Studies on the in Vivo Biotransformation of the TobaccoAlkaloid â-Nicotyrine†

Xin Liu,‡ Kay Castagnoli,‡ Cornelis J. Van der Schyf,‡,§ andNeal Castagnoli, Jr.*,‡,§

Harvey W. Peters Center, Department of Chemistry, and Department of Biomedical Sciences andPathobiology, VA-MD Regional College of Veterinary Medicine, Virginia Tech,

Blacksburg, Virginia 24061

Received July 8, 1999

This paper reports the results of studies on the in vivo metabolic fate of the tobacco alkaloid1-methyl-2-(3-pyridinyl)pyrrole (â-nicotyrine) in New Zealand white rabbits. Two previouslycharacterized metabolites, 5-hydroxy-1-methyl-5-(3-pyridinyl)-2-pyrrolidinone (5-hydroxycoti-nine) and 2-hydroxy-1-methyl-5-(3-pyridinyl)-3-pyrrolin-2-one, were present in low concentra-tions in the urine of the treated animals. The major urinary metabolite of â-nicotyrine wasidentified as cis-3′-hydroxy-1-methyl-5-(3-pyridinyl)-2-pyrrolidinone (cis-3′-hydroxycotinine),the diastereoisomer of the major urinary metabolite of (S)-nicotine. The pathway leading tocis-3′-hydroxycotinine is proposed to proceed via autoxidation of 2-hydroxy-1-methyl-5-(3-pyridinyl)pyrrole, a postulated cytochrome P450-generated metabolite of â-nicotyrine, followedby reduction of the carbon-carbon double bond present in the resulting 3-hydroxy-3-pyrrolin-2-one species. This proposal is supported by the in vivo biotransformation of 2-acetoxy-1-methyl-5-(3-pyridinyl)pyrrole, a latent form of the putative hydroxypyrrole intermediate, to cis-3′-hydroxycotinine. The in vivo conversion of 5-hydroxy-1-methyl-5-(3-pyridinyl)-3-pyrrolin-2-one to 5-hydroxycotinine is offered as evidence that supports the proposed reduction step.

Introduction

The use of tobacco products is known to cause serioushealth problems (1-3). While the pharmacology andtoxicology of (S)-nicotine (1), the principal tobacco alka-loid, have been examined extensively, relatively littlework has focused on the minor tobacco alkaloids. We havebeen interested in the metabolic fate of 1-methyl-2-(3-pyridinyl)pyrrole [â-nicotyrine (2)] in part because ofreports that structurally related five-membered het-eroarenes may be biotransformed to reactive metabolitesthat contribute to their toxic properties (4-6). Resultsfrom preliminary studies on â-nicotyrine employing cy-tochrome P450 rich Clara cells isolated from rabbit lungsuggest that this alkaloid is bioactivated in an NADPH-dependent process to form pneumotoxic metabolites (7).

Previous in vivo studies have established that â-nico-tyrine is biotransformed extensively in the rabbit and

human to unknown metabolites (8, 9). NADPH-supple-mented rabbit lung and liver microsomal preparationsconvert â-nicotyrine to four metabolites (10). The primarymetabolites (Scheme 1) were shown to be an equilibriummixture of two unstable pyrrolinone species, 1-methyl-5-(3-pyridinyl)-4-pyrrolin-2-one (3) and 1-methyl-5-(3-pyridinyl)-3-pyrrolin-2-one (4). These pyrrolinones un-dergo autoxidation, to form 5-hydroxy-1-methyl-5-(3-pyridinyl)-3-pyrrolin-2-one (5), and hydration, to form5-hydroxycotinine (6) (11). This paper summarizes theresults of our in vivo metabolic studies on â-nicotyrinein New Zealand (NZ) white rabbits.

Materials and Methods

Chemistry. All reactions were carried out under a nitrogenatmosphere. Reagents, chemicals, and analytical standards notdescribed elsewhere were purchased from Aldrich Chemical Co.(Milwaukee, WI). HPLC grade solvents, acetonitrile, triethyl-amine, and acetic acid were purchased from Fisher Scientific(Pittsburgh, PA). Water was obtained from a Milli-Q system(Millipore, Bedford, MA). The hydroperchlorate salt of (3R,5S)-trans-3′-hydroxycotinine (7) was a gift from P. Jacob, III (12).â-Nicotyrine (13), 2-acetoxy-1-methyl-5-(3-pyridinyl)pyrrole (8)-(10), and racemic cis-3′-hydroxycotinine (9) (14) were synthe-sized according to published methods.

GC/electron ionization mass spectrometry (GC/EIMS)1 analy-ses were performed on a Hewlett-Packard (HP) 5970 massselective detector interfaced with an HP 5890 gas chromato-graph using an HP-1 capillary column (12.5 m × 0.2 mm × 0.33µm). Samples were injected into a split/splitless injector port

† A preliminary report of this work appeared in Nicotine Safety andToxicity (Benowitz, N. L., Ed.), N. Castagnoli et al., 1998, pp 57-65,Oxford University Press, New York and Oxford.

* To whom correspondence should be addressed: Department ofChemistry, Virginia Tech, Blacksburg, VA 24061-0212. Telephone:(540) 231-8200. Fax: (540) 231-8890. E-mail: ncastagnoli@chemserver.chem.vt.edu.

‡ Harvey W. Peters Center, Department of Chemistry, VirginiaTech.

§ Department of Biomedical Sciences and Pathobiology, VA-MDRegional College of Veterinary Medicine, Virginia Tech.

1 Abbreviations: GC-EIMS, gas chromatography/electron ionizationmass spectrometry; HP, Hewlett-Packard; DA, diode array; ICR, albinomice descended from animals from Charles River Laboratories, Wilm-ington, MA; NZ, New Zealand.

336 Chem. Res. Toxicol. 2000, 13, 336-341

10.1021/tx990124t CCC: $19.00 © 2000 American Chemical SocietyPublished on Web 04/20/2000

(250 °C) operating in the splitless mode. The column headpressure was 28 psi. The temperature program used for the GC/EIMS analyses, except as noted, was as follows. The initialtemperature of 100 °C was maintained for 1 min and then wasincreased at a rate of 25 °C/min to 275 °C where it wasmaintained for 2 min before recycling (program 1). For the GC/EIMS analysis of the chloro diastereoisomers derived from the3′-hydroxycotinines, we used the following temperature pro-gram. The initial temperature of 70 °C was maintained for 1min and then was increased at a rate of 20 °C/min to 270 °Cwhere it was maintained for 2 min before recycling (program2) (12, 15). HPLC UV-diode array (DA) analyses were performedusing an HP 1040A detector, Beckman 114 pumps, a Beckman421A gradient controller, a Rheodyne injector, and an AlltechEconosil C8 column (25 cm × 4.6 mm, 10 µm). The followinggradient elution profile was employed at a flow rate of 1 mL/min using mobile phase A [water/acetonitrile/acetic acid/tri-ethylamine (90:10:0.5:0.1, v:v:v:v)] and mobile phase B (100%acetonitrile): from 0 to 10 min, 100% A; from 10 to 20 min, 0 to60% B; from 20 to 30 min, 60% B; from 30 to 35 min, 60 to 0%B; from 35 to 45 min, 0% B (10). Typical retention times(detection wavelengths) were as follows: cis- and trans-3′-hydroxycotinine, 7.2 min (260 nm); 5-hydroxycotinine, 8.5 min(260 nm); the 5-hydroxypyrrolinone 4, 11.7 min (260 nm); andâ-nicotyrine (2), 23.0 min (280 nm).

Syntheses of tert-Butyldimethylsilyl Ether Derivatives10 and 11 of (3R,5S)-trans-3′-Hydroxycotinine (7) andRacemic cis-3′-Hydroxycotinine (9), respectively (15).Solutions of (3R,5S)-trans-3′-hydroxycotinine (7) or racemic cis-3′-hydroxycotinine (9), as their free bases (2 mg), in 0.5 mL ofanhydrous N,N-dimethylacetamide containing 12 mg of tert-butyldimethylsilyl chloride and 12 mg of imidazole were vor-texed for 1 min in a capped incubation tube and then wereplaced on a heating block at 80 °C for 1 h. After the mixtureshad cooled to room temperature, 0.4 mL of a toluene/butanolsolution (90:10) and 1 mL of water were added. The resultingmixtures were vortexed for 1 min, centrifuged, and placed in adry ice/acetone bath to freeze out the aqueous layer. The upperorganic layers were analyzed by GC/EIMS. The same procedureas described above was followed to analyze the residue of urinesamples obtained from â-nicotyrine-treated rabbits (see belowfor details).

(3S,5S)-trans-3-Chlorocotinine (12) and Racemic cis-3-Chlorocotinine (13). These compounds were synthesizedstarting from (3R,5S)-trans-3′-hydroxycotinine (7) and racemiccis-3′-hydroxycotinine (9) by reaction with thionyl chloride inthe presence of pyridine as previously described (12, 16). Theresulting diastereomeric chlorocotinines had different GC reten-tion times (program 2, 5.30 min for the trans-isomer derivedfrom cis-3′-hydroxycotinine and 5.45 min for the cis-isomerderived from trans-3′-hydroxycotinine) and nearly identical GC/EIMS data: m/z (abundance), 212 (12%, M+•, 37Cl), 210 (35%,M+•, 35Cl), 175 (15%), 153 (10%), 134 (10%, 37Cl), 132 (28%,

35Cl), 118 (100%), 91 (20%), 78 (8%). This reaction also was usedto convert metabolic 3′-hydroxycotinine to the corresponding3-chloro analogue as follows. The residue obtained from theSoxhlet extraction of the lyophilized rabbit urine sample (500µg) in 2 mL of dichloromethane was heated for 2 h at 80 °Cwith 200 µL of thionyl chloride and 2 drops of pyridine. Afterremoval of the volatile components, the residue in 0.5 mL ofsaturated aqueous sodium carbonate was vortexed with 1 mLof ethyl acetate for 1 min. After standing, the mixture was cooledin a dry ice/acetone bath to freeze out the aqueous phase andthe upper organic layer was analyzed by GC/EIMS.

Metabolism. All animals were housed in the LaboratoryAnimal Resource at Virginia Tech. Care and use of labora-tory animals followed the guidelines issued by the NationalResearch Council. All protocols had prior approval from theVirginia Tech Animal Care Committee in accordance with NIHrequirements.

ICR male mice (20-30 g) were used to establish dosingregimens for subsequent in vivo â-nicotyrine metabolism stud-ies. NZ white male rabbits (1.8-2.4 kg) were used for the invivo studies of â-nicotyrine, (3R,5S)-trans-3′-hydroxycotinine,the 5-hydroxypyrrolinone 5, and the 5-acetoxypyrrole 8. Animalswere placed individually in metabolic cages, and control urinesamples were collected for 24 h. After intraperitoneal (ip)administration of the appropriate compound [50 mg/kg exceptfor â-nicotyrine (100 mg/kg)], urine was collected over a periodof 24 h (treated urine samples). All urine samples werecentrifuged at 10000g for 15 min. The clear supernatants wereremoved and lyophilized, and the resulting residues (approxi-mately 2 g/sample) were extracted for 2 h with 150 mL ofbenzene using a Soxhlet extractor. Following evaporation of thebenzene, the extracts were reconstituted in 2 mL of methanol.The resulting samples were analyzed by HPLC and GC/EIMS.

Results and Discussion

(S)-Nicotine is highly toxic to mammals (17, 18) dueto its potent cholinergic stimulating properties that aremediated by the N-methylpyrrolidinium conjugate acidof the alkaloid (19, 20). We anticipated that the acutetoxicity of â-nicotyrine should be much lower since thepyrrolyl group will not be protonated at physiological pH.At ip doses of up to 100 mg/kg, neither mice nor rabbitsshowed any abnormal behavior. Therefore, our in vivometabolic studies employed an ip dose of 100 mg/kg.

The high polarity of the in vitro metabolites 5 and 6 ofâ-nicotyrine suggested the need for an efficient extractionprocedure. Consequently, urine samples were lyophilized,and the residues were subjected to continuous Soxhletextraction with benzene for 2 h. The HPLC UV-diodearray assay employed in our in vitro studies (10) was

Scheme 1. Metabolic Fate of â-Nicotyrine in NADPH-Supplemented Rabbit Liver Microsomal Preparations

In Vivo Biotransformation of â-Nicotyrine Chem. Res. Toxicol., Vol. 13, No. 5, 2000 337

utilized for these in vivo studies except that the samples,which also were to be analyzed by GC/EIMS, werereconstituted in methanol prior to analysis. Under theseconditions, 100% of â-nicotyrine, more than 90% of the5-hydroxypyrrolinone 5, and 80% of 5-hydroxycotinine (6)could be recovered from spiked control urine. The HPLCtracings of extracts obtained from control (A) and treated(B) urine samples are shown in Figure 1. One major(tR ) 8 min) and two minor (tR ) 10 and 13 min)metabolite-related peaks were present in the tracing ofthe extract from the treated urine samples. The back-ground at the retention time (23 min) of â-nicotyrine wascomplex, but no significant increase in signal was ob-served in the treated urine-extracted sample. Anyâ-nicotyrine present was below the levels of detection ina chromatogram obtained at 280 nm, the λmax for thiscompound (data not shown). Assignments for the twominor peaks were made by comparisons of their HPLC-DA characteristics (retention times and UV spectra) withthose of synthetic 5-hydroxycotinine (6, tR ) 10 min,Figure 1D) and the 5-hydroxypyrrolinone (5, tR ) 13 min,Figure 1C). When the urine samples from treated rabbitswere spiked with synthetic 5 and 6, the resulting peakintensities increased accordingly; no evidence of separa-tion of the metabolites from the standards was evident.Both 5 and 6 were anticipated metabolites since both areformed in rabbit lung and liver microsomal incubationsof â-nicotyrine (11). It should be noted that 5-hydroxy-cotinine also has been reported to be an in vivo metaboliteof (S)-nicotine (22) and (S)-cotinine (23). The major peakeluting at 8 min had not been observed in the HPLC

chromatograms of the microsomal incubation extracts.The HPLC-DA spectrum of this metabolite (Figure 1E)was similar to the corresponding spectra of 5 and 6.

The total ion chromatogram obtained by GC/EIMSanalysis of the urinary extracts from the treated animalsalso displayed a major peak (tR ) 5.0 min) that was notpresent in the corresponding control tracing (data notshown and ref 21). The full mass spectrum exhibited anapparent molecular ion at m/z 192, which is consistentwith a molecular formula of C10H12N2O2 correspondingto cotinine + 16 atomic mass units. The possibility thatthis unknown metabolite was 5-hydroxycotinine (6) couldbe ruled out by comparison of its LC and GC/EIMSproperties with those of synthetic 6 (21).

A structure isomeric with 6 is trans-3′-hydroxycotinine(7), the major metabolite detected in the urine of smokers(24) and of animals treated with (S)-cotinine (23, 25). TheHPLC-DA and GC/EIMS properties of synthetic (3R,5S)-trans-3′-hydroxycotinine were indistinguishable fromthose of the metabolite. Furthermore, the GC/EIMScharacteristics of the tert-butyldimethylsilyl ether deriva-tive(s) 10 prepared from 7 and from the metabolite alsowere identical (21). Unfortunately, the HPLC-DA (Figure1E) and GC/EIMS characteristics of synthetic cis-3′-hydroxycotinine (9) and its tert-butyldimethylsilyl etherderivative 11 (15) also were indistinguishable from thecorresponding characteristics observed with the metabo-lite. Consequently, although the structure of the metabo-lite undoubtedly was a 3′-hydroxycotinine, it was notpossible to assign the relative stereochemistry at C-3 withthe available evidence since we were unable to separate

Figure 1. (Left) HPLC chromatogram (monitored at 260 nm) of a representative urine sample collected from a control rabbit (A)and a rabbit treated with â-nicotyrine (B). (Right) HPLC-DA spectra showing the spectral characteristics of the peaks identified as5 (C), 6 (D), and a mixture of 9 (E) from the same sample (solid lines) and the corresponding spectral characteristics of the standardcompounds (dotted lines). Note that the HPLC-DA characteristics of synthetic trans-3′-hydroxycotinine (7) were indistinguishablefrom those of synthetic cis-3′-hydroxycotinine (9). See the text for evidence supporting the assignment of the metabolite as the cis-isomer.

338 Chem. Res. Toxicol., Vol. 13, No. 5, 2000 Liu et al.

these diastereomers or their tert-butyldimethylsilyl etherderivatives in either system.2

We next turned our attention to the 3-chlorocotininesthat may be prepared in yields of >90% by treating thecorresponding 3′-hydroxycotinines with thionyl chloride(16). Treatment of highly purified synthetic (3R,5S)-trans-3′-hydroxycotinine and synthetic racemic cis-3′-hydroxycotinine with thionyl chloride gave the corre-sponding 3-chlorocotinine species with GC retentiontimes of 5.30 and 5.45 min, respectively. This reaction,which is known to proceed by SN2 attack of Cl- on thechlorosulfonyloxy intermediate 14 as shown in Scheme2 for (3R,5S)-trans-3′-hydroxycotinine (7), leads to inver-

sion of configuration at C-3 (16). Therefore, the 3-chlo-rocotinine produced from trans-3′-hydroxycotinine mustbe the cis-isomer 13, and the 3-chlorocotinine producedfrom cis-3′-hydroxycotinine must be the trans-isomer 12.The GC/EIMS characteristics of the product (i.e., theretention time and mass spectrum) obtained by treatingthe 3′-hydroxycotinine derived metabolically from â-nico-tyrine with thionyl chloride were found to be identicalto those of trans-3-chlorocotinine. Consequently, the 3′-hydroxycotinine metabolite itself must have the cis-configuration 9 rather than the trans-configuration of the(S)-nicotine metabolite. The amounts of the trans-isomer,if present, were below the levels of detection.

The metabolic pathway leading to the formation of cis-3′-hydroxycotinine from â-nicotyrine is not apparent. Thepossibility that trans-3′-hydroxycotinine undergoes inver-sion of configuration to form the cis-isomer was ruled outsince we could detect only the trans-isomer in the urineobtained from a rabbit treated ip with (3R,5S)-trans-3′-hydroxycotinine. A second possible pathway proceeds viathe P450-generated pyrrolinones 3 and 4 (Scheme 3). Theconversion of these pyrrolinones to the 5-hydroxypyrroli-none 5 has been proposed to proceed via autoxidation of2-hydroxy-1-methyl-5-(3-pyridinyl)pyrrole (15), a tau-tomer of 3 and 4 (11). Proton loss and single-electrontransfer from 15 to dioxygen lead to the resonance-

2 Voncken et al. (26), O’Doherty et al. (27), and, more recently,Murphy et al. (28) have reported the separation of cis- and trans-3′-hydroxycotinine by GC and by HPLC. Our attempts to resolve thesecompounds chromatographically, however, were unsuccessful.

Scheme 2. Conversion of trans-3′-Hydroxycotinine to cis-3-Chlorocotinine

Scheme 3. Proposed in Vivo Biotransformation Pathway of â-Nicotyrine (2) to cis-3′-Hydroxycotinine (9)

In Vivo Biotransformation of â-Nicotyrine Chem. Res. Toxicol., Vol. 13, No. 5, 2000 339

stabilized radical 16. A second oxidative step coupledwith the introduction of the hydroxyl group at C-5(radical recombination with hydroxyl radical and/or withhydroperoxyl radical followed by a hydrolytic sequence)yields 5. The isomeric 3′-hydroxypyrrolinone 17 also couldbe formed via radical recombination at C-3 of 16. Rear-rangement of 17 to 18, analogous to the known equilib-rium between 3 and 4 (10), followed by reduction of thecarbon-carbon double bond of 18 would result in 9.Although neither 17 nor 18 was characterized in ourearlier in vitro studies on the metabolism of â-nicotyrine(10, 11), we did report the presence in the GC/EIMStracing of a hydroxypyrrolinone with a mass fragmenta-tion pattern almost identical to that observed for 5 (11).We also know from earlier synthetic efforts (14) that 18is chemically unstable, behavior which could explain ourinability to detect this compound as an in vitro metaboliteof â-nicotyrine.

These considerations led us to examine the in vivometabolic fate of the hydroxypyrrole intermediate 15.Since 15 itself was not available synthetically, thecorresponding 2-acetoxy derivative 8, which undergoeshydrolysis to yield 15 (10), was used as a precursor of 15in these studies. HPLC-DA analysis of a urine sampleobtained from a rabbit treated with 8 showed a majorpeak corresponding to 3′-hydroxycotinine, a minor peakcorresponding to the 5-hydroxypyrrolinone 5, and asecond minor peak corresponding to 5-hydroxycotinine(6). The GC/EIMS analysis confirmed the presence of the3′-hydroxycotinine that was shown, by conversion to itschloro analogue, to have the expected cis-configuration(21). Consequently, we conclude that â-nicotyrine isbiotransformed in vivo via the corresponding 2-hydroxy-pyrrole intermediate 15 to cis-3′-hydroxycotinine (9). Thekey second step proposed for the metabolic productionof cis-3′-hydroxycotinine from â-nicotyrine involves car-bon-carbon double bond reduction of 17 and/or 18. Wehave tested this hypothesis by examining the in vivo fateof the 5-hydroxy-3-pyrrolinone 5 which served as a modelfor such a reduction. HPLC-DA analysis of the urinaryextract obtained following ip administration of 5 led tothe detection of 5-hydroxycotinine as the dominant UVabsorbing peak in the HPLC-DA chromatogram (21). Thisbehavior is consistent with the proposed pathway to cis-3′-hydroxycotinine and may account for the 5-hydroxy-cotinine (6) found in the urine of rabbits treated withâ-nicotyrine. The results of these in vivo studies clearlyshow that cis-3′-hydroxycotinine is the principal urinarymetabolite of â-nicotyrine in the rabbit. This compound,however, is not present in rabbit lung and liver mito-chondrial incubation mixtures even though convincingevidence is available to suggest that the formation of cis-3′-hydroxycotinine proceeds via the 5-hydroxypyrrolespecies 15 which, in turn, is derived from the P450-generated pyrrolinones 3 and 4. The oxidation-reductionsequence outlined in Scheme 3 suggests that the reduc-tion of 17 and/or 18 to cis-3′-hydroxycotinine is mediatedby an extrahepatic pathway. Voncken et al. have reportedthe excretion of cis-3′-hydroxycotinine in the urine ofsmokers (26). The possibility that (S)-nicotine may beconverted to cis-3′-hydroxycotinine via â-nicotyrine isunder investigation.

Acknowledgment. We thank Dr. David Moore, Mr.David Gemmell, and the staff of the Laboratory of AnimalResources for their support and assistance and Dr.

Peyton Jacob, III, for his gift of (3R,5S)-trans-3′-hydroxy-cotinine. This work was supported by National Instituteon Drug Abuse Grant R01 DA11089, Council for TobaccoResearch Grant 4728R1, and the Harvey W. PetersCenter for the Study of Parkinson’s Disease and Disor-ders of the Central Nervous System.

References

(1) U.S. Department of Health and Human Services (1984) Thehealth consequence of smoking: Cardiovascular disease, a reportof the surgeons general 1984. U.S. Department of Health andHuman Services DHHS Publication (PHS) 84-50204, p 384, U.S.Government Printing Office, Washington, DC.

(2) Vial, W. C. (1986) Cigarette smoking and lung disease. Am. J.Med. Sci. 291, 130-142.

(3) Nieburg, P., Marks, J. S., McLaren, N. M., and Remington, P. L.(1985) The fetal tobacco syndrome. JAMA, J. Am. Med. Assoc.253, 2998-2999.

(4) Wilcox, A. L., Bao, Y. T., and Loeppky, R. N. (1991) Pyrroles aseffective agents for blocking amine nitrosation. Chem. Res.Toxicol. 4, 373-381.

(5) Wolson, D. M., Goldsworthy, T. L., Popp, J. A., and Betterworth,B. E. (1992) Evaluation of genotoxicity, pathological lesion, andcell proliferation in livers of rats and mice treated with furan.Environ. Mol. Mutagen. 19, 209-222.

(6) Fransson-Steen, R., Goldsworthy, T. L., Kedderis, G. L., andMaronpot, R. R. (1997) Furan-induced liver cell proliferation andapoptosis in female B6C3F1 mice. Toxicology 118, 195-204.

(7) Nichols, W. K., Covington, M. O., Yost, G. S., Liu, X., andCastagnoli, N., Jr. (1990) Bioactivation of the tobacco alkaloidâ-nicotyrine in isolated Clara cells (unpublished data).

(8) Beckett, A. H., Gorrod, J. W., and Jenner, P. (1972) A possiblerelation between pKa1 and lipid solubility and the amountsexcreted in urine of some tobacco alkaloids given to man. J.Pharm. Pharmacol. 24, 115-120.

(9) Jenner, P., and Gorrod, J. W. (1973) Comparative in vitro hepaticmetabolism of some tertiary N-methyl tobacco alkaloids in variousspecies. Res. Commun. Chem. Pathol. Pharmacol. 6, 829-843.

(10) Liu, X., Zang, L., Van der Schyf, C. J., Igarashi, K., Castagnoli,K., and Castagnoli, N., Jr. (1999) Studies on the pyrrolinonemetabolites derived from the tobacco alkaloid 1-methyl-2-(3-pyridinyl)pyrrole (â-nicotyrine). Chem. Res. Toxicol. 12, 508-512.

(11) Shigenaga, M. K., Kim, B. H., Caldera-Munoz, P., Cairns, T.,Jacob, P., III, Trevor, A. J., and Castagnoli, N., Jr. (1989) Liverand lung microsomal metabolism of the tobacco alkaloid â-nico-tyrine. Chem. Res. Toxicol. 2, 282-287.

(12) Jacob, P., III, Shulgin, A. T., and Benowitz, N. L. (1990) Synthesisof (3′R,5′S)-trans-3′-hydroxycotinine, a major metabolite of nico-tine. Metabolic formation of 3′-hydroxycotinine in humans ishighly stereoselective. J. Med. Chem. 33, 1888-1891.

(13) Shigenaga, M. K. (1989) Study on the metabolism and bioacti-vation of (S)-nicotine and â-nicotyrine. Ph.D. Thesis, Chapter V,University of California, San Francisco, CA.

(14) Dagne, E., and Castagnoli, N., Jr. (1972) Structure of hydroxy-cotinine, a nicotine metabolite. J. Med. Chem. 15, 356-360.

(15) Jacob, P., III, Shulgin, A. T., Yu, L., and Benowitz, N. L. (1992)Determination of the trans-3′-hydroxycotinine in urine of smokersusing gas chromatography with nitrogen-selective detection orselected ion monitoring. J. Chromatogr. 583, 145-154.

(16) Dagne, E. (1968) Biotransformation Studies on S-(-)-Cotinine.Ph.D. Dissertation, pp 97-98, University of California, SanFrancisco, CA.

(17) The Merck Index, 11th ed. (1989) p 1030, Merck & Co. Inc.,Rahway, NJ.

(18) Gilman, A. G., Goodman, L. S., and Murad, F. (1990) in ThePharmacological Basis of Therapeutics, 8th ed., Chapter 9,Pergamon Press, New York.

(19) Clarke, P. B. S. (1990) The central pharmacology of nicotine:Electrophysiological approaches. In Nicotine Psychopharmacology.Molecular, Cellular, and Behavioural Aspects (Wannacott, S.,Russell, M. A. H., and Stolerman, I. P., Eds.) pp 158-193, OxfordScience Publications, Oxford, U.K.

(20) Wannacott, S. (1990) Characterization of nicotine receptor sitein the brain. In Nicotine Psychopharmacology. Molecular, Cel-lular, and Behavioural Aspects (Wannacott, S., Russell, M. A. H.,and Stolerman, I. P., Eds.) pp 226-277, Oxford Science Publica-tions, Oxford, U.K.

(21) Liu, X. (1995) In vitro and in vivo studies on the biotransformationof â-nicotyrine, a minor tobacco alkaloid. Ph.D. Thesis, VirginiaTech, Blacksburg, VA.

340 Chem. Res. Toxicol., Vol. 13, No. 5, 2000 Liu et al.

(22) McKennis, H., Jr., Bowman, E. R., Turnbull, L. B., and Schwartz,S. L. (1962) The correct structure of a ketoamide arising fromthe metabolism of (-)-nicotine. J. Am. Chem. Soc. 84, 4598-4599.

(23) Dagne, E., Gruenke, L., and Castagnoli, N., Jr. (1974) Deuteriumisotope effects in the in vivo metabolism of cotinine. J. Med. Chem.17, 1330.

(24) Neurath, G. B., and Pein, F. G. (1987) Gas chromatographydetermination of trans-3′-hydroxycotinine, a major metabolite ofnicotine in smokers. J. Chromatogr. 415, 400-406.

(25) Cundy, K. C., and Crooks, P. A. (1987) Biotransformation ofprimary nicotine metabolites II. Metabolism of [3H]-S-(-)-cotininein the guinea pig: Determination of in vivo urinary metabolitesby high-performance liquid-radiochromatography. Xenobiotica 17,785-792.

(26) Voncken, P., Rustemeier, K., and Schepers, G. (1990) Identifica-tion of cis-3′-hydroxycotinine as a urinary nicotine metabolite.Xenobiotica 20, 1353-1356.

(27) O’Doherty, S., Revans, A., Smith, C. L., McBride, M., and Cooke,M. (1988) Determination of cis-3′-hydroxycotinine and trans-3′-hydroxycotinine by high-performance liquid-chromatography. J.High Resolut. Chromatogr. 11, 723-725.

(28) Murphy, S. E., Johnson, L. M., and Pullo, D. A. (1999) Charac-terization of multiple products of cytochrome P450 2A6-catalyzedcotinine metabolism. Chem. Res. Toxicol. 12, 639-645.

TX990124T

In Vivo Biotransformation of â-Nicotyrine Chem. Res. Toxicol., Vol. 13, No. 5, 2000 341