Contribution of Nicotine and Nornicotine toward the Production of N′-Nitrosonornicotine in Air-Cured Tobacco (Nicotiana tabacum)

Article abstract of DOI:10.1021/acs.jnatprod.5b00678

N′-Nitrosonornicotine (6) is a potent and organ-specific carcinogen found in tobacco and tobacco smoke in substantial amounts. Nicotine (1) and nornicotine (2) are proposed to be the precursors of 6 in tobacco. Since 1 can be rapidly demethylated to 2 in tobacco, to distinguish between the direct formation of 6 from these potential precursors is difficult. A gas chromatography/thermal energy analyzer method using two columns in series was developed to separate the enantiomers of 6, N′-nitrosoanabasine (7), and N′-nitrosoanatabine (8). Tobacco lines with different combinations of three nicotine demethylases inhibited were grown in the field. Air-cured leaves were analyzed for the enantiomeric composition of four main alkaloids and their corresponding tobacco-specific nitrosamines. The percentage of (R)-6 of total 6 varied from 7% to 69% in mutant lines. The measured 6 had the same enantiomeric composition as 2, rather than 1, even when the level of 2 was reduced to 0.6% of 1 in a triple mutant line. The pattern of the enantiomeric composition of 1, 2, and 6 demonstrated that the direct formation of 6 from 1, if it occurs, is negligible in air-cured tobacco. Since (S)-6 is more highly carcinogenic than its R form, the reduction of (S)-2 should be a priority for the reduction of 6.

Full text of DOI:10.1021/acs.jnatprod.5b00678

Article
pubs.acs.org/jnp
Contribution of Nicotine and Nornicotine toward the Production of
N′‑Nitrosonornicotine in Air-Cured Tobacco (Nicotiana tabacum)
†
,‡
‡
‡
,‡
Bin Cai, Huihua Ji, Franklin F. Fannin, and Lowell P. Bush*
†
Guizhou Academy of Tobacco Science, Guiyang, Guizhou 550081, People’s Republic of China
‡
Department of Plant and Soil Sciences, University of Kentucky, Lexington, Kentucky 40546-0312, United States
*
S Supporting Information
ABSTRACT: N′-Nitrosonornicotine (6) is a potent and
organ-specific carcinogen found in tobacco and tobacco
smoke in substantial amounts. Nicotine (1) and nornicotine
(
2) are proposed to be the precursors of 6 in tobacco. Since 1
can be rapidly demethylated to 2 in tobacco, to distinguish
between the direct formation of 6 from these potential
precursors is difficult. A gas chromatography/thermal energy
analyzer method using two columns in series was developed to
separate the enantiomers of 6, N′-nitrosoanabasine (7), and
N′-nitrosoanatabine (8). Tobacco lines with different combi-
nations of three nicotine demethylases inhibited were grown in the field. Air-cured leaves were analyzed for the enantiomeric
composition of four main alkaloids and their corresponding tobacco-specific nitrosamines. The percentage of (R)-6 of total 6
varied from 7% to 69% in mutant lines. The measured 6 had the same enantiomeric composition as 2, rather than 1, even when
the level of 2 was reduced to 0.6% of 1 in a triple mutant line. The pattern of the enantiomeric composition of 1, 2, and 6
demonstrated that the direct formation of 6 from 1, if it occurs, is negligible in air-cured tobacco. Since (S)-6 is more highly
carcinogenic than its R form, the reduction of (S)-2 should be a priority for the reduction of 6.
2
,4,5
6
obacco-specific nitrosamines (TSNAs) are a class of
toxicants formed by nitrosation of tobacco alkaloids. The
genotypes,
tobacco types, different positions of the leaf on
7
T
the stalk or within a single leaf, or external factors such as the
5
,8,9
2
four main TSNAs identified in tobacco are N′-nitrosonornico-
tine (6; NNN), 4-(methylnitrosamino)-1-(3-pyridyl)-1-buta-
none (5; NNK), N′-nitrosoanabasine (7; NAB), and N′-
nitrosoanatabine (8; NAT). These are nitrosated products of
the corresponding alkaloid and mostly formed during the
use of nitrogen fertilizer,
curing conditions, storage
1
0,11
12
conditions,
and nitrite scavengers.
The presence of 6 in tobacco (Nicotiana tabacum L.,
Solanaceae) was first suggested by Druckrey and Preussmann
in 1962 and then confirmed in 1974 in tobacco smoke.
Since then, many researchers have endeavored to study the
formation mechanisms of 6 (Table S1, Supporting Informa-
tion). On the basis of feeding assays, Hecht and his colleagues
proposed that both nicotine (1) and nornicotine (2) are
13,14
1
,2
tobacco curing process (Figure 1). Compound 6 generally
occurs in greater quantities than the others and is considered
the second most potently carcinogenic of the TSNAs. The
3
compound is formed primarily during tobacco curing, and its
15
alkaloid precursors for 6. Compound 2 has been suggested as
the major precursor based on the nitrosation rate in vitro.
levels may be influenced by endogenous factors such as
16,17
The major role that 2 may play in the formation of 6 was
confirmed by suppression of nicotine demethylase activities
18
through RNA interference (RNAi) technology. Since 1 can
be rapidly demethylated to 2 in tobacco, to distinguish between
the formation of 6 directly from 1 or through 2 is difficult
technically.
Compound 6 can induce tumors of the oral cavity,
19
esophagus, and nasal mucosa when administered to rats.
This is a chiral compound, and the carcinogenic potential of the
two enantiomers of 6 is different, with the S enantiomer having
20−23
a greater carcinogenicity than the R enantiomer.
Balbo et
al. reported 14 times more oral cavity tumors in rats treated
Figure 1. Overview of the nitrosation reactions leading to the TSNA
formation in tobacco leaf. All compounds were quantified in this study.
The carbon with an asterisk (*) is the asymmetric carbon atom.
Received: August 1, 2015
©
XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.5b00678
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 1. Nicotine (1), Nornicotine (2), and N′-Nitrosonornicotine (6) Levels in Air-Cured Lamina Affected by Different
a
Nicotine Demethylase Mutations
treatment
1
2
6
mg g−
1
mg g−1
μg g−1
E4
E5
E10
2
010
E4E5E10
e4E5E10
E4e5E10
E4E5e10
e4e5E10
e4E5e10
E4e5e10
e4e5e10
+
+
+
16.2 ± 3.5
29.7 ± 7.7
11.5 ± 1.7
15.0 ± 2.1
31.7 ± 2.5
37.6 ± 3.8
14.0 ± 0.5
44.2 ± 1.2
70.5 ± 8.5
11.8 ± 3.5
50.1 ± 5.2
8.8 ± 4.6
24.80 ± 2.28
0.88 ± 0.22
20.86 ± 2.35
19.30 ± 4.00
0.90 ± 0.11
0.95 ± 0.11
18.34 ± 1.95
0.48 ± 0.04
1.86 ± 0.12
44.99 ± 8.20
1.63 ± 0.15
42.55 ± 2.68
43.10 ± 4.92
1.44 ± 0.21
1.53 ± 0.12
41.01 ± 2.73
0.40 ± 0.03
12.41 ± 3.75
1.16 ± 0.68
8.99 ± 3.42
10.18 ± 3.33
1.65 ± 0.50
1.59 ± 0.54
13.63 ± 3.91
0.75 ± 0.11
0.96 ± 0.26
8.40 ± 3.03
1.20 ± 0.38
9.60 ± 1.84
8.16 ± 1.42
0.72 ± 0.34
1.52 ± 0.37
10.32 ± 3.16
0.44 ± 0.14
−
+
+
+
−
−
−
+
+
+
−
+
−
−
+
−
−
−
+
−
−
+
−
+
b
s
2011
TN 90 LC
E4E5E10
e4E5E10
E4e5E10
E4E5e10
e4e5E10
e4E5e10
E4e5e10
e4e5e10
+
+
+
−
+
+
+
−
−
−
+
+
+
−
+
2.9 ± 0.9
−
−
+
52.1 ± 3.8
61.0 ± 8.5
9.6 ± 4.6
−
−
−
−
−
−
63.1 ± 3.9
a
Tobacco mutants with different demethylase genes silenced were grown in the field for two years. The air-cured laminas were analyzed for alkaloid
and TSNA level (see Table S2 in the Supporting Information for the rest of the alkaloid and TSNA data). All results are the average of three (2010)
b
or four (2011) replicates ± standard deviation. Each replicate consists of five middle leaves, one from each of five plants. TN 90LC is a commercial
variety and used as control only in 2011.
1
with (S)-6 when compared to the incidence in rats treated with
13.63 μg g− (double mutant line with a functional E4 gene).
2
2
(
R)-6. Since (S)-6 was found to be the predominant form of
The range of content of 6 in this study was comparable with a
previous report, in which the same parent and its derived
24
18
the compound in tobacco products, the S form accounts for
the greater degree of harmfulness.
−1
RNAi lines were used and 6 ranged from 0.19 to 12.32 μg g .
Nitrosation formation from secondary amines is generally
much more rapid than that from tertiary amines. In this study,
it was found the nitrosation reaction rate of 1 is 20- to 200-fold
lower than the 2 nitrosation (Figure S1, Supporting
Information). The nitrosation reaction occurred in the
decreasing order 2 > 4 > 3 > 1. These results are consistent
with previous aqueous results showing nitrosation of 2 is more
rapid than that of 3. Therefore, the nitrosation reaction in air-
cured tobacco is much slower for tertiary amines than for
secondary amines.
Among the three demethylases, CYP82E4 accounts for 96%
of the 2 and 90% of the 6 formed subsequently (see the
Experimental Section for the calculations). CYP82E5v2 and
CYP82E10 accounted for 2% and 5% of 2 and 6, respectively
Compounds 1, 2, and 6 are all chiral, with a stereocenter at
the 2′-position (Figure 1). Recently, it was demonstrated that
the enantiomeric composition of 2 varies in nicotine
demethylase mutant lines, while the composition of 1 remained
2
5,26
unchanged.
To determine the role of 1 in the formation of
6
, the enantiomeric composition was compared among 6, 1,
and 2 in tobacco mutant lines, but no sign of involvement of 1
in the formation of 6 in tobacco was evident.
17
RESULTS AND DISCUSSION
■
Effects of Individual Demethylases on N′-Nitro-
sonornicotine (6). Through RNAi-mediated suppression of
nicotine demethylation, 2 has been shown to be the major
1
8
precursor for 6. The relative contribution of each of the three
27
(
Table 1). These results confirmed the major role played by
nicotine demethylases in the formation of 6, as proposed by
demethylases toward the 2 contents has been evaluated. To
evaluate the effects of each nicotine demethylase on the
formation of 6, a burley line (with a high propensity for
nicotine demethylation), along with seven derived mutant lines,
was grown in the field at Rocky Mount, North Carolina (2010),
and Lexington, Kentucky (2011). The middle leaves were
sampled after being air-cured and analyzed for alkaloids and
TSNAs (Table 1 and Table S2, Supporting Information). TN
27
Lewis et al. Since the three nicotine demethylases are
associated with about 95% of the 6 in the air-cured leaf, the
question was posed as to the contribution of 1 and 2 to the
accumulation of 6 present in a triple mutant line (e4e5e10).
The answer found may facilitate the further reduction of 6.
Development of a Simple Quantification Method for
Enantiomers of TSNAs. To further explore the precursor for
the remaining 5% of 6 in the triple mutant line (e4e5e10), the
enantiomeric compositions of 1, 2, and 6 were measured. The
hypothesis was that the composition of 6 would reflect the
composition of its alkaloid precursor. Carmella and his
colleagues used this strategy before by comparing the 6 from
90LC is a commercial variety and was added to the 2011
experiment.
Alkaloid and TSNA production is sensitive to the environ-
ment. Although the general effects of the three demethylase
genes on alkaloids and TSNAs were similar between the two
years, the average concentration levels were different in the
alkaloids and TSNAs. Besides environmental effects, it is
obvious that nicotine demethylase has significant effects on the
content of 6. During the two-year experiment, the content of 6
28
in vitro nicotine demethylation with that in tobacco.
Methods for enantiomeric separation of 6 by capillary
electrophoresis and gas chromatography/thermal energy
29
−1
28
in mutant lines ranged from 0.44 μg g (triple mutant line) to
analyzer (GC/TEA) have been reported before. Since the
B
DOI: 10.1021/acs.jnatprod.5b00678
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
content of 6 has been reduced significantly in mutant lines, a
more sensitive method for TSNA enantiomer measurement
was modified and developed from a previously published
method. The major change in the method was that instead of
single chiral column, a serial combination of a chiral column
and a nonpolar column (Zebron ZB-5 in this study) was
installed in the GC, and therefore the step of collecting the
HPLC elute was avoided. All the enantiomers of 6, 7, and 8
could be baseline-separated by this method (Figure 2). Samples
cigarette tobacco was consistent with the composition of
cigarette tobacco and smokeless tobacco products, which
ranged from 34% to 43% of R form for total 6, as reported
by Stepanov et al. When compared with previous methods,
the present newly developed method involves a simpler
procedure, a better separation, and lower compound loss
during measurement.
Nicotine (1) Is Not Involved in the Formation of N′-
Nitrosonornicotine (6) in the Tobacco Leaf. After
development of the above method, all the mutant lines from
the 2011 growing season were analyzed for their alkaloid and
TSNA composition. Not only the total contents (Table 1) but
mutations in nicotine demethylase levels created a population
containing variable enantiomeric compositions of 1, 2, and 6
2
8
24
(
Table 2). Despite the large variation in the levels of 2 spanning
−1
from 0.44 to 10.32 μg g and the (R)-2 portion from 5.9% to
6.6% of the total amount of 2, in all samples the enantiomeric
6
composition of 6 was similar to that of 2. Also in the triple
mutant line, the enantiomeric composition of 6 was the same as
the enantiomeric composition of 2, suggesting that even for the
residual 6 formed in the triple mutant line, 2 is the major
contributing precursor alkaloid. On the basis of the
enantiomeric composition of 1, 2, and 6, no evidence was
found for the involvement of 1 in the formation of 6 in air-
cured tobacco leaf.
During the treatment of wastewater from a cigarette factory,
compound 6 is also formed from 2, not from 1, as in tobacco
3
0
Figure 2. (Top) Gas chromatography of standard solution (mixtures
of racemic N′-nitrosonornicotine (6), N′-nitrosoanabasine (7), and
N′-nitrosoanatabine (8)) and (bottom) a TSNA extract from
reference cigarette 2R1 tobacco. The enantiomers of three TSNAs
are eluted sequentially: (S)-6 (43.6 min), (R)-6 (44.7 min), (S)-8
leaf. Therefore, to further reduce the level of 6 in the triple
nicotine demethylase mutant line, future investigators should
continue to focus on reducing the level of 2 and nitrosation
agents (nitrate and nitrite).
Compounds 7 and 8 consist of approximately 40% and 15%
of the R form, respectively, and the composition was not
affected by the mutation of nicotine demethylase genes (Table
(
46.6 min), (R)-8 (47.7 min), (S)-7 (48.4 min), (R)-7 (49.1 min).
These two chromatographs were converted from printed chromato-
grams using the software Adobe Illustrator CS4.
2
). These two substances have the same enantiomeric
31−33
composition as 3 and 4 from previous reports,
suggesting
2
R1 and 1R4F were two reference cigarette tobaccos used as
controls to validate the improved method (Table 2). Similar
amounts of 2, 6, and 8 in 2R1 were measured as described by
Carmella et al. The composition of 6 in the reference
that there is no enantiomeric composition change in air-cured
tobacco leaf during or after the nitrosation reaction of the
respective secondary amine alkaloids. Therefore, compounds 6,
2
8
Table 2. Enantiomeric Compositions of Nicotine (1), Nornicotine (2), NNN (6), NAB (7), and NAT (8) in Air-Cured Mutant
a
Lamina from a 2011 Field Trial
treatment
b
E4
E5
E10
(R)-1%
0.5
(R)-2%
30.0
(R)-6%
29.5
(R)-8%
15.2
(R)-7%
2
1
R1
41.3
40.5
42.4
39.9
41.7
41.0
36.3
40.8
44.8
37.5
40.8
R4F
0.3
29.0
26.6
16.5
TN 90LC
parent
+s
+
+
+
0.0 ± 0.0
0.1 ± 0.1
0.0 ± 0.0
0.1 ± 0.0
0.4 ± 0.3
0.4 ± 0.0
0.1 ± 0.0
0.2 ± 0.1
3.8 ± 0.3
66.7 ± 2.2
6.0 ± 0.7
70.5 ± 3.5
6.5 ± 0.4
5.9 ± 0.2
65.2 ± 8.3
66.6 ± 3.6
8.0 ± 0.4
18.1 ± 2.5
63.5 ± 2.6
7.3 ± 1.3
69.0 ± 3.2
7.0 ± 0.2
6.0 ± 1.0
68.1 ± 0.4
66.5 ± 1.2
7.3 ± 1.1
19.3 ± 3.2
15.5 ± 0.6
17.3 ± 3.8
18.4 ± 2.7
15.6 ± 1.4
14.7 ± 1.2
17.1 ± 1.1
15.4 ± 0.4
13.5 ± 0.8
16.4 ± 0.9
+
+
e4E5E10
E4e5E10
E4E5e10
e4e5E10
e4E5e10
E4e5e10
e4e5e10
−
+
+
+
−
−
−
+
+
+
−
+
−
−
+
−
−
−
−
−
−
a
Tobacco mutants with different demethylase genes silenced were grown in the field in 2011. The air-cured laminas were analyzed for the
enantiomeric composition of alkaloids and TSNAs. Two reference cigarette tobaccos, 2R1 and 1R4F, were used as controls. TN 90LC has a
s
functional CYP82E4 gene, but the gene expression is suppressed (+ ). All results are the average of four replicates ± standard deviation. Each
replicate consists of five middle leaves, one from each of five plants. Generally, a higher percentage of the R form of NNN was found in tobacco
b
containing a functional CYP82E4 gene. Samples were combined to get a quantifiable response for (R)-7%.
C
DOI: 10.1021/acs.jnatprod.5b00678
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
, and 8 are all formed from secondary amine alkaloid
Article
7
nitrosation.
The enantiomeric composition of TSNAs in cut filler and
cigarette smoke of the reference cigarette 3R4F was analyzed.
The R% of compounds 6, 7, and 8 was 26%, 40%, and 16% in
the cut filler, and that in cigarette smoke was 27%, 44%, and
19%, respectively. Therefore, the 6 in cigarette smoke is either
the nitrosation product of nornicotine or the direct transfer
from the cut filler.
Nicotine (1) May Not Be the Direct Precursor of 4-
Methylnitrosamino)-1-(3-pyridyl)-1-butanone (5) in To-
(
bacco Leaf. Our knowledge on the mechanism of the
formation of 5 and 6 (Figure 1) comes mainly from the
34
study of the in vitro reaction between 1 and sodium nitrite.
Since it has been demonstrated in this study that 1 is unlikely
the direct precursor for 6 in tobacco, it is suspected that this
compound may not serve as a direct precursor in the formation
of 5 in vivo. Pseudooxynicotine, as an autoxidation product of
35
1
, is another possible candidate of the precursor for 5. The
dependence of 5 on pseudoxynicotine in tobacco supports the
1
role of the latter compound in the formation of 5. However,
this was not found in another study, in which pseudoxynicotine
and 5 were measured in 50 air-cured burley and 44 flue-cured
36
tobacco samples from different countries. Visualization of the
pseudoxynicotine and 5 data from the paper of Land and
Vuarnoz suggests there is no significant correlation between
pseudoxynicotine and compound 5 in tobacco (Figure S2,
Supporting Information), which could be due to variable nitrite
36
levels in the samples. Therefore, additional investigations are
needed to determine whether pseudoxynicotine serves as a
direct alkaloid precursor in tobacco.
Figure 3. Relationships between nicotine (1), anatabine (4), and
anabasine (3) and their corresponding nitrosation products associated
with nicotine demethylase CYP82E4 in air-cured burley tobacco leaf.
Lower NNK (5), NAT (8), and NAB (7) were associated with
tobacco containing functional CYP82E4 (square vs triangle). Each
data point is the average of three (2010) or four (2011) replicates.
Each replicate consists of five middle leaves, one from each of five
plants.
Relationship between N′-Nitrosonornicotine (6) and
Its Potential Precursors. Many studies concerning the
relationship between 6 and its potential precursors have been
reported (Table S1, Supporting Information). There are large
variations in these reported results, and some of them
contradict one another. The reasons for this could be due to
many uncontrolled factors in these studies that influence the
formation of compound 6: alkaloid content, nitrosating agents,
the environment, culture practices, and differences in nicotine
demethylation during curing.
Also examined was the correlation between 6 and its putative
precursors in the nicotine demethylase mutant (Figure S3,
Supporting Information). After perturbation of the demethy-
lation of 1, this alkaloid in air-cured leaves exhibits a
significantly negative correlation with 6 (r = −0.84; Figure
S3A, Supporting Information), and 2 has a significantly positive
correlation with 6 (r = 0.91; Figure S3B, Supporting
Information), as was expected. There was no significant
correlation between nitrate, nitrite, and 6 found in this study.
Competition of Four Alkaloids for the Nitrosation
Reaction. Besides the effects on compound 6, CYP82E4 also
has a great influence on the level of the other three TSNAs. In
this study, it was found that a dramatic increase of 2 (which
increased over 10-fold, Table 1) changed the relationship
between the other three alkaloids and their corresponding
nitrosation products (Figure 3). Whenever there was a
functional CYP82E4 present in the tobacco, levels of
compounds 5, 7, and 8 were not influenced by their
corresponding precursors, which is different from the situation
when the CYP82E4 function is inhibited in tobacco (Figure 3).
Lewis et al. found that the RNAi-mediated suppression of
nicotine demethylation increased the levels of 5, consistent with
18
the results in the present study. It is proposed that this could
be due to the competition among 2 and the three other main
alkaloids in tobacco leaves for nitrosating agents. However,
there were no significant differences in the nitrite and nitrate
contents in air-cured tobacco leaf (Table S3, Supporting
Information), which are potential sources of nitrosating agents.
The nitrosation competition phenomenon found in nicotine
demethylase mutant lines may not apply to commercial tobacco
varieties.
Further Changes of the Composition of N′-Nitro-
sonornicotine (6). Due to the enantioselectivity of three
demethylases, the mutant tobacco lines with different
combinations of nicotine demethylase gene suppressed were
found to contain different compositions of 2 and 6. A tobacco
line with the mutation in CYP82E4 (e4E5E10) was shown to
26
have dramatically reduced (S)-2 and (S)-6 levels, but not
26
those of (R)-2 and (R)-6 (Figure 4). Further loss of function
of the other two demethylases in e4e5e10 reduces the (R)-2 and
(R)-6 levels, but not for (S)-2 and (S)-6. Currently, most seeds
of commercial tobacco varieties, such as TN 90LC, come from
parent plants that have been screened for low nicotine
37
demethylation. These varieties contain similar concentration
levels of 2 and 6 to those found in e4E5E10. Previous
investigators have incorporated the triple mutant phenotype
D
DOI: 10.1021/acs.jnatprod.5b00678
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
experiment. After curing, the fourth leaf from the top of each stalk was
sampled and used to quantify the alkaloid contents and TSNAs and
their enantiomeric composition. All the samples were oven-dried (55
°
C) and ground to pass through a 1 mm sieve.
Alkaloid Quantification and Separation of Enantiomers of
Nicotine (1) and Nornicotine (2). The procedure of quantification
of the four tobacco main alkaloids and of separation of the
26
enantiomers of 1 and 2 has been described in a previous study.
Briefly, ground tobacco leaves were extracted by methyl tert-butyl
alcohol and aqueous sodium hydroxide. A portion of the methyl tert-
butyl alcohol extract was injected into a GC/FID instrument, and
quantification of the alkaloids was determined against alkaloid
standards and a quinoline internal standard. Contribution of nicotine
demethylase CYP82E4 to the contents of 2 and 6 is calculated as
follows: V = (C
− Cx, e4E5E10)/ Cx,E4E5E10. V represents the
x
x,E4E5E10
x
contribution of nicotine demethylase CYP82E4 to the accumulation of
compound x (2 or 6), Cx,E4E5E10 represents the concentration of
compound x (2 or 6) in the parent line E4E5E10, and Cx,e4E5E10
represents the concentration of compound x (2 or 6) in mutant line
e4E5E10.
The remainder of methyl tert-butyl alcohol extract was used for the
analysis of the enantiomeric composition of 1 and 2. The chlorophyll
in the methyl tert-butyl alcohol extract was removed through an acid
wash, and 1 and 2 in the extract were separated and purified by thin-
layer chromatography. Purified 1 was further separated into its R and S
forms by HPLC/UV on a Chiralcel OD-H column. Compound 2 was
methylated to 1 by incubating with formic acid and formaldehyde at
110 °C. The enantiomeric composition of 1 and 2 was calculated
from the peak areas of the two isomers with absolute amounts of the
isomers calculated based on the total amount of each alkaloid present
and their R/S ratio.
Figure 4. N′-Nitrosonornicotine (6) enantiomer accumulation in air-
cured mutant lamina from a 2011 field trial. Each data point is an
average of four bulk samples, and each bulk sample consists of five
middle leaves, one from each of five plants. The error bars represent
standard deviations, and +/− below the x-axis indicates the presence/
absence of a functional demethylase gene. The data were analyzed with
one-way ANOVA by Sigmaplot 12.0. Values with different letters are
significantly different (Holm−Sidak test, p < 0.05).
2
6
Measurement and Enantiomeric Separation of TSNAs.
Measurements of compounds 5−8 were performed according to
4
0
38,39
method 1 of Morgan et al. Ground tobacco leaves were extracted
with methylene chloride and aqueous sodium hydroxide. The
methylene chloride extracts were filtered, and an aliquot of the
extracts was injected into the GC/TEA for quantification of TSNAs.
The rest of the solution was used for the measurement of the
enantiomeric composition of TSNAs.
into commercial tobacco varieties,
so there should be
another lowering in the levels of 6 in tobacco in the future.
Compared with current commercial varieties, the triple mutant
lines reduce the levels of 6, but mainly (R)-6 relative to (S)-6.
20−23
Since the S form is the more carcinogenic,
future
endeavors should focus on the specific reduction of (S)-2.
On the basis of a previous report, a method was developed with
baseline separation of the three chiral TSNAs and with similar
2
8
EXPERIMENTAL SECTION
pretreatment. Briefly, the residual extracts mentioned above were
cleaned and concentrated using extraction buffer (20 mM ascorbic
acid, 54.8 mM citric acid, and 90.4 mM sodium hydrogen phosphate)
and the use of a SEP-Pak Plus silica gel solid-phase extraction
cartridge. Differing from the previous report in which further HPLC
purification was required, the eluate containing the TSNAs was
directly injected into the GC/TEA for the separation of enantiomers
■
General Experimental Procedures. Nicotine (1) and nornico-
tine (2) were purified and separated by thin-layer chromatography on
silica gel 60 F254 plates (EMD Chemicals Inc.). The alkaloid contents
were measured by gas chromatography coupled with a flame ionization
detector (PerkinElmer Autosystem XL with Prevent). The enantio-
meric composition of compound 1 was measured by high-performance
liquid chromatography equipped with ultraviolet detector (PerkinElm-
er series 200) on a Chiralcel OD-H column (Chiral Technologies
Inc.). The quantification and enantiomeric separation of TSNAs were
performed by gas chromatography (HP 5890 GC, Hewlett Packward)
coupled with a thermal energy analyzer (TEA model 543, Thermedics
Inc.).
(
Figure 2). For this purpose, a tandem column arrangement was
installed in the GC/TEA. The first was a Cyclosil-B column (30 m ×
50 μm × 0.25 μm, J & W Scientific), and the second was a Zebron
2
ZB-5 column (30 m × 530 μm × 1.5 μm, Phenomenex) connected via
a standard column connector. The carrier gas was He, and the
injection temperature 225 °C. The column oven temperature program
−1
was 60 °C for 2 min, 20 °C min to 180 °C, held for 10 min, 0.5 °C
Plant Material. TN 90LC is a widely used commercial burley
−1
−1
min to 203 °C, increased at 6 °C min to 215 °C, and then held for
10 min. The enantiomeric composition of 6−8 was calculated from the
peak areas of the two isomers with absolute amounts of the isomers
calculated based on total amount of each TSNA present and their R/S
ratio.
tobacco cultivar and has been selected from the TN90 cultivar as a low
37
nicotine converter. The burley tobacco breeding line DH98-325-6
with a very high nicotine demethylation capacity was used as parent to
develop genetic stocks possessing different combinations of knockout
2
7
mutations in the CYP82E4, CYP82E5 v2, and CYP82E10 genes. In
this report, small letters represent a homozygous knockout mutation
and capital letters represent a homozygous functional gene. For
example, e4E5E10 represents the mutant line with a homozygous
knockout mutation in the CYP82E4 gene and functional CYP82E5 v2
and CYP82E10 genes.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.5b00678.
■
*
S
The burley parent line (DH98-325-6) along with seven derived
mutant lines were grown at the Upper Coastal Plain Research Station
Figure S1 Nitrosation reactions in nicotine demethylase
mutant lines; Figure S2 Relationship between pseudoox-
ynicotine and total NNK in tobacco for 50 air-cured
(
Rocky Mount, NC) during 2010 and also at Spindletop Research
Farm (Lexington, KY) in 2011, both using standard production and
curing practices for burley tobacco. TN 90LC was added to the 2011
E
DOI: 10.1021/acs.jnatprod.5b00678
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
burley and 24 flue-cured tobacco samples; Figure S3
Correlation relationships between NNN and its putative
precursors influenced by nicotine demethylase CYP82E4
in air-cured lamina based on two-year studies; Table S1
Correlations between NNN and its putative precursors
(21) Zhao, L.; Balbo, S.; Wang, M.; Upadhyaya, P.; Khariwala, S. S.;
Villalta, P. W.; Hecht, S. S. Chem. Res. Toxicol. 2013, 26, 1526−1535.
(22) Balbo, S.; James-Yi, S.; Johnson, C. S.; O’Sullivan, M. G.;
Stepanov, I.; Wang, M.; Bandyopadhyay, D.; Kassie, F.; Carmella, S.;
Upadhyaya, P.; Hecht, S. S. Carcinogenesis 2013, 34, 2178−2183.
(
23) McIntee, E. J.; Hecht, S. S. Chem. Res. Toxicol. 2000, 13, 192−
(
nicotine, nornicotine, nitrite, and nitrate) from the
1
99.
24) Stepanov, I.; Yershova, K.; Carmella, S.; Upadhyaya, P.; Hecht,
literature; Table S2 Alkaloid and TSNA concentrations
in cured mutant lamina from 2010 and 2011 field trials;
Table S3 Nitrite and nitrate levels in cured mutant
lamina from 2011 field trial; Experiment procedure:
Measurement of nitrate and nitrite (PDF)
(
S. S. Nicotine Tob. Res. 2013, 15, 1305−1310.
(25) Cai, B.; Siminszky, B.; Chappell, J.; Dewey, R. E.; Bush, L. P. J.
Biol. Chem. 2012, 287, 42804−42811.
(26) Cai, B.; Jack, A. M.; Lewis, R. S.; Dewey, R. E.; Bush, L. P.
Phytochemistry 2013, 95, 188−196.
27) Lewis, R. S.; Bowen, S. W.; Keogh, M. R.; Dewey, R. E.
Phytochemistry 2010, 71, 1988−1998.
28) Carmella, S. G.; McIntee, E. J.; Chen, M.; Hecht, S. S.
Carcinogenesis 2000, 21, 839−843.
29) McCorquodale, E. M.; Boutrid, H.; Colyer, C. L. Anal. Chim.
(
AUTHOR INFORMATION
■
(
Corresponding Author
Tel: +1 859 2180764. Fax: +1 859 3231077. E-mail: lpbush@
email.uky.edu.
*
(
Acta 2003, 496, 177−184.
Notes
(30) Chen, B.; Qian, Y.; Wu, M.; Zhu, L.; Hu, B.; Li, X.-F. Environ.
Sci. Technol. 2015, 49, 459−466.
The authors declare no competing financial interest.
(31) Armstrong, D. W.; Wang, X. D.; Lee, J. T.; Liu, Y. S. Chirality
1
(
999, 11, 82−84.
ACKNOWLEDGMENTS
■
32) Liu, B.; Chen, C.; Wu, D.; Su, Q. J. Chromatogr. B: Anal.
This work was supported by Altria Client Services and Philip
Morris International.
Technol. Biomed. Life Sci. 2008, 865, 13−17.
(33) Perfetti, T. A.; Coleman, W. M. I. Beitr. Tabakforsch. Int. 1998,
1
8, 35−42.
34) Hecht, S. S.; Chen, C. B.; Ornaf, R. M.; Jacobs, E.; Adams, J. D.;
Hoffmann, D. J. Org. Chem. 1978, 43, 72−76.
35) Caldwell, W. S.; Greene, J. M.; Dobson, G. P.; DeBethizy, J. D.
(
REFERENCES
■
(
1) Bush, L. P.; Cui, M.; Shi, H.; Burton, H. R. Recent Adv. Tob. Sci.
001, 27, 23−46.
(
2
Ann. N. Y. Acad. Sci. 1993, 686, 213−227.
(
(
(
2) Djordjevic, M. V.; Gay, S. L.; Bush, L. P.; Chaplin, J. F. J. Agric.
Food Chem. 1989, 37, 752−756.
3) Andersen, R. A.; Fleming, P. D.; Burton, H. R.; Hamilton-Kemp,
T. R.; Sutton, T. G. J. Agric. Food Chem. 1989, 37, 44−50.
4) MacKown, C. T.; Eivazi, F.; Sims, J. L.; Bush, L. P. J. Agric. Food
Chem. 1984, 32, 1269−1272.
5) Lewis, R. S.; Parker, R. G.; Danehower, D. A.; Andres, K.; Jack, A.
M.; Whitley, D. S.; Bush, L. P. J. Agric. Food Chem. 2012, 60, 6454−
461.
36) Lang, G.; Vuarnoz, A. J. Nat. Prod. 2015, 78, 85−92.
37) Jack, A.; Bush, L. http://www.uky.edu/Ag/Tobacco/Pdf/LC-
(
Protocol.pdf, 2007.
(
(
(
38) Lewis, R. S.; Dewey, R. E. U.S. Patent 20140196730 A1, 2014.
39) Lewis, R. S.; Dewey, R. E. U.S. Patent 20140251353 A1, 2014.
40) Morgan, W. T.; Reece, J. B.; Risner, C. H.; Bennett, C. B.;
(
(
Midgett, C. H.; Johnson, K. S.; Burton, H. R. Beitr. Tabakforsch. Int.
013, 21, 192−203.
2
6
(
6) Spiegelhalder, B.; Bartsch, H. Eur. J. Cancer Prev. 1996, 5, 33−38.
(
7) Burton, H. R.; Dye, N. K.; Bush, L. P. J. Agric. Food Chem. 1992,
4
(
0, 1050−1055.
8) Hecht, S. S.; Chen, C. B.; Dong, M.; Ornaf, R. M.; Hoffmann, D.;
Tso, T. C. Beitr. Tabakforsch. Int. 1977, 9, 1−6.
9) Chamberlain, W. J.; Chortyk, O. T. Beitr. Tabakforsch. Int. 1992,
5, 87−92.
10) Andersen, R. A.; Fleming, P. D.; Burton, H. R.; Hamilton-Kemp,
T. R.; Sutton, T. G. J. Agric. Food Chem. 1991, 39, 1280−1287.
11) Shi, H.; Wang, R.; Bush, L. P.; Zhou, J.; Yang, H.-J.; Fannin, N.;
Bai, R. J. Agric. Food Chem. 2013, 61, 11588−11594.
12) Rundlof, T.; Olsson, E.; Wiernik, A.; Back, S.; Aune, M.;
Johansson, L.; Wahlberg, I. J. Agric. Food Chem. 2000, 48, 4381−4388.
13) Hoffmann, D.; Rathkamp, G.; Liu, Y. Y. In N-Nitroso Compounds
(
1
(
(
(
̈
(
in the Environment; Bogovski, P., Walker, E. A., Eds.; International
Agency for Research on Cancer: Lyon, 1974; Vol. 9, pp 159−165.
(
14) Druckrey, H.; Preussmann, R. Naturwissenschaften 1962, 49,
4
(
98−499.
15) Hecht, S. S.; Chen, C. B.; Hirota, N.; Ornaf, R. M.; Tso, T. C.;
Hoffmann, D. J. Natl. Cancer Inst. 1978, 60, 819−824.
16) Caldwell, W. S.; Greene, J. M.; Plowchalk, D. R.; DeBethizy, J.
D. Chem. Res. Toxicol. 1991, 4, 513−516.
17) Mirvish, S. S.; Sams, J.; Hecht, S. S. J. Natl. Cancer Inst. 1977, 59,
211−1213.
18) Lewis, R. S.; Jack, A. M.; Morris, J. W.; Robert, V. J. M.;
(
(
1
(
Gavilano, L. B.; Siminszky, B.; Bush, L. P.; Hayes, A. J.; Dewey, R. E.
Plant Biotechnol. J. 2008, 6, 346−354.
(
(
19) Hecht, S. S. Chem. Res. Toxicol. 1998, 11, 559−603.
20) Lao, Y.; Yu, N.; Kassie, F.; Villalta, P. W.; Hecht, S. S. Chem. Res.
Toxicol. 2007, 20, 246−256.
F
DOI: 10.1021/acs.jnatprod.5b00678
J. Nat. Prod. XXXX, XXX, XXX−XXX