Abstract
Muraglitazar (Pargluva), a dual α/γ peroxisome proliferator-activated receptor (PPAR) activator, has both glucose- and lipid-lowering effects in animal models and in patients with diabetes. This study describes the in vivo and in vitro comparative metabolism of [14C]muraglitazar in rats, dogs, monkeys, and humans by quantitative and qualitative metabolite profiling. Metabolite identification and quantification methods used in these studies included liquid chromatography/mass spectrometry (LC/MS), LC/tandem MS, LC/radiodetection, LC/UV, and a newly described mass defect filtering technique in conjunction with high resolution MS. After oral administration of [14C]muraglitazar, absorption was rapid in all species, reaching a concentration peak for parent and total radioactivity in plasma within 1 h. The most abundant component in plasma at all times in all species was the parent drug, and no metabolite was present in greater than 2.5% of the muraglitazar concentrations at 1 h postdose in rats, dogs, and humans. All metabolites observed in human plasma were also present in rats, dogs, or monkeys. Urinary excretion of radioactivity was low (<5% of the dose) in all intact species, and the primary route of elimination was via biliary excretion in rats, monkeys, and humans. Based on recovered doses in urine and bile, muraglitazar showed a very good absorption in rats, monkeys, and humans. The major drug-related components in bile of rats, monkeys, and humans were glucuronides of muraglitazar and its oxidative metabolites. The parent compound was a minor component in bile, suggesting extensive metabolism of the drug. In contrast, the parent drug and oxidative metabolites were the major components in feces, and no glucuronide conjugates were found, suggesting that glucuronide metabolites were excreted in bile and hydrolyzed in the gastrointestinal tract. The metabolites of muraglitazar resulted from both glucuronidation and oxidation. The metabolites in general had greatly reduced activity as PPARα/γ activators relative to muraglitazar. In conclusion, muraglitazar was rapidly absorbed, extensively metabolized through glucuronidation and oxidation, and mainly eliminated in the feces via biliary excretion of glucuronide metabolites in all species studied. Disposition and metabolic pathways were qualitatively similar in rats, dogs, monkeys, and humans.
The peroxisome proliferator-activated receptors (PPARs) are a set of nuclear hormone receptors. The two most intensively investigated subtypes have been PPARα (expressed primarily in the liver and plays a critical role in lipid metabolism) and PPARγ (predominantly expressed in adipose tissue and implicated in insulin sensitization as well glucose and fatty acid utilization). PPARα is the target of the fibrate class of hypolipidemic drugs such as fenofibrate (Balfour et al., 1990; Despres, 2001) and gemfibrozil (Packard et al., 2002), whereas PPARγ is the target of the thiazolidinedione (Mudaliar and Henry, 2001) class of antidiabetic drugs such as rosiglitazone (Balfour and Plosker, 1999; Cheng-Lai and Levine, 2000; Goldstein, 2000) and pioglitazone (Gillies and Dunn, 2000). Muraglitazar (N-[(4-methoxyphenoxy)carbonyl]-N-[[4-[2-(5-methyl-2-phenyl-4-oxazolyl)ethoxy]-phenyl]methyl]glycine), a novel dual PPARα/γ activator, is an oxybenzylglycine analog (nonthiazolidinedione) and has both glucose- and lipid-lowering effects when tested in animal models of diabetes and dyslipidemia and in patients with diabetes (Devasthale et al., 2005; Harrity et al., 2006; Kendall et al., 2006). Seventeen human oxidative metabolites of muraglitazar were identified after oral administration of [14C]muraglitazar (Zhang et al., 2006), and glucuronidation was the major clearance pathway of muraglitazar in humans (Wang et al., 2006).
This study was conducted to determine comparative metabolite profiles quantitatively and qualitatively in plasma, urine, bile, and feces from rats, dogs, monkeys, and humans after oral administration of [14C]muraglitazar and in incubations with hepatocytes from mice, rats, monkeys, and humans. Seven identified metabolites were synthesized for verification and tested for their PPARα/γ activation activity. Qualitative and quantitative profiling provides identities, amounts (or concentrations), and changes versus time of muraglitazar and its metabolites in plasma and in excreta (urine, bile, and feces) of rats, dogs, monkeys, and humans. The recently published biotransformation of [14C]muraglitazar in mice (Li et al., 2006) was included in some sections for species comparison.
Materials and Methods
Materials. [14C]Muraglitazar was prepared at Bristol-Myers Squibb and had a radiochemical purity of 99.5%. Metabolite standards M1, M5, M10, M11, and M15 were prepared as described previously (Wang et al., 2006; Zhang et al., 2006); M9 and M16 were also prepared at Bristol-Myers Squibb. Ecolite liquid scintillation cocktail was purchased from MP Biomedicals (Irvine, CA). Deepwell LumaPlate 96-well plates were purchased from PerkinElmer (Boston, MA). For solid phase extraction, Oasis HLB C-18 cartridge columns (5 ml) were obtained from Waters (Milford, MA). Polyethylene glycol-400 and sodium hydroxide were purchased from Sigma-Aldrich (St. Louis, MO). Acetonitrile and trifluoroacetic acid were purchased from EM Scientific (Gibbstown, NJ). All other chemicals used were of reagent grade or better. Cryopreserved hepatocytes from male cynomolgus monkeys and three human donors (pooled before use) were purchased from In Vitro Technologies (Baltimore, MD). Hepatocytes were stored in liquid nitrogen before use. Fresh hepatocytes from male Sprague-Dawley rats and male ICR/CD-1 mice were prepared in house using a literature method (Berry and Friend, 1969).
Study Subjects, Dosing, and Sample Collection. All animal housing and care conformed to the standards recommended by the Guide for the Care and Use of Laboratory Animals. Dogs and monkeys were non-naive animals. Animals completed at least 4 days of quarantine or conditioning before dose administration. LabChow or LabDiet (Purina Mills, St. Louis, MO) was fed to dogs and monkeys daily. LabDiet was available to rats and mice ad libitum. Tap water was provided to animals ad libitum. Mice were fasted for 4 h and rats, dogs, monkeys, and humans were fasted overnight before dose administration. Food was returned at 4 h postdosing for rats, dogs, and monkeys. All human subjects were required to give informed and written consent before participation in this study. The extrapolated radiation dose equivalent for tissues in humans after a 100-μCi oral dose ranged from 0 to 33.2 millirem based on the tissue distribution study in rats (data not shown). This level is >75 times lower than the annual allowable exposure limit. Therefore, the administration of 100 μCi of [14C]muraglitazar to healthy male subjects was considered safe. The dose was prepared by dissolving [14C]muraglitazar in polyethylene glycol-400. Plasma, urine, bile, and fecal samples were obtained after a single oral administration of [14C]muraglitazar to rats, dogs, monkeys, or humans.
Rat. After oral administration of [14C]muraglitazar to male Sprague-Dawley rats (11 mg/kg, 8.4 μCi/mg), blood (1-, 4-, and 12-h) was obtained from rats (three rats per collection time) by terminal bleeding via cardiac puncture after carbon dioxide anesthesia. Urine (0- to 12-, 12- to 24-, and 24-h intervals thereafter for 0–168 h) and feces (24-h intervals for 0–168 h) were collected from three rats on dry ice. In a separate biotransformation study, bile, urine, and feces (0- to 24-h) were collected from three bile duct-cannulated (BDC) male rats for 0 to 24 h after oral administration of [14C]muraglitazar (10 mg/kg, 5 μCi/mg). For bile collection, a bile salt solution (18 mg/ml cholic acid and 1.1 mg/ml sodium bicarbonate in saline, pH 7.2) was infused via the duodenal cannula at 1 ml/h for BDC rats. For the pharmacokinetic study, blood (300 μl) was collected from three rats via a jugular vein cannula at 0, 0.25, 0.5, 1, 2, 4, 6, 8, 12, 24, 48, 72, 96, 120, 144, and 168 h postdose.
Mouse. Bile, urine, and feces were collected (0–24 h) from 5 bile ductcannulated male CD-1 mice after an oral dose of [14C]muraglitazar (1 and 40 mg/kg, 11.2 μCi/mg). For bile collection, a solution of taurocholic acid (2.3 mg/ml in 0.9% saline) was infused via the duodenal cannula at 0.3 ml/h for BDC mice (Li et al., 2006).
Dog. Blood (10 ml at 1, 4, and 12 h) was collected from the indwelling venous catheter from three male beagle dogs after oral doses of [14C]muraglitazar (1.9 mg/kg, 2.8 μCi/mg). Urine (0- to 12-, 12- to 24-, and 24-h intervals thereafter for 0–168 h) and feces (24-h intervals for 0–168 h) were also collected on dry ice. In addition, another series of blood samples (2 ml) was collected at 0, 0.25, 0.5, 1, 2, 4, 6, 8, 12, 24, 48, 72, 96, 120, 144, and 168 h postdose for pharmacokinetic analysis.
Monkey. Blood (5 ml at 1, 4, and 12 h) was collected from the indwelling venous catheter from three male cynomolgus monkeys after oral doses of [14C]muraglitazar (2 mg/kg, 6.6 μCi/mg). Urine (0- to 12-, 12- to 24-, and 24 h-intervals thereafter for 0–168 h) and feces (24-h intervals for 0–168 h) were also collected on dry ice. In a separate study, bile, urine, and feces were collected (0–8, 8–24, and 24–48 h for 0–48 h) from three bile ductcannulated male cynomolgus monkeys after an oral dose of [14C]muraglitazar (5 mg/kg, 6 μCi/mg). For bile collection, a bile salt solution (18 mg/ml cholic acid and 1.1 mg/ml sodium bicarbonate in saline, pH 7.2) was administered via a distal (flushing) catheter at 1 ml/kg/h for BDC monkeys. In addition, another series of blood samples (2 ml) was collected at 0, 0.25, 0.5, 1, 2, 4, 6, 8, 12, 24, 48, 72, 96, 120, 144, and 168 h postdose for pharmacokinetic analysis.
Human. Blood (10 ml at 1, 4, and 12 h) was obtained by venipuncture from six healthy male subjects after a single 10-mg oral dose of 100 μCi. Urine (0- to 8-, 8- to 24-, and 24-h interval thereafter for 0–240 h) and feces (24-h intervals for 0–240 h) were obtained. In addition, another series of blood samples (2 ml) was collected at 0, 0.25, 0.5, 1, 2, 4, 6, 8, 12, 24, 48, 72, 96, 120, 144, and 168 h postdose for pharmacokinetic analysis. Bile was collected over 3 to 8 h postdose from four subjects after an oral 20-mg (100-μCi) dose of [14C]muraglitazar (Wang et al., 2006). An intravenous dose of cholecystokinin was used to stimulate gallbladder contraction at 7 h postdose in humans.
To stabilize acylglucuronide metabolites in bile samples, acetic acid was added to 2 or 5% (v/v) after bile sample collection before sample freezing at –20°C. The blood was collected in tubes containing EDTA and centrifuged within 30 min of collection to harvest plasma (10 min, 1300g at 4°C). Animal fecal homogenate was prepared by mixing water/ethanol (50:50 v/v) and feces (5:1 w/w) followed by homogenization. A paste of human feces was prepared as homogenate before analysis for radioactivity and sample pooling. Urine and fecal homogenate were pooled by combining urine and feces obtained during each collection interval (1% for feces and 2% for urine). A pooled plasma sample was prepared by combining equal volumes of plasma (1–2 ml) from all subjects for each collection time point. Pooled plasma, urine, and feces were analyzed for radioactivity distribution and metabolite identification by LC/MS/MS. A description of the studies and recovery of radioactive doses in animals and humans are shown in Table 1. The dose recovery data from humans with bile collection were presented previously and are included here for comparison (Wang et al., 2006).
Incubations of [14C]Muraglitazar in Hepatocytes. Cryopreserved hepatocytes from monkeys and humans were thawed as recommended by the manufacturer. [14C]Muraglitazar (25.6 μg in 10 μl of methanol) was added to 1-ml portions (final drug concentration = 25.6 μg/ml) of cell suspensions in Krebs-Henseleit buffer in open 22-ml glass vials. The cell concentrations were 3, 3, 2, and 2 million cells/ml for hepatocytes from mice, rats, monkeys, and humans, respectively. The cell viability before incubation was 65, 92, 78, and 82% for mouse, rat, monkey, and human hepatocytes, respectively. The reaction vials were incubated with shaking (120 rpm) for 4 h in a 37°C chamber. The chamber was maintained under an oxygen/carbon dioxide (95:5%) atmosphere. A 4-h incubation of [14C]muraglitazar (25.6 μg/ml) in buffer served as a negative control. At the end of the incubations, acetonitrile (2 ml) was added to each sample, and the mixtures were sonicated for 5 min in a Branson 5210 sonicator. The samples were centrifuged at 3000g for 10 min. The supernatants were removed and aliquots were taken for radioactivity counting. The radioactivity recoveries were 95.6, 96.5, 99.2, and 98.7%, respectively, from mouse, rat, monkey, and human hepatocyte incubations. The remainder of supernatant was evaporated to near dryness under a stream of nitrogen. The residues were reconstituted in 0.2 ml of water/acetonitrile (2:1 v/v) for high-performance liquid chromatography (HPLC)/UV, LC/MS/MS, and LC/radiodetection.
Radioactivity Analysis and HPLC. The radioactivity detection and HPLC analysis were performed as described previously (Zhang et al., 2006). Under these conditions, the synthetic metabolite standards, M1, M5, M9, M10, M11, M15, and M16 each had the same retention times (monitored by UV at 275 nm) as the corresponding metabolites in the human fecal extract (monitored by radioactivity detection). These metabolite standards also had the same MS fragmentation patterns as the in vivo metabolites.
Preparation of Biological Samples for Metabolite Profiling.Plasma. Pooled plasma was extracted by addition of 2 ml of acetonitrile to 1 ml of plasma, vortexed for 1 min, and centrifuged at 2000g for 10 min (Eppendorf 5810 R centrifuge, Eppendorf Company, Hamburg, Germany). The supernatant was removed and saved. The pellet was treated with an additional three extractions with 2 ml of acetonitrile/water (2:1 v/v) and all supernatants were combined. Aliquots of 100 to 300 μl of this solution were mixed with Ecolite cocktail and counted for radioactivity. After the solvent was evaporated to dryness under nitrogen, the residue was redissolved in 0.20 ml of water/acetonitrile (2:1 v/v) and centrifuged at 2000g for 10 min before injecting onto the HPLC column.
Urine. Pooled urine (5 ml) was concentrated either by passing through a 5-ml C18 cartridge (Oasis; Waters) or by direct evaporation under a stream of nitrogen. The cartridge was eluted with methanol after washing with a half-column volume of water and the eluate was concentrated under a stream of nitrogen. The residue from direct evaporation of the urine or from the evaporation of the methanol eluate was suspended in 0.20 ml of water/acetonitrile (2:1, v/v) and centrifuged at 2000g for 10 min, and the supernatant (50 μl) was subjected to HPLC analysis.
Feces. Pooled fecal homogenate was extracted by addition of 4 ml of acetonitrile to approximately 1 g of fecal homogenate, vortexed for 5 min, sonicated for 5 min, and centrifuged at 2000g for 30 min. The supernatant was removed and saved. The extraction was repeated twice with 2 ml of acetonitrile/water (2:1, v/v) and the supernatants were combined. Aliquots of 0.5 ml of the combined supernatant were mixed with Ecolite cocktail and counted for radioactivity. The combined supernatant was evaporated to dryness under a stream of nitrogen. The residue was suspended in 0.20 ml of water/acetonitrile (2:1 v/v) and centrifuged at 2000g for 10 min before subjecting 50-μl aliquots to HPLC analysis.
Bile. Pooled bile samples were mixed with 1 volume of acetonitrile and centrifuged at 2000g for 10 min. The supernatant (10–30 μl) was analyzed directly.
LC/MS for Quantification of Muraglitazar in Plasma. The method used trazodone as the internal standard (IS). After the addition of 0.2 ml of acetonitrile containing 100 ng/ml IS to 0.1 ml of plasma sample, methylene chloride (0.1 ml) was added. The samples were vortexed for 2 min and then centrifuged at 14,000 rpm for 3 min. The organic phase was removed and evaporated to dryness under nitrogen at 40°C. The residue was reconstituted with 0.2 ml of HPLC mobile phase A and 20 μl of the sample was injected onto the LC/MS apparatus. Chromatographic separation was achieved on a Waters 2690 HPLC system using a BDS Hypersil C18 analytical column (2 × 50 mm, 3 μm) at a flow rate of 0.3 ml/min. An isocratic system was used with 90% mobile phase A (acetonitrile/0.1% formic acid, 50:50 v/v) and 10% mobile phase B (acetonitrile/0.1% formic acid, 95:5 v/v). Analysis (m/z 517.1 for muraglitazar and m/z 372.0 for IS) was accomplished by positive turbo ion spray on a Sciex API 100 mass spectrometer (Applied Biosystems, Foster City, CA).
Pharmacokinetic Analysis. The plasma concentration versus time data for radioactivity and unchanged muraglitazar were analyzed by a noncompartmental method (Gibaldi and Perrier, 1982). The peak plasma concentration, Cmax, and the time to reach peak concentration as the first occurrence, Tmax, were recorded directly from experimental observations. The area under the plasma concentration versus time curve (AUC) was calculated by a combination of the trapezoidal methods. The AUC was calculated from time 0 to the time, T, of the last measurable concentration (AUC0-T). The first-order rate constant of decline of radioactivity concentrations and unchanged muraglitazar, expressed as equivalents of muraglitazar, in the terminal phase of each plasma concentration versus time profile, K, was estimated by log-linear regression (using no weighting factor) of at least three data points, which yielded a minimum mean square error. The absolute value of K was used to estimate the apparent terminal elimination half-life, t1/2.
LC/MS and LC/MS/MS for Identification of Metabolites. A YMC ODS AQ C18 column (2.0 × 150 mm, 5 μm) was used for LC/MS analysis. The flow rate was 0.28 or 0.3 ml/min. LC/MS analysis was performed on an LCQ, an LTQ (Thermo Electron Corporation, Waltham, MA) or a Micromass Q-TOF Ultima mass spectrometer (Waters) with an electrospray ionization source. Samples were analyzed in the positive ion mode. The capillary temperature used for the LCQ and LTQ was set at 230°C. The desolvation temperature used on the Q-TOF was 300°C. The nitrogen gas flow rate, and spray and cone voltages were adjusted to give maximum sensitivity for muraglitazar. The Q-TOF was tuned to 18,000 resolution at half-peak height using an insulin tuning solution, and was calibrated up to 1500 Da using a polyalanine calibration solution. For accurate mass measurement, the m/z 556.2771 of an infused 20 ng/μl leucine enkephalin solution was used as lock mass. The experimentally obtained masses were within 5 mDa compared with their respective calculated values.
Total Ion Chromatograms after Mass Defect Filtering Analysis. The data files from the Q-TOF high resolution mass analysis of fecal extracts of humans, monkeys, and rats were processed using the mass defect filtering method (Zhang et al., 2003). The centroid raw data were processed using a mass defect filter designed to retain ion species with mass defects between 0.006 and 0.106 Da in a mass range of 250 to 850 Da. The filter was based on the apparent mass of muraglitazar (517.1975 Da) in the uncorrected centroid raw data file.
Radioactivity Profiles and Quantification of Metabolites. The average cpm value from a baseline section (2–3 min) of the chromatogram was subtracted from the cpm value of each fraction. Biotransformation profiles were prepared by plotting the cpm values against time after injection using Microsoft Excel software (Microsoft, Redmond, WA). Radioactive peaks in the biotransformation profiles were reported as a percentage of the total radioactivity collected during the entire HPLC run. The relative distributions of radioactive metabolites in urine, bile, and feces were calculated from the percentage of dose excreted in the matrix (based on dpm by scintillation counting) multiplied by the percentage of distribution of metabolites in chromatograms of the matrix (based on cpm counting by TopCount; PerkinElmer).
Determination of PPARα/γ Binding and Transactivation Activities of Muraglitazar Metabolites. The in vitro PPAR biological activity of the seven synthetic muraglitazar metabolites was evaluated and PPARα/γ binding and transactivation activity values were determined as described previously (Devasthale et al., 2005).
Results
Excretion of Radioactive Dose. The recovery values of radioactive dose in urine, bile, and feces of rats, mice, dogs, monkeys, and humans are shown in Table 1. Urinary excretion of radioactivity was low (<3.7% of dose) in all intact species, and the primary route of excretion was feces (87.7, 89.0, 80.4, and 61.6% of dose for intact rats, dogs, monkeys, and humans, respectively). The bile contained the majority of radioactivity from rats, mice, monkeys, and humans during the given collection periods, representing the major elimination pathway of [14C]muraglitazar in these four species. Based on recovered dose in urine and bile, the absorption of muraglitazar represented at least 66, 82, 41, and 44% in rats, mice, monkeys, and humans, respectively, after oral administration. The overall recovery of radioactivity in the human study of 10-mg dose, in which fecal paste was prepared, was low (64.2%) and the reason for the low recovery was not known. An additional study was conducted to address the radioactivity recovery issue by preparing fecal homogenate as slurry in 50:50 (v/v) ethanol/water containing 2% acetic acid after an oral 20-mg dose of [14C]muraglitazar. The radioactive dose recovery from that study was 3.7% in urine, 50.7% in the feces, and 39.9% of dose in the 3- to 8-h bile collection (Wang et al., 2006). The fecal metabolite profiles were qualitatively similar between the two human C-14 studies. It was, therefore, concluded that the metabolite profiling results from the samples of the 10-mg dose study were valid despite the lower than an anticipated recovery. Radioactivity excretion profiles of [14C]muraglitazar in mice after oral administration were similar to other species (Li et al., 2006).
Figure 1 shows the proportion of the doses excreted in urine and feces in each of the collection periods in rats, dogs, monkeys, and humans. The significant portions of the excretion of radioactivity in urine occurred during the first 8 to 12 h after the dose administration. The concentrations of muraglitazar equivalents (parent drug plus metabolites, calculated using the actual volumes of urine collected) in rat, monkey, and human urine during this period of urinary excretion are shown in Table 2 (Urine). The concentrations of drug-related radioactivity in rat, dog, and monkey urine were approximately 9, 1, and 3 times that in human urine. The rate of fecal excretion in rats and dogs was rapid, with approximately 85% recovered within 24 h and nearly 95% within 48 h postdose (Fig. 1B). In contrast, the rate of fecal excretion in monkeys was slower, with approximately 30% recovered within 48 h and 75% within 96 h postdose. The rate of fecal excretion in humans was similar to that in monkeys, with 25% within 48 h and 80% within 96 h postdose.
The majority of the excretion of radioactivity in bile probably occurred during the first 24 to 48 h after dose administration. The average radioactive dose recovery value in 0- to 24-h bile of BDC rats was 65.1% and in the corresponding feces of these rats was 0.55%. The radioactive dose recovery values in 0- to 48-h bile of three BDC monkeys were 15.4, 16.9, and 75.2%, and in corresponding feces of these three monkeys were 30.6, 6.7, and 7.7%. The concentrations of muraglitazar equivalents (parent drug plus metabolites) in rat, mouse, monkey, and human bile within specified collection periods are shown in Table 2 (Bile). The concentrations of drug-related radioactivity in rat, mouse, and monkey bile were approximately 4, 5.5 to 190, and 0.1 times that in human bile. Human subjects seemed to show a fast biliary elimination since approximately 40% of dose was found in bile during the 3- to 8-h collection after dose.
Pharmacokinetic Results. The pharmacokinetic parameters for the unchanged drug and radioactivity are presented in Table 3. The animal and human plasma concentration-time profiles for radioactivity and muraglitazar are presented in Fig. 2. After oral administration, the radioactivity and muraglitazar concentrations reached a maximum at 0.5 to 1 h in rats, dogs, monkeys, and humans. Although the trends declined in dogs and humans after reaching Cmax, the radioactivity and muraglitazar concentrations reached a second peak at approximately 6 to 8 h in rats and monkeys, suggesting the potential of enterohepatic recycling of muraglitazar and radioactivity in rats and monkeys. The mean terminal elimination t1/2 values of plasma radioactivity and muraglitazar were longer in monkeys and humans (18–35 h) than those from rats and dogs (4–10 h). The longer t1/2 values for the parent compound relative to total radioactivity in rats, dogs, and humans were probably due to truncation of the radioactivity AUC curves as a result of lower assay sensitivity of radioactivity compared with LC/MS-based parent determination. The overall kinetic profiles and AUC values for radioactivity are similar to those of muraglitazar in all species.
Identification of [14C]Muraglitazar Metabolites.Table 4 shows the proposed structures for the muraglitazar metabolites identified in rats, dogs, monkeys, and humans after oral administration of [14C]muraglitazar. Most of the metabolite identification was done from LC/MS/MS analysis. Seven of the muraglitazar metabolite standards were also synthesized and matched the metabolites in human feces by retention times and mass fragmentation patterns. Detailed LC/MS/MS identification of the metabolites in mice and humans has been presented previously (Wang et al., 2006; Zhang et al., 2006), and identification of muraglitazar metabolites in rats, dogs, and monkeys matched those identified in mice and humans. Typical fragmentation patterns of muraglitazar in full-scan MS/MS analysis showed a cleavage adjacent to the carbamate to give a fragment at m/z 292 and cleavage at the ether bond to give a fragment at m/z 186. Similar fragmentation patterns were observed with muraglitazar metabolites M2 to M16. All monohydroxylation metabolites (M2, M5, M6, M7, M8a, M10, M11, and M14) showed a fragmentation ion at m/z 202 (202 = 186 + 16) or its dehydrated ion at m/z 184 (184 = 202–H2O), indicating that oxygenation was on the oxazole-ring side of the molecules.
Metabolite M1. This radioactive peak is a minor metabolite in in vivo samples but a major metabolite in hepatocyte incubations. M1 had poor ionization properties in a mass spectrometer (the parent and all other metabolites do not show ionization of this part of the molecule). M1 was assigned based on comparing retention time of the radioactive peak with the metabolite standard as O-dealkyl muraglitazar.
Metabolite M22. M22 had an accurate mass of 549.1864 Da and a formula of C29H29N2O9, consistent with a dioxygenation of muraglitazar. Since M22 was only observed in in vitro incubations and not observed in in vivo samples, M22 was not further characterized.
Metabolite M23. From hepatocyte incubations, 5-methyl-2-phenyl oxazoyl-4-acetic acid was also identified as a nonradiolabeled metabolite resulting from the dealkylation of muraglitazar. This metabolite had a measured accurate mass of 218.0820 and a formula of C12H12NO3 (expected mass of 218.0817). M23 was identified as 5-methyl-2-phenyl oxazoyl-4-acetic acid. The same dealkylation reaction also produced M1.
Metabolite Profiles in Plasma. The extraction recovery of radioactivity from all plasma samples was >98%. The HPLC radiochromatographic profiles of 1-h plasma from rat, dog, monkey, and human are shown in Fig. 3. The relative distribution of radioactive metabolites in pooled plasma is summarized in Table 5. The metabolic profiles of plasma were qualitatively similar at different time points within species and across species. Muraglitazar was the major component at 1, 4, and 12 h, accounting for more than 85% of plasma radioactivity at 1 h in all four species.
The prominent metabolite in rat plasma was M15, which accounted for 2 to 4% of the sample radioactivity. Several metabolites were in rat plasma and each accounted for less than 0.4% of the radioactivity. The parent compound in dog plasma represented 90.5% at 1 h, 81.3% at 4 h, and 33.0% at 12 h of the sample radioactivity. Prominent metabolites in dog plasma included M1, M5, M15, and M20. Major metabolites in monkey plasma were M5, M11, and M18. M11 accounted for 8.0% of the plasma radioactivity at 1 h, and increased to 16.3% at 4 h and 21.4% at 12 h. M5 accounted for 1.5 to 2.9% of the plasma radioactivity. Metabolite M18 represented 3 to 4% of plasma radioactivity at 4 and 12 h in monkeys. Other metabolites observed in monkey plasma were minor. M13 was a major metabolite in human plasma, but no metabolite accounted for more than 4.5% of the plasma radioactivity at all time points. Similarly, the parent drug was the major circulating drug-related component in mice (Li et al., 2006). All metabolites (M1, M2, M5, M8, M8a, M10-13, M15, and M18) observed in human plasma were also present in rats, dogs, or monkeys.
Metabolite Profiles in Urine. Urinary excretion was a minor pathway, representing less than 5% of the dose in all studied intact species after oral administration of [14C]muraglitazar. The HPLC radiochromatographic profiles of pooled urine from rats, dogs, monkeys, and humans are shown in Fig. 4, and metabolite distribution in rat, dog, monkey, and human urine is listed in Table 6. Several very minor components (0.3–0.8% of the dose) were not sufficiently abundant for characterization. Rat urine contained many radioactive peaks. The prominent rat urinary metabolites included M1, M2, M10, M15, and M21. One major radioactive peak, M1, was observed in dog urine. Metabolites M1, M21, and M11 were prominent radioactive peaks in monkey urine, accounting for approximately 1, 0.3, and 0.6% of the dose, respectively. The low level of radioactivity in human urine was distributed in several radioactive peaks, of which the glucuronide M13 accounted for 1% of the dose and the parent accounted for 0.3% of the dose. All human urinary metabolites were found in animal urine samples.
Metabolite Profiles in Feces. Feces was the major elimination route after oral administration of [14C]muraglitazar to intact species. The HPLC radiochromatographic profiles of pooled feces are shown in Fig. 5 and the metabolite distribution in rat, dog, monkey, and human feces is listed in Table 6. Muraglitazar was one of the abundant components present in feces from all species. The most abundant metabolite was different in each species; however, each of these metabolites was hydroxylated, and O-demethylated products were found in the plasma of that species.
Rat feces contained prominent radioactive peaks M2, M5, M8a, M9, M15, and parent. The parent drug (P) accounted for 13% of the dose in the rat feces. M15, the most abundant peak, accounted for approximately 34% of the dose. Dog feces contained two major radioactive peaks, M15 and P, representing 25 and 53% of the dose, respectively. No other radioactive peaks accounted for more than 2% of the dose in dog feces. Prominent metabolites in monkey feces included M6, M8, M10, M11, M15, and P, each representing 5 to 26% of the dose. Major metabolites in the 0- to 240-h pooled human feces included M4, M5, M10, M11, M15, and M16, each of which accounted for 3 to 11% of the dose. The parent compound accounted for 9.6% of the dose. Radioactivity distribution of 0- to 48-h pooled human feces was also determined and the parent compound represented approximately 40% of the sample radioactivity that was higher than that in 0- to 240-h pooled human feces (data not shown).
Metabolite Profiles in Bile Samples. The HPLC-radiochromatograms of pooled bile samples from rats, monkeys, and humans are shown in Fig. 6. Within 24 h after oral administration of [14C]muraglitazar to BDC rats, 65% of the dose was excreted in bile and <1% of the dose was excreted in feces. The parent drug represented <5% of the dose in the rat bile, suggesting extensive metabolism of muraglitazar in rats. The glucuronide of O-demethyl muraglitazar (M20), the glucuronide of muraglitazar (M13), and O-demethyl muraglitazar (M15) were the major metabolites in rat bile, each accounting for 11 to 40% of the dose (Fig. 6). Under the conditions used for sample collection and storage (2–5% acetic acid), the acylglucuronide metabolites appeared to be chromatographically separated as single peaks, which suggested that minimal acyl migration occurred.
Within 48 h after oral administration of [14C]muraglitazar to BDC monkeys, the average excretion of radioactive dose was 36% (range of 15–75%) in bile and 15% in feces. Metabolism of [14C]muraglitazar was extensive with small amounts of unchanged [14C]muraglitazar in the monkey bile. M13 was the prominent metabolite in bile, accounting for approximately 14% of the dose; all other metabolites were less than 5% of the dose in bile. M15 and [14C]muraglitazar were the major radioactive components, corresponding to approximately 2.6 and 3.3% of the dose in the BDC monkey feces. Glucuronides were not present in feces; the major metabolites in bile were glucuronide conjugates. Metabolites M9, M10, and M11 were also prominent components in feces (data not shown), suggesting bile leakage to intestines in which the glucuronides were hydrolyzed to muraglitazar and oxidative metabolites.
After oral administration of [14C]muraglitazar to human subjects, approximately 40% of the dose was found in the bile collected over 3 to 8 h postdose using an oral-gastro-duodenal tube (Wang et al., 2006). More than 60% of the radioactivity in bile was present as the glucuronide of [14C]muraglitazar (M13). Fecal samples contained mostly oxidative metabolites and little glucuronide of the parent drug.
Similarly, muraglitazar glucuronide (M13) was also a major metabolite in the mouse bile but was not found in the BDC mouse feces (Li et al., 2006). In addition to those metabolites (M1-M21) identified in other species (rats, dogs, monkeys, and humans), unique metabolites identified in mice included the taurine conjugates of muraglitazar and its metabolites formed from O-demethylation, hydroxylation and dihydroxylation, and glutathione conjugates (Li et al., 2006).
Metabolite Profiles in Hepatocyte Incubations. The distribution of radioactive peaks presented in each hepatocyte incubation is summarized in Table 7. The parent drug was the major component in all incubations. The control incubation of [14C]muraglitazar also contained M1, M22, and a minor peak that had the same retention time as M10. Identification of in vitro metabolites was confirmed by LC/MS/MS analyses. Twenty-one radioactive metabolites (M1-M8, M10-M15, M17-M22, and M8a) were detected in the hepatocyte incubations, and the most prominent metabolites were M1, M6, M10, M11, M12, M13, M14, and M15. Eleven radioactive peaks, M1, M6-7, M10-15, M18, and M22, were observed from the incubation with human hepatocytes. Qualitatively similar metabolic profiles were obtained from the incubations with hepatocytes from mice, rats, and monkeys. Figure 7 shows the LC/UV profiles of [14C]muraglitazar metabolites in hepatocytes of mice, rats, monkeys, and humans. The LC/UV profiles were qualitatively similar to the radioactivity profiles of these hepatocyte incubations. All human metabolites of [14C]muraglitazar were detected in incubations in hepatocytes from at least one toxicological animal species. Subsequent studies examining the stability of M13 in neutral pH aqueous buffer systems demonstrated that the metabolite hydrolyzed (data not shown). Although the rate of hydrolysis was not measured under the exact conditions of the hepatocyte incubations, it is likely that there would be significant hydrolysis of M13 leading to under-representation of this pathway.
Summary of Metabolite Profiling Results. Parent compound was the major circulating drug-related component in rats, dogs, monkeys, and humans after oral administration of [14C]muraglitazar. Fecal excretion of [14C]muraglitazar and its metabolites was the major elimination pathway (>95% of the recovered dose) in intact species, principally via biliary elimination when with bile collections. The results suggested that biliary elimination (as glucuronides) was the major clearance pathway of [14C]muraglitazar and that the glucuronide metabolites were hydrolyzed in the intestine before excretion in feces. Urinary excretion represented a minor elimination pathway (<5% of the recovered dose) in intact species. Biotransformation profiles of [14C]muraglitazar were qualitatively similar in rats, dogs, monkeys, and humans. Major metabolic pathways of muraglitazar were glucuronidation, aliphatic/aryl hydroxylation, O-dealkylation, and a combination of O-demethylation and hydroxylation, and oxazole-ring opening. Biotransformation profiles of [14C]muraglitazar were qualitatively similar in hepatocytes of mice, rats, monkeys, and humans. In vitro metabolism of [14C]muraglitazar was qualitatively similar to its in vivo metabolism.
LC/MS Profiling of Muraglitazar Metabolites.Figure 8 shows the ion chromatograms of total mass spectral signals versus HPLC run time after mass defect filtering treatment of LC/accurate mass analysis of human, monkey, and rat feces. The relative intensities of the total ion peak of each metabolite and the parent resemble qualitatively those peaks in the radiochromatograms of these samples. Full-scan mass spectra and MS/MS spectra were derived from these ion chromatograms for identification of metabolites. Mass defect filter-treated total ion chromatograms were especially useful to identify molecular ions of each metabolite because of the effective removal of endogenous interfering ions (Zhang et al., 2003).
PPARα/γ Activities of Muraglitazar Metabolites.Table 8 shows the PPARα/γ activities of muraglitazar metabolites. Based on these data from the PPARα and γ binding and transactivation assays, the muraglitazar metabolites M1, M5, M9, M15, and M16 show little or no functional PPARα or PPARγ agonist activity in the PPARα and PPARγ receptor transactivation assays in the human embryonic kidney cell line in comparison with the parent compound muraglitazar. Metabolite M10 shows no functional PPARγ agonist activity in the transactivation assay, but it did have partial PPARα agonist activity in the PPARα transactivation assay. Metabolite M11, which also had minimal functional PPARγ agonist activity in the transactivation assay, showed low levels of reasonable PPARα agonist activity in the PPARα transactivation assay, with an EC50 value of 0.260 μM, 7-fold less potent than muraglitazar (PPARγ EC50 = 0.035 μM). In conclusion, none of the seven muraglitazar metabolites showed significant PPARγ agonist functional activity, and only one metabolite showed any significant PPARα agonist functional activity (M11, which is 7-fold less potent than muraglitazar based on a comparison of EC50 values from the transactivation assay). The lack of significant PPARα/γ activities for these seven synthetic metabolites suggests that these metabolites will have little or no contribution to the clinical efficacy of muraglitazar.
Discussion
After oral administration of [14C]muraglitazar, the parent compound was the major circulating component in plasma of rats, dogs, monkeys, and humans. AUC values for muraglitazar and total radioactivity are similar in rats, dogs, and humans, although the parent AUC represented approximately 40% of the radioactivity AUC in monkeys, suggesting that the radioactivity was mainly attributed to parent compound. HPLC/radioactivity profiling of selected plasma samples confirmed that muraglitazar was the major circulating component, and there were limited amounts of circulating metabolites. Biliary elimination followed by fecal excretion was the major elimination pathway in all species, whereas urinary elimination was minor. All human metabolites were found in samples from at least one of the toxicological animals (rats or monkeys). The metabolic pathways of muraglitazar included glucuronidation, aliphatic/aryl hydroxylation, O-dealkylation, a combination of O-demethylation and hydroxylation, and oxazole-ring opening. Although the mechanism is not known, oxazole-ring opening appeared to lead to the formation of two stable metabolites, M9 and M16, in all species tested. Both muraglitazar glucuronide (M13) and O-demethyl muraglitazar glucuronide (M20) were major metabolites in the bile of BDC rats. Muraglitazar glucuronide (M13) was the major metabolite in the bile of BDC monkeys. However, oxidative metabolites were predominant in feces of intact animals, suggesting that the glucuronides were hydrolyzed in the intestine. Similarly, most of the radioactive components in human bile were glucuronide metabolites, whereas oxidative metabolites were found in human feces (Wang et al., 2006). The biliary excretion and intestinal hydrolysis of the glucuronide metabolites provide a potential for enterohepatic recycling of muraglitazar and its oxidative metabolites. However, no obvious evidence for this process was observed in the plasma radioactivity time profiles for humans and dogs, although a second peak for both muraglitazar and total radioactivity appeared at 6 to 8 h postdose in rats and monkeys. The second peak for the parent and radioactivity in rats is probably not due to enterohepatic recirculation inasmuch as the rat is a continuous bile secretor. The effect may be due to regional differences in absorption or in changes of drug solubility in the rat intestines after food was reintroduced to the fasted animals at approximately 4 h postdose. Enterohepatic recycling may lead to the second peak in monkeys and may occur in all species, but it appeared to be minor in humans (Wang et al., 2006).
The biotransformation of [14C]muraglitazar was also investigated in intact and BDC CD-1 mice after oral administration of 1 and 40 mg/kg (Li et al., 2006). Similar to that in rats, monkeys, and humans, biliary excretion in mice was the major route of elimination, accounting for >66% of the dose during a 0- to 48-h collection period. Urinary excretion accounted for 7 to 16% of the dose in mice. Besides those metabolites (M1-M21) identified in other species (rats, dogs, monkeys, and humans), metabolites identified only in mice included the taurine conjugates of muraglitazar and its metabolites formed from O-demethylation, hydroxylation and dihydroxylation, and glutathione conjugates. None of these taurine and glutathione conjugates were found in the bile samples of rats, monkeys, or humans. The major drug-related component circulating in plasma was the parent compound for up to 4 h postdose.
When muraglitazar was incubated with hepatocytes of mice, rats, monkeys, and humans at a concentration of 2 to 10 times Cmax values determined in these species, the majority of metabolites resulted from oxidation pathways including hydroxylation and O-demethylation, which were also observed in vivo. A relatively small fraction of metabolites resulted from glucuronidation to form M13 and M18 in these hepatocyte incubations, which underestimated the in vivo importance of the glucuronidation pathway. This is likely due to the instability of the glucuronide metabolite in the hepatocyte incubations. In contrast, the glucuronide metabolites in plasma and bile samples were relatively stable under the acidified and low temperature conditions.
In general, both quantitative and qualitative metabolite profiles are important to understand the relative abundance and identities of metabolites in various matrices from different species. The metabolites of muraglitazar in plasma, urine, feces, and bile as well as hepatocyte incubations were profiled quantitatively and qualitatively through LC/radioactivity detection, LC/UV, LC/MS, and LC/MS/MS analysis. Metabolites identified in each species were assigned as the same structure based on HPLC retention time and MS/MS spectral properties. Qualitatively and quantitatively similar metabolite profiles of muraglitazar in rats, dogs, monkeys, and humans support the idea that these animal species were valid models to evaluate the safety profile of muraglitazar.
Mass defect filtering analysis not only generates metabolite identification data (mass spectra with minimal endogenous interferences) (Zhang et al., 2003), but also provides total ion chromatograms that qualitatively resemble the metabolite radiochromatograms of fecal extracts of humans, monkeys, and rats (Fig. 8). Although total ion chromatograms obtained after the mass defect filtering analysis of high-resolution MS data, representing mass spectral signals or total ion signals acquired at each time point plotted against HPLC run time, resemble the radiochromatograms of human, monkey, and rat feces, these ion chromatograms should not be used to fully quantify the amounts of metabolites. However, the close resemblance to the radiochromatograms of the fecal samples provides the first potentially useful LC/MS-based estimation method for relative abundance of metabolites. This metabolite quantity estimation method would be especially useful when a validated quantitation method is not available, yet quantity information is important to know in certain phases of drug development, such as the first-in-man study for metabolite profile in plasma. Therefore, LC/high-resolution MS and LC/MS/MS methods were complementary to quantitative profiling using radioactivity detection when the existence of a metabolite needed to be qualified.
In summary, the current study describes the disposition and comparative metabolism of [14C]muraglitazar in rats, mice, dogs, monkeys, and humans by qualitative and quantitative metabolite profiling. [14C]Muraglitazar was rapidly absorbed and circulated as the major drug-related component in all species. The compound was extensively metabolized by both oxidative and conjugative pathways, and eliminated mainly through biliary excretion as glucuronide conjugates of the parent and its oxidative metabolites.
Acknowledgments
We thank Kamelia Behnia for preparation of rat and mouse hepatocytes and Geoffrey Kinnel for technical assistance.
Footnotes
-
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
-
doi:10.1124/dmd.106.012450.
-
ABBREVIATIONS: PPAR, peroxisome proliferator-activated receptor; BDC, bile duct-cannulated; HPLC, high-performance liquid chromatography; IS, internal standard; AUC, concentration versus time curve; LC/MS/MS, liquid chromatography/tandem mass spectrometry; LC/MS, liquid chromatography/mass spectrometry.
- Received August 11, 2006.
- Accepted October 20, 2006.
- The American Society for Pharmacology and Experimental Therapeutics