Preparation of pyridine-N-glucuronides of tobacco-specific nitrosamines

Article abstract of DOI:10.1021/tx000262e

Nicotine and cotinine are metabolized to pyridine-N-glucuronides in humans. This suggests that the analogous metabolites of the carcinogenic nicotine-related nitrosamines N′-nitrosonornicotine (NNN), 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK), and 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) should also be formed in people exposed to these compounds via tobacco products. We describe the synthesis of the appropriate pyridine-N-glucuronides: pyridyl-N-β-D-glucopyranuronosyl-N′-nitrosonornicotinium inner salt (NNN-N-Gluc, 8), 4-(methylnitrosamino)-1-(3-pyridyl-N-β-D-glucopyranuronosyl)-1 -butanonium inner salt (NNK-N-Gluc, 9), and 4-(methylnitrosamino)-1-(3-pyridyl-N-β-D-glucopyranuronosyl)-1 -butanolonium inner salt (NNAL-N-Gluc, 10). The starting material, methyl 2,3,4-tri-O-acetyl-1-bromo-1-deoxy-α-D-glucopyranuronate (1), is prepared in two steps from glucuronolactone. Reactions of 1 with racemic NNN (2), NNK (3), or racemic NNAL (4) are carried out with no solvent and the crude products are deprotected by treatment with base, giving the desired N-glucuronides 8-10 in 5-7% overall yield after HPLC purification. The N-glucuronides were characterized by ¹H NMR, including COSY and NOESY spectra, and by MS and MS/MS. NNN-N-Gluc exists as a 52:48 ratio of (E)- and (Z)-rotamers, which were partially separated by HPLC. This ratio was surprisingly similar to the (E):(Z) ratio for NNN itself suggesting hydrogen bonding of the (Z)-nitroso oxygen atom to the 2″-hydroxyl group of the glucuronide moiety. Partial HPLC separations of the (E)- and (Z)-rotamers of NNK-N-Gluc and the (E)- and (Z)-rotamers as well as the (R)- and (S)-diastereomers of NNAL-N-Gluc were also achieved. The standards prepared in this study as well as the HPLC conditions developed for their separation will be important for analysis of these compounds in human urine.

Full text of DOI:10.1021/tx000262e

Chem. Res. Toxicol. 2001, 14, 555-561
555
P r ep a r a tion of P yr id in e-N-glu cu r on id es of
Toba cco-Sp ecific Nitr osa m in es
Pramod Upadhyaya, Edward J . McIntee, and Stephen S. Hecht*
University of Minnesota Cancer Center, Minneapolis, Minnesota 55455
Received December 27, 2000
Nicotine and cotinine are metabolized to pyridine-N-glucuronides in humans. This suggests
that the analogous metabolites of the carcinogenic nicotine-related nitrosamines N′-nitroso-
nornicotine (NNN), 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK), and 4-(methyl-
nitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) should also be formed in people exposed to these
compounds via tobacco products. We describe the synthesis of the appropriate pyridine-N-
glucuronides: pyridyl-N-â-D-glucopyranuronosyl-N′-nitrosonornicotinium inner salt (NNN-N-
Gluc, 8), 4-(methylnitrosamino)-1-(3-pyridyl-N-â-D-glucopyranuronosyl)-1-butanonium inner
salt (NNK-N-Gluc, 9), and 4-(methylnitrosamino)-1-(3-pyridyl-N-â-D-glucopyranuronosyl)-1-
butanolonium inner salt (NNAL-N-Gluc, 10). The starting material, methyl 2,3,4-tri-O-acetyl-
1-bromo-1-deoxy-R-D-glucopyranuronate (1), is prepared in two steps from glucuronolactone.
Reactions of 1 with racemic NNN (2), NNK (3), or racemic NNAL (4) are carried out with no
solvent and the crude products are deprotected by treatment with base, giving the desired
N-glucuronides 8-10 in 5-7% overall yield after HPLC purification. The N-glucuronides were
characterized by 1H NMR, including COSY and NOESY spectra, and by MS and MS/MS. NNN-
N-Gluc exists as a 52:48 ratio of (E)- and (Z)-rotamers, which were partially separated by
HPLC. This ratio was surprisingly similar to the (E):(Z) ratio for NNN itself suggesting
hydrogen bonding of the (Z)-nitroso oxygen atom to the 2′′-hydroxyl group of the glucuronide
moiety. Partial HPLC separations of the (E)- and (Z)-rotamers of NNK-N-Gluc and the (E)-
and (Z)-rotamers as well as the (R)- and (S)-diastereomers of NNAL-N-Gluc were also achieved.
The standards prepared in this study as well as the HPLC conditions developed for their
separation will be important for analysis of these compounds in human urine.
Among the tobacco-specific nitrosamines, 4-(methylni-
trosamino)-1-(3-pyridyl)-1-butanone (NNK) and N′-ni-
trosonornicotine (NNN) are the strongest carcinogens and
are believed to play a significant role in cancer induction
in people who use tobacco products (Figure 1) (2, 7-12).
NNK is extensively reduced in vivo to 4-(methylnitro-
samino)-1-(3-pyridyl)-1-butanol (NNAL, Figure 1), which
is also a potent carcinogen and is detected in the urine
of people exposed to NNK (9, 10, 13). O-Glucuronidation
of NNAL is a well-established pathway of NNK/NNAL
metabolism in rodents and humans (9, 14). However,
there have been no reports on pyridine-N-glucuronidation
of NNK, NNAL, or NNN. In view of the metabolic
conversion of nicotine and cotinine to pyridine-N-glucu-
ronides in humans, we hypothesize that this would also
be a metabolic pathway of tobacco-specific nitrosamines.
Therefore, in this study, we have prepared and charac-
terized the pyridine-N-glucuronides of NNK, NNAL, and
NNN and have developed methods for their separation.
These standards and methods will be important for
analysis of human urine for these metabolites.
In tr od u ction
A large number of xenobiotic compounds undergo
N-glucuronidation, catalyzed by UDP-glucuronosyltrans-
ferases (1, 2). Substrates include aryl- and alkylamines,
sulfonamides, hydroxylamines, pyridines, and other het-
erocycles of various types (1). The major tobacco alkaloid
nicotine as well as its principal metabolite cotinine are
metabolized to pyridine-N-glucuronides (3). Pyridyl-N-
â-D-glucopyranuronosyl-(S)-(-)-nicotinium inner salt (nic-
otine-N-Gluc,1 Figure 1) and pyridyl-N-â-D-glucopyranu-
ronosyl-(S)-(-)-cotininium inner salt (cotinine-N-Gluc)
account for 3-5 and 12-17%, respectively, of total
urinary nicotine metabolites in humans (4-6).
Tobacco-specific nitrosamines are an important class
of carcinogens found in tobacco products (7-9). These
nitrosamines are structurally related to nicotine and
other tobacco alkaloids from which they are derived.
* To whom correspondence should be addressed. Phone: (612) 624-
7604. Fax: (612) 626-5135. E-mail: hecht002@umn.edu.
1 Abbreviations: cotinine-N-Gluc, pyridyl-N-â-D-glucopyranuronosyl-
(S)-(-)-cotininium inner salt; GC-TEA, gas chromatography with
nitrosamine selective detection; nicotine-N-Gluc, pyridyl-N-â-D-gluco-
pyranuronosyl-(S)-(-)-nicotinium inner salt; NNAL, 4-(methylnitro-
samino)-1-(3-pyridyl)-1-butanol; NNAL-N-Gluc, 4-(methylnitrosamino)-
1-(3-pyridyl-N-â-D-glucopyranuronosyl)-1-butanolonium inner salt;
NNAL-O-Gluc, 4-(methylnitrosamino)-1-(3-pyridyl)-1-(O-â-D-glucopy-
ranuronosyl)butane; NNK, 4-(methylnitrosamino)-1-(3-pyridyl)-1-bu-
tanone; NNK-N-Gluc, 4-(methylnitrosamino)-1-(3-pyridyl-N-â-D-glu-
copyranuronosyl)-1-butanonium inner salt; NNN, N′-nitrosonornicotine;
NNN-N-Gluc, pyridyl-N-â-D-glucopyranuronosyl-N′-nitrosonornicotin-
ium inner salt.
Exp er im en ta l Section
Ap p a r a tu s. HPLC analyses were carried out with a Waters
Associates system equipped with a Model 440 UV-VIS detector
set to 254 nm. The following solvent elution programs were
used: system 1, a 300 × 7.8 mm semipreparative Phenomenex
Bondclone C18 10 µm column (Phenomenex, Torrance CA) was
eluted with 10 mM ammonium acetate (pH 6.5) for 15 min, then
by a gradient of 0 to 30% methanol in 10 mM ammonium acetate
10.1021/tx000262e CCC: $20.00 © 2001 American Chemical Society
Published on Web 04/05/2001

556 Chem. Res. Toxicol., Vol. 14, No. 5, 2001
Upadhyaya et al.
F igu r e 1. Structures of some of the compounds discussed in the text.
(pH 6.5) in 60 min at a flow rate of 3 mL/min; system 2, two
analytical C18 columns (Luna C18, 250 × 4.6 mm, 5 µm and
Phenomenex C18 Bondclone 300 × 3.9, 10 µm) with elution
conditions identical to system 1 except the flow rate was 1 mL/
min; system 3, a 250 × 4.6 mm Luna C18 5 µm column
(Phenomenex) with elution identical to system 2. LC-MS and
LC-MS/MS analyses were carried out on a Finnigan MAT LCQ
Deca instrument (Thermoquest LC/MS Division, San J ose, CA)
in the positive ion electrospray ionization mode. The MS was
interfaced with a Waters Alliance HPLC equipped with an SPD-
10A UV-Vis detector. The HPLC was operated as in system 3.
FAB MS were obtained at Michigan State University (East
Lansing, MI) with a J EOL HX-110 double focusing instrument
(J EOL USA, Peabody, MA) operated in the EI positive ion mode.
Ions were produced by bombardment with a beam of Xe atoms
(6 keV). The accelerating voltage was 10 kV and the resolution
was set at 10 000. The instrument was scanned in 20 s from
m/z 0 to 1500, saving data from 50 to 1500. NMR data were
obtained on either a 600 or 800 MHz instrument (Varian, Inc.,
Palo Alto, CA).
Ch em ica ls a n d En zym es. NNK, NNAL, and NNN were
synthesized as described elsewhere (15-17). Glucuronolactone,
acetic anhydride, pyridine, 30% HBr, acetic acid, and NaOH
were obtained from Aldrich Chemical Co. (Milwaukee, WI).
â-Glucuronidase (type-IXA from Escherichia coli) was purchased
from Sigma Chemical Co. (St. Louis, MO). Methyl 2,3,4-tri-O-
acetyl-1-bromo-1-deoxy-R-D-glucopyranuronate and the (R)- and
(S)-diastereomers of 4-(methylnitrosamino)-1-(3-pyridyl)-1-(O-
â-D-glucopyranuronosyl)butane (NNAL-O-Gluc) were prepared
as previously described (14, 18).
P yr id yl-N-â-D-glu cop yr a n u r on osyl-N′-n it r oson or n ico-
tin iu m In n er Sa lt (NNN-N-Glu c, 8). A mixture of NNN (2)
(0.18 g, 1.0 mmol) and methyl 2,3,4-tri-O-acetyl-1-bromo-1-
deoxy-R-D-glucopyranuronate (0.8 g, 3.0 mmol) was stirred and
heated at 55-60 °C under an N2 atmosphere for 3 days. The
resulting mixture was cooled to room temperature and the
viscous mass was sonicated in 10 mL of H2O and extracted with
CHCl3 (3 × 10 mL). The CHCl3 layer was discarded. The
aqueous layer was concentrated to dryness and dissolved in 1
mL of 1 M NaOH. The mixture was stirred for 16 h at room
temperature. The aqueous solution was adjusted to pH 7.0 with
acetic acid and the crude NNN-N-Gluc was purified by HPLC
using system 1. Retention times were 5-10 min. The overall
yield was 5-7%. The 5-10 min peaks were collected and
reanalyzed using HPLC system 2. Peak 1a,b, retention time 7-8
min: 1H NMR (D2O) δ 9.09 (m, 0.52H, E-pyr-2H), 8.97 (d, J )
5.49 Hz, 0.52H, E-pyr-6H), 8.87 (dd, J ) 5,5 Hz, 0.48H, Z-pyr-
6H), 8.85 (d, J ) 10.4 Hz, 0.48H, Z-pyr-2H), 8.55 (m, 0.52H,
E-pyr-4H), 8.32 (m, 0.48H, Z-pyr-4H), 8.09 (m, 0.52H, E-pyr-
5H), 8.00 (m, 0.48H, Z-pyr-5H), 5.84 (dd, J ) 8,8 Hz, 0.26H,
E-2′-H), 5.82 (dd, J ) 8,8 Hz, 0.26H, E-2′-H), 5.7 (d, J ) 8.5 Hz,
0.26H, E-1′′H), 5.69 (d, J ) 8.5 Hz, 0.26H, E-1′′-H), 5.65 (d, J )
8.5 Hz, 0.24H, Z-1′′-H), 5.64 (d, J ) 8.6 Hz, 0.24H, Z-1′′-H), 5.32
(dd, J ) 7.6,7.3 Hz, 0.48H, Z-2′-H), 4.59 (m, 0.48H, Z-5′-H), 4.44
(m, 0.48H, Z-5′-H), 3.99 (d, J ) 9.8 Hz, 0.52H, E-5′′-H), 3.97 (d,
J ) 9.2 Hz, 0.48H, Z-5′-H), 3.84 (m, 0.52H, E-5′-H), 3.70 (m,
0.52H, E-5′-H), 3.68-3.61 (m, 2H, 3′′-H, and 4′′-H), 3.55-3.48
(m, 1H, 2′′-H), 2.63-2.54 (m, 1H, 3′-H), 2.2-1.95 (m, 3H, 3′-H,
and 4′-H); ESI-MS m/z (rel intensity) 354 (100, M + 1), 178 (55);
high-resolution FAB-MS, calculated for C15H20N3O7 354.1301,
found 354.1292. Peak 2, retention time 9-10 min: ESI-MS m/z
(rel intensity) 336 (100, M + 1), 178 (20).
4-(Meth yln itr osa m in o)-1-(3-p yr id yl-N-â-D-glu cop yr a n u -
r on osyl)-1-bu ta n on iu m In n er Sa lt (NNK-N-Glu c, 9). The
same procedure was followed as for the preparation of NNN-
N-Gluc (8). Crude NNK-N-Gluc was purified by HPLC using
system 1. Retention times were 10-15 min. The overall yield
was 5-7%: 1H NMR (D2O) δ 9.43 (s, 1H, pyr-2H), 9.16 (d, J )
6.4 Hz, 1H, pyr-6H), 9.01 (d, J ) 8.1 Hz, 1H, pyr-4H), 8.21 (dd,
J ) 6.4,8.1 Hz, 1H, pyr-5H), 5.78 (d, J ) 9 Hz, 1H, 1′′-H), 4.17
(t, J ) 6.8 Hz, 1.7H, E-4H), 3.99 (d, J ) 9.4 Hz, 1H, 5′′-H), 3.7-
3.63 (m, 2.8H, 3′′-H, 4′′-H, Z-4H, and Z-CH3), 3.58 (dd, J ) 9,8.5
Hz, 1H, 2′′-H), 3.16 (dt, J ) 6,6.8 Hz, 1.7H, E-2-H), 3.06-3.04
(m, 2.8H, Z-2H, and E-CH3), 2.15 (qui, J ) 6.8 Hz, 1.7H, E-3H),
1.92 (qui, J ) 6.8 Hz, 1.7H, Z-3H); ESI-MS m/z (rel intensity)
384 (100, M + 1), 208 (10); high-resolution FAB-MS, calculated
for C16H22N3O8 384.1407, found 384.1415.
4-(Meth yln itr osa m in o)-1-(3-p yr id yl-N-â-D-glu cop yr a n u -
r on osyl)-1-bu ta n olon iu m In n er Sa lt (NNAL-N-Glu c, 10).
Using the procedures described above, crude NNAL-N-Gluc was
obtained and purified using HPLC system 1. Retention times
were 10-15 min. Overall yield was 5-7%. The individual
NNAL-N-Gluc diastereomers were separated and collected using
HPLC system 2. Individual isomers were tested for purity with
HPLC system 3. 1H NMR (D2O) (R)-NNAL-N-Gluc δ 9.09 (s,
0.83H, E-pyr-2H), 9.08 (s, 0.17H, Z-pyr-2H), 9.03 (d, J ) 6 Hz,
1H, pyr-6H), 8.67 (d, J ) 8 Hz, 1H, pyr-4H), 8.18 (dd, J ) 7,7
Hz, 1H, pyr-5H), 5.82 (d, J ) 8.6 Hz, 1H, 1′′-H), 5.11 (dd, J )
4.8,8 Hz, 0.83H, E-1H), 5.07 (dd, J ) 4.8,8 Hz, 0.17H, Z-1H),
4.25 (m, 1.7H, E-4H), 4.11 (d, J ) 10.2 Hz, 1H, 5′′-H), 3.77 (m,
2.5H, 3′′-H, 4′′-H and Z-CH3), 3.69 (m, 1.3H, H2′′ and Z-4H),
3.14 (s, 2.5H, E-CH3), 1.71-1.91 (m, 4H, 2H, and 3H); ESI-MS
m/z (rel intensity) 386 (100, M + 1), 210 (22); (S)-NNAL-N-
Gluc: δ 9.07 (s, 0.83H, E-pyr-2H), 9.06 (s, 0.17H, Z-pyr-2H),
9.04 (d, J ) 6 Hz, 1H, pyr-6H), 8.68 (d, J ) 7.8 Hz, 1H, pyr-
4H), 8.19 (dd, J ) 6.6,6.6 Hz, 1H, pyr-5H), 5.82 (d, J ) 9 Hz,
1H, 1′′-H), 5.10 (dd, J ) 4.8,8 Hz, 0.83H, E-1H), 5.06 (dd, J )
4.8,8 Hz, 0.17H, Z-1H), 4.25 (m, 1.7H, E-4H), 4.11 (d, J ) 10.2
Hz, 1H, 5′′-H), 3.77 (m, 2.5H, 3′′-H, 4′′-H, and Z-CH3), 3.68 (m,
1.3H, 2′′-H, and Z-4H), 3.14 (s, 2.5H, E-CH3), 1.70-1.92 (m, 4H,
2-H, and 3-H); ESI-MS m/z (rel intensity) 386 (100, M + 1), 210
(22); high-resolution FAB-MS (racemic mixture), calculated for
C16H24N3O8 386.1563, found 386.1565.
En zym atic an d Ch em ical Hydr olysis of N-Glu cu r on ides.
N-Glucuronide samples were heated with 1 mL of 1 N NaOH
for 1 h at 80 °C or incubated with 30 000 units of â-glucu-
ronidase for 12 h at 37 °C. For NaOH treated samples, the pH
was adjusted to 7.0 and the mixture was extracted with CHCl3
(3 × 2 mL). The extracts were concentrated and analyzed by
HPLC using system 3. For â-glucuronidase treated samples, the

Tobacco-Specific Nitrosamine N-Glucuronides
Chem. Res. Toxicol., Vol. 14, No. 5, 2001 557
Sch em e 1. Syn th esis of NNN-N-Glu c (8), NNK-N-Glu c (9), a n d NNAL-N-Glu c (10)
aqueous layer was extracted with CHCl3 (3 × 2 mL), and the
extracts were concentrated and analyzed by HPLC using system
3.
GC-TEA An a lysis of N-Glu cu r on id es. The CHCl3 extracts
from either â-glucuronidase or 1 N NaOH-treated samples were
concentrated to dryness, dissolved in 100 µL of CH3CN, and
analyzed by chiral column gas chromatography with nitro-
samine selective detection (GC-TEA) as described previously (13,
19). NNAL-containing samples were derivatized with bis-
(trimethylsilyl)trifluoroacetamide prior to GC-TEA analysis.
Resu lts a n d Discu ssion
The method of synthesis is summarized in Scheme 1.
This is based on the procedure reported previously for
synthesis of cotinine-N-Gluc (18). The starting material,
bromo ester 1, was prepared in two steps from glucurono-
lactone. Reactions of 1 with racemic NNN (2), NNK (3),
or racemic NNAL (4) were carried out with no solvent.
The crude products 5-7 were deprotected by treatment
with base, giving the desired N-glucuronides 8-10, which
were isolated by HPLC. Yields of N-glucuronides 8-10
were 5-7% from 2-4. Temperature had a significant
influence on yield in the reactions of 1 with 2-4. For
example, NNK-N-Gluc (9) was the major product ob-
served by HPLC when the reaction temperature was 50-
55 °C, whereas it was barely detectable when the
temperature was 60-65 °C.
HPLC analysis (system 1) of the reaction mixture
containing NNN-N-Gluc (8) gave the chromatogram
illustrated in Figure 2. The peaks eluting at 5-10 min
had retention times expected for NNN-N-Gluc. The peaks
eluting at 46-54 min were identified as unreacted NNN.
The 5-10 min peaks were collected and reanalyzed in
HPLC system 2, giving the chromatogram illustrated in
the inset of Figure 2. Analysis of peaks 1a and 1b by ESI-
MS gave base peaks of m/z 354, which is M + 1 of NNN-
N-Gluc. A peak at m/z 178, corresponding to replacement
of the sugar ring by a proton, was also observed (relative
intensity, 55). Analysis of peak 2 by ESI-MS gave a base
peak of m/z 336 and a peak at m/z 178 (relative intensity,
20). Peaks 1a, 1b, and 2 all yielded NNN upon hydrolysis
with 1 N NaOH, 80 °C, 1 h. These data are consistent
with the identities of peaks 1a and 1b as isomers of NNN-
N-Gluc. The structure of peak 2 was not pursued, but
apparently dehydration occurred in the sugar ring.
F igu r e 2. Chromatogram obtained upon HPLC analysis
(system 1) of the crude reaction mixture from the preparation
of NNN-N-Gluc (8). (Inset) Analysis of the peaks eluting at 5-10
min using HPLC system 2.
1H NMR spectral data for NNN-N-Gluc are sum-
marized in Table 1. All assignments were confirmed by
COSY spectra. Since the starting material was racemic
NNN, there are four possible products; two diastereomers
as a consequence of the chiral center at the 2′-carbon,
with each diastereomer consisting of two rotamers, in
which the N-nitroso group is (E) or (Z) with respect to
the pyridyl-N-glucuronide moiety. The structures of these
four isomers are illustrated in Figure 3A. Protons cor-
responding to the (E)- and (Z)-rotamers were readily
assigned by comparison to the spectrum of NNN itself
(Table 1). The (E):(Z) ratio was 52:48. Distinct resonances
were observed corresponding to the (R)- and (S)-diaster-
eomers, e.g., the 1′′ protons at 5.69 and 5.70 of (E)-NNN-
N-Gluc and 5.64 and 5.65 of (Z)-NNN-N-Gluc. However,
these resonances could not be assigned specifically to the
(R)- or (S)-diastereomer. Each pyridyl proton was shifted
downfield compared to the corresponding chemical shifts
in the spectrum of NNN itself. These data are consistent
with the presence of a full positive charge on the pyridyl

558 Chem. Res. Toxicol., Vol. 14, No. 5, 2001
Ta ble 1. 1H NMR Da ta for NNN a n d NNN-N-Glu ca
Upadhyaya et al.
Pyridyl Protons
4
2
5
6
(E)-NNN
8.3(m)
7.59(dt)
7.29(dd)
8.3(m)
J ) 12.8, 2.4 Hz
7.40(dt)
J ) 12.8, 8 Hz
7.22(dd)
(Z)-NNN
8.14(d)
8.23(dd)
J ) 3.2 Hz
J ) 12.8, 3.2 Hz
8.55(m)
J ) 12.8, 8 Hz
8.09(m)
J ) 8.8, 2.4 Hz
(E)-NNN-N-Gluc
(Z)-NNN-N-Gluc
9.09(m)
8.97(d)
J ) 5.49 Hz
8.87(dd)
8.85(d)
8.32(m)
8.00(m)
J ) 5, 5 Hz
Pyrrolidine Protons
2′
3′
4′
5′
(E)-NNN
5.47(dd)
J ) 11.2, 12.0 Hz
5.10(dd)
J ) 11.2, 12.0 Hz
5.84(dd)a
J ) 8, 8 Hz;
5.82(dd)
2.4(m)
1.76-2.06(m)
2.4(m)
1.76-2.06(m)
2.6(m)
1.95-2.2(m)
1.76-2.06(m)
3.75(m)
3.62(m)
4.45(m)
4.34(m)
3.84(m)
3.70(m)
(Z)-NNN
1.76-2.06(m)
1.95-2.2(m)
(E)-NNN-N-Gluc
J ) 8, 8 Hz
5.32(m)
J ) 7.6, 7.3 Hz
(Z)-NNN-N-Gluc
2.6(m)
1.95-2.2(m)
1.95-2.2(m)
4.59(m)
4.44(m)
Glucuronide Protons
1′′
2′′
3′′
4′′
5′′
(E)-NNN-N-Gluc
(Z)-NNN-N-Gluc
5.69(d)b
J ) 9.2 Hz;
5.70(d)
J ) 8.5 Hz
5.64(d)b
3.5(m)
3.6(m)
3.6(m)
3.99(d)
J ) 9.8 Hz
3.5(m)
3.6(m)
3.6(m)
3.97(d)
J ) 8.5 Hz;
5.65(d)
J ) 9.2 Hz
J ) 8.6 Hz
a
b
Spectra were obtained in D2O. Values are chemical shifts (δ) and coupling constants (J ). Separate resonances observed for (R)- and
(S)-diastereomers.
nitrogen. A NOESY spectrum was also consistent with
the presence of the N-glucuronide moiety, clearly showing
NOEs between the 1′′ proton of the sugar ring and the
two and six protons of the pyridine ring. The â configu-
ration of the N-glucuronide moiety was confirmed by a
NOESY spectrum, which demonstrated connectivity
among the 1′′, 3′′, and 5′′-protons of the sugar ring.
Moreover, the coupling constants of 8-9 Hz for H 1′′-2′′
and H 4′′-5′′ were consistent with the â configuration of
the N-glucuronide ring (20).
bonding of the N-nitroso oxygen atom to the 2′′ hydroxyl
group of the glucuronide ring, which could increase the
proportion of the (Z)-isomer.
HPLC analysis of the reaction mixture containing
NNK-N-Gluc (9) produced the chromatogram shown in
Figure 4. The material eluting at 10-13 min had a base
peak at m/z 384, which is M + 1 of NNK-N-Gluc. The
1
peak eluting at 43 min was unreacted NNK. H NMR
data are summarized in Table 2. The (E)- and (Z)-
rotamers of NNK-N-Gluc were observed, and the (E):(Z)
ratio was 84:16. As in the case of NNN-N-Gluc, the
pyridyl protons were shifted downfield relative to their
positions in NNK itself. The attachment of the sugar ring
to the pyridyl nitrogen was also confirmed by a NOESY
spectrum, as was the â configuration of the sugar ring.
NNK-N-Gluc was readily converted to NNK by treatment
with 1 N NaOH, 80 °C, for 1 h or by incubation with
â-glucuronidase for 12 h.
Treatment of peaks 1a and 1b with â-glucuronidase
resulted in complete conversion to NNN, consistent with
the assigned structure. Each of peaks 1a and 1b was
collected from HPLC and allowed to stand at 40 °C. This
resulted in reconversion to a mixture of peaks 1a and
1b, with a half-life for the reconversion of approximately
10 min. These data, together with the NMR data sum-
marized above, demonstrate that peaks 1a and 1b are
the E- and Z-isomers of NNN-N-Gluc, respectively. As
shown in Figure 2 (inset), peak 1a, assigned as the
E-isomer, is slightly larger than peak 1b, consistent with
HPLC analysis of the reaction mixture for preparation
of NNAL-N-Gluc (10) gave the chromatogram illustrated
in Figure 5. The material eluting at 9-11 min had a base
peak at m/z 386, which is M + 1 of NNAL-N-Gluc. MS/
MS analysis of this peak gave a base peak of m/z 210,
corresponding to loss of the sugar moiety. Unreacted
NNAL eluted at 43 min. Similar to NNN-N-Gluc, there
1
the 52:48 ratio observed in the H NMR spectrum. The
(R)- and (S)-diastereomers of NNN-N-Gluc were not
separated under these conditions.
The 52:48 (E):(Z) ratio observed for NNN-N-Gluc was
unexpected. The (E):(Z) ratio of NNN itself at pH 7.0 is
68:32. Addition of the N-glucuronide moiety to the
pyridine ring would be expected to increase steric bulk
in this portion of the molecule compared to NNN itself,
thereby increasing the (E):(Z) ratio. Instead, a decrease
was observed. We propose that this is due to hydrogen
1
are four isomers of NNAL-N-Gluc (Figure 3B). H NMR
data are analogous to those described above for NNN-
N-Gluc and NNK-N-Gluc. NNAL-N-Gluc was completely
converted to NNAL by treatment with 1 N NaOH at
80 °C for 1 h, or â-glucuronidase for 12 h. NNAL-O-Gluc
was not observed in these reactions.

Tobacco-Specific Nitrosamine N-Glucuronides
Chem. Res. Toxicol., Vol. 14, No. 5, 2001 559
F igu r e 3. Structures of isomers of (A) NNN-N-Gluc and (B) NNAL-N-Gluc.
F igu r e 4. Chromatogram obtained upon HPLC analysis
(system 1) of the crude reaction mixture from the preparation
of NNK-N-Gluc (9).
When NNAL-N-Gluc was collected (from system 1,
Figure 5) and reanalyzed using HPLC system 2, the
chromatogram obtained in the inset in Figure 5 was
F igu r e 5. Chromatogram obtained upon HPLC analysis
(system 1) of the crude reaction mixture from the preparation
of NNAL-N-Gluc (10). (Inset) Analysis of the peaks eluting at
8-12 min using HPLC system 2.
obtained. Each of the two large peaks was collected for
1
MS and H NMR analysis. The MS are essentially the
same as that described above. The 1H NMR data are
summarized in Table 2. These data clearly demonstrate
that these two peaks are diastereomers. Each peak was
treated with â-glucuronidase. The released NNAL was
silylated and analyzed by chiral column GC-TEA (19).
The first HPLC peak gave mainly the second eluting GC-
TEA peak (retention time 74.8 min) and was therefore
derived from (R)-NNAL while the second HPLC peak
gave mainly the first eluting GC-TEA peak (retention
time 74.0 min) and was therefore derived from (S)-NNAL.
These data confirm the assignments shown in the inset
in Figure 5.
HPLC separation of the glucuronides of NNN, NNK,
and NNAL is illustrated in Figure 6. The identity of each
peak was confirmed by LC-MS. In this system, NNN-N-
Gluc eluted first, with partial separation of the (E)- and

560 Chem. Res. Toxicol., Vol. 14, No. 5, 2001
Ta ble 2. 1H NMR Da ta for NNK, NNAL, NNK-N-Glu c, a n d NNAL-N-Glu c
Upadhyaya et al.
Pyridyl Protons
4
2
5
6
NNKa
9.14(s)(E)-
9.13(s)(Z)-
9.43(s)
8.20(d)
J ) 8 Hz
9.01(d)
J ) 8.1 Hz
7.67(d)
7.42(dd)
J ) 7,7 Hz
8.21(dd)
J ) 6.4,8.1 Hz
7.26(m)
8.78(m)
NNK-N-Gluc
NNALa
9.16(d)
J ) 6.4 Hz
8.42(d)
8.46(s)
J ) 8 Hz
J ) 4 Hz
9.03(d)
J ) 6 Hz
9.04(d)
(R)-NNAL-N-Gluc
(S)-NNAL-N-Gluc
9.09(s)(E)-
9.08(s)(Z)-
9.07(s)(E)-
9.06(s)(Z)-
8.67(d)
8.18(dd)
J ) 7, 7 Hz
8.19(dd)
J ) 8 Hz
8.68(d)
J ) 7.8 Hz
J ) 6.6, 6.6 Hz
J ) 6 Hz
Butyl Protons
1
2
3
4
CH3
NNKa
3.06(t)(E)-
J ) 7.2 Hz;
2.94(t)(Z)-
J ) 7.2 Hz
3.16(dt)(E)-
J ) 6,6.8 Hz;
3.05(m)(Z)-
2.23(qui)(E)-
J ) 7.2 Hz;
1.98(qui)(Z)-
J ) 7.2 Hz
2.15(qui)(E)-
J ) 6.8 Hz;
1.92(qui)(Z)-
J ) 6.8 Hz
1.9(m),1.8(m)(E)-
1.6-1.7(m)(Z)-
1.71-1.91(m)
4.26(t)(E)-
J ) 6.8 Hz;
3.71(t)(Z)-
J ) 7.2 Hz
4.17(t)(E)-
J ) 6.8 Hz;
3.66(m)(Z)-
3.08(s)(E)-
3.79(s)(Z)-
NNK-N-Gluc
3.05(s)(E)-
3.68(s)(Z)-
NNALa
4.76(m)(E)-
4.71(m)(Z)-
1.8(m),1.7(m)(E)-
1.6-1.7(m)(Z)-
1.71-1.91(m)
4.15(m)(E)-
3.58(m),3.59(m)(Z)-
4.25(m)(E)-
3.00(s)(E)-
3.71(s)(Z)-
3.14(s)(E)-
3.77(s)(Z)-
(R)-NNAL-N-Gluc
5.11(dd)(E)-
J ) 4.8, 8 Hz;
5.07(dd)(Z)-
J ) 4.8, 8 Hz
5.10(dd)(E)-
J ) 4.8, 8 Hz;
5.06(dd)(Z)-
J ) 4.8, 8 Hz
3.70(m)(Z)-
(S)-NNAL-N-Gluc
1.70-1.92(m)
1.70-1.92(m)
4.25(m)(E)-
3.69(m)(Z)-
3.14(s)(E)-
3.77(s)(Z)-
Glucuronide Protonsa
1′′
2′′
3′′
4′′
5′′
NNK-N-Gluc
5.78(d)
3.58(dd)
3.66(m)
3.66(m)
3.99(d)
J ) 9 Hz
J ) 9, 8.5 Hz
J ) 9.4 Hz
(R)-NNAL-N-Gluc
(S)-NNAL-N-Gluc
5.82(d)
3.68(m)
3.77(m)
3.79(m)
3.77(m)
3.79(m)
4.11(d)
J ) 8.6 Hz
5.82(d)
J ) 10.2 Hz
4.11(d)
3.68(m)
J ) 9 Hz
J ) 10.2 Hz
a
NNK and NNAL spectra were obtained in CDCl3; glucuronides were run in D2O. Values are chemical shifts (δ) and coupling constants
(J ).
(Z)-rotamers, but no separation of the (R)- and (S)-
diastereomers. NNAL-N-Gluc eluted next, with partial
resolution of its four isomers. Only [R-(Z)]NNAL-N-Gluc
and [S-(E)]NNAL-N-Gluc coeluted. The compound eluting
next, NNK-N-Gluc, was clearly separated into its (E)- and
(Z)-rotamers. The compound eluting last was NNAL-O-
Gluc, which was resolved into its four isomers. We
anticipate that this HPLC system will be useful for LC-
MS analysis of these glucuronides, as formed in microso-
mal preparations, or in urine after appropriate prepuri-
fication steps.
It is likely that pyridine-N-glucuronides will be me-
tabolites of tobacco-specific nitrosamines in humans. Both
nicotine and cotinine are excreted in part as nicotine-N-
Gluc and cotinine-N-Gluc in human urine (3-6). A major
metabolite of cotinine, trans-3′-hydroxycotinine, is ex-
creted in part as an O-glucuronide in human urine (3-
6). Nicotine and cotinine are structurally related to NNN
and NNK while trans-3′-hydroxycotinine has polarity
similar to that of NNAL. NNAL-O-Gluc diastereomers
were identified in rodent and monkey urine by NMR and
MS and in human urine by collection of HPLC fractions
of the appropriate retention times followed by conversion
to NNAL using â-glucuronidase (21-24). These results,
however, do not exclude the presence of NNAL-N-Gluc
F igu r e 6. Separation of glucuronides of NNN, NNK, and
NNAL using HPLC system 3.

Tobacco-Specific Nitrosamine N-Glucuronides
Chem. Res. Toxicol., Vol. 14, No. 5, 2001 561
(13) Upadhyaya, P., Carmella, S. G., Guengerich, F. P., and Hecht, S.
S. (2000) Formation and metabolism of 4-(methylnitrosamino)-
1-(3-pyridyl)-1-butanol enantiomers in vitro in mouse, rat and
human tissues. Carcinogenesis 21, 1233-1238.
in human urine. We are currently using the standards
prepared here to analyze human urine for NNN-N-Gluc,
NNK-N-Gluc, and NNAL-N-Gluc.
(14) Upadhyaya, P., Kenney, P. M. J ., Hochalter, J . B., Wang, M., and
Hecht, S. S. (1999) Tumorigenicity and metabolism of 4-(meth-
ylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) enantiomers and
metabolites in the A/J mouse. Carcinogenesis 20, 1577-1582.
(15) Hecht, S. S., Lin, D., and Castonguay, A. (1983) Effects of
R-deuterium substitution on the mutagenicity of 4-(methylnitro-
samino)-1-(3-pyridyl)-1-butanone (NNK). Carcinogenesis 4, 305-
310.
(16) Hecht, S. S., Young, R., and Chen, C. B. (1980) Metabolism in
the F344 rat of 4-(N-methyl-N-nitrosamino)-1-(3-pyridyl)-1-bu-
tanone, a tobacco specific carcinogen. Cancer Res. 40, 4144-4150.
(17) Hu, M. W., Bondinell, W. E., and Hoffmann, D. (1974) Synthesis
of carbon 14-labeled myosmine, nornicotine, and N′-nitrosonor-
nicotine. J . Labelled Cmpds. 10, 79-88.
Ack n ow led gm en t. This study was supported by
grant CA-81301 from the National Cancer Institute.
Stephen S. Hecht is an American Cancer Society Re-
search Professor, supported by ACS Grant RP-00-138. We
thank Peter W. Villalta, Analytical Chemistry and Biom-
arkers Facility, for MS analyses. The facility is supported
in part by NCI grant CA-77598.
Refer en ces
(1) Chiu, S. H. L., and Huskey, S. R. W. (1998) 1996 ASPET
N-Glucuronidation of Xenobiotics Symposium: Species differences
in N-glucuronidation. Drug Metab. Dispos. 26, 838-847.
(2) Tukey, R., and Strassburg, C. P. (2000) Human UDP-glucurono-
syltransferasses: metabolism, expression, and disease. Annu. Rev.
Pharmacol. Toxicol. 40, 581-616.
(18) Caldwell, W. S., Greene, J . M., Byrd, G. D., Chang, K. M., Uhrig,
M. S., deBethizy, J . D., Crooks, P. A., Bhatti, B. S., and Riggs, R.
M. (1992) Characterization of the glucuronide conjugate of
cotinine: a previously unidentified major metabolite of nicotine
in smokers’ urine. Chem. Res. Toxicol. 5, 280-285.
(3) Gorrod, J . W., and Schepers, G. (1999) Biotransformation of
nicotine in mammalian systems. In Analytical Determination of
Nicotine and Related Compounds and their Metabolites (Gorrod,
J . W., and J acob, P., III, Eds.) pp 45-67, Elsevier, Amsterdam.
(4) Byrd, G. D., Chang, K. M., Greene, J . M., and deBethizy, J . D.
(1992) Evidence for urinary excretion of glucuronide conjugates
of nicotine, cotinine, and trans-3′-hydroxycotinine in smokers.
Drug Metab. Dispos. 20, 192-197.
(19) Carmella, S. G., Ye, M., Upadhyaya, P., and Hecht, S. S. (1999)
Stereochemistry of metabolites of a tobacco-specific lung carcino-
gen in smokers' urine. Cancer Res. 59, 3602-3605.
(20) Feil, V. J . (1986) Identification of xenobiotic conjugates by nuclear
magnetic resonance spectrometry. In Xenobiotic Conjugation
Chemistry (Paulson, G. D., Caldwell, J ., Huston, D. H., and Menn,
J . J ., Eds.) pp 177-202, American Chemical Society, Washington,
DC.
(5) Benowitz, N. L., J acob, P., Fong, I., and Gupta, S. (1994) Nicotine
metabolic profile in man: comparison of cigarette smoking and
transdermal nicotine. J . Pharmacol. Exp. Ther. 268, 296-303.
(6) Hecht, S. S., Carmella, S. G., and Murphy, S. E. (1999) Effects of
watercress consumption on urinary metabolites of nicotine in
smokers. Cancer Epidemiol., Biomarkers Prev. 8, 907-913.
(7) Hoffmann, D., and Hecht, S. S. (1985) Nicotine-derived N-
nitrosamines and tobacco related cancer: current status and
future directions. Cancer Res. 45, 935-944.
(8) Hecht, S. S., and Hoffmann, D. (1988) Tobacco-specific nitro-
samines, an important group of carcinogens in tobacco and tobacco
smoke. Carcinogenesis 9, 875-884.
(9) Hecht, S. S. (1998) Biochemistry, biology, and carcinogenicity of
tobacco-specific N-nitrosamines. Chem. Res. Toxicol. 11, 559-603.
(10) Hecht, S. S. (1999) Tobacco smoke carcinogens and lung cancer.
J . Natl. Cancer Ins. 91, 1194-1210.
(21) Morse, M. A., Eklind, K. I., Toussaint, M., Amin, S. G., and Chung,
F. L. (1990) Characterization of a glucuronide metabolite of
4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) and its
dose-dependent excretion in the urine of mice and rats. Carcino-
genesis 11, 1819-1823.
(22) Hecht, S. S., Trushin, N., Reid-Quinn, C. A., Burak, E. S., J ones,
A. B., Southers, J . L., Gombar, C. T., Carmella, S. G., Anderson,
L. M., and Rice, J . M. (1993) Metabolism of the tobacco-specific
nitrosamine 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone in
the Patas monkey: pharmacokinetics and characterization of
glucuronide metabolites. Carcinogenesis 14, 229-236.
(23) Carmella, S. G., Akerkar, S., and Hecht, S. S. (1993) Metabolites
of the tobacco-specific nitrosamine 4-(methylnitrosamino)-1-(3-
pyridyl)-1-butanone in smokers’ urine. Cancer Res. 53, 721-724.
(24) Murphy, S. E., Carmella, S. G., Idris, A. M., and Hoffmann, D.
(1994) Uptake and metabolism of carcinogenic levels of tobacco-
specific nitrosamines by Sudanese snuff dippers. Cancer Epide-
miol., Biomarkers Prev. 3, 423-428.
(11) Spiegelhalder, B., and Bartsch, H. (1996) Tobacco-specific nitro-
samines. Eur. J . Cancer Prev. 5, 33-38.
(12) Magee, P. N. (1996) Nitrosamines and human cancer: introduc-
tion and overview. Eur. J . Cancer Prev. 5, 7-10.
TX000262E