Abstract
Induction of drug-metabolizing enzymes (DMEs) is highly species-specific and can lead to drug-drug interaction and toxicities. In this series of studies we tested the species specificity of the antidiabetic drug development candidate and mixed peroxisome proliferator-activated receptor (PPAR) α/γ agonist (S)-4-O-tolylsulfanyl-2-(4-trifluormethyl-phenoxy)-butyric acid (EMD 392949, EMD) with regard to the induction of gene expression and activities of DMEs, their regulators, and typical PPAR target genes. EMD clearly induced PPARα target genes in rats in vivo and in rat hepatocytes but lacked significant induction of DMEs, except for cytochrome P450 (P450) 4A. CYP2C and CYP3A were consistently induced in livers of EMD-treated monkeys. Interestingly, classic rodent peroxisomal proliferation markers were induced in monkeys after 17 weeks but not after a 4-week treatment, a fact also observed in human hepatocytes after 72 h but not 24 h of EMD treatment. In human hepatocyte cultures, EMD showed similar gene expression profiles and induction of P450 activities as in monkeys, indicating that the monkey is predictive for human P450 induction by EMD. In addition, EMD induced a similar gene expression pattern as the PPARα agonist fenofibrate in primary rat and human hepatocyte cultures. In conclusion, these data showed an excellent correlation of in vivo data on DME gene expression and activity levels with results generated in hepatocyte monolayer cultures, enabling a solid estimation of human P450 induction. This study also clearly highlighted major differences between primates and rodents in the regulation of major inducible P450s, with evidence of CYP3A and CYP2C inducibility by PPARα agonists in monkeys and humans.
The liver is the major site of biotransformation of xenobiotics, and biotransformation is divided into three main phases: activation (phase I), conjugation (phase II), and drug transport (phase III). Phase I reactions, including microsomal cytochrome P450 (P450)-dependent oxidation pathways, and phase II reactions like UDP-glucuronosyltransferase (UGT)-dependent conjugation are involved in detoxification and elimination of endogenous and exogenous substances, formation of pharmacologically active drugs from prodrugs, but also generation of toxic metabolites (Parkinson, 2001).
Exposure to drugs, occupational and industrial chemicals, or environmental pollutants can lead to either the induction or the inhibition of biotransformation (Coecke et al., 2006). Because of their inducibility, drug-metabolizing enzymes (DMEs) such as P450s can be involved in various side effects such as profound endogenous hormonal disturbances, increased liver weight, drug-drug interactions, and exacerbated toxic effects. Therefore, evaluation of the inducing potential of a given chemical on these enzymes is invaluable for human safety assessment (Madan et al., 2003).
Because of major species differences, both in the catalytic activities and regulation of this group of enzymes, the evaluation of a compound's effect can be accurately performed only with human tissue (Silva et al., 1998). During the past decade, primary cultures of isolated human hepatocytes have proven to be a valuable model to study the inducing potential of drugs on different P450 isozymes (e.g., LeCluyse et al., 2000, 2005; Richert et al., 2003, 2006; Hewitt et al., 2007). Major families of inducers have been identified, and transcription factors involved in specific induction pathways have been discovered (Waxman, 1999), such as the arylhydrocarbon receptor (AhR), pregnane X receptor (PXR), constitutive androstane receptor (CAR), and peroxisome proliferator-activated receptor (PPAR) α.
PPARs are nuclear receptors that control a variety of genes involved in several pathways of lipid metabolism (Devergne and Wahli, 1999). In humans, PPARα and PPARγ are important regulators of lipid and lipoprotein metabolism, cellular differentiation, and glucose homeostasis. PPARα mainly acts on lipid and lipoprotein catabolism genes, predominantly in the liver (e.g., β-oxidation of fatty acids), whereas PPARγ plays an active role in the regulation of lipid storage and contributes to insulin action. Consequently, the development of PPARα/γ agonists represents an opportunity to produce tailored compounds that can treat both perturbation of lipid metabolism and insulin resistance (Staels and Fruchart, 2005; Harrity et al., 2006).
EMD 392949 (EMD) is a new chemical entity activating both PPARα and PPARγ, and it has been shown to ameliorate hyperglycemia and hyperinsulinemia in db/db mice (unpublished data). By combining the pharmacological properties of a PPARα and a PPARγ activator, EMD would be an ideal candidate for the treatment of “metabolic syndrome” and type 2 diabetes.
Recently, drug-drug interactions have been observed in humans after administration of PPARα ligands such as fenofibrate. This is most probably related, at least in part, to P450 induction (Prueksaritanont et al., 2005). Therefore, the aim of the present study was to compare the effects of EMD administration on DME regulators and DME mRNA expression in conjunction with related monooxygenase activities in different animal species and humans.
Materials and Methods
Chemicals and Reagents. Chemicals used in this study were obtained from Sigma-Aldrich (St. Quentin-Fallavier, France), and reagents for cell culture were from Invitrogen (Cergy Pontoise, France) unless stated otherwise. Cell culture plastics were purchased from Becton Dickinson (Grenoble, France). EMD (batch 00195C0, >95% purity) was from Merck Santé (Lyon, France).
Animals and Treatment. All the animal experiments were approved by the local authorities and were conducted in compliance with the principles of Good Laboratory Practice of the Organization for Economic Cooperation and Development, the European Union, and the Food and Drug Administration Good Laboratory Practice regulation 21, Code of Federal Regulations Part 58, as well as the local animal welfare regulations.
In a rat toxicity study, a group of three male HsdCpb:WU Wistar rats with a mean age of 7 weeks at the beginning of treatment were dosed for 13 weeks with EMD by daily p.o. (gavage) administration of 0 (control, 0.25% aqueous hydroxypropyl methylcellulose), 3, or 100 mg/kg b.wt. Approximately 1 h after the last treatment the animals were sacrificed, and portions of livers were immediately frozen in liquid nitrogen until required. The study was performed by Merck KGaA (Darmstadt, Germany).
In a monkey ex vivo P450 induction study (performed by Covance, Harrogate, Yorkshire, UK), a group of three male cynomolgus monkeys (Macaca fascicularis), which were at least 24 months old, were treated by daily p.o. (gavage) dosing with 0 (vehicle control, 0.25% aqueous hydroxypropyl methylcellulose), 30, 100, or 300 mg/kg b.wt./day EMD for 4 weeks. On the day of necropsy, approximately 1 h after the last treatment, portions of livers were immediately frozen in liquid nitrogen until required.
In a monkey toxicity study (performed by MDS Pharma Services, Sevres, France), a group of three male cynomolgus monkeys with an age of 26 to 33 months at the beginning of treatment were treated daily by p.o. (gavage) administration of 0 (vehicle control, 0.25% aqueous hydroxypropyl methylcellulose), 15, or 150 mg/kg b.wt./day EMD for 17 weeks. At the end of the treatment the animals were sacrificed, and portions of livers were frozen in liquid nitrogen until use.
Source of Human Livers. Liver samples were taken from patients undergoing liver resection for different pathologies (Table 1). All the experimental procedures were performed in compliance with French law and regulations after approval by the National Ethics Committee (Paris, France). Informed consent was obtained from all the patients for the use of liver tissue for research purposes.
Hepatocyte Isolation. Rat hepatocytes were isolated from male Wistar rat livers by a two-step collagenase perfusion method as previously described (Viollon-Abadie et al., 2000). Human hepatocytes were isolated based on a modification of a two-step collagenase digestion method, according to a recently described protocol (Richert et al., 2004; LeCluyse et al., 2005).
Hepatocyte Culture and Treatment. Rat and human hepatocytes were plated in 60-mm dishes at a density of 3.5 × 106 cells/dish or in 6-well BD BioCoat plates (BD Biosciences, San Jose, CA) at a density of 1.5 × 106 cells/well in 3 or 2 ml of attachment medium, respectively, and cultured under aCO2/air (5%/95%) humidified atmosphere at 37°C. Attachment medium consisted of Dulbecco's modified Eagle's medium containing 5% fetal calf serum, 50 mg/l gentamicin, 4 mg/l insulin, and 10–5 M hydrocortisone. After a 24-h attachment period, the medium was discarded and replaced by incubation medium, consisting of Dulbecco's modified Eagle's medium, supplemented with 50 mg/l gentamicin, 4 mg/l insulin, and 10–5 M hydrocortisone and containing 0 [vehicle control, 0.1% (v/v) dimethyl sulfoxide], 30, or 100 μM EMD or 100 μM fenofibrate (five dishes for each group). Medium in all the dishes was renewed daily.
Microsome Preparation.Liver microsomes. The livers (n = 3 per species and dose level) were thawed in ice-cold 50 mM Tris-HCl, pH 7.4, containing 0.25 M sucrose, scissor-minced, and homogenized. Microsomal suspensions were prepared by differential centrifugation as described previously (Richert et al., 2002). Briefly, liver homogenates were sonicated and centrifuged for 20 min at 9000g and 4°C. Supernatant fractions were collected and centrifuged for 60 min at 100,000g and 4°C. The resulting microsomal pellets were resuspended in 80 to 120 μl of 0.25 M sucrose. The protein concentration of each sample was determined by the Lowry assay (Lowry et al., 1951). The microsomes were snap-frozen and stored at –80°C until evaluated.
Hepatocyte microsomes. At the end of the 72-h incubation period, cells from dishes were harvested for microsome preparation. Culture dishes within individual treatment groups were scraped, pooled, and frozen at –80°C as previously described (Richert et al., 2004; LeCluyse et al., 2005). After thawing of cell homogenates, microsomes were prepared by differential centrifugation as described above. The protein concentration of each sample was determined by the bicinchoninic acid protein assay kit, according to the manufacturer's instructions (Sigma-Aldrich), and bovine serum albumin was used as a standard.
Microsomal Enzyme Activity Assays. Microsomal activities of 7-ethoxyresorufin O-deethylase (EROD, CYP1A), 7-pentoxyresorufin O-depentylase (PROD, CYP2B) or 7-benzyloxyresorufin O-debenzylase (CYP2B), bupropion-hydroxylase (CYP2B), testosterone 6β- and 16β-hydroxylases (CYP3A and CYP2B, respectively), and lauric acid 12-hydroxylases (CYP4A) were determined as previously described (Okita et al., 1991; Faucette et al., 2000; Robertson et al., 2000; Richert et al., 2002). Acyl-CoA Oxidase Enzyme Activity. Acyl-CoA oxidase (ACOX) activity was measured in rat and monkey liver samples. Fifty-milligram portions of frozen livers were homogenized for 30 s on ice in 1 ml of sucrose solution [10% (w/v), 3 mM imidazole, pH 7.4] using a rotor stator homogenizer. Liver homogenates were frozen in liquid nitrogen and stored at –80°C until analyzed. After thawing on ice and centrifugation (10 min, 4°C, 7000g), the protein concentration was determined in the supernatants using the Bradford method (Bradford, 1976) with bovine serum albumin as standard. Palmitoyl-CoA oxidase activity was determined in supernatants according to a modification of a previously described method (Small et al., 1985; Ammerschlaeger et al., 2004). Absorption was recorded at 502 nm for 4 min every 12 s.
mRNA Preparation and Analysis.Liver mRNA. Frozen portions of rat or monkey livers were fragmented in liquid nitrogen. Pieces of 40 to 100 mg were immersed in TRI Reagent (Sigma, Taufkirchen, Germany) and immediately homogenized for 45 s on ice using a rotor stator homogenizer. Total RNA was isolated following the TRI Reagent standard protocol provided by the manufacturer. RNA pellets were dissolved in nuclease-free water and stored at –80°C until further use.
Hepatocyte mRNA. At the end of the incubation periods (24 and 72 h), each well was rinsed twice with ice-cold phosphate-buffered saline, and 500 μl of TRI Reagent was added to each well. Cells were scraped, and the three wells from individual treatment groups were pooled. Total RNA was isolated following the TRI Reagent standard protocol provided by the manufacturer. RNA pellets were dissolved in nuclease-free water and stored at –80°C until further use.
mRNA analysis. Quality and concentration of total RNA were determined using the NanoDrop spectrophotometer (Kisker, Steinfurt, Germany) and the Agilent Bioanalyzer 2100 applying the Total RNA Nano Assay (Agilent Technologies, Waldbronn, Germany) according to the manufacturer's protocols.
cDNA Synthesis and Analysis. Five micrograms of total RNA was reverse-transcribed to cDNA using random hexamer primers with the Transcriptor First Strand cDNA Synthesis Kit (Roche, Mannheim, Germany) according to the protocol provided by the manufacturer. cDNA quality and concentration were determined using the Agilent Bioanalyzer 2100 applying the mRNA Pico Assay (Agilent Technologies).
Real-Time Polymerase Chain Reaction. Real-time polymerase chain reaction (PCR) analysis was essentially performed as described by Tuschl and Mueller (2006). Briefly, single gene real-time PCR primers and probes were delivered as TaqMan Gene Expression Assays (Applied Biosystems, Darmstadt, Germany) for the rat and human genes listed in Table 2. Assays targeting human genes were also applied to analyze mRNA isolated from cynomolgus monkey liver samples. The amplicon sequences for the human assays in Tables 2 and 3 were used to search for sequence similarities in homologous genes of cynomolgus monkey (M. fascicularis) or other nonhuman primate species. The Basic Local Alignment Search Tool (Altschul et al., 1990) results are shown in Table 4. Real-time PCR was performed on Applied Biosystems ABI Prism 7000 Sequence Detection System with ABI Prism 7000 SDS software 1.0. Two nanograms of cDNA was used per reaction, and 18S rRNA control (4310893E) (Applied Biosystems) was used for normalization. Reactions were performed in triplicate for each sample. Analysis of gene expression values was performed using the efficiency-corrected comparative CT method. Gene expression ratios were calculated using the following formula:
TaqMan Low-Density Array. In addition to single gene real-time PCR measurements, TaqMan low-density arrays (Applied Biosystems) were used to analyze rat, human, and monkey mRNA. Fifty nanograms of cDNA was used per sample and loaded into a single sample loading port. Tables 3 and 5 list human and rat genes with the corresponding gene expression assays present on the respective TaqMan low-density arrays. Assays targeting human genes were also applied to analyze mRNA isolated from M. fascicularis liver samples (see above). Thermal cycling and fluorescence detection were performed on Applied Biosystems ABI Prism 7900HT Sequence Detection System with ABI Prism 7900HT SDS software 2.1. Analysis of gene expression values was performed using the efficiency-corrected comparative CT method (see above).
Statistical Analysis. Statistical significance of alterations in enzyme activity or gene expression was analyzed using Origin software (OriginLab Corporation, Northampton, MA). An analysis of variance with Tukey's post hoc test was applied to analyze each experimental group. Statistical analysis was not used on human hepatocyte data because each donor is presented individually and no mean value of biological replicates was calculated. Statistically significant results are labeled with capital letters (p < 0.01) or lower-case letters (p < 0.05) in Figs. 1, 2, 3, 4 and Table 6. The letter a/A stands for significantly different from control and b/B for significantly different from control and other dose(s) labeled with b/B or c/C. The letter c/C indicates no significant difference from control but significant difference from other dose(s) labeled with c/C or b/B.
Results
Effects of EMD on Hepatic Xenobiotic Metabolizing Enzymes following Repeated in Vivo Administration in Rats and Monkeys. Male Wistar rats were treated p.o. with EMD at 0, 3, or 100 mg/kg b.wt./day for 13 weeks. At the end of the treatment period, livers were assessed for mRNA expression of various DMEs, relevant nuclear receptors, and transcription factors (Fig. 1A), as well as for selected microsomal P450-dependent monooxygenase activities (Fig. 1B). Following repeated p.o. administration of EMD, neither Cyp1A2 mRNA expression nor Cyp1A-specific EROD monooxygenase activity was affected at 3 mg/kg/day but was decreased with statistical significance at the high dose of 100 mg/kg/day. The latter effect was associated with a slight decrease, although not statistically significant, in the abundance of AhR mRNA, the main regulator of Cyp1A expression. Cyp2B mRNA expression was strongly increased at both doses but did not reach statistical significance because of strong interindividual variation in the magnitude of induction. Moreover, the related PROD monooxygenase-dependent activity was moderately and significantly increased. Cyp3A mRNA expression and activity were increased about 2-fold at 3 mg/kg, whereas at 100 mg/kg this effect had disappeared at the enzyme activity level and was even reduced at the gene expression level. In addition, multidrug resistance (MDR) 1 gene expression was significantly repressed by EMD. Although PXR mRNA abundance was almost unchanged, there was a significant elevation of about 2-fold in CAR expression at both dose levels. At 100 mg/kg/day EMD, a strong and dose-dependent increase (>15-fold) in Cyp4A activity was observed, along with a 6-fold induction of corresponding Cyp4A3 mRNA, both being highly significant and typical features of PPARα activation. In line with these observations, we observed significant and dose-dependent inductions of ACOX, carnitine-palmitoyl transferase (CPT1a), and PPARα mRNAs by EMD. ACOX enzyme activity was also markedly and significantly increased by EMD treatment in rats (Table 6). Overall, these findings strongly confirm that EMD is a potent PPARα agonist in the rat.
We then compared the effects of EMD observed in rats with those in monkeys. In an ex vivo P450 induction study, male cynomolgus monkeys were treated by daily p.o. dosing with 0, 30, 100, or 300 mg/kg EMD for 4 weeks. At the end of the treatment period, livers were assessed for mRNA expression of various DMEs, their regulators, and typical PPAR target genes (Fig. 2A), as well as for selected microsomal P450-dependent monooxygenase activities (Fig. 2B). EMD strongly and significantly decreased CYP1A2 mRNA in a dose-dependent manner, again correlating with a reduction in AhR expression, especially at the highest dose tested. Additionally, CYP2B6 mRNA was significantly repressed. In contrast, CYP2C9 was slightly induced at 30 mg/kg EMD, whereas CYP3A4 and CYP4A were induced only at 300 mg/kg. Enzyme activities of CYP2B6, CYP3A, and CYP4A were moderately increased (maximum of 2-fold), whereas CYP1A-dependent EROD activity was almost unchanged or weakly decreased at all three dose levels.
Two of the main regulators of DMEs, the nuclear receptors PXR and CAR, were regulated in an opposite direction. There was a weak reduction in PXR expression, whereas CAR was slightly induced at 30 mg/kg. The transcription factor hepatic nuclear factor 1α was induced at 300 mg/kg. ACOX, a hallmark marker of peroxisome proliferators in rodents, was repressed at the mRNA level by EMD (Fig. 2A), but no significant change in ACOX enzyme activity was noted compared with the vehicle-treated control (Table 6).
In a second study, cynomolgus monkeys were treated p.o. with 0, 15, or 150 mg/kg/day for 17 weeks, followed by a 4-week recovery period for a group of the high-dose animals. Livers were assessed for mRNA expression as described above (Fig. 3). Similar to the 4-week study, CYP1A2 was significantly repressed after 17 weeks of treatment but only at the high dose. In contrast to the repression of CYP2B6 in the short-term study, there were only minor changes after 17 weeks. A very distinct increase in CYP2C9, CYP3A4, and MDR1 mRNA was observed after 17 weeks (Fig. 3) that was less apparent in the 4-week study (Fig. 2A). Interestingly, CYP4A11 was weakly but notably induced, in line with a distinct induction of ACOX and PPARα (Fig. 3), although ACOX enzyme activity was not increased (Table 6). Contrary to the 4-week study (Fig. 2A), mRNA levels for the transcription factors hepatic nuclear factor 1α, AhR, and PXR but not CAR were higher after 17 weeks of treatment (Fig. 3). At the end of the recovery period, most of the gene expression changes were significantly reversed.
In Vivo versus in Vitro Effects of EMD on Hepatic Xenobiotic Metabolizing Enzymes in Rats. For comparison of in vivo with in vitro effects, male Wistar rat hepatocytes were treated with EMD at 0, 30, or 100 μM for 24 and 72 h. The doses were chosen based on peak plasma concentrations observed in the rat toxicity study (peak plasma concentrations were in the range of 30–470 μM) and on PPARα/γ activity in vitro (3–100 μM; data not shown). Fenofibrate was included as a reference PPARα activator. After 24- and 72-h incubation with the compounds, hepatocyte cultures were assessed for mRNA expression of DMEs, transcription factors, and PPAR marker genes as described above (Fig. 4, A and B). Additionally, selected microsomal P450 activities were tested after 72 h of treatment with EMD (Fig. 4C).
After 24- and 72-h treatment, the effects of EMD on mRNA expression were very similar to that of fenofibrate, especially at the corresponding dose of 100 μM, where there was no statistically significant difference between both compounds' profiles. The changes in gene expression after in vitro treatment of cultured rat hepatocytes with EMD were altogether equivalent to those observed after in vivo administration (Fig. 1A) for all the genes measured. The same was true for the effects of EMD on microsomal monooxygenase activities: decrease of Cyp1A-dependent activity, slight increase of Cyp3A-dependent activity, and a strong increase in Cyp4A-dependent activity observed after in vitro treatment (Fig. 4C), although only Cyp4A activity was significantly changed. Overall, the gene expression profiles, as well as the DME activities, were in agreement with the observations after in vivo treatment with EMD (Fig. 1).
Effects of EMD on Hepatic Xenobiotic Metabolizing Enzymes in Human Hepatocytes. Finally, we assessed EMD for its P450-inducing capacity in human hepatocytes to allow an extrapolation to humans. Fresh human hepatocytes from three different donors (Table 1) were treated with EMD at 0, 30, or 100 μM for 24 and 72 h, and cultures were assessed for mRNA expression (Figs. 5, 6, 7). Again, fenofibrate was included as a reference PPARα activator. From one donor (donor 3; Table 1) microsomal P450 activities were measured after 72-h treatment (Fig. 7C).
The effects of EMD on gene expression were comparable with that of fenofibrate, although not as similar as seen in rat hepatocytes (Fig. 4, A and B). Depending on the donor and on the gene of interest, effects were maximal after 24 or 72 h of treatment. CYP1A1 was consistently repressed after 24 h and induced after 72-h treatment, especially at 100 μM EMD, in all three donors. Similar to the results in monkeys (Fig. 3), AhR was induced by EMD and to a lesser extent by fenofibrate. CYP1A2 mRNA expression was decreased by the treatment at both time points in hepatocytes from two of the three donors but was induced after 72 h in donor 3 (Fig. 7B). In hepatocytes from donor 3, CYP1A1/2-dependent EROD activity was not affected by EMD treatment (Fig. 7C).
A distinct increase in CPT1a, a typical PPARα marker in human hepatocytes (Ammerschlaeger et al., 2004), was detected in all three human hepatocyte cultures at 24 and 72 h. PPARα was weakly induced by EMD and fenofibrate similar to the effects on AhR (Figs. 5, 6, 7). Interestingly, CYP4A11 and ACOX mRNA were more strongly induced after 72 h (Figs. 5, 6, 7), indicating a delayed induction of these classic rodent PPARα markers. This finding was confirmed by a 2-fold increase in CYP4A-dependent lauric acid hydroxylase activity in donor 3 (Fig. 7C).
Strikingly, CYP2C8, CYP3A4, and MDR1 were consistently and strongly induced on the mRNA level (Figs. 5, 6, 7). Furthermore, CYP3A activity was distinctively and dose-dependently increased (Fig. 7C) in hepatocytes from donor 3, confirming the gene expression data. The mRNA expression of PXR and CAR, regulators of CYP3A and/or CYP2C, was not consistently deregulated. However, there was a slight repression of PXR and CAR by EMD in the majority of cases.
Discussion
In regulatory animal toxicity and toxicokinetic studies the drug development candidate EMD showed marked species-specific differences in its kinetic properties: exposure after repeated dosing was not dose-proportional in cynomolgus monkeys, whereas these effects were minor in rats (unpublished data). These observations are indicative of induction of metabolism of the parent drug predominantly in monkeys. To better characterize the species-specific properties of EMD with respect to DME induction, in particular P450s, we compared its effect on specific DME mRNA expression and activity. In addition, the expression of major regulators of DMEs in vivo in rats and monkeys and in rat hepatocytes was studied. We then further investigated how these effects might translate to humans by using primary human hepatocyte cultures.
As expected from its pharmacological activity (unpublished data), EMD markedly induced Cyp4A-dependent lauric acid ω-hydroxylation activity and related mRNA, as well as genes from the fatty acid β-oxidation pathways in Wistar rats. This was also observed after in vitro exposure of rat hepatocytes to EMD or fenofibrate, a prototypical PPARα agonist, consistent with the known effects of PPARα ligands in rodents (for a review, see Johnson et al., 2002). Interestingly, PPARα was induced in vivo but not in vitro in rats. This lack of PPARα induction by peroxisome proliferators in vitro has been previously reported by our laboratory (Ammerschlaeger et al., 2004).
In line with the well documented species-specific actions of PPARα agonists in vivo (e.g., Richert et al., 1996; Johnson et al., 2002) and in vitro (e.g., Perrone et al., 1998; Ammerschlaeger et al., 2004), the induction of CYP4A and ACOX mRNA expression were much less pronounced in monkey than in rat livers. In fact, these PPARα markers were repressed (in the case of ACOX) in the 4-week monkey study, but after treatment for 17 weeks, a notable induction of ACOX, CYP4A, and PPARα was detectable. This indicated a time-dependent induction of these typical rodent peroxisome proliferative genes in nonhuman primates.
Treatment of male Wistar rats with EMD suppressed Cyp1A1/2 metabolic activity and Cyp1A2-related mRNA. This was also observed after treatment of male rat hepatocytes with EMD or fenofibrate. These observations are in line with reports on Cyp1A1/2 metabolic activity suppression in rats after fenofibrate administration (Shaban et al., 2004). These authors concluded that this effect was PPARα-dependent because of an inhibitory effect on AhR function. In the present study we provide evidence that AhR mRNA expression was repressed by EMD in rats and by EMD and fenofibrate in rat hepatocytes. Male cynomolgus monkeys also responded by decreases in CYP1A-related activity and mRNA expression after repeated EMD treatment. However, AhR expression was slightly induced after long-term treatment with EMD but not after 4 weeks. This suggests a different long-term regulation of CYP1A in nonhuman primates.
The most striking differences between rats and monkeys included the consistent induction of CYP2C, CYP3A, and MDR1 mRNAs in monkeys but repression and/or marginal effects on these DMEs in rats in vivo and in vitro. CYP2B was also regulated in an opposite manner in monkeys and rats. The present results supported the assumption that EMD is an inducer of CYP2C and CYP3A in monkeys but not in rats. In monkeys, PXR, the major regulator of CYP3A (Reschly and Krasowski, 2006), was induced after treatment with EMD for 17 weeks, indicating that PXR induction leads to increased expression and activity of CYP3A and/or CYP2C. In contrast, CAR (Reschly and Krasowski, 2006), another important P450 regulator, was not induced in monkeys after 17 weeks but only slightly after 4 weeks of treatment. Contrary to the effects in monkeys, Cyp2C mRNA expression was repressed in rat livers by EMD and in rat hepatocytes by both EMD and fenofibrate treatment, correlating well with the known effects of PPARα agonists on Cyp2C in rats (Fan et al., 2004). Cyp3A mRNA and activity were only marginally affected in rats, consistent with the minor effects on PXR expression. CAR was induced by EMD, correlating with strong increases in Cyp2B mRNA expression in rats in vivo and in vitro. It has been previously reported that PPARα agonists have the potential to induce Cyp2B and lead to the suppression of Cyp2C11 in rats both on protein and mRNA levels (Shaban et al., 2005), further confirming that EMD is a potent PPARα agonist in rats. In monkeys, CYP2B6 was repressed after 4 weeks and remained unchanged after 17 weeks. Contrasting effects on P450s and their regulators in monkeys compared with rats indicate major differences in the mechanisms of regulation of P450s in nonhuman primates compared with rats.
The present study showed an excellent correlation between the in vivo effects of EMD on rat livers and the in vitro effects on cultured rat hepatocytes in terms of specific P450 induction, which is in line with our previous results obtained from PPARα ligands (Richert et al., 1996; Goll et al., 1999). Our results further confirmed that primary cultures of hepatocytes can be considered as the gold standard for DME induction studies in vitro (Richert et al., 2003; Castell et al., 2006; Tuschl and Mueller, 2006). Therefore, we extended the evaluation of the response of P450s and nuclear receptor expression to fenofibrate and EMD to primary cultures of human hepatocytes.
EMD and fenofibrate induced CPT1a in all three human hepatocyte cultures, suggesting that EMD activated PPARα in human hepatocytes, which is in accordance with previous observations (Richert et al., 2003; Ammerschlaeger et al., 2004; Raucy et al., 2004). The slightly increased PPARα mRNA levels in human hepatocyte cultures further support this assumption, although there was considerable variation between donors. The classic rodent PPARα markers CYP4A and ACOX were also induced in human hepatocytes, and in most cases this was strongest after 72-h treatment. Interestingly, induction of ACOX and CYP4A in monkey livers was only apparent after long-term treatment with EMD (see above). Taken together, the results in human and nonhuman primates—species that are in general refractory to peroxisome proliferation—suggest that a minor induction of peroxisome proliferation markers may occur in these species after prolonged exposure. Nevertheless, it has to be stressed here that the maximum levels of induction after EMD or fenofibrate treatment were only about 2- to 4-fold in human hepatocytes or monkey livers compared with 8- to 80-fold in rat hepatocytes. This is consistent with the well established difference in susceptibilities of human versus rodent hepatocytes (Richert et al., 2003). Thus, the present results further corroborated that PPARα ligands, including EMD, although effective in human hepatocytes and monkey livers, are much less powerful inducers of the peroxisomal fatty acid metabolism pathways in primates than in rodents.
CYP3A and CYP2C, the major drug-metabolizing P450s in humans, were strongly induced by fenofibrate and EMD in human hepatocytes. The induction of activity and mRNA expression was comparable with that seen in vivo in monkeys, indicating that monkeys are, in this particular case, predictive for the P450 induction of EMD. An induction in CYP3A4 and CYP2C in human hepatocytes by PPARα agonists has been previously shown for clofibric acid (Richert et al., 2003; Prueksaritanont et al., 2005). In addition, MDR1, which is also regulated by PXR and correlates well with CYP3A4 expression (Reschly and Krasowski, 2006), was induced by EMD in monkeys and humans but not in rats. Taken together, this indicates that PPARα agonists in general may be CYP3A inducers, an assumption that should be confirmed by analysis of a broader variety of PPARα agonists. The marginal effect on Cyp3A and repression of Cyp2C11 in rat hepatocytes, both by fenofibrate and EMD treatment, compared with the strong CYP2C8/9 and CYP3A4 induction in monkey livers and human hepatocytes highlights the marked difference between rodents and primates in the regulation of these P450s. PXR was not induced by EMD or fenofibrate in human hepatocytes, which is in agreement with previous studies that showed that fibrates failed to activate human PXR in a reporter gene assay (Prueksaritanont et al., 2005). As CYP3A4 and CYP2C8 expression can also be mediated by the glucocorticoid receptor and CAR (Dvorak et al., 2003; Sugatani et al., 2004, Faucette et al., 2006), it is possible that the latter pathway could be involved in the induction of these enzymes. Further work is necessary to explore this possibility.
In summary, we confirmed in the present study the well established differences in typical PPARα activities between rodent and nonrodent species. More interestingly, we have discovered, to our knowledge for the first time, that PPARα agonists are able to significantly induce CYP4A and ACOX in monkeys after extended treatment duration. An even more important finding was the observation that CYP2C and CYP3A mRNAs were strongly induced in monkey livers and human hepatocytes, whereas they were repressed in rat livers. In conclusion, these data show an excellent correlation between in vivo data on gene expression and activity level of DMEs with results generated in hepatocyte monolayer culture, enabling a reliable estimation of human P450 induction by EMD. This study also clearly highlighted major differences between primates and rodents in the regulation of all the major inducible liver P450s, with evidence of CYP3A and CYP2C inducibility by PPARα agonists in monkey and humans.
Acknowledgments
We thank Drs. Bernhard Ladstetter and Peter-Jürgen Kramer (Merck Serono) for supporting this study and Dr. Phil Hewitt (Merck Serono) for editing the manuscript. We also thank Drs. Francis Contard and Gilles Chavernac (Merck Serono) for providing pharmacological data on EMD, MDS Pharma Services (France) and Covance (UK) for performing the monkey studies, and Dr. Peter Tempel (Merck Serono) for performing the rat study. We thank Jean-Philippe Guenzi for technical assistance in performing the experiments.
Footnotes
-
Supported by European Centre for the Validation of Alternative Methods Grant 19471-2002-05-F1 ED ISP FR. L.R. and G.T. contributed equally to this work.
-
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
-
doi:10.1124/dmd.107.018358.
-
ABBREVIATIONS: P450, cytochrome P450; UGT, UDP-glucuronosyltransferase; DME, drug-metabolizing enzyme; AhR, aryl hydrocarbon receptor; PXR, pregnane X receptor; CAR, constitutive androstane receptor; PPAR, peroxisome proliferator-activated receptor; EMD, EMD 392949, (S)-4-O-tolylsulfanyl-2-(4-trifluormethyl-phenoxy)-butyric acid; EROD, 7-ethoxyresorufin-O-deethylase; PROD, 7-pentoxyresorufin-O-depentylase; ACOX, acyl-CoA oxidase; PCR, polymerase chain reaction; MDR, multidrug resistance; CPT, carnitine-palmitoyl transferase; CT, threshold cycle.
-
↵1 Current affiliation: AstraZeneca, Mölndal, Sweden.
- Received August 17, 2007.
- Accepted January 22, 2008.
- The American Society for Pharmacology and Experimental Therapeutics