In Vitro and in Vivo Metabolism of the Antianxiolytic Agent Fenobam in the Rat

W. N. WU*~, L. A. MCKOWN*, AND P. J. O'NEILL~ Received June 20, 1994, from the *Department of Drug Metabolism, The R. W Johnson Pharmaceutical Research Institute, Spring House, PA 19477-0776, and #Ethicon, Inc., Somer/ille, NJ 08876. Accepted for publication November 11, 1994@.

Abstract 0 Fenobam [(Fn); N-(3-chlorophenyl)-N-(4,5-dihydro-l -methyl- 4-0x0-1 Kimidazole-2-yl)urea] sulfate is a novel agent with potent anxiolytic activity in rats. [14C]Fn sulfate was administered as an oral solution (250 mglkg) to male Wistar rats, and 52% of the administered dose was excreted in urine (0-5 days). In vitro metabolism of Fn was studied by incubating [l4C1Fn with rat hepatic 9000 x g supernatant preparations. Unchanged Fn and a total of six metabolites were isolated, quantified, and identified from the urine and liver 9000 x g supernatant samples by column chromatography; TLC; UV, IR, and NMR spectroscopy; MS; and comparison with synthetic samples. Four metabolic pathways for Fn are proposed: (1) hydroxylation at the phenyl ring to form 4-hydroxyphenyl- Fn, a major pathway in vivo (12% of the sample radioactivity) but a minor pathway in vitro (4% of the sample radioactivity); (2) hydroxylation at the creatinine ring to form 5-hydroxy-Fn (19%) of the sample radioactivity), a dominant pathway in vitro but not in vivo; (3) oxidative cleavage at the creatinine ring (loss of a ketene unit), a minor pathway for Fn but an important pathway for 4-hydroxyphenyl-Fn in vivo; and (4) Ndemethylation, a minor pathway for Fn in vivo.

Fenobam [N-(3-chlorophenyl)-N'-(4,5-dihydro-l-methyl-4- oxo-1H-imidazole-2-yl)urea is a novel agent with anxiolytic activity in rats.lS2 It has been shown to possess selective anxiolytic properties in animal models and in clinical trials, with a mode of action that differentiates it from the benzodiazepines.'-l0 Previous studies of fenobam disposition indicated that fenobam is rapidly and extensively metabolized by rats, dogs, and humans.4J1 The objective of this work was to identify in vitro and in vivo metabolites of fenobam in the rat to gain an understanding of the biotransformation path- ways involved.

Experimental Section Chemi~als-['~C1Fenobam sulfate (1.96 pCi/mg) was synthesized

with the I4C-label in the carbonyl position as shown in Figure 1. The radiochemical and chemical purities were determined to be >97%. RWJ-34892 (Metabolite 1) and RWJ-34734 (Metabolite 2), synthesized at The R. W. Johnson Pharmaceutical Research Institute (Spring House, PA), were used as reference samples for TLC, NMR spectro- scopy, and MS.12 Glusulase, a mixture of arylsulfatase and /3-glucu- ronidase (1:4, v/v) from Helix pomatia, was purchased from Endo Laboratories, Inc., Garden City, NY. Glucose-6-phosphate, NADP, and Trizma base [tris-(hydroxymethyl)aminomethane] were obtained from the Sigma Chemical Company, St. Louis, MO. Amberlite XAD-2 was purchased from Rohm and Haas, Philadelphia, PA. Sephadex LH-20 (25-100 micron) was obtained from Pharmacia, Upsala, Sweden. Silica gel PF 254 was obtained from E. Merck, Darmstadt, Germany. Diazald (N-methyl-N-nitroso-p-toluenesulfonamide) was obtained from the Aldrich Chemical Company, Milwaukee, MI. Biofluor (New England Nuclear, Boston, MA) was used as a scintil- lation solution. Glass distilled solvents (ethyl acetate, chloroform, methylene dichloride, hexane, and methanol) were obtained from Burdick & Jackson Laboratories, Inc., Muskegon, MI. Other standard chemicals used were reagent grade, obtained from commercial sources, and used without further purification. All solvent mixture ratios were VJV.

@Abstract published in Advance ACS Abstracts, January 1, 1995.

Animals-Male CR Wistar rats, weighing 200-250 g, were ob- tained from Charles River Breeding Laboratories, Inc., Wilmington, MA.

Hepatic 9000 x g Fraction Preparation, Isolation, and Purification f o r Analysis-Male CR Wistar rats were sacrificed, and their livers were removed and homogenized in three volumes of cold 0.05 M Tris HC1 buffer (pH 7.5) containing 1.15% KCl. The homogenate was centrifuged at 9000 x g at 2 "C in a refrigerated Sorvall RC-5B centrifuge (DuPont, Newtown, CT). Incubation mix- tures were then prepared using 1 mL of this supernatant (containing 250 mg wet weight equivalent of liver) and 0.5 mM NADP, 5 mM glucose-6-phosphate, 5 mM magnesium chloride, and 0.1 mM [l4C1- fenobam (added as 20 pL of an acetone solution) in 5 mL of TrisKCl. Mixtures were incubated at 37 "C for 30 min in air. The reaction was terminated by placing the flasks on ice. The contents of the flasks were pooled and extracted once with an equal volume of ethyl acetate. Aliquots of the extracts were chromatographed on silica gel GF TLC plates (20 x 5 cm, 250 pm; Analtech, Newark, DE) with hexand methanovethylene dichloride (34: 14:52) as the mobile phase. The TLC plates were visualized under short-wave W light, and four drug- related spots (M-1, M-2, and two unknowns) were located. The spots were scraped into individual vials and extracted with a mixture of methanol and methylene chloride (1:l) for spectroscopic analysis.

Extraction and Initial Isolation Procedure for the Urinary Metabolites from Rat-Rat urine (270 mL) was collected from 0 to 24 h from five male rats given a single oral 250-mgkg dose fenobam. This urine was combined with rat urine (50 mL, 5.86 x lo7 dpm) obtained from male rats given [l4CIfenobam (250 mgkg, 1.96 pCi/ mg). Acetate buffer (50 mL, 0.1 N, pH 5.1) was added to the combined urine, and the mixture was adjusted to pH 5.1 with glacial acetic acid. The acidified urine was incubated with 3.25 mL of Glusulase at 37 "C for 2 h and shaken with 165 g of Amberlite XAD-2 resin. After filtration and washing with water (18% of the total reactivity was present in the aqueous portion), the resin was extracted twice with methanol (178 and 58 mL). The combined methanol extract (67% of total radioactivity present in the original sample) was evaporated to dryness to yield a brown residue (0.45 9). The residue was dissolved in 60 mL of hot watedmethanol (5:l) and chromatographed on a Sephadex LH-20 column (340 g) with water, watedmethanol, metha- nol, and methanovmethylene chloride as eluents, and 50-mL fractions were collected. A total of 48% of the applied radioactivity eluted off the column.

Isolation, Purification, and Identification of 4-Hydroxy- phenylfenobam (M-2)-The residue from fraction nos. 143-180 of the Sephadex LH-20 column was recrystallized as a beige crystalline product (5 mg) from methanol. The product was analyzed by W, IR, and NMR spectroscopy, and MS. Final structural assignment was confirmed by comparison with synthetic material.I2

Preparation of N-Methyl-4-methogyphenylfenobam (M-2a)- The crystalline residue (Rj, 0.31; 1.5 mg) of 4-hydroxyphenylfenobam was suspended in 1.5 mL of methanol. The solution was reacted with an excess of amount of ethereal diazomethane solution (5 mL) and kept at room temperature overnight. The excess ether and diaz- omethane were evaporated under nitrogen to yield a pale yellow residue that was dissolved in 0.5 mL of methanol and spotted as a band on four silica gel GF plates. These plates were developed in a solvent system of ethylene dichloride/hexandmethanol(52:34:14). The major radioactive band (Rj, 0.42) was scraped from the plates and suspended in 2 mL of hot ethyl acetate/methanol (80:20). The silica gel was removed by filtration, and the solution was evaporated under nitrogen to yield a colorless residue that was analyzed by chemical ionization-mass spectroscopy (CI-MS).

Second Extraction and Isolation Procedure for the Urinary Metabolites-Fresh urine (250 mgkg of [14C]fenobam dosed to male rates; 24-h collection, 250 mL) was mixed with 250 mL of 1 M sodium

0 1995, American Chemical Sociefy and American Pharmaceutical Association

0022-3549/95/3 184-0 185$09.00/0 Journal of Pharmaceutical Sciences / 185 Vol. 84, No. 2, February 1995

Compound Structure CI-MS & El-MS m/z(% relative abundance)

M-2

M-3

M-4

CI-MS 351(MC&*. 0.3). 339(MC2H5+, 1). 313(MW. Clis. 3). 311(MH-. 9). 186(cliS. 35). lW(l00). 160(Clia. 4). lSe(9). 154(38). 128(W)

M-2 a dimethyl derivative

128, : rc$N& H3

nNJgN&

HN, HN & --N 3 $ . 4 on

O N I ,

I [at! 7 i 4 l 58 CI-MS l4

C Y

271(MH*. Clis. 0.1).269(MW. 0.3). 172(Clis. no). 17C(l00). 146(Clis. lo). 14402). 126(M). loo(98)

O N I ,

CI-MS 283(MC3H5*, 0.1). 271(MC2H5*, 2). 245(MH'. Clis. 7). 243(MW. 22). 172(Clis. 30). 170(100). 1 46(Clis. 8). 144(26). lOO(2). 75(95)

170 144

Y

L L H 170 144 Clis: chlorine isotope (M.W.: 37)

M-5

M-6

Clis: chlorine isotope (M.W.: 37) Figure 1-Structures and a summary of MS data for fenobam and its metabolites and derivatives.

186 /Journal of Pharmaceutical Sciences Vol, 84, No. 2, February 1995

acetate buffer (pH 5.1) and adjusted to pH 5.1 with glacial acetic acid. The acidic urine (505 mL, 165.22 pCi, 48.2% of total dose) was hydrolyzed with Glusulase at 37 "C overnight and adsorbed onto XAD-2 resin (16 g). The material was then eluted off the resin with two methanol washes (215 and 184 mL). The combined methanol extract (65.0% of total radioactivity present in the original sample) was evaporated to dryness to yield 205 mg of total residue. The residue was suspended in hot methylene chloride/methanol(2:1) and chromatographed on a silica gel PF 254 column (30 g). "he column was eluted with methylene chloride with increasing amounts of methanol and 5-mL fractions were collected. All of the applied radioactivity was eluted from this column.

Isolation, Purification, and Identification of N-Desmethyl- 4-hydrosyphenylfenobam (M-3)-After evaporation of fraction no. 14 from the silica gel column, the residue was dissolved in hot methanol (0.5 mL). The methanol solution was concentrated to a small volume and formed a white crystalline substance. The crystal- line product was collected and analyzed by CI-MS.

Isolation, Purification, and Identification of N43-Chloro-4- hydroxypheny1)-N'-[imino(methylamino)methyl]urea (M-4)- Fraction nos. 98-100 of the silica gel column were combined and concentrated to a small volume. After filtration, a beige crystalline material formed. The mother liquor was applied as bands to silica gel GF plates and developed twice in ethylene dichloridehexane/ methanol (52:34:14). The major band at Rf 0.27, which showed a positive test with phosphomolybdic acid test solution (i.e., phenolic), was removed from each plate and extracted twice with 2 mL of ethyl acetate/methanol(4:1). The first extract was evaporated to dryness and the residue was resuspended in hot methanol. The methanol solution was applied as bands to silica gel GF plates and developed twice in the solvent system just described. The band at Rf 0.3 on each plate was removed and extracted with 2 mL of ethyl acetate/ methanol (4:l). The ethyl acetate extract was evaporated to yield a residue which was analyzed by electron impact-(EI-) and CI-MS.

Preparation of N-(3-Chloro-4-methoxyphenyl)-N'-[imino- (methylamino)methyl]urea (M-4akMetabolite 4 was dissolved in 0.5 mL of methanol to which an excess amount of ethereal diazo- methane was added. The ether solution was maintained for several hours at room temperature and evaporated to dryness under nitrogen to yield a methylated product that was analyzed by MS.

Isolation, Purification, and Identification of Two Minor Products (M-5, M-6)-The three minor bands [at Rf 0.18, 0.34 (M- 6), and 0.4 (M-5)1 removed from two silica gel GF plates described for fraction nos. 98-100 were extracted individually with ethyl acetatelmethanol (4:l) and analyzed by EI- and CI-MS. No structural information was gained for the isolate a t Rf 0.18.

Spectroscopic Analysis-The melting points were taken in a capillary tube on a Hoover capillary melting point apparatus and were corrected. The UV spectra were obtained on a Perkin-Elmer 552 spectrophotometer. The IR spectra were taken on a Perkin-Elmer IR grating spectrophotometer (model 283) with 1% KBr wafers of the compounds. The NMResonance spectra were determined in DMSO- d6 on a Perkin-Elmer R32 NMR spectrometer or a JEOLPourier transfer NMR spectrometer (model FX609), with tetramethylsilane as internal standard and chemical shifts reported in ppm (6) downfield from TMS (0 ppm). The EI-MS spectra were obtained on a Hitachi Perkin-Elmer RMU-6 spectrometer by direct inlet a t 70 eV. The CI- MS spectra were obtained on a Finnigan model 9500-3300-6100 GC/ MSDS by direct inlet. The mass spectrometer was operated a t an electron energy of 100 eV, source temperature a t 100 "C, and reagent gas source pressure of 1 torr for methane.

Total Radioactivity Determination-Total radioactivity in each urine sample or column fraction was determined by directly counting 0.1-mL aliquots of each sample in 10 mL of Biofluor. The samples were counted in a refrigerated liquid scintillation counter (Beckman LS-2OOB or Searle Analytic 81), and counting efficiency was deter- mined with an external standard.

TLC Conditions for Analytical Work-The radioactive metabo- lites were separated by direct thin-layer radiochromatographic pro- cedures. The samples were applied as bands or spots 2.0 cm from the bottom of 5 x 20 cm silica gel GF TLC plates. Reference standards (-10-20 pg) of nonradioactive fenobam, 4-hydroxyphen- ylfenobam (M-2, RWJ-34734), and 5-hydroxyimidazolfenobam (M-1, RWJ-34892) were chromatographed in the reference channel. The plates were removed when the solvent front had migrated 14 cm above the origin. The reference standards and metabolites were visualized

in vifro

' P

Untreatment

Rat Liver Preparation

Figure 2-TLC metabolic profiles of fenobam metabolite extract of rat liver preparation.

Table 1-Percent of Sample Radioactivity for Fenobarn and Its Metabolites

Compound In Vitro In Vivo

Fenobam M-1 M-2 M-3 M-4 M-5 M-6 Unknowns Total % Sample Radioactivity Identified

55 19 4

- 22 78

-a

12 -

a 2a 5 5

42 58

a-, Not detected. with a short-wave UV light and/or detected by a radioscanner (Berthold model 600, Varian Instruments). The TLC plates were also analyzed by zone removal of the silica gel corresponding to reference samples. Each zone was scraped and transferred to a liquid scintil- lation vial containing 0.5 mL of water, and counted in 10 mL of Biofluor.

Results and Discussion In Vitro Metabolism-The extraction and TLC separation

of the incubation mixtures resulted in the isolation of four metabolic products. A representative TLC metabolic profile is shown in Figure 2. The major component was confirmed by TLC and CI-MS (Figure 1) to be unmetabolized fenobam (55% of the sample radioactivity; Table 1). Metabolite 1 (19% of the sample radioactivity; Table 1) was identified as 5-hy- droxyfenobam on the basis of its MS fragment ions (Figure 1). Final confirmation was obtained with a synthetic sample (RWJ-34892)12 for comparison by TLC and MS. Metabolite 2 (4% of the sample radioactivity; Table 1) was identified as 4-hydroxyphenylfenobam on the basis of its MS data (Figure 1). The third metabolite (19% of the sample radioactivity) has not been assigned a structure.

Journal of Pharmaceutical Sciences / 107 Vol. 84, No. 2, Februaty 1995

In Vivo Metabolism-After Glusulase hydrolysis and XAD-2-resin adsorption, the residue from the rat urine was separated by Sephadex LH-20 and silica gel column chro- matographies, which subsequently led to the isolation of five metabolites (Figure 1 ); they are, the two major metabolites 4-hydroxyphenylfenobam (M-2, 12% of the sample radioac- tivity) and N-(3-chloro-4-hydroxyphenyl)-N'-[imino(methyl- amino)methyl]urea (M-4, 28% of the sample radioactivity), N-desmethyl-4-hydroxyphenylfenobam (M-3,8% of the sample radioactivity), and two metabolic cleavage products (M-5 and M-6, each 5% of the sample radioactivity) as minor metabolites (Table 1).

4-Hydroxyphenylfenobam (M-2) was first isolated from the rat liver 9000 x g supernatant in vitro study as a minor metabolite (4% of the sample radioactivity; Table 1). The structure of M-2 was proposed on the basis of CI-MS, and it was obtained as a beige crystalline product with mp of 231- 232 "C (232-233 "C for RWJ-34734) from rat urine. The UV spectrum of the product indicated a typical benzenoid absorp- tion at 271 nm. A slightly bathochromic shift to 282 nm, observed under basic conditions, was suggestive of the pres- ence of a phenolic hydroxy group. This conclusion was also supported by a positive result with phosphomolybdic acid test solution. The IR spectrum indicated the presence of a broad hydrogen-bonded hydroxy band at 3200 cm-l, with a shoulder from 3500 cm-l, and the presence of an amide band at 3120, 1670, and 1570 cm-'. The IH NMR spectrum clearly revealed two singlet resonances at 62.98 for N-CH3 and 63.98 for CH:! protons, and three aromatic protons appeared as an Al3X system at 66.83,7.30, and 7.73, which supported substitution of an hydroxy group at the C4 position of the phenyl ring. The 'H NMR also showed one proton singlet a t 69.27 arising from the exchangeable phenolic hydroxy group. The CI-MS spec- trum exhibited a MH+ ion at rnlz 283, consistent with the formula C11H11N403C1-H+. The two most intense and impor- tant fragments ions at rnlz 170 (100%) and 114 agreed with the proposed structure (Figure 1). Direct comparison of the urinary product with synthetic material (RWJ-34734P con- firmed the structural assignment.

Methylation of 4-hydroxyphenylfenobam with diazomethane yielded a dimethylated product (M-2a) for the latter. The structure was assigned by 'H NMR (2N-CH3 groups at 62.98 and 3.10) and CI-MS data (Figure 1). N-Desmethyl-4-hydroxyphenylfenobam (M-3, 8% of the

sample radioactivity) was isolated as a minor metabolite from rat urine (Table 1). The CI-MS spectrum (Figure 1) showed a weak but observable molecular ion at rnlz 269, corresponding to CloH9N403C1~H+. The MS data suggested the presence of a phenolic hydroxy group that was further confirmed by a positive result with phosphomolybdic acid test solution. The CI-MS data also contained two very intense peaks at rnlz 170 (100%) and 100 that could be assigned to the fragment ions C7Ha02C1 and C3Ha30, thereby locating the hydroxy group on the phenyl ring and indicating that the imidazole nitrogen lacked the methyl group. The overall fragment ions in CI- MS confirmed the structure of N-desmethyl-4-hydroxyphen- ylfenobam. N-~3-Chloro-4-hydroxyphenyl)-N'-[imino~methylamino )-

methyllurea (M-4, 28% of the radioactive sample) was the major metabolite isolated from rat urine (Table 1). The metabolite gave a positive result with phosphomolybdic acid test solution. The compound did not exhibit a molecular ion in the EI-MS spectrum because decomposition, but the CI- MS data (Figure 1) revealed a protonated molecular ion at mlz 243, corresponding to the formula CgH11NO&lH+, with two adduct-molecular ions at rnlz 283 (MC3H7+) and 271 (MCzHs+). The prominent fragment ion at mlz 170 (100%) indicated substitution of a hydroxy group on the phenyl ring. The 0-methylated product was obtained by treatment of M-4

100 75 ~ = c = ~ - ~ c m x m/z 205 ( M C2H5+)

m/z 257 ( MH+) I,!'

257 c . , I

100 150 200 250 mlz

Figure 3-CI-MS (methane) for metabolite 4 methyl ether.

,rJ2-dC / FeMLIm

I' -L, 1 4= I \ \ .$J& M 2 \ I con,ulatss

Scheme 1-Proposed biotransformation pathways for fenobam in the rat. (The brackets indicate metabolites which have not been identified.)

with diazomethane. The EI- and CI-MS spectra of this methyl ether (M-4a) revealed molecular and protonated molecular ions at rnlz 256 (M+) and 257 (MH+), respectively, correspond- ing to the formula C10H13N402C1, which is 14 amu higher than the parent substance (the important fragment ions and a representative spectrum are shown in Figures 1 and 3).

The MS spectral data for two additional minor metabolites, metabolites 5 and 6 (each 5% of the sample radioactivity) were obtained (Table 1). Metabolite 5 gave the CI-MS data shown in Figure 1. Thus, it was assumed that metabolite 5 is probably a cleavage product. It could, however, be deduced that metabolite 5 was nonphenolic. The CI- and EI-MS data supporting the structure of metabolite 6 (shown in Figure 1) was obtained for another minor metabolite.

Direct spot TLC assays of untreated and Glusulase hydro- lyzed urine from rats are shown in Figure 2. Rats excrete little or no unchanged fenobam. The majority of the radio- activity in rats is accounted for by polar materials (M-3, M-4, M-5, and M-6, and possibly additional unidentified metabo- lites). Metabolites 1 and 2 appear to be less polar and are excreted as conjugates by rats.

The use of the in vitro system allowed an assessment of structures for two metabolites, indicating that oxidation of the molecule occurred on both the phenyl ring and the creatinine ring. These biotransformations were demonstrated to occur in vivo in rats and were further supported by the identification of phenolic metabolites (M-2, M-3, and M-4) and metabolites (M-5 and M-6) that indicated that oxidation had initially occurred on the five-membered ring in the urine (Scheme 1). The major urinary metabolite (M-4) appeared to be a product of oxidation on both portions of the molecule. The four proposed biotransformation pathways for fenobam in the rat are shown in Scheme 1; they are, (1) hydroxylation at the phenyl ring to form 4-hydroxyphenyl fenobam, a major pathway in vivo but a minor in vitro; (2) hyroxylation at the creatinine ring to form 5-hydroxyfenobam, a dominant path- way in vitro but not in vivo; (3) oxidative cleavage a t the creatinine ring and loss of a ketene unit via water hydrolysis

188 / Journal of Pharmaceutical Sciences Vol. 84, No. 2, February 1995

of the open forms of carbinolamine analogue, an important pathway for 4-hydroxyphenylfenobam in vivo but, a minor pathway for fenobam; and (4) N-demethylation, a minor pathway for both fenobam and 4-hydroxyphenylfenobam in vivo.

Fenobam was extensively metabolized by rats in vivo, but not in vitro. Fenobam may be highly protein bound to rat liver S-9 fraction, which would hinder in vitro metabolism. The very polar fraction of the Glusulase-treated urine (42% of the sample radioactivity) was extremely water soluble and solvent unextractable. "his fraction has not been isolated in sufficient quantities nor with sufficient purity for spectroscopic characterization and, therefore, its concents remains un- known.

References and Notes 1. Rasmussen, C. R.; Gardocki, J. R. Abstracts, 178th ACS National

Meeting, 1979; Division of Medicinal Chemistry Abstract 24. 2. Rasmussen, C. R. US. Patent 3 983 135, 1976. 3. Fabre, F. L. Clin. Res. 1977,25, 269A. 4. Itil, T. M.; Seaman, P. A.; Hugue, M.; Mukhopadhyay, S.;

Biasucci, D.; Ng, K T.; Ciccone, P. E. Curr. Ther. Res. Clin. Exp., 1978,24, 708-724.

5. Pecknold, J. C.; McClure, D. J.; Appeltauer, L. Curr. Ther. Res.

6. Friedman, C. T. H.; Davis, L. J.; Ciccone, P. E.; Rubin, R. T.

7. Patel, J. B.; Martin, C.; Malick, J. B. Eur. J . Pharmacol. 1982,

Clin. Exp. 1980,27,119-123.

Curr. Ther. Res. Clin. Exp. 1980,27, 144-151.

86.295-298. 8. Pecknold, J. C.; McClure, D. J.; Appletauer, L. J . Clin. Psychop-

9. Lapierre, Y. D.; Osewumi, L. K. Curr. Ther. Clin. EXD. 1982. harmacol. 1982,2, 129-133.

31,-95-101. 10. Goldberg, M. E.; Salama, A. I.; Patel, J. B.; Malick, J. B.

Neuropharmacology 1983, 22, 1499-1504. 11. Wu, W. N.; Ng, K. T.; Mutter, M. S.; ONeill, P. J. Abstracts,

35th ASMS Annual Conference on Mass Spectrometry and Allied Topics, American Society of Mass Spectrometry. Denver, CO, 1987; Abstract TPB14.

12. Rasmussen, C. R., The R. W. Johnson Pharmaceutical Research Institute, personal communication.

Acknowledgments We thank Dr. C. R. Rasmussen for synthesizing fenobam metabo-

lites, Dr. L. E. Weaner for synthesizing ['*C]fenobam, Mr. J . Kalbron and Mr. M. Mutter for the MS and NMR analysis of fenobam metabolites, and Drs. A. R. Takacs and B. L. Ferraiolo for reviewing the manuscript.

J89404777

Journal of Pharmaceutical Sciences / 189 Vol. 84, No. 2, Februaty 7995