Abstract
The role of nitric oxide in the inhibition of the cytochrome P450 system produced by interferon-γ in human hepatocytes has been examined. Nitric oxide exogenously released from S-nitroso-N-acetylpenicillamine produced a dose-dependent decrease in cytochrome P4501A2 activity, assessed as 7-ethoxy resorufin O-deethylation. After 24 hr of treatment with 300 U/ml interferon-γ, a rise in nitric oxide release (200% over control cells) and a parallel inhibition in 7-ethoxyresorufin O-deethylase activity (50% of control) were observed in human hepatocytes. This inhibition was concentration-dependently prevented by NG-monomethyl-l-arginine, a competitive inhibitor of nitric oxide biosynthesis. Comparable results were observed for cytochrome P4502A6 (7-coumarin hydroxylation), 2B6 (7-benzoxyresorufin O-dealkylation) and 3A4 (testosterone 6β-hydroxylation) activities. Decreases in CYP1A2 activity found after exposure of 3-methylcholanthrene-treated hepatocytes to interferon-γ were also reversed in the presence of NG-monomethyl-l-arginine. Down-regulation of cytochrome P4501A2 and 3A4 expression by interferon-γ was observed in parallel. This study suggests that at least some of the interferon-γ effects on human hepatocyte cytochrome P450 isoenzymes are mediated by nitric oxide biosynthesis.
It has long been recognized that IFN, either administered in its recombinant form or produced in vivo in response to a variety of stimuli, down-regulated the expression of hepatic CYP enzymes in rodents (Renton and Knickle 1990;Morgan and Norman 1990; Craig et al., 1992). However, some important differences in vivo and in vitro in the down-regulatory effects on specific CYPs of INF-α and IFN-γ have been reported (Craig et al., 1990; Chen et al., 1995). In addition, clinical observation indicated that giving IFN to patients resulted in a reduction in the clearance of antypirine (Williams and Farrell 1986; Brockmeyer et al., 1992), theophylline (Williams et al., 1987; Israel et al., 1993) and erythromycin (Craig et al., 1993) and a decrease in hepatic monooxygenase activities (Okuno et al., 1990, 1993).
Direct inhibitory effects of INF-α and IFN-γ on hepatic CYP-dependent drug metabolism was demonstrated in primary cultures of human hepatocytes (Donato et al., 1993a; Abdel-Razzaket al., 1993, 1994). This phenomenon was also observed after treatment of human hepatocytes with other cytokines (interleukin-1β, interleukin-6, tumor necrosis factor-α and transforming growth factor-β) (Abdel-Razzak et al., 1993, 1994;Muntané-Relat et al., 1995). However, the mechanism by which IFN modulates drug-metabolizing enzymes in human hepatocytes remains unclear. Different mechanisms for the suppression of CYP enzymes by IFNs and other cytokines have been proposed. They include an induction of heme oxygenase activity, which generates oxygen radicals that destroy CYP apoproteins (Ghezzi et al., 1985; Manneringet al., 1988), and a decrease in CYP mRNA levels and subsequent synthesis of CYP apoproteins (Craig et al., 1990;Sakai et al., 1992; Cribb et al., 1994). Recent studies suggest that NO could be involved in the inhibition of CYP1A1 and 2B1 activities in rat hepatocytes (Khatsenko et al., 1993; Wink et al., 1993; Stadler et al., 1994;Carlson and Billings 1996). NO is a short-lived mediator that is synthesized from l-arginine in a variety of cell types, producing important metabolic changes in target cells (Moncada et al., 1991; Star, 1993). It has been demonstrated that hepatocytes, as with other cells, produce NO in vivo during chronic inflammation (Billiar et al., 1990) and that rat and human hepatocytes in culture increased NO production in response to a combination of inflammatory cytokines (Nussler et al., 1992,1994; Geller et al., 1993). It has also been suggested that the decrease in CYP1A1 activity produced by incubation of rat hepatocytes with a mixture of lipopolysaccharide and cytokines could be primarily due to functional inhibition of CYP by NO (Stadler et al., 1994). The purpose of our study was to investigate the effects of NO on the inhibition of the CYP system produced by IFN-γ in human hepatocytes. Specific monooxygenase activities were used as probes of different CYP isozymes, and the relative levels of specific CYP apoproteins and mRNAs were also determined. The results of this work provide evidence indicating that the reduction in CYP activities induced by IFN-γ in human hepatocytes is due, in part, to NO inactivation of CYP.
Material and Methods
Materials.
Human recombinant IFN-γ, 7-benzoxyresorufin, 7-ethoxyresorufin, 7-pentoxyresorufin, collagenase and β-glucuronidase/arylsulfatase were purchased from Boehringer Mannheim (Mannheim, Germany); SNAP was obtained from Research Biochemicals International (Natick, MA); NMA, MC, coumarin, 7-hydroxycoumarin, resorufin and testosterone were purchased from Sigma Chemical Co. (St. Louis, MO); 6β-hydroxytestosterone was supplied by Steraloids Inc. (Wilton, NH); trans 35S-label (specific activity 1138 Ci/mmol) was obtained from ICN Pharmaceuticals Inc (Irvine, CA); newborn calf serum was obtained from Gibco (Paisley, UK); Ham’s F-12 and Leibovitz L-15 culture media were from Flow (Irvine, Stoctland, UK); RPMI 1640 methionine-free medium was purchased from Seromed (Berlin, Germany); all other reagents used in this study were of analytical grade.
Isolation and culture of human hepatocytes.
Surgical liver biopsies (1–5 g) were taken from patients undergoing cholecystectomy after informed consent was obtained. Patients had no known liver pathology nor did they receive medication during the weeks before surgery. None of the patients was habitual consumers of alcohol or other drugs. A total of 15 liver biopsies (5 males and 10 females) were used. Patients’ ages ranged from 26 to 73 yr (table 1). Human hepatocytes were isolated using a two-step perfusion technique (Gómez-Lechón et al., 1990) and seeded on 24-well or on 3.5-cm diameter fibronectin-coated plates (3.6 μg/cm2) at a density of 8 × 104cells/cm2 in an appropriate volume of medium. Culture medium was Ham’s F-12/Leibovitz L-15 (1/1, v/v) supplemented with 2% newborn calf serum, 5 mM glucose, 50 U/ml penicillin, 50 μg/ml streptomycin, 0.2% bovine serum albumin and 10− 8 M insulin. Medium was changed 1 hr later to remove unattached hepatocytes. By 24 hr cultures were shifted to serum-free medium containing 10− 8 M dexamethasone.
Characteristics of donors and of liver cell preparations
Treatment of cultures.
Treatments were started by 24 hr of culture after medium renewal, and cells were incubated with the different agents for 24 hr. SNAP (0.01–2 mM) was added to cultures dissolved in dimethyl sulfoxide. The final concentration of the solvent in culture medium was less than 0.5% (v/v) and the control cultures were treated with the same amount of solvent. Unless otherwise indicated, human hepatocyte cultures were exposed to 300 U/ml IFN-γ. NMA (0.001–1 mM) was used as competitive inhibitor ofl-arginine-dependent NO biosynthesis. For CYP induction experiments, MC was dissolved in dimethylsulfoxide and added to 24-hr-old cultured human hepatocytes at a final concentration of 2 μM (concentration of solvent in culture medium was 0.5%, v/v).
Determination of nitrite concentration.
To determine the amount of NO synthetized by hepatocytes, the culture supernatants were assayed for nitrite content, as a stable end product of NO oxidation. Nitrite accumulation was measured by adding 100 μl of culture supernatant to 100 μl of Griess reagent (Stein and Strejan, 1993). The assay was performed in 96-well plates and the absorbance at 540 nm was measured in a microplate reader.
Monooxygenase activity assays.
EROD, PROD and BROD activities were assayed by incubating intact hepatocytes cultured on 24-well culture plates with 8 μM 7-ethoxyresorufin, 15 μM 7-pentoxyresorufin or 15 μM 7-benzoxyresorufin, respectively, and the resorufin formed was quantified fluorimetrically as described (Donatoet al., 1993b). CH activity was measured directly in intact hepatocytes cultured on 24-well culture plates incubated with 100 μM coumarin for 30 min at 37°C. Aliquots of 200 μl of medium supernatants were incubated with 40 U of β-glucuronidase and 30 U of arylsulfatase in 50 μl of 0.1 M sodium acetate buffer (pH 4.5). After 2 hr of incubation at 37°C, each sample was diluted (1:3) in 0.1 M Tris pH 9. The 7-hydroxycoumarin formed was quantified fluorimetrically by means of a Cytofluor 2350 microplate reader (Millipore Iberica, Barcelona, Spain) using 355 and 460 nm excitation and emission filters, respectively. 6β-OHT activity was measured by incubating intact hepatocytes cultured on 24-well plates for 30 min with 300 μl of culture medium containing 250 μM testosterone. Metabolites were extracted and analyzed by high-performance liquid chromatography as described (Donato et al., 1993a). All activity values are expressed as pmol of product formed per mg of total cellular protein and per min. Cellular protein was measured according to the method ofLowry et al. (1951).
Western blot analysis and immunoprecipitation assays.
Polyclonal antibodies against recombinant CYP1A2 and CYP3A4 were kindly provided by Dr. F. P. Guengerich (Nashville, TN). Liver S-9 fractions (30 μg protein/lane) from human cultured hepatocytes were electrophoresed in a sodium dodecyl sulfate-polyacrylamide gel. Proteins were transferred to Immobilon membranes (Millipore) and sheets were incubated with rabbit antiserum raised against recombinant CYP3A4. After washing, blots were developed with horseradish peroxidase-labeled goat anti-rabbit IgG, using 0.05% diaminobenzidine (w/v) and 0.001% H2O2 (v/v) in phosphate buffered saline. The relative intensities of the bands were estimated from densitometric analysis of the blot with an image analyzer (Visilog 3).
For immunoprecipitation assays, hepatocytes cultured on 24-well plates were pulse-labeled for 4 hr with 50 μCi/ml of [35S]-methionine (trans 35S-label) in methionine-free RPMI-1640 culture medium. Radiolabeled CYP isozymes present in the cellular lysate were immunoprecipitated with specific polyclonal antibodies against recombinant CYP1A2 and CYP3A4, subjected to sodium dodecyl sulfate-polyacrylamide electrophoresis under reducing conditions and to fluorography. Radioactivity in the individual protein bands was counted.
Analysis of mRNA by semiquantitative RT-PCR.
Total cellular RNA was extracted as described (Chomczynski and Sacchi, 1987). RNA (1 μg) was reverse transcribed by incubating for 60 min at 37°C in 30 μl of 50 mM Tris-HCl (pH 8.3) containing 75 mM KCl, 10 mM dithiothreitol, 3 mM MgCl2, 200 μM each deoxynucleotide triphosphate, 300 U Moloney Murine leukemia virus reverse transcriptase (Gibco BRL, Grand Island, NY), 30 U RNAsin (Promega, Madison, WI) and 2 μM oligo-dT14 (A/C/C) primer. The reaction was stopped by heating at 95°C for 5 min. For CYP3A4 cDNA (Gonzalez et al., 1988) the forward primer was from 1353 to 1379 nt (5′-CCT TAC ACA TAC ACA CCC TTT GGA AGT-3′) and the reverse primer was from 1705 to 1734 nt (5′- AGC TAC ATB CAT GTA CAG AAT CCC CGG TTA-3′). For human β-actin cDNA (Ponte et al., 1984) the forward primer was from 480 to 499 nt (5′-CGT ACC ACT GGC ATC GTG ATT-3′) and the reverse primer from 911 to 931 nt (5′-GTG TTG GCG TAC AGG TCT TTG-3′). Diluted cDNA (3 μl) was amplified in 30 μl of 10 mM Tris-HCl (pH 9) containing 50 mM KCl, 1.5 mM MgCl2, 50 μM each deoxynucleotide triphosphate, 1 U AmpliTaq DNA polymerase (Perkin Elmer, Norwalk, CT) and 0.2 μM of each primer. After denaturing for 4 min at 94°C, amplification was performed by 27 cycles of 45 sec 94°C, 45 sec 60°C and 45 sec 72°C, and a final extension of 5 min at 72°C. For quantitative analysis, 0.1 μCi of α32P-dATP (3000 Ci/mmol, Amersham, Buckinghamshire, England) was included in the PCR reaction. Aliquots (20 μl) of the PCR reaction were subjected to electrophoresis on 1.2% agarose gel and exposed to phosphor storage screens for up to 1 hr. After scanning the screens with a Posphorimager, the intensity of the bands was quantified with the Imagequant software package (Molecular Dynamics, Sunnyvale, CA). The results were plotted on a log-log scale against the dilution factor of the cDNA. The PCR reaction was considered to be exponential if 2-fold amplification was detected. PCR results were normalized by analysis of β-actin from the same cDNA dilution series.
Statistical analysis.
Data were analyzed by using the Student’s t test. Values of P < .05 were considered as significant.
Results
Role of NO in inhibition of CYP1A2 activity by IFN-γ.
The EROD activity of freshly isolated hepatocytes (6.1 ± 1.4 pmol/mg/min, n = 3) decreased during the first 24 hr in culture to 43% (2.6 ± 0.4 pmol/mg/min, n = 3). After this decrease, probably due to the adaptation of cells to culture conditions, the enzyme was stabilized, and 37% (2.3 ± 0.3 pmol/mg/min, n = 3) of the initial EROD activity was still detected after 72 hr of culture. Therefore, our standard procedure for studying IFN-γ effects on CYP activity is to maintain cells in control culture conditions for 24 hr before starting treatments with IFN-γ. After 24 hr of incubation of human hepatocytes with increasing concentrations of IFN-γ, both an increase in nitrite levels in the culture medium and a decrease in CYP1A2 (assessed as EROD) activity were observed (fig. 1). Maximal EROD inhibition (60% of the activity of untreated cells) was produced by 300 U/ml of IFN-γ. Parallel to this decrease, the nitrite concentration in the culture medium was doubled by IFN-γ. At this point, a series of experiments was conducted to investigate the possible correlation between the effects observed on EROD activity and the induction of NO production produced by IFN-γ. To examine the inhibitory effects of NO on CYP1A2 activity, hepatocytes were incubated with SNAP, a compound that spontaneously generates NO in culture medium, and EROD was measured. Figure 2 shows the nitrite levels and EROD activity measured 24 hr after addition of different SNAP concentrations to the culture medium. Addition of SNAP to human hepatocyte cultures resulted in a rise in the nitrite concentration in the culture medium. Parallel to this, a concentration-dependent inhibition of EROD activity was observed, and the highest effect (30% of activity of untreated cells) was reached with 2 mM SNAP. Beyond this concentration, cytotoxicity was clearly observed (50% of control, SNAP 5 mM, assessed by 3-[4, 5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide test). The effects of blocking NO production on the inhibition of EROD activity produced by IFN-γ were studied in human hepatocytes treated with NMA, a competitive inhibitor of l-arginine in NO biosynthesis, both in absence and presence of 300 U/ml IFN-γ (fig.3). As expected, the basal as well as IFN-γ-induced NO formation decreased when NMA was added to the incubation medium. A complete inhibition of NO biosynthesis was produced by 300 μM NMA, both in controls and IFN-γ-treated cells (fig. 3A). NMA by itself did not produce any alteration in EROD activity, but it partially reversed the inhibitory effect of IFN-γ on EROD (fig. 3B). NMA did not fully restore EROD activity and IFN-γ continue to produce about a 20% reduction (vs. cells not treated with IFN) in EROD even in the presence of 300 μM NMA.
Concentration-dependent effects of IFN-γ on endogenous NO synthesis and EROD activity. After 24 hr in culture, human hepatocytes were exposed to increasing concentrations of IFN-γ for an additional 24 hr. NO production (•) was quantified by measuring nitrite levels. EROD activity (♦) was determined spectrofluorimetrically on cell monolayers. Values are means ± S.D. of three different experiments (donors F, G and H). * P < .05 respect to untreated cells.
NO release from SNAP and its effects on EROD activity. After 24 hr in culture, human hepatocytes were exposed to increasing concentrations of SNAP for an additional 24 hr. The NO release in culture medium (•) was quantified by measuring nitrite levels. EROD activity (♦) was determined spectrofluorimetrically on cell monolayers. Values are means ± S.D. of four cultures from donors A, B, C and D. * P < .05 respect to control hepatocytes.
Concentration-dependent effects of NMA on NO synthesis and EROD activity in presence or absence of IFN-γ. After 24 hr in culture, human hepatocytes were exposed to increasing concentrations of NMA for an additional 24 hr in absence (○) or presence (•) of 300 U/ml IFN-γ. A, NO production was quantified by measuring nitrite levels. B, EROD activity was determined spectrofluorimetrically on cell monolayers. Values are means ± S.D. of three different cell preparations from donors C, D and E. * P < .05 respect to cells not treated with NMA. †P < .05 respect to the corresponding hepatocytes not treated with IFN-γ.
The effects of IFN-γ on induced EROD activity were also examined. Incubation with 2 μM MC for 24 hr led to a 5-fold increase in EROD activity with respect to untreated hepatocytes (fig.4B). The reduction in EROD activity induced in MC-treated cells by treatment with IFN-γ was comparable to that observed in basal activity. As in noninduced hepatocytes, 300 μM NMA produced a full inhibition in NO production (fig. 4A) and partially reversed the effects on EROD activity (fig. 4B).
Inhibitory effect of NO biosynthesis on EROD activity of MC-treated human hepatocytes. After 24 hr in culture, human hepatocytes were exposed to 2 μM MC and 300 U/ml IFN-γ for an additional 24 hr in absence (□) or presence (▨) of 300 μM NMA. A, NO production was quantified by measuring nitrite levels. B, EROD activity was determined spectrofluorimetrically on cell monolayers. Values are means ± S.D. of three different experiments (donors F, G and J). * P < .05 respect to MC-treated hepatocytes. †P < .05 respect to the corresponding cells not treated with NMA.
NO effects on other CYP activities.
In an attempt to determine whether the observed effects on human CYP1A2 were exclusive to this CYP isozyme, we studied the effects of IFN-γ on activities catalyzed by other CYP isozymes. Table 2 shows that IFN-γ produced a decrease (56–66% as compared to control activities) in CYP-dependent monooxygenase activities in human cultured hepatocytes. In all the activities studied, the effects of IFN-γ were prevented by the presence of NMA, and a 82 and 89% of control activities was reached in cells treated with IFN-γ and NMA.
NO-mediated effects of IFN-Y on different monooxygenase activities
NO-mediated effects of IFN-γ on CYP1A2 and 3A4 apoproteins.
In the subsequent experiments, the effects of IFN-γ on the expression of CYP isozymes were investigated. Immunoblot analysis revealed that accumulation of the CYP3A4 apoprotein of human hepatocytes was decreased by IFN-γ (fig. 5). Quantification of the dark areas with an image analyzer showed a drop in the CYP3A4 levels of IFN-γ-treated hepatocytes to 52% that of the control cells. Again, IFN-γ-mediated down-regulation of protein concentration was apparently reduced by NMA (to 80% of levels in control cells), although NMA alone did not modify protein levels (95% of untreated cells). The effects of IFN-γ and/or NMA on de novosynthesis of CYP proteins, determined by a immunoprecipitation, are shown in figure 6. As in the case of CYP apoprotein levels, human hepatocytes treated with IFN-γ showed a reduction inde novo synthesis of CYP1A2 and CYP3A4 to 72 and 65% of the control values, respectively. Again, NMA reduced the effects of IFN-γ (to 84 and 90% of controls for CYP1A2 and CYP3A4, respectively).
Immunoblot analysis of CYP3A4 protein after IFN-γ treatment in human hepatocytes. After 24 hr in culture, human hepatocytes were exposed to 300 U/ml IFN-γ and/or 300 μM NMA for an additional 24 hr. Cellular lysates (40 μg of protein per lane) were analyzed by Western blot with an anti-CYP1A2 polyclonal antibody. The results correspond to a representative culture (donor L). Lane 1, control; lane 2, NMA; lane 3, IFN-γ; lane 4, IFN-γ + NMA.
Effect of IFN-γ on de novosynthesis of CYP1A2 and CYP3A4 in human hepatocytes. After 24 hr in culture, hepatocytes were maintained in control conditions (□) or exposed to 300 μM NMA (░), 300 U/ml IFN-γ (▧) or both (▩) for an additional 24 hr. Cells were then pulse-labeled for 4 hr with 50 μCi/ml of [35S]-methionine in methionine-free RPMI-1640 culture medium. Samples were processed as described in “Materials and Methods.” Results are expressed as percentage of total proteinde novo synthesis, determined by trichloroacetic acid precipitation of a 10-μl aliquot of cell lysate. Values are means ± S.D. of three different experiments (donors H, I and M). * P < .05 respect to control cells.
NO-mediated effects of IFN-γ on CYP3A4 mRNA.
To overcome the detection problem in the RNA obtained from the limited number of human liver cells, and the problem of cross-detection of closely related isoforms of the CYP3A subfamily (CYP3A5 and 3A7), we evaluated the relative changes in CYP3A4 mRNA levels by semiquantitative RT-PCR. The yield of the PCR product is proportional to the input cDNA, when the cDNA is the only limiting factor of the reaction. We determined that in control samples, CYP3A4 and β-actin (internal control) amplification was in the exponential phase of the reaction when the input cDNA was diluted more than 8- and 250-fold, respectively, and was amplified for 27 cycles.
Once the optimal conditions for semiquantitative RT-PCR were established, we measured the effect of IFN-γ and/or NMA on the expression levels of CYP3A4 mRNA. IFN-γ produced a reduction in the accumulation of specific messages to 72% (66–75%, n= 3) of that measured in untreated cells (fig. 7). This down-regulation caused by IFN-γ was restored to control levels (94–98%, n = 2) in the presence of NMA, although NMA alone did not have a significant effect on CYP3A4 mRNA.
Effect of IFN-γ on the expression of CYP3A4 mRNA of human hepatocytes. After 24 hr in culture, hepatocytes were maintained in control conditions (□) or exposed to 300 μM NMA (░), 300 U/ml IFN-γ (▧) or both (▩) for an additional 24 hr. CYP3A4 mRNAs from three different experiments (donors M, N and O) were quantified by RT-PCR analysis as described in “Materials and Methods.” Values were calculated from confirmed three to four RT-PCR bands and are expressed as percentage of mRNA CYP3A4/mRNA β-actin in control cells.
Discussion
In previously reported experiments we observed a marked reduction in CYP-dependent activities after treatment of human hepatocytes with IFN-γ (Donato et al., 1993a). Although a decrease in specific CYP1 mRNA produced by IFN-γ and other cytokines has been described in human hepatocytes (Abdel-Razzak et al., 1993; Muntane-Relat et al., 1995), the mechanism involved in this phenomenon remains unknown. A possible correlation between NO level induction and inhibition of CYP expression during rat hepatocyte treatment with a combination of endotoxin and different cytokines, including IFN-γ, has recently been reported (Stadler et al., 1994; Carlson and Billings 1996). Therefore, it was of interest to confirm the possible role of NO on inhibitory effects of human CYP activities produced by IFN-γ.
Using human hepatocytes as an experimental model, we have demonstrated that exogenously added NO inhibited the CYP1A2 activity (assessed as EROD activity, Gonzalez 1990) of human hepatocytes in a dose-dependent manner (fig. 2). This finding is in agreement with previous experiments using genetically engineered V79-derived cell lines constitutively expressing rat and human CYP1A1/2 (Stadler et al., 1994) and microsomal preparations from rat hepatocytes (Wink et al., 1993) or rat liver (Khatsenko et al., 1993).
It has been reported that induction of NO synthesis in human cultured hepatocytes requires stimulation with at least two different cytokines (Geller et al., 1993). We have observed that human hepatocytes increased NO release in culture supernatants in response to IFN-γ stimulus alone (fig. 1). However, NO production was lower than when cells were stimulated with a mixture of cytokines (data not shown). Inhibition of NO biosynthesis by treatment with NMA, a competitor of l-arginine, partially reversed the effects of IFN-γ on CYP1A2 activity both in control and MC-induced human hepatocytes (figs. 3 and 4). Similarly, IFN-γ inhibition of CH and 6β-OHT activities, which specifically identifies human CYP2A6 (Yunet al., 1991) and CYP3A4 (Waxman et al., 1991) respectively, and PROD and BROD activities, linked to CYP2B1 in rat liver (Burke et al., 1985) and probably to the same subfamily in humans, was reduced in the presence of NMA (table 2). These results indicate that endogenous NO, like the exogenously produced NO, had a direct inhibitory effect on CYP activities and that the increased release of NO induced by IFN-γ can explain, at least in part, the effects observed on the CYP system in human hepatocytes. It was reported that the NO synthase inhibitor l-nitroarginine failed to protect mice from changes in CYP activities after administration of an IFN inducer (Hodgson and Renton 1995). In addition, no decreases in CYP activities were observed in animals treated with NO-generating drugs. These results indicated that NO is not a mediator of CYP down-regulation in mouse liver, and they are clearly different from the results obtained in rat hepatocytes (Winket al., 1993; Stadler et al., 1994).
The physiological significance of NO biosynthesis in the liver is only beginning to be understood and the molecular basis of the inhibitory effects of NO on CYP enzymes is not yet clear. Reactivity of NO with hemo iron to yield nitrosyl-heme adducts may convert hemoproteins into the primary target of NO within cells (Ignarro, 1990; Chamulitratet al., 1995). It has been postulated that the NO inhibition of CYP-mediated O-dealkylase activities in microsomal preparations involves both binding of NO to the prosthetic group in the catalytic center and a destruction of the integrity of the primary structure of the hemoprotein, possibly resulting from the action of nitrogen oxides derived from the oxidation of NO by oxygen (Khatsenko et al., 1993; Wink et al., 1993). However, functional inhibition of CYP enzymes by NO could not be the only mechanism involved. In vitro hepatocyte NO production has been associated with a decrease in hepatocyte total protein synthesis (Billiar et al., 1989). In particular, induction of endogenous NO biosynthesis after rat hepatocyte stimulation with lipopolysaccharide and cytokines leads to a decrease in CYP1A1 and CYP1A2 protein and mRNA expression, which can be up-regulated by NMA treatment only in the case of CYP1A1 (Stadler et al., 1994). In our study, we observed that treatment of human hepatocytes with IFN-γ produced a reduction in specific CYP apoprotein and mRNA levels (figs. 5, 6, 7), which could explain the decreases observed in CYP activities. Whether IFN-γ reduces mRNA levels by interfering with the transcriptional activation of the genes or by increasing the rate of mRNA degradation is not known. The fact that IFN-γ effects on CYP expression were reversed by NMA indicates that down-regulation of CYP isozymes by this cytokine could be mediated by NO synthetized by hepatocytes in response to IFN-γ stimulation. However, it is interesting to note that NMA produced a complete inhibition of NO synthesis, but did not fully prevent inhibition of CYPs. This suggests that the suppressive effects induced by IFN-γ were only partially attributable to NO biosynthesis induction and that other non-NO-dependent mechanisms could also be involved.
In summary, our results support the idea that the action of IFN-γ on human hepatocyte CYP-dependent metabolism must be due to at least two different mechanisms, an NO-mediated one and/or another that is NO-independent, which produce both inhibition of activity of CYP isozymes belonging to at least four different subfamilies (1A, 2A, 2B and 3A) and decreases in enzyme content. This down-regulation of the CYP system by IFN-γ may be clinically important in understanding the altered pharmacokinetics of drugs in patients who suffer pathological processes that involve a release of endogenous IFN-γ or are undergoing IFN-γ treatment (i.e., antiviral or antitumoral therapy). As the therapeutic use of IFN-γ will only increase, it is important to identify the individual modulating agents and mechanisms responsible for decreasing the capacity of the liver to metabolize and subsequently eliminate drugs.
Acknowledgments
The authors thank to Dr. F. P. Guengerich, Center in Molecular Toxicology, Vanderbilt University, Nashville, TN for providing anti-CYP1A2 and 3A4 antibodies..The expert technical assistance of E. Belenchón, T. Hualde and M. C. Lorenzo is gratefully acknowledged.
Footnotes
-
Send reprint requests to: Dr. M. JoséGómez-Lechón, Unidad de Hepatologı́a Experimental, Centro de Investigación, Hospital Universitario La Fe, Avda. Campanar 21, 46009-Valencia, Spain.
-
↵1 This work was supported by the European Union (Biomed I, Project Nr. BMH1–1097 and AIR, Project Nr. CT93–0860), the Spanish Fondo de Investigaciones Sanitarias (Project Nr. 94/1084) and The ALIVE Foundation.
- Abbreviations:
- BROD
- 7-benzoxyresorufin O-dealkylase
- CH
- coumarin 7-hydroxylase
- CYP
- cytochrome P450
- EROD
- 7-ethoxyresorufin O-deethylase
- IFN
- interferon
- MC
- 3-methylcholanthrene
- NMA
- NG-monomethyl-l-arginine
- NO
- nitric oxide
- PROD
- 7-pentoxyresorufin O-depentylase
- RT-PCR
- reverse transcriptase-polymerase chain reaction
- SNAP
- S-nitroso-N-acetylpenicillamine
- 6β-OHT
- testosterone 6β-hydroxylase
- Received June 5, 1996.
- Accepted December 5, 1996.
- The American Society for Pharmacology and Experimental Therapeutics
