Rational design of novel CYP2A6 inhibitors

Article abstract of DOI:10.1016/j.bmc.2014.10.001

Inhibition of CYP2A6-mediated nicotine metabolism can reduce cigarette smoking. We sought potent and selective CYP2A6 inhibitors to be used as leads for drugs useful in smoking reduction therapy, by evaluating CYP2A6 inhibitory effect of novel formyl, alk

Full text of DOI:10.1016/j.bmc.2014.10.001

Bioorganic & Medicinal Chemistry 22 (2014) 6655–6664
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Rational design of novel CYP2A6 inhibitors
a
a
a
a
b
Niina Tani a, , Risto O. Juvonen , Hannu Raunio , Muluneh Fashe , Jukka Leppänen , Bin Zhao ,
⇑
Rachel F. Tyndale b, Minna Rahnasto-Rilla a,
⇑
a School of Pharmacy, Faculty of Health Sciences, University of Eastern Finland, POB 1627, 70211 Kuopio, Finland
b Departments of Pharmacology and Toxicology and Psychiatry, University of Toronto, Campbell Family Mental Health Research Institute, M5S 1A8 Toronto, Ontario, Canada
a r t i c l e i n f o
a b s t r a c t
Article history:
Inhibition of CYP2A6-mediated nicotine metabolism can reduce cigarette smoking. We sought potent and
selective CYP2A6 inhibitors to be used as leads for drugs useful in smoking reduction therapy, by evalu-
ating CYP2A6 inhibitory effect of novel formyl, alkyl amine or carbonitrile substituted aromatic core
structures. The most potent CYP2A6 inhibitors were thienopyridine-2-carbaldehyde, benzothienoph-
ene-3-ylmethanamine, benzofuran-5-carbaldehyde and indole-5-carbaldehyde, with IC50 values below
Received 18 June 2014
Accepted 2 October 2014
Available online 13 October 2014
Keywords:
Cytochrome P450 (CYP)
CYP2A6
Nicotine
Inhibition
0.5 lM for coumarin 7-hydroxylation. Nicotine oxidation was effectively inhibited in vitro by two alkyl
amine compounds and benzofuran-5-carbonitrile. Some of these molecules could serve as potential lead
molecules when designing CYP2A6 inhibitory drugs for smoking reduction therapy.
Ó 2014 Elsevier Ltd. All rights reserved.
Smoking reduction therapy
1. Introduction
CYP2A6 activity in vivo are more likely to be non-smokers than
their counterparts with normal CYP2A6 activity;5 smoking cessa-
Tobacco smoking increases the risk for numerous types of can-
cer, as well as cardiovascular and respiratory diseases, duodenal
and gastric ulcers, osteoporosis and diabetes, resulting in a signif-
icantly shorter life expectancy in smokers relative to non-smok-
ers.1,2 Most smokers would like to quit smoking and many
attempt to do so, but they often fail to quit, and/or resume smoking
after some time.1 The primary addictive agent in cigarettes and
smokeless tobacco (chewing tobacco, snuff, nicotine gum) is nico-
tine. Nicotine is a bicyclic alkaloid with pyridine and pyrrolidine
rings. It binds to nicotine acetylcholine receptors in the brain evok-
ing psychoactive effects, including pleasure and feelings of relaxa-
tion. Smoking also helps to control mood and concentration.3
Nicotine metabolism is two-step reaction, where nicotine is first
tion trials have also demonstrated that smokers who have slower
CYP2A6 activity have better cessation rates.6,7 It has been sug-
gested that inhibition of CYP2A6 activity, that is, phenocopying
the reduced activity genotype, could help smokers to both reduce
how much they smoke and help them quit smoking. Pilot studies
with nonspecific CYP2A6 inhibitors have yielded promising
results.8–11
Most compounds that inhibit CYP enzymes are reversible com-
petitive inhibitors. In this case the interaction between the inhibi-
tor and enzyme occurs via non-covalent bonds, and the inhibition
is transient. Mechanism based irreversible CYP inhibition causes
longer duration enzyme inactivation. Here stable covalent bonds
are formed between the inhibitor and the enzyme active site,
and the recovery of enzyme activity requires de novo synthesis
of enzyme. Irreversible inhibitors contain specific substructures
that are transformed to reactive electrophilic functional groups in
the enzyme active site, reacting with nucleophilic amino acid res-
idues or protoporphyrin IX (heme).
Numerous compounds with CYP2A6 inhibiting properties have
been discovered.12–16 We have used several different chemical
scaffolds to obtain compounds with more potent and selective
inhibitory properties. Thus far, we have obtained these compounds
from commercial sources, including extensive chemical libraries
(e.g. Maybridge). We have shown previously that the benzothio-
phene core with acceptor atoms evoked potent CYP2A6
inhibition.17,18
oxidized in a rate-limiting step to nicotine-D
50(10)-iminium ion,
mainly by the cytochrome P450 (CYP) 2A6 enzyme, and then to
cotinine by cytosolic aldehyde oxidase.4
The CYP2A6 gene is polymorphic, and several loss-of-function
alleles lead to inactivity or reduced activity of the enzyme
(www.cypalleles.ki.se/cyp2a6.htm). Genetic variation in this
enzyme alters many smoking behaviors. For example, individuals
with reduced or loss of function variants associated with 50%
⇑
Corresponding authors. Tel.: +358 40 355 2011; fax: +358 17 162424 (N.T.); tel.:
+358 40 355 3786; fax: +358 17 162424 (M.R.-R.).
E-mail addresses: Niina.Tani@uef.fi (N. Tani), Minna.Rahnasto@uef.fi
(M. Rahnasto-Rilla).
http://dx.doi.org/10.1016/j.bmc.2014.10.001
0968-0896/Ó 2014 Elsevier Ltd. All rights reserved.

6656
N. Tani et al. / Bioorg. Med. Chem. 22 (2014) 6655–6664
benzo[b]thiophen-3-yl)acetonitrile
(21),
5-chloro-3-meth-
ylbenzo[b]thiophene (22), 1-benzothiophene-3-carbonyl chloride
(23), 1-benzofuran-5-carbaldehyde (24), benzofuran-5-carbonitrile
(25) 1-benzofuran-2-carbaldehyde (26), 2-benzofuran-carbonitrile
(27) 5-bromo-1-benzofuran-2-carbaldehyde (28), Benzo[b]furan-
2-carboxaldehyde (29), (5-bromobenzo[b]furan-2-yl)methylamine
(30), 2-(benzo[b]thiophen-3-yl)acetaldehyde (31), 2-bromo-1-ben-
zofuran (32), were all purchased from Maybridge, Great Britain. 1H-
indole-5-carbaldehyde (33), 1-methyl-1H-indole-5-carbaldehyde
(34), 1H-indole-6-carbaldehyde (35), 5-bromo-1H-indole (36), 1-
methyl-1H-indole-6-carbaldehyde (37), 1H-indole-2-carbaldehyde
(38), 4-bromo-1H-indole-3-carbaldehyde (39), 2-methyl-1H-indol-
5-amine (40), 6-bromo-1H-indole-3-carbaldehyde (41), 1H-indole-
3-carbaldehyde (42), 4-ethylbenzonitrile (43) and p-tolunitrile (44)
were all purchased from Sigma Aldrich, USA.
Thieno[3,2-c]pyridine-2-carbaldehyde (1), thieno[2,3-c]pyri-
dine-2-carbaldehyde (2), thieno[3,2-c]pyridin-2-ylmethanamine
(7), N,N-dimethyl-1-(thieno[3,2-c]pyridin-2-yl)methanamine (8),
N-methyl-1-(thieno[3,2-c]pyridin-2-yl)methanamine (9) and thie-
no[2,3-c]pyridin-2-ylmethanamine (14) were synthesized in-house
(see Scheme 1). The synthesis reactions were monitored by thin-
layer chromatography using aluminum sheets coated with silica
gel 60 F245 (0.24 mm) with UV and ninhydrin visualization. Purifi-
cations by flash chromatography were performed either on a Combi-
flash Companion Instrument with RediSep Columns (Teledyne ISCO,
Lincoln, CA, USA) or by normal column chromatography with silica
gel 60 (0.063–0.200 mm mesh). 1H nuclear magnetic resonance
(NMR) spectra were recorded on a Bruker Avance AV 500 spectrom-
eter (Bruker Biospin, Fallanden, Switzerland) operating at
500.13 MHz using the solvent peak as an internal standard. Further-
more, the products were characterized by mass spectrometry with a
Finnigan LCQ quadrupole ion trap mass spectrometer (Finnigan
MAT, San Jose, CA) equipped with an electrospray ionization source,
and the purity was determined by elemental analysis (C, H, N) with a
ThermoQuest CE Instruments EA 1110-CHNS-O elemental analyzer
(CE Instruments, Milan, Italy). All compounds tested in the assays
were characterized by combustion analysis to be of at least 95% of
purity unless otherwise stated.
Thieno[3,2-c]pyridine-2-carbaldehyde (1). To the cooled to
ꢀ78 °C mixture of thieno[3,2-c]pyridine (1.3 g, 9.62 mmol) in dry
tetrahydrofuran (40 ml), n-butyllithium (1.6 M in hexane,
7.51 ml) was added dropwise. Dimethylformamide (0.88 g,
12 mmol) was added to the solution after 20 min stirring and the
mixture was allowed to warm up to room temperature and was
treated with saturated ammonium chloride solution. The mixture
was extracted with diethyl ether and the organic phase was dried
with sodium sulfate. The product was purified by column chroma-
tography using 1:1 petroleum ether/ethyl acetate as an eluent to
give 1.2 g (76%) of a yellowish solid. 1H NMR CDCl3: d 7.83 (d,
1H), 8.13 (s, 1H), 8.57 (d, 1H), 9.25 (s, 1H), 10.14 (1H, s). ESI-MS:
164.07 (M+1). Anal. Calcd for C8H5NOS: C, 58.88; H, 3.09; N,
8.58; found C, 58.92; H, 3.23; N, 8.29.
Figure 1. Refining of the novel compounds. Hydrogen bonding and
p–p-stacking
sites are illustrated as light blue and light green half-moon shapes, respectively. The
main interacting amino acid residues (Phe107, Phe118, Asn297) and the heme
moiety are illustrated as spheres. The reference compound: R = NH2, X = S. Main
scaffolds: R = NH2, NHCH3, N(CH3)2, CH2O, X = S, O, NH.
The main aims of this study were to expand the chemical spec-
trum of known CYP2A6 inhibitors, and to clarify the optimal steric
and electrostatic interactions occurring between the inhibitor and
the enzyme active site. A further goal was to discover irreversible
inhibitors of CYP2A6. We used the previously identified benzothio-
phene, benzofuran and benzoindole cores and used molecular mod-
eling to detect optimal interactions with the enzyme active site
(Fig. 1). Novel thienopyridine derivatives were synthesized based
on this information. Two molecules showed the desired properties
and will be taken to further development as lead compounds.
2. Materials and methods
2.1. Chemicals
5-Chloro-1-benzothiophen-3-yl)methylamine
hydrochloride
(compound 3), N-(1-benzothien-2-ylmethyl)-N-methylamine
hydrochloride (4), 5-bromo-1-benzothiophen-3-yl)methylamine
(5), 1-benzothiophene-5-carbaldehyde (6), N-methyl-1-(thie-
no[2,3-b]pyridin-2-ylmethyl)amine (10), 1-benzothiophene-2-
carbaldehyde (11), 5-methyl-1-benzothiophene-2-carbaldehyde
(12), 5-bromobenzo[b]thiophene-3-carbaldehyde (13), thieno
[3,2-c]pyridine-3-carbothioamide (15), thieno[2,3-b]pyridine-2-
carbaldehyde (16), benzo[b]thiophene-7-carbaldehyde (17), 1-ben-
zothiophen-2-ylmethanol (18), 1-benzothiophen-3-ylmethanol
(19), 5-chloro-1-benzothiophene-3-carbonitrile (20), 2-(5-chloro-
R1
b or c
d,e
N
H
R2
O
H
Y
Y
Z
Y
Z
a
Z
S
S
S
7 Y=N, Z=CH, R1, R2 =CH3
8 Y=N, Z=CH, RI=CH3, R2=H
9 Y=N, Z=CH, R1, R2 =H
10 Y=CH, Z=N, R1, R2=H
1 Y=N, Z=CH
2 Y=CH, Z=N
Scheme 1. Synthesis of 1,2: (a) n-BuLi, DMF; 7: (b) Me2NH2, NaCNBH3, AcONa, MeOH; 8: (c) MeNH2, NaBH4 EtOH; 9, 10: (d) NH2OH*HCl, EtOH/Pyridine (e) Zn, AcOH/MeOH.

N. Tani et al. / Bioorg. Med. Chem. 22 (2014) 6655–6664
6657
Thieno[2,3-c]pyridine-2-carbaldehyde (2). Thieno[2,3-c]pyri-
dine (330 mg, 2,44 mmol) was reacted and purified similar method
to (1) to give 200 mg (50%) of a yellowish solid. 1H NMR CDCl3: d
7.79 (d, 1H), 8.03 (s, 1H), 8.58 (d, 1H), 9.23 (s, 1H), 10.18 (1H, s).
ESI-MS: 164.07 (M+1). Anal. Calcd for C8H5NOS: C, 58.88; H,
3.09; N, 8.58; found C, 58.82; H, 3.14; N, 8.28.
N,N-Dimethyl-1-(thieno[3,2-c]pyridin-2-yl)methanamine (8).
Thieno[3,2-c]pyridine-2-carbaldehyde (73 mg, 0.45 mmol), sodium
cyanoborohydride (31 mg, 0.49 mmol), dimethylamine hydrochlo-
ride (77 mg, 0.94 mmol) and sodium acetate (55 mg, 0.67 mmol)
were dissolved in methanol (8 ml) and stirred for 24 h. Ethyl acetate
(20 ml) was added and the mixture was acidified with 6 N HCl and
washed twice with saturated NaHCO3 solution. Organic phase was
dried and evaporated and the product was purified by column chro-
matography using dichloromethane/MeOH 20–100% to yield 70 mg
(59%) of a white solid. 1H NMR CD3OD: d 2.3 4 (s, 6H), 4.91 (s, 2H),
7.43 (d, 1H), 7.95 (dd, 1H), 8.34 (m, 1H), 8.97 (s, 1H). ESI-MS:
193.06 (M+1). Anal. Calcd for C10H12N2S*2HCl: C, 45.29; H, 5.32; N,
10.56; found C, 46.95; H, 4.40; N, 8.50.
The characterization by 1H NMR, mass spectrometry (MS), and
high-performance liquid chromatography (HPLC) methods
confirmed that the purities were greater than 95%.
(ꢀ)-Nicotine hydrogen tartrate and (ꢀ)-cotinine were pur-
chased from Sigma-Aldrich (St. Louis, MO). The internal standard
5-methylcotinine was custom-made by Toronto Research Chemi-
cals (Toronto, ON). Tris–hydrochloric acid (MP Biomedicals LLC,
USA), Tris (MP Biomedicals LLC, USA), magnesium chloride hexahy-
drate (Riedel-de HaenÒ, Germany), coumarin (Sigma Aldrich, USA),
7-hydroxycoumarin (Aldrich), trichloroacetic acid (TCA) (Sigma–
Aldrich), nicotinamide adenine dinucleotide phosphate-oxidase
(NADPH) (Darmstadt, Germany), dimethyl sulfoxide (DMSO) (J.T.
Baker).
2.2. Biological materials
Human liver samples were obtained from patients undergoing
surgery to remove hepatic tumors.19 The use of surplus tissue
was approved by the Ethics Committee of the University of Kuopio.
Liver samples were frozen in liquid nitrogen and stored at ꢀ70 °C.
Only tumor-free tissue was used for the experiments. Baculovirus-
insect cell-expressed human CYP2A6 was purchased from BD
Biosciences Discovery Labware (Bedford, MA, USA).
N-Methyl-1-(thieno[2,3-c]pyridin-2-yl)methanamine, (9) Thie-
no[2,3-c]pyridine-2-carbaldehyde (43 mg, 0.26 mmol) and methyl-
amine 8 M in EtOH (65 ll, 0.52 mmol) were dissolved in ethanol
(8 ml) and stirred for 18 h in sealed vial. Sodium borohydride
(11 mg, 0.31 mmol) was added to the vial and the stirring was con-
tinued for 2 h. Ethyl acetate (20 ml) was added and the mixture
was washed twice with saturated NaHCO3 solution. The organic
phase was dried, evaporated and the product was purified by col-
umn chromatography using dichloromethane/MeOH from 0% to
20% to yield 41 mg (63%) of a white solid. 1H NMR CD3OD: d 2.5
4 (s, 3H), 4.11 (s, 2H), 7.20 (s, 1H), 7.57 (d, 1H), 8.48 (d, 1H), 9.07
(s, 1H). ESI-MS: 179.06 (M+1). Anal. Calcd for C9H10N2S*2HCl: C,
42.42; H, 4.79; N, 10.99; found C, 42.78; H, 4.44; N, 10.48.
Thieno[3,2-c]pyridin-2-ylmethanamine (7). Thieno[3,2-c]pyri-
dine-2-carbaldehyde (50 mg, 0.31 mmol) and hydroxylamine
hydrochloride (32 mg, 0.46 mmol) were dissolved in ethanol
(4 ml) and pyridine (1 ml) reacted at 80 °C for 20 min. The mixture
was evaporated to dryness and the solid was dissolved in MeOH
(10 ml) and AcOH (10 ml) and reacted with zinc powder (98 mg)
for 1 h. The mixture was filtered and evaporated to dryness. The
product was purified by column chromatography using dichloro-
methane/MeOH 20% and was washed with hexane to yield 55 mg
(82%) of a white solid. 1H NMR CD3OD: d 1.9 4 (s, 3H, AcOH),
4.31 (s, 2H), 7.55 (s, 1H), 7.97 (d, 1H), 8.36 (d, 1H), 9.00 (s, 1H).
ESI-MS: 164.98 (M+1). Anal. Calcd for C8H8N2Sꢁ1.3*AcOH: C,
52.55; H, 5.49; N, 11.56; found C, 52.52; H, 5.12; N, 11.59.
Thieno[2,3-c]pyridin-2-ylmethanamine (14), thieno[2,3-c]pyri-
dine-2-carbaldehyde (80 mg, 0.49 mmol) was treated similar
method as (7) to give 90 mg (65%) of a white solid. 1H NMR CD3OD:
d 1.84 (s, 3H, AcOH), 4.21 (s, 2H), 7.37 (s, 1H), 7.69 (d, 1H), 8.30 (d,
1H), 8.97 (s, 1H). ESI-MS: 165.07 (M+1). Anal. Calcd for C8H8N2-
S*AcOH: C, 53.55; H, 5.39; N, 12.49; found C, 53.33; H, 5.31; N,
12.34.
2.3. Determination of IC50 and Ki values
The IC50 values for inhibition of coumarin 7-hydroxylation were
determined as described previously.20 The Ki values were mea-
sured or calculated (Ki = IC50/(1 + S/Km)) for the most potent
CYP2A6 inhibitors. The concentrations of the inhibitors used in this
experiment were the same as those for the IC50 assay. The reaction
was initiated with the addition of NADPH (final concentration
0.33 mM) and the reaction mixture was preincubated for 10 min
at 37 °C and stopped with the addition of 10% TCA. The fluores-
cence of the product 7-hydroxycoumarin was measured immedi-
ately after addition of 140 lL of 1.6 M glycine-NaOH at excitation
and emission wavelengths of 355 nm 450 nm, respectively.
2.4. Determination of inhibition type
The IC50 shift assay detects both reversible and time-dependent
inhibitors. The effect of preincubation on the IC50 values of all the
compounds was determined (Fig. 2). The reaction pool (50 mM
Tris–HCl, 5 mM MgCl2, 0.33 mM NADPH and 1 pmol cDNA
expressed CYP2A6 enzyme), was preincubated (without or with
the NADPH-generating system) for 0–60 min at 37 °C. After prein-
cubation, coumarin (final concentration 10
mixture was further incubated for 10 min with cofactors at 37 °C.
The reactions were stopped with 60 l of 10% TCA. The remaining
lM) was added and the
l
enzyme activity in the presence of inhibitor was measured in a
Victor2 fluorometer (Perkin Elmer) immediately after adding
140
l
L of 1.6 M (pH 10.4) glycine buffer.
Preincubaꢁon
test II: irreversible
inacꢁvaꢁon
Inacꢁvaꢁon:
No. 16, 24, 25, 38
IC50-values:
No.1-44
Preincubaꢁon test I:
IC50 Shiꢀ:
No. 11, 13, 16,
24, 25, 38
IC50 shiꢀ
Catalyꢁc reacꢁon
study
Slow-binding
No.11-13
Figure 2. Scheme of the tests to elucidate mechanism of inhibition.

6658
N. Tani et al. / Bioorg. Med. Chem. 22 (2014) 6655–6664
For determining an irreversible mode of inhibition, five con-
2.7. Modeling
centrations of the test compounds (plus a vehicle control) were
preincubated for various times (including 0 min) with 5–20 pmol
cDNA-expressed CYP2A6 and NADPH (Fig. 2). The reaction mix-
ture was diluted 10–40 fold with 50 mM Tris–HCl buffer (pH
The crystal structure of CYP2A623 was obtained from the Pro-
tein Data Bank (PDB) database (pdb: 2FDW). The structure and
the test compounds were further refined as described previously.18
All compounds were docked to the active site of this refined
CYP2A6 structure CYP2A6 using the Sybyl X 1.224 with Surflex-
Dock module.25 The docking procedure was allowed to produce
up to 30 different docking poses for each inhibitor. The poses were
then ranked based on CScore (consensus score)26 ranking values
and visual inspection.
7.4), 5 mM MgCl2, 10 lM coumarin, and 0.3 mM NADPH, and
the incubation was continued for 10 min at 37 °C. The reaction
was terminated and measured as described for the IC50 assay.
From the data generated, two kinetic constants, the maximum
inactivation rate constant (Kinact) and the inhibitor concentration
that produced half maximal rate of inactivation (Ki) were deter-
mined using linear regression of double reciprocal of values of
observed inactivation rate constant (Kobs) against [I] where Ki
is the negative reciprocal value of x-intercept and Kinact is the
reciprocal of the value of y-intercept. Kobs values were
determined from the initial slope of the linear regression of
the natural log (ln) of enzyme activity remaining against
preincubation time.
3. Results
3.1. Selection of compounds
Selection of novel inhibitor molecules was initiated by docking
the previously identified inhibitors, including benzothiophene
derivatives17,18 to the active site of CYP2A6. The main interactions
between the inhibitors and the active site residues were hydrogen
2.5. Nicotine oxidation
bonding with Asn297 and
p–p stacking with phenylalanines 107
Nicotine metabolizing activity was analyzed as previously
described21 where drug concentrations are expressed as the free
base. For expressed CYP2A6 (containing P450 reductase and cyto-
chrome b5) (BD Biosciences, Mississauga, Canada), the linear con-
ditions of nicotine metabolism were obtained under assay
conditions of 10 pmol CYP enzyme and 10 pmol cytochrome
b5/ml with an incubation time of 15 min. All incubation mixtures
contained 1 mM NADPH and 50 ll cytosol in 50 mM Tris–HCl buf-
fer, pH 7.4 and were performed at 37 °C in a final volume of
0.5 ml. The reactions were stopped with a final concentration of
4% v/v Na2CO3. After incubation, 5-methylcotinine (70 lg) was
added as the internal standard and the samples were prepared
and analyzed for nicotine and metabolites by HPLC system as
described previously.22 The limits of quantification were 5 ng/ml
for nicotine, 12.5 ng/ml for cotinine and 10 ng/ml for
30-hydroxycotinine.
and 118 (Fig. 1). To identify potential irreversible inhibitors of
CYP2A6, compounds with known inactivating substructures were
screened. These substructures were expected to form covalent
interactions and thus inactivate the enzyme. Figure 3A illustrates
the docking pose of the reference compound benzo[b]thiophen-
2-ylmethanamine18 that has otherwise an optimal structure but
lacks a proper acceptor atom and thus did not form a hydrogen
bond with Asn297. Many of the purchased novel compounds did
not form desired hydrogen bonding interactions.
To identify potent inhibitors with a properly placed acceptor
atom, compounds 1, 2, 7, 8, 9, and 14 were synthesized (Table 1).
To optimize the position of the acceptor atom, the initial benzo-
thiophene core was replaced with a thienopyridine with formyl
or alkyl amine substitutions at position two, and a nitrogen atom
at position five or six. The docking poses showed that the nitrogen
atom of the thienopyridine with formyl substitution formed a
hydrogen bond with residue Asn297 (Fig. 3B). The alkyl amine sub-
stituent in most of the thienopyridine derivatives was coordinated
towards heme although for most compounds Asn297 was too far
for hydrogen bonding interaction (Fig. 3C).
2.6. Data analysis
The data were analyzed using GraphPad Prism software v5.01.
Enzyme inactivation kinetic values (IC50, Ki, Km, Vmax) were deter-
mined with the GraphPad Prism software using nonlinear modes
and various enzyme kinetics equations. For the estimation of IC50
values, the fluorescence was measured at different concentrations
of the inhibitors. Control incubations (blanks) were carried out
without the substrate, enzyme, or inhibitor. The blank values were
deducted from all groups and the percentage of remaining activity
was calculated at each inhibitor concentration taking control group
values as 100% enzyme activity.
3.2. Inhibition of coumarin 7-hydroxylation
Basic coumarin 7-hydroxylation inhibition kinetic parameters
were determined for 44 compounds (Table 1). The IC50 values var-
ied from 0.3
the type and position of substituents affected the IC50 value. The
M) were the formyl derivatives
lM to more than 100 lM. The core structure as well as
most potent inhibitors (IC50 60.5
l
of thiophene, benzofuran and indole, for example, compounds 1, 2,
A
B
C
PHE
118
PHE
107
PHE
107
PHE
107
PHE
118
PHE
118
ASN
297
2.3 Å
3-4 Å
ASN
297
HEME
HEME
HEME
Figure 3. Main interactions between CYP2A6 inhibitors and residues at the enzyme active site. Light blue sphere with dotted line, hydrogen bond interaction; light green
sphere, -stacking interaction; grey sphere, heme interaction. (A) Reference compound; (B) Compound 8; C, compound 1. The images are 2-dimensional projections of the
3-dimensional docking poses; the distances are not in correct scale. The average distance between the nitrogen atom and heme Fe was 2.3 Å for 1 and 3–4 Å for 8.
p–p

N. Tani et al. / Bioorg. Med. Chem. 22 (2014) 6655–6664
6659
M)
Table 1
Structures of compounds and IC50 values for coumarin 7-hydroxylation
a
No.
Structure
IC50
(l
M)
Effect of preincubationb
No effect
Ki (l
NH2
Reference compound
0.10 0.04
0.01
S
Benzothiophenes and thienopyridines
O
N
1*
0.30 0.03
0.40 0.03
No effect
No effect
0.07
0.06
S
S
O
2*
N
Cl
NH2
3
4
0.40 0.1
No effect
No effect
0.09
ND
S
NH
0.50 0.06
S
NH2
Br
5
0.70 0.1
No effect
0.2
S
O
N
6
0.90 0.4
1.0 0.6
No effect
No effect
ND
ND
S
NH2
7*
S
N
N
N
8*
1.4 0.2
No effect
0.6
S
HN
9*
1.5 0.3
2.2 0.8
No effect
No effect
ND
0.8
S
NH
10
S
N
O
11
12
3.2 1.5
3.4 0.4
0.8;
ND
ND
S
O
No effect
S
O
Br
13
5.0
5.2
0.5;
ND
ND
S
NH2
14*
No effect
N
N
S
H2N
S
15
6.6 1.6
No effect
ND
S
(continued on next page)

6660
N. Tani et al. / Bioorg. Med. Chem. 22 (2014) 6655–6664
Table 1 (continued)
a
No.
Structure
IC50
(lM)
Effect of preincubationb
Ki (lM)
O
16
17
8.0 1.4
0.4 0.05;
ND
S
N
S
11 1.1
No effect
ND
O
OH
18
19
18 1.6
37 4.2
No effect
No effect
ND
ND
S
OH
S
N
Cl
20
41 17
No effect
ND
S
N
Cl
21
22
51 15
88 10
No effect
No effect
ND
ND
S
Cl
S
O
Cl
23
270
ND
ND
S
Benzofurans
O
24
0.40 0.09
1.6 0.3
0.05;
0.6;
0.07
0.1
O
O
25
N
O
26
27
17 25
52 13
ND
ND
ND
ND
O
O
N
Br
Br
O
28
29
72 33
<100
ND
ND
ND
ND
O
OH
O

N. Tani et al. / Bioorg. Med. Chem. 22 (2014) 6655–6664
6661
M)
Table 1 (continued)
a
No.
Structure
IC50
(l
M)
Effect of preincubationb
ND
Ki (l
Br
NH2
30
31
<100
ND
O
O
<100
ND
ND
ND
ND
O
32
140 60
Br
O
Indoles
33
O
O
0.40 0.06
0.68 0.68
No effect
No effect
0.07
ND
N
H
34
N
35
36
1.5 0.2
7.4 0.6
No effect
No effect
ND
ND
O
N
H
Br
N
H
O
37
38
8.4 2.0
11 1.5
No effect
ND
ND
N
O
1.8;
N
H
Br
O
39
12 1.8
No effect
8.6
N
H
H2N
40
41
34 8.0
57 8.0
No effect
No effect
ND
3.3
N
H
O
N
H
Br
O
42
79 18
No effect
11
N
H
Others
N
43
7.4 1.5
ND
ND
(continued on next page)

6662
N. Tani et al. / Bioorg. Med. Chem. 22 (2014) 6655–6664
Table 1 (continued)
a
No.
Structure
IC50
(lM)
Effect of preincubationb
Ki (lM)
44
8.1 2.1
ND
ND
N
ND, not determined.
*
Novel synthesized compounds.
a
10
lM coumarin concentration.
b
30-minute preincubation, ; = IC50 shifts down.
24, 33, and the alkyl amine (primary and secondary) thiophene
derivatives, for example, 3 and 4. A formyl substitution at position
two increased the inhibition potency (1, 2) compared with the
amine thienopyridines (7, 8, 9, 14).
The Ki values for the benzothiophene alkyl amine derivatives 5 and
8 and the nitrile derivative 25 were similar for nicotine oxidation
and coumarin 7-hydroxylation. In contrast, the Ki values were
more than 15-fold higher for nicotine oxidation for the compounds
with a formyl substituent (1, 2, 24, 33).
A crucial difference between these two assays is that cytosolic
fraction containing aldehyde oxidase is included in the nicotine
oxidation assay to mediate the formation of cotinine from the nic-
otine iminium ion, while the cytosolic fraction is not included in
the coumarin 7-hydroxylation assay. To test the effect of cytosolic
The IC50 shift assay was performed to detect potential irrevers-
ible inhibition (Fig. 2). A 30-minute preincubation decreased IC50 of
compounds 11, 13, 16, 24, 25 and 38, all containing formyl or
nitrile substituents. Preincubation did not affect IC50 of the other
compounds. To test whether these compounds inactivate the
enzyme by binding covalently to the active site, further studies
(preincubation followed by dilution) with recombinant CYP2A6
enzyme were performed (Fig. 2). In irreversible inhibition, the
inactivation of enzyme is dependent on time, substrate concentra-
tion and NADPH. Such a trend was detected for compounds 16, 24,
enzymes on the stability of the inhibitors, up to 100
liver cytosolic fraction was added to the reaction mixture of the
coumarin 7-hydroxylation assay. Inclusion of 100 l cytosol in
ll of mouse
l
the assay totally abolished the CYP2A6 inhibition capacity of the
compounds with formyl substituent (data not shown).
25 and 38. The Kinact values ranged from 0.08 to 0.22 minꢀ1
whereas the Ki values were 2.4–22 M for these compounds.
,
l
For compounds 11 and 13 the progress curves of the catalytic
reaction did not display a simple linear product-versus-time rela-
tionship when the reaction was initiated by enzyme (data not
shown). This means that the equilibrium between the free enzyme
and enzyme-bound forms becomes established slowly, and calcu-
lation of inhibitor affinity values by classical steady-state methods
is inappropriate, since they significantly underestimate true inhibi-
tion potency. Therefore, the Kiapp values of slow-binding inhibitors
4. Discussion
The main aim of this study was to identify novel potent inhib-
itors of CYP2A6 based on the previously detected scaffolds.18 The
previously identified lead compound benzo[b]thiophen-2-ylme-
thanamine is not suitable for further development due to its reac-
tive and irritating properties. The inhibitors of CYP2A6 discovered
thus far are relatively small, planar molecules with one or two
acceptor atoms.27,28 In this study, the most potent novel CYP2A6
inhibitors were thienopyridine derivatives with hydrogen bond
acceptor atoms either in the core structure and/or in the substitu-
ent. These inhibitors are among the most potent identified to
date.13–16,23,29–31 A nitrogen atom in the core of the formyl thio-
phenes at positions five and six (1, 2) was favorable for CYP2A6
inhibition. The docking analysis showed that the compounds with
a formyl group (1, 2) had an optimal orientation interacting with
Asn297, Phe 107 and Phe118. Several docking poses of the formyl
substituted indan and furan derivatives included a hydrogen bond
with Asn297. In contrast, most of the docking poses of the thieno-
pyridine derivatives with an alkyl amine substituent were without
hydrogen bonding indicating a more flexible orientation (Fig. 4).
Most compounds that inhibit CYP enzymes are transient revers-
ible competitive inhibitors forming non-covalent bonds with the
enzyme. In contrast, irreversible CYP inhibitors are mechanism
based and cause longer duration inhibition. Because covalent bonds
are formed between the inhibitor and the enzyme, the recovery of
enzyme activity requires the synthesis of new enzyme. Irreversible
inhibitors contain specific substructures that are transformed to
reactive electrophilic functional groups in the enzyme active site,
reacting with nucleophilic amino acid residues or heme.32
were determined. The Kiapp values were 0.13
and 13, respectively.
lM and 1.5 lM for 11
3.3. Inhibition of nicotine oxidation
Individual coumarin 7-hydoxylation inhibitory compounds
were selected from all major cores for testing in the in vitro nico-
tine oxidation inhibition assay. These compounds included three
synthesized compounds (1, 2, 8) and the most potent inhibitors
of coumarin 7-hydroxylation in each core (3, 5, 24, 25, 33). These
compounds had formyl, alkyl amine or nitrile substituents. Com-
pounds 1, 2, 3, 8, 24 and 33 were competitive inhibitors of nicotine
oxidation, whereas 5 and 25 were mixed type inhibitors (Table 2).
Table 2
Inhibition of nicotine hydroxylation by selected compounds
Compound Nicotine
oxidation Kia
M)
Coumarin 7-
hydroxylation Ki
(lM)
Type of inhibition for
nicotine oxidation
(l
1
2
3
5
8
24
25
33
1.1
4.2
6.0
0.8
0.5
5.5
0.2
1.8
0.07
0.06
0.09
0.2
0.2
0.07
0.2
Competitive
Competitive
Mixed
The coumarin 7-hydroxylation assay is
a standard high
Mixed
throughput screening method to evaluate CYP2A6 inhibition. As
our goal was to identify compounds that particularly inhibit
nicotine metabolism, the most potent coumarin 7-hydroxylation
inhibitors were subsequently tested in the nicotine oxidation
assay. The formyl thiophene derivatives did not inhibit nicotine
metabolism although they were very potent inhibitors of coumarin
Competitive
Competitive
Irreversible
Competitive
0.07
a
Assay using CYP2A6 supersomes with two nicotine concentration (30 and
M).
300
l

N. Tani et al. / Bioorg. Med. Chem. 22 (2014) 6655–6664
6663
Figure 5. Docking pose of the irreversible inhibitor 25 (orange) in the CYP2A6
active site. Blue line indicates the average distance between the functional nitrogen
atom and heme Fe.
Figure 4. Docking poses of compounds 1 (magenta) and 8 (orange) in the CYP2A6
active site; dotted line indicates hydrogen bonding between 1 and Asn297.
The alkyl amine derivatives 5 and 8 were potent reversible
inhibitors and the nitrile derivative 25 was an irreversible inhibitor
of both nicotine and coumarin oxidations. They were also stable in
the presence of cytosolic enzymes. We will next investigate the
inhibitory properties of these compounds for nicotine metabolism
in the mouse in vivo model.21,22,36
7-hydroxylation. The nicotine assay contains cytosol as a source of
aldehyde oxidase. Inclusion of cytosol to the coumarin 7-hydroxyl-
ation assay totally abolished the inhibition by the formyl thio-
phene derivatives. This indicated that cytosolic enzymes
(possibly aldehyde dehydrogenases) metabolize formyl groups
into carboxylic acids, greatly decreasing their inhibitory potency.
Thus, further development of formyl thiophene derivatives as
in vivo inhibitors would likely be futile.
Compounds 16, 24, 25 and 38 were found to be irreversible
inhibitors of CYP2A6 with coumarin as the substrate. In the thieno-
pyridines the position of the nitrogen atom affected the inhibition
mechanism (e.g. 16 vs 1). In 16 the nitrogen at position 7 caused
slow binding inhibition whereas 1, with nitrogen at position 5,
was a reversible inhibitor. For the benzofuran derivatives, the inhi-
bition mechanism was affected by inclusion of a formyl group at
position five (24). Since the formyl derivatives turned out to be
unstable when exposed to cytosolic enzymes, we will use the infor-
mation obtained about their structure–activity relationship to
design more stable derivatives with similar inhibitory properties.
Replacement of the formyl substituent (24) at position five with
nitrile (25) produced potent inhibition of CYP2A6 in both the cou-
marin and nicotine assays. Docking analysis showed that the furan
oxygen in 24 and the nitrile in 25 coordinated towards heme
(Fig. 5). The functional N-atom of the nitrile group was 1.7 Å from
the heme Fe; the oxygen atom of furan was somewhat further
away (3.1 Å).
5. Conclusions
We evaluated CYP2A6 inhibition by chemically diverse com-
pounds using benzothiophene, benzofuran, indole and thienopyri-
dine cores. Formyl and alkyl amine derivatives of these cores were
identified as potent inhibitors of CYP2A6 in vitro. The formyl deriv-
atives are less promising as lead compounds than amine derivatives,
as they are labile when exposed to cytosolic enzymes. Compounds 8
(N,N-dimethyl-1-(thieno[3,2-c]pyridine-2-yl)methanamine) and
25 (benzofuran-5-carbonitrile) efficiently inhibited nicotine
metabolism in vitro, and will be characterized further.
Acknowledgements
This research was funded by the Academy of Finland (Grant
118434) and supported by the University of Eastern Finland. We
acknowledge the support of the Endowed Chair in Addiction for
the Department of Psychiatry, CIHR Grant MOP86471 and TMH-
109787, the Campbell Family Mental Health Research Institute,
Centre for Addiction and Mental Health and the Canada Foundation
for Innovation (#20289 and #16014), the CAMH Foundation and
the Ontario Ministry of Research and Innovation. We also thank
Hannele Jaatinen for expert technical assistance and Dr. Ewen Mac-
Donald for revising the language of the manuscript.
Previous studies have also shown differences in the CYP2A6
inhibitory effect of some neurotransmitter and steroid hormones33
and isoflavones34 depending on whether coumarin or nicotine was
used as a substrate. Thus, the inhibitory characteristics of a com-
pound may vary depending on the substrate used, a well-known
phenomenon for other CYP enzymes, especially CYP3A4.35 In the
two-step reaction from nicotine to cotinine, some endogenous
compounds inhibit the first step, CYP2A6-dependent nicotine-
References and notes
1. Benowitz, N. L. N. Engl. J. Med. 2010, 362, 2295.
2. Chandler, M. A.; Rennard, S. I. Chest 2010, 137, 428.
3. Benowitz, N. L. Clin. Pharmacol. Ther. 2008, 83, 531.
4. Hukkanen, J.; Jacob, P., 3rd; Benowitz, N. L. Clin. Pharmacol. Ther. 2006, 80, 522.
5. Schoedel, K. A.; Hoffmann, E. B.; Rao, Y.; Sellers, E. M.; Tyndale, R. F.
Pharmacogenetics 2004, 14, 615.
D
50(10)-iminium ion formation, while others inhibit the second
step, cotinine formation from nicotine-
lyzed by cytosolic aldehyde oxidase.33
D
50(10)-iminium ion cata-

6664
N. Tani et al. / Bioorg. Med. Chem. 22 (2014) 6655–6664
6. Ho, M. K.; Mwenifumbo, J. C.; Al Koudsi, N.; Okuyemi, K. S.; Ahluwalia, J. S.;
Benowitz, N. L.; Tyndale, R. F. Clin. Pharmacol. Ther. 2009, 85, 635.
7. Patterson, F.; Schnoll, R. A.; Wileyto, E. P.; Pinto, A.; Epstein, L. H.; Shields, P. G.;
Hawk, L. W.; Tyndale, R. F.; Benowitz, N.; Lerman, C. Clin. Pharmacol. Ther. 2008,
84, 320.
8. Sellers, E. M.; Kaplan, H. L.; Tyndale, R. F. Clin. Pharmacol. Ther. 2000, 68, 35.
9. Tyndale, R. F.; Sellers, E. M. Ther. Drug Monit. 2002, 24, 163.
10. Tyndale, R. F.; Sellers, E. M. Drug Metab. Dispos. 2001, 29, 548.
11. Siu, E. C.; Tyndale, R. F. Annu. Rev. Pharmacol. Toxicol. 2007, 47, 541.
12. Honkakoski, P.; Maenpaa, J.; Leikola, J.; Pasanen, M.; Juvonen, R.; Lang, M. A.;
Pelkonen, O.; Raunio, H. Pharmacol. Toxicol. 1993, 72, 107.
21. Siu, E. C.; Tyndale, R. F. J. Pharmacol. Exp. Ther. 2008, 324, 992.
22. Siu, E. C.; Wildenauer, D. B.; Tyndale, R. F. Psychopharmacology 2006, 184, 401.
23. Yano, J. K.; Denton, T. T.; Cerny, M. A.; Zhang, X.; Johnson, E. F.; Cashman, J. R. J.
Med. Chem. 2006, 49, 6987.
24. SYBYL-X 1.2., T. I.: 1699 South Hanley Rd., St. Louis, Missouri, 63144, USA.
25. Jain, A. N. J. Med. Chem. 2003, 46, 499.
26. Clark, R. D.; Strizhev, A.; Leonard, J. M.; Blake, J. F.; Matthew, J. B. J. Mol. Graph.
Model. 2002, 20, 281.
27. Lewis, D. F. V.; Ito, Y. Xenobiotica 2009, 39, 625.
28. Juvonen, R. O.; Gynther, J.; Pasanen, M.; Alhava, E.; Poso, A. Xenobiotica 2000,
30, 81.
13. Draper, A. J.; Madan, A.; Parkinson, A. Arch. Biochem. Biophys. 1997, 341, 47.
14. Zhang, W.; Kilicarslan, T.; Tyndale, R. F.; Sellers, E. M. Drug Metab. Dispos. 2001,
29, 897.
15. Di, Y. M.; Chow, V. D.; Yang, L. P.; Zhou, S. F. Curr. Drug Metab. 2009, 10, 754.
16. Raunio, H.; Rahnasto-Rilla, M. Drug Metabol. Drug Interact. 2012, 2, 73.
17. Rahnasto, M.; Wittekindt, C.; Juvonen, R. O.; Turpeinen, M.; Petsalo, A.;
Pelkonen, O.; Poso, A.; Stahl, G.; Holtje, H. D.; Raunio, H. Pharmacogenomics J.
2008, 8, 328.
29. Maenpaa, J.; Sigusch, H.; Raunio, H.; Syngelma, T.; Vuorela, P.; Vuorela, H.;
Pelkonen, O. Biochem. Pharmacol. 1993, 45, 1035.
30. Taavitsainen, P.; Juvonen, R.; Pelkonen, O. Drug Metab. Dispos. 2001, 29, 217.
31. Kinonen, T.; Pasanen, M.; Gynther, J.; Poso, A.; Jarvinen, T.; Alhava, E.; Juvonen,
R. O. Br. J. Pharmacol. 1995, 116, 2625.
32. Fowler, S.; Zhang, H. J. AAPS J. 2008, 10, 410.
33. Higashi, E.; Nakajima, M.; Katoh, M.; Tokudome, S.; Yokoi, T. Drug Metab.
Dispos. 2007, 35, 508.
18. Rahnasto, M. K.; Raunio, H. A.; Wittekindt, C.; Salminen, K. A.; Leppanen, J.;
Juvonen, R. O.; Poso, A.; Lahtela-Kakkonen, M. K. Bioorg. Med. Chem. 2011, 19,
7186.
19. Rahnasto, M.; Raunio, H.; Poso, A.; Juvonen, R. O. Xenobiotica 2003, 33, 529.
20. Aitio, A. Anal. Biochem. 1978, 85, 488.
34. Nakajima, M.; Itoh, M.; Yamanaka, H.; Fukami, T.; Tokudome, S.; Yamamoto, Y.;
Yamamoto, H.; Yokoi, T. J. Clin. Pharmacol. 2006, 46, 337.
35. Hutzler, J. M.; Tracy, T. S. Drug Metab. Dispos. 2002, 30, 355.
36. Siu, E. C.; Tyndale, R. F. Mol. Pharmacol. 2007, 71, 826.