Jump to:

RJ Reynolds

Characterization of the Glucuronide Conjugate of Cotinine: A Previously Unidentified Major Metabolite of Nicotine in Smokers' Urine.

Date: 1992
Length: 6 pages
508677954-508677959
Jump To Images
snapshot_rjr 508677954-508677959

Fields

Request
Castano
1rfp1
Burton
2rfp16
1rfp76
1rfp77
1rfp78
4rfp9
Minnesota
1rfp71
Referenced Document
List of Footnotes.
Box
Rjr1212
Site
R&D
Biological Research
Caldwell Ws
Mgr
Date Loaded
27 Feb 1998
Type
PUBLISHED DOC
Author
Chem Res Toxicol
Caldwell, W.S.
Greene, J.M.
Byrd, G.D.
Chang, K.M.
Uhrig, M.S.
Debethizy, J.D.
Rjr
Crooks, P.A.
Bhatti, B.S.
Riggs, R.M.
Univ, O.F. Ky
Named Person
Aldrich Chemical
Sigma Chemical
Burdick & Jackson Labs
Fisscher Scientific
Millipore
Cambridge Isotope Laboratories
Whatman Laboratory Sales
Analytica, O.F. Branford
Hewlett Packard
Phrasor Scientific
Sage Instruments
Bollenback
Gilson Medical Electronics
Vydac
Waters
Isco
Savant Instruments
Rheodyne
Kyerematen
Anderson
UCSF Legacy ID
jrl04d00

Document Images

Text Control

Highlight Text:

OCR Text Alignment:

Image Control

Image Rotation:

Image Size:

Page 1: jrl04d00
4 @ ri I 280 Characterization of the Glucuronide Conjugate of Cotinine: A Previously Unidentified Major Metabolite of Nicotine in Smokers' Urine I 11 W. S. Caldwell,* J. M. Greene, G. D. Byrd, K. M. Chang, M. S. Uhrig, and J. D. deBethizy Research and Development, R. J. Reynolds Tobacco Co., Winston-Salem, North Carolina 27102 P. A. Crooks, B. S. Bhatti, and R. M. Riggs College of Pharmacy, University of Kentucky, Lexington, Kentucky 40536-0082 Received November 1, 1991 Chem. Res. Toxicol. 1992, 5, 280-285 Recent studies in our laboratories have confirmed that a major unidentified metabolite of nicotine in smokers' urine was susceptible to enzymatic degradation by P-glucuronidase to afford (S)-(-)-cotinine. In order to establish the identity of this metabolite, the quaternary ammonium conjugate, viz., (S)-(-)-cotinine N-glucuronide, was synthesized. Reaction of methyl 2,3,4-tri- 0-acetyl-l-bromo-l-deoxy-a-D-glucopyranuronate with (S)-(-)-cotinine at 60 °C for 3 days affords the fully protected conjugate as the bromide salt. Deprotection was accomplished in 1 M NaOH overnight at 25 °C. The deprotected inner salt was isolated by Dowex-50W cation-exchange chromatography. Electrospray mass spectra of the inner salt revealed the presence of ions with m/z 353 (M + H)+, 375 (M + Na)+, and 391 (M + K)+ as well as ions resulting from loss of water and cleavage of the glyeosidic bond. Proton and carbon nuclear magnetic resonance spectra established that the position of glucuronidation was the pyridyl nitrogen. The magnitude of the coupling between H1. and Hr of the sugar ring (8.71 Hz) and nuclear Overhauser en- hancements were consistent with the #-isomer of the glucuronide conjugate. The synthetic (S)-(-)-cotinine N-glucuronide was susceptible to enzymatic hydrolysis by fl-glucuronidase to afford (S)-(-)-cotinine. Application of a cation-exchange high-performance liquid chromato- graphic method enabled the collection of a fraction containing (S)-(-)-cotinine N-glucuronide frorn a smoker's urine. The electrospray mass spectrum of this fraction contained ions consistent with the presence of (S)-()-cotinine N-glucuronide: The concentrated fraction-was subjected--------- ------ -- to enzymatic hydrolysis by fl-glucuronidase to afford (S)-(-)-cotinine. A thermospray ion-exchange liquid chromatographic mass spectral method was developed for the direct determination of (S)-(-)-cotinine N glucuronide in smokers' urine. The data herein reported established the identity of a major urinary metabolite of nicotine as N-fl-D-glucopyranuronosyl-(S)-(-)-cotininium inner salt. Introduction Due to the widespread use of tobacco products, the disposition of nicotine and its metabolites has been the subject of intense study throughout this century (1). Nicotine and 'cotinine have been used to assess exposure to mainstream and environmental tobacco smoke (2-4); however, in smokers, urinary nicotine and cotinine account for only 10-40% of the nicotine absorbed (5). The amount of nicotine excreted depends on.amoking status, with smokers excreting 2 times more unmetabolized nicotine than nonsmokers (6). Accurate assessment of nicotine exposure therefore requires quantification of all urinary nicotine and nicotine metabolites. Methods enabling the determination of a wide variety of nicotine metabolites have been developed (7, 8), and recent reports of glucuronide conjugates of nicotine and two of its major metabolites, cotinine and 3'-hydroxy- cotinine, in the urine of smokers have appeared (6, 9, 10). Studies in our laboratories have confirmed that a major unidentified metabolite of nicotine in smokers' urine was susceptible to enzymatic degradation by fl-glucuronidase to afford (S)-(-)-cotinine (11-14). Over the last two decades, there have been several re- ports of urinary excretion of quaternary ammonium-linked glucuronide conjugates of xenobiotics by humans (15-20). These metabolites have been isolated and characterized by FAB' mass spectrometry (17-20) and NMR spectros- copy (15;16,19, 20). The enzymatic preparation of qua- ternary ammonium-linked glucuronides of several xeno- biotics (21, 22) and methods for the chemical synthesis of the glucuronide of pyridine (N-S-D-glucopyranuronosyl- pyridinium bromide) have been reported (23, 24). These reports, and consideration of the structure of (S)-(-)-co- tinine, led us to the conclusion that the glucuronide con- jugte of (S)-(-)-cotinine is most likely N-S-D-glucopyran- uronosyl-(S)-(-)-cotininium inner salt. In this study, we prepared and characterized this (S)-(-)-cotinine N-gluc- uronide (CG) by the method outlined in Scheme I and compared it to the metabolite, isolated from a smoker's urine, which liberates (S)-( )-cotinine upon enzymatic hydrolysis by fl-glucuronidase. Materials and Methods Chemicals. Glucuronolactone, (S)-(-)-cotinine, Dowex 50X1-100, acetic anhydride, pyridine, 30% hydrogen bromide, 1 Abbreviations: FAB, fast atom bombardment; NMR, nuclear mag- netic resonance; CG, (S)-(-)-cotinine N-glucuronide; TMS, tetra- methylsilane; TSP, 3-(trimethylsilyl)propionio-2,2,3,3-d4 acid, sodium salt; TLC, thin-layer chromatography; HPLC, high-performance liquid chro- matography; ES/MS, electrospray mass spectrometry; LC/MS, liquid chromatography/mass spectrometry; NOE, nuclear Overhauser effect; RIA, radioimmunoassay; ELISA, enzyme-linked immunosorbent assay; dpwr, decoupler power. I
Page 2: jrl04d00
f I Glucuronide Conjugate of Cotinine Scheme I. Synthesis of N-,6-o-Glucopyranuronosyl-(S)-(-)-cotininium Inner Salt (CG)° HO 1 COOCH3 O 8r H 1. NaOH / MeOH AcO -i 2. Pyr; AcZ0 AcO 2 4°C I H8r / AcOH ato coocH, / (s).(-) cotmine AcO ~ Aeo ~~ 9Ac 4 HO so°C 4 coo N• // ` 5 0 3 7• OAc (S)-(-)-Cotinine N-glucuronide ° Compounds shown are as follows: 1, glucuronolactone; 2, tet- ra-0-acetyl-,e-D-glifcopyranuronate; 3, methyl 2,3,4-tri-0-acetyl-l- bromo-l-deory-a-D-giucopyranuronate; 4, N-(2,3,4-tri-O-acetyl-6- methyl-S-n-glucopyranuronosyl)-(S)-(-)-contininium bromide; 5, N-P-D-glucopyranuronosyl-(S)-(-)-cotininium inner salt (CG). chloroform, formic acid, ammonium formate, sodium hydroxide, and acetic acid were obtained from Aldrich Chemical Co. (Mil- waukee, WI) and were of the highest purity available. Sodium acetate, n-saccharic acid 1,4-lactone, and A-glucuronidase (bovine liver, 624000 Fishman units/g) were purchased from Sigma Chemical Co. (St. Louis, MO). High-purity methanol and ace- tonitrile were obtained from Burdick and Jackson Labs (Musk-. egan, MI). Diethyl ether, sodium bicarbonate, magnesium sulfate, and ammonium hydroxide were obtained from Fischer Scientific (Pittsburgh, PA). All water used for this study was purified with a Milli-Q Water System (Millipore Corp., Bedford, MA) and had a resistance of at least 18 Mil-cm. Deuterated NMR solvents were purchased from Aldrich Chemical Co. Methyl-d3-cotinine (98 atom % D) was purchased from Cambridge Isotope Laboratories (Woburn, MA). Synthetic Procedures. Melting points were recorded on a Fisher Johns melting point apparatus and are uncorrected. Silica gel plates (2.5 X 7.5 cm), fluorescent at 254 nm, were purchased from Whatman Laboratory Sales (Hillsboro, OR). Solvent systems utilized for TLC monitoring of reaction products were as follows: (a) CHC13/MeOH (9:1); (b) CHC13/MeOH/Hz0 (65:30:10). 1H- NMR,13C-NMR, DEPT, and two-dimensional NMR spectra were obtained on either a Varian VXR-400S or a Varian Unity 300- MHz spectrometer (Palo Alto, CA). Spectra were run at 21 °C in either CDCl3 or DMSO-ds using TMS as internal standard, or in D20 using TSP as internal standard. NOE difference spectra were measured at 25 °C. Fast atom bombardment mass spectra (FAB-MS) were recorded on a Kratos Concept 1H spectrometer using glycerol as the matrix. Electrospray Mass Spectral Analysis. The electrospray ion source was obtained from Analytica of Branford (Branford, CT) for operation on an HP-5988A MS (Hewlett-Packard Co., Palo Alto, CA) with an HED detector (Phrasor Scientific, Duarte, CA) and an RTE A data system. The following conditions were H3C-N 2 3% 6' HO y \ OH Chem. Res. Toxicol., Vol. 5, No. 2, 1992 281 used for the electrospray source: electrospray needle (26 gauge), 25.0 V; cylindrical electrode, -3.1 kV; skimmer, -4.2 kV; capillary entrance, -5.1 kV. The ion source conditions were as follows: distance between needle and capillary entrance, 1.5 cm; capillary esit,107.0 V; skimmer 1, 19.0 V; skimmer 2, 11.0 V; first, second, and third focusing lens, 8.4, 18.0, and,49 V, respectively. A source pressure of 5 X 10' Torr was maintained with a sample liquid flow rate of 1 pL/min, delivered by an Orion Sage syringe pump, Mode1352 (Sage Instruments, Boston, MA). The drying gas was N2 maintained at 55 cm3/min and 170 °C. Mass spectra were acquired by scanning 2000-80 u. Cyto- chrome c and gramidicin S standards were purchased from Sigma Chemical Co. and used to tune and calibrate the mass spec- trometer. All samples were introduced into the electrospray system in a solution of 1:1 (v:v) methanol/water containing 2% acetic acid at concentrations of 20 pmol/µL. Synthesis of (S)-(-)-Cotinine N-Glucuronide (Scheme I). (A) Methy12,3,4-Tri-O-acetyl-l-bromo-l-deozy-a-D-gluco- pyranuronate (3). The title compound was prepared using a modification of the procedure reported by Bollenback et aL (25). Glucuronolactone (1) (20 g, 0.114 mol) was stirred into 1.5 L of methanol containing sodium hydroxide (60 mg) for I h at ambient temperature. The methanol was removed under reduced pressure on a rotary evaporator, keeping the bath temperature below 50 °C. The resulting syrup was treated with acetic anhydride (750 mL) in pyridine (500 mL), keeping the temperature of the mixture below 40 °C. The reaction was left to stand at 40 °C overnight, and the resulting crystals were isolated by filtration to afford 12.7 g (73% yield) of methyl tetra-O-acetyl-(3-D-glucopyranuronate (2), mp 177 °C [lit. 176.5-178 °C (25), 178 °C (26)]. A sample of 2 (2.5 g, 6.7 mmol) was dissolved in 30% HBr in acetic acid (i2 mL) at 0-5 °C and the mixture allowed to stand overnight at 4°C. The solvent was removed under reduced pressure, the residue dissolved in CHC13 (5 mL), and the organic solution extracted with saturated NaHCO3 solution (5 mL) and then water (5 mL). The resulting chloroform layer was dried (MgSO4), the solvent evaporated, and the residue crystallized from absolute ethanol to afford 3 as a cream-colored crystalline solid (2.39 g, 90%): mp 102-105 °C [lit.104-105 °C (27);-106-107-°G (25)];1H-NMR (CDC13) 5 6.77 (1 H, d, H-1), 5.62 (1 H, t, H-3), 5.22 (1 H, t, H-2), 4.85 (1 H, dd, H-4), 4.60 (1 H, d, H-5), 3.75 (3 H, s, COOCH3), 2.12 [3 H, s, C(2)-OCOCH3], 2.05 [3 H, s, C- (4)-OCOCH3], 2.02 [3 H, s, C(3)-OCOCH3]. Note: This product readily decomposes if left at ambient temperature; it is best stored at -20 °C. (B) N-(2,3,4-Tri-O-acetyl-6-methyl-,6-D-glucopyranuro- nosyl)-(S)-(-)-contininium Bromide (4). A mixture of (S)- (: )-cotinine (28) (0.176 g, 1.0 mmol) and 3 (2.38 g, 6 mmol) was heated at 60 °C under N2 for 3 days. The resulting melt was cooled to room temperature.and the viscous mass'partitioned between . water (10 mL) and diethyl ether (5 X 20 mL). The ether extracts containing unreacted 3 and its degradation products were dis- carded, and the aqueous layer was eluted through an XAD-4 column (3 X 8 cm) followed by elution with 2 bed volumes of distilled water. Lyophilization of the combined aqueous eluate at high vacuum afforded 4 as an amorphous glass that was ho- mogeneous by TLC (system a, Rf 0.21) (430 mg, 75% yield): 1H-NMR (DMSO-d6) S 9.45-9.32 (2 H, m, pyridinium H-2 and H-6), 8.84-8.80 (1 H, d, pyridinium H-4), 8.40-8.30 (1 H, t, py- ridinium H-5), 6.40 (1 H, d, H-1"), 5.80-5.70 (2 H, m, H-2" and H-5'), 5.58-5.48 (1 H, t, H-4"), 4.83-4.95 (3 H, m, H-3". H-4', and H-5'), 3.70 (3 H, s, C02CH3), 2.52 (3 H, s, NCH3), 2.23-2.43 (3 H, m, H-3'a, H-3'b, and H-4'b), 2.12-2.20 (1 H, m, H-4'a),1.8, 1.9, 2.1(9 H, 3 x s, 3 X OCOCH3); FAB-MS (positive mode) m/z 493 (M-Br)+, 365 [M+H-Br-(3 X COCH3)]', 317 (C13H170y)*, 281 [Ci3Hi,0y - (2 X H20)]+, 257 [C13H,09 - (4 x CH3)]+r 177 (C10H12N20 + H)+, 155 [C13H170y - (3 x COCH3) + CH30H + H]+, 127 (155 - CO)+, 115, 106, 98, and 93. (C) N-,6-D-Glucopyranuronosyl-(S)-(-)-cotininium Inner Salt (5). Compound 4 (200 mg, 0.35 mmol) was dissolved in 1 M NaOH (1.0 mL), and the solution was left at ambient tem- perature for 16 h. Sample pH was adjusted to 7.0 with dilute aqueous acetic acid, and the mixture was applied to a column containing strong cation-exchange resin (Dowex 50X 1, 50-100 mesh, 7 X 12 cm). The column was eluted with 4 bed volumes of distilled water, followed by 2 M aqueous ammonia solution.
Page 3: jrl04d00
282 Chern. Res. Toxicol., Vol. 5, No. 2, 1992 The first UV-absorbing basic fractions were collected and lyo- philized under high vacuum to afford 5 as a light brown amorphous solid (98 mg, 80% yield), homogeneous by TLC (solvent system b, Rt 0.38). The product was recrystallized from methanol/water to afford a white solid: 1H-NMR (D20) S 9.18-9.12 (2 H, m, pyridinium H-2 and H-6), 8.64 (1 H, d, pyridinium H-4), 8.23-8.20 (1 H, m, pyridinium H-5), 5.82 (1 H, d, H-1"), 5.18-5.12 (1 H, m, H-5'), 4.19,4.14 (1 H, m, H-5"), 3.84-3.74 (2 H, m, H-3" and H-4"), 3.70-3.66 (1 H, t, H-2"), 2.82 (3 H, s, N-CH3), 2.69-2.55 (3 H, m, H-4'a, H-3'a, and H-3'b), 2.02-1.95 (1 H, in, H-4'b); 13C-NMR (D20) S 182.07 (C-2'), 177.23 (C-6"), 149.18 (G4), 145.16 (C-3), 144.70 (G6),143.76 (C-2),131.49 (G5), 97.82 (G1"), 81.82 (G5"), 78.02 (G3"), 76.82 (C-2"), 73.99 (C-4"), 64.93 (G5'), 32.49 (G31, 31.17 (N-CH), 29.83 (C-4'). 1H-NMR and 13GNMR assignments were confirmed by DEPT, COSY, and HETCOR spectral analysis. FAB-MS (positive mode) m/z 353 (M + H)+, 335 (M + H - H20)+, 307 (M + H - H20 - CO)+, 177 (M + H - CsHs06)+, 147, 131, 115. Enzymatic Hydrolysis. A 500 µg/mL solution of synthetic CG in 50 mM sodium acetate buffer (pH 5.0) was incubated with 5000 Fishman units of fl-glucuronidase for 3 h at 37 °C. Parallel incubations were conducted in•the absence of P-glucuronidase (control) and in the presence of 10 mM n-saccharic acid 1,4- lactone, a specific inhibitor of fl-glucuronidase. After incubation, the solutions were placed on ice and 10-,c.L samples of each reaction mixture were analyzed by ion-exchange HPLC with a Gilson System 42 analytical HPLC (Gilson Medical Electronics, Inc., Middleton, WI) equipped with a Vydac 10-µm cation-exchange column (4.6 X 250 mm) (Vydac, Hesperi, CA). The solvents used for HPLC were as follows: solvent A, 25 mM ammonium formate, pH 3.8,; solvent B, acetonitrile. Elution was accomplished at a flow rate of 1.0 mL/min with the following gradient profile: 100% solvent A for 12 min; 3-min linear gradient to 55% solvent B; isocratic elution for 13 min; and return via linear gradient to initial conditions over a 3-min period. Detection was by ultraviolet absorbance at 266 nm with a Gilson Model 116 UV detector. Under these conditions, CG and cotinine eluted at 8.9 and 24.1 min, respectively. Isolation of Urinary CG. Urine was collected from-a 44 year- old male smoker (46.cigarettes/day) and stored at -20 °C prior to use. The smoker's urine was fractionated by ion-exchange HPLC using a Waters analytical system composed of Mode1510 pumps, WISP Mode1712 autosampler maintained at 4°C, and Model 680 gradient controller (Waters, Milford, MA). The chromatograph was equipped with a Vydac 10-µm cation-exchange column (4.6 X 250 mm). The solvents used for HPLC were as follows: solvent A, 50 mM ammonium formate, pH 3.4; solvent B, acetonitrile. Elution was accomplished with the following gradient profile: 100% solvent A for 13 min at a flow rate of LO mL/min; 0.1-min gradient (curve 10) to 60% solvent B at a flow rate of 1.3 mL/min; isocratic elution for 3.9 min; and return via curve 10 to initial conditions over a 1.1-min period, for a total run time of 35 min. Detection was by ultraviolet absorbance at 266 nm with a Waters Model 490E variable-wavelength detector. A fraction, collected at the same retention time as authentic CG, was collected from 40 injections (10 EcL%injection) using a Foxy Model 2130 fraction collectoi (Isco, Lincoln, NE). The fraction was taken to dryness by lyophilization. The residue was dissolved in 0.20 mL of methanol and analyzed by ES/MS. The fraction collected from ion-eachange HPLC of the smoker's urine was also hydrolyzed by fl-glucuronidase as described above. After incubation, the reaction mixtures were evaporated to dryness on a SpeedVac concentrator (Savant Instruments, Inc., Fram- ingdale, NY) (P = 0.500 Torr, ambient temperature). The residues were dissolved in 50 pL of methanol containing methyl-d3-cotinine as internal standard and analyzed for cotinine by thermospray LC/MS using a previously reported method (11, 13,14). Thermospray LC/MS Analysis of Urinary CG. Ther- mospray LC/MS analyses were performed on a Hewlett-Packard 5988A mass spectrometer equipped with a thermospray interface and a Vydac 4.6-mm x 25-cm cation-exchange column. The solvents used for HPLC were as follows: solvent A, 50 mM ammonium formate, pH 3.4; solvent B, acetonitrile. Elution was accomplished at a flow rate of 1.0 mL/min with the following gradient profile: 100% solvent A for 10 min; 0.1-min linear gradient to 70% solvent B; isocratic elution for 12 min; and return Caldwell et al. 2 \ 6 - HDO - 6' A 1 4 5 3". 4" B 2' 362 5 ~ 4 1I/ I 2"4" 3"I 5" ~ ~ 5' 6" 180 160 140 120 100 80 60 40 PPm Figure 1. 300-MHz proton NMR spectrum (A) and 75-MHz carbon NMR spectrum (B) of synthetic (S)-(-)-cotinine N-gluc- uronide. via linear gradient to initial conditions over a 0.1-min period. Additional water for the thermospray interface was added po- stcolumn at 0.5 mL/min through a tee connection at the end of the column. The thermospray was operated in the filament off, positive mode with a stem temperature of 124 °C to give a tip temperature n1.194_°-G and_a vapor temperature of.248 °C. The mass spectrometer source temperature was 250 °C. The mass spectrometer scanned 150-400 u with a cycle time of approxi- mately 2 s. Prior to injection, urine samples were passed through a 2ycm filter. A 10-µL aliquot was injected onto the column using a Rheodyne Model 7161 injector (Rheodyne Inc., Cotati, CA). . Results (S)-(-)-Cotinine N-glucuronide (5) was obtained in an overall yield of 60% from the reaction of the acetyl-pro- tected a-bromo sugar 3 with cotinine free base. The re- action was allowed to proceed for 3 days at 60 °C under N2 in the absence of the solvent and was followed by iso-. lation and deprotection of the coupled product 4 in aqueous alkali and purification of the inner salt of 5 via cation-exchange chromatography (see Scheme I). The proton and carbon NMR spectra of CG are shown in Figure 1. The proton assignments shown in Figure 1A were made with the assistance of a two-dimensional J- correlated (COSY) spectrum of synthetic CG (data not shown). The proton spectrum contains four aromatic signals in the region 9.2-8.2 ppm corresponding to the four pyridyl ring protons. All four aromatic signals are shifted downfield by 0.6-0.9 ppm relative to the pyridyl signals in the proton spectrum of cotinine measured under the same conditions (data not shown). The signal from the H5, proton of the pyrrolidine ring is shifted downfield by 0.3 ppm, and the N-methyl protons are shifted downfield by only 0.1 ppm. These data suggest the presence of a positive charge on the pyridine ring in the glucuronide conjugate. No signals corresponding to free cotinine are evident in this spectrum. The carbon NMRR spectrum of CG is shown in Figure 1B. The spectrum contains the 16 lines expected for CG. The carbon assignments were made with the assistance of
Page 4: jrl04d00
r Glucuronide Conjugate of Cotinine Table I. Nuclear Overhauser Enhancements Resulting from Selective Irradiation of HL. and H6, of CG at 25 °C in Dy0° proton irradiated chemical shift (ppm) proton affected chemical shift (ppm) % enhancement Hl.. 5.84 H2 9.15 11 H6 9.12 11 HS.. 4.13 14 H3.. 3.78 10 H6. 5.12 HZ 9.15 7 116 9.12 2 H4 8.65 7 H5 8.25 2 ° NOEs were determined using the following instrument param- eters: 45° proton pulse, acquisition time = 3 a, dpwr = 27, d1 = 1 s, d2 = 9 a, 32 transients. Enhancements are relative to control spectrum with irradiation at 7.1 ppm. DEPT and two-dimensional heteronuclear correlation (HETCOR) spectra of CG (data not shown). Relative to the analogous signals in the carbon spectrum of cotinine measured under the same conditions (data not shown), the signals of carbons 3, 4, and 5 are shifted downfield by 5.6, 10.6, and 3.9 ppm, respectively, and the signals of carbons 2 and 6 are shifted upfield by 6.5 and 6.7 ppm, respectively. The magnitude and direction of these chemical shift changes are strikingly similar to the chemical shift changes observed when the pyridyl nitrogen of nicotine is proton- ated by strong mineral acids (29). These data are con- sistent with conjugation of the pyridyl nitrogen of cotinine. To confirm the site of conjugation, a series of nuclear Overhauser effect (NOE) egperiments was performed. The saturation of one resonance in an NMR spectrum will often lead to changes in the intensities of other resonances in this spectrum. This phenomenon, called the nuclear Ov- erhauser effect, arises from the population changes brought about by dipole-dipole cross-relaxation of nuclei which are close together in space (30). When irradiation of one proton signal leads to a change in intensity of another proton signal, the two protons are close to one another. The NOE enhancements resulting from irradiation of Hi. and Hby of CG are shown in Table I. The NOEs observed between the anomeric proton Hl., and the pyridyl protons H2 and H6 are of the same magnitude as the NOEs ob- served between the pyrrolidinyl H5, and the pyridyl protons H2 and H4. These results confirm that the glucuronide moiety of CG is linked to the pyridyl nitrogen. The large NOEs observed between Hr and Hyr and Hi.r and Hs. result from 1,3 diaxial interaction and are strong evidence for the proposed #-configuration at the anomeric carbon of the glucuronide. Additional evidence for this proposal comes from the vicinal coupling constants, 3J, measured from the proton spectrum of CG and confirmed by computer simulation of the spectrum. The coupling constants are 3J(H1..,Hr), 8.71 Hz, 3J(H2-,H3. ) 9.20 Hz, 3J(H3..,H4.) 9.80 Hz, and 3J(H4..,Hr) 9.40 Hz. Vicinal protons in S-glucuronides typically have coupling constants of 7-12 Hz while vicinal.protons in a-glucuronides have coupling constants between 2 and 4 Hz (31). The vicinal coupling constants for the glucuronide protons of CG are clearly consistent with the proposed #-configuration. Final confirmation of the anomeric configuration was obtained by treatment of CG with bovine A-glucuronidase. CG was hydrolyzed by the enzyme to form free cotinine (reaction followed by ion-exchange HPLC), and the reac- tion was completely inhibited by D-saccharic acid, 1,4- lactone, a specific inhibitor of 6-glucuronidase. An ion-exchange HPLC method, developed using the authentic CG as a retention time standard, was used to isolate a fraction-eluting at the same retention time as 0 d 100 ;~ Chem.. Res. Toxicol., Vol. 5, No. 2, 1992 283 177 (M+H-C,H.O,)' 353 (M+H)' 335 (M+H•H,O)' 391 (M+K)' [_ J~ 1 375 (M+Na)' 80 120 160 200 240 280 320 360 400 mlz Figure 2. Electrospray mass spectrum of synthetic (S)-(-)-co- tinine N-glucuronide. authentic CG from multiple injections of a smoker's urine. The HPLC fraction was subjected to hydrolysis by fl-glu- curonidase under the same conditions used for enzymatic hydrolysis of authentic CG. Parallel incubations were conducted in the absence of fl-glucuronidase (control) and in the presence of D-saccharic acid 1,4-lactone. The three reaction mixtures were analyzed for the presence of coti- nine by thermospray LC/MS using a previously reported method (11,13,14). The thermospray LC/MS chroma- tograms (not shown) revealed 0-glucuronidase-specif'ic liberation of cotinine from the urinary fraction. The ES/MS of authentic CG shown in Figure 2 is an average of several spectra over a 3-min acquisition. The major peak at m/z 353 represents the protonated molec- ular ion of CG in which the carboxylate of the glucuronic acid moiety is neutralized by a proton. The peaks at m/z 375 and 391 correspond to the sodium and potassium ad- ducts (M + Na)+ and (M + K)+, respectively. The peak at m/z 335 results from loss of water from the protonated molecular ion. The ion at m/z 177 corresponds to pro- tonated cotinine resulting from cleavage of the glycosidic bond. This interpretation is supported by both HPLC and NMR data which revealed no evidence for the presence of free cotinine in the standard. The ES/MS of the HPLC fraction collected from the smoker's urine contained an ion with m/z 375, consistent with (M + Na)+ for CG. This same ion was observed in the ES/MS of authentic CG (Figure 2). The presence of this ion was not unexpected since a natriated molecular ion, (M + Na)+, was observed in the FAB spectrum of the quaternary N-glucuronide of lamotrigine isolated from human urine (20). A thermospray ion-exchange LC/MS method was de- veloped for the direct determination of CG in smokers' urine using authentic CG as a retention time standard. Thermospray analysis of authentic CG using positive ion evaporation (filament off) gave almost exclusively the protonated aglycon (m/z 177) with none of the intact molecular ion observed. This suggests that CG decomposes under these conditions to produce the free aglycon. Figure 3 shows thermospray LC/MS chromatograms of a non- smoker's urine, a nonsmoker's urine spiked with authentic CG, a smoker's urine, and a smoker's urine treated with P-glucuronidase. The chromatogram of the smoker's urine contains a peak with m/z 177 which coelutes with au- thentic CG and is not present in the nonsmoker's urine. Upon treatment of the smoker's urine with fl-gluc- uronidase, this peak disappears. These results represent the first direct determination of CG in a smoker's urine. Discussion Attempts to account for all nicotine absorbed by sum- mation of urinary nicotine and its metabolites have met
Page 5: jrl04d00
~ 284 Chem. Res. Toxicol., Vol. 5, No. 2, 1992 aD U c ~ v ~ ~ ¢ ~ n : E 8000 1 12000 4000 0 12000 8000 4000 0 12000 8000 4000 0 12000 8000 4000 0 A 6 7 8 9 10 Retention Time (min) Figure 3. Thermospray ion-exchange LC/MS chromatograms of a nonsmoker's urine (A), a nonsmoker's urine spiked with authentic CG (B), a smoker's urine (C), and a smoker's urine treated with fl-glucuronidase (D). with limited success (5,32). Kyerematen et al. (6), using HPLC with radiometric detection, were able to account for all of an intravenous dose of radiolabeled, racemic nicotine excreted in urine (ca. 80% of administered nico- tine) by summation of nicotine and eight of its metabolites (including 3'-hydrogycotinine glucuronide) in both smokers and nonsmokers. This study was instrumental in estab- lishing which metabolites could be expected in the urine of humans and that at least 80% of -the dose would be- ezpected to be excreted in the urine. However, the limi- tati_Qn_s_impnsed by th-e iiaa. n_f a-radir31-Pbv.l-vr_i nimtin-e and the need to quantify nicotine uptake in smokers and nonsmokers exposed to environmental tobacco smoke necessitate the development of methods which do not rely on radiolabeL Recently, there have been several reports of quantification of nicotine metabolites in smokers' urine using S-glucuronidase coupled with analytical determina- tion of the free metabolites (9-11, 13, 14). Using )4-gluc- uronidase-thermospray LC/MS, Byrd et al. (13,14). re- ported that the glucuronide conjugates of nicotine, coti- nine, and 3'-hydrozycotinine account for 29% of total nicotine metabolites excreted in urine. Consequently, these glucuronides represent a major class of nicotine metabolites, and convenient methods for their direct de- termination in urine are desirable. The characterization of CG and the other nicotine-dreived• glucuronide conju- gates will aid in the development of improved methods for the quantification of nicotine uptake in smokers and nonsmokers. The synthesis of CG outlined in Scheme I is based on methodology developed for the preparation of pyridine N-glucuronide (23, 24). Prior to the synthesis of CG, we prepared pyridine N-glucuronide as a model for synthetic and analytical method development. We characterized the pyridine conjugate using the same mass spectral and NMR techniques used for the characterization of CG. Our successful preparation of pyridine N-glucuronide and its rigorous characterization by FAB-MS and high-field NMR have confirmed and extended the work of previous in- vestigators. The synthesis of CG has demonstrated the utility of the method outlined in Scheme I for the prepa- ration of glucuronide conjugates of nicotine metabolites. Caldwell et al. Recently, Anderson et al. (33) reported that serum co- tinine levels measured by radioimmunoassay were 60% higher than serum cotinine levels measured by a gas chromatographic technique for samples containing >50 ng mL-i cotinine. They speculated that one possible expla- nation for the discrepancy is cross-reactivity of cotinine glucuronide with the cotinine antibodies. Authentic co- tinine glucuronide will aid in the determination of the extent of cross-reactivity with cotinine antibodies in both RIA and ELISA and enable the validation of these im- munoassays for the estimation of nicotine exposure. Using cation-exchange HPLC, we have isolated, from a smoker's urine, a fraction which coelutes with authentic CG and liberates cotinine upon fl-glucuronidase hydrolysis. Application of a thermospray ion-exchange LC/MS meth- od has enabled us to perform the first reported direct determination of CG is a smoker's urine. The data herein reported establish the identity of CG, a major urinary metabolite of nicotine, as N-P-D-glucopyranuronosyl- (S)-(-)-cotininium inner salt. References (1) Kyerematen, G. A., and Vessl, E. S. (1991) Metabolism of nico- tine. Drug Metab. Rev. 23, 3-41. (2) Neurath, G. B., Dunger, M., Orth, D., and Pein, F. G. (1987) Trans-3'-hydroxycotinine as a main metabolite in urine of smokers. Int. Arch. Occup. Environ. Health 59, 199-201. (3) Davis, R. A. (1986) The determination of nicotine and cotinine in plasma. J. Chromatogr. Sci. 24, 134-141. (4) Jarvis, J. M. (1987) Uptake of environmental tobacco smoke. In Environmental Carcinogens-Methods of Analysis and Exposure Measurement Passive Smoking (O'Neill, L K., Brunnemann, K. D., Dodet, B., and Hoffmann, D., Eds.) pp 43-58, International Agency for Research on Cancer, Lyon, France. (5) Neurath, G. B., Dunger, M., Krenz, 0., Orth, D., and Pein, F. G. (1988) Trans-3'-hydrosycotinine-a main metabolite in smokers. - _Klin. Wochenschr. 66, 2-4. (6) Kyerematen, G. A., Morgan, M. L., Chattopadhyay, B., de- Bethizy, J. D., and Vessel, E. S. (1990) Disposition of nicotine and eight-snetabolites- in smokers-and-nonsmokars: -Identification in smokers of two metabolites that are longer lived than cotinine. Clin. Pharmacol. Ther. 48, 641-651. (7) Kyerematen, G. A., Taylor, L. H., deBethizy, J. D., and Vesell, E. S. (1987) Radiometric-high-performance liquid chromato- graphic assay for nicotine and twelve of its metabolites. J. Chromatogn 419, 191-203. (8) McManus, K. T., deBethizy, J. D., Garteiz, D. A., Kyerematen, G. A., and Vesell, E. S. (1990) A new quantitative thermospray LC-MS method for,_ nicotine and its metabolites in biological fluids. J. Chromatogr. Sci. 28, 510-516. (9) Curvall, M., Vala, E. K., Englund, G., and Enzell, C. R. (1989) Urinary excretion of nicotine and its major metabolites. 43rd Tobacco Chemists' Research Conference, Richmond, VA, October 1989. (10) Haley, N. J., Leutzinger, Y., Azelrad, C. M., and Richie, J. P., Jr. (1990) Contribution of trans-3'-hydrozycotinine and gluc- uronide conjugates of nicotine metabolites to the measurement of cotinine by RIA. Symposium of XIth International Congress of Pharmacology (International Symposium on Nicotine), Hamburg, June 28-30, 1990. (11) Byrd, G. D., Chang, K. M., Greene, J. M., and deBethizy, J. D. (1990) Determination of nicotine and its metabolites in urine by thermospray LC/MS. 44th Tobacco Chemists' Research Con- ference, Winston-Salem, NC, October 1990. (12) Chang, K. M., McManus, K. T., Greene, J. M., Byrd, G. C., and deBethizy, J. D. (1991) Glucuronidation as a metabolic pathway for nicotine metabolism. Toxicologist 11, 48. (13) Byrd, G. D., Chang, K. M., Greene, J. M., and deBethizy, J. D. (1992) Determination of nicotine and its metabolites in urine by thermospray LC/MS. In The Biology of Nicotine (Lippiello, P. M., Collins, A. C., Gray, J. A., and Robinson, J. H., Eds.) pp 71-83, Raven Press, New York. (14) Byrd, G. D., Chang, K. M., Greene, J. M., deBethizy, J. D. (1992) Evidence for urinary excretion of glucuronide conjugates of nicotine, cotinine, and trans-3'-hydroxycotinine in smokers. Drug. Metab. Dispos. (in press).
Page 6: jrl04d00
Glucuronide Conjugate of Cotinine (15) Porter, C., Arison, B. H., Gruber, V. F., Titus, D. C., Vanden- heuvel, W. J. A. (1975) Human metabolism of cyproheptadine. Drug Metab. Dispos. 3, 189-197. (16) Chaudhuri, N. K., Servando, 0. A., Manniello, M. J., Luders, R. C., Chao, D. K., and Bartlett, M. F. (1976) Metabolism of tripelennamine in man. Drug Metab. Dispos. 4, 372-378. (17) Lehman, J. P., Fenselau, C., and Depaulo, J. R. (1983) Qua- ternary ammonium-linked glucuronides of amitriptyline, imipr- amine, and chlorpromazine. Drug. Metab. Dispos. 11, 221-225. (18) Chaudhary, A. K., Hubbard, J. W., McKay, G., and Midha, K. K. (1988) Identification of a quaternary ammonium-linked gluc- uronide of chlorpromazine in the urine of a schizophrenic patient treated with chlorpromazine. Drug Metab. Dispos. 16, 506-508. (19) Macrae, P. V., Kinns, M., Pullen, F. S., and Tarbit, M. H. (1990) Characterization of a quaternary, N-glucuronide metabolite of the imidazole antifungal, tioconazole. Drug Metab. Dispos. 16, 1100-1102. (20) Sinz, M. W., and Remmel, R. P. (1991) Isolation and charac- terization of a novel quaternary ammonium-linked glucuronide of lamotrigine. Drug Metab. Dispos. 19, 149-153. (21) Lehman, J. P., and Fenselau, C. (1982) Synthesis of quaternary ammonium-linked glucuronides by rabbit hepatic microsomal UDP-glucuronyltransferase and analysis by fast-atom bombard- ment mass spectrometry. Drug Metab. Dispos. 10, 446-449. (22) Dulik, D. M., and Fenselau, C. (1987) Species-dependent glu- curonidation of drugs by immobilized rabbit, rhesus monkey, and human UDP-glucuronyltransferases. Drug Metab. Dispos. 15, 473-477. (23) Aboul-Enein, H. Y. (1977) Synthesis of ft-D-glucurono- pyranosylpyridinium bromide. J. Carbohydr. Nucleosides Nu- cleotides 4, 77-81. (24) Dalgaard, L. (1983) Preparation and hydrolysis of N-gluc- Chem. Res. Toxicol., Vol. 5, No. 2, 1992 285 uronides of pyridine and piperidine. Acta Chem. Scand. B 37, 923-928. (25) Bollenback, G. N., Long, J. W., Benjamin, D. G., and Lindquist, J. A. (1955) The synthesis of aryl-D-glucopyranosiduronic acids. J. Am. Chem. Soc. 77, 3310-3315. (26) Goebel, W. F., and Barbers, F. H. (1934) Derivatives of gluc- uronic acid. IV. The synthesis of a- and S-tetraacetylglucuronic~ acid methyl ester and of 1-chlorotriacetylglucuronic acid methyl ester. J. Biol. Chem. 106, 63-69. (27) Goebel, W. F., and Barbers, F. H. (1935) Derivatives of gluc- uronic acid. IV. The preparation of a-chloro- and a-bromotri- acetylglucuronic acid methyl ester, and the synthesis of fl-gluc- uronides. J. Biol. Chem. 111, 347-353. (28) Bowman, E. R., and McKennis, H., Jr. (1963) (-)-Cotinine. Biochem. Prep. 10, 36-39. (29) Slaven, R. W. (1984) The carbon-13 and proton NMR spectra of nicotine in aqueous media J. Heterocycl. Chem. 21,1329-1332. (30) Neuhaus, D., and Williamson, M. (1989) Chapter 1- Background. In The Nuclear Ouerhauser Effect in Structural and Conformational Analysis, pp 3-22, VCH, New York. (31) Feil, V. J. (1986) Identification of xenobiotic conjugates by nuclear magnetic resonance spectrometry. In Xenobiotic Conju- gation Chemistry (Paulson, G. D., Caldwell, J., Huston, D. H., and Menn, J. J., Eds.) pp 177 202, American Chemical Society, Washington, DC. (32) Zhang, Y., Jacob, P., and Benowitz, N. L. (1990) Determination of nornicotine in smokers' urine by gas chromatography following reductive alkylation to N=propylnornicotine. J. Chromatogn 525, 349-357. (33) Anderson, L G. M., Proctor, C. J., and Husager, L. (1991) Comparison of the measurement of serum cotinine levels by gas chromatography and radioimmunoassay. Analyst 116. 691-693.

Text Control

Highlight Text:

OCR Text Alignment:

Image Control

Image Rotation:

Image Size: