Chem. Res. Toxicol. 1993,6, 341-344
341
Identification of a Pyridinium Metabolite in Human Urine following a Single Oral Dose of 1-[2-[Bis[4-(trifluoromethyl)phenyllmethoxy]ethyl] - 1,2,5,6-tetrahydro-3pyridinecarboxylic Acid Monohydrochloride, a y-Aminobutyric Acid Uptake Inhibitor Louis Radulovic,*gt Thomas Woolf,+Susan Bjorge,? Charles Taylor,l Michael Reily,e Howard Bockbrader,t and Tsun Changt Pharmacokinetics and Drug Metabolism, Pharmacology, and Chemistry Departments, Parke-Davis Pharmaceutical Research Division, Ann Arbor, Michigan 48105 Received June 23, 1992
Single-dose administration of 50 mg of 1-[2- [bis[4-(trifluoromethyl)phenyllmethoxy]ethyl]1,2,5,6-tetrahydro-3-pyridinecarboxylic acid monohydrochloride resulted in temporary neurological and psychological symptoms in two subjects. Because of the nature of adverse effects, urine from a subject who received CI-966 orally was extracted t o investigate the metabolism of CI-966 in man. An unknown urinary component was identified as a pyridinium metabolite of CI-966 based on HPLC-MS and 'H and 19FNMR. Structural confirmation was achieved by chromatographic and spectroscopic comparisons t o a reference standard. In several in vitro screens and preclinical studies, the pyridinium metabolite appears t o possess minimal pharmacological activity.
Introduction CI-966l is a novel anticonvulsant which differs structurally and mechanistically from currently marketed anticonvulsants ( 1 ) . CI-966 acts by inhibiting the synaptic uptake of GABA, which is a major inhibitory neurotransmitter in the mammalian central nervous system ( 2 ) . Inadequate GABAergic inhibition may be the underlying mechanism of some seizure disorders in man (3-5). Proposed indications were for treatment of partial seizures and generalized tonic-clonic seizures. A randomized, double-blind, escalating single-dosetolerance, and pharmacokinetic study comparing the effects of CI-966 capsules with those of placebo was conducted in healthy volunteers (6, 7). The study was terminated at the 50-mg dose due to adverse effects experienced by two female subjects. Physical signs included unresponsiveness to nonpainful sensory stimuli, tremor, myoclonus, increased muscle rigidity, and cogwheeling. Short- and long-term memory impairment was also observed. Severity of symptoms appeared to correlate with plasma CI-966 concentration and all physical signs had resolved within 24 h following dosing. However, psychiatric symptoms resembling those observed in manic and schizophrenic patients lasted for 5 to 6 days in both subjects, suggesting that CI-966 elimination from brain may be slower than that from plasma (harmonic mean t l p
* Address correspondence to this author at the Pharmacokinetics and Drug Metabolism Department, Parke-Davis Pharmaceutical Research Division, 2800 Plymouth Rd., Ann Arbor, MI 48105. Pharmacokinetics and Drug Metabolism Department. Pharmacology Department. Chemistry Department. I Abbreviations: CI-966,l[2-[bis[4-(trifluoromethyl)phenyll methoxylethyl]-l,2,5,6-tetrahydro-3-pyridinecarboxylicacid monohydrochloride; GABA, y-aminobutyric acid; HPP+, 4-(4-chlorophenyl)-l-(4-fluorophenyl)-4-oxobutylpyridinium;MAO, monoamine oxidase; MPTP, l-methyl-4-phenyl-l,2,3,64etrahydropyridine; MPP+, l-methyl-4-phenylpyriMPP-, dinium; MPDP+, l-methyl-4-phenyl-2,3-dihydropyridinium; l-methyl-4-phenylpyridinium radical. +
f
0893-228x/93/2706-0341$04.00/0
of 16.8 h) or biotransformation of CI-966 to one or more active metabolites occurs. It is possible that some of the neurological and psychological symptoms observed following administration of 50 mg of CI-966 may be unrelated to inhibition of GABA uptake and could be attributed to an unknown pharmacological response to CI-966 or a metabolite (8). In fact, CI-966 undergoes reductive and oxidative metabolism in rata (9). A piperidine and pyridinium metabolite of CI-966 were identified in pooled rat urine. Because of the severity of adverse events and our knowledge of the metabolism of CI-966 in rata, it was deemed important to examine the metabolic fate of CI966 in man.
Materials and Methods Isolation Procedure. Approximately 1 L of human urine (0-6 and 24-48 h collections)from a single male subject given 25 mg of CI-966 was processed by solid-phase extraction using Analytichem (Harbor City, CA) 3-mL CN end-capped cartridges. Urine was adjusted to pH 9 with base and aliquots (6 mL) were applied to individual CN cartridges, previously conditioned with 3 X 2 mL of acetonitrile and 3 X 2 mL of 50 mM ammonium phosphate dibasic (pH 9) (buffer A). Each cartridge was sequentially washed with 2 X 2 mL of buffer A, 2 mL of 20% acetonitrile in buffer A, 3 X 2 mL of 50 mM ammonium acetate (pH 2.1) (buffer B), and 2 X 1 mL of 25% acetonitrile in buffer B. Analytes eluted with 2 X 750 pL of 60% acetonitrile in buffer B. The eluent was evaporated to dryness and the residue was reconstituted in a mobile phase/95% ethanol (955) mixture. Analytes were resolved using a LC-18-DB 5 - ~ m semipreparative HPLC column (10 mm i.d. X 25 cm) by Supelco, Inc. (Bellefonte, PA), maintained at 40 "C. Mobile phase consisted of 45% acetonitrile in 50 mM ammonium phosphate (pH 2.1). Flow rate was 4.5 mL/min. Absorbance was monitored a t 225 nm. An ISCO Foxy fraction collector (Isco, Inc., Lincoln, NE) with peak detection capability was used to collect the peaks of interest. Isolates were desalted using the same 3-mL CN end-capped cartridges according to the following procedure. Samples were 0 1993 American Chemical Society
Radulovic et al.
342 Chem. Res. Toxicol., Vol. 6, No. 3, 1993 A.
470
0
I . ,
200
300
I1
I.
400 mlz
B.
. 500
600
700
500
600
700
474
I
%I loo]
0 0
5
1'0 min
15
20
Figure 1. HPLC-UV analysis (225 nm) of an extracted urine sample from a male subject given a 25-mg oral dose of CI-966 (peak A, unknown metabolite of CI-966). diluted to 10% acetonitrile in buffer A and applied to cartridges previously conditioned with 3 X 2 mL of acetonitrile and 3 x 2 mL of buffer A. Cartridges were washed with 3 X 2 mL of buffer A and 3 X 2 mL of water (HPLC grade). Analytes were eluted with 2 X 750 pL of methanol and evaporated to dryness prior to mass spectroscopy. HPLC-MS Analysis. For thermospray ionization analysis, a Waters 600MS multisolvent delivery system was interfaced to a VG Trio-2 mass spectrometer using a plasma spray/ thermospray probe assembly and associated ion source. Separation was achieved using a 3-pm Pecosphere C18CR column (4.6 mm i.d. X 3.3 cm) by Perkin-Elmer (Norwalk, CT) and a mobile phase consisting of 50% acetonitrile in 100 mM ammonium acetate (pH 2) at a flow rate of 0.7 mL/min. The mass spectrometer was operated in the plasma spray mode with a source temperature of 200 "C, capillary temperature of 190 "C, and discharge current of 300 MA. Full scans (700-150 D) were taken repetitively a t a scan rate of 1 s decade-'. NMR Analyses. lH and I9FNMR analyses were performed in deuterated methanol on a Varian XL300 NMR equipped with a M68000 computer. Reference Standards. Details of the chemical synthesis of the pyridinium metabolite and CI-966 reference standards were previously reported (9).
200
300
400
m/z Figure 2. HPLC-MS analyses of unknown isolated from human urine following a 25-mg dose of CI-966 (A) and of standard CI966 (B). A.
E.
I " " I " " l " " I 1 " ' I " " l
-50
-55
-60
-65
-75 PPM
-70
Figure 3. Fluorine NMR (282 MHz) of unknown isolated from
human urine following administration of CI-966 (A) and of standard CI-966 (B) obtained in methanol-&. A.
Results and Discussion HPLC-UV analysis of urine from a male subject who received a single 25-mg dose revealed the presence of a major unknown UV-absorbing peak (peak A of Figure 1). Positive ion HPLC-MS analysis of the isolated unknown gave a mass spectrum with a presumed molecular ion at m/z 470, three mass units less than that of CI-966, and a major fragment ion at mlz 426 (base peak) corresponding to loss of 44 mass units (-Con) (Figure 2). A single fluorine resonance with the same chemical shift as the bis(trifluoromethyl) groups of CI-966 was observed in the NMR spectrum (Figure 31, suggesting that 1-[2-[bis[4(trifluoromethyl)phenyllmethoxylethyll group remains unchanged in the metabolite. The lH NMR spectrum of CI-966 (Figure 4B) shows a resonance assigned to the vinyl proton at 7.2 ppm. This signal is not present in the 'H NMR spectrum of the metabolite, but four new peaks appeared between 8.0 and 9.5 ppm (Figure 4A). These observations are consistent with a 3-substituted pyridinium moiety. Structural confirmation of the metabolite as a pyridinium derivative of CI-966 was made by chromatographic and spectroscopic comparisons to synthetic reference material. Presence of unchanged CI-966 in
I
cF3
0
I J " ' I " " I " " I " " l " " 9.5 9.0 8.5 8.0
7.5
PPM
Figure 4. Proton NMR (300 MHz) of unknown isolated from
human urine following administration of CI-966 (A) and standard CI-966 (B) obtained in methanol-&.
human urine was confirmed by coinjection with reference standard on HPLC-MS.
Chem. Res. Toxicol.,Vol. 6, No.3, 1993 343
Tetrahydropyridine-3-carboxylateMetabolism r
d"
Enzymatic/
Nonenzymatic
Enzymat 1 c
Hooco-o L
/
Ze'oxidation
te'oxidatlon
CI-966
Hooc~ CF3
L
D l h y d r o p y r l d l n l um
P y r i d 1 n l um
Figure 5. Metabolic scheme for CI-966. Metabolism of CI-966 to a pyridinium metabolite represents an overall four-electron oxidation. A proposed metabolic scheme is shown in Figure 5. Recently, haloperidol was found to undergo a similar four-electron oxidation to the putative neurotoxin HPP+ (10). The similarity in acute neurological side-effects, including Parkinsonism, of haloperidol and CI-966 raised concern whether the conversion of CI-966, a l,&disubstituted tetrahydropyridine analog, to a pyridinium species may lead to neurotoxicity. Preclinical toxicological studies conducted in rhesus monkeys with a structurally similar compound, 1-(4,4bis(4-fluorophenyl)butyl)-4-(4-fluorophenoxy)-l,2,3,6-tetrahydropyridine, resulted in nigrostriatal abnormalities (11)similar to those seen with MPTP (12).The authors proposed a possible structure-activity relationship of tetrahydropyridine analogs with neurologicand associated neuropathologic manifestations. Formation of a pyridinium species a priori may not be sufficient to elicit neurotoxicity. In fact, this has been shown with two nonneurotoxic MPTP analogs, 4'-methyl-MPTP and 4'fluoro-MPTP (13). Further studies are necessary to elucidate the structure-activity relationships of the conversion of 1,3-disubstituted tetrahydropyridine compounds, such as CI-966,to pyridinium species and potential neurotoxicity. Of the two forms of monoamine oxidase (A and B), MA0 B appears to be primarily responsible for the in vivo bioactivation of MPTP to MPP+for the following reasons. In vitro catalytic activity of MA0 B toward MPTP is approximately 14 times that of MA0 A (14),and both oxidation products, MPDP+ and MPP+, are more inhibitory to MA0 A than B. Ki values of 2.4 pM and 3.0 WM were reported for MPDP+ and MPP+, respectively, with MA0 A compared to 230 pM (MPDP+) and 300 pM (MPP+) with MA0 B (15,16). Unlike MPTP, MA0 B-mediated biotransformation of CI-966to the pyridinium metabolite seems highly unlikely. MA0 B structureactivity relationships with MPTP derivatives suggest a requirement for small substituents (Le. methyl-, ethyl-, propyl-) on the tertiary nitrogen for conversion to a dihydropyridinium species (17,181. Furthermore, compounds bearing substituents on the 3-position as opposed to the 4-position of the tetrahydropyridine moiety are not good substrates for oxidation by MA0 B (19). Therefore, oxidation of (3-966 to the dihydropyridine derivative most likely occurs through a non-MA0 B-mediated oxidation. CI-966 may be metabolized by the cytochrome P-450 system which has been shown to catalyze the ring a-carbon oxidation of various cyclic tertiary amines (20-24),including MPTP (25),to iminium species. Subsequent oxidation of the CI-966 dihydropyridine derivative to the pyridinium metabolite may be enzymatic, i.e. catalyzed by the cytochrome P-450 system as is haloperidol (261,or nonenzymatic via bimolecular redox
disproportionation in a manner similar to that observed for the conversion of MPDP+ to MPP+ (27). Recently, Gerlach et al. (13)have proposed a direct hydride transfer from MPDP+ to a cellular two-electron oxidant. Additionally, a sequence of two one-electron oxidations of MPDP+ via a MPP' intermediate is feasible under physiological condition, with further oxidation to MPP+ occurring within the dopaminergic neuron (13). Chronic administration of high doses of CI-966 to beagle dogs in a 13-weektoxicologystudy did not cause protracted movement disorders nor did it result in nigrostriatal lesions as determined by histology or enzymatic assays for tyrosine hydroxylase or glutamic acid decarboxylase activity of brain tissue. Furthermore, the pyridinium metabolite of CI-966was only active as an inhibitor of in vitro dopamine uptake into rat neocortical slices at high concentrations (IC5o 100 pM) (8)unlike the MPTP metabolite, MPP+ (IC5,, = 0.14 pM)(28).The pyridinium metabolite of CI966 was not active in a battery of 35 receptor binding assays including subtypes of dopamine, glutamate, and GABA receptors (8). Administration of the pyridinium metabolite to mice did not cause any behavioral effects at intravenous doses up to 30 mg/kg (8). Finally, an explanation for the neurological and psychological symptoms seen in healthy subjects receiving CI-966 is that they result from the effects of increased GABA concentration in many areas of the brain (8). Although available data with CI-966 do not suggest any neurotoxic action of CI-966 or its metabolites, additional studies are needed to more thoroughly explore the possibility. Appropriate studies to assess neurotoxic potential of the pyridinium metabolite of CI-966 would include mitochondrial respiration inhibition, intracerebral microdialysis, and neuronal toxicity in cell culture.
Conclusions CI-966,which contains a tetrahydropyridine moiety with a bulky substituent on the tertiary nitrogen, undergoes biotransformation to a pyridinium derivative in man. To the best of our knowledge, this is the first report of a 1,3disubstituted tetrahydropyridine which is metabolized to a pyridinium metabolite in man. Investigators working with such tetrahydropyridine analogs should be cognizant of potential bioactivation of these compounds.
References (1) Taylor, C. P.,Vartanian,M. G.,Schwarz,R. D.,Rock,D.M.,Callahan, M. J., and Davis, M. D. (1990) Pharmacology of CI-966: a potent GABA uptake inhibitor, in vitro and in experimentalanimals. Drug Deu. Res. 21, 195-216. (2) Bloom, F. E. (1985) Neurohumoral transmission and the central nervous system. In Pharmacological Basis of Therapeutics, 7th ed.; (Gilman, A. G., Goodman, L. S., Rall, T. W., and Murad, F., Eds.), pp 236-259, MacMillan, New York. (3) Roberts,E.,Chase,T. N.,and Tower,D. B. (1976). GABA inNeruous System Function, Raven Press, New York.
344 Chem. Res. Toxicol., Vol. 6, No. 3, 1993 (4) Krogsgaard-Larsen,P.,Scheel-Kruger, J.,andKofod,J. (1979) GABA Neurotransmitters: Pharmacochemical, Biochemical, and Pharmacological Aspects, Munksgaard, Copenhagen. ( 5 ) Taylor, C. P. (1990) GABA receptors and GABAergic synapses as targets for drug development. Drug Deu. Res. 21, 151-160. (6) Sedman, A. J., Gilmet, G. P., Sayed, A. J., and Posvar, E. L. (1990) Initial human safety and tolerance study of GABA uptake inhibitor, CI-966: potential role of GABA as a mediator in the pathogenesis of schizophrenia and mania. Drug Deu. Res. 21, 235-242. (7) Busch, J. A., Radulovic, L. L., Sedman, A. J., Posvar, E., and Chang, T. (1989) HPLC assay for CI-966 in human plasma and preliminary pharmacokinetic results obtained following administration of single oral rising doses to healthy subjects. Pharm. Res. 6, 5-236. (8) Taylor,C.P.,andSedman,A. J. (1991)Pharmacologyofthegammaaminobutyric acid-uptake inhibitor CI-966 and its metabolites: preclinical and clinical studies. In Transmitter Amino Acid Receptors: Structures, Transduction and Models for Drug Deuelopment (Barnard, E. A., and Costa, E., Eds.) pp 251-271, Thieme Medical Publishers, Inc., New York. (9) Black, A,, Bjorge, S., Bockbrader, H., Chang, T., Gregor, V. E., Lobbestael, S. J., Nugiel, D., Pavia, M. R., Radulovic, L., and Woolf, T. (1990) The synthesis and metabolic profile of CI-966: a potent, orally-active inhibitor of GABA uptake. Drug Deu. Res. 21, 189193. (IO) Subramanyam, B., Pond, S. M., Eyles, D. W., Whiteford, H. A., Fouda, H. G., and Castagnoli, N., Jr. (1991) Identification of potentially neurotoxic pyridinium metabolite in the urine of schizophrenic patients treated with haloperidol. Biochem. Biophys. Res. Commun. 181, 573-578. (11) Barsoum, N. J., Gough, A. W., Sturgess, J. M., and de la Igelesia, F. A. (1986) Parkinson-like syndrome in nonhuman primates receivinga tetrahydropyridine derivative. NeuroToxicology7,119126. (12) Langston, J. W., Ballard, P., Tetrud, J. W., and Irwin, I. (1984) Chronic Parkinsonism in humans due to a product of meperidineanalog synthesis. Science 219, 979-980. (13) Gerlach,G., Riederer, P., Przuntek, H.,andYoudim, M. B. H. (1991) M P T P mechanisms of neurotoxicity and their implications for Parkinson's disease. Eur. J . Pharmacol. 208, 273-286. (14) Salach, J. I., Singer, T. P., Castagnoli, N., Jr., and Trevor, A. (1984) Oxidation of the neurotoxic amine l-methyl-4-phenyl-l,2,3,6-tetrahydropyridine (MPTP) by monoamine oxidases A and B and suicide inactivation of the enzymes by MPTP. Biochem. Biophys. Res. Commun. 125, 831-35. (15) Singer, T. P., Salach, J. I., and Crabtree, D. (1985) Reversible inhibition and mechanism-based irreversible inactivation of monamine oxidases by l-methyl-l-phenyl-l,2,3,6-tetrahydropyridine (MPTP). Biochem. Biophys. Res. Commun. 127, 707-712.
Radulouic et al. (16) Singer, T. P., Salach, J. I., Castagnoli, N., Jr., and Trevor, A. J. (1986) Interactions of the neurotoxic amine 1-methyl-4-phenyl1,2,3,6-tetrahydropyridine with monoamine oxidases. Biochem. J . 235, 785-89. (17) Markey, S. P., and Schmuff, N. R. (1986) The pharmacology of the Parkinsonian syndrome producing neurotoxic MPTP (l-methyl-4phenyl-1,2,3,6-tetrahydropyridine)and structurally related compounds. Med. Res. Reu. 6, 389-429. (18) Heikkila, R. E., Manzino, L., Cabbat, F. S., and Duvoisin, R. C. (1985) Studies on the oxidation of the dopaminergic neurotoxin l-methyl-4-phenyl-l,2,5,6-tetrahydropyridine by monoamine oxidase B. J. Neurochem. 45, 1049-1054. (19) Abell, C. W. (1987) Monoamine oxidase A and B from human liver and brain. Methods Enzymol. 142, 638-650. (20) Hoag, M. K. P., Schmidt-Peetz, M., Lampen, P., Trevor, A., and Castagnoli, N., Jr. (1988) Metabolic studies on phencyclidine: characterization of a phencyclidine iminium ion metabolite. Chem. Res. Toxicol. 1, 128-131. (21) Hoag, K., Trevor, A., Kalir, A., and Castagnoli, N. (1987) Phenycyclidine iminium ion: NADPH-dependent metabolism, covalent binding to macromolecules, and inactivation of cytochrome(s) P-450. Drug Metab. Dispos. 15,485-490. (22) Shigenaga, M., and Castagnoli, N. (1988) Metabolism dependent covalent binding of (S)-[5-3H]nicotineto liver and lung microsomal macromolecules. Drug Metab. Dispos. 16, 397-402. (23) Peterson, L., and Castagnoli, N. (1988) Regio- and stereochemical studies on the a-carbon oxidation of @)-nicotine by cytochrome P-450 model systems. J . Med. Chem. 31, 637. (24) Peterson, L., Trevor, A., and Castgnoli, N. (1987) Stereochemical studies on the cytochrome P-450 catalyzed oxidation of @)-nicotine to the (SI-nicotine A' l5')-iminium species. J. Med. Chem. 30,249254. (25) Ottoboni, S., Carlson, T. J., Trager, W. F., Castagnoli, K., and Castagnoli, N., Jr. (1990) Studies on the cytochrome P-450 catalyzed ring a-carbon oxidation of the nigrostriatal toxin 1-methyl-4-phenyl1,2,3,6-tetrahydropyridine (MPTP). Chem. Res. Toxicol. 3, 423427. (26) Subramanyam, B., Woolf, T., and Castagnoli, N., Jr. (1991) Studies on the in vitro conversion of haloperidol t o a potentially neurotoxic pyridinium metabolite. Chem. Res. Toricol. 4, 123-128. (27) Singer, T. P., Castagnoli, N.,Ramsay, R. R., andTrevor, A. J. (1987) Biochemical events in the development of Parkinsonism induced by l-methyl-4-phenyl-l,2,3,6-tetrahydropyridine. J . Neurochem. 49, 1-8. (28) Javitch, J. A., D'Amato, R. J., Strittmatter, S. M., and Snyder, S. H. (1985) Parkinsonism-inducing neurotoxin, N-methyl-4-phenyl1,2,3,6-tetrahydropyridineby dopamine neurons explains selective toxicity. Proc. Natl. Acad. Sci. U.S.A. 82, 2173-2177.
