Metabolism of Lamotrigine to a Reactive Arene Oxide Intermediate

glucuronidation (35), causes an apparent competitive inhibition of lamotrigine's N-glucuronidation and, with valproate, more frequent skin reactions. ...
0 downloads 0 Views 83KB Size
NOVEMBER 2000 VOLUME 13, NUMBER 11 © Copyright 2000 by the American Chemical Society

Articles Metabolism of Lamotrigine to a Reactive Arene Oxide Intermediate James L. Maggs, Dean J. Naisbitt, Justice N. A. Tettey, Munir Pirmohamed, and B. Kevin Park* Department of Pharmacology and Therapeutics, The University of Liverpool, Liverpool L69 3BX, United Kingdom Received April 11, 2000

Lamotrigine [3,5-diamino-6-(2,3-dichlorophenyl)-1,2,4-triazine] is an antiepileptic drug associated with hypersensitivity reactions which are thought to be an immunological response to metabolically generated drug-protein adducts. The o-dichlorophenyl moiety is a potential site for bioactivation of the drug to an arene oxide. The metabolites of [14C]lamotrigine (78 µmol/kg, iv) in adult male Wistar rats were characterized with particular reference to thioether derivatives of an epoxide intermediate. Biliary recovery of radioactivity from anesthetized and cannulated animals was 7.3 ( 3.0% (mean ( SD, n ) 4) of the dose over 4 h; 5.5 ( 0.5% was recovered in bladder urine after 4 h. Bile contained [14C]lamotrigine (1.4 ( 0.3%), a glutathione adduct of [14C]dihydrohydroxylamotrigine (1.8 ( 0.3%), i.e., an adduct of an arene oxide, and the glutathione (1.5 ( 0.7%), cysteinylglycine (1.9 ( 0.5%), and N-acetylcysteine (0.4 ( 0.2%) adducts of [14C]lamotrigine. Formation of the thioether metabolites was partially blocked by the cytochrome P450 inhibitor, ketoconazole. Urine contained [14C]lamotrigine (4.5 ( 0.5%) and [14C]lamotrigine N-oxide (0.9 ( 0.2%). The radiolabeled material in skin (15.6 ( 1.4%) was almost entirely [14C]lamotrigine. Isolated rat hepatocytes achieved a low rate of turnover to the glutathione adduct and N-oxide. However, neither rat nor human liver microsomes catalyzed NADPH-dependent irreversible binding. Lamotrigine can be bioactivated to an arene oxide by rat hepatocytes in the absence of a major competing pathway such as N-glucuronidation. Inhibition of N-glucuronidation has been associated with an increased risk of skin reactions in patients.

Introduction Lamotrigine [Lamictal, 3,5-diamino-6-(2,3-dichlorophenyl)-1,2,4-triazine; Scheme 1] is used increasingly as an adjunctive and monotherapy in the management of a wide range of epilepsy syndromes (1). It has also been assessed for the treatment of various unrelated neurological conditions (2-4). While lamotrigine is generally

well tolerated by patients, it is associated with a relatively high incidence (g10%) of allergic skin rashes (5) and with rare serious events which include severe cutaneous reactions, neutropenia, and leucopenia (6, 7). Isolated cases of acute hepatic failure (8) and agranulocytosis (9) have also been reported. The immunological perturbations seen in patients with severe cutaneous toxicity are considered to be indicative

10.1021/tx0000825 CCC: $19.00 © 2000 American Chemical Society Published on Web 10/13/2000

1076

Chem. Res. Toxicol., Vol. 13, No. 11, 2000

Scheme 1. [14C]Lamotrigine (*, position of the radiolabel) and the Proposed Pathway for Its Bioactivation in Rat Hepatocytesa

Maggs et al.

epoxides, namely, glutathione adducts, are usually found at their highest concentrations in bile. Metabolic degradation of these adducts may yield diverse mixtures of urinary thioether products (19), the complexity of which could hinder recognition and quantification of bioactivation when, as in the case of lamotrigine, it cannot be more than a minor pathway in vivo. This paper describes studies on the disposition and metabolism of lamotrigine in rats, with particular emphasis on the formation of arene oxide intermediates.

Materials and Methods

a Positions of epoxidation and glutathione (GS) addition were not determined; those that are shown are illustrative. Roman numerals refer to peaks in Figure 2.

of an immune-mediated hypersensitivity reaction (7, 10). Similar adverse reactions, collectively termed the anticonvulsant hypersensitivity syndrome, are associated with other aromatic antiepileptics, such as phenytoin and carbamazepine, and are generally thought to depend on bioactivation of the drug to reactive intermediates which covalently modify cellular proteins (11). These haptenated proteins, it is supposed, are recognized by antigen presenting cells, processed, and presented to the effector arm of the immune response, resulting ultimately in damage or destruction of cells containing the modified proteins. Arene oxide metabolites contribute to the covalent binding of carbamazepine to hepatic microsomal protein (12). Formation of the epoxides in rats is revealed by biliary excretion of glutathione adducts (13). The odichlorophenyl moiety of lamotrigine suggests a potential for bioactivation to an arene oxide. The parent compound, 1,2-dichlorobenzene, a metabolism-dependent hepatotoxicant, is oxidized in rat and human liver to epoxides which are deactivated primarily by glutathione conjugation (14-16). However, lamotrigine is chemically unrelated to other aromatic antiepileptics. The diaminotriazine substituent is the dominant site of biotransformation in most species. Humans eliminate lamotrigine principally as N-glucuronides in urine; N-oxide and N-methyl derivatives are minor metabolites (17, 18). Rats, lacking significant N-glucuronidation and methylation pathways, eliminate lamotrigine unchanged and as small amounts of the N-oxide (17). Epoxidation of the o-dichlorophenyl moiety appears to be blocked; neither phenolic nor thioether derivatives of an epoxide intermediate have been found in the urine of either humans or experimental animals. Although urinary elimination of epoxide-derived metabolites of lamotrigine by rats has not been reported, this species might yet serve as a model of bioactivation because the most distinctive derivatives of biogenic

Chemicals. Lamotrigine, [14C]lamotrigine [3,5-diamino-6(2,3-dichlorophenyl)-[6-14C]-1,2,4-triazine; specific activity, 52.9 mCi/mmol; radiochemically homogeneous as determined by HPLC], and lamotrigine N-2-oxide were donated by Glaxo Wellcome plc (Stevenage, Hertfordshire, U.K.). Ketoconazole was provided by Jansen Pharmaceuticals (Beerse, Belgium). All general reagents were purchased from either BDH (Poole, Dorset, U.K.) or Sigma Chemical (Poole, Dorset, U.K.) and were analytical-grade. Chromatography-grade solvents were products of Fisher Scientific (Loughborough, Leicestershire, U.K.). Metabolites in Human Urine. Samples of urine were provided by adult male epileptic patients taking oral lamotrigine (250 mg/day). They were analyzed by LC/MS. Animal Experiments. Male Wistar rats (180-250 g) obtained from a colony maintained by the Biomedical Services Unit, The University of Liverpool, were anesthetized with urethane (1.4 g/mL of isotonic saline; 1 mL/kg, ip). Cannulae were inserted into the jugular vein and common bile duct, and the penis was ligated. Drug-blank bile was collected for approximately 20 min before treatment. [14C]Lamotrigine (78 µmol/ kg, 10 µCi) dissolved in dimethyl sulfoxide (50 µL) was administered iv. Some of the rats were treated with ketoconazole (94 µmol/kg, iv) in dimethyl sulfoxide (50 µL) 1 h before administration of the [14C]lamotrigine. Bile was collected hourly for 4 h at room temperature. Aliquots were analyzed immediately by LC/ MS with parallel radiometric detection; the remainder was stored at -30 °C. Urine was aspirated from the bladder 4 h after treatment. Radioactivity in the bile and urine was assayed by liquid scintillation counting. Portions (ca. 100 mg) of the major internal organs and of skin were solubilized in 1 mL of OptiSolv (Wallac UK, Milton Keynes, Buckinghamshire, U.K.) at 50 °C over 16 h. The solutions were decolorized with hydrogen peroxide (200 µL), neutralized with glacial acetic acid (50 µL), left in darkness overnight, and finally mixed with 20 mL of Ultima Gold (Packard Bioscience BV, Groningen, The Netherlands) for scintillation counting. The complete skin was stripped from each carcass, and the remainder was immersed in 3 M KOH (500 mL) in a capped plastic bottle which was maintained at 50 °C for 24 h. Aliquots of the resulting liquid fraction (1 mL) were neutralized with concentrated HCl. They were assayed for radioactivity by scintillation counting. Extraction of Radiolabeled Material from Tissues. Rat tissues, stored at - 30 °C for no longer than 2 days, were portioned (liver and kidneys, 2-3 g; whole skin, 8-12 g), immersed in ethyl acetate (50 mL), and stirred at room temperature for 16 h. The solvent was decanted, centrifuged to sediment the suspended material, and concentrated at 40 °C under a stream of nitrogen. The resulting residue was extracted with methanol (5 mL). The methanolic extract was reduced to dryness under nitrogen, and reconstituted in methanol (200 µL) for radiochromatographic and LC/MS analyses. Microsomal Incubations. Microsomes were prepared from histologically normal human livers and from the livers of male Wistar rats as described previously (20). They were stored at 80 °C. Some of the rats had received single doses of sodium phenobarbital (60 mg/kg, ip) in isotonic saline on the three preceding days. Microsomal protein concentrations were measured by a standard method (21). [14C]Lamotrigine (final

Bioactivation of Lamotrigine concentration, 2-10 µM; 0.1 µCi) in dimethyl sulfoxide (10 µL) was incubated at 37 °C with either human or rat microsomes (final protein concentration, 2 mg/mL) in 67 mM phosphatebuffered saline (pH 7.4) for 30 min. The volume of the final mixture, in 25 mL Erlenmeyer flasks, was 1 mL. Enzymatic activity was initiated by the addition of NADPH (1 mM). Reactions were terminated with ethyl acetate (2 mL). Precipitated protein was separated by centrifugation. The supernatant was evaporated to dryness under a stream of nitrogen and reconstituted in a 1:1 (v/v) ethanol/water mixture (150 µL) for analysis by HPLC with on-line radiometric detection. The protein was subjected to exhaustive solvent extraction to determine if any radiolabeled material was bound irreversibly (22). Hepatocyte Incubations. Freshly isolated hepatocytes were obtained from three male Wistar rats (190-210 g) and suspended (2.4 × 106 viable cells/mL) separately in Krebs-Henseleit buffer (pH 7.6) containing 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (12.5 mM) as described previously (23). The viability of the cells, determined by trypan blue exclusion, was g90%. [14C]Lamotrigine (final concentration, 20 µM; 0.2 µCi) in dimethyl sulfoxide (10 µL) was incubated with 5 mL aliquots of the hepatocyte suspension in rotating 50 mL round-bottomed flasks at 37 °C under an O2/CO2 (95:5, v/v) atmosphere for 4 h. Either the drug or hepatocytes were omitted from control incubations. Each incubation was terminated with 2 volumes of ice-cold methanol, vortexed, and centrifuged (4000 rpm, 5 min). The resulting supernatant was reduced to dryness under nitrogen at 40 °C. It was resuspended in 1 mL of a 1:1 (v/v) methanol/water mixture for radiochromatographic analysis. Supernatants were concentrated to approximately 0.7 mL for analysis by LC/MS with parallel radiometric detection. Chromatographic and Mass Spectrometric Analyses. [14C]Lamotrigine and its metabolites formed by isolated hepatocytes were resolved for radiometric quantification on a 5 µm HyPURITY Elite C18 column (150 mm × 4.6 mm i.d.; Hypersil, Runcorn, U.K.). The eluent was acetonitrile (from 10 to 50% over the course of 30 min) in 25 mM ammonium acetate (pH 4.0). The flow rate was 1.0 mL/min. Aliquots of bile (40-70 µL), urine (100 µL), tissue extracts (50-100 µL), and concentrated supernatants from microsomal incubations (50-100 µL) were eluted from a 5 µm Columbus C8 column (250 mm × 4.6 mm i.d.; Phenomenex, Macclesfield, U.K.) with a gradient of acetonitrile (from 5 to 40% over the course of 40 min) in 0.1 M ammonium acetate (pH 6.9). The flow rate was 0.9 mL/min. Mass spectral data were obtained with a Quattro II quadrupole instrument fitted with an in-line electrospray source (Micromass, Manchester, U.K.). The rate of eluate split flow to the mass spectrometer was approximately 40 µL/min. The remainder was directed to a radioactivity flow detector. Positive-ion spectra were acquired between m/z 100 and 1050 at one scan per 5 s. The interface temperature was 70 °C; the capillary voltage was 3.9 × 103 V. Analyte fragmentation was enhanced by increasing the cone voltage from 30 V. Spectra (two to four) within a chromatographic peak were averaged, and the averaged adjacent (background) spectra were subtracted from them. Selected ion monitoring data in 4-13 channels were acquired with a dwell time of 200 ms and an interchannel delay of 20 ms. All data were processed via MassLynx 2.0 software (Micromass). Daughter spectra of material resolved by HPLC were generated by collision-induced decomposition in a Micromass Q-Tof hybrid tandem instrument using a collision energy of 20 eV and a gas (argon) pressure of 0.5 × 10-3 mbar. Radiometric HPLC. Radiolabeled analytes were quantified using a Radiomatic Flo-Oneβeta A-250 flow detector (Packard, Pangbourne, Berkshire, U.K.). The eluate was mixed with Ultima-Flo AP scintillant (Packard Bioscience) at a rate of 1 mL/min. Peak areas were computed after background subtraction using A250-1.6 software. Statistical Analysis. Data were analyzed for normality by the Shapiro-Wilk test. Groups were compared using the Mann-

Chem. Res. Toxicol., Vol. 13, No. 11, 2000 1077

Figure 1. Biliary excretion of radioactivity by anesthetized and cannulated male Wistar rats treated with [14C]lamotrigine (78 µmol/kg, iv) with (b) or without (O) previous administration of ketoconazole (94 µmol/kg, ip, 1 h earlier). Results are means ( SD (n ) 4). Asterisks indicate values significantly different from that of control rats (p < 0.05).

Figure 2. HPLC radiochromatogram of the biliary metabolites (0-1 h bile collection) of [14C]lamotrigine (78 µmol/kg, iv) in male rats. Whitney test, accepting a p value of