that CYP3A4 down-regulation by IL-6 does not proceed through the Jak/STAT pathway, and IL-6 down-regulates CYP3A4 through translational induction of the liver-enriched transcriptional inhibitory protein of CCAAT/enhancer binding protein-β, which competes with and antagonizes constitutive CCAAT/enhancer binding protein transactivators (Jover et al., 2002). Thus, it remains to be determined whether translocation of STAT1/3 into the nucleus is involved in down-regulation of CYP3A11 and 2C29 mRNAs following BCG administration.
Moreover, since BCG-mediated CYP2C29 mRNA down-regulation in TNFα KO was not so marked as compared with that seen in wild-type mice, TNFα might also be involved in the P450 mRNA down-regulation, especially in CYP2C29 mRNA. However, TNFα was not detected in all mice treated with BCG at a dose of 40 mg/kg. Although it is difficult to explain this phenomenon, it is suggested that BCG released a small amount of TNFα, which may be involved in CYP3A11 and 2C29 mRNA down-regulation in wild-type mouse livers. BCG at a dose of 80 mg/kg released TNFα at approximately 100 pg/ml in serum (data not shown).
This study shows that LPS down-regulated CYP3A11 and CYP2C29 mRNAs in IL-6 KO mice. As shown in Fig. 8, in our current study using IL-6 KO mice, TNFα was released after LPS administration but not after BCG at the dosage used. It is generally accepted that the TNFα signal is transduced by nuclear factor κB. Morgan et al. (2002) reported that the CYP2C11 gene contains a binding site for the cytokine-activated transcription factor nuclear factor κB that confers down-regulation of a linked reporter gene by cytokines. A possible involvement of IFNγ and MAPKs, which play roles in signal transduction of IL-1 or TNFα, has been suggested. However, this possibility will be low because serum IFNγ was not increased by LPS administration, and the magnitudes of phosphorylation of MAPKs showed no differences between wild-type and IL-6 KO mouse livers (data not shown). Nevertheless, we observed that the phosphorylation of LPS-mediated MAPKs was stronger than that of BCG (data not shown). Taken together, these findings suggest that CYP3A11 and 2C29 mRNAs are down-regulated by LPS in IL-6 KO mice and that this down-regulation involves TNFα. This study showed that the effect of LPS on CYP3A11 and 2C29 mRNA down-regulation in wild-type mice was stronger than that of BCG. Since TNFα was generated by LPS, but not by BCG, it is suggested that the difference between the effect of LPS and that of BCG is dependent upon TNFα levels.
BCG-mediated suppression of CYP3A enzyme activity lasted even after the mRNA returned to control levels. It has been shown that P450 enzyme activity is inhibited by nitric oxide (Morgan, 1997; Takemura et al., 1999). BCG is known to produce nitric oxide via inducible nitric-oxide synthase (Zhang et al., 2002). These observations suggest that suppressed enzyme activity in the early and late period after BCG treatment is probably due to tissue gases produced during BCG treatment.
P450 is a heme-containing protein, and therefore, P450 content of tissues is influenced not only by protein synthesis, but also by heme synthesis and degradation. Thus, induction of heme oxygenase-1 (HO-1), a rate-limiting enzyme for heme degradation, by LPS (Oguro et al., 2002) or BCG (data not shown) treatment resulted in the decreases in P450 content (Kobayashi et al., 2000). In addition, an electron produced by NADPH-P450 reductase-mediated reductive reaction is necessary to activate both P450 and HO (Dulak and Jozkowicz, 2003) and may present an ever further rate-limiting step in metabolism. It has been shown that P450 enzyme activity is inhibited by carbon monoxide, which is a degradation product of heme (Morgan, 1997). Based on these findings, it is expected that HO-1 induction consumes more electrons, which may further lead to decrease of P450 activity.
In conclusion, the present study has revealed that BCG-mediated down-regulation of CYP3A11 and 2C29 mRNAs is due to the BCG-induced increase of circulating IL-6 concentrations. This event probably leads to suppression of CYP3A enzyme activity. Furthermore, TNFα plays an important role in LPS-mediated P450 down-regulation in IL-6 KO mice. However, it is not clear whether the cytokine-induced P450 mRNA down-regulation reflects a transcriptional or a post-transcriptional step.
Acknowledgments
We thank S. Kenmotsu and Y. Arima for technical assistance. We also thank Dr. Manfred Kopf (Molecular Biomedicine, Swiss Federal Institute of Technology, Zurich, Switzerland) for kindly supplying IL-6 KO mice, and Drs. S. Imaoka and Y. Funae (Laboratory of Chemical Biology, Osaka City University Medical School, Osaka, Japan) for kindly providing anti-rat polyclonal CYP3A2 and 2C11 antibodies.
Footnotes
-
A part of this study was supported by a grant-in-aid from the Ministry of Education, Science and Culture, Japan.
-
ABBREVIATIONS: P450, cytochrome P450; IL, interleukin; BCG, bacillus Calmette-Guérin; LPS, lipopolysaccharide; TNF, tumor necrosis factor; IFN, interferon; KO, knockout; STAT, signal transducer and activator of transcription; MAPK, mitogen-activated protein kinase; PAGE, polyacrylamide gel electrophoresis; HPLC, high-performance liquid chromatography; HO, heme oxygenase.
- Received December 12, 2003.
- Accepted March 23, 2004.
- The American Society for Pharmacology and Experimental Therapeutics