Structure activity relationship studies of 3-arylsulfonyl-pyrido[1,2-a] pyrimidin-4-imines as potent 5-HT6 antagonists

Article abstract of DOI:10.1016/j.bmc.2014.01.003

Comprehensive structure activity relationship (SAR) studies were conducted on a focused screening hit, 2-(methylthio)-3-(phenylsulfonyl)-4H-pyrido[1,2-a] pyrimidin-4-imine (1, IC₅₀: 4.0 nM), as 5-HT₆ selective antagonists. Activity was improved some 2-4 fold when small, electron-donating groups were added to the central core as observed in 19, 20 and 26. Molecular docking of key compounds in a homology model of the human 5-HT₆ receptor was used to rationalize our structure-activity relationship (SAR) findings. In pharmacokinetic experiments, compound 1 displayed good brain uptake in rats following intra-peritoneal administration, but limited oral bioavailability.

Full text of DOI:10.1016/j.bmc.2014.01.003

Bioorganic & Medicinal Chemistry 22 (2014) 1782–1790
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Structure activity relationship studies of 3-arylsulfonyl-pyrido
[1,2-a]pyrimidin-4-imines as potent 5-HT6 antagonists
a
a
a
a
Shuanghua Hu a, , Yazhong Huang , Yong-Jin Wu , Huan He , Katherine A. Grant-Young ,
Robert L. Bertekap b, Valerie Whiterock c, Patrick Brassil c, Kimberley Lentz c, Prasanna Sivaprakasam d,
David R. Langley d, Ryan S. Westphal b, Paul M. Scola a
⇑
a Bristol-Myers Squibb Research and Development, Discovery Chemistry, 5 Research Parkway, Wallingford, CT 06492, USA
b Bristol-Myers Squibb Research and Development, Biology Neuroscience, 5 Research Parkway, Wallingford, CT 06492, USA
c Bristol-Myers Squibb Research and Development, Pharmaceutical Candidate Optimization, 5 Research Parkway, Wallingford, CT 06492, USA
d Bristol-Myers Squibb Research and Development, Computer Assisted Drug Design, 5 Research Parkway, Wallingford, CT 06492, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Comprehensive structure activity relationship (SAR) studies were conducted on a focused screening hit,
2-(methylthio)-3-(phenylsulfonyl)-4H-pyrido[1,2-a]pyrimidin-4-imine (1, IC50: 4.0 nM), as 5-HT6 selec-
tive antagonists. Activity was improved some 2–4 fold when small, electron-donating groups were added
to the central core as observed in 19, 20 and 26. Molecular docking of key compounds in a homology
model of the human 5-HT6 receptor was used to rationalize our structure–activity relationship (SAR)
findings. In pharmacokinetic experiments, compound 1 displayed good brain uptake in rats following
intra-peritoneal administration, but limited oral bioavailability.
Received 18 October 2013
Revised 20 December 2013
Accepted 1 January 2014
Available online 9 January 2014
Keywords:
5-HT6 antagonist
Cognition
Ó 2014 Elsevier Ltd. All rights reserved.
Alzheimer’s disease
Feeding behavior
Obesity
1. Introduction
patients with AD and schizophrenia.7–11 In preclinical studies,
pharmacological blockade of 5-HT6 receptors using selective
Serotonin, or 5-hydroxytryptamine (5-HT), is one of several ma-
jor neurotransmitters in the central nervous system. The effects of
this chemical messenger are mediated through interaction with a
family of serotonin receptors (5HT1–5HT7). These seven receptors
function cooperatively in the regulation of a variety of biological
processes including those associated with learning, memory, satia-
tion and mood modulation.1,2 The 5-HT6 receptor, first cloned in
1993,3 is almost exclusively found in the central nervous system
with the highest expression levels found in areas associated with
learning and memory.
In individuals afflicted with Alzheimer’s disease (AD), the loss of
cognitive function correlates with a decrease in cholinergic neuro-
transmission. Disruption of 5-HT6 receptor expression, or blockade
of 5-HT6 receptor signaling, increases regional cholinergic tone and
produces a behavioral syndrome that is reversed by centrally-act-
ing muscarinic antagonists, suggesting that the 5-HT6 receptor is
involved in the negative regulation of acetylcholine-mediated
neurotransmission.4–6 Correspondingly, it is anticipated that
5-HT6 receptor antagonists may improve cognitive function in
antagonists has been shown to improve the performance of
animals in a variety of cognition models. SB-742457 (Fig. 1),12 an
orally active 5-HT6 antagonist has advanced into phase II clinical
studies in mild to moderate AD patients. This compound has dem-
onstrated significant improvement in patients’ global function.
Moreover, the AD assessment scale-cognitive (ADAS-Cog) score of
patients treated with SB-742457 was comparable to patients trea-
ted with donepezil.13 The 5-HT6 antagonist Lu-AE58054 (Fig. 1),
has also entered phase II clinical trials for the treatment of cogni-
tive disorders,14 and was recently reported to significantly improve
cognitive performance in patients with moderate AD disease after
treatment for six months as add-on to donepezil.15
Many evidences also suggest that the 5-HT6 receptor plays an
important role in feeding behavior by affecting within- and post-
meal satiety, and 5-HT6 receptor has become an emerging target
for the treatment of obesity.16–18 For example, 5-HT6 receptor
knock-out mice were resistant to dietary-induced obesity via
reduction of food intake.19 A number of small molecule 5-HT6
receptor antagonists have been shown to dose-dependently reduce
food intake in rodents when dosed acutely, and produce sustained
reduction in food intake and weight loss when dosed chroni-
cally.19–21 Two small molecule 5-HT6 antagonists, PRX-07034
⇑
Corresponding author. Tel.: +1 203 677 6314; fax: +1 203 677 7702.
E-mail address: shuanghua.hu@bms.com (S. Hu).
0968-0896/$ - see front matter Ó 2014 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2014.01.003

S. Hu et al. / Bioorg. Med. Chem. 22 (2014) 1782–1790
1783
O
S
accompanied by concomitant generation of solvolysis derivative
2, the amount of which increased with reaction time. In the present
study, we found that formation of undesired product 2 could be
prevented by conducting the reaction in an inert solvent and in
the presence of a strong base. Specifically, the conjugate base
formed upon treatment of 2-aminopyridine with sodium bis(tri-
methylsilyl)amide in dioxane, was added to 4 (R1 = Ph) and the
resulting reaction mixture was heated at a temperature of 80 °C
overnight. Under these conditions, analogs of 1 with structural
variations at the benzenesulfonyl moiety (7–13), or with substitu-
ents on the fused pyridine ring (23–38) were formed in high yield.
It should be noted that the arylsulfonyl-3,3-bis-methylsulfanyl-
acrylonitriles 4, required for this methodology were prepared by
O
N
F
O
O
S
F
O
N
N
N
Lu-AE58054
SB-742457
N
H
HN
OMe
OMe
O
NH
H
O
OMe
N
N
H
N
Cl
S
S
O
O
O
O
adding the sodium salts of a-cyanoarylsulfones 3, to CS2 and sub-
sequently di-methylating the intermediate dithiocarboxylates
BVT-74316
PRX-07034
in situ using methyliodide.
Figure 1. 5-HT6 receptor antagonists that have entered clinical trials.
Further structural modifications to 1 could be achieved using
the methods depicted in Scheme 2.
(Fig. 1), and BVT-74316 (Fig. 1), have both advanced into clinical
evaluation for the treatment of obesity.16–18
Several groups have sought to develop novel 5-HT6 receptor
antagonists and numerous reports describing their discovery and
development have appeared in recent years.22–29 Drawing on this
literature, a 4-point pharmacophore model was constructed30
For example, the thiomethyl moiety in 1 could be readily dis-
placed with simple nucleophiles such as alcohols or amines which
served to provide alkoxy analogs 2 and 18, as well as amine deriv-
atives 19 and 20, respectively. Moreover, the imine of 1 was readily
functionalized by treatment with acetyl chloride in the presence of
pyridine to provide N-acetyl derivative 15, while N-methyl deriva-
tive 14, was obtained by treating 1 with trifluoromethanesulfo-
nate. Finally, oxidation of 1 with m-chloroperoxybenzoic acid
(mCPBA) afforded sulfoxide derivative 17. Upon synthesis, analogs
were evaluated in a radioligand binding assay that involved dis-
placement of tritium labeled Lysergic acid diethylamide [3H]LSD
bound to cloned 5-HT6 human receptors stably expressed in HeLa
cells and the results from these experiments are shown in Tables
1–5.
Structural modifications to the phenyl group in 1 were explored
first. As shown in Table 1 mono- or di-substitution of the benzene-
sulfonyl group of 1 by either a halogen, and or a methyl group, as in
analogs 7, 8, 9 and 11, was detrimental to 5-HT6 activity. Mono-
substitution by a bulky t-butyl group as in 10, or tri-substitution
as found in 11, further reduced affinity, as did replacement of the
phenyl ring by the larger 2-naphthyl group, as found in analog
13. Significant reduction in affinity was also observed when a tri-
fluoromethoxy group was introduced at the p-position of the phe-
nyl ring, as noted in 12. Moreover, replacement of the phenyl ring
in 1 by either a benzyl, or an alkyl moiety gave compounds with no
detectable binding activity (data not shown). This early SAR
suggested that direct attachment of a simple phenyl moiety to
the sulfonyl group was optimal for activity.
and subsequently employed in
a virtual screen of in-house
compounds. Based on these findings, selected compounds were as-
sayed for their activity at the 5-HT6 receptor and as described in an
earlier communication30 this exercise resulted in the identification
of 1, which proved to be a very potent 5-HT6 antagonist (IC50
:
4.0 nM). Importantly, 1 was highly selective for the 5-HT6 receptor,
as this compound was devoid of activity in a Cerep screen which
included a full panel of 5-HT receptors. Hence, with 1 as a starting
point, our objective was to develop SAR in this series with a focus
on further driving potency. In addition, we sought to understand
the pharmacokinetic profile of compounds in this series.
2. Results and discussion
The synthesis of 1 as previously described,30 invoked the meth-
odology illustrated in Scheme 1. Specifically, 2-aminopyridine
added to 2-benzenesulfonyl-3,3-bis-methylsulfanyl-acrylonitrile
4 (R1 = Ph), generating 1 in a single step. This reaction was
performed in ethanol and formation of the desired product was
O
S
NH
O
S
NH
O
O
N
N
+
H
H
S
N
O
N
O
S
N
O
S
N
O
O
1
R1R2NH
b
2
ROH
a
N
N
a
1
O
S
NH
O
R
N
N
N
O
R1
R2
S
S
CN
d
b,c
R1
N
2, 18
19-20
NC
SO2R1
R2
mCPBA
d
SO2R1
CH3COCl
MeOTf
S
N
3
e
4
c
7-13, 23-38
H
O
S
NMe
N
O
O
N
O
S
NC(O)Me
O
O
S
O
O
O
S
e
S
S
CO2Me
SO2Ph
N
b,c
N
N
N
MeO
SO2Ph
S
N
N
S
S
N
O
S
O
14
15
16
5
6
17
Scheme 1. Reagents and conditions: (a) 2-Aminopyridine, EtOH, reflux; (b)
CS2,(1 equiv), NaH (2 equiv), DMSO; (c) CH3I (2 equiv); (d) 2-aminopyridines,
NaN(Si(CH3)3)2 (1 equiv), Dioxane, 80 °C, overnight; (e); 2-aminopyridine, K2CO3,
dioxane, 80 °C, overnight.
Scheme 2. Reagents and conditions: (a) 90 °C, MeOH, 4 days, 50%; (b) RNH2
(1 equiv), Hg(OAc)2 (1 equiv), DMF, rt, 10 min, 27%; (c) CH3COCl, Py, CH2Cl2, rt,
overnight, 70%; (d) MeOTf (1.5 equiv), CH2Cl2, rt, ꢀ60%; (e) mCPBA (0.7 equiv),
CH2Cl2, 0 °C, 2 h, 50%.

1784
S. Hu et al. / Bioorg. Med. Chem. 22 (2014) 1782–1790
Table 1
Table 4
Radioligand binding assay results from displacement of [3H]LSD bound to cloned 5-
HT6 human receptors stably expressed in HeLa cells
Radioligand binding assay results from displacement of [3H]LSD bound to cloned 5-
HT6 human receptors stably expressed in HeLa cells
H
H
O
S
N
O
N
O
O
R1
R2
6
S
N
R
N
S
N
S
N
9
a
R1
R2
IC50 (nM)
a
Compound
Compound
R
IC50 (nM)
1
2330
24
25
26
H
7-F
H
H
H
H
H
H
H
H
H
H
H
H
9-Me
9-Br
9-Me
4.0
28
28
1
7
8
Ph
4-F-Ph
2-Cl-Ph
3,4-Di-Cl-Ph
4-t-Bu-Ph
3,6-Di-Me-4-Cl-Ph
4-CF3O-Ph
2-Naphthyl
4
22
11
18
65
50
420
59
7-Cl
7-Br
7-Me
7-CONH2
7-CO2Me
8-Me
8-Et
8-n-Pr
9-Me
9-Bn
7-Br
7-Me
7-Bn
56
9
1.3
8.8
470
41
10
11
12
1330
27
28
29
3030
31
80
a
IC50 values are the means of 2–4 experiments. [In this assay, the IC50 of cloza-
pine was 17 nM with a SEM of 2 nM.]
700
3.4
330
12
126
60
32
33
34
3530,b
36
Table 2
Radioligand binding assay results from displacement of [3H]LSD bound to cloned 5-
HT6 human receptors stably expressed in HeLa cells
a
IC50 values are the means of 2–4 experiments. [In this assay, the IC50 of cloza-
O
S
X
pine was 17 nM with a SEM of 2 nM.]
The poor solubility of 35 caused the variability in IC50 values between this work
and Ref. 30.
O
b
N
S
N
a
Table 5
Compound
X
IC50 (nM)
Radioligand binding assay results from displacement of [3H]LSD bound to cloned 5-
HT6 human receptors stably expressed in HeLa cells
1
14
15
16
NH
4.0
120
83
N–Me
NAc
O
H
N
O
S
O
52
A
B
N
a
IC50 values are the means of 2–4 experiments. [In this assay, the IC50 of cloza-
pine was 17 nM with a SEM of 2 nM.]
S
N
A
Table 3
a
Compound
IC50 (nM)
Radioligand binding assay results from displacement of [3H]LSD bound to cloned 5-
HT6 human receptors stably expressed in HeLa cells
B
H
1
4.0
55
N
O
S
O
N
3730
Y
N
2
a
Compound
Y
IC50 (nM)
1
17
18
2
19
20
21
–SMe
–S(O)Me
–OMe
–OEt
–NHMe
–NHEt
4.0
250
21
38
57
N
12
a
IC50 values are the means of 2–4 experiments. [In this assay, the IC50 of cloza-
pine was 17 nM with a SEM of 2 nM.]
2.4
2.0
42
–NH(CH2)2OH
H
N
22b
60
Structure–activity relationships at the 2-position of the pyr-
ido[1,2-a]pyrimidin-4-imine core were then considered (Table 3).
In the event, oxidation of the thiomethyl ether in 1 to the corre-
sponding sulfoxide 17, resulted a significant reduction in potency.
A more favorable outcome was observed with analogs bearing
lower alkoxy groups at this 2-position (18 and 2), however these
compounds were nonetheless 3- to 5-fold less active than 1. Addi-
tionally, analogs bearing functionalized amines at this position lost
activity (21, 22), however potency was enhanced 2-fold (vs 1) with
the employment of lower alkylamino moieties as in derivatives 19
and 20. The observation that simple amines could serve as a
replacement for the thio methyl group of 1, was significant given
our concerns a priori, regarding the metabolic liability of this thio
functionality. Moreover, with regard to the immediate SAR obser-
vations, the differential in activity between these simple amines
N
a
IC50 values are the means of 2–4 experiments. [In this assay, the IC50 of cloza-
pine was 17 nM with a SEM of 2 nM.]
b
Compound 22 had an additional 7-methyl group on the fused pyridine ring.
We next explored modifications to the imine functionality in 1
(Table 2). As previously described,30 the N–H group of 1 partici-
pates in an intramolecular hydrogen bond to the pendant sulfonyl
and this interaction proved essential for 5-HT6 potency. For exam-
ple, N-methylation or acetylation, as shown in analogs 14 and 15,
led to an approximately 16-fold and 30-fold decrease in affinity,
respectively. Alternatively, replacement of the imine functionality
with an oxo group as in 16, resulted in a 10-fold decrease in
potency.

S. Hu et al. / Bioorg. Med. Chem. 22 (2014) 1782–1790
1785
(19, 20) and the corresponding alkoxy analogues (18, 2) was of
interest and rationalized using the molecular modeling model
(vide infra).
We then turned to the development of SAR on the fused
pyridine ring of compound 1 as summarized in Table 4.
residue resulted in a 36-fold decrease in the antagonist activity of
SB-258585. The methyl group of the thiomethyl in 1 was forced
slightly out of plane of the central core to favorably interact with
hydrophobic residues in TM3 and TM5. However, many experi-
mental and theoretical studies in literature suggest a conforma-
tional preference of thiomethyl to be planar versus perpendicular
in thioanisoles.35 Docked pose of 19 was very similar to 1 with
an additional intramolecular hydrogen bond of the second sulfonyl
oxygen and the donor NH functionality of NHCH3 group. This in
turn would stabilize the binding conformation of 19 and may ac-
count for the 2-fold increase in potency of 19 when compared to 1.
Several of the most potent analogs in Tables 1–5 were subse-
quently evaluated in a functional assay in which 5-HT-stimulated
adenylyl cyclase activity was determined by measuring the conver-
A methyl substituent at the 7-position, as in 26, improved the
potency 4-fold, but related halogen-substituted analogs 23–25
gave compounds with about 10-fold less affinity. Comparable
affinity was observed with 9-methyl analogue 32, but a significant
decrease was seen with the corresponding 9-benzyloxy analog 33.
Methyl and ethyl substitutions at the 8-position as in derivatives
29 and 30, resulted in a 10- and 20-fold decrease in affinity respec-
tively, while further increases in substituent size, as in 31, resulted
in a further loss in activity. Interestingly, the activity of the 7-car-
boxamide derivative 27 was comparable to 1, whereas the corre-
sponding 7-methoxycarbonyl analogue was significantly less
active. From this data set it appeared that positions 7 and 9 were
amenable to further exploration as a means to drive potency.
Lastly, we investigated the impact of additional rings directly
fused onto the pyridine ring by synthesizing the analogs shown
in Table 5. Fused naphthyl or naphthyridine replacements as in
37 and 38 respectively, led to a 14-fold reduction in affinity. These
observations are consistent with the above findings, as substitu-
ents at the C8 position were not tolerated.
In order to understand the SAR in the context of the 5-HT6
receptor, docking studies were performed using 1 and 19. A homol-
ogy model of human 5-HT6 receptor was built from the sequence
(NP_000862.1)31 using the X-ray structure of b1-adrenergic G-pro-
tein-coupled receptor (pdb entry: 2VT4) as a template.32 Among
the several docked poses generated, the most highly populated
top scoring pose that agreed with the SAR and available mutagen-
esis studies in the literature33,34 is shown in Figure 2. The intramo-
lecular hydrogen bond between the imine –NH and the first
oxygen of sulfonyl group as observed in the X-ray structure of
130 was retained in the docked poses. Notably, residues capable
of complementing a hydrogen bond with the donor –NH function-
ality of imine were not found in proximity of the central core. The
intramolecular hydrogen bond in 1 favors a conformation which
places the phenyl group to optimally interact with hydrophobic
sion of [a-
33P]ATP to [33P]cAMP in HeLa cells expressing the cloned
human 5-HT6 receptor.37 In this assay, all compounds tested were
shown to be full antagonists, consistent with previous observa-
tions30 made with respect to derivative 1.
The pharmacokinetic parameters of compound 1 in rat were
evaluated. For example, intraperitoneal injection of compound 1
to rats at a dose of 10 mg/kg, provided brain levels of 1.9 lM and
a brain to plasma (B/P) ratio of 1.2, which was achieved thirty min-
utes post dosing. However, compound 1 demonstrated poor expo-
sure after oral dosing and an oral bioavailability of less than 1%.
This low exposure after PO dosing was likely due to a first pass ef-
fect, as the half life of 1 was only 5.2 min when incubated with
human liver microsomes in the presence of NADPH (Table 6), while
the permeability was acceptable (PC value of 170 nm/s). It should
be noted that in general the stability of compounds in this series
was similar in both rat and human microsomes.
Metabolic identification studies conducted on compound 1
indicated that a sulfoxide was identified as the major metabolite;
the result of oxidation of the thiomethyl moiety in 1. Interestingly,
only modest improvements in microsomal stability were observed
with analogs in which this metabolically labile thiomethyl group
was replaced with an alkoxy or amine, as in 18 or 19. Interestingly
however, a solution to metabolic stability in this series was found
with analogues in which the imine group was further funtionalized
or replaced. For example, compound 15 in which the imine is acet-
ylated, had a T1/2 of 43 min in the human liver microsomal assay,
while 16 the corresponding keto analog of 1, had a half life of
40 min. The impact of these changes on microsomal stability was
interesting as it suggested that oxidative stability could be
achieved by a structural modification remote to the position
oxidized. Unfortunately, both compounds 15 and 16 were only
modestly active as inhibitors of the 5-HT6 receptor and as a conse-
quence efforts in this immediate series were discontinued.
Attention then turned to the identification of replacements for
the pyrido[1,2-a]pyrimidin-4-imine core and these results will be
subsequently reported.
residues in the extracellular-2 (EC2) loop. An edge-to-face, CH–p,
of the fused pyridyl ring of 1 and F284 of transmembrane-6
(TM6) was observed. This may explain why the electron donating
groups when incorporated to the central core improved the activity
(cf: 26). It is noteworthy to mention that in the studies reported by
Harris et al.,34 the only point mutation that caused major effects of
their tested non-basic ligands is F284A. W281 (TM6) and F285
(TM6) are the other residues that were in proximity to the central
core of 1. The second lone pair of the first oxygen atom of the
sulfonyl group showed a hydrogen bond with N288 (TM6). Studies
by de la Fuente et al.,33 demonstrated that alanine mutation of this
Figure 2. Docked binding modes of 1 and 19 in the homology model of human 5-HT6 receptor. Ligands are shown in green carbon stick and critical residues that make
hydrogen bonds or hydrophobic contacts are highlighted in gray carbon stick. Intra- and inter-molecular hydrogen bonds observed in/with ligands are denoted as yellow
dotted lines. Non-polar hydrogen atoms are not shown for clarity. Graphics are created using Pymol.36

1786
S. Hu et al. / Bioorg. Med. Chem. 22 (2014) 1782–1790
Table 6
4.3. Syntheses of intermediates and final compounds
In vitro stability of compounds in human liver microsomes
a
a
2-(4-Chlorobenzenesulfonyl)-3,3-bis-methylsulfanyl-acryloni-
trile (4, 4-chlorobenzenesulfonyl-). To a solution of 2-(4-chloro-
benzenesulfonyl)acetonitrile (600 mg, 2.78 mmol) in anhydrous
DMSO (8 ml) under an argon atmosphere was added carbon
disulfide (212 mg, 2.78 mmol). The reaction mixture was cooled
to approximately 0 °C and 2 equiv of sodium hydride (5.56 mmol)
were added in one portion. The resulting thick reaction mixture
was swirled until homogeneous and allowed to warm to room
temperature for one hour with stirring. The reaction mixture was
cooled again to 0 °C and 2 equiv of iodomethane (5.56 mmol) were
added under argon. The reaction mixture was allowed to gradually
warm to room temperature and deionized water was added
(30 ml) after 18 h of stirring. The resulting slurry was stirred for
another 15 min. The solid was collected by filtration, washed with
deionized water, and dried. The crude solid product thus obtained
was dissolved in 25% hexanes/chloroform and purified via flash
chromatography to yield the titled compound as a light yellow
Compound
T1/2 (min)
Compound
T1/2 (min)
1
7
9
10
11
12
14
5.2
3.1
3.1
3.9
4.1
2.8
2.8
15
16
18
19
25
33
37
43
40
13.7
7.2
7.5
5.5
7.9
a
Compounds (0.5 lM) were incubated in human liver microsomes (1 mg/ml)
supplemented with NADPH (1 mM) for 0–45 min. The reactions were terminated
and the remaining parent compounds in the supernatants were measured by LC–
MS/MS. T1/2 is obtained from the plot of % remaining versus time. [In this assay, T1/2
of omeprazole was 26 1.4 min.]
3. Conclusion
In conclusion, we report here the synthesis of analogs and
derivatives of 2-(methylthio)-3-(phenylsulfonyl)-4H-pyrido[1,2-
a]pyrimidin-4-imine (1). These compounds have been used to de-
velop SAR at multiple sites, leading to improved inhibitory potency
for several compounds, among which the 7-methyl compound 26
(IC50: 1.3 nM), was the most potent. Docking studies were instruc-
tive in rationalizing the SAR in this novel series and deepened our
understanding of receptor–ligand interactions. In further charac-
terization of 1, a B/P ratio of 1.2 was observed in rat, however
the oral bioavailability was poor, limited by a first pass effect.
Structural solutions to the observed metabolic instability of 1 were
identified using a human microsomal assay however, modifica-
tions that provided oxidative stability, compromised 5-HT6
potency. These observations indicated that the discovery of an
orally active 5-HT6 receptor antagonist would require in-depth
structural changes to the core of analog 1, and efforts to this end
will be reported in due course.
l
solid in 65% yield. H NMR (300 MHz, CDCl3): d 8.0–7.9 (d, 2H),
7.6–7.5 (d, 2H), 2.7 (s, 3H), 2.6 (s, 3H).
4.3.1. 2-Benzenesulfonyl-3,3-bis-methylsulfanylacrylonitrile (4,
benzenesulfonyl-)
The titled compound was prepared as described above starting
l
from 2-(benzenesulfonyl)acetonitrile. H NMR (300 MHz, CDCl3): d
8.0 (d, 2H), 7.6 (t, 1H), 7.5 (t, 2H), 2.7 (s, 3H), 2.6 (s, 3H).
4.3.2. 3,3-Bis-methylsulfonyl-2-(naphthalene-2-sulfonyl)-
acrylonitrile (4, naphthalene-2-sulfonyl-)
The title compound was prepared as described above starting
from 2-(naphthalene-2-sulfonyl)acrylonitrile. lH NMR (300 MHz,
CDCl3): d 8.6 (s, 1H), 8.0–7.9 (m, 4H), 7.7–7.6 (m, 2H), 2.7 (s, 3H),
2.6 (s, 3H).
4.3.3. 2-(4-Chloro-2,5-dimethylbenzenesulfonyl)-3,3-bis-
methyl-sulfonylacrylonitrile (4, 4-chloro-2,5-
4. Materials and methods
dimethylbenzenesulfonyl-)
The title compound was prepared as described above starting
from 2-(4-chloro-2,5-dimethylbenzenesulfonyl)acetonitrile. lH
NMR (300 MHz, CDCl3): d 8.0 (s, 1H), 7.3 (s, 1H), 7.2 (s, 1H), 2.7
(s, 3H), 2.6 (s, 3H), 2.5 (s, 3H), 2.4 (s, 3H).
4.1. Radioligand binding assay at the 5-HT6 receptor
Compounds of the present invention were evaluated using a cell
line that stably expresses the human 5-HT6 receptor. Membranes
prepared from the cell line were incubated with 3H-LSD (1.6 nM)
in the presence of increasing concentrations (10À10 to 10À6 M) of
test compounds for 60 min at 37 °C. The reaction was terminated
by vacuum filtration of the membranes and washing of the filters
with assay buffer. Radioactivity retained on the filters was deter-
mined using liquid scintillation spectrometry. The potency of the
test compound to inhibit 3H-LSD binding to the 5-HT receptor
was defined as the concentration of compound required to inhibit
50% (IC50) of radioligand binding as determined by computer
assisted analysis of the data. Clozapine was used as an internal
standard for comparison.
4.3.4. 2-(4-Fluoro-benzenesulfonyl)-3,3-bis-methylsulfonyl-
acrylonitrile (4, 4-fluorobenzenesulfonyl)
The title compound was prepared as described above starting
l
from 2-(4-fluoro-benzenesulfonyl)acetonitrile. H NMR (300 MHz,
CDCl3): d 8.0 (m, 2H), 7.2 (m, 2H), 2.7 (s, 3H), 2.6 (s, 3H).
4.3.5. 2-(3,4-Dichloro-benzenesulfonyl)-3,3-bis-methylsulfonyl-
acrylonitrile (4, 3,4-dichlorobenzenesulfonyl)
The title compound was prepared as described above starting
from 2-(3,4-dichlorobenzenesulfonyl)acetonitrile. lH NMR (CDCl3):
d 8.1 (s, 1H), 7.9 (d, 1H), 7.6 (d, 1H), 2.7 (s, 3H), 2.6 (s, 3H).
4.2. Molecular modeling
4.3.6. 2-(2-Chloro-benzenesulfonyl)-3,3-bis-methylsulfonyl-
acrylonitrile (4, 2-chlorobenzenesulfonyl-)
Prime38 and Glide39 from Schrödinger suite of programs were
used for homology modeling and docking studies, respectively.
X-ray structure of 1 was used as starting geometry to build 19.
Blast search in Prime was used for finding homologous protein
and the GPCR-specific alignment module within Prime was used
for alignment. For docking, an active site around the key residues
D106, N288, F284, and W281 was considered and 10 poses of each
ligand were allowed to evolve. Default settings from Prime and
Glide were used unless otherwise specified.
The title compound was prepared as described above starting
from 2-(2-chlorobenzenesulfonyl)acetonitrile. H NMR (CDCl3): d
8.3 (d, 1H), 7.6–7.4 (m,3H), 2.7 (s, 3H), 2.4 (s, 3H).
l
4.3.7. 2-(4-t-Butyl-benzenesulfonyl)-3,3-bis-methylsulfonyl-
acrylonitrile (4, 4-t-butyl-benzenesulfonyl-)
The title compound was prepared as described above starting
l
from 2-(4-t-butylbenzenesulfonyl)acetonitrile. H NMR (CDCl3): d
8.0 (d, 2H), 7.6 (d, 2H), 2.7 (s, 3H), 2.6 (s, 3H), 1.4 (s, 9H).

S. Hu et al. / Bioorg. Med. Chem. 22 (2014) 1782–1790
1787
4.3.8. 2-(4-Trifiuoromethoxybenzenesulfonyl)-3,3-
4.3.11. 3-(4-FIuorobenzenesulfonyl)-2-
bismethylsulfonyl-acrylonitrile (4, 4-
trifiuoromethoxybenzenesulfonyl-)
The title compound was prepared as described above starting
from 2-(4-trifiuoromethoxybenzenesulfonyl)acetonitrile. lH NMR
(CDCl3): d 8.1 (d, 2H), 7.4 (d, 2H), 2.7 (s, 3H), 2.6 (s, 3H).
methylsulfanylpyrido[1,2-a]pyrimidin-4-ylideneamine (7)
The title compound was prepared by method D. lH NMR (CDCl3)
d 9.4 (d, 7.3, 1H), 9.0 (s, br, 1H), 8.1 (m, 2H), 7.8 (ddd, 8.7, 6.9, 1.6,
1H), 7.3 (d, 7.3, 1H), 7.2 (m, 2H), 7.1 (td, 7.0, 1.5, 1H), 2.5 (s, 3H).
4.3.12. 3-(2-Chlorobenzenesulfonyl)-2-
4.3.9. 2-Benzenesulfonyl-3,3-bis-methylsulfanyl-acrylic acid
methyl ester (6)
methylsulfanylpyrido[1,2-a]pyrimidin-4-ylideneamine (8)
The title compound was prepared by method D. lH NMR
(CDCl3): d 9.4 (d, 1H), 9.1 (s, 1H), 8.4 (d, 1H), 7.8 (t, 1H), 7.5 (t,
3H), 7.3 (d, 1H), 7.0 (t, 1H), 2.4 (s, 3H).
Methyl phenylsulfonylacetate (5 g, 23.3 mmol) was mixed with
carbon disulfide (1.77 g, 23.3 mmol) in anhydrous DMSO (100 ml)
at room temperature. Sodium hydride (60% in mineral oil, 1.2 g,
48.9 mmol) was added portionwise. After stirring 1 h at room tem-
perature, neat iodomethane (6.9 g, 48.9 mmol) was added and the
reaction mixture was stirred overnight. The reaction mixture was
quenched by water, and the resulting mixture extracted with
dichloromethane. The organic layer was concentrated and purified
via flash chromatography to provide crude 2-benzenesulfonyl-3,3-
bis-methylsulfanyl-acrylic acid methyl ester, which was further
purified by recrystallization from methanol to yield 3.7 g of white
solid. lH NMR (500 MHz, CDCl3): d 8.1 (d, 2H), 7.6 (t, 1H), 7.5 (t, 2H),
3.9 (s, 3H), 2.4 (s, 3H), 2.2 (s, 3H).
4.3.13. 3-(3,4-Dichlorobenzenesulfonyl)-2-methylsulfanyl-
pyrido[1,2-a]pyrimidin-4-ylideneamine (9)
The title compound was prepared by method D. lH NMR
(CDCl3): d 9.6 (d, 1H), 8.1 (s, 1H), 7.9–7.8 (m, 3H), 7.6 (d, 1H), 7.4
(d, 1H), 7.1 (t, 1H), 2.5 (s, 3H).
4.3.14. 3-(4-t-Butylbenzenesulfonyl)-2-
methylsulfanylpyrido[1,2-a]pyrimidin-4-ylideneamine (10)
The title compound was prepared by method D. lH NMR
(CDCl3): d 9.4 (dd, 7.3, 1.5, 1H), 9.1 (s, br, 1H), 8.0 (d,7.3, 2H), 7.7
(ddd, 8.7, 6.9, 1.6, 1H), 7.5 (d, 7.0, 2H), 7.3 (d, 9.0, 1H), 7.0 (td,
7.1, 1.5, 1H), 2.5 (s, 3H), 1.3 (m,9H).
Compounds 7–13 and 23–38 were prepared according to meth-
ods A–D described below.
Method A: In an one dram vial fitted with a TEFLONÒ lined cap,
substituted 2-aminopyridine (0.12 mmol) was dissolved or sus-
4.3.15. 3-(2,5-Dimethyl-4-chlorobenzenesulfonyl)-2-
pended in 200
bis(trimethylsilyl)amide (120
After 15 min, 2-arylsulfonyl-3,3-bis-methylsulfanyl-acrylonitrile
4 in dioxane (120 l, 0.12 mmol) was added to the reaction vial,
l
l of dioxane at room temperature and sodium
methylsulfanyl-pyrido[1,2-a]pyrimidin-4-ylideneamine (11)
The title compound was prepared by method D. lH NMR
(CDCl3): d 9.5 (s, br, 1H), 8.1 (s, br, 1H), 7.8 (t, 7.9, 1H), 7.4 (d,
8.8, 1H), 7.2 (d, 9.0, 2H), 7.1 (t, 6.8, 1H), 2.5 (s, 3H), 2.4 (s, 6H).
ll in THF, 0.12 mmol) was added.
l
which was then subjected to heating at 80 °C in a metal heating
block with shaking overnight. The reaction mixture was concen-
trated, re-dissolved in CH2CI2, and washed with aqueous NH4Cl
solution. The organic layer was concentrated in vacuo; the residue
was then taken up in a DMF/MeOH mixture (1:1) and subjected to
reverse phase preparative HPLC (XTerra™ MS C18 5.0 mM) using
MeOH/H2O mixture as the eluent. The product was collected after
fractions were concentrated in vacuo.
4.3.16. 3-(4-Trifluoromethoxy-benzenesulfonyl)-2-
methylsulfanyl-pyrido[1,2-a]pyrimidin-4-ylideneamine (12)
The title compound was prepared by method D. lH NMR
(CDCl3): d 9.3 (dq, 7.3, 0.7, 1H), 9.0 (s, br, 1H), 8.1 (m, 2H), 7.7
(ddd, 8.8, 6.8, 1.8, 1H), 7.3 (dt, 8.8, 0.8, 1H), 7.3 (m, 2H), 7.0 (td,
7.0, 1.5, 1H), 2.4 (s, 3H).
Method B: In an one dram vial fitted with a TEFLONÒ lined cap,
4.3.17. 2-Methylsulfanyl-3-(naphthalene-2-sulfonyl)-
pyrido[1,2-a]pyrimidin-4-ylideneamine (13)
substituted 2-aminopyridine (0.6 mmol) was dissolved in 400
DMF at room temperature and 2-arylsulfony1-3,3-bis-meth-
ylsulfanyl-acrylonitrile in dimethylformamide (120 l,
ll
The title compound was prepared by method D. lH NMR
(CDCl3): d 9.6 (d, 6.2, 1H), 8.6 (s, br, 1H), 8.0 (d, 7.0, 2H), 7.9 (m,
1H), 7.8 (m, 4H), 7.4 (d, 7.3, 2H), 7.1 (t, 7.1, 1H), 2.5 (s, 3H).
4
l
0.12 mmol) was added. The reaction mixture was subjected to
heating at 80 °C in a metal heating block with shaking overnight.
After cooling to room temperature, the reaction mixture was
further diluted with DMF and/or MeOH and subjected directly to
reverse phase preparative HPLC using TFA (trifiuoroacetic acid,
0.1%) buffered MeOH/H2O mixture as the eluent. The product
was collected after fractions were concentrated in vacuo.
Method C: Same as in method B except that the reaction was
carried out in anhydrous ethanol.
4.3.18. 3-Benzenesulfonyl-7-fluoro-2-methylsulfanyl-
pyrido[1,2-a]pyrimidin-4-ylideneamine (23)
The title compound was prepared by method A. lH NMR
(500 MHz, CDCl3): d 9.33 (1H, dd, 5.2 Hz, 3.0 Hz), 9.11 (1H, s),
8.07 (2H, dd, 8.3 Hz, 1.0 Hz), 7.66 (1H, m), 7.59 (1H, t, 7.3 Hz),
7.49 (2H, t, 7.6 Hz), 7.35 (1H, dd, 9.5 Hz, 5.2 Hz), 2.48 (3H, s).
Method D: To a one dram vial was added 20 mg of 2-arylsulfo-
nyl-3,3-bis-methylsulfanyl-acrylonitrile along with 2 equiv of
4.3.19. 3-Benzenesulfonyl-7-chloro-2-methylsulfanyl-
pyrido[1,2-a]pyrimidin-4-ylideneamine (24)
substituted 2-aminopyridine in 750
l
l of DMF. The vial was fitted
The title compound was prepared by method A. lH NMR
(500 MHz, CDCl3): d 9.41 (1H, d, 2.4 Hz), 9.13 (1H, br, s), 8.06
(2H, dd, 8.3 Hz, 1.0 Hz), 7.67 (1H, dd, 9.5 Hz, 2.4 Hz), 7.60 (1H, t,
7.3 Hz), 7.49 (2H, t, 7.6 Hz), 7.28 (1H, d, 9.5 Hz), 2.49 (3H, s). The
mono TFA salt was prepared by method B. Mono TFA salt: lH
NMR (500 MHz, CDCl3): d 9.56 (IH, s), 8.06 (2H, d, 7.3 Hz), 7.91
(1H, dd, 9.5 Hz, 1.8 Hz), 7.67 (1H, t, 7.6 Hz), 7.52–7.58 (3H, m),
2.57 (3H, s).
with a TEFLONÒ lined cap and heated to 80 °C in a sand bath for
2 days. The crude product was allowed to return to room temper-
ature, diluted to 2 ml with methanol or acetonitrile and purified on
a reverse phase preparative HPLC column (XTerra™ MS C18
5.0 mM, MeOH/H2O as the eluent) to give the title compound.
4.3.10. 3-Benzenesulfonyl-2-methylsulfanyl-pyrido[1,2-
a]pyrimidin-4-ylideneamine (1)
The title compound was prepared by method C. lH NMR
(300 MHz, CDCl3): d 9.43–9.41 (1H, br d), 8.09–8.06 (2H, m),
7.80–7.75 (1H, m), 7.58–7.56 (1H, m), 7.51–7.46 (2H, m), 7.37-
7.34 (1H, m), 7.26 (1H, s), 7.08–7.04 (1H, m), 2.49 (3H, s).
4.3.20. 3-Benzenesulfonyl-7-bromo-2-methylsulfanyl-
pyrido[1,2-a]pyrimidin-4-ylideneamine (25)
The title compound was prepared by method A. lH NMR
(500 MHz, CDCl3): d 9.50 (1H, d, 2.2 Hz), 9.11 (1H, br, s), 8.06

1788
S. Hu et al. / Bioorg. Med. Chem. 22 (2014) 1782–1790
(2H, dd, 8.3 Hz, 1.0 Hz), 7.77 (1H, dd, 9.3 Hz, 2.2 Hz), 7.59 (1H, t,
7.3 Hz), 7.49 (2H, m), 7.21 (1H, d, 9.3 Hz), 2.48 (3H, s). The mono
TFA salt was also prepared by method B. Mono TFA salt: H NMR
(500 MHz, CDCl3): d 9.64 (1H, s), 8.08 (2H, d, 7.6 Hz), 8.05 (1H, d,
9.4 Hz), 7.69 (1H, t, 7.6 Hz), 7.56 (2H, t, 7.6 Hz), 7.53 (1H, d,
9.2 Hz), 2.59 (3H, s).
4.3.28. 3-Benzenesulfonyl-9-benzyloxy-2-methylsulfanyl-
pyrido[1,2-a]pyrimidin-4-ylideneamine, mono TFA salt (33)
The title compound was prepared by method B. lH NMR
(500 MHz, CDCl3): d 9.14 (1H, d, 7.0 Hz), 8.11(2H, d, 7.3 Hz), 7.71
(1H, t, 7.6 Hz), 7.58 (3H, m), 7.35–7.49 (6H, m), 5.30 (2H, s), 2.64
(3H, s).
l
4.3.29. 3-Benzenesulfonyl-7-bromo-9-methyl-2-
4.3.21. 3-Benzenesulfonyl-7-methyl-2-methylsulfanyl-
pyrido[1,2-a]pyrimidin-4-ylideneamine (26)
methylsulfanyl-pyrido[1,2-a]pyrimidin-4-ylideneamine (34)
The title compound was prepared by method A. lH NMR
(500 MHz, CDCl3): d 9.40 (1H, d, 1.8 Hz), 9.08 (1H, s), 8.06 (2H,
dd, 8.3 Hz, 1.0 Hz), 7.68 (1H, d, 2.1 Hz), 7.59 (1H, t, 7.6 Hz), 7.49
(2H, t, 7.6 Hz), 2.50 (3H, s), 2.43 (3H, s).
The title compound was prepared by method A. lH NMR
(500 MHz, CDCl3): d 9.19 (1H, s), 9.04 (1H, br, s), 8.07 (2H, dd,
8.3 Hz, 1.0 Hz), 7.63 (1H, dd, 9.0 Hz, 2.0 Hz), 7.57 (1H, t, 7.3 Hz),
7.48 (2H, t, 7.6 Hz), 7.29 (1H, d, 8.9 Hz), 2.49 (3H, s), 2.37 (3H, s).
The mono TFA salt was prepared by method B: lH NMR
(500 MHz, CDCl3): d 9.25 (IH, s), 8.09 (2H, d, 7.6 Hz), 7.99 (1H, d,
9.1 Hz), 7.71 (1H, d, 8.9 Hz), 7.70 (1H, t, 7.0 Hz), 7.57 (2H, t,
7.6 Hz), 2.60 (3H, s), 2.55 (3H, s).
4.3.30. 3-Benzenesulfonyl-9-bromo-7-methyl-2-
methylsulfanyl-pyrido[1,2-a]pyrimidin-4-ylideneamine (35)
The title compound was prepared by method A. lH NMR
(500 MHz, CDCl3): d 9.17 (1H, s), 9.12 (1H, s), 8.06 (2H, dd,
8.3 Hz, 1.2 Hz), 7.98 (1H, d, 1.8 Hz), 7.58 (1H, t, 7.6 Hz), 7.48 (2H,
t, 7.6 Hz), 2.57 (3H, s), 2.35 (3H, s).
4.3.22. 3-Benzenesulfonyl-4-imino-2-methylsulfanyl-4H-
pyrido[1,2-a]pyrimidine-7-carboxylic acid amide, mono TFA
salt (27)
4.3.31. 3-Benzenesulfonyl-7-benzyl-9-methyl-2-
The title compound was prepared by method B. lH NMR
(500 MHz, CDCl3): d 9.96 (1H, s), 8.87 (1H, s), 8.71 (1H, dd,
9.1 Hz, 1.5 Hz), 8.11 (2H, dd, 8.2 Hz, 1.2 Hz), 7.81 (1H, d, 9.1 Hz),
7.74 (1H, t, 7.6 Hz), 7.61 (2H, m), 5.92 (1H, s), 2.66 (3H, s).
methylsulfanyl-pyrido[1,2-a]pyrimidin-4-ylideneamine (36)
The title compound was prepared by method A. lH NMR
(500 MHz, CDCl3): d 9.19 (1H, s), 9.01 (1H, br, s), 8.09 (2H, dd,
8.3 Hz, 1.0 Hz), 7.57 (1H, t, 7.6 Hz), 7.48 (2H, t, 7.6 Hz), 7.43 (1H,
s), 7.31 (2H t, 7.6 Hz), 7.25 (1H, t, 7.3 Hz), 7.17 (2H, d, 7.6 Hz),
3.93 (2H, s), 2.49 (3H, s), 2.38 (3H, s).
4.3.23. 3-Benzenesulfonyl-4-imino-2-methylsulfanyl-4H-
pyrido[1,2-a]pyrimidine-7-carboxylic acid methyl ester (28)
The title compound was prepared by method A. lH NMR
(500 MHz, CDCl3): d 9.97 (1H, d, 2.1 Hz), 9.28 (1H, s), 8.19 (1H,
dd, 9.2 Hz, 1.5 Hz), 8.07 (2H, d, 7.9 Hz), 7.60 (1H, t, 7.6 Hz), 7.50
(2H, t, 7.6 Hz), 7.29 (1H, d, 9.2 Hz), 3.95 (3H, s), 2.50 (3H, s).
4.3.32. 3-Benzenesulfonyl-2-methylsulfanyl-pyrimido[2,1-
a]isoquinolin-4-ylideneamine, mono TFA salt (37)
The title compound was prepared by method B. lH NMR
(500 MHz, CDCl3): d 9.14 (1H, d, 7.3 Hz), 9.00 (1Hd, 8.2 Hz), 8.14
(2H, d, 7.6 Hz), 8.05 (1H, t, 7.0 Hz), 7.94 (1H, d, 7.9 Hz), 7.89 (1H,
t, 7.7 Hz), 7.85 (1H, d, 7.0 Hz), 7.72 (1H, t, 7.3 Hz), 7.59 (2H, t,
7.6 Hz), 2.82 (3H, s).
4.3.24. 3-Benzenesulfonyl-8-methyl-2-methylsulfanyl-
pyrido[1,2-a]pyrimidin-4-ylideneamine (29)
The title compound was prepared by method A. lH NMR
(500 MHz, CDCl3): d 9.26 (1H, d, 7.3 Hz), 9.02 (1H, br, s), 8.07
(2H, dd, 8.3 Hz, 1.0 Hz), 7.57 (1H, t, 7.3 Hz), 7.48 (2H, m), 7.15
(1H, s), 6.88 (1H, dd, 7.3 Hz, 1.8 Hz), 2.48 (3H, s), 2.44 (3H, s).
The title compound was also prepared by method B. Mono TFA
4.3.33. 7-Benzenesulfonyl-6-methylsulfanyl-4,5,8atriaza-
phenanthren-8-ylideneamine, mono TFA salt (38)
The title compound was prepared by method B. lH NMR
(500 MHz, CDCl3): d 9.81 (1H, d, 7.6 Hz), 9.22 (1H, dd, 4.6 Hz,
1.6 Hz), 8.26 (1H, dd, 8.2 Hz, 1.5 Hz), 8.13 (2H, dd, 8.5 Hz, 1.2 Hz),
7.91 (1H, dd, 8.2 Hz, 4.6 Hz), 7.83 (1H,d, 7.6 Hz), 7.73 (1H, t,
7.3 Hz), 7.59 (2H, t, 7.6 Hz), 2.90 (3H, s).
l
salt: H NMR (500 MHz, CDCl3): d 9.42 (IH, d, 6.7 Hz), 8.08 (2H, d,
7.3 Hz), 7.70 (1H, t, 7.6 Hz), 7.52–7.62 (4H, m), 2.62 (3H, s),
2.60(3H, s).
4.3.25. 3-Benzenesulfonyl-8-ethyl-2-methylsulfanyl-pyrido[1,2-
a]pyrimidin-4-ylideneamine (30)
4.3.34. 3-Benzenesulfonyl-2-ethoxy-pyrido[1,2-a]pyrimidin-4-
ylideneamine, mono TFA salt (2)
The title compound was prepared by method A. lH NMR
(500 MHz, CDCl3): d 9.27 (1H, d, 7.3 Hz), 9.02 (1H, br, s), 8.07
(2H, dd, 8.3 Hz, 1.3 Hz), 7.57 (1H, t, 7.6 Hz),7.48 (2H, t, 7.6 Hz),
7.15 (1H, s), 6.91 (1H, dd, 7.3 Hz, 2.0 Hz), 2.73 (2H, q, 7.6 Hz),
2.49 (3H, s), 1.30 (3H, t, 7.6 Hz).
The title compound was produced as a minor product when the
preparation of 3-benzenesulfonyl-2-methylsulfanyl-pyrido[1,2-
a]pyrimidin-4-ylideneamine was carried out in ethanol (method
l
C). H NMR (500 MHz, CDCl3): d 9.67 (1H, br), 8.16 (1H, br), 8.03
(2H, d, 7.6 Hz), 7.69 (3H, m), 7.58 (2H, t, 8.2 Hz), 4.55 (2H, q,
7.3 Hz), 1.30 (3H, t, 7.3 Hz).
4.3.26. 3-Benzenesulfonyl-2-methylsulfanyl-8-propyl-
pyrido[1,2-a]pyrimidin-4-ylideneamine (31)
4.3.35. 3-Benzenesulfonyl-2-methylsulfanyl-pyrido[1,2-
a]pyrimidin-4-ylidene)-methylamine (14)
The title compound was prepared by method A. lH NMR
(500 MHz, CDCl3): d 9.27 (1H, d, 7.6 Hz), 9.02 (1H, br, s), 8.07
(2H, dd, 8.3 Hz, 1.0 Hz), 7.57 (1H, t, 7.6 Hz), 7.47 (2H, t, 7.6 Hz),
7.13 (1H, s), 6.89 (1H, dd, 7.3 Hz, 1.8 Hz), 2.66 (2H, t, 7.6 Hz),
2.49 (3H, s), 1.71 (2H, m), 0.98 (3H, t, 7.6 Hz).
3-Benzenesulfonyl-2-methylsulfanyl-pyrido[1,2-a]pyrimidin-
4-ylideneamine (100 mg, 0.30 mmol) was dissolved in anhydrous
dichloromethane (1 ml). Methyl trifluoromethanesulfonate
(74 mg, 0.45 mmol) was added at room temperature. The reaction
was quenched with saturated aqueous NaHCO3 solution at about
65% conversion. The crude product was extracted with dichloro-
methane and purified by preparative HPLC (XTerra™ MS C18
5.0 mM, MeOH/H2O as the eluent) to yield 41 mg of the product
as an orange solid. lH NMR (300 MHz, CDCl3): d 8.91 (1H, d,
7.0 Hz), 7.90–7.88 (2H, m), 7.77–7.71 (1H, m), 7.49–7.39 (3H, m),
7.05–7.00 (1H, m), 3.43 (3H, s), 2.33 (3H, s).
4.3.27. 3-Benzenesulfonyl-9-methyl-2-methylsulfanyl-
pyrido[1,2-a]pyrimidin-4-ylideneamine (32)
The title compound was prepared by method A. lH NMR
(500 MHz, CDCl3): d 09.27 (1H, d, 6.4 Hz), 9.05 (1H, s), 8.09 (2H,
dd, 8.3 Hz, 1.0 Hz), 7.64 (1H, d, 7.0 Hz), 7.58(1H, t, 7.6 Hz), 7.49
(2H, t, 7.6 Hz), 6.96 (1H, t, 7.0 Hz), 2.51 (3H, s), 2.46 (3H, s).

S. Hu et al. / Bioorg. Med. Chem. 22 (2014) 1782–1790
1789
4.3.36. N-(3-Benzenesulfonyl-2-methylsulfanyl-pyrido[1,2-
4.3.41. 3-(Benzenesulfony 1-4-imine-4H -pyrido [1,2-
a]pyrimidin-4-ylidene)-acetamide (15)
a]pyrimidin-2-yl)-ethylamine (20)
3-Benzenesulfonyl-2-methylsulfanyl-pyrido[1,2-a]pyrimidin-
4-ylideneamine (100 mg, 0.30 mmol) was dissolved in anhydrous
dichloromethane (2 ml) to which pyridine (48 mg, 0.6 mmol) was
added. Acetyl chloride was then added at room temperature and
the reaction mixture was stirred overnight. The reaction was
quenched with saturated aqueous NaHCO3 solution, extracted with
dichloromethane, and purified by flash chromatography to yield
79 mg of a yellow solid product. lH NMR (300 MHz, CDCl3): d
8.99 (1H, d), 8.15–8.12 (1H, m), 7.94–7.88 (1H, m), 7.60–7.46
(3H, m), 7.26–7.17 (3H, m), 2.53 (3H, s), 2.30 (3H, s).
The title compound was prepared in an analogous fashion to 19
using ethylamine in place of methylamine. H NMR (CDCl3): d 9.2
(d, 1H), 8.3 (s, br, 1H), 8.1 (s, br, 1H), 8.0 (d, 1H), 7.6–7.4 (m, 5H),
7.1 (d, 1H), 6.8 (t, 1H), 3.6 (m, 2H), 1.2 (t, 3H).
l
4.3.42. 2-(3-Benzenesulfony 1-4-imino-4H-pyrido[1,2-
a]pyrimidin-2-ylamino)ethanol (21)
To an one dram vial was added 20 mg (0.06 mmol) of 3-(ben-
zenesulfonyl)-2-methylsulfanyl-pyrido[1,2-a]pyrimidin-4-ylid-
eneamine and 1.1 equiv of ethanolamine in 750 ml of DMF. The
vial was fitted with a TEFLONÒ lined cap and heated to 55 °C over
night. The resulting product was then purified on a reverse phase
preparative HPLC (XTerra™ MS C18 5.0 mM, MeOH/H2O as the elu-
ent) to give 23 mg of the title compound as a brown solid. lH NMR
(300 MHz, CDCl3): d 9.3 (d, 1H), 8.5 (s, br, 2H), 8.0 (d, 2H), 7.7 (t,
1H), 7.6 (t, 1H), 7.5 (t, 2H), 7.2 (d,1H), 6.9 (t, 1H), 3.9 (m, 2H), 3.8
(m, 2H).
4.3.37. 3-Benzenesulfonyl-2-methylsulfanyl-pyrido[1,2-
a]pyrimidin-4-one (16)
2-Benzenesulfonyl-3,3-bis-methylsulfanyl-acrylic acid methyl
ester (6, 670 mg, 2.1 mmol) was mixed with 2-aminopyridine
(296 mg, 3.15 mmol) and K2CO3 (290 mg, 2.1 mmol) in DMF (2 ml).
The reaction mixture was heated at 100 °C overnight and then cooled
down to room temperature. The reaction mixture was diluted with
dichloromethane (10 ml), washed with saturated aqueous NaCl
solution, dried with Na2SO4 and concentrated. The desired product
was obtained (30 mg) as an off-white solid after purifications by
4.3.43. 3-Benzenesulfonyl-4-imino-7-methyl-4H-pyrido[1,2-
a]pyrimidin-2-yl)-(5-methylpyridin-2-yl)amine, mono TFA
salt(22)
l
preparative HPLC and flash chromatography. H NMR (300 MHz,
The title compound was obtained as a minor side product dur-
ing the preparation, by method B, of 3-benzenesulfonyl-7-methyl-
2-methylsulfanyl-pyrido[1,2-a]pyrimidin-4-ylideneamine, mono
CDCl3): d 8.89 (1H, d, J = 7.0 Hz), 8.21–8.18 (2H, d), 7.89–7.86 (1H,
m),7.57–7.47 (4H, m), 7.18–7.16 (1H, m), 2.58 (3H, s).
l
TFA salt (26). H NMR (500 MHz, CDCl3): d 8.99 (1H, s), 8.35 (1H,
4.3.38. 3-Benzenesulfonyl-2-methanesulfinyl-pyrido[1,2-
a]pyrimidin-4-ylideneamine (17)
s), 8.07 (2H, d, 7.6 Hz), 8.05 (1H, d, 8.5 Hz), 7.84 (1H, d, 9.1 Hz),
7.67–7.73 (2H, m), 7.57 (2H, t, 7.6 Hz), 7.52 (1H, d, 8.9 Hz), 2.46
(3H, s), 2.41 (3H, s).
3-Benzenesulfonyl-2-methylsulfanyl-pyrido[1,2-a]pyrimidin-
4-ylideneamine (50 mg, 0.15 mmol) was dissolved in anhydrous
dichloromethane (2 ml) to which 3-chloroperoxybenzoic acid
powder (32 mg, 50–60%, ꢀ0.11 mmol) was added at 0 °C and
stirred for 2 h. 2,6-tert-butyl-4-methylpyridine (38 mg, 90%,
0.17 mmol) was added to neutralize the reaction mixture. After
concentration, the crude mixture was purified by preparative HPLC
(XTerra™ MS C18 5.0 mM, MeOH/H2O as the eluent) to yield 26 mg
of product as a pale yellow solid. lH NMR (300 MHz, CDCl3): d 9.52
(1H, d), 9.34 (1H, br s), 8.04–7.94 (3H, m), 7.86–7.83 (1H, m), 7.64–
7.61 (1H, m), 7.56–7.51 (2H, m), 7.29–7.28 (1H, m), 3.06 (3H, s).
Acknowledgments
The authors would like to thank Drs. Robert Gentles and Louis
Chupak for their help discussions of this manuscript.
References and notes
1. Codony, X.; Burgueno, J.; Ramirez, M. J.; Vela, J. M. Int. Rev. Neurobiol. 2010, 94,
89.
2. Dawson, L. A. Int. Rev. Neurobiol. 2011, 96, 1.
3. Monsma, F. J., Jr.; Shen, Y.; Ward, R. P.; Hamblin, M. W.; Sibley, D. R. Mol.
Pharmcol. 1993, 43, 320.
4. Bourson, A.; Borroni, E.; Austin, R. H.; Monsma, F. J., Jr.; Sleight, A. J. J.
Pharmacol. Exp. 1996, 274, 173.
5. Sleight, A. J.; Monsma, F. J., Jr.; Borroni, E.; Austin, R. H.; Bourson, A. Behav. Brain
Res. 1995, 73, 245.
6. Bentley, J. C.; Bourson, A.; Boess, F. G.; Fone, K. C. F.; Marsden, C. A. Br. J. Pharm.
1999, 126, 1537.
4.3.39. 3-Benzenesulfonyl-2-methoxy-pyrido[1,2-a]pyrimidin-
4-ylideneamine (18)
3-Benzenesulfonyl-2-methylsulfanyl-pyrido[1,2-a]pyrimidin-
4-ylideneamine (50 mg, 0.15 mmol) was dissolved in methanol
(2 ml) and heated at 90 °C for 4 days. When the conversion reached
about 50% the product was isolated from the starting material by
preparative HPLC (XTerra™ MS C18 5.0 mM, MeOH/H2O as the elu-
7. Liu, K.; Robichaud, A. Drug Dev. Res. 2009, 70, 145.
8. Da Silva Costa-Aze, V.; Dauphin, F.; Boulouard, M. Behav. Brain Res. 2011, 222,
134.
l
ent) to yield 21 mg of bright yellow product. H NMR (300 MHz,
9. Codony, X.; Vela, J. M.; Ramirez, M. J. Curr. Opin. Pharmacol. 2011, 11, 94.
10. Chaudhary, S.; Mahesh, R.; Gautam, B.; Jindal, A.; Bhatt, S. Int. Res. J. Pharm.
2010, 1, 7.
11. Meneses, A.; Perez-Garcia, G.; Ponce-Lopez, T.; Castillo, C. Int. Rev. Neurobiol.
2011, 96, 27.
CDCl3): d 9.55 (1H, d), 9.45 (1H, br s), 8.01–7.98 (2H, m), 7.83–
7.77 (1H, m), 7.55–7.45 (3H, m), 7.30 (1H, d, 8.8 Hz), 7.13–7.08
(1H, m), 3.89 (3H, s).
12. de Bruin, N. M. W. J.; Kloeze, B. M.; McCreary, A. C. Neuropharmacology 2011,
61, 451.
4.3.40. 3-Benzenesulfonyl-4-imino-4H-pyrido[1,2-a]pyrimidin-
2-yl-methylamine (19)
13. Hedgecock, C. Presented at the Fourth Annual Anglo-Swedish Medicinal
Chemistry Symposium (Part II), Lund, Sweden, 15–18 March, 2009.
14. Anrt, J.; Bang-Anderson, B.; Grayson, B.; Bymaster, F. P.; Cohen, M. P.; Delapp,
N. W.; Giethlen, B.; Kreilgaard, M.; Mckinzie, D. L.; Neill, J. C.; Nelson, D. L.;
Nielsen, S. M.; Poulsen, M. N.; Schaus, J. M.; Witten, L. M. Int. J.
Neuopsychopharmacol. 2010, 13, 1021.
15. Wilkinson J., et al. ‘A clinical Phase II study of Lu AE58054 Added to Stable
Donepezil Treatment in Patients with Moderate Alzheimer’s Disease’. Abstract
(O3-06-01) Presented at the Alzheimer’s Association International Conference,
2013, July 13–18, Boston, MA.
16. Halford, J. C. G.; Harrold, J. A.; Boyland, E. J.; Lawton, C. L.; Blundell, J. E. Drugs
2007, 67, 27.
17. Sargent, B. J.; Henderson, A. J. Curr. Opin. Pharmacol. 2011, 11, 52.
18. Heal, D. J.; Smith, S. L.; Fisas, A.; Codony, X.; Buschmann, H. Pharmacol. Ther.
2008, 117, 207.
3-Benzenesulfonyl-2-methylsulfanyl-pyrido[1,2-a]pyrimidin-
4-ylideneamine (100 mg, 0.30 mmol) was dissolved in DMF (2 ml).
Methylamine (0.3 mmol, 2 M in THF) was added at room tempera-
ture, followed immediately by addition of mercury(II) acetate
(96 mg, 0.30 mmol) in one portion. After 10 min, the solvent was
evaporated under a stream of nitrogen and the crude product
was purified by preparative HPLC (XTerra™ MS C18 5.0 mM,
MeOH/H2O as the eluent). 26 mg (27%) of the product was ob-
l
tained as a pale yellow solid. H NMR (500 MHz, CDCl3): d 9.2 (d,
1H), 8.3 (s, br, 1H), 8.0 (s, br, 1H), 7.9 (d, 2H), 7.6–7.4 (m, 4H),
7.1 (d, 1H), 6.8 (t, 1H), 3.1 (d, 3H).

1790
S. Hu et al. / Bioorg. Med. Chem. 22 (2014) 1782–1790
19. Caldirola, P. S. Presented at SMI Inaugural Conference on Obesity and Related
Disorders, London, 2003.
20. Shacham, S.; Marantz, Y.; Senderowitz, H., et al Obes. Res. 2005, 13, A192.
21. Woolley, M. L.; Bentley, J. C.; Sleight, A. J.; Marsden, C. A.; Fone, K. C. F.
Neuropharmacology 2001, 41, 210.
22. Hayat, F.; Cho, S.; Rhim, H.; Viswanath, A. N. I.; Pae, A. N.; Lee, J. Y.; Choo, D. J.;
Choo, H.-Y. P. Bioorg. Med. Chem. 2013, 21, 5573.
23. Ivachtchenko, A.; Golovina, E.; Kadieva, M.; Mitkin, O.; Tkachenko, S.; Okun, I.
Bioorg. Med. Chem. 2013, 21, 4614.
29. Liu, K. G.; Robichaud, A. J. Int. Rev. Neurobiol. 2010, 94, 1.
30. Wu, Y.-J.; He, H.; Hu, S.; Huang, Y.; Scola, P. M.; Grant-Young, K.; Bertekap,
Robert L.; Wu, Dedong; Gao, Qi; Li, Yi; Klakouski, Cheryl; Westphal, R. S. J. Med.
Chem. 2003, 46, 4834.
31. Pruitt, K. D.; Tatusova, T.; Brown, G. R.; Maglott, D. R. Nucleic Acids Res. 2011, 40,
D130.
32. Warne, T.; Serrano-Vega, M. J.; Baker, J. G.; Moukhametzianov, R.; Edwards, P.
C.; Henderson, R.; Leslie, A. G. W.; Tate, C. G.; Schertler, G. F. X. Nature 2008,
454, 486.
24. Nirogi, R.; Shinde, A.; Daulatabad, A.; Kambhampati, R.; Gudla, P.; Shaik, M.;
Gampa, M.; Balasubramaniam, S.; Gangadasari, P.; Reballi, V.; Badange, R.;
Bojja, K.; Subramanian, R.; Bhyrapuneni, G.; Muddana, N.; Jayarajan, P. J. Med.
Chem. 2012, 55, 9255.
25. Ivachtchenko, A. V.; Golovina, E. S.; Kadieva, M. G.; Kysil, V. M.; Mitkin, O. D.;
Tkachenko, S. E.; Okun, I. Bioorg. Med. Chem. 2011, 19, 1482.
26. Liu, K. G.; Robichaud, A. J.; Greenfield, A. A.; Lo, J. R.; Grosanu, C.; Mattes, J. F.;
Cai, Y.; Zhang, G.; Zhang, J. Y.; Kowal, D. M.; Smith, D. L.; Di, L.; Kerns, E. H.;
Schechter, L. E.; Comery, T. A. Bioorg. Med. Chem. 2011, 19, 650.
27. van Loevezijn, A.; Venhorst, J.; Bakker, W. I. I.; de Korte, C. G.; de Looff, W.;
Verhoog, S.; van Wees, J.-W.; van Hoeve, M.; van de Woestijne, R. P.; van der
Neut, M. A. W.; Borst, A. J. M.; van Dongen, M. J. P.; de Bruin, N. M. W. J.; Keizer,
H. G.; Kruse, C. G. J. Med. Chem. 2011, 54, 7030.
33. de la Fuente, T.; Martín-Fontecha, M.; Sallander, J.; Benhamú, B.; Campillo, M.;
Medina, R. A.; Pellissier, L. P.; Claeysen, S.; Dumuis, A.; Pardo, L.; López-
Rodríguez, M. L. J. Med. Chem. 2010, 53, 1357.
34. Harris Iii, R. N.; Stabler, R. S.; Repke, D. B.; Kress, J. M.; Walker, K. A.; Martin, R.
S.; Brothers, J. M.; Ilnicka, M.; Lee, S. W.; Mirzadegan, T. Bioorg. Med. Chem. Lett.
2010, 20, 3436.
35. Nagasaka-Hoshino, M.; Isozaki, T.; Suzuki, T.; Ichimura, T.; Kawauchi, S. Chem.
Phys. Lett. 2008, 457, 58. and references cited therein.
36. PyMOL: the PyMOL Molecular Graphics System, version 1.3, Schrödinger LLC.
37. Wu, Y.-J.; Hu, S.; Scola, P. M.; Huang, Y.; Grant-Young, K. A. U.S. 20,040,019,064,
2004.
38. Suite 2012: Prime, version 3.1, Schrödinger, LLC: New York, NY, 2012.
39. Suite 2012: Glide, version 5.8, Schrödinger, LLC: New York, NY, 2012.
28. Holenz, J.; Pauwels, P. J.; Diaz, J. L.; Merce, R.; Codony, X.; Buschmann, H. Drug
Discovery Today 2006, 11, 283.