New potent human acetylcholinesterase inhibitors in the tetracyclic triterpene series with inhibitory potency on amyloid β aggregation

Article abstract of DOI:10.1016/j.ejmech.2011.02.073

New acetylcholinesterase inhibitors in the tetracyclic triterpene series were synthesized, tested in vitro for the inhibition of cholinesterases (different sources of AChE and BuChE) and for the ability to prevent AChE-induced Aβ aggregation. Some compoun

Full text of DOI:10.1016/j.ejmech.2011.02.073

European Journal of Medicinal Chemistry 46 (2011) 2193e2205
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
New potent human acetylcholinesterase inhibitors in the tetracyclic triterpene
series with inhibitory potency on amyloid
b aggregation
*
Julien Rouleau, Bogdan I. Iorga, Catherine Guillou
Institut de Chimie des Substances Naturelles, Bt 27, CNRS Avenue de la Terrasse, 91198 Gif-sur-Yvette, France
a r t i c l e i n f o
a b s t r a c t
Article history:
New acetylcholinesterase inhibitors in the tetracyclic triterpene series were synthesized, tested in vitro
Received 10 November 2010
Received in revised form
24 February 2011
Accepted 26 February 2011
Available online 8 March 2011
for the inhibition of cholinesterases (different sources of AChE and BuChE) and for the ability to prevent
AChE-induced A
showed ability to block the AChE-induced A
b
aggregation. Some compounds have hAChE IC50 values in the nanomolar range and
aggregation. The mode of interaction between EeAChE and
b
compounds 1 and 36e was investigated using docking and molecular dynamics simulations. These
studies suggested that both compounds interact simultaneously with the catalytic and the peripheral
sites of AChE, and the nature of protein-ligand interactions is mainly hydrophobic.
Ó 2011 Elsevier Masson SAS. All rights reserved.
Keywords:
Acetylcholinesterase inhibitors
N-3-Isobutyrylcycloxobuxidine-F
derivatives
Molecular modeling
Alzheimer’s disease
b
-amyloid aggregation
Butyrylcholinesterase
Synthesis
1. Introduction
addition to its catalytic activity, AChE exerts secondary non-
cholinergic functions related to its peripheral binding site on
Alzheimer disease (AD) is a progressive neurodegenerative
disorder of the central nervous system (CNS) that affects mainly
aged population. AD is characterized by profound memory
impairments, emotional disturbance, and also personality changes.
The main pathological changes in the AD brain are extracellular
amyloid plaques [1], intracellular neurofibrillary tangles containing
abnormally hyperphosphorylated tau protein [2], and loss of
neurons in the nucleus basalis of Meynert and the hippocampus.
AD is characterized by a pronounced alteration of the cholinergic
system and other neurotransmitter systems (glutamate and sero-
tonine). The cholinergic hypothesis postulates that memory
impairments in patients with AD result from a deficit of cholinergic
function in the brain [3]. Most currently prescribed AD drugs aim to
increase the level of acetylcholine (ACh) in the brain by inhibiting
acetylcholinesterase (AChE). However, clinical use of AChE inhibi-
tors is sometimes limited mainly due to their adverse effects and
modest benefits to AD patients. Therefore, novel more effective
therapies, including AChE inhibitors, need to be developed. In
differentiation, cell adhesion, in mediating the processing and
deposition of -amyloid peptide (A ) [2]. It was postulated that
AChE binds through its peripheral site to the non amyloidogenic
form of -amyloid protein acting as a chaperone protein and
inducing conformational change to the amyloidogenic form with
the subsequent amyloid fibril formation. Moreover, it has been
shown that molecules which are able to interact with both the
active and peripheral sites of AChE could prevent the aggregating
b
b
b
activity of AChE toward Ab besides the inhibitory activity [3e5].
Therefore, inhibitors with dual binding to AChE represent a new
therapeutic strategic option [6,7].
In continuation of our on-going research on new AChE inhibitors
[8], we will present our recent studies on acetylcholinesterase
inhibitor in the tetracyclic triterpene series. We previously reported
that compound 1 was identified by high-throughput screening of
The Institut de Chimie des Substances Naturelles’ chemical library.
Some hemisynthetic analogs of 1 (i.e. 2a and 2b) were found
to be very active on Electrophorus electricus acetylcholinesterase
(EeAChE) and Torpedo californica acetylcholinesterase (TcAChE) but
less potent on human acetylcholinesterase (hAChE) (Fig. 1). The aim
of the present study is to develop new potent hAChE inhibitors in
this series.
* Corresponding author. Tel.: þ33 1 69 82 30 75; fax: þ33 1 69 07 72 47.
E-mail address: catherine.guillou@icsn.cnrs-gif.fr (C. Guillou).
0223-5234/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2011.02.073

2194
J. Rouleau et al. / European Journal of Medicinal Chemistry 46 (2011) 2193e2205
21
20
18
N
N
O
O
O
H
12
13
14
17
16
15
11
19
OH
H
8
H
OH
C
1
4
N
9
H
2
10
5
O
H
D
A
3
7
28
B
6
N
N
OH
N
H
HO
H
H
30
1'
29
R
N
H
O
2a : R1 = iPr
2b: R1 = (S)-(CH3)CH(C2H5)
3 : N-3-isobutyrylcycloxobuxidine-F
1: oxazine
Fig. 1. Structures of compounds 1e3.
Herein we describe the synthesis, pharmacological evaluations
and molecular modeling of a novel class of analogs.
heating in methanol, and treatment under acid conditions provided
a 90/10 mixture of diamine 18 (major) and the amino amide 19
(minor). Finally, the dihydro-1H-imidazole 20 was prepared by
reaction of 18 and methyl orthoisobutyrate in the presence of ytter-
bium triflate (Scheme 3).
The dihydropyrimidine 25 was synthesized from cycloxob-
uxidine-F 4. First, the primary amine was protected with tert-butyl
dicarbonate (i.e. 21) and the 16-hydroxy function was transformed
into a benzoate (i.e. 22). Oxidation of the alcohol group at position 29
with Dess-Martin periodinane afforded the corresponding aldehyde
23. Aldolisation of 23 with 3-methyl-2-butanone led to the keto
alcohol 24. Subsequent cyclization under acid conditions followed by
deprotection of the benzoate group furnished the expected dihy-
dropyridine 25 (Scheme 4).
These new inhibitors contain a modified heterocycle which is
supposed to interact with the catalytic site and new substituents
connected to the nitrogen atom N 20, which is supposed to be
responsible for the binding to the peripheral site of the enzyme.
Newly synthesized compounds were tested in vitro for AChE and
BuChE inhibition potency. Their selectivity for hAChE was also
evaluated. The ability of compounds 36e, 36f, 36i, and 36k to
inhibit the AChE-induced A
b aggregation compared with propi-
dium iodide and tacrine was assayed by means of a thioflavin T-
based fluorometric assay [9]. A molecular modeling study (docking
and molecular dynamics simulations) of 1 and 36e was also per-
formed on EeAChE.
The tetrahydropyridine 27 was synthesized from the keto
alcohol 24. Dehydratation of 24 was first performed using potas-
2. Chemistry
sium carbonate to afford the a,b-unsaturated ketone 26. The tet-
Introduction of fluorine atoms on a molecule can modify its
properties [10]. In addition, the presence of a triflurometyl group on
m-(N,N,N,-trimethylammonio) trifluoroacetophenone (TMTFA) or
zifrosilone greatly improved their acetylcholinesterase inhibition
potency [11]. Replacement of the isopropyl group present in the 10
position of the triterpene tetracyclic scaffold of 1 by other substit-
uents bearing one, three or seven fluorine atoms was first envis-
aged. The new fluoro analogs 9aec were synthesized from
cyclobuxidine-F 4 obtained by hydrolysis of N-3-isobutyryl-cyclo-
xobuxidine-F 3 (Scheme 1) [12]. Preparation of amides 6aec was
carried out by coupling trifluoroacetic anhydride or 2-fluoro-2-
methylpropanoic acid or methyl heptafluoroisobutyrate with
cycloxobuxidine-F 4. The 16-hydroxy function was first protected as
an acetate (i.e. 7aeb) or as a benzoate (i.e. 7c). The nitrogen atom at
position 20 was protected as a salt by reaction with APTS before the
transformation of the alcohol group at position 29 to the corre-
sponding tosylates (i.e. 8aec). Then the acetate and benzoate
protective groups were then removed under basic conditions to
afford compounds 8aec. Finally, treatment of the resulting tosy-
lates with sodium diformylamide in DMF provided the expected
fluoro oxazines 9aec (Scheme 1).
rahydropyridine 27 was obtained after hydrogenation of 26 over
palladium hydroxide and cyclization under acid conditions in the
presence of trifluoroacetic acid (Scheme 5).
The synthesis of compound 30 was achieved following the
strategy summarized in Scheme 6. Reaction of the known tosylate
28 [13] with potassium thioacetate afforded a thioacetate inter-
mediate that was directly hydrolyzed under basic conditions to
provide the thiol 29. The 5,6-dihydro-4H-1,3-thiazine 30 was
obtained by cyclization of 29 in the presence of trimethylalumi-
nium (Scheme 6).
Synthetic routes to N-substituted thiazine analogs 35aeb and
carbamate 35c were also developed. The alcohol group at position
29 of 14 was transformed into the corresponding tosylate (i.e. 31)
and then substituted with potassium thioacetate to give the cor-
responding thioacetate. This later was converted into its N-oxide by
oxidation with m-chloroperbenzoic acid. Subsequent treatment of
this N-oxide with a mixture of ferrous sulfate heptahydrate and
iron chloride hexahydrate provided N-demethylated amine 32.
Hydrolysis of the thioacetate 32 provided the thiol 33 which was
cyclized in the presence of trimethylaluminium (i.e. 34). Acylation
or alkylation of the amine 34 furnished the expected amides 35aec
and the tertiary amines 36aek (Scheme 7).
The N-substituted tetrahydropyrimidines derivatives 13aeb
were synthesized following the reactions depicted in Scheme 2
using the aldehyde 10 [8] as key intermediate, which was con-
verted into the corresponding amines 12aeb by aminoreduction.
Subsequent heating under microwaves irradiation furnished 13a
and 13b having an open cyclopropane ring (Scheme 2).
3. Results and discussion
3.1. AChE and BuChE inhibitory activity and AChE/BuChE selectivity
The new dihydro-1H-imidazole analog 20 was synthesized from
N-3-isobutyrylcycloxobuxidine-F 3. The 16-hydroxy function was
first protected as a benzoyl (i.e.14). Oxidation of the alcohol group at
position 29 with Jones’ reagent afforded the acid (i.e.15). Reaction of
15 with oxalyl chloride provided the corresponding acid chloride
which reacted with sodium azide to give the carbonyl azide 16.
Curtius rearrangement of 16 under microwaves irradiation furnished
the isocyanate 17. Subsequent hydrolysis with lithium hydroxide,
To determine the potential interest of the newly synthesized
compounds for the treatment of AD, their AChE inhibitory activity
was assayed by the method of Ellman [14] using E. electricus and
human recombinant AChE and tacrine, 1 and 3 as reference
compounds. Moreover, to further study of the biological profiles of
the novel compounds, their butyrylcholinesterase (BuChE) inhibi-
tory activity on human serum butyrylcholinesterase (hBuChE) was
also determined by the same method. Recent studies have shown

J. Rouleau et al. / European Journal of Medicinal Chemistry 46 (2011) 2193e2205
2195
N
N
O
H
O
H
H
OH
H
OH
a
O
N
H
HO
H2N
HO
4
(94%)
H
3
: N-3-isobutyrylcycloxobuxidine-F
N
O
O
OH
b or c or d
4
R2
O
R1
O
R1
N
H
H
HO
5a R1 = CF3; R2 = COCF3
5b R1 = (CH3)2CF; R2 = H
5c R1 = (CF3)2CF; R2 = OCH3
6a R1 = CF3 (84%)
6b R1 = (CH3)2CF (86%)
6c R1 = (CF3)2CF (73%)
N
N
O
O
e or f
g
H
OR3
H
OH
O
O
R1
N
H
R1
N
H
H
H
HO
TsO
7a R1 = CF3; R3 = COCH3 (72%)
7b R1 = (CH3)2CF; R3 = COCH3(86%)
7c R1 = (CF3)2CF; R3 = COPh (95%)
8a R1 = CF3 (50%)
8b R1 = (CH3)2CF (66%)
8c R1 = (CF3)2CF (31%)
N
OH
O
H
h
N
H
R1
O
9a R1 = CF3 (68%)
9b R1 = (CH3)CF (32%)
9c R1 = (CF3)2CF (14%)
Scheme 1. Synthesis of oxazines 9aec.
that in AD patients with severe pathology, BuChE increases while
AChE is reduced in specific brain regions [15] this is not in agree-
ment with another recent study which shows a decrease of BuChE
activity in vivo [16]. The results are summarized in Table 1.
The inhibitory potency depends on numerous factors. Intro-
duction of fluoro groups at position 10 of the oxazine scaffold (i.e.
9ae9c) decreased the inhibitory activity for AChE and BuChE.
Replacement of the oxazine ring by different heterocycle rings
affected the affinity for EeAChE and hAChE. An order of the inhib-
itory potency for AChE can be established among the various
heterocycles: alkylated tetrahydropyrimidines (i.e. 13aeb) z im-
idazole (i.e. 20) < tetrahydropyrimidine (i.e. 27) < thiazine (i.e.
30) < dihydropyrimidine (i.e. 25). The dihydropyrimidine 25 is the
most potent inhibitor in these series (12 fold more active than 1)
and selective for hAChE. Compounds 13aeb and 20 are more potent
on BuChE than on AChE. Thus, alkylation of the nitrogen atom of the
tetrahydropyrimidine ring (i.e. 13a and 13b) increases the selec-
tivity for BuChE (i.e. 2a).
The analogs of second generation with modified nitrogen atom
at position 20 were also evaluated. This series was envisaged in
order to increase the binding of the inhibitors with the peripheral
site of AChE and thereby increasing the inhibitory potency.
N-demethylation of the nitrogen atom of the thiazine (i.e. 34,
hAChE IC50 ¼ 76 nM) produced no significant change on AChE

2196
J. Rouleau et al. / European Journal of Medicinal Chemistry 46 (2011) 2193e2205
N
N
OH
RNH2
O
O
H
OH
H
O
O
a
N
H
N
H
H
H
O
HN
R
H
11a : R = Me
11b: R = Et
10
12a : R = Me (90 %)
12b : R = Et (85%)
c
b
O
N
O
H
N
H
OH
H
N
OH
H
N
H
N
N
13a (80%)
13b (40%)
Scheme 2. Synthesis of tetrahydropyrimidines 13aeb.
inhibition (i.e. 30, hAChE IC50 ¼ 60 nM) but decreased the selec-
tivity for AChE (i.e. 34, 17; i.e. 30, 89).
was fully flexible. No hydrogen bond could be identified between
oxazine 1 and the protein, and only one was present in the AChE-
36e complex, between OH group in position 16 and OH group of
TYR70. Furthermore, in order to take into account the protein
flexibility and a possible induced-fit that could take place upon
ligand binding, molecular dynamics simulations were carried out,
using the docking results as starting conformations.
Three molecular dynamics simulations, of 10 ns each, were set
up, one with the apo protein (Sim1), for comparison purposes, and
the other two with oxazine 1 (Sim2) and compound 36e (Sim3). All
three structures were generally stable during the simulation, the
root mean square deviation (RMSD) for protein reaching a plateau
in less than 5 ns, at 1.8 Å for Sim3 and at about 2 Å for Sim1 and
Sim2. In Sim2 and Sim3, the ligands attained very quickly an
equilibrium state with an RMSD of 0.7 Å, which was conserved
throughout the simulation (Figure S1, Supporting Information). All
simulations showed three main regions with higher flexibility
(around residues 73, 341 and 378), but none of them were in the
binding site region (Figure S2, Supporting Information).
In the amide series (i.e. 35ae35c) introduction of a phenyl
group (i.e. 35b) increased the inhibition activity likely as a result of
a better interaction with the peripheral site of the enzyme prob-
ably via hydrophobic interactions. We previously reported that
compound 1 was able to bind simultaneously to both catalytic and
peripheral sites of AChE. Its oxazine scaffold interacted with the
active site and the nitrogen atom at position 20 with the periph-
eral site [8].
The AChE inhibitory potency within the N-alkylated series of
thiazine derivatives is related to the substituent, to the substitution
of R and to the length of the linker between R and the nitrogen
atom. The inhibition activity follows this order: (CH2)2Bn (i.e.
36c) < (CH2)2Pht (i.e. 36j) < CH2Bn (i.e. 36b) < (CH2)3Pht (i.e.
36k) < o-CF3Bn (i.e. 36g) < p-MeOBn (i.e. 36f) < Bn (i.e. 36a) z o-
MeOBn (i.e. 36d) < p-CF3Bn (i.e. 36i) z m-CF3Bn (i.e. 36h) z m-
MeOBn (i.e. 36e). Introduction of a CF3 substituent on the benzyl
ring increased the selectivity for hAChE versus BuChE.
Generally speaking, very few proteineligand hydrogen bonds
were identified in Sim2 and Sim3, the interactions being mostly
hydrophobic. In Sim2, only the interaction between the OH group
in position 16 of the steroid skeleton and the OH group of TYR70
was scarcely observed during the simulation (less than 10% of the
simulation time). In Sim3, three hydrogen bonds were present all
along the simulation: i) the same interaction mentioned above, for
about 50% of the simulation time; ii) a hydrogen bond between the
OH group in position 16 of compound 36e and the carboxyl group of
ASP72, present all the time; iii) an interaction between the oxygen
in position 11 of steroid and the OH group of TYR121, present for
about 25% of the simulation time. Therefore, it seems that
compound 36e can establish better interactions than oxazine 1
within the binding site of AChE and these interactions may be
3.2. Molecular modeling studies
In order to gain insight into the mode of interaction between this
new class of inhibitors and AChE, molecular modeling studies were
performed using the compound 36e, the most potent in the series, as
a representative member. The results were compared with those
obtained for the apo AChE and for the complex with oxazine 1.
Molecular docking calculations showed that compound 36e is
positioned in a conformation similar to those proposed previously
for oxazine 1 [8], and that it is able to interact with both the cata-
lytic and peripheral sites of AChE. The binding site was defined as
a sphere with a radius of 20 Å around the OH group of SER200 and
the protein was kept rigid during the docking, whereas the ligand

J. Rouleau et al. / European Journal of Medicinal Chemistry 46 (2011) 2193e2205
2197
order to accommodate its presence and to maximize the hydro-
phobic interactions. Unexpectedly, no catione
and interac-
tions could be evidenced between the TRP279 at the peripheral site
and the positively charged nitrogen in position 20 and/or its
aromatic substituent (in the case of compound 36e).
N
N
p
pep
O
O
H
OBz
H
OBz
O
b
O
a
3
N
H
N
H
HO
H
H
O
HO
O
3.3. Inhibition of AChE-Induced Ab aggregation
14
15
Three compounds, 36e, 36h and 36k, were selected to assess
their abilities to inhibit A aggregation induced by AChE using
a thioflavin T-based fluorometric assay [9], compared with the
reference compound propidium iodide [9], a known specific
peripheral site-binding inhibitor and tacrine a known specific
inhibitor of the active site [9]. The N-substituted thiazines 36e, 36h
N
N
b
O
H
OBz
H
OBz
O
O
d
c
N
N
H
O
H
O
H
O
H
N
C
N3
and 36k inhibit, at 100 mM, the AChE-induced Ab aggregation with
17
16
percentages of inhibition ranging from 42% to 69% (Table 2). By
observation of the data reported in Table 2, it appears that intro-
duction of meta CF3 group on the N-benzyl terminal group (i.e. 36h)
or a phtalimide terminal group (i.e. 36k) decrease the anti aggre-
gating action. In contrast, introduction of a m-OMeBn (i.e. 36e)
favors the anti-aggregating action. The best compound 36e allows
a better positioning of the methoxy aromatic ring with the
peripheral site of the enzyme.
e
N
N
O
+
H
OH
H
OH
O
H2N
N
H
H
H
NH2
NH2
18
19
18/19 : 90/10
Thiazines 36e could be classified as good inhibitor of the AChE-
induced A
inhibit, at 100
b
aggregation. The most potent described inhibitors
M, the AChE-induced aggregation with
f
m
Ab
percentages of inhibition ranging from 82% to 98% [17e21].
N
O
H
4. Conclusion
H
OH
We have synthesized new inhibitors in the tetracyclic triterpene
series. These compounds have been synthesized from the natural
product N-3-isobutyrylclyclobuxidine-F. Variations of the hetero-
cycle led us to identify that presence of a thiazine or a dihy-
dropyrimidine ring enhanced the hAChE inhibitory potency.
Alkylation of the nitrogen atom at position 20 with functionalized
benzyl groups in the thiazine series also increased the hAChE
inhibitory activity. Molecular modeling studies showed that 1 and
36e interact simultaneously with both the catalytic and peripheral
sites of EeAChE, with the nature of proteineligand interactions
being mainly hydrophobic. Futhermore, 36e significantly pre-
N
NH
20
Scheme 3. Synthesis of Dihydro-1H-imidazole 20.
responsible, at least in part, for the good inhibition observed for the
compound 36e.
vented the AChE-induced Ab aggregation. Taken together, all the
The evolution of the proteineligand contact surface during the
simulations Sim2 and Sim3 is shown in Figure S3 (Supporting
Information). Whereas during the first 5 ns the surfaces are
rather comparable, during the last 5 ns the protein interaction with
compound 36e is clearly more favored than those with oxazine 1. A
possible explanation for this behavior may be that AChE better
accommodates the cyclopropyl-derived scaffold of compound 36e
than the seven-membered ring-derived core of oxazine 1. This
result is also in agreement with the lower RMSD observed for
protein in Sim3 (see above) and possibly could explain the good
biological activity measured experimentally for compound 36e.
In the apo simulation (Sim1), the residues forming the binding
site did not show important movements, with the exception of
TRP84, which underwent a 3.6 Å movement in the direction of the
binding site opening, and of the TYR334-PHE330-PHE331 triplet,
which underwent a concerted movement resulting in the complete
closure of the gorge (Fig. 2). Thus, it seems that in the absence of
a ligand this closed conformation is more stable than the open
conformation originally present in the X-ray structure.
reported results suggested that new dual binding AChE inhibitors
in the thiazine tetracyclic triterpene series are promising leads for
the development of disease-modifying drugs for the future treat-
ment of AD.
5. Experimental section
5.1. Chemistry
All commercially available reagents were used without further
purification. All experiments involving water-sensitive compounds
were carried out underargon and scrupulously dryconditions, using
anhydrous solvents. MeOH was distilled from Mg/I2, CH2Cl2 from
P2O5, 1,4-dioxane from LiAlH4, NEt3 and pyridine from KOH. All
separations were carried out under flash chromatographic condi-
tions on Merck silica gel 60 (70e230 mesh) at medium pressure
(200 mbar). TLC was performed on Merck silica gel plates (60F254
)
with a fluorescent indicator. NMR spectra were determined on
Bruker Avance-300, Bruker AC-400, Bruker AC-500 or Bruker-600
instruments and using tetramethylsilane (TMS) relative to TMS.
Chemical shifts are reported in parts per million (ppm) relative to
TMS. High resolution mass spectra (HRMS) were obtained on a mass
maldi-toff spectrometer. The assignment of the 1H and 13C NMR
In both Sim2 and Sim3 simulations, the ligands move about
2.8 Å toward the peripheral site (Figs. 3 and 4), to reach a stable,
common equilibrium position (Figure S4, Supporting Information).
In the meantime, the residues surrounding the ligand move apart in

2198
J. Rouleau et al. / European Journal of Medicinal Chemistry 46 (2011) 2193e2205
N
N
N
O
O
O
H
OH
H
OH
H
OBz
b
a
H2N
BocHN
BocHN
H
H
H
HO
HO
HO
4
21
22
N
N
N
O
O
O
H
OBz
H
OH
H
OBz
d
c
e
BocHN
H
OH
N
BocHN
H
H
H
O
O
23
24
25
Scheme 4. Synthesis of dihydropyridine 25.
N
N
OH
N
O
O
O
H
H
OBz
H
H
OH
a
b
BocHN
BocHN
H
H
N
OH
O
O
24
26
27
Scheme 5. Synthesis of tetrahydropyridine 27.
spectra of the different compounds has been made by 1He1H,
1He13C, COSY, HMBC, HMQC and by analogy with previously des-
cribed compounds. The numbering indicated on molecule is only
mentioned for the NMR spectra interpretation. Infrared (IR) spec-
tra were recorded on a Fourier PerkineElmer Spectrum BX FT-IR
instruments. Elemental analyses were performed by the micro-
analysis laboratory of the ICSN, CNRS, Gif-sur-Yvette.
5.1.1. N-[20S-(dimethylamino)-16
,14 -dimethyl-11-oxo-9,19-cyclo-5
trifluoroacetamide (6a)
To 68 mg (0.158 mmol,1 eq.) of cycloxobuxidine-F 3 solubilized in
DCM/pyridine (mixture 3/1, 4 mL) were added at 0 ꢀC 25
a
-hydroxy-4
a
-(hydroxymethyl)-
4b
a
a
,9 -pregnan-3
b
b
-yl]-
mL
(0.174 mmol, 1.1 eq.) of trifluoroacetic anhydride. The mixture
was stirred at 0 ꢀC for 1 h then at RT for further 12 h. After 3
N
OH
N
N
O
O
O
H
H
OAc
a
H
OH
b
O
O
N
N
H
N
H
H
H
H
S
TsO
HS
28
29
30
Scheme 6. Synthesis of 5,6-dihydro-4H-1,3-thiazine 30.

J. Rouleau et al. / European Journal of Medicinal Chemistry 46 (2011) 2193e2205
2199
N
N
O
O
H
OBz
H
OBz
O
O
a
N
H
N
H
H
H
HO
TsO
14
31
NH
OBz
NH
OH
O
O
H
H
O
O
b
c
N
H
N
H
H
H
AcS
HS
33
32
R
O
NH
OH
N
O
O
H
H
H
OH
d
e
N
N
H
S
S
34
35a R = Me (92%)
35b R = Ph (83%)
35c R = O -Bu (83%)
t
f
R
N
O
H
H
OH
36a R = Bn (93%)
36b R= CH2Bn (84%)
36c R= (CH2)2Bn (72%)
N
36d R =
36e R =
36f R =
36g R =
36h R =
o
-OMeBn (83%)
m
-OMeBn (80%)
S
p
-OMeBn (70%)
o-CF3Bn (55%)
m
-CF3Bn (62%)
36i R = p-CF3Bn (75%)
36j R=(CH2)2Pht (72%)
36k R= (CH2)3Pht (75%)
Scheme 7. Synthesis of compounds 35aec and 36aek.
co-evaporations with 3 ꢁ 10 mL of 1,2-dichloroethane, the residue
was dissolved in DCM, washed with 20 mL of a saturated solution of
sodium hydrogen carbonate then extracted with 3 ꢁ 20 mL of DCM.
Combined organic layers were washed with brine (20 mL), dried
over magnesium sulfate and concentrated under vacuum. The crude
product was purified by silica gel column chromatography (eluent:
DCM/MeOH/NH4OH: 96/3/1) to afford the title compound as an
amorphous white solid (70 mg, yield 84%). IR (cmꢂ1): 3276, 2944,
1.57 (2H, m, H19
(1H, m, H1
),1.22 (3H, s, H28),1.06 (1H, d, Jgem ¼ 3.8 Hz, H19
(4H, m, J21e20 ¼ 6.6 Hz, H21 þ H6 ), 0.86 (3H, s, H18), 0.64 (3H, s,
H30); 13C NMR (75.4 MHz, CDCl3): (ppm) 211.3 (C11), 158.0 (C10),
a
þ H7
b
),1.52 (1H, m, H15
a
),1.43 (1H, m, H7
a
),1.31
a
b), 0.92
b
d
113.9 (C20), 78.0 (C16), 64.2 (C29), 62.4 (C20), 55.8 (C17), 52.4 (C3), 51.4
(C12), 47.1 (C14), 44.6 (C4 þ C13), 43.1 (C15), 41.5 (C8), 41.2 (C5), 37.2
(C10), 34.3 (C9), 30.5 (C19), 27.4 (C1), 27.2 (C2), 24.3 (C7), 20.7 (C28),
18.4 (C6),17.9 (C18),11.2 (C30),10.1 (C21). MS(ESI, m/z): 529.2 (M þ H);
HRMS (ESI, m/z): calcd for C28H44N2O4F3: 529.3253, found: 529.3254.
2872, 1698, 1555, 1454, 1157; 1H NMR (300 MHz, CDCl3):
6.43 (1H, d, JNAHeH3 ¼ 9.2 Hz, NAeH), 4.16 (1H, m, H16 ), 4.12 (1H, m,
H3
), 3.42 (1H, d, Jgem ¼ 12.9 Hz, H29b), 3.03 (1H, d, Jgem ¼ 12.9 Hz,
H29a), 2.70 (1H, dq, J20e17 ¼ 10.8 Hz, J20e21 ¼6.7 Hz, H20), 2.54 (1H,
d (ppm)
b
a
5.1.2. 20S-(dimethylamino)-4
11-oxo-3 -[(trifluoro acetyl) amino]-9,19-cyclo-5
-yl acetate (7a)
To 144 mg (0.27 mmol, 1 eq.) of 6a solubilized in 5 mL of
mixture DCM/pyridine (3/1) were added at
ꢀC 27
a-(hydroxymethyl)-4
b,14
a-dimethyl-
b
a,9b-pregnan-
a
d, Jgem ¼ 17.2 Hz, H12
J1 ), 2.32 (1H, d, Jgem ¼ 17.2 Hz, H12
¼ 3.4 Hz, H1
NBMe2), 2.10 (1H, dd, J5 ¼ 3.6 Hz, H5
¼ 12.5 Hz, J5
a
), 2.50 (1H, ddd, Jgem ¼ 13.2 Hz, J1
¼ 3.6 Hz,
16
a
b
e2
b
b
), 2.31 (6H, bs,
), 2.01 (3H,
), 1.71 (2H, m, H2 ),
þ H6
b
e2
a
a
0
m
L
ae6b
ae6a
m, H17
a
þ H15
b
þ H8
b
), 1.80 (1H, m, H2
a
b
a
(0.286 mmol, 1.05 eq.) of acetic anhydride. The mixture was stirred

2200
J. Rouleau et al. / European Journal of Medicinal Chemistry 46 (2011) 2193e2205
Table 1
AChE and BuChE Inhibitory activities of new analogs.a
Compounds
IC50 (nM)
EeAChE
hAChE
hBuChE
AChE selectivityb
1
2a
9a
9b
31 ꢃ 4c
299c
nd
>1000c
>3
>10 000
>10 000
336 ꢃ 41
1701 ꢃ 260
>10 000
>10 000
>10 000
16 ꢃ 1.4
294 ꢃ 36
38 ꢃ 4
>10 000c
>10 000
>10 000
>10 000
2948 ꢃ 329
3326 ꢃ 143
7224 ꢃ 742
679 ꢃ 6.9
>10 000
5375 ꢃ 430
1343 ꢃ 120
>10 000
>1000
ndd
nd
>10 000
744 ꢃ 103
3003 ꢃ 292
>10 000
>10 000
>10 000
25 ꢃ 1.4
618 ꢃ 87
60 ꢃ 7.9
76 ꢃ 14.3
298 ꢃ 19.2
43 ꢃ 10.0
159 ꢃ 46.7
7 ꢃ 0.9
>13
>3
9c
13a
13b
20
25
27
0.29
0.33
0.72
25
16c
89
17
30
34
45 ꢃ 3.8
139 ꢃ 11.9
37 ꢃ 2.9
186 ꢃ 13.7
13 ꢃ 0.8
14 ꢃ 0.9
79 ꢃ 7.1
12 ꢃ 1.7
6 ꢃ 0.5
35a
35b
35c
36a
36b
36c
36d
36e
36f
36g
36h
36i
36j
36k
Tacrine
>33
>23
>63
79
25
18
55
159
24
>909
>3333
>3333
9
>10 000
554 ꢃ 36
501 ꢃ 40
1430 ꢃ 162
333 ꢃ 25
477 ꢃ 35
217 ꢃ 17
>10 000
>10 000
>10 000
653 ꢃ 140
224 ꢃ 19
73 ꢃ 7
20 ꢃ 1.6
76 ꢃ 6.9
6 ꢃ 1.2
Fig. 3. Selected residues of the protein binding site in complex with compound 1 at
the beginning (gray) and at the end (green) of the 10 ns molecular dynamics simu-
lation. (For interpretation of the references to colour in this figure legend, the reader is
referred to the web version of this article.)
3 ꢃ 0.2
18 ꢃ 3.0
24 ꢃ 2.6
9 ꢃ 1.2
9 ꢃ 0.4
11 ꢃ 0.9
3 ꢃ 1.1
crude product was purified by neutral alumina column chroma-
tography (eluent: DCM/MeOH: 99/1 to 97/3) to afford the title
compound as an amorphous white solid (112 mg, yield 72%). IR
(cmꢂ1): 3296, 2966, 2935, 2873, 2832, 1716, 1698, 1668, 1558, 1456,
7 ꢃ 0.8
3 ꢃ 0.4
24 ꢃ 1.7
14 ꢃ 4.5
136 ꢃ 10.8
70 ꢃ 7.1
14 ꢃ 5.3
484 ꢃ 40
16
0.15
1251,1208,1185, 1156; 1H NMR (300 MHz, CDCl3):
d
(ppm) 6.90 (1H,
a
Values are expressed as the mean ꢃ standard error of the mean of three
experiments. IC50 inhibitory concentration (nM) of AChE from Electrophorus elec-
tricus (EeAChE) or human recombinant (hAChE) or BuChE from human serum
(hBuChE).
d, JNAHeH3 ¼ 9.0 Hz, NAeH), 5.11 (1H, m, H16
b
), 4.07 (1H, m, H3 ),
a
3.40 (1H, d, Jgem ¼ 12.9 Hz, H29b), 3.00 (1H, d, Jgem ¼ 12.9 Hz, H29a),
2.53 (1H, d, Jgem ¼ 17.2 Hz, H12
(1H, d, Jgem ¼ 17.2 Hz, H12 ), 2.21 (1H, dd, J17
J17 a), 2.13 (6H, bs, NBMe2), 2.00e2.09 (3H, m,
a
), 2.46 (2H, m, H1
b
þ H20), 2.36
b
IC50 (hBuChE)/IC50 (hAChE).
Values from Ref. [8].
Not determined.
c
b
¼ 11.1 Hz,
ae20
d
¼ 5.8 Hz, H17
a
e16b
H5
H2
H7
a
a
b
þ H8
þ H2
), 1.42 (1H, m, H15
b
þ H15
þ H6 ), 1.56 (1H, d, Jgem ¼ 3.8 Hz, H19
), 1.25e1.36 (2H, m, H7
þ H1
), 0.90 (1H, m, H6b
b
), 1.98 (3H, s, H32), 1.62e1.79 (3H, m,
), 1.51 (1H, m,
), 1.14 (3H, s,
), 0.83 (3H, s,
b
a
b
at 0 ꢀC for 1 h then at RT for further 12 h. After 3 co-evaporations
with 3 ꢁ 10 mL of 1,2-dichloroethane, the residue was dissolved in
DCM, washed with 20 mL of a saturated solution of sodium
hydrogen carbonate then extracted with 3 ꢁ 20 mL of DCM.
Combined organic layers were washed with brine (20 mL), dried
over magnesium sulfate and concentrated under vacuum. The
a
a
a
H28),1.05 (1H, d, Jgem ¼ 3.8 Hz, H19
b
H18), 0.80 (3H, d, J21e20 ¼ 6.3 Hz, H21), 0.62 (3H, s, H30); 13C NMR
(75.4 MHz, CDCl3):
d
(ppm) 211.3 (C11), 170.6 (C31), 158.4 (q,
JC1 eF ¼ 37.3 Hz, C10), 115.8 (q, JC2 eF ¼ 287.6 Hz, C20), 78.8 (C16), 63.9
(C29), 59.4 (C20), 54.9 (C17), 52.2 (C3), 51.8 (C12), 47.3 (C14), 44.6
(C4), 44.4 (C13), 42.9 (C15), 41.3 (C8), 41.2 (C5), 40.1 (NBMe2), 37.3
(C10), 33.8 (C9), 30.5 (C19), 27.4 (C1), 27.0 (C2), 24.2 (C7), 21.8 (C32),
19.5 (C28), 18.2 (C6), 17.8 (C18), 11.1 (C30), 9.77 (C21); MS (ESI, m/z):
571.3 (M þ H), 529.3; HRMS (ESI, m/z): calcd for C30H46N2O5F3:
571.3359, found: 571.3385.
0
0
5.1.3. {20S-(dimethylamino)-16
oxo-3 -[(trifluoroacetyl) amino]-9,19-cyclo-5
methyl tosylate (8a)
a
-hydroxy-4
b
,14
,9
a
-dimethyl-11-
-pregnan-4 -yl}
b
a
b
a
To 110 mg (0.19 mmol, 1 eq.) of 7a solubilized in 5 mL of
anhydrous MeOH were added 37 mg (0.19 mmol, 1 eq.) of p-tol-
uenesulfonic acid and a small amount of sodium sulfate. The
mixture was stirred at RT for 1 h and filtered through a pad of celite.
The solvent was removed and the residue was dried several
minutes under vacuum. 1.5 mL of pyridine and 110 mg (0.19 mmol,
1 eq.) of tosyl chloride were added. The mixture was stirred at 90 ꢀC
for 3 h. After 3 co-evaporations with 3 ꢁ 10 mL of 1,2-dichloro-
ethane, the residue was dissolved in 10 mL of a mixture MeOH/
water (3/1) and a small amount of potassium carbonate was added.
The mixture was stirred at RT for 3 h and the MeOH was removed
under vacuum. The mixture was extracted with 3 ꢁ 10 mL of DCM.
Combined organic layers were washed with brine, dried over
magnesium sulfate and concentrated under vacuum. The crude
product was purified by silica gel column chromatography (eluent:
Fig. 2. Selected residues of the protein binding site (apo form) at the beginning (gray)
and at the end (orange) of the 10 ns molecular dynamics simulation. (For interpreta-
tion of the references to colour in this figure legend, the reader is referred to the web
version of this article.)

J. Rouleau et al. / European Journal of Medicinal Chemistry 46 (2011) 2193e2205
2201
Fig. 4. Selected residues of the protein binding site in complex with compound 36e at the beginning (gray) and at the end (magenta) of the 10 ns molecular dynamics simulation.
(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
DCM/MeOH/NH4OH: 98/1/1) to afford the title compound as an
amorphous light yellow solid (70 mg, yield 50%). IR (cmꢂ1): 3321,
2924, 2869, 1719, 1667, 1551, 1454, 1367, 1159, 1189, 1208, 1175; 1H
5.1.4. 20S-(dimethylamino)-16
cyclo-{20-trifluoromethyl-50,60-dihydro-40
[40,50:3,4]}-5
,9 -pregnan-11-one (9a)
a
-hydroxy-4
b,14a-dimethyl-9,19-
a
H-[10,30]oxazino
a
b
NMR (300 MHz, CDCl3):
7.34 (d, 2H, J ¼ 8.3 Hz, H33 þ H35), 5.92 (1H, d, JNAHeH3 ¼ 9.8 Hz,
NAeH), 4.27e4.08 (2H, m, H3 ), 3.86 (1H, d, Jgem ¼ 9.9 Hz,
þ H16
H29b), 3.43 (1H, d, Jgem ¼ 9.9 Hz, H29a), 2.65 (1H, m, H20), 2.53 (1H,
d, Jgem ¼ 17.2 Hz, H12 ), 2.45 (1H, m, H1 ), 2.43 (3H, s, H37), 2.30
(1H, d, Jgem ¼ 17.2 Hz, H12 ), 2.28 (6H, bs, NBMe2), 2.01 (4H, m,
H17 þ H8 þ H15 ), 1.79 (1H, m, H2 ), 1.56 (1H, m, H2 ),
þ H5
1.54 (1H, d, Jgem ¼ 3.9 Hz, H19 ), 1.51 (1H, m, H7 ), 1.50 (2H, m,
H6 ), 1.34 (1H, m, H7 ), 1.29 (1H, m, H1 ), 1.19 (3H, s, H28),
þ H15
), 0.89 (3H, d,
d
(ppm) 7.77 (2H, d, J ¼ 8.3 Hz, H32 þ H36),
To 34 mg (0.05 mmol, 1 eq.) of 8a solubilized in 2 mL of anhy-
drous DMF were added 14 mg (0.15 mmol, 3 eq.) of sodium difor-
mylamide. The mixture was stirred at 120 ꢀC for 5 h, washed with
20 mL of a sodium hydrogen carbonate saturated solution and
extracted with 5 ꢁ 30 mL of DCM. Combined organic layers were
washed with brine, dried over magnesium sulfate and concentrated
under vacuum. The crude product was purified by preparative TLC
(eluent: DCM/MeOH/NH4OH: 89/10/1) to afford the title compound
as an amorphous colorless solid (17 mg, yield 68%). IR (cmꢂ1): 3669,
2924, 1719, 1698, 1455, 1151, 1209; 1H NMR (300 MHz, CDCl3):
a
b
a
b
b
a
a
b
b
a
b
b
b
a
a
a
a
1.01 (1H, d, Jgem ¼ 3.9 Hz, H19
b), 0.94 (1H, m, H6b
J21e20 ¼ 6.5 Hz, H21), 0.84 (3H, s, H18), 0.75 (3H, s, H30); 13C NMR
d
(ppm) 4.26 (1H, m, H16
(1H, d, Jgem ¼ 10.3 Hz, H29a), 3.16 (1H, m, H3
2.53 (1H, d, Jgem ¼ 17.1 Hz, H12 ), 2.43 (1H, m, H1
NBMe2), 2.34 (1H, d, Jgem ¼ 17.1 Hz, H12 ), 2.13 (1H, m, H5
(1H, m, H15 ), 2.04 (1H, m, H17 ), 1.99 (1H, m, H2 ), 1.68 (1H, d,
Jgem ¼ 4.1 Hz, H19 ), 1.58 (1H, m, H7 ), 1.55 (1H, m, H2 ), 1.54 (1H,
m, H8 ), 1.51 (1H, m, H15 ), 1.50 (1H, m, H6 ), 1.27 (1H, m, H1
1.23 (1H, m, H7
), 1.22 (3H, s, H28), 1.12 (1H, d, Jgem ¼ 4.1 Hz, H19
1.02 (1H, m, H6
H18), 0.87 (3H, s, H30); 13C NMR (75.4 MHz, CDCl3):
b
), 4.17 (1H, d, Jgem ¼ 10.3 Hz, H29b), 3.99
), 2.79 (1H, m, H20),
), 2.40 (6H, bs,
), 2.03
0
(75.4 MHz, CDCl3):
d
(ppm) 211.1 (C11), 156.4 (q, JC1 eF ¼ 36.8 Hz,
a
C10), 145.1 (C34), 131.9 (C31), 129.8 (C33 þ C35), 128.1 (C32 þ C36),
a
b
115.7 (q, JC2 eF ¼ 288.2 Hz, C20), 78.0 (C16), 70.2 (C29), 62.1 (C20),
b
a
0
55.8 (C17), 51.4 (C12), 51.3 (C3), 47 (C14), 44.4 (C13), 43.0 (C4), 42.9
(C15), 42.3 (C8), 41.8 (C5), 36.6 (C10), 34.1 (C9), 30.4 (C19), 27.6 (C2),
27.0 (C1), 24.3 (C7), 21.6 (C37), 20.9 (C28), 18.5 (C6), 17.9 (C18), 11.1
(C30), 10.0 (C21); MS (ESI, m/z): 683.3 (M þ H); HRMS (ESI, m/z):
calcd for C35H50N2O6SF3: 683.3342, found: 683.3305.
b
a
a
b
b
b
b
a
a
a
b
),
),
a
b
), 0.98 (3H, d, J21e20 ¼ 6.4 Hz, H21), 0.85 (3H, s,
d
(ppm) 211.1
(C11), 147.0 (C10), 78.2 (C16), 77.4 (C29), 63.0 (C20), 59.9 (C3), 55.9
(C17), 51.4 (C12), 47.2 (C14), 44.9 (C13), 46.1 (C8), 43.2 (C15), 40.6
(C5), 36.7 (C10), 35.1 (C4), 33.9 (C9), 29.8 (C19), 28.0 (C1), 27.7 (C2),
23.7 (C7), 20.3 (C28), 18.2 (C6), 17.9 (C18), 10.9 (C30), 10.4 (C21). MS
(ESI, m/z): 511.3 (M þ H); HRMS (ESI, m/z): calcd for C28H42N2O3F3:
511.3148, found: 511.3129.
Table 2
Ab
aggregation of 36e, 36h and 36k, propidium iodide and tacrine.a
Compdb
AChE-induced Ab1e40 aggregation (%)
36e
36h
36k
69 ꢃ 2
56 ꢃ 1
42 ꢃ 1
5.1.5. 20S-(dimethylamino)-16
cyclo-{20-isopropyl-10-methyl-10,40
[40,50:3,4]}-5
,9 -pregnan-11-one (13a)
To 35 mg (0.068 mmol, 1 eq.) of 12a solubilized in 1 mL of n-
a
-hydroxy-4
a
b,14a-dimethyl-9,19-
Propidium iodide
Tacrine
82 ꢃ 2.5c
,50,60-tetrahydro-pyrimidino
7 ꢃ 0.21c
a
b
a
Values are expressed as mean ꢃ standard error of the mean from two inde-
pendent measurements.
b
BuOH were added 33 mL (0.34 mmol, 5 eq.) of TEA. The mixture was
A 100
Data from Ref. [9].
mM concentration of the inhibitor was used.
heated at 200 ꢀC in a microwave (Pmax ¼ 300 W) for 2 h. The
c

2202
J. Rouleau et al. / European Journal of Medicinal Chemistry 46 (2011) 2193e2205
solvent was removed under vacuum and the residue was purified
by amine functionalized silica gel column chromatography (eluent:
DCM/MeOH: 90/10) to afford the title compound as an amorphous
white solid (27 mg, yield 80%). IR (cmꢂ1): 3379, 2966, 2940, 2871,
1731, 1666, 1621, 1572, 1461, 1227, 1188, 1150, 1095; 1H NMR
3 ꢁ 20 mL of DCM. Combined organic layers were washed with
brine, dried over magnesium sulfate and concentrated under
vacuum. The crude product was purified by silica gel column
chromatography (eluent: DCM/MeOH/NH4OH: 90/10/1) to afford
the title compound as an amorphous white solid (17 mg, yield
85%). IR (cmꢂ1): 3316, 2965, 2937, 2871, 2836, 1669, 1595, 1461,
(300 MHz, CDCl3):
d
(ppm) 4.09 (1H, m, H16
b
), 3.29 (1H, d,
), 3.14 (1H,
Jgem ¼ 12.5 Hz, H29b), 3.22 (3H, s, H31), 3.18 (1H, m, H3
a
1252, 1223; 1H NMR (300 MHz, CDCl3):
2bs, NAH or NCeH), 4.09 (1H, m, H16
), 3.17 (1H, dm, J ¼ 13.2 Hz,
d
(ppm) 5.53 þ 5.37 (1H,
m, H20), 3.02 (1H, d, Jgem ¼ 12.5 Hz, H29a), 2.60 (1H, dq,
b
J20e17 ¼ 11.3 Hz, J20e21 ¼ 6.5 Hz, H20), 2.49 (1H, d, Jgem ¼ 17.2 Hz,
H3
a
), 2.61 (1H, dq, J20e17 ¼ 10.8 Hz, J20e21 ¼ 6.5 Hz, H20), 2.53
a
a
H12
a
), 2.36 (1H, m, H1
b
), 2.31 (1H, d, Jgem ¼ 17.2 Hz, H12
b), 2.25
(1H, m, H20), 2.51 (1H, d, Jgem ¼ 16.6 Hz, H12
a), 2.33 (1H, d,
(6H, bs, NBMe2), 2.18e1.92 (4H, m, H2
a
þ H8
b
þ H15
b
þ H17), 1.68
Jgem ¼ 16.6 Hz, H12
b), 2.24 (6H, bs, NBMe2), 2.18 (2H, m,
(1H, m, H2
1.51 (1H, m, H5
H6
þ H7
b
), 1.63 (1H, d, Jgem ¼ 3.8 Hz, H19
a
), 1.56 (1H, m, H7
b
),
H1
(1H, m, H5
1.61 (2H, m, H6
H6
b
þ H8
b
a
), 2.01 (1H, m, H15
), 1.66 (1H, m, H2
þ H7 ), 1.47 (2H, m, H15
b
b
), 2.00 (2H, m, H2
), 1.64 (1H, d, Jgem ¼ 4.2 Hz, H19
þ H7 ), 1.43 (1H, m,
a
þ H17
a
), 1.98
a
), 1.47 (7H, m, H30 þ H40 þ H15
a
), 1.37 (2H, m,
a
),
a
a
), 1.26 (1H, m, H1
a), 1.20 (3H, s, H28), 1.12 (1H, m, H6
b
),
b
a
a
a
b
1.06 (1H, d, Jgem ¼ 3.8 Hz, H19
0.84 (3H, s, H18), 0.82 (3H, s, H30); 13C NMR (75.4 MHz, CDCl3):
(ppm) 210.1 (C11), 167.5 (C10), 78.2 (C16), 62.5 (C29), 62.0 (C20),
b
), 0.87 (3H, d, J21e20 ¼ 6.5 Hz, H21),
a
), 1.27 (3H, s, H28), 1.25 (1H, m, H1
), 1.18 þ 1.17 (6H, 2d,
J ¼ 7.0 Hz, H30 þ H40), 1.04 (1H, d, Jgem ¼ 4.2 Hz, H19
b), 0.88 (3H,
d
d, J21e20 ¼ 6.6 Hz, H21), 0.87 (3H, s, H30), 0.81 (3H, s, H18); 13C
58.0 (C3), 55.9 (C17), 51.3 (C12), 47.1 (C14), 46.7 (C5), 44.5 (C13),
42.6 (C15), 40.9 (C8), 39.7 (NCMe), 36.2 (C10), 35.0 (C4), 33.9 (C9),
30.5 (C20), 29.6 (C19), 27.1 (C1), 23.9e23.8 (C2eC7), 20.6 (C28),
19.5e18.9 (C30eC40), 18.6 (C6), 17.9 (C18), 11.9 (C30), 9.9 (C21); MS
(ESI, m/z): 498.3 (M þ H); HRMS (ESI, m/z): calcd for C31H52N3O2:
498.4060, found: 498.4058.
NMR (75.4 MHz, CDCl3): d
(ppm) 210.2 (C11), 173.1 (C10), 78.3
(C16), 73.8 (C3), 69.2 (C4), 62.0 (C20), 55.9 (C17), 51.4 (C12), 48.2
(C5), 47.5 (C14), 44.9 (C13), 42.1 (C15), 39.1 (C8), 37.7 (C10), 35.4
(C9), 29.9 (C20), 28.9 (C1), 28.2 (C19), 24.1 (C2), 22.7 (C7), 20.1
(C28), 19.8 (C6), 17.8 (C18), 14.2 (C30), 9.9 (C21); MS (ESI, m/z):
470.3 (M þ H); HRMS (ESI, m/z): calcd for C29H48N3O2: 470.3747,
found: 470.3758.
5.1.6. 20S-(dimethylamino)-16
(10 / 19)abeo-{20-isopropyl-10-methyl-10,40
pyrimidino[40,50:3,4]}-5
,9 -pregnan-1(10)-en-11-one (13b)
To 34 mg (0.064 mmol, 1 eq.) of 12b solubilized in 1 mL of n-
BuOH were added 31 L (0.32 mmol, 5 eq.) of TEA. The mixture
a
-hydroxy-4
b
,14
a-dimethyl-9
a
,50,60-tetrahydro-
5.1.8. 20S-(dimethylamino)-16
a
-hydroxy-4
b
,14
a
-dimethyl-9,19-
a
b
cyclo-{60-isopropyl-20
a
,30-dihydropyridino[20,30:3,4]}-5
a,9b-
pregnan-11-one (25)
m
To 30 mg (0.042 mmol, 1 eq.) of 24 solubilized in 1 mL of DCM
was added 1 mL of TFA. The mixture was stirred at RT for 2 h then
concentrated under vacuum. The residue was dissolved in 6 mL of
a mixture MeOH/water (ratio 3/1) and a small amount of potassium
carbonate was added. The mixture was refluxed for 3 h and the
MeOH was removed under reduced pressure. The mixture was
diluted with 20 mL of a saturated sodium hydrogen carbonate
solution, extracted with 3 ꢁ 20 mL of DCM. Combined organic
layers were washed with brine, dried over magnesium sulfate and
concentrated under vacuum. The crude product was purified by
silica gel column chromatography (eluent: DCM/MeOH/NH4OH:
95/5/1) to afford the title compound as an amorphous white solid
(13 mg, yield: 63%). IR (cmꢂ1): 3352, 2962, 2932, 2868, 1668, 1639,
was heated at 250 ꢀC in a microwave (Pmax ¼ 300 W) for 5 h. The
solvent was removed under vacuum and the residue was purified
by amine functionalized silica gel column chromatography
(eluent: DCM/MeOH: 90/10) to afford the title compound as an
amorphous light yellow solid (13 mg, yield 40%). IR (cmꢂ1): 3180,
2965, 2934, 2875, 1694, 1621, 1580, 1451; 1H NMR (500 MHz,
CDCl3):
Jgem ¼ 14.6 Hz, J31e32 ¼ 7.3 Hz, H31
J ¼ 11.0 Hz, H3 ), 3.39 (1H, dq, Jgem ¼ 14.6 Hz, J31e32 ¼ 7.3 Hz,
H31 ), 3.38 (1H, d, Jgem 15.6 Hz, H19 ), 3.32 (1H, d,
d
(ppm) 5.60 (1H, bs, H1), 4.09 (1H, m, H16
b), 3.61 (1H, dq,
a
), 3.51 (1H, dd, J ¼ 5.8 Hz,
a
b
¼
a
Jgem ¼ 12.8 Hz, H29b), 3.23 (1H, d, Jgem ¼ 12.8 Hz, H29a), 3.05 (2H,
m, H20 þ H2
), 2.61 (1H, dq, J20e17 ¼ 10.7 Hz, J20e21 ¼ 6.7 Hz,
a
a
H20), 2.57 (1H, d, Jgem
¼
15.6 Hz, H12
), 2.24 (6H, bs, NBMe2), 2.01 (2H, m,
), 1.98 (2H, m, H15 H19 ), 1.93 (3H, m,
), 1.54 (6H, d, J(3 þ4 )
a
), 2.36 (1H, d,
1576, 1462, 1261, 1224, 1158; 1H NMR (500 MHz, CDCl3):
6.55 (1H, d, J ¼ 9.5 Hz, H29), 5.98 (1H, d, J ¼ 9.5 Hz, H31), 4.18 (1H,
d (ppm)
Jgem ¼ 15.6 Hz, H12
H17 H9
H2
b
a
þ
a
b
þ
b
m, H16
b
), 2.94 (1H, dd, J ¼ 13.1 Hz, 4.0 Hz, H3
a), 2.72 (1H, m, H20),
b
þ H5
a
þ H7
b
), 1.68 (2H, m, H6
a
þ H8
b
2.59 (1H, sept, J ¼ 7.0 Hz, H20), 2.52 (1H, d, Jgem ¼ 16.8 Hz, H12
a),
0
0
¼ 7.0 Hz, H30 and H40), 1.50 (1H, m, H15
a
), 1.40 (2H, m,
2.33 (8H, m, H1
b
þ H12
b
þ NBMe2), 2.16 (1H, m, H8
b), 2.05 (2H, m,
20
H6
b
þ H7
a
), 1.29 (3H, t, J32e31 ¼ 7.3 Hz, H32), 1.26 (3H, s, H28),
H2
a
þ H15
), 1.68 (2H, m, H5
), 1.51 (1H, dd, J ¼ 13.7 Hz, J ¼ 2.1 Hz, H15
b
), 2.02 (1H, m, H17
a
),1.89 (1H, qd, J ¼ 13.1 Hz, J ¼ 3.7 Hz,
), 1.66 (1H, d, J ¼ 4.3 Hz, H19 ), 1.63
), 1.40 (1H, m,
0.87 (3H, d, J21e20 ¼ 6.4 Hz, H21), 0.78 (3H, s, H30), 0.69 (3H, s,
H2b
a
þ H6
a
a
H18); 13C NMR (75.4 MHz, CDCl3): (ppm) 210.8 (C11), 167.7 (C10),
d
(1H, m, H7
b
a
136.1 (C10), 120.5 (C1), 78.2 (C16), 62.0 (C20), 58.5 (C29), 55.5
(C17), 54.5 (C3), 50.5 (C5), 49.9 (C12), 49.8 (C8), 49.5 (C9), 48.3
(C14), 47.0 (C31), 46.2 (C13), 42.4 (C15), 36.7 (C19), 33.5 (C7), 32.8
(C4), 31.1 (C2), 30.1 (C20). 25.1 (C6), 20.0 and 19.4 (C03 þ C04), 18.8
(C28), 17.7 (C18), 13.7 (C32), 10.8 (C30), 10.0 (C21); MS (ESI, m/z):
512.3 (M þ H); HRMS (ESI, m/z): calcd for C32H54N3O2: 512.4216,
found: 512.4210.
H7a
), 1.25 (1H, m, H1
a
), 1.24 (3H, s, H28), 1.12 þ 1.09 (6H, 2d,
J ¼ 7.0 Hz, H30 þ H40), 1.08 (1H, m, H19
b), 1.04 (1H, m, H6b), 0.93
(3H, d, J21e20 ¼ 6.4 Hz, H21), 0.84 (3H, s, H18), 0.67 (3H, s, H30); 13
C
NMR (75.4 MHz, CDCl3):
d
(ppm) 210.4 (C11), 170.6 (C10), 147.4
(C29), 118.9 (C31), 77.6 (C16), 64.9 (C3), 62.4 (C20), 55.8 (C17), 51.4
(C12), 47.3 (C14), 45.4 (C5), 44.9 (C13), 42.8 (C15), 40.4 (C8), 38.3
(C4), 36.6 (C20), 34.4 (C10), 30.9 (C9), 29.3 (C19), 28.3 (C1 þ C2), 23.9
(C7), 20.3 (C28), 19.9 (C30 þ C40), 17.8 (C18), 17.7 (C6), 10.2 (C30),10.1
(C21); MS (ESI, m/z): 481.4 (M þ H); HRMS (ESI, m/z): calcd for
C31H49N2O2: 481.3794, found: 481.3784.
5.1.7. 20S-(dimethyamino)-16
a
-hydroxy-4 -dimethyl-9,19-
b
,14
a
cyclo-{20-isopropyl-40
a
,50-dihydro-10H-imidazolo[40,50:3,4]}-5
a,9b-
pregnan-11-one (20)
To 18 mg (0.043 mmol, 1 eq.) of 18 solubilized in 2 mL of
dioxane were added 8 L (0.052 mmol, 1.2 eq.) of methyl
5.1.9. 20S-(dimethylamino)-16
a
-hydroxy-4 -dimethyl-9,19-
b
,14
a
m
cyclo-{60-isopropyl-20
a
,30,40,50-tetrahydropyridino[20,30:3,4]}-5
a,9b-
orthoisobutyrate and 8 mg (0.013 mmol, 0.3 eq.) of ytterbium
pregnan-11-one (27)
triflate. The mixture was refluxed for 12 h, washed with 10 mL of
To 53 mg (0.088 mmol, 1 eq.) of 26 dissolved in 0.5 mL of MeOH
a
saturated sodium carbonate solution and extracted with
were added 9.3 mg (0.013 mmol, 0.15 eq.) of palladium hydroxide

J. Rouleau et al. / European Journal of Medicinal Chemistry 46 (2011) 2193e2205
2203
(20%). The mixture was stirred at RT under hydrogen atmosphere
(2 bar) in a Parr apparatus for 24 h. The mixture was filtered
through a pad of celite and the filtrate was concentrated under
reduced pressure. The residue was dissolved in 1 mL of DCM and
1 mL of TFA was slowly added. The mixture was stirred at RT for 2 h
then the solvents were removed under vacuum. The residue was
dissolved in 10 mL of DCM, washed with 10 mL of a saturated
hydrogen carbonate solution. The aqueous layer was extracted with
2 ꢁ 10 mL of DCM. Combined organic layers were washed with
brine, dried over magnesium sulfate and concentrated under
vacuum. The title compound was obtained without further purifi-
cation as an amorphous white solid (29 mg, yield: 68% for 2 steps).
IR (cmꢂ1): 3363, 2961, 2932, 2868, 1695, 1666, 1462, 1224, 1166; 1H
5.1.11. 16
cyclo-{20-isopropyl-50,60-dihydro-40
,9 -pregnan-11-one (34)
191 mg (0.38 mmol, 1 eq.) of 33, 945
trimethylaluminium (2 M in toluene) and 10 mL of 1,2-dichloro-
ethane were used by following similar procedure described for the
preparation of 30 to afford the title compound as an amorphous
white solid (128 mg, yield: 70%). IR (cmꢂ1): 3309, 2963, 2928, 2864,
1665, 1633, 1463, 1224, 1174, 1145, 1094; 1H NMR (300 MHz, CDCl3):
a
-Hydroxy-4
b
,14
a
-dimethyl-20S-(methylamino)-9,19-
a
H-[10,30]thiazino[40,50:3,4]}-
5a b
mL (1.9 mmol, 5 eq.) of
d
(ppm) 4.15 (1H, m, H16b), 2.85 (1H, m, H3a), 2.84 (1H, d,
Jgem ¼ 11.7 Hz, H29b), 2.74 (1H, d, Jgem ¼ 11.7 Hz, H29a), 2.59 (1H, m,
H20), 2.54 (1H, d, Jgem ¼ 17.1 Hz, H12
a
), 2.48 (1H, m, H20), 2.46 (4H,
), 2.07 (1H, m,
), 1.72 (1H, m, H17 ), 1.70
), 1.65 (1H, d, Jgem ¼ 4.5 Hz, H19 ), 1.56 (2H, m,
b), 1.51 (1H, m, H15a), 1.42 (1H, m, H5a), 1.34 (1H, m,
m, NBMe þ H1
b
), 2.37 (1H, d, Jgem ¼ 17.1 Hz, H12
b
NMR (500 MHz, CDCl3):
d
(ppm) 4.14 (1H, m, H16
b), 2.87 (1H, dm,
H8b
), 2.05 (1H, m, H2
a
), 2.01 (1H, m, H15
b
a
J ¼ 12.6 Hz, H3
a
), 2.66 (1H, dq, J ¼ 10.7 Hz, J ¼ 6.4 Hz, H20), 2.53
(1H, m, H2
b
a
(1H, m, H20), 2.51 (1H, d, Jgem ¼ 17.0 Hz, H12
a), 2.37 (1H, m, H1
b
),
H6
H7
a
a
þ H7
2.32 (1H, d, J ¼ 17.0 Hz, H12
b), 2.28 (6H, bs, NBMe2), 2.21 (2H, m,
), 1.31 (1H, m, H1
a
), 1.20 (3H, s, H28), 1.17 (6H, m, H30 þ H40),
H31), 2.10 (1H, m, H8
H2 ), 1.71 (1H, m, H29a), 1.66 (1H, d, J ¼ 4.0 Hz, H19
þ H17
(1H, m, H7 ), 1.58 (1H, m, H6 ), 1.50 (1H, m, H2 ), 1.49 (1H, m,
H15 ), 1.40 (1H, m, H5
(1H, m, H1
b
), 2.03 (1H, m, H15
b
), 2.02 (2H, m,
1.12 (1H, d, Jgem ¼ 4.5 Hz, H19
1.02 (1H, m, H6
(75.4 MHz, CDCl3):
b
), 1.10 (3H, d, J21e20 ¼ 6.0 Hz, H21),
a
a
a
), 1.60
b
), 0.85 (3H, s, H18), 0.76 (3H, s, H30); 13C NMR
b
a
b
d
(ppm) 210.8 (C11), 165.8 (C10), 77.9 (C16), 62.7
a
a
), 1.32 (1H, m, H7
a), 1.30 (1H, m, H29b), 1.29
(C3), 60.8 (C17), 58.5 (C20), 51.5 (C12), 49.8 (C5), 47.1 (C14), 44.7
(C13), 42.8 (C15), 41.4 (C8), 40.2 (C20), 38.7 (C29), 38.4 (C10), 34.4
(C9), 33.8 (NBMe), 33.4 (C4), 30.6 (C2), 30.3 (C19), 28.0 (C1), 24.4
(C7), 21.0 (C30 þ C40), 20.4 (C28),18.7 (C21), 18.0 (C6),17.9 (C18), 11.3
(C30); MS (ESI, m/z): 487.3 (M þ H); HRMS (ESI, m/z): calcd for
C29H47N2O2S: 487.3358, found: 487.3364.
a
), 1.23 (3H, s, H28), 1.09 (6H, d, J ¼ 6.8 Hz, H30 þ H40),
1.08 (1H, m, H19
b
), 0.98 (1H, m, H6
b
), 0.90 (3H, d, J21e20 ¼ 6.4 Hz,
H21), 0.84 (3H, s, H18), 0.63 (3H, s, H30); 13C NMR (75.4 MHz,
CDCl3):
d
(ppm) 210.7 (C11), 175.6 (C10), 77.9 (C16), 65.0 (C3), 62.3
(C20), 55.9 (C17), 51.4 (C12), 49.2 (C5), 47.3 (C14), 44.8 (C13), 42.8
(C15), 40.7 (C8), 38.5 (C10), 38.4 (C20), 36.4 (C4), 34.5 (C9), 32.2
(C29), 30.1 (C19), 29.3 (C2), 28.5 (C1), 24.2 (C7), 23.1 (C31), 20.5
(C28), 20.3 þ 20.2 (C30 þ C40), 17.8 (C18), 17.6 (C6), 11.4 (C30), 10.0
(C21); MS (ESI, m/z): 483.4 (M þ H); HRMS (ESI, m/z): calcd for
C31H51N2O2: 483.3951, found: 483.3959.
5.1.12. 20S-[(3-methoxybenzyl)(methyl)amino]-16
,14
-dimethyl-9,19-cyclo-{20-isopropyl-50,60-dihydro-40
thiazino[40,50:3,4]}-5
,9 -pregnan-11-one (36e)
16 mg (0.033 mmol, 1 eq.) of 34, 14 mg (0.099 mmol, 3 eq.) of
potassium carbonate, 6 L (0.043 mmol,1.3 eq.) of 3-methoxybenzyl
a
-hydroxy-
4b
a
a
H-[10,30]
a
b
m
5.1.10. 20S-(dimethylamino)-16
cyclo-{20-isopropyl-50,60-dihydro-40
,9 -pregnan-11-one (30)
To 133 mg (0.256 mmol, 1 eq.) of 29 dissolved in 5 mL of 1,2-
dichloroethane were added 640 L (1.28 mmol, 5 eq.) of trimethy-
a
-hydroxy-4
a
b
,14
a
-dimethyl-9,19-
chloride and 0.5 mL of DMF were used by following similar proce-
dure described for the preparation of 36a to give the title compound
as an amorphous white solid (16 mg, yield: 80%). IR (cmꢂ1): 3438,
2962, 2930, 2866, 1667, 1633, 1600, 1586, 1455, 1264, 1223, 1155,
H-[10,30]thiazino[40,50:3,4]}-
5
a b
m
1042,1018; 1H NMR (500 MHz, CD3CN):
d
(ppm) 7.16 (1H, t, J ¼ 7.9 Hz,
H36), 6.82 (1H, d, J ¼ 7.9 Hz, H37), 6.78 (1H, s, H33), 6.73 (1H, dd,
J ¼ 7.9 Hz, J ¼ 2.1 Hz, H35), 3.98 (1H, bs, H16 ), 3.73 (3H, s, OMe), 3.53
), 2.76 (1H, d,
laluminium (2 M in toluene). The mixture was refluxed for 3 h then
cooled to RT. The mixture was quenched with 3 mL of sodium
hydroxide (1 N), diluted with 10 mL of a saturated potassium
hydrogen carbonate solution and extracted with 3 ꢁ 10 mL of DCM.
Combined organic layers were washed with brine, dried over
magnesium sulfate and concentrated under vacuum. The crude
product was purified by silica gel column chromatography (gradient
elution: DCM/MeOH/NH4OH: from 99/0/1 to 90/10/1) to afford the
title compound as an amorphous white solid (76 mg, yield: 59%). IR
(cmꢂ1): 3330, 2962, 2933, 2866, 1661, 1633, 1462, 1260, 1124, 1157,
b
(2H, bs, H31), 2.77 (1H, dd, J ¼ 12.5 Hz, J ¼ 4.0 Hz, H3
a
J ¼ 11.9 Hz, H29b), 2.74 (1H, m, H20), 2.65 (1H, d, Jgem ¼ 11.9 Hz,
H29a), 2.51 (1H, septuplet, J ¼ 7.0 Hz, H20), 2.45 (1H, d, Jgem ¼ 16.8 Hz,
H12
Jgem ¼ 16.8 Hz, H12
(2H, m, H2 ),1.92 (1H, m, H15
þ H8
Jgem ¼ 4.3 Hz, H19 ), 1.50 (2H, m, H6
a
), 2.37 (1H, dt, Jgem ¼ 13.7 Hz, J ¼ 3.7 Hz, H1
b
), 2.25 (1H, d,
), 1.98
),1.56 (1H, d,
b
), 2.09 (3H, s, NBMe), 2.01 (1H, m, H17
),1.61 (1H, m, H2
þ H7 ), 1.45 (1H, m, H15
), 1.23 (2H, m, H1
a
a
b
b
b
a
a
b
a
a
),
),
1.35 (1H, dd, J ¼ 12.2 Hz, J ¼ 3.4 Hz, H5
a
a
þ H7
1088; 1H NMR (300 MHz, CDCl3):
(1H, m, H3
), 2.83 (1H, d, Jgem ¼ 11.6 Hz, H29b), 2.73 (1H, d,
d
(ppm) 4.10 (1H, m, H16
b
), 2.84
1.14 (3H, s, H28),1.11 þ1.10 (6H, 2d, J ¼ 7.0 Hz, H30 þ H40),1.03 (1H, d,
a
J ¼ 4.3 Hz, H19
b), 0.95 (1H, m, H6b), 0.89 (3H, bs, H21), 0.70 (3H, s,
Jgem ¼ 11.6 Hz, H29a), 2.63 (1H, dq, J20e17 ¼ 9.9 Hz, J20e21 ¼ 6.6 Hz,
H18), 0.69 (3H, s, H30); 13C NMR (75.4 MHz, CD3CN):
d (ppm) 211.0
a
H20), 2.58 (1H, m, H20), 2.51 (1H, d, Jgem ¼ 17.0 Hz, H12
a
), 2.45 (1H,
), 2.25
), 2.01 (1H, m,
),1.64 (1H, d, Jgem ¼ 3.9 Hz,
), 1.49 (1H, m, H15 ), 1.42
), 1.21 (3H, s, H28),
(C11), 165.8 (C10), 159.7 (C34), 140.3 (C32), 129.5 (C36), 121.3 (C37),
114.4 (C33), 113.0 (C35), 77.9 (C16), 62.7 (C3), 61.3 (C31), 59.5 (C20),
55.9 (C17), 55.2 (OMe), 51.5 (C12), 49.8 (C5), 47.2 (C14), 44.7 (C13),
42.8 (C15), 41.4 (C8), 40.3 (C20), 38.8 (C29), 38.4 (C10), 34.5
(NBMe þ C9), 33.4 (C4), 30.6 (C2), 30.3 (C19), 28.0 (C1), 24.4 (C7),
21.0 þ 20.8 (C30 þ C40), 20.6 (C28), 18.0 (C6), 17.8 (C18), 11.3 (C30),
10.7 (C21); MS (ESI, m/z): 607.3 (M þ H); HRMS (ESI, m/z): calcd for
C37H55N2O3S: 607.3933, found: 607.3926.
dt, J ¼ 13.4 Hz, J ¼ 3.5 Hz, H1
b
), 2.32 (1H, d, Jgem ¼ 17.0 Hz, H12
b
(6H, bs, NBMe2), 2.10 (1H, m, H8
b
), 2.05 (1H, m, H2
),1.71 (1H, m, H2
), 1.53 (1H, m, H6
), 1.31 (1H, m, H1
a
H15
b
),1.98 (1H, m, H17
), 1.58 (1H, m, H7
), 1.34 (1H, m, H7
a
b
b
a
H19a
a
(1H, m, H5
a
a
a
1.17 and 1.18 (6H, 2d, J ¼ 6.9 Hz, H30 þ H40),1.11 (1H, d, Jgem ¼ 3.9 Hz,
H19b
),1.01 (1H, m, H6
b
), 0.88 (3H, d, J21e20 ¼ 6.6 Hz, H21), 0.85 (3H, s,
H18), 0.76 (3H, s, H30); 13C NMR (75.4 MHz, CDCl3):
d (ppm) 210.8
(C11), 165.7 (C10), 78.3 (C16), 62.7 (C3), 62.0 (C20), 55.7 (C17), 51.4
(C12), 49.8 (C5), 47.1 (C14), 44.5 (C13), 42.8 (C15), 41.4 (C8), 40.2
(C20), 38.7 (C29), 38.4 (C10), 34.5 (C9), 33.4 (C4), 30.6 (C2), 30.2 (C19),
28.0 (C1), 24.4 (C7), 20.8 and 21.0 (C30 þ C40), 20.7 (C28), 18.0 (C6),
17.9 (C18), 11.3 (C30), 9.9 (C21); MS (ESI, m/z): 501.3 (M þ H); HRMS
(ESI, m/z): calcd for C30H49N2O2S: 501.3515, found: 501.3537.
5.2. Biochemical methods
5.2.1. In vitro AChE inhibition assay
Enzymes: EeAChE from E. electricus (reference C 2888) and
human recombinant hAChE (reference C 1682) were purchased
from Sigma.

2204
J. Rouleau et al. / European Journal of Medicinal Chemistry 46 (2011) 2193e2205
Inhibition of AChE activity was determined by the spectroscopic
intensities obtained for Ab plus AChE in the presence and in the
method of Ellman et al. [14], using acetylthiocholine iodide as
substrate, in 96-well microtiter plates. All solutions were brought
absence of inhibitor, respectively, after subtracting the fluorescence
of respective blanks. Each assay was run in triplicate.
to room temperature prior to use. Aliquots of 200
containing 640 L (10 mM) dithiobisdinitrobenzoate (DTNB) in
0.1 M sodium phosphate, pH 8.0, 19.2 mL of the same buffer, and
13 L of a solution of AChE (100 U/mL) in water, were added to
each well, followed by 2 L of a DMSO solution of the inhibitor. The
reaction was initiated by adding 20 L of acetylthiocholine iodide
(7.5 mM) to each well, and followed by monitoring the appearance
of the thiolate dianion produced by reduction of DTNB at 412 nm
for 120 s at 25 ꢀC in a Molecular Devices Spectra Max 384 Plus
plate reader. Each inhibitor was evaluated at several concentra-
tions in the range of 10ꢂ9e2 ꢁ 10ꢂ5 M. Percentage inhibition was
calculated relative to a control sample (DMSO), and IC50 values
displayed represent the mean ꢃ standard deviation for triplicate
assays.
mL of a solution
m
5.3. Computer-aided molecular modeling
m
Molecular docking was carried out using GOLD 4.0 [22] with
standard parameters. Crystal structure of EeAChE (pdb code 1EEA)
has been chosen in order to facilitate the comparison with the
inhibition assay results, the binding site being defined as a 20 Å
radius sphere around the OH group of Ser200. Although this protein
has been crystallized in the apo form, the residues that form the
anionic site are positioned in the open form of the gorge, thus
making this structure appropriate for docking compounds
presumed to interact with both the catalytic and peripheral sites of
EeAChE. The 3D structures of ligands were constructed with COR-
INA 3.44 (Molecular Networks GmbH).
m
m
Molecular dynamics simulations were carried out with GRO-
MACS version 4.0.2 [23] using the OPLS-AA [24] force field. Each
system was energy-minimized until convergence using a steepest
descents algorithm. Molecular dynamics with position restraints
for 200 ps was then performed followed by the production run of
10 ns. During the position restraints and production runs, the
ParinelloeRahman method [25] was used for pressure coupling,
and the temperature was coupled using the Nosé-Hoover method
[26] at 300 K. Electrostatics were calculated with the particle mesh
Ewald method [27]. The P-LINCS algorithm [28] was used to
constrain bond lengths, and a time step of 2 fs was used
throughout. Ligand topologies for the OPLS-AA force field were
obtained using an in-house developed script. All calculations were
performed using the HPC facilities at the ICSN.
5.2.2. In vitro BuChE inhibition assay
Enzyme: hBuChE from human serum (reference C 9971) was
purchased from Sigma.
Inhibition of BuChE activity was determined by the spectro-
scopic method of Ellman et al. [14], using butyrylthiocholine iodide
as substrate, in 96-well microtiter plates. All solutions were
brought to room temperature prior to use. Aliquots of 200
a solution containing 640 L (10 mM) DTNB in 0.1 M sodium
phosphate, pH 8.0, 19.2 mL of the same buffer, and 13 L of a solu-
tion of BuChE (100 U/mL) in water, were added to each well, fol-
lowed by 2 L of a DMSO solution of the inhibitor. The reaction was
initiated by adding 20 L of butyrylthiocholine iodide (7.5 mM) to
mL of
m
m
m
m
each well, and followed by monitoring the appearance of the thi-
olate dianion produced by reduction of DTNB at 412 nm for 120 s at
25 ꢀC in a Molecular Devices Spectra Max 384 Plus plate reader.
Each inhibitor was evaluated at several concentrations in the range
of 10ꢂ9e2 ꢁ 10ꢂ5 M. Percentage inhibition was calculated relative
to a control sample, and IC50 values displayed represent the
mean ꢃ standard deviation for triplicate assays.
Images were generated with Chimera [29] and raytraced with
the POV-ray module.
Acknowledgment
We thank the CNRS and ICSN for financial support. J. Rouleau
was supported by fellowship from the Institut de Chimie des
Substances Naturelles (ICSN-CNRS). Professor J-Y. Lallemand is
gratefully acknowledged for his interest in our work. We also thank
Dr. Jordi Molgo for fruitful discussions.
5.2.3. Inhibition of AChE-induced Ab40 aggregation assay
Thioflavine T (T3516), 1,1,1,3,3,3,-hexafluoro-2-propanol (HFIP)
and human recombinant hAChE (reference C 1682) were purchased
from Sigma.
purchased form Bachem AG (Bubendorf, Switzerland). Aliquots of
L of (Ab1e40), lyophilized from 2 mg/mL HFIP and dissolved in
DMSO at a final concentration of 230 M, were incubated for 24 h at
room temperature in 0.215 M sodium phosphate buffer (pH 8.0).
For coincubation experiments, aliquots of hAChE (2.30 M, ratio
100:1) and hAChE in the presence of the tested compound (100 M)
were added. Blanks containing A , hAChE, A plus the tested
compound, and hAChE plus the tested compound in 0.215 M
sodium phosphate buffer (pH 8.0) were prepared. The final volume
of each vial was 20 mL. To quantify amyloid fibril formation, the
thioflavin T fluorescence method was used [9] and monitored with
a spectrofluorometer (Hitachi F-2500). Thioflavin T binds to
amyloid fibrils, giving rise to an intense specific emission band at
490 nm in its fluorescent emission spectrum. Therefore, after
incubation, the samples were diluted to a final volume of 2 mL with
b
-amyloid1e40 (A
b
1ꢂ40) trifluoroacetate salt was
Appendix. Supplementary material
2
m
m
Supplementary data related to this article can be found online at
doi:10.1016/j.ejmech.2011.02.073.
m
m
References
b
b
[1] J. Hardy, D.J. Selkoe, Science 297 (2002) 353e356.
[2] M. Tolnay, A. Probst, Neuropathol. Appl. Neurobiol. 25 (1999) 171e187.
[3] R.T. Bartus, R.L. Dean 3rd, B. Beer, A.S. Lippa, Science 217 (1982) 408e414.
[4] S.B. Dunnett, H.C. Fibiger, Prog. Brain Res. 98 (1999) 413e420.
[5] R.T. Bartus, Exp. Neurol. 163 (2000) 495e529.
[6] D. Munoz-Torrero, P. Camps, Curr. Med. Chem. 13 (2006) 399e422.
[7] D. Munoz-Torrero, Curr. Med. Chem. 15 (2008) 2433e2455.
[8] T. Sauvaître, M. Barlier, D. Herlem, N. Gresh, A. Chiaroni, D. Guenard,
C. Guillou, J. Med. Chem. 50 (2007) 5311e5323.
[9] M. Bartolini, C. Bertucci, V. Cavrini, V. Andriasino, Biochem. Pharmacol. 65
(2003) 407e416.
[10] W.K. Hagmann, J. Med. Chem. 51 (2008) 4259e4369.
[11] H.K. Nair, K. Lee, D.M. Quinn, J. Am. Chem. Soc. 115 (1993) 9939e9941.
[12] J.T. Shaw, W.L. Corbett, D.L. Layman, G.D. Cuny, J. Kerschner, J. Heterocycl.
Chem. 25 (1988) 1837e1840.
50 mM glycineeNaOH buffer (pH 8.5) containing 1.5
T. A 300 s time scan of fluorescence intensity was carried out
exc ¼ 446 nm, em ¼ 490 nm), and values at the plateau were
averaged after subtraction of the background fluorescence of the
1.5 M thioflavin T solution. The percent inhibition of the AChE-
mM thioflavin
(l
l
[13] C. Guillou, T. Sauvaitre, D. Guénard, J.Y. Lallemand, D. Herlem, J. Molgo, F.
Khuong-Huu, WO 2006 082 126.
[14] G.L. Ellman, K.D. Courtney, V. Andres, R.M. Featherstone, Biochem. Pharmacol.
7 (1961) 88e95.
[15] E. Giacobini, Butyrylcholinesterase Its Functions and Inhibitors. Martin Dunitz,
London, 2003.
m
induced aggregation due to the presence of increasing concentra-
tions of the inhibitor was calculated by the following expression:
100 ꢂ [(IFi/IF0) ꢁ 100] where IFi and IFo were the fluorescence

J. Rouleau et al. / European Journal of Medicinal Chemistry 46 (2011) 2193e2205
2205
[16] D.E. Kuhl, R.A. Koeppe, S.E. Snyder, S. Minoshima, K.A. Frey, M.R. Kilbourn,
Ann. Neurol. 59 (2006) 13e20.
[17] M.L. Bolognesi, V. Andrisano, M. Bartolini, R. Banzi, C. Melchiorre, J. Med.
Chem. 48 (2005) 24e27.
[21] Q. Xie, H. Wang, Z. Xia, M. Lu, W. Zhang, X. Wang, W. Fu, Y. Tang, W. Sheng,
W. Li, W. Zhou, X. Zhu, Z. Qiu, H. Chen, J. Med. Chem. 51 (2008) 2027e2036.
[22] M.L. Verdonk, J.C. Cole, M.J. Hartshorn, C.W. Murray, R.D. Taylor, Proteins 52
(2003) 609e623.
[18] P. Munoz-Ruiz, L. Rubio, E. Garcıa-Palomero, I. Dorronsoro, M. del Monte-
Millan, R. Valenzuela, P. Usan, C. de Austria, M. Bartolini, V. Andrisano,
A. Bidon-Chanal, M. Orozco, F.J. Luque, M. Medina, A. Martınez, J. Med. Chem.
48 (2005) 7223e7233.
[23] B. Hess, C. Kutzner, D. van der Spoel, E. Lindahl, J. Chem. Theory Comput. 4
(2008) 435e447.
[24] W.L. Jorgensen, D.S. Maxwell, J. Tirado-Rives, J. Am. Chem. Soc. 118 (1996)
11225e11236.
[19] A. Cavalli, M.L. Bolognesi, S. Capsoni, V. Andrisano, M. Bartolini, E. Margotti,
A. Cattaneo, M. Recanatini, C. Melchiorre, Angew. Chem. Int. Ed. 46 (2007)
3689e3692.
[20] M.L. Bolognesi, R. Banzi, M. Bartolini, A. Cavalli, A. Tarozzi, V. Andrisano,
A. Minarini, M. Rosini, V. Tumiatti, C. Bergamini, R. Fato, G. Lenaz,
P. Hrelia, A. Cattaneo, M. Recanatini, C. Melchiorre, J. Med. Chem. 50
(2007) 4882e4897.
[25] M. Parinello, A. Rahman, J. Appl. Phys. 52 (1981) 7182e7190.
[26] S. Nosé, Mol. Phys. 52 (1984) 255e268.
[27] U. Essman, L. Perela, M.L. Berkowitz, T. Darden, H. Lee, L.G. Pedersen, J. Chem.
Phys. 103 (1995) 8577e8592.
[28] B. Hess, J. Chem. Theory Comput. 4 (2008) 116e122.
[29] E.F. Pettersen, T.D. Goddard, C.C. Huang, G.S. Couch, D.M. Greenblatt,
E.C. Meng, T.E. Ferrin, J. Comput. Chem. 25 (2004) 1605e1612.