Nicotine metabolism and urinary elimination in mouse: In vitro and in vivo

Article abstract of DOI:10.1080/00498250701708539

This study aimed at elucidating the in vivo metabolism of nicotine both with and without inhibitors of nicotine metabolism. Second, the role of mouse CYP2A5 in nicotine oxidation in vitro was studied as such information is needed to assess whether the mouse is a suitable model for studying chemical inhibitors of the human CYP2A6. The oxidation of nicotine to cotinine was measured and the ability of various inhibitors to modify this reaction was determined. Nicotine and various inhibitors were co-administered to CD2F1 mice, and nicotine and urinary levels of nicotine and four metabolites were determined. In mouse liver microsomes anti-CYP2A5 antibody and known chemical inhibitors of the CYP2A5 enzyme blocked cotinine formation by 85-100%, depending on the pre-treatment of the mice. The amount of trans-3-hydroxycotine was five times higher than cotinine N-oxide, and ten times higher than nicotine N-1-oxide and cotinine. Methoxsalen, an irreversible inhibitor of CYP2A5, significantly reduced the metabolic elimination of nicotine in vivo, but the reversible inhibitors had no effect. It is concluded that the metabolism of nicotine in mouse is very similar to that in man and, therefore, that the mouse is a suitable model for testing novel chemical inhibitors of human CYP2A6.

Full text of DOI:10.1080/00498250701708539

Xenobiotica, Vol. 38, No. 1, January 2008, 34–47
Nicotine metabolism and urinary elimination in mouse:
in vitro and in vivo
H. RAUNIO1, N. POKELA1, K. PUHAKAINEN1, M. RAHNASTO1,
T. MAURIALA2, S. AURIOLA2, & R. O. JUVONEN1
2
1Department of Pharmacology and Toxicology and Department of Medicinal Chemistry,
University of Kuopio, Kuopio, Finland
(Received 20 September 2007; accepted 26 September 2007)
Abstract
This study aimed at elucidating the in vivo metabolism of nicotine both with and without inhibitors of
nicotine metabolism. Second, the role of mouse CYP2A5 in nicotine oxidation in vitro was studied as
such information is needed to assess whether the mouse is a suitable model for studying chemical
inhibitors of the human CYP2A6. The oxidation of nicotine to cotinine was measured and the ability
of various inhibitors to modify this reaction was determined. Nicotine and various inhibitors
were co-administered to CD2F1 mice, and nicotine and urinary levels of nicotine and four metabolites
were determined. In mouse liver microsomes anti-CYP2A5 antibody and known chemical inhibitors
of the CYP2A5 enzyme blocked cotinine formation by 85–100%, depending on the pre-treatment
of the mice. The amount of trans-3-hydroxycotine was five times higher than cotinine N-oxide, and
ten times higher than nicotine N-1-oxide and cotinine. Methoxsalen, an irreversible inhibitor of
CYP2A5, significantly reduced the metabolic elimination of nicotine in vivo, but the reversible
inhibitors had no effect. It is concluded that the metabolism of nicotine in mouse is very similar to that
in man and, therefore, that the mouse is a suitable model for testing novel chemical inhibitors of
human CYP2A6.
Keywords: Nicotine oxidation, CYP2A5, human/mouse correlation
Introduction
Tobacco is one of the worst threats to human health. It damages the lungs, blood
circulation, and heart, and causes cancer (Ezzati et al. 2003). Nicotine is the key
compound promoting the use of tobacco due to its highly addictive properties (George and
O’Malley 2004) and several lines of evidence indicate that smokers titrate their cigarette
consumption to maintain steady levels of nicotine in the brain (Benowitz 1996).
Correspondence: R. O. Juvonen, Department of Pharmacology and Toxicology, University of Kuopio, POB 1627, FIN-70211,
Kuopio, Finland. Tel: 358-17-162010. Fax: 358-17-162424. E-mail: risto.juvonen@uku.fi
ISSN 0049-8254 print/ISSN 1366-5928 online ß 2008 Informa UK Ltd.
DOI: 10.1080/00498250701708539

Nicotine metabolism and elimination
35
Thus, factors that influence nicotine plasma levels, by either intake or elimination, would be
anticipated to affect smoking behaviour (Malaiyandi et al. 2005).
In humans, the main elimination route of nicotine is oxidation to cotinine which is
catalysed by hepatic CYP2A6 and aldehyde oxidase enzymes (Brandange and Lindblom
1979; Nakajima et al. 1996), reviewed by Hukkanen et al. (2005) (Figure 1), although less is
known about the enzymes that catalyse nicotine metabolism in other species. Guinea pig and
hamster hepatocytes show the highest total metabolism of nicotine, followed by mouse, rat,
and human hepatocytes (Hukkanen et al. 2005). A recent study by Murphy et al. (2005)
revealed that mouse CYP2A5 and rat CYP2A3 enzymes are efficient catalysts of cotinine
formation in vitro. CYP2A5 is an abundant and active enzyme in mouse liver, whereas
CYP2A3 is present in minor quantities in extrahepatic tissues in rat (Honkakoski and
Negishi 1997; Su and Ding 2004). The mouse has a much higher CYP2A activity in vivo
than the rat and thus the mouse appears to be the preferred in vivo model species for nicotine
metabolism studies.
CYP2A6 genetic variations, causing reduced or absent enzyme activity, are associated
with a reduced risk for smoking and population studies have indicated that individuals with
H
N
Other nicotine primary metabolites
N
CYP2A6
H
H
N+
N+
-O
N
N
Nicotine iminium ion
Aldehyde
oxidase
Nicotine 1-N-oxide
O
H
N
Other cotinine metabolites
N
Cotinine
OH
H
O
O
H
H
N
N
N+
N
O
Trans-3-hydroxy
cotinine
Cotinine N-oxide
Figure 1. Major metabolic pathways of nicotine in humans. CYP2A6 is the major catalysing enzyme
of the oxidation of nicotine to nicotine iminium ion, which is further oxidized to cotinine by
aldehyde oxidase.

36
H. Raunio et al.
an inactive CYP2A6 enzyme are less prone to develop cigarette smoking-associated cancers
(Oscarson 2001; Raunio et al. 2001; Malaiyandi et al. 2005). This is especially prominent
in the Japanese population, which has a high frequency of inactive CYP2A6 alleles
(Kamataki et al. 2005). It has been suggested by Sellers and Tyndale (2000) that inhibition
of the CYP2A6 enzyme would slow nicotine inactivation, prolong the duration that nicotine
remains in the brain, and lead to a reduction in the number of cigarettes smoked. Such
inhibition could be achieved by chemical agents that interfere with the catalytic activity of the
CYP2A6 enzyme. Pilot studies with CYP2A6 inhibitors suggest that a reduction in cigarette
consumption can be achieved (Sellers et al. 2003).
The goals of the current study were (1) to elucidate the metabolic pathways of nicotine in
mouse in vivo, (2) to study the role of the CYP2A5 enzyme in nicotine-to-cotinine formation
in mouse liver microsomes in vitro, and (3) to test whether known potent inhibitors of
CYP2A5 could reduce the metabolic elimination of nicotine in vivo. This kind of
information is crucial when deciding which animal model to use in testing novel CYP2A6
inhibitors for potential use in man.
Materials and methods
Reagents
(10S,20S)-nicotine-10oxide (>98% purity), (S)-cotinine N-oxide (98% purity), (þ)-cotinine
(98% purity), [(4S)-(–)-1 methyl-5-(pyridin-3-yl)]-2-pyrroliline
(98% purity),
R-(þ)-cotinine, (3S,5S)-30-hydroxycotinine (98% purity), and nicotine (96% purity) were
from TRC (Toronto Research Chemicals, Toronto, Canada). (–)-Cotinine (98% purity)
and (–)-nicotine (99% purity), ꢀ-glucuronidase, coumarin, methoxsalen, ꢁ-nonanoic
lactone, 2,7-dimethylnaphthalene, 4-methylnaphthalene, 2-bromonaphthalene, and
4-methoxybenzaldehyde were obtained from Sigma/Aldrich (St Louis, MO, USA).
NADPH was from Roche (Mannheim, Germany).
Animals
Female DBA/2N/Kuo and CD2F1 mice were obtained from the National Laboratory
Animal Centre, Kuopio University. For the in vitro experiments, liver microsomes were
prepared from livers of DBA/2N and CD2F1 mice (7–12 weeks old, 15–25 g body weight) as
described previously (Lang et al. 1981; Rahnasto et al. 2003). The CD2F1 mice used in the
in vivo experiments were 7 weeks old (17–23 g body weight) at the beginning of the
experiments. The mice had unrestricted access to water and standard chow (Lactamin R36,
Lactamin AB, So¨derta¨lje, Sweden). The Ethics Committee for Animal Experiments,
University of Kuopio approved these experiments.
In vitro experiments
For the in vitro studies, the mice were administered pyrazole (150 mg kg–1 intraperitoneally
(i.p.) for three consecutive days) or phenobarbital (60 mg kg–1 i.p. for four consecutive days).
The control animals received vehicle only (0.9% saline). The livers were removed into
ice-cold saline and microsomes were prepared, pooled, and stored at –80ꢀC until used
(Lang et al. 1981). Pyrazole is known to maximize CYP2A5 oxidizing capacity

Nicotine metabolism and elimination
37
(Juvonen et al. 1985), and phenobarbital induces several other CYP forms in mouse liver
(Honkakoski and Negishi 1997). Microsomal protein content was determined by the
method of Smith et al. (1985). The preparation and characterization of the anti-CYP2A5
antibody has been described earlier by Raunio et al. (1998). This antibody does not inhibit
CYP2B, CYP2D, and CYP3A activities in mouse liver microsomes.
In vivo experiments
A total of 65 CD2F1 mice were used for the in vivo experiments. The test inhibitors
were administered as single doses by gavage at 2 mg kg–1, followed by nicotine at
10 mg kg–1 (average 1.2 mmol/mouse) 30 min later. The mice were housed in metabolic
cages for 24 h for collection of urine. After measuring the volume, the urine was
centrifuged and the clear supernatant was stored at –20ꢀC until assayed. The inhibitors
(methoxsalen, coumarin, ꢁ-nonanoic lactone, 2,7-dimethylnaphthalene, 2-bromo-
naphthalene, and 4-methoxybenzaldehyde) were chosen because they are potent
CYP2A5 inhibitors in vitro (Ma¨enpa¨a¨ et al. 1993; Juvonen et al. 2000; Rahnasto
et al. 2003, 2005).
High-pressure liquid chromatography (HPLC) assay (in vitro experiments)
Nicotine 50-oxidation was measured indirectly via the formation of cotinine from
nicotine. The reaction mixture consisted of 50 mM potassium phosphate (pH 7.4), 5 mM
MgCl2, 200 mM nicotine, 2 mg liver cytosol (as aldehyde oxidase source), and 33–215 mg
microsomes in a total volume of 0.5 ml. The reaction was initiated by adding NADPH
(1 mM final concentration) and allowed to proceed for 20 min at 37ꢀC. The reaction was
terminated by adding 100 ml of 20% Na2CO3 and mixing on a Vortex. The samples were
extracted with 2 ml of 2-propanol and dichloromethane (1:2 v/v) and shaken vigorously.
A total of 1 ml of the organic phase was transferred to another tube, evaporated by
centrifugation, and the residual was dissolved in 250 ml of the mobile phase solution for
HPLC analysis. Nicotine-to-cotinine formation was determined with a modification of
the HPLC method published by Messina et al. (1997). The HPLC system (Shimadzu)
consisted of a LC-10AD VP pump, a SIL-9A injector, and an SPD-10A VP multi-
wavelength detector. The stationary phase was a reverse phase Supelcosil LC-18-DB
column (C18, 250 ꢁ 4.6 mm). The mobile phase was a mixture of 27% acetonitrile, 73%
20 mM KH2PO4, and 1 mM Na-1-heptane sulfonate. The isocratic flow rate was
0.5 ml minꢂ1 at 60–70 kgf pressure; the sample volume was 50 ml. Based on spectral
analysis of nicotine and cotinine, 261 nm was chosen as the detection wavelength.
For the preparation of standard curves, cotinine (0.125–50 mM) was mixed with the
reaction mixture, extracted, and dissolved in 250 ml of the HPLC mobile phase solution.
The standard curve was linear up to 50 mM cotinine. Cotinine was measured as peak
height and compared with the standard curve, enabling peak heights to be converted to
cotinine concentrations. Nicotine-to-cotinine kinetic studies were carried out by
incubating 10, 20, 40, 100, 20, 400, and 1000 mM nicotine with microsomes from the
variously treated mice. Inhibition studies consisted of incubation of 200 mM nicotine with
various concentrations of the anti-CYP2A5 antibody and the chemical inhibitors
ꢁ-nonanoic lactone and coumarin.

38
H. Raunio et al.
Mass spectrometry assay for nicotine metabolism (in vivo experiments)
Liquid chromatography-atmospheric chemical ionization mass spectrometer (LC-
APCI-MS) consisted of a Thermo Finnigan LTQ (USA) linear ion-trap mass spectrometer
coupled with a Surveyor LC pump. The instrument was operated in positive mode using
multiple reaction monitoring (MRM). A Phenomenex Gemini (C18, 250 ꢁ 2 mm) column
was used. Flow rate was 200 ml minꢂ1 and injection volume was 10 ml. Solvent A was 0.01%
ammonia solution and solvent B was methanol. A linear gradient from 2% B to 100% B
in 20 min was used. The MS settings were: APCI vaporizer temperature, 300ꢀC; capillary
temperature, 200ꢀC; source voltage, 6000 V; collision energy, 46; sheath gas flow, 48
arbitrary units; and AUX gas flow, 10 arbitrary units. XCalibur software was used to control
LC-MS system.
The calibration curves covered the range from 0.01 to 10 mM and their correlation
coefficients ranged from 0.990 to 0.999. All of results were calculated using 1/x weighted
linear regression. All the peak area ratios, regression coefficients, and parameters were
calculated from the peak area data by the LCquan program version 2.0 (Thermo Finnigan).
Stock solutions of all six analytes were prepared in methanol (100 mM). Working solutions
(10 mM) was prepared by mixing stock solutions of all six analytes and further diluting with
water. Calibration curve samples were prepared at concentrations of 0.01, 0.05, 0.1, 0.5,
1.0, 5.0, and 10 mM by diluting the working solution with water. A total of 10 ml of 100 mM
acetanilide solution were added as an internal standard before analysis. The pH of a 250 ml
aliquot of urine was adjusted to pH 5 with phosphate and incubated with 0.1 U of
ꢀ-glucuronidase (Sigma G7017, contains sulfatase activity) for 24 h at room temperature,
centrifuged at 3000g for 20 min, and diluted at 1:15 with water for subsequent analysis.
Calculation of kinetic parameters
Non-linear (Michaelis–Menten) kinetic parameters for nicotine-to-cotinine formation were
calculated using Prism 3.0 software (San Diego, CA, USA).
Results
In vitro metabolism and inhibition of nicotine metabolism
DBA/2N mice were used in the in vitro experiments because they have a high basal hepatic
CYP2A5 activity which is further inducible by chemicals such as pyrazole and phenobarbital
(Juvonen et al. 1985) and CD2F1 mice possess somewhat lower basal CYP2A5 activity
(Wood and Conney 1974; Honkakoski and Negishi 1997). The in vitro nicotine-to-cotinine
formation conditions were optimized with respect to the amount of microsomes in the
reaction mixture and incubation time (data not shown). In addition, different HPLC
conditions were examined to obtain an optimal separation of nicotine and cotinine.
The oxidation of nicotine to cotinine followed Michaelis–Menten kinetics, as a hyperbolic
curve was obtained (Figure 2). The Vmax of the reaction was markedly increased
in microsomes from pyrazole (7.5-fold) and phenobarbital (5.1-fold) pre-treated mice.
The Vmax in CDF1 control mouse microsomes was 63% of the Vmax in DBA/2 control
mouse microsomes (Table I). These results suggest that CYP2A5 is an efficient catalyst of
the reaction since pyrazole and phenobarbital increase the Vmax of the oxidation reaction.

Nicotine metabolism and elimination
39
Figure 2. Michaelis–Menten kinetics of oxidation of nicotine to cotinine. The experiment was
undertaken with microsomes from pyrazole-treated DBA/2 mice. Data are examples of one
determination; the dotted lines represent the 95% confidence limits.
Table I. Kinetic parameters (Michaelis–Menten) of nicotine-to-cotinine formation by mouse
liver microsomes. Oxidation of nicotine to cotinine was measured at 0–1 mM nicotine
concentrations with 30–250 mg microsomes, liver cytosol as a source for aldehyde oxidase, and
1 mM NADPH at 50 mM phosphate buffer pH 7.4. Non-linear one binding site analysis was
used to determine Km and Vmax values.
Km mM
(95% CI)
Vmax (95% CI)
nmol mg–1 protein min–1
Microsomes
DBA/2 control
33
(6.8–59)
23
1.6
(1.3–1.9)
12
DBA/2 pyrazole
(14–33)
38
(11–13)
8.1
DBA/2 phenobarbital
CD2F1 control
(0.0–80)
68
(5.8–10)
1.0
(37–99)
(0.9–1.1)
Note: Data are the average of two determinations.
The role of CYP2A5 in nicotine-to-cotinine formation in vitro was further assessed by
adding anti-CYP2A5 antibody to the reaction mixture. As illustrated in Figure 3, the
antibody efficiently blocked cotinine formation, the inhibition being more than 90% in
microsomes obtained from pyrazole-treated mice. The inhibition was less efficient in
microsomes from phenobarbital-treated mice (70%). The potent in vitro chemical inhibitor
of CYP2A5, ꢁ-nonanoic lactone (Juvonen et al. 2000), was also efficient in reducing
cotinine formation. As shown in Figure 4, up to 100% inhibition of the reaction was
obtained with this latter compound in microsomes from control and pyrazole-treated DBA/2
mice. In addition, another CYP2A5 substrate, coumarin, completely abolished the reaction,
but a concentration as high as 400 mM was required for 100% inhibition (data not shown).
These inhibition results confirm that CYP2A5 is the major enzyme involved in the oxidation
of nicotine to cotinine in mouse liver microsomes.
In vivo inhibition of nicotine metabolism
Nicotine was administered orally to CD2F1 mice, urine collected for 24 h and nicotine and
its metabolites cotinine, cotinine N-oxide, 3-hydroxycotinine, and nicotine 1-N-oxide

40
H. Raunio et al.
Figure 3. Inhibition of cotinine formation by anti-CYP2A5 antibody in mouse liver microsomes.
Microsomes were from D2 mice treated with 0.9% NaCl (1), pyrazole (2), phenobarbital (3), and
untreated CD2F1 (4) mice. Data are the mean ꢃ standard deviation derived from three determina-
tions. The absolute value of the 100% uninhibited activity is 1.4 pmol/(min ꢁ mg protein) for NaCl
D2, 11.8 pmol/(min ꢁ mg protein) for pyrazole D2, 7.1 pmol/(min ꢁ mg protein) for phenobarbital D2
and 0.86 pmol/(min ꢁ mg protein) for CDF1 mice.
Figure 4. Concentration-dependent inhibition of cotinine formation by ꢁ-nonanoic lactone. Data are
the mean ꢃ standard deviation derived from two determinations. The absolute value of the 100%
uninhibited activity is 1.3 pmol/(min ꢁ mg protein) for D2 control and 12.9 pmol/(min ꢁ mg protein)
for D2 pyrazole mice.
determined by the APCI-LC-MS analysis. All these analytes were well ionized and yielded a
protonated molecule M þ Hþ. To increase the specificity of the assay, the precursor ions
were fragmented by colliding them with helium and monitoring the fragment ions created.
MRM chromatograms of all analytes are shown in Figure 5. No significant interfering peaks
were observed on the ion channels used and a good chromatographic separation was
achieved. Cotinine N-oxide and 3-hydroxycotinine were separated by chromatography and
had different fragmentation patterns.
To determine the effect of CYP2A5 inhibitors on nicotine elimination, CD2F1 mice were
administered orally with nicotine (10 mg kg–1) alone or together with the inhibitors
methoxsalen, 2,7-dimethylnaphthalene, 2-bromonaphthalene, and 4-methylnaphthalene
(each 2 mg kg–1). Urine was collected for 24 h in metabolic cages.

Nicotine metabolism and elimination
41
RT: 17.13
100
50
Acetanilide 136.3 → 94.3
Nicotine 163.2 → 84.5
0
RT: 19.22
100
50
0
RT: 13.64
100
Cotinine 177.2 → 80.3
50
0
RT: 9.74
100
Nicotine-1-oxide 179.3 → 84.3
Cotinine-N-oxide 193.2 → 96.3
3-Hydroxycotinine 193.2 → 149.2
50
0
RT: 8.55
100
50
0
RT: 11.46
100
50
0
8
10
12
14
16
18
20
Time (min)
Figure 5. APCI-LC-MS MRM chromatograms of nicotine and its metabolites and MRM transitions
monitored. Acetanilide was used as the internal standard.
There was a considerable variation in the total amount of urine excreted during 24 h
among individual animals (mean volume ¼ 0.64 ml, range ¼ 0.25–3.1 ml). The average yield
of nicotine and four major metabolites from the total nicotine dose (1.2 mmol/mouse) was
39% (range ¼ 6–81%). On average, less than 0.6% of the administered nicotine dose was
excreted unchanged in urine (Figure 6). The major measured metabolite in urine was trans-
3-hydroxycotinine, which represented 77% of the total amount of nicotine-related material.
Cotinine N-oxide, nicotine 1-N-oxide, and cotinine represented 16, 3–5, and 3–4% of the
administered dose, respectively. These results confirm earlier findings (Hukkanen et al.
2005) that in the mouse metabolism is the major elimination route for nicotine with the main
metabolite being trans-3-hydroxycotinine.
Methoxsalen was the only inhibitor that increased the amount of nicotine and
nicotine N-oxide and reduced the amount of cotinine derived metabolites in the urine.
In contrast, 2,7-dimethylnaphthalene had no effect (Figure 7). In addition, 2-
bromonaphthalene and 4-methylnaphthalene did not affect the levels of nicotine and
its metabolites in urine (results not shown). Methoxsalen increased the amount of
nicotine in urine from 1–3 to 21 ꢃ 11 nmol and the amount of nicotine 1-N-oxide to

42
H. Raunio et al.
Trans-3-hydroxycotinine
Cotinine N-oxide
Cotinine
Nicotine 1-N-oxide
Nicotine
0
100
200
300
400
500
600
Amount excreted [nmol/24 h]
Figure 6. Excretion of nicotine and its metabolites into urine. A total of 10 mg kg–1 (average 1.21 mmol/
mouse) nicotine was administered per orally to CD2F1 mice, urine was collected for 24 h, and nicotine
and its major metabolites were determined by HPLC-MS. Data are the mean ꢃ standard deviation
derived from four determinations.
40 ꢃ 18 nmol from the control level of 14–31 nmol. The cotinine level was decreased
by methoxsalen co-administration from 26 ꢃ 12 to 13 ꢃ 6 nmol, cotinine N-oxide from
95 ꢃ 42 to 11 ꢃ 8 nmol, and trans-3-hydroxycotinine from 500 ꢃ 220 to 35 ꢃ 26 nmol.
The yield of nicotine and the four metabolites in urine was markedly decreased by
methoxsalen co-administration as it was only 10% ꢃ 4% in mice given methoxsalen
and 58% ꢃ 25% in mice treated with the other inhibitors or without inhibitor
treatment.
To analyse further the elimination pattern of nicotine, different metabolite/nicotine
ratios were calculated to assess if they could be used as an indicator of inhibitory effects
(Table II). All these ratios varied between 0.2 and 4 in mice given methoxsalen and
between 2 and 840 in control mice. Methoxsalen had the most prominent effect on the
cotinine N-oxide/nicotine and trans-3-hydroxy cotinine/nicotine ratios but only a minor
effect on the nicotine 1-N-oxide/nicotine ratio. In summary, methoxsalen efficiently
inhibited the metabolic elimination of nicotine and metabolite excretion into urine
whereas 2,7-dimethylnaphthalene, 2-bromonaphthalene and 4-methylanaphthalene failed
to do so.
Discussion
The main aim of this study was to assess if the mouse could be used as a model for the
testing of chemical inhibitors of CYP2A6. To achieve this aim, three different types of
experiments were carried out: (1) in vivo metabolism of nicotine, (2) in vitro oxidation of
nicotine to cotinine, and (3) in vivo inhibition of nicotine metabolism by CYP2A5 specific
inhibitors. The results of this study show that: (1) nicotine metabolism is very similar
between mouse and human; (2) CYP2A5 is the main catalyst of nicotine-to-cotinine
formation in mouse liver microsomes in vitro; and (3) orally administered methoxsalen, an
irreversible inhibitor of CYP2A5, reduced the metabolic elimination of nicotine in mice
in vivo and, therefore, leads to a prolongation of the presence of nicotine in the body.

Nicotine metabolism and elimination
43
Trans-3-hydroxycotinine
Cotinine N-oxide
Cotinine
Methoxalen
2.7-dimethylnaphthalene
Control
Nicotine 1-N-oxide
Nicotine
1
10
100
1000
Amount excreted [nmol/24 h]
Figure 7. Effect of CYP2A5 inhibitors on the excretion of nicotine and its metabolites into urine.
Methoxsalen and 2,7-dimethylnaphthalene (2 mg kg–1) were administered per orally to CD2F1 mice
and 10 mg kg–1 nicotine were given 30 min later. Urine was collected for 24 h, and nicotine and its
major metabolites were determined by HPLC-MS. Data are shown as the mean ꢃ standard deviation
derived from four to six determinations.
Table II. Effect of methoxsalen on the metabolite/nicotine ratios in urine.
Metabolite/nicotine ratio
Metabolite
Control
Methoxsalen
Nicotine 1-N-oxide
Cotinine
11.5 ꢃ 6.2
11.2 ꢃ 5.6
62.8 ꢃ 28
297 ꢃ 120
2.0 ꢃ 0.4
0.7 ꢃ 0.1
0.6 ꢃ 0.5
2.0 ꢃ 1.6
Cotinine N-oxide
Trans-3-hydroxy cotinine
Note: Data are the mean ꢃ standard deviation derived from four to six determinations.
This is the first study to assess nicotine metabolism in detail in the intact mouse. Nicotine
and four major metabolites were assayed in this study and the metabolic pattern in mice
resembled that in humans. Cotinine was the major primary metabolite in mouse as it is
also in humans, and trans-3-hydroxy cotinine is the major overall metabolite in urine
(Hukkanen et al. 2005). A total of 40–60% of nicotine is eliminated as trans-3-hydroxycotine
in human smokers, whereas in mouse it represented 75% of nicotine elimination. In mouse,
the second most abundant metabolite was cotinine N-oxide (16% of total), whereas in
humans who smoke, cotinine together with cotinine glucuronide constitute 22–35% of
nicotine elimination. In mice, cotinine and nicotine N-oxide were excreted in urine in equal
amounts. Interestingly, urinary nicotine constitutes less than 1% of its own elimination in
mice whereas in smokers, 8–10% of nicotine is eliminated as such in urine. Three main
reasons may explain this difference. First, smokers are exposed via the lung whereas mice
were exposed orally. Second, smokers are exposed continuously to low nicotine doses and
the mice obtained one high, bolus dose, resulting in different blood and tissue
concentrations. Third, mice seem to be more efficient nicotine metabolizers than humans.
In conclusion, the present study confirms that nicotine oxidation to cotinine is the major
primary metabolic pathway in mouse as it is in humans.
In humans CYP2A6 is the key enzyme in elimination of nicotine (Tyndale and
Sellers 2001). Although CYP2A6 shares only 82% similarity in its amino acid sequence with

44
H. Raunio et al.
CYP2A5, both enzymes efficiently catalyse many common substrates and are inhibited by
the same inhibitors. For example, both enzymes efficiently catalyse coumarin 7-hydro-
xylation, and methoxsalen is a potent inhibitor of both enzymes (Pelkonen et al. 2000). The
current study expands previous knowledge on nicotine-to-cotinine formation in the mouse
and confirms the key role of CYP2A5 in this reaction in the mouse both in vitro and in vivo.
Siu and Tyndale (2007) recently showed that CYP2A5 works equally efficiently in oxidation
of nicotine to cotinine in DBA/2N and C57BL/6 mouse strains, but C57BL/6 is more
efficient in oxidation of cotinine. Mouse liver microsomes were used as the source of the
CYP2A5 enzyme. This approach has the clear advantage that the relative contribution of
CYP2A5 to cotinine formation can be estimated. According to the results obtained with the
specific anti-CYP2A5 antibody, 80% of cotinine formation was attributable to CYP2A5 in
control DBA/2 mouse microsomes. In pyrazole-induced microsomes, the contribution of
CYP2A5 was 90%. As expected, the reduction in cotinine formation was least efficient in
microsomes from phenobarbital-treated mice. This is due to the fact that phenobarbital
induces several CYP forms simultaneously, including CYP2A, CYP2B, and CYP3A forms
(Honkakoski and Negishi 1997). Some of these forms appear to be able to contribute to
cotinine formation in mouse liver. Up to 100% reduction in cotinine formation was obtained
with the chemical inhibitor ꢁ-nonanoic lactone (Juvonen et al. 2000) and the CYP2A5
substrate coumarin. Previous studies have shown that methoxsalen blocks CYP2A5
completely (Ma¨enpa¨a¨ et al. 1993, 1994; Damaj et al. 2007). These data strongly argue
for the crucial contribution of CYP2A5 in nicotine-to-cotinine conversion in mouse liver.
The observed kinetic parameters of nicotine-to-cotinine formation in CD2F1 mouse liver
microsomes are strikingly similar to those in human liver microsomes. The apparent Km and
V
max values for cotinine formation in human liver microsomes are approximately 65 mM and
470 pmol mg–1 protein minꢂ1
,
respectively, with substantial inter-individual variation
(Messina et al. 1997). Coumarin can inhibit up to 80% of nicotine-to-cotinine formation
in human liver microsomes (Messina et al. 1997). Recently, Murphy et al. (2005) showed
´
that CYP2A5 is an efficient catalyst of nicotine 5-oxidation, with a Km ¼ 7.7 mM and a
Vmax ¼ 1.5 pmol product minꢂ1 pmol–1 P450; and Siu and Tyndale (2007) showed that there
exist the high- and low-affinity active site for this reaction in mouse liver microsomes.
The catalysing efficiency expressed as Vmax/Km is much higher for mouse CYP2A5
(0.199 l minꢂ1) than the human CYP2A6 (0.009 l minꢂ1). These latter parameters were
obtained using heterologously expressed enzymes (Murphy et al. 2005). Thus, with regard
to nicotine-to-cotinine reaction kinetics, the CD2F1 mice resemble humans more than the
high activity DBA/2 mice. A recent study (Siu et al. 2006) indicated that increased nicotine
self-administration in mice was associated with increased rates of nicotine metabolism as a
result of greater hepatic CYP2A5 protein levels.
In our in vivo experiments with CD2F1 mice showed that only orally administered
methoxsalen significantly reduced the conversion of nicotine to cotinine. The lack of effect
by the other inhibitors used is not altogether surprising since methoxsalen is an irreversible
inhibitor capable of destroying the CYP2A enzyme (Pelkonen et al. 2000). In contrast,
1,7-dimethylnaphthalene, 2-bromonaphthalene, and 4-methylnaphthalene are thought to
be competitive and reversible inhibitors based on their structural similarity to naphthalene
(Asikainen et al. 2003) and, therefore, they have a transient mode of action. Higher
doses could be more efficient in vivo. However, as they are competitive inhibitors, it is
possible that they are rapidly metabolized and therefore do not have prolonged inhibitory
effects on CYP2A5.
There is considerable interest in the development of an inhibitor of the human CYP2A6
enzyme. Inhibition of CYP2A6 and slower nicotine metabolism have been proposed to

Nicotine metabolism and elimination
45
decrease (1) the risk of smoking initiation and dependence, (2) the number of cigarettes
smoked, and (3) the risk of tobacco-related cancers and mutations. Proof-of-concept studies
have been carried out with the CYP2A6 inhibitors methoxsalen (used in the treatment of
psoriasis) and tranylcypromine (a monoamine oxidase inhibitor used in the treatment of
depression) (Sellers et al. 2000; Tyndale and Sellers 2001; Zhang et al. 2001). A major
disadvantage of these CYP2A6 inhibitors is that CYP enzymes are not the only target
molecules, the other targets being monoamino oxidase for tranylcypromine (Frieling and
Bleich 2006) and methoxsalen being photoactive and hepatotoxic (McNeely and Goa 1998).
Secondly, they are not selective for the various CYP forms and thus inhibit several other
CYPs, especially CYP2E1 and CYP3A4 (Ma¨enpa¨a¨ et al. 1994; Pelkonen et al. 1998;
Taavitsainen et al. 2001) in addition to CYP2A6.
The present authors are currently developing potent and selective CYP2A6 inhibitors
(Poso et al. 1995, 2001; Rahnasto et al. 2005). Usually an inhibitor of the human CYP2A6
is also an inhibitor of the mouse CYP2A5, but there are exceptions such as
2-phenylethylamine, which is a much more potent inhibitor of CYP2A6 (Rahnasto et al.
2003). Initially, the candidate inhibitors are tested with in vitro methods, typically using
heterologously expressed CYP2A6 enzyme. Since it is not possible to test these compounds
in humans (unless they are registered drugs), an animal model is needed. The crucial
prerequisite for such a model is that the conversion of nicotine to cotinine occurs in a similar
fashion as is the case in humans. The mouse in vivo model characterized in the current study
appears to fulfil this basic tenet. The effect of inhibitors on the elimination of nicotine
can well be evaluated by determining nicotine and its metabolites in urine and the ratio of
trans-3-hydroxycotine and cotinine N-oxide to nicotine are the most sensitive indicators for
inhibitory effects.
In conclusion, robust and simple HPLC and HPLC/MS methods have been developed
for the quantitation of nicotine-to-cotinine biotransformation by mouse liver, both in vitro
and in vivo. The relative contribution of liver CYP2A5 enzyme to this reaction was found to
be large, up to 100% after CYP2A5 induction by pyrazole. The main features of urine
metabolic pattern of nicotine are similar in mouse and human. Therefore this in vivo model
can be used to test the performance of chemical inhibitors of CYP2A5/CYP2A6.
Acknowledgements
The authors thank Ms Hannele Jaatinen for expert technical help; and Dr Ewen MacDonald
for assistance in preparation of the manuscript.
References
Asikainen A, Tarhanen J, Poso A, Pasanen M, Alhava E, Juvonen RO. 2003. Predictive value of comparative
molecular field analysis modelling of naphthalene inhibition of human CYP2A6 and mouse CYP2A5 enzymes.
Toxicology In Vitro 17:449–455.
Benowitz NL. 1996, Pharmacology of nicotine: Addiction and therapeutics. Annual Review of Pharmacology and
Toxicology 36:597–613.
Brandange S, Lindblom L. 1979. The enzyme ‘aldehyde oxidase’ is an iminium oxidase. Reaction with nicotine
delta 10(50) iminium ion. Biochemical and Biophysical Research Communications 91:991–996.

46
H. Raunio et al.
Damaj MI, Siu EC, Sellers EM, Tyndale RF, Martin BR. 2007. Inhibition of nicotine metabolism by methoxsalen:
Pharmacokinetic and pharmacological studies in mice. Journal of Pharmacology and Experimental Therapeutics
320:250–257.
Ezzati M, Hoorn SV, Rodgers A, Lopez AD, Mathers CD, Murray CJ. 2003. Comparative Risk Assessment
Collaborating Group. Estimates of global and regional potential health gains from reducing multiple major risk
factors. Lancet 362:271–280.
Frieling H, Bleich S. 2006. Tranylcypromine: New perspectives on an ‘old’ drug. European Archives of Psychiatry
and Clinical Neuroscience 256:268–273.
George TP, O’Malley SS. 2004. Current pharmacological treatments for nicotine dependence. Trends in
Pharmacological Sciences 25:42–48.
Honkakoski P, Negishi M. 1997. The structure, function and regulation of cytochrome P450 2A enzymes.
Drug Metabolism Reviews 29:977–996.
Hukkanen J, Jacob P, Benowitz NL. 2005. Metabolism and disposition kinetics of nicotine. Pharmacological
Reviews 57:79–115.
Juvonen RO, Gynther J, Pasanen M, Alhava E. 2000. Pronounced differences in inhibition potency of lactone and
non-lactone compounds for mouse and human coumarin 7-hydroxylases. Xenobiotica 1:81–92.
Juvonen RO, Kaipainen PK, Lang MA. 1985. Selective induction of coumarin 7-hydroxylase by pyrazole in D2
mice. European Journal of Biochemistry 152:3–8.
Kamataki T, Fujieda M, Kiyotani K, Iwano S, Kunitoh H. 2005. Genetic polymorphism of CYP2A6 as one of the
potential determinants of tobacco-related cancer risk. Biochemical and Biophysical Research Communications
338:306–310.
Lang MA, Gielen JE, Nebert DW. 1981. Genetic evidence for many unique liver microsomal P-450 mediated
monooxygenase activities in heterogenic stock mice. Journal of Biological Chemistry 256:12068–12075.
Ma¨enpa¨a¨ J, Juvonen R, Raunio H, Rautio A, Pelkonen O. 1994. Metabolic interactions of methoxsalen and
coumarin in humans and mice. Biochemical Pharmacology 48:1363–1369.
Ma¨enpa¨a¨ J, Sigusch H, Raunio H, Syngelma¨ T, Vuorela P, Vuorela H, Pelkonen O. 1993. Differential inhibition of
coumarin 7-hydroxylase activity in mouse and human liver microsomes. Biochemical Pharmacology
45:1035–1042.
Malaiyandi V, Sellers EM, Tyndale RF. 2005. Implications of CYP2A6 genetic variation for smoking behaviors and
nicotine dependence. Clinical Pharmacology and Therapeutics 77:145–158.
McNeely W, Goa KL. 1998. 5-Methoxypsoralen. A review of its effects in psoriasis and vitiligo. Drugs 56:667–690.
Messina ES, Tyndale RF, Sellers EM. 1997. A major role for CYP2A6 in nicotine C-oxidation by human liver
microsomes. Journal of Pharmacology and Experimental Therapeutics 282:1608–1614.
Murphy SE, Raulinaitis V, Brown KM. 2005. Nicotine 50-oxidation and methyl oxidation by P450 2A enzymes.
Drug Metabolism and Disposition 33:1188–1173.
Nakajima M, Yamamoto T, Nunoy K, Yokoi T, Nagashima K, Inoue K, Funae Y, Shimada N, Kamataki T,
Kuroiwa Y. 1996. Characterization of CYP2A6 involved in 30-hydroxylation of cotinine in human liver
microsomes. Journal of Pharmacology and Experimental Therapy 277:1010–1015.
Oscarson M. 2001. Genetic polymorphism in the cytochrome P450 2A6 (CYP2A6) gene: Implications for
interindividual differences in nicotine metabolism. Drug Metabolism and Disposition 29:91–95.
Pelkonen O, Ma¨enpa¨a¨ J, Taavitsainen P, Rautio A, Raunio H. 1998. Inhibition and induction of human P450
(CYP) enzymes. Xenobiotica 28:1203–1253.
Pelkonen O, Rautio A, Raunio H, Pasanen M. 2000. CYP2A6: A human coumarin 7-hydroxylase. Toxicology
144:139–147.
Poso A, Gynther J, Juvonen RO. 2001. A comparative molecular field analysis of cytochrome P450 2A5 and 2A6
inhibitors. Journal of Computer Aided Molecular Design 15:195–202.
Poso A, Juvonen R, Gynther J. 1995, Comparative molecular field analysis of compounds with CYP2A5 binding
affinity. Quantitative Structure–Activity Relationship 14:507–511.
Rahnasto M, Raunio H, Poso A, Juvonen R. 2003. More potent inhibition of human CYP2A6 than mouse
CYP2A5 enzyme activities by derivatives of phenylethylamine and benzaldehyde. Xenobiotica 33:529–539.
Rahnasto M, Raunio H, Poso A, Wittekindt C, Juvonen RO. 2005. Quantitative structure–activity relationship
analysis of inhibitors of the nicotine metabolizing CYP2A6 enzyme. Journal of Medicinal Chemistry 48:440–449.
Raunio H, Juvonen R, Pasanen M, Pelkonen O, Pa¨a¨kko¨ P, Soini Y. 1998. Cytochrome P4502A6 (CYP2A6)
expression in human hepatocellular carcinoma. Hepatology 27:427–432.
Raunio H, Rautio A, Gullsten H, Pelkonen O. 2001. Polymorphisms of CYP2A6 and its practical consequences.
British Journal of Clinical Pharmacology 52:357–363.
Sellers EM, Kaplan HL, Tyndale RF. 2000. Inhibition of cytochrome P450 2A6 increases nicotine’s oral
bioavailability and decreases smoking. Clinical Pharmacology and Therapeutics 68:35–43.

Nicotine metabolism and elimination
47
Sellers EM, Tyndale RF. 2000. Mimicking gene defects to treat drug dependence. Annals of the New York
Academy of Sciences 909:233–246.
Sellers EM, Tyndale RF, Fernandes LC. 2003. Decreasing smoking behaviour and risk through CYP2A6
inhibition. Drug Discovery Today 8:487–493.
Siu ECK, Tyndale RF. 2007. Characterization and comparison of nicotine and cotinine metabolism in vitro and
in vitro in DBA/2 and C57BL/6 mice. Molecular Pharmacology 71:826–834.
Siu ECK, Wildenauer DB, Tyndale RF. 2006. Nicotine self-administration in mice is associated with rates of
nicotine inactivation by CYP2A5. Psychopharmacology 184:401–408.
Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, Provenzano MD, Fujimoto EK, Goeke NM,
Olson BJ, Klenk DC. 1985. Measurement of protein using bicinchoninic acid. Analytical Biochemistry
150:76–85.
Su T, Ding X. 2004. Regulation of the cytochrome P450 2A genes. Toxicology and Applied Pharmacology
199:285–294.
Taavitsainen P, Juvonen R, Pelkonen O. 2001. In vitro inhibition of cytochrome P450 enzymes in human liver
microsomes by
a potent CYP2A6 inhibitor, tranyl-2phenylcyclopropylamine (tranylcypromine), and its
nonamine analog, cyclopropylbenzene. Drug Metabolism and Disposition 29:217–222.
Tyndale RF, Sellers EM. 2001. Variable CYP2A6-mediated metabolism alters smoking behaviour and risk. Drug
Metabolism and Disposition 29:548–552.
Wood AW, Conney AH. 1974. Genetic variation in coumarin hydroxylase activity in the mouse (Mus musculus).
Science 185:612–614.
Zhang W, Kilicarslan T, Tyndale RF, Sellers EM. 2001. Evaluation of methoxsalen, tranylcypromine, and
tryptamine as specific and selective CYP2A6 inhibitors in vitro. Drug Metabolism and Disposition 29:897–902.