Positive allosteric modulators of 2-amino-3-(3-hydroxy-5-methylisoxazol-4-yl)propionic acid receptors belonging to 4-cyclopropyl-3,4-dihydro-2 H -1,2,4-pyridothiadiazine dioxides and diversely chloro-substituted 4-cyclopropyl-3,4-dihydro-2 H -1,2,4-benzot

Article abstract of DOI:10.1021/jm501268r

Two 4-ethyl-substituted pyridothiadiazine dioxides belonging to α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor positive allosteric modulators were cocrystallized with the GluA2 ligand binding domain in order to decipher the impact of

Full text of DOI:10.1021/jm501268r

Article
pubs.acs.org/jmc
Positive Allosteric Modulators of 2‑Amino-3-(3-hydroxy-5-
methylisoxazol-4-yl)propionic Acid Receptors Belonging to
4‑Cyclopropyl-3,4-dihydro‑2H‑1,2,4-pyridothiadiazine Dioxides and
Diversely Chloro-Substituted 4‑Cyclopropyl-3,4-dihydro‑2H‑1,2,4-
benzothiadiazine 1,1-Dioxides
Pierre Francotte,†,∥ Ann-Beth Nørholm,‡,∥ Taru Deva,‡ Lars Olsen,‡ Karla Frydenvang,‡ Eric Goffin,†
Pierre Fraikin,† Pascal de Tullio,† Sylvie Challal,§ Jean-Yves Thomas,§ Fabrice Iop,§ Caroline Louis,§
Iuliana Botez-Pop,§ Pierre Lestage,§ Laurence Danober,§ Jette S. Kastrup,‡,∥ and Bernard Pirotte*,†,∥
†Department of Medicinal Chemistry, Center for Interdisciplinary Research on Medicines (CIRM), University of Liege, Avenue de
̂ ̀
l’Hopital, 1, B36, B-4000 Liege, Belgium
‡Department of Drug Design and Pharmacology, Faculty of Health and Medical Sciences, University of Copenhagen,
Universitetsparken, 2, DK-2100 Copenhagen, Denmark
§Institut de Recherches Servier, 125 Chemin de Ronde, F-78290 Croissy-sur-Seine, France
S
* Supporting Information
ABSTRACT: Two 4-ethyl-substituted pyridothiadiazine dioxides
belonging to α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid
(AMPA) receptor positive allosteric modulators were cocrystallized
with the GluA2 ligand binding domain in order to decipher the
impact of the position of the nitrogen atom on their binding mode at
the AMPA receptors. The latter was found to be very similar to that
of previously described benzothiadiazine-type AMPA receptor
modulators. The affinity of the two compounds for the receptor
was determined by isothermal titration calorimetry. Accordingly, the
synthesis and biological evaluation of novel 4-cyclopropyl-substituted pyridothiadiazine dioxides was performed and completed
with the synthesis of the corresponding chloro-substituted 4-cyclopropyl-3,4-dihydro-2H-benzothiadiazine 1,1-dioxides. The “8-
aza” compound 32 was found to be the most potent pyridothiadiazine-type AMPA receptor potentiator in vitro, whereas the 7-
chloro-substituted compound 36c emerged as the most promising benzothiadiazine dioxide. Due to proper drug-likeness and low
in vivo acute toxicity in mice, 36c was chosen for a more complete preclinical evaluation. The compound was able to easily cross
the blood−brain barrier. In an in vivo object recognition test with CD1 mice, oral administration of 36c was found to
significantly improve cognition performance at doses as low as 1 mg/kg.
the ligand binding domain (LBD),7 three membrane-spanning
hydrophobic domains (M1, M3, and M4), and one intra-
INTRODUCTION
■
L-Glutamate is the major excitatory neurotransmitter in the
membrane re-entrant loop (M2), which constitutes part of the
brain, acting through two distinct types of receptors, the
ion channel pore,8 as well as an intracellular domain comprising
ionotropic (ligand-gated ion channel, iGluRs) and the
motifs for binding and phosphorylation of transmembrane
metabotropic (G-protein coupled, mGluRs) receptors. On the
AMPA receptor regulatory proteins (TARPs).9
basis of their sensitivity to selective agonists, iGluRs have been
classified into three subclasses, the N-methyl-D-aspartic acid
iGluRs not only contain the binding site for the natural
ligand L-glutamate but also include several allosteric sites for
(NMDA), the α-amino-3-hydroxy-5-methyl-4-isoxazolepro-
pionic acid (AMPA), and the kainic acid (KA) receptors.1
different kinds of modulators, among which are AMPAR-
positive allosteric modulators (AMPApams).10 By slowing the
AMPA receptors (AMPARs) mediate fast excitatory
rate of receptor deactivation (the process of channel
postsynaptic currents at a great majority of the synapses and
inactivation following the dissociation of L-glutamate) and/or
desensitization (the process of ion channel inactivation with L-
glutamate remaining bound to the receptor), AMPApams
are involved in the expression of long-term potentiation, a
phenomenon closely linked to learning and memory processes.2
These are tetrameric assemblies of subunits GluA1−4, most
being symmetric dimer-of-dimers of GluA2 and either GluA1,
GluA3, or GluA4.3−6 Each subunit comprises a large
extracellular domain harboring the L-glutamate binding site in
Received: August 18, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm501268r | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 1. Chemical structure of previously described positive allosteric modulators of AMPA receptors belonging to benzo-, pyrido-, and
thienothiadiazine dioxides [diazoxide (1), cyclothiazide (2), IDRA 21 (3), S18986 (4), BPAM 97 (7), BPAM 121 (8), and BPAM 344 (9)]. The
positions of atoms in the benzothiadiazine ring system have been labeled on the structure of diazoxide.
potentiate the effect of the neurotransmitter on its postsynaptic
receptors. Compared to direct AMPAR agonists, allosteric
modulators are anticipated to exert finer tuning to potentiate
glutamatergic function, since they have no effects in the absence
of the natural ligand in the synapse.11 Direct AMPAR agonists
that interact with the L-glutamate receptor binding site may
cause severe neurotoxicity in case of overstimulation,11 while
AMPApams were found to have relatively few adverse effects at
therapeutically relevant doses.12
A dysfunction of glutamatergic neurotransmission has been
implicated in a number of neurological and psychiatric
diseases.13 Enhancement of AMPAR signals by AMPAR
potentiators is therefore expected to be beneficial in the
management of neurological disorders such as depression,
schizophrenia, Parkinson’s disease, Alzheimer’s disease, atten-
tion deficit hyperactivity disorder, and mood disorders.14−18
Such therapeutic promise is linked, at least in part, to the
noteworthy property of AMPAR potentiators to increase the
expression of the brain derived neurotrophic factor
(BDNF).13,19−22
Different chemical series have been investigated for a
potential activity as AMPApams, among which are benzothia-
diazine dioxides (BTDs).11 The first representatives of BTD-
type AMPApams were the antihypertensive agent (KATP
channel opener) diazoxide (1) and the diuretic agent
cyclothiazide (CTZ) (2) (Figure 1).11,23,24 A compound
resulting from saturation of the nitrogen−carbon double
bond of diazoxide, namely IDRA 21 (3), was further
demonstrated as a more potent AMPAR potentiator than the
parent compound.25 Despite being a weak AMPAR potentiator
in vitro compared to CTZ, the small-sized IDRA 21 attracted
great clinical interest, as it was found to be active in vivo,
enhancing cognition performance in animal models of cognitive
impairments.26,27
(PTDs) and benzothiadiazine dioxides (BTDs) and, more
specifically, on compounds structurally related to IDRA 21 and
S18986 (4) (Figure 1).28−32 The first series of pyridothiadia-
zine isosteres of BTDs led to the establishment of coherent
structure−activity relationships and to the identification of the
“8-aza” compound 5 as one of the most potent PTD-type
AMPAR potentiators.28 The corresponding “7-aza” analogue 6
was less active, indicating the critical impact of the position of
the nitrogen atom on biological activity (rank order of potency:
8-aza > 7-aza > 5-aza).28 Coming back to BTDs, removal of the
methyl group at the 3-position of IDRA 21 and introduction of
a short alkyl chain at the 4-position was responsible for a
marked improvement of in vitro and in vivo activity, as
observed with compound 7 (Figure 1).29 Refinement of the
pharmacokinetic properties (metabolic stability) was subse-
quently achieved by introducing a monofluoroalkyl chain at the
4-position of the heterocycle, providing compounds like 8
(Figure 1).31 A further improvement of the pharmacokinetic
properties as well as of the biological activity was reached by
introducing a cyclopropyl side chain at the 4-position, leading
to BTD compounds such as 9 (Figure 1).32 Another recent
evolution occurred with the discovery of thienothiadiazine
dioxides (i.e., compounds 10 and 11; Figure 1) as isosteres of
benzo- and pyridothiadiazine dioxides.33
Two of these newly synthesized BTDs were previously
cocrystallized with the ligand binding domain of the AMPAR
subunit GluA2 (GluA2-LBD).32,34 It was observed that
compound 7, as well as its 4-cyclopropyl-substituted analogue
9, at the level of the dimer interface, interact with the same
binding site and through a similar binding mode as IDRA 21,
the allosteric site being very close to that of cyclothiazide
(CTZ). However, the binding mode of CTZ was found to be
quite different.32 Representatives of other families of
AMPApams (i.e., benzamides and N-biaryl(cyclo)alkyl-2-
propanesulfonamides) also bind within the same dimer
In recent years, we have performed extensive research on
original AMPApams belonging to pyridothiadiazine dioxides
B
dx.doi.org/10.1021/jm501268r | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 1. X-ray Data Collection and Refinement Statistics of GluA2-LBD-L483Y-N754S with (S)-Glutamate in Complex with 5
and 6, Respectively
Compound 5
Compound 6
Data Collection
space group
P21212
P21212
a, b, c (Å)
98.29, 121,57, 47.36
97.83, 121.37, 47.33
34.87−1.90 (2.00−1.90)
45 062
a
resolution (Å)
no. of unique reflections
redundancy
29.38−1.56 (1.64−1.56)
81 474
3.8 (3.8)
99.8 (100)
4.4 (34.1)
10.1 (1.8)
4.7 (4.7)
completeness (%)
99.7 (100)
b
Rmerge (%)
8.9 (30.8)
I/σI
5.7 (2.3)
Refinement
c
Rwork/Rfree (%)
15.0/17.5
15.9/20.8
no. of non-hydrogen atoms
4852
4644
no. of residues (molecules A/B)
263/259
258/263
no. of (S)-glutamate/modulator/water/glycerol/acetate/chloride/sulfate/PEG
2/2/590/3/4/−/5/2
2/2/402/4/8/2/3/1
RMS Deviations
bond lengths (Å)
bond angles (deg)
0.01
1.3
0.007
1.0
d
no. of residues in allowed areas of Ramachandran plot (%)
100
100
Average B Values (Å2)
molecules A/B
22.0/18.1
23.7/29.7
(S)-glutamate/modulator/water/glycerol/acetate/chloride/sulfate/PEG
14.3/15.2/33.2/76.2/59.0/−/35.2/55.4
12.8/17.6/29.3/47.2/40.6/62.9/48.7/50.9
a
b
Values for the outermost resolution shell are denoted in parentheses. Rmerge = ΣhklΣi|Ii(hkl) − I(hkl)|/ΣhklΣi|Ii(hkl)|, with Miller indices hkl, the
c
intensity of an individual measurement of the reflection (Ii(hkl)), and the intensity from multiple observations (I(hkl)). Rwork = Σhkl|Fo − Fc|/Σhkl|Fo|,
with observed (|Fo|) and calculated (|Fc|) structure factor amplitudes. Rfree is calculated with 5% reflections omitted from the refinement process.
PROCHECK60 was used to calculate the Ramachandran plot.
d
interface, although on a distinct allosteric binding site close to
the two neighboring CTZ and BTD binding sites.35,36
The present work aims at exploring the interaction of
representative PTD-type AMPApams (i.e., compounds 5 and
6) with the GluA2-LBD in order to decipher the impact of the
position of the nitrogen atom on their binding mode and
affinity for the AMPARs. This study was completed by the
synthesis and biological evaluation of new pyridothiadiazines
belonging to the different heterocyclic subclasses and bearing
the cyclopropyl group at the 4-position, as well as their
corresponding 4-cyclopropyl-substituted BTDs with a chlorine
atom at the different possible positions on the benzene ring.
The positions of the nitrogen atom in the pyridine ring of
PTDs and of the chlorine atom on the benzene ring of BTDs
are expected to similarly influence the electronic distribution of
the molecule (electronic imprint) and therefore their
interaction with the BTD allosteric binding site of AMPARs.
Finally, the most promising compound was selected for a
more complete preclinical evaluation (in vitro and in vivo
testing) in the perspective of development as a new drug
candidate.
respectively. Data collection and refinement statistics are shown
in Table 1. We have previously used the dimeric GluA2-LBD-
L483Y-N754S for crystallization with allosteric modulators, and
the mutations were shown not to interfere with the modulator
binding pocket.32,34 For both 5 and 6, a single dimer is seen in
the crystal structure composed of molecules A and B. The
complexes crystallize with (S)-glutamate bound at the agonist
binding site and two molecules of the respective modulator at
the dimer interface (Figure 2A). Upon (S)-glutamate binding, a
rotation of the D2 lobe of ∼20° compared to the apo GluA2-
LBD structure results in a closed clamshell-like conformation.7
The rotation angle measured using DynDom to molecule A was
20.4° and molecule B was 21.4° for 5 and to molecule A was
21.4° and molecule B was 20.5° for 6, compared to the apo
structure (PDB code 1FTO, molecule A). These rotation
angles are comparable to those of the glutamate bound
structure without modulators bound.36 Therefore, 5 and 6
stabilize the dimer interface without affecting the GluA2
clamshell structure.
Compounds 5 and 6 bind in the same region in GluA2-LBD-
L483Y-N754S as the previously crystallized compounds 7 and
9.32,34 The N4 ethyl moiety is directed toward the protein
backbone residues Phe495, Met496, and Ser497 (Figure 2B,C).
This orientation was also previously reported for 7.34 For both
5 and 6, a hydrogen bond is seen from the N2 hydrogen atom
of the modulator to the carbonyl oxygen atom of Pro494, which
was also observed for 7 and 9. Thus, the interactions between
the four modulators crystallized to date and the GluA2-LBD are
similar.
RESULTS AND DISCUSSION
■
Structure Determination. The aim of the structural
investigations on PTDs was to determine if differences in the
position of the nitrogen atom in the pyridine ring of PTDs
could be responsible for differences in binding mode at the
level of the allosteric binding site. Two previously described 4-
ethyl-substituted pyridothiadiazine dioxides,28 i.e., compounds
5 and 6, were chosen as representatives of the 8-aza series of
very active compounds and of the less active 7-aza compounds,
respectively.
Compound 6 forms a hydrogen bond from one oxygen atom
of the sulfonamide to a water molecule, which is seen for both
copies of the modulators with distances of 3.2 Å (W1) and 2.3
Å (W2), respectively (Figure 2C). W1 further forms hydrogen
bonds to two water molecules and to the carbonyl oxygen atom
We determined two high-resolution structures of GluA2-
LBD-L483Y-N754S in complex with (S)-glutamate and 5 or 6,
C
dx.doi.org/10.1021/jm501268r | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
five to eight water molecules relative to the GluA2-LBD
structure with glutamate bound (PDB code 2UXA).37 These
water molecules are located near the N2, N4, N7/N8, the
sulfonamide group and the ethyl group. In conclusion, the
dimer interface of GluA2-LBD is solvated, but this solvation
only to a minor extent involves the modulator molecules
directly.
The side chain of Ser497 was found in two conformations in
the structure with 6 (refined to occupancies of 0.36/0.64 and
0.33/0.67 for molecules A and B, respectively) (Figure 2C),
whereas Ser497 was only seen in one conformation in the
structure with 5 with the side chain hydroxyl group directed
toward the modulator (Figure 2B). The conformation of
Ser497 pointing toward the modulator is the conformation
observed in the glutamate bound structure without modulator
(PDB code 2UXA). In the structure with 5, the side chains of
Ser497 in the two molecules of the dimer are connected via two
water molecules. In the structure with 6, the side chains of
Ser497, when directed toward the modulator, are connected via
a single water molecule, whereas the side chain conformations
directed away from the modulator form hydrogen bonds to two
putative acetate ions. The conformation of Ser497 directed
away from the modulator has previously been observed in the
structure with 7 and 9, where the side chains of Ser497 were
also found to stabilize the GluA2-LBD dimer interface through
an interaction via two water molecules. Both 7 and 9 have a
fluorine atom at the C7 position, which forms a weak fluorine
hydrogen bond to the side chain hydroxyl group of Ser497,
when the side chain is directed toward the modulator. The
distance between the hydroxyl group of Ser497 and the N7
atom of 6 (4.4 Å) and the N8 atom of 5 (5.0 Å) is too long for
a hydrogen bond to be formed. The lack of a hydrogen bond to
the Ser497 hydroxyl group might contribute to a lower binding
affinity of PTD compounds relative to the BTD compounds.
Both the single (in structure with 6) and the two (in structure
with 5) water molecules connecting the Ser497 residues have
different solvation patterns relative to the structure without any
modulator bound (PDB code 2UXA). In the latter structure,
two water molecules are found further away from the Ser497
residues (both 3.7 Å), which results in weak hydrogen bonds.
Thus, the solvation pattern involving the residue Ser497 differs
not only between the structures with 5 and 6, but also in the
structure without modulators, even though the solvation does
not directly involve the modulators.
A comparison of the binding mode of 5 and 6 with that of 7
and 9 (Figure 2D) reveals a slight tilt of the bicyclic ring system
of 6, but not of the N4 ethyl moiety, for both copies of the
modulator, whereas 5 shows almost complete overlap with both
7 and 9 of the bicyclic ring system and the N4 ethyl moiety. A
comparison with the intermodulator distances at the dimer
interface shows 6 to have a different distance pattern (N7−N7,
4.5 Å; C8−C8, 5.7 Å) compared to 5 (C7−C7, 3.8 Å; N8−N8,
5.0 Å), 7 (C7−C7,4.0 Å; C8−C8, 5.2 Å), and 9 (C7−C7, 4.0
Å; C8−C8, 5.3 Å), and 6 is moved ∼0.5 Å relative to 5, 7, and
9. Hence, 6 has a slightly different binding mode compared to
what has previously been observed for these modulators. We
previously compared the binding mode and potency of IDRA
21, belonging to the shifted thiazide class, with those of
compounds 7 and 9.32,34 These compounds showed similar
binding modes, although the bicyclic ring system of IDRA-21 is
slightly shifted compared to that of 7 and 9, and the increased
potency of compounds 7 and 9 relative to that of IDRA 21 was
attributed to increased van der Waals interactions. CTZ
Figure 2. X-ray structures of the GluA2-LBD-L483Y-N754S dimer in
complex with 5 and 6, respectively. (A) Cartoon representation of
GluA2-LBD with two molecules of compound 5 at the dimer interface
and (S)-glutamate bound at the agonist binding site (stick
representation). Water molecules are shown as red spheres. (B)
Closeup of the modulator binding site of 5. The side chain of Ser497
was found in one conformation directed toward the modulator. A
simple 2Fo − Fc OMIT electron density map contoured at 1σ and
carved 2.0 Å around the modulators is shown in light gray. Selected
hydrogen bonds are shown as black dotted lines. (C) Closeup of the
modulator binding site of 6. The side chain of Ser497 was found in
two conformations. (D) Side view of the modulators 5, 6, 7, and 9.
The structures were superimposed on protein molecules A.
Compound 6 (cyan stick) is slightly tilted compared to 5 (green
stick), 7 (yellow stick, PDB code 3TDJ), and 9 (magenta stick, PDB
code 4N07).
of Ile481 (A), and W2 forms hydrogen bonds to the carbonyl
oxygen atom of Ile481 (B) and has close van der Waals
contacts to the side chain of Lys730 (B). Compound 5 does
not form hydrogen bonds to any water molecules in the crystal
structure. Additionally, we have been able to locate 9
(compound 6) and 14 (compound 5) water molecules within
4 Å from the modulators, which, however, do not form direct
hydrogen bonds to the BPAMs but form a dimer interface
water network involving residues Ile481, Ser497, Ser729,
Lys730, Gly731, Ser754, and Asp760. The modulators displace
D
dx.doi.org/10.1021/jm501268r | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 3. Isothermal titration calorimetry study of binding of 5 (A) and 6 (B) to GluA2-LBD-L483Y-N754S. Raw data (top panel) and isotherm
(bottom panel) are shown. Heat is developed after each injection (exothermic reaction), and the signal is reduced when the protein is saturated with
5 and 6, respectively. The experiments shown are representatives of three experiments.
Table 2. Isothermal Titration Calorimetry Using GluA2-LBD-L483Y-N754S
a
b
5
6
7
9
c
a
Kd (μM)
9.8 0.7
−3.4 0.1
−3.4 0.1
2.1 0.2
1200 40
5.6 0.9
0.35 0.02
−7.5 0.2
−1.3 0.2
2.7 0.1
ΔH (kcal/mol)
−TΔS (kcal/mol)
nH
−3.2 0.1
−0.8 0.1
−4.9 0.4
−2.3 0.4
2.3 0.1
d
2.0
a
b
c
d
Values from ref 34. Values from ref 32. Standard deviation. nH of compound 6 was fixed at 2.00 during fitting.
belongs to the classical thiazide class, and it showed a different
binding mode from that of compounds 7 and 9.32 Through
these comparisons with compounds 7 and 9, it follows that
compounds 5 and 6 show a similar binding mode compared to
that of IDRA 21 and a different binding mode from that of
CTZ.
Isothermal Titration Calorimetry. The structures of
GluA2-LBD-L483Y-N754S in complex with either 5 or 6
showed different numbers of coordinated water molecules and
differences in intermodulator distances. To understand the
thermodynamic differences between binding of 5 and 6, we
measured the thermodynamics of binding using isothermal
titration calorimetry (ITC). GluA2-LBD-L483Y-N754S was
also the choice for these studies, as it exists as a preformed
dimer in solution, thus mimicking the endogenous binding
environment of the dimer of dimers structure found in the full-
length GluA2 receptor. Therefore, modulator binding can be
measured independently of dimer formation.
ΔH = −3.2 kcal/mol and −TΔS = −0.8 kcal/mol (Table 2).
Whereas the enthalpies of 5 and 6 are comparable, 6 has higher
entropy of binding than 5, which explains the better binding
affinity of 5 compared to 6. A bound water molecule is
associated with higher order and thus higher entropy of
binding. Compounds 5 and 6 bind in an overall similar binding
mode, but differences in the water network surrounding the
modulators were observed. Therefore, it is likely that water
molecules coordinating directly to the sulfonamide of 6 may
partly explain the entropy penalty upon binding of 6 to GluA2-
LBD. A comparison of the thermodynamic details of binding of
5 and 6 to those of 7 shows a slight loss of enthalpy for 5 and 6
(ΔH = −3.4/−3.2 versus −4.9 kcal/mol), presumably due to
the lack of the weak hydrogen−fluorine bond seen between 7
and GluA2-LBD-L483Y-N754S.
Chemistry. The binding mode of compounds 5 and 6,
representatives of two distinct series of PTDs, was found to be
very similar to that of BTDs 7 and 9 previously described.32,34
In these previous studies, we found that the ethyl group forms
favorable van der Waals interactions to primarily Met49634 and
that replacement of the ethyl group with a cyclopropyl group
increased van der Waals interactions between the modulator
and the backbone of residues Phe495 and Met496, as well as
stacking interactions with the peptide backbone due to the π-
The ITC measurements showed that the binding of both 5
and 6 to GluA2-LBD-L483Y-N754S is an exothermic process,
meaning that heat is generated during binding of the
modulators to GluA2-LBD (Figure 3). For 5, the binding
affinity (Kd) was measured to be 9.8 μM with ΔH = −3.4 kcal/
mol and −TΔS = −3.4 kcal/mol, and for 6, Kd = 1200 μM with
E
dx.doi.org/10.1021/jm501268r | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
character of the cyclopropyl group.32 It was therefore expected
that the replacement of the ethyl chain by a cyclopropyl chain
at the 4-position of PTDs was allowed and could be responsible
for an improvement of the activity on AMPARs. As a result, it
was decided to explore the synthesis and the biological activity
of 4-cyclopropyl-substituted compounds belonging to four
distinct series of PTDs, namely, 5-aza, 5-aza-7-chloro, 7-aza,
and 8-aza compounds. In addition, the corresponding chloro-
substituted BTDs bearing the halogen atom at the different
possible positions on the benzene ring were prepared in order
to compare the impact of the position of the electronegative
halogen atom with the position of the electronegative nitrogen
atom.
Reduction of the nitro group of 17 by means of iron powder
in the presence of ammonium chloride provided the resulting
3-amino-5-chloro-2-fluoropyridine 18, which was engaged in
the Meerwein variation of the Sandmeyer reaction to give the
corresponding o-fluoropyridinesulfonamide 19. The three last
steps were similar to those described for 16. In the case of a
nucleophilic aromatic substitution, the o-fluoro-substituted
pyridinesulfonamide 19 was expected to be more sensitive to
nucleophilic addition than the o-chloro-substituted intermedi-
ate 13.
The target 7-aza compound 4-cyclopropyl-3,4-dihydro-2H-
pyrido[4,3-e]-1,2,4-thiadiazine 1,1-dioxide 26 was easily
obtained in three steps starting from the key intermediate 4-
chloropyridine-3-sulfonamide 23 (Scheme 3).
The synthesis of the 5-aza compound 4-cyclopropyl-3,4-
dihydro-2H-pyrido[2,3-e]-1,2,4-thiadiazine 1,1-dioxide 16 is
described in Scheme 1. Starting from the commercially available
a
Scheme 3
a
Scheme 1
a
Reagents: (i) Cyclopropylamine, dioxane, 100 °C, 24 h (77%). (ii)
HC(OEt)3, 120 °C, 1 h (86%). (iii) NaBH4, 2-propanol, 50 °C, 10
min (83%).
a
Reagents: (i) (1) HNO2, −5 °C; (2) SO2, Cu2Cl2, 15 min; (3) NH3
(60%). (ii) Cyclopropylamine, dioxane, 110 °C, 24 h (85%). (iii)
HC(OEt)3, 150 °C, 3 h (84%). (iv) NaBH4, 2-propanol, 50 °C, 5 min.
(50%).
In the case of the corresponding 8-aza compound 4-
cyclopropyl-3,4-dihydro-2H-pyrido[3,2-e]-1,2,4-thiadiazine 1,1-
dioxide 32, the key intermediate 3-fluoropyridine-2-sulfona-
mide 29 was prepared in two steps from the commercially
available 2-chloro-3-fluoropyridine 27 by conversion into the
corresponding isothiouronium salt by reaction with thiourea
and then its decomposition under alkaline conditions into the
corresponding thiol 28 (Scheme 4). The latter was used to
2-chloro-3-aminopyridine 12, diazotization under the con-
ditions of the Meerwein variation of the Sandmeyer reaction38
provided the corresponding 2-chloropyridine-3-sulfonamide 13.
Nucleophilic substitution of the chlorine atom of 13 by
cyclopropylamine, giving intermediate 14, was greatly facilitated
by the fact that the halogen atom was located at the ortho
position of both the pyridinic nitrogen atom and the electron-
withdrawing sulfonamide group. Ring closure of 14 by means of
triethyl orthoformate provided compound 15, which was
converted with sodium borohydride (saturation of the N2−
C3 bond) into the corresponding target compound 16.
a
Scheme 4
The corresponding 5-aza-7-chloro compound 7-chloro-4-
cyclopropyl-3,4-dihydro-2H-pyrido[2,3-e]-1,2,4-thiadiazine 1,1-
dioxide 22 was obtained following a quite similar synthetic
pathway (Scheme 2). In this case, the key intermediate 5-
chloro-2-fluoropyridine-3-sulfonamide 19 was obtained in two
steps starting from 5-chloro-2-fluoro-3-nitropyridine 17.
a
Scheme 2
a
Reagents: (i) (1) Thiourea, EtOH, reflux, 7 h (isothiouronium, 63%);
(2) Na2CO3, NaOH 10% m/v, H2O, pH 12 (95%). (ii) (1) 12 N HCl,
Cl2, H2O, −15 °C; (2) NH3/H2O 25% m/v. (iii) Cyclopropylamine,
dioxane, 90 °C, 48 h. (iv) HC(OEt)3, 130 °C, 24 h (ii + iii + iv: 8%).
(v) NaBH4, 2-propanol, 50 °C, 5 min (84%).
prepare the pyridinesulfonamide 29 after treatment of 28 with
gaseous chlorine in concentrated hydrochloric acid followed by
the reaction of the formed pyridinesulfonyl chloride with
aqueous ammonia.
Access to the chloro-substituted 3,4-dihydro-2H-1,2,4-
benzothiadiazine 1,1-dioxides 36 is described in Scheme 5.
The appropriate o-fluorobenzenesulfonamides 33 are engaged
in the same chemical pathway as described for the synthesis of
22 and 32.
a
Reagents: (i) Fe, NH4Cl, EtOH/H2O (30%). (ii) (1) HNO2, −5 °C;
(2) SO2, Cu2Cl2, 15 min; (3) NH3 (60%). (iii) Cyclopropylamine,
dioxane, reflux, 24 h. (iv) HC(OEt)3, 120 °C, 1 h. (v) NaBH4, 2-
propanol, 50 °C, 5 min (69%).
F
dx.doi.org/10.1021/jm501268r | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 5
a
Reagents: (i) cyclopropylamine, dioxane, 100−110 °C, 24 h (80−90%). (ii) HC(OEt)3, 130 °C, 2−24 h (70−80%). (iii) NaBH4, 2-propanol, 50−
55 °C, 5−10 min (80−85%).
A critical point with PTDs is the possibility of protonation of
7-aza compounds in weak acidic conditions [see Figures 4 and
weak acid character linked to the presence of the sulfonamide
function with a pKa value of 9.15 determined by UV
spectrophotometry. In summary, it can be stated that the 7-
aza compounds like 6 may exist at the physiological pH of 7.4
as their nonionic species, which is the case for the other PTDs
and BTDs. Consequently, the nonionic species of 7-aza PTDs
is expected to interact with the allosteric binding site of the
GluA2-LBD.
Figure 4. Deprotonation process of compound 6 in water solution
revealing the existence of two ionization constants (pKa1 and pKa2).
Biological Results. The original 4-cyclopropyl-substituted
3,4-dihydro-2H-1,2,4-pyrido/benzothiadiazine 1,1-dioxides 16,
22, 26, 32, and 36a−d were evaluated as AMPAR potentiators
in an in vitro fluorescence assay. The experiments were
performed on primary cultures of neurons from rat embryonic
cortex, measuring the effect of the modulators on AMPA-
evoked membrane depolarization, as previously described.31,32
For each compound, the EC2× value was determined
(concentration of modulator giving a 2-fold increase of the
depolarization induced by 300 μM AMPA), as well as their
maximum effect (Emax value normalized to unity for AMPA-
evoked response taken as 1×). For compounds 32 and 36c, the
EC50 value (concentration of modulator responsible for 50% of
the maximal effect) was also calculated (Table 3).
S1 (Supporting Information)]. In fact, PTDs like the 7-aza
compound 6 belong to 4-aminopyridine derivatives, which are
known to express a more pronounced basic character than the
corresponding 2-aminopyridine (i.e., the 5-aza compounds)
and 3-aminopyridine derivatives (i.e., the 8-aza compounds).
The pKa value of the protonated form of compound 6
determined by UV spectrophotometry was found to be 5.68
(pKa1). The second pKa value corresponding to the
deprotonation of the sulfonamide group at the 2-position of
PTDs was determined to be 9.03 (pKa2). The corresponding 8-
aza compound 5 belonging to 3-aminopyridine derivatives
expressed a less basic character (pKa1 < 3) and only showed a
Table 3. Effects of 4-Cyclopropylpyrido/benzothiadiazine Dioxides on the Fluorescence Induced by 300 μM AMPA in Primary
Neuronal Cultures from Rat Embryonic Cortex and on AMPA-Mediated Presynaptic Noradrenaline (NA) Release on Rat
Hippocampal Slices
a
b
c
d
e
compd
(N)
R5
R6
R7
R8
EC2× (μM)
Emax
>×4 (3)
EC50 (μM)
NA rel (%)
av log P
f
16
5-N
7-N
8-N
5-N
−
−
−
−
−
−
H
H
−
Cl
H
H
H
H
H
H
H
H
H
Cl
H
H
H
H
−
H
Cl
H
H
Cl
H
F
H
H
−
H
H
H
H
Cl
H
13.9 [2.7; 73.0] (3)
4.4 [1.9; 9.8] (3)
0.35 [0.06; 2.2] (2)
2.4 [0.19; 32.2] (2)
27.1 [7.3; 101.0] (3)
5.2 [1.9; 14.5] (3)
0.68 [0.25; 1.8] (4)
2.0 [0.45; 8.5] (3)
0.27 0.03 (3)
ND
221
330
323
276
196
327
411
352
ND
0.60
0.13
0.33
1.28
1.86
1.85
1.86
1.88
1.41
26
>×11 (3)
×10 (2)
>×7 (2)
>×3 (2)
>×5 (3)
×11 (4)
>×9 (3)
ND
32
1.5 [1.5, 1.5] (2)
22
ND
36a
36b
ND
ND
g
36c
2.9 [1.3; 6.3] (4)
ND
36d
h
9
×15.0 2.6 (3)
0.90 0.10 (3)
a
EC2×: concentration of modulator giving a 2-fold increase of the depolarization induced by 300 μM AMPA using fluorescent membrane potential
dyes and imaging based microplater reader; results are expressed as the geometric mean with its 95% confidence intervals in brackets and n value in
b
parentheses. Emax: maximal effect obtained by compounds in the presence of 300 μM AMPA, normalized to unity for response evoked in the
c
presence of AMPA alone taken as 1×. EC50: concentration of modulator responsible for 50% of the maximal effect; results are expressed as the
d
geometric mean with its 95% confidence intervals in brackets and n value in parentheses. NA rel: noradrenaline release; 100% corresponds to
e
AMPA alone. av log P: log P value calculated using the VCCLAB online software (Virtual Computational Chemistry Laboratory, http://www.
f
g
h
vcclab.org, 2005). ND: not determined. Published compound (see ref 32). Published compound and published results expressed as the arithmetic
mean SEM, with the n value in parentheses (see ref 32).
G
dx.doi.org/10.1021/jm501268r | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Considering that the EC2× values generated in the previous
voltage clamp assay on Xenopus oocytes28 and those obtained in
the fluorescence assay are expected to be very similar, as
discussed in a recent publication,32 it is noteworthy to observe
that the replacement of the ethyl chain by a cyclopropyl chain
at the 4-position of PTDs was constantly responsible for a
strong improvement of the biological activity on AMPARs. For
example, the 4-ethyl-substituted 8-aza compound 5 was
reported to express an EC2× value (voltage clamp) of 7.9
μM,28 while its corresponding 4-cyclopropyl-substituted 8-aza
compound 32 exhibited an EC2× value (fluorescence assay) of
0.35 μM (Table 3), corresponding to more than 20-fold
improvement of the activity on AMPARs. Such activity of the
PTD compound 32 on AMPARs was not far from that of the
previously reported reference BTD compound 9 (EC2× = 0.27
μM; arithmetic mean value).32 The 4-ethyl-substituted 5-aza-7-
chloro compound was previously reported with an EC2× value
(voltage clamp) of 20 μM,30 while its corresponding 4-
cyclopropyl-substituted PTD 22 was found to express an EC2×
value (fluorescence assay) of 2.4 μM (Table 3). Thus, the rank
order of potency is given as follows: 8-aza > 5-aza-7-chloro > 7-
aza > 5-aza, which is in accordance with the previous structure−
activity relationships.28,30
A similar rank order of activity was not observed for the
chloro-substituted 4-cyclopropyl-BTDs. In that case, the order
is 7-chloro > 8-chloro > 6-chloro > 5-chloro. However, the 8-
position for an electronegative group (chlorine atom or
pyridinic nitrogen atom) on ring-fused thiadiazine dioxides
(BTDs and PTDs) appeared to be favorable for activity.
Several AMPApams, such as CTZ (2), S18986 (4), and
compound 7, have been reported to potentiate noradrenaline
(NA) release in rat hippocampal slices, an effect linked to their
interaction with presynaptic AMPA receptors.39,40 This activity
could be responsible for the cognition-enhancing properties of
AMPAR potentiators. Potentiation of noradrenaline release was
measured on rat hippocampal slices with the PTDs and BTDs
listed in Table 3. At 300 μM, most of the compounds induced a
strong enhancement of the (S)-AMPA (10 μM) evoked
[3H]NA release [100% representing the effects shown by (S)-
AMPA alone]. This result clearly indicated that the new
compounds were able to act on presynaptic AMPARs. The
most potent compound was found to be compound 36c, with
an enhancement of NA release corresponding to 411% of the
basal value given by (S)-AMPA alone, while the most potent
PTD as AMPAR potentiator, compound 32, was also quite
active with a 323% increase of NA release. For compound 36c,
a concentration−response curve was established and an EC2×
value (drug concentration giving a 200% increase of NA
release) of 4.3 μM [3.1; 6.0] (n = 3) was determined, indicating
that compound 36c expressed a much more potent effect than
our previous hit compound 7 (EC2× value established at 30 μM
under the same experimental conditions29).
acute toxicity.41 After oral administration of a high dose of the
tested compound, the changes in the behavioral and
physiological states of the mice were recorded. It was found
that compound 36c was devoid of acute toxicity in mice after
acute oral administration up to 30 mg/kg.
The rapid and easy distribution of compound 36c into the
brain was confirmed by measuring the cerebral and plasma
concentration of the compound after oral administration (3
mg/kg) in mice using LC−MS/MS evaluation. The following
modulator concentrations measured 20 min after adminis-
tration were obtained (n = 3): [brain] = 72.7 23.4 ng/g and
[plasma] = 64.7
2.7 ng/mL, providing a [brain]/[plasma]
ratio (mL/g) of 1.12. This ratio supports the view that the
compound after oral administration is equally distributed into
the two compartments and therefore easily crosses the blood−
brain barrier.
It should be noted that compound 36c, as well as the other
compounds listed in Table 3, responds perfectly to the “rule of
three” (MW < 300 Da; number of hydrogen bond donors ≤3;
number of hydrogen bond acceptors ≤3; clogP ≤ 3; number of
rotatable bonds ≤3), a much more stringent set of structural
criteria compared to the well-known Lipinski’s “rule of five” for
expecting appreciable drug-likeness and oral bioavailability for a
new compound.42 Thus, compound 36c, with a molecular
weight of 258.73 Da, a calculated log P value of 1.86 (see Table
3), and a perfect compliance to the other criteria possesses the
required properties to be orally bioavailable and appropriately
distributed even into the central nervous system (CNS).
Compound 36c was also examined in an ex vivo electro-
physiological test performed on rat hippocampal slices, as
previously described.29,31 The effects of the compound at 3 and
10 μM were measured on the excitatory postsynaptic field
potentials (EPSfPs) evoked in the CA1 region after electrical
stimulation of the Schaffer collateral. Because of the presence of
1.2 mM Mg2+ in the perfusion medium, EPSfPs recorded in this
assay are primarily mediated by AMPAR activation. As shown
in Table 4, compound 36c markedly increased the signals at the
Table 4. Effects of Compound 36c (3 and 10 μM) on the
AMPA-Mediated Postsynaptic Response Recorded in CA1
Area of Rat Hippocampal Slices
a
mean SEM (n)
concn (μM)
area (%)
amplitude (%)
half-width (%)
3
10.0 3.7 (3)
145 35 (4)
5.3 3.2 (3)
58 13 (4)
9.0 1.5 (3)
113 24 (4)
10
a
Results are expressed as a percentage of increase versus baseline
(mean and standard deviation of the mean). Area, amplitude, and half-
width of the AMPA-mediated EPSfP measured in the presence of the
compound were normalized as a percentage of those recorded over the
10 min baseline period taken as 100%.
Considering the calculated log P values obtained with the
new compounds (see Table 3), it is expected that the more
pronounced hydrophilic character of compound 32 (average
log P = 0.33) compared to compound 36c (average log P =
1.86) should be less favorable for brain penetration and drug
distribution into the CNS. Therefore, and taking into account
the first biological results, compound 36c was selected for
further biological evaluation. The safety of the compound was
assessed first using the Irwin’s test, a commonly used
observation test for detecting untoward effects of new
compounds on general behavior and for evaluating their
concentration of 10 μM (a concentration-dependent effect was
observed), suggesting that the compound was able to interact
with postsynaptic AMPARs located on hippocampal CA1
neurons, as was also demonstrated previously for the BTD
compound 8.31 However, 36c demonstrated a much more
pronounced impact on EPSfPs than 8, since the latter
compound tested at 50 μM induced an area, amplitude, and
half-width increase of 66%, 23%, and 47%, respectively.31 Some
epileptiform activity was also observed after application of 36c
at 10 μM, which could reflect additional effects on other targets
as previously reported for 831 and CTZ.43
H
dx.doi.org/10.1021/jm501268r | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
The next test performed in vivo with 36c was electro-
physiological recordings of the postsynaptic response in the
hippocampus of anesthetized Wistar rats. This was performed
to confirm the in vitro effects observed on EPSfPs on
hippocampal slices and also to measure its effects on long-
term potentiation (LTP). The LTP of synaptic response is
considered to be one of the synaptic plastic mechanisms that
underlies learning and memory processes.44,45 Within 1 h after
intraperitoneal injection of 36c (3 and 10 mg/kg), an increase
of the amplitude of the postsynaptic response was observed in
the dentate gyrus (p = 0.074, post hoc analysis of treatment
effect for all time levels with Dunnett test after ANOVA),
corroborating the initial observations made in vitro on
hippocampal slices (Figure 5). After 1 h, LTP was induced
was inspired by the model developed by Ennaceur and
Delacour46 and was based on the fact that animals
remembering a familiar object spend less time exploring it
compared to exploring a new object. As can be seen in Figure 6,
Figure 6. Effect of 36c (po) on the recognition phase in an object
recognition task in CD1 mice. 36c was administered by oral route 1 h
before habituation, familiarization, and recognition sessions, given at
24 h intervals. Results are means + SEM of the durations of
exploration of new and familiar objects. ***p ≤ 0.001 versus familiar
object (paired Student’s t test).
oral administration of 36c 1 h before each sessions significantly
increased the performance of mice at doses as low as 1 mg/kg.
This effect of 36c in vivo on object recognition test in mice
strongly suggests that 36c is absorbed in mice after oral
administration and is able to reach the CNS.
Figure 5. Effect of 36c (3 and 10 mg/kg ip) on the potentiation of the
synaptic response evoked in the dentate gyrus of the hippocampus by
a high-frequency stimulation (HFS) delivered in the perforant path. In
each rat, synaptic responses were averaged over eight successive
stimulations (every 2 min) and normalized to the average amplitude of
synaptic responses obtained during the baseline 60 min period
(control value taken as 100%). Results are expressed as mean SEM.
*p ≤ 0.05 versus vehicle [Dunnett test after a significant two way
(time × treatment) ANOVA]. Closed symbols: p ≤ 0.05 versus the
average amplitude of synaptic response obtained during the 6 min-
period preceding the tetanus, analyzed for each group separately on
data expressed as a percentage of the 6 min-period preceding the
tetanus.
CONCLUSIONS
■
In conclusion, the 3D structures of compounds 5 and 6, two
PTD-type AMPApams, cocrystallized with GluA2-LBD-L483Y-
N754S were determined in order to decipher the impact of the
position of the nitrogen atom of the heterocycle on their
binding mode and affinity at the AMPA receptors. It was found
that the two PTDs bind in the same region as the previously
crystallized BTDs 7 and 9. The 8-aza compound 5 showed a
complete overlap with both BTD compounds 7 and 9, whereas
the 7-aza analogue 6 had a slightly different binding mode
compared to the three others. Such a difference could explain in
part the lower activity on AMPARs of the 7-aza compound 6
compared to its 8-aza counterpart 5.
by a tetanic high-frequency stimulation, and the EPSfPs
continued to be measured during 3 h. Compound 36c at 10
mg/kg significantly increased both the induction and the
maintenance of LTP compared to control (p ≤ 0.05 versus
vehicle, Dunnett test after significant (time × treatment)
ANOVA interaction, p ≤ 0.01). This effect was not observed at
3 mg/kg (Figure 5). The potentiation of the synaptic response
evoked by the tetanus was significant during the first 6 min after
tetanus in vehicle rats (n = 8; p ≤ 0.001), the first 18 min in rats
treated with 3 mg/kg 36c (n = 8; p ≤ 0.01) and during 42 min
in rats treated with 10 mg/kg 36c (n = 8; p ≤ 0.001) (Figure
5). Such an effect on LTP is a key feature for a potential
cognitive enhancer. Moreover, these in vivo results clearly
highlighted the ability of 36c to cross the blood−brain barrier
and to reach the CNS.
The affinity of the two compounds for the receptor was
determined by isothermal titration calorimetry and highlighted
the better binding affinity of 5 compared to 6.
Due to the positive impact of the replacement of the ethyl
chain by a cyclopropyl chain at the 4-position of the
heterocycle, the synthesis and biological evaluation of new
PTDs bearing such a 4-cyclopropyl moiety were performed and
completed with the synthesis of their corresponding 4-
cyclopropyl-substituted BTDs with a chlorine atom at the
different possible positions on the benzene ring. In accordance
with previous structure−activity relationships, the 8-aza
compound 32 (4-cyclopropyl-3,4-dihydro-2H-pyrido[3,2-e]-
1,2,4-thiadiazine 1,1-dioxide) was found to be the most potent
PTD-type AMPA receptor potentiator in vitro, whereas the 7-
chloro-substituted compound 36c (7-chloro-4-cyclopropyl-3,4-
dihydro-2H-1,2,4-benzothiadiazine 1,1-dioxide) emerged as the
most promising chloro-substituted BTD.
Lastly, effects of 36c were evaluated in vivo in an object
recognition test with CD1 mice. This three-session test,
considered as a paradigm for episodic memory in rodents,
I
dx.doi.org/10.1021/jm501268r | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Isothermal Titration Calorimetry. ITC was performed using an
ITC200 MicroCalorimeter (GE Healthcare) with a cell volume of 200
μL at 25 °C. Protein concentration was determined by UV absorption.
The protein solution consisted of either 96 μM (for compound 6
titration) or 50 μM (for compound 5 titration) GluA2-LBD-L483Y-
N754S in 100 mM HEPES, 100 mM NaCl, 2 mM KCl, 5 mM L-Glu,
pH 7.0. The buffer used for the protein solution was also used to
dissolve compounds 5 and 6. Both experiments were direct titrations
with 8.2 mM 6 and 1.9 mM 5, using 20 injections of 2 μL at 3 min
intervals, except for the first injection, which was 0.4 μL. The Origin
7.0 software (MicroCal) was used for data analysis (single binding site
model). Prior to fitting, the first data point was discarded and the heat
of dilution of injecting ligand into buffer was subtracted. The reported
ΔH, −TΔS, and Kd are mean values of three independent titrations.
Chemistry. All commercial chemicals (Sigma-Aldrich, Appolo
Scientific, and Fluorochem) and solvents were reagent grade and used
without further purification. Melting points were determined on a
Stuart SMP3 apparatus in open capillary tubes and are uncorrected.
NMR spectra were recorded on a Bruker Avance 500 spectrometer
(1H, 500 MHz; 13C, 125 MHz) using DMSO-d6 as solvent and
tetramethylsilane (TMS) as internal standard; chemical shifts are
reported in δ values (ppm) relative to internal TMS. The abbreviation
s = singlet, d = doublet, t = triplet, q = quadruplet, m = multiplet and
bs = broad signal are used throughout. Elemental analyses (C, H, N, S)
were carried out on a Thermo Flash EA 1112 series elemental analyzer
and were within 0.4% of the theoretical values. This analytical
process ensured for each target compound a purity equal or greater
than 95%. All reactions were followed by TLC (silica gel 60F254
Merck) and visualization was accomplished with UV light (254 or 366
nm).
Due to favorable physicochemical properties (drug-likeness)
and a low in vivo acute toxicity observed in mice with 36c
(Irwin’s test), this BTD compound was selected for a more
complete preclinical evaluation. Measuring the cerebral and
plasma concentration of the compound (LC−MS/MS)
confirmed the easy distribution of 36c into the brain after
oral administration (3 mg/kg) in mice. In an ex vivo
electrophysiological test performed on rat hippocampal slices,
the effects of 36c were measured on the excitatory postsynaptic
field potentials (EPSfPs) evoked in the CA1 region after
electrical stimulation of the Schaffer collateral. A marked
increase of the signals when tested at 10 μM suggested that the
compound was able to interact with postsynaptic AMPARs
located on hippocampal CA1 neurons. An in vivo test
conducted with 36c (electrophysiological recordings of the
postsynaptic response in the hippocampus of anesthetized
Wistar rats) confirmed the in vitro effects observed on EPSfPs
on hippocampal slices. In this test, compound 36c at 10 mg/kg
significantly increased both the induction and the maintenance
of LTP compared to control. Such an effect on LTP is a key
feature for a potential cognitive enhancer. Lastly, in an in vivo
object recognition test with CD1 mice, oral administration of
36c was found to significantly improve cognition performance
at doses as low as 1 mg/kg. Taken as a whole, the present study
highlighted compound 36c and its analogues as promising
small-sized molecules for a future development as new drug
candidates.
2-Chloropyridine-3-sulfonamide (13). Glacial acetic acid (30 mL)
was introduced in a round-bottom flask (500 mL) and saturated for 30
min with gaseous sulfur dioxide. To this solution an aqueous solution
of cupric chloride (1.5 g in 10 mL) was added, giving a suspension of
cuprous chloride (suspension A). In another flask, 3-amino-2-
chloropyridine (12) (5 g) was dissolved in a mixture of glacial acetic
acid (30 mL) and 12 N HCl (15 mL), and the resulting solution
(solution B) was cooled on an ice/salt bath (−5 °C). A solution of
sodium nitrite (2.5 g) in water (10 mL) was added dropwise to
solution B, and the resulting mixture was slowly added to suspension A
and stirred on an ice bath for 15 min. The reaction mixture was then
poured on a mixture of water (200 mL) and diethyl ether (200 mL).
The organic layer was separated, washed with water (100 mL), and
concentrated to dryness under reduced pressure. The residue of 2-
chloropyridine-3-sulfonyl chloride was dissolved in dioxane (25 mL),
and this solution was poured under stirring on a cooled mixture of
concentrated ammonia (25 mL) and water (10 mL). After 30 min, the
reaction mixture was concentrated under reduced pressure and the
resulting precipitate of the final compound was collected by filtration,
washed with water, and recrystallized in methanol−water to give a
white solid (yield 60%): mp 187−189 °C (lit.57 mp 189−190 °C); 1H
NMR (DMSO-d6) δ 7.65 (dd, J = 7.8 Hz/4.8 Hz, 1H, 5-H), 7.85 (s,
2H, SO2NH2), 8.37 (dd, J = 7.8 Hz/1.8 Hz, 1H, 4-H), 8.61 (dd, J = 4.8
Hz/1.8 Hz, 1H, 6-H); 13C NMR (DMSO-d6) δ 123.6 (C-5), 138.0 (C-
3), 138.4 (C-4), 146.5 (C-2), 152.2 (C-6).
2-Cyclopropylaminopyridine-3-sulfonamide (14). The solution of
2-chloropyridine-3-sulfonamide (13) (3 g) in dioxane (30 mL) and
cyclopropylamine (3 mL) was heated in a closed vessel at 110 °C for
24 h. The solvent and the excess of amine were removed by distillation
under reduced pressure, and the residue was dissolved in methanol (20
mL). The methanolic solution was cooled and mixed with water (60
mL) under stirring. The resulting precipitate was collected by
filtration, washed with water, dried, and used in the next step without
further purification (yield 85%): 1H NMR (DMSO-d6) δ 0.51 (m, 2H,
CH(CH2)2), 0.75 (m, 2H, CH(CH2)2), 2.82 (m, 1H, CH(CH2)2),
6.43 (s, 1H, NH), 6.74 (dd, J = 7.7 Hz/4.8 Hz, 1H, 5-H), 7.38 (bs, 2H,
SO2NH2), 7.87 (dd, J = 7.7 Hz/1.8 Hz, 1H, 4-H), 8.29 (dd, J = 4.8
Hz/1.8 Hz, 1H, 6-H); 13C NMR (DMSO-d6) δ 6.8 (CH(CH2)2), 24.2
(CH(CH2)2), 111.8 (C-5), 121.1 (C-3), 136.5 (C-4), 151.7 (C-6),
153.9 (C-2).
EXPERIMENTAL SECTION
■
Structural Biology. Protein Expression and Purification. GluA2-
LBD-L483Y-N754S was expressed and purified as previously
described.34 Briefly, Escherichia coli Origami B (DE3) cells were
transformed with the GluA2-LBD-L483Y-N754S pET-22b(+) plas-
mid. Cells were grown in LB medium to an OD600 of 0.9 and
subsequently cooled on ice to 20 °C, where protein expression was
induced by 0.5 mM isopropyl β-D-thiogalactopyranoside. Eighteen
hours later, cells were harvested and lysed, and Ni2+-affinity
chromatography was used to purify the soluble protein. To remove
the N-terminal His-tag, a tryptic digest was performed. The cleaved
protein was further purified by anion-exchange chromatography and
size-exclusion chromatography.
Crystallization. A solution of 6 mg/mL GluA2-LBD-L483Y-N754S
in 10 mM HEPES, 20 mM NaCl, 1 mM EDTA, 1 mM (S)-glutamate,
pH 7.0 was used for crystallization, and 100 μL of protein solution was
incubated with 0.5−1 mg of solid compound 5 or 6 at 7 °C for at least
24 h prior to crystallization at 7 °C. For the hanging drop vapor
diffusion method, drops consisted of 1 μL of protein solution and 1 μL
of reservoir solution, and the reservoir volume was 0.5 mL.
Crystallization conditions were as follows for 5: 24% PEG4000, 0.1
M ammonium sulfate, 0.1 M phosphate−citrate, pH 4.5. Crystal-
lization conditions were as follows for 6: 22% PEG4000, 0.2 M
ammonium sulfate, 0.1 M phosphate−citrate, pH 4.5. Crystals were
cryoprotected by fast soaking in reservoir solution with 20% glycerol
and flash-cooled in liquid nitrogen.
X-ray Structure Determination. Diffraction data were collected at
beamline I911-3, MAX-Lab in Lund, Sweden. XDS47 and SCALA
within CCP448 were used for data processing. The structures were
solved in Phaser49 by molecular replacement using the structure of
GluA2-LBD-L483Y (PDB code 1LB8)50 as search model. For initial
model building, ARP/wARP51 was used and the program Coot52 was
used for manual model building. PRODRG53 and eLBOW54 were used
to obtain topology and parameter files for 5 and 6. Refinement was
performed using Phenix.55 DynDom56 was used to calculate domain
closures, and Figure 2 was prepared using PyMOL (The PyMOL
Molecular Graphics System. Version 1.5.0.5, Schrodinger, LLC).
̈
J
dx.doi.org/10.1021/jm501268r | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
4-Cyclopropyl-4H-pyrido[2,3-e]-1,2,4-thiadiazine 1,1-Dioxide
(15). In an open vessel, a mixture of 2-cyclopropylaminopyridine-3-
sulfonamide (14) (2 g) and triethyl orthoformate (20 mL) was heated
at 150 °C for 3 h. The resulting suspension was cooled on an ice bath
and the insoluble material was collected by filtration, washed with
diethyl ether, dried, and crystallized in methanol to give a white solid
(s, 1H, 6-H); 13C NMR (DMSO-d6) δ 6.8 (CH(CH2)2), 31.9
(CH(CH2)2), 119.2 (C-8a), 128.6 (C-7), 133.4 (C-8), 145.4 (C-4a),
151.4 (C-6), 151.5 (C-3).
7-Chloro-4-cyclopropyl-3,4-dihydro-2H-pyrido[2,3-e]-1,2,4-thia-
diazine 1,1-Dioxide (22). The title compound was obtained as
described for 16 starting from 7-chloro-4-cyclopropyl-4H-pyrido[2,3-
e]-1,2,4-thiadiazine 1,1-dioxide (21) (0.1 g) to give a white solid (yield
69%): mp 197−199 °C; 1H NMR (DMSO-d6) δ 0.68 (m, 2H,
CH(CH2)2), 0.86 (m, 2H, CH(CH2)2), 2.68 (m, 1H, CH(CH2)2),
4.75 (s, 2H, 3-CH2), 8.09 (m, 1H, 8-H), 8.17 (bs, 1H, NH), 8.45 (s,
1H, 6-H); 13C NMR (DMSO-d6) δ 7.8 (CH(CH2)2), 29.4
(CH(CH2)2), 59.3 (C-3), 119.0 (C-8a), 119.7 (C-7), 131.9 (C-8),
150.6 (C-6), 151.5 (C-4a).
4-Cyclopropylaminopyridine-3-sulfonamide (24). The title com-
pound was obtained as described for 14 starting from 4-
chloropyridine-3-sulfonamide58 (1 g), except for the heating
conditions (100 °C, 24 h) (yield 77%). The resulting white solid
was used in the next step without further purification: 1H NMR
(DMSO-d6) δ 0.57 (m, 2H, CH(CH2)2), 0.85 (m, 2H, CH(CH2)2),
2.55 (m, 1H, CH(CH2)2), 6.55 (s, 1H, NH), 7.05 (d, J = 5.9 Hz, 1H,
5-H), 7.54 (s, 2H, SO2NH2), 8.29 (d, J = 5.8 Hz, 1H, 6-H), 7.65 (s,
1H, 8-H); 13C NMR (DMSO-d6) d 7.1 (CH(CH2)2), 24.1
(CH(CH2)2), 107.5 (C-3), 121.8 (C-1), 148.1 (C-6), 150.1 (C-2),
152.3 (C-4).
1
(yield 84%): mp 230−232 °C; H NMR (DMSO-d6) δ 0.99 (m, 2H,
CH(CH2)2), 1.08 (m, 2H, CH(CH2)2), 3.39 (m, 1H, CH(CH2)2),
7.62 (dd, J = 7.9 Hz/4.7 Hz, 1H, 7-H), 8.30 (s, 1H, 3-H), 8.40 (dd, J =
7.9 Hz/1.7 Hz, 1H, 8-H), 8.85 (dd, J = 4.7 Hz/1.7 Hz, 1H, 6-H); 13C
NMR (DMSO-d6) δ 6.8 (CH(CH2)2), 31.8 (CH(CH2)2), 118.7 (C-
8a), 123.0 (C-7), 134.2 (C-8), 146.7 (C-4a), 151.6 (C-6), 152.8 (C-3).
4-Cyclopropyl-3,4-dihydro-2H-pyrido[2,3-e]-1,2,4-thiadiazine
1,1-Dioxide (16). A mixture of 4-cyclopropyl-4H-pyrido[2,3-e]-1,2,4-
thiadiazine 1,1-dioxide (15) (0.8 g) and 2-propanol (10 mL) was
supplemented with finely divided sodium borohydride (0.7 g) and
then heated at 50 °C for 5 min. The solvent was removed by
distillation under reduced pressure and the residue was taken up with
water (10 mL). The resulting suspension was slightly acidified by the
addition of 6 N HCl. The title compound was extracted three times
with dichloromethane (3 × 30 mL). The combined organic layers
were dried over magnesium sulfate, and the filtrate was concentrated
under reduced pressure. The residue was crystallized in methanol−
1
water to give a white solid (yield 50%): mp 141−143 °C; H NMR
(DMSO-d6) δ 0.67 (m, 2H, CH(CH2)2), 0.84 (m, 2H, CH(CH2)2),
2.67 (m, 1H, CH(CH2)2), 4.73 (s, 2H, 3-CH2), 6.86 (dd, J = 7.6 Hz/
4.8 Hz, 1H, 7-H), 7.91 (dd, J = 7.6 Hz/1.8 Hz, 1H, 8-H), 8.00 (bs, 1H,
NH), 8.39 (dd, J = 4.8 Hz/1.8 Hz, 1H, 6-H); 13C NMR (DMSO-d6) δ
7.9 (CH(CH2)2), 29.3 (CH(CH2)2), 59.4 (C-3), 113.3 (C-7), 119.4
(C-8a), 132.7 (C-8), 152.3 (C-6), 152.8 (C-4a).
3-Amino-5-chloro-2-fluoropyridine (18). 5-Chloro-2-fluoro-3-ni-
tropyridine (17) (5 g) was dissolved in a hot 1:1 mixture of
ethanol−water (100 mL). Ammonium chloride (2.5 g) and iron
powder (10 g) were added to the solution, and the reaction mixture
was refluxed for 10 min. The insoluble material was removed by
filtration and the hot filtrate was treated with charcoal. The filtrate was
concentrated to dryness and the residue was taken up with water (20
mL) and cooled on an ice bath. The precipitate was collected by
filtration, dried, and used in the next step without further purification
(yield 30%): 1H NMR (DMSO-d6) δ 5.84 (bs, 2H, NH2), 7.16 (dd, J =
9.2 Hz/2.4 Hz, 1H, 4-H), 7.32 (t, J = 2.3 Hz, 1H, 6-H); 13C NMR
(DMSO-d6) δ 121.8 (d, J = 7 Hz, C-4), 127.9 (d, J = 3 Hz, C-5), 128.7
(d, J = 14 Hz, C-6), 133.0 (d, J = 31 Hz, C-3), 149.2−151.0 (d, J = 231
Hz, C-2).
4-Cyclopropyl-4H-pyrido[4,3-e]-1,2,4-thiadiazine 1,1-Dioxide
(25). The title compound was obtained as described for 15 starting
from 4-cyclopropylaminopyridine-3-sulfonamide (24) (0.5 g), except
for the heating conditions (120 °C, 1 h), to give a white solid (yield
86%): mp 192−194 °C; 1H NMR (DMSO-d6) δ 1.05 (m, 2H,
CH(CH2)2), 1.16 (m, 2H, CH(CH2)2), 3.36 (m, 1H, CH(CH2)2),
7.74 (d, J = 5.9 Hz, 1H, 5-H), 8.25 (s, 1H, 3-H), 8.86 (d, J = 5.9 Hz,
1H, 6-H), 9.03 (s, 1H, 8-H); 13C NMR (DMSO-d6) δ 7.1
(CH(CH2)2), 31.9 (CH(CH2)2), 110.8 (C-5), 118.4 (C-8a), 142.8
(C-4a), 146.0 (C-8), 152.4 (C-3), 152.9 (C-6).
4-Cyclopropyl-3,4-dihydro-2H-pyrido[4,3-e]-1,2,4-thiadiazine
1,1-Dioxide (26). The title compound was obtained as described for
16 starting from 4-cyclopropyl-4H-pyrido[4,3-e]-1,2,4-thiadiazine 1,1-
dioxide (25) (0.25 g) to give a white solid (yield 83%): mp 217−219
1
°C; H NMR (DMSO-d6) δ 0.71 (m, 2H, CH(CH2)2), 0.95 (m, 2H,
CH(CH2)2), 2.62 (m, 1H, CH(CH2)2), 4.78 (s, 2H, 3-CH2), 7.13 (d, J
= 6 Hz, 1H, 5-H), 8.08 (bs, 1H, NH), 8.33 (d, J = 6 Hz, 1H, 6-H), 8.50
(s, 1H, 8-H); 13C NMR (DMSO-d6) δ 8.0 (CH(CH2)2), 29.4
(CH(CH2)2), 60.8 (C-3), 108.9 (C-5), 119.8 (C-8a), 144.5 (C-8),
148.8 (C-4a), 151.9 (C-6).
3-Fluoropyridine-2-thiol (28). 2-Chloro-3-fluoropyridine (27) (5
g) was added to a solution of thiourea (2 g) in hot ethanol (40 mL).
The reaction mixture was heated to 80 °C for 7 h. After cooling,
diethyl ether (20 mL) was added and the precipitate of the
isothiouronium salt was collected by filtration and washed with
diethyl ether (yield 63%). The salt (1.5 g) was dissolved in water (10
mL) and supplemented with sodium carbonate (1.5 g). A solution of
10% m/v NaOH in water was added dropwise under stirring until pH
12. The persistent insoluble material was removed by filtration and the
filtrate was acidified by means of 12 N HCl. The title compound,
which precipitated, was collected by filtration, washed with water, and
5-Chloro-2-fluoropyridine-3-sulfonamide (19). The title com-
pound was obtained as described for 13 starting from 3-amino-5-
chloro-2-fluoropyridine (18) (1 g) to give a white solid (yield 60%):
mp 146−148 °C; 1H NMR (DMSO-d6) δ 8.02 (s, 2H, SO2NH2), 8.35
(m, 1H, 6-H), 8.59 (s, 1H, 4-H); 13C NMR (DMSO-d6) δ 127.5 (d, J
= 33 Hz, C-3), 128.3 (d, J = 5 Hz, C-5), 139.1 (d, J = 2 Hz, C-4),
149.4 (d, J = 16 Hz, C-6), 155.2−157.2 (d, J = 242 Hz, C-2).
5-Chloro-2-cyclopropylaminopyridine-3-sulfonamide (20). The
title compound was obtained as described for 14 starting from 5-
chloro-2-fluoropyridine-3-sulfonamide (19) (0.5 g), except for the
heating conditions (reflux overnight in an open vessel). After removal
of the solvent by distillation under reduced pressure, the residue was
1
dried to give a white solid (yield 95%): mp 169−171 °C; H NMR
1
(DMSO-d6) δ 6.76 (m, 1H, 5-H), 7.48 (t, J = 7.6 Hz, 1H, 4-H), 7.60
(d, J = 5 Hz, 1H, 6-H), 13.95 (s, 1H, SH); 13C NMR (DMSO-d6) δ
112.3 (d, J = 6 Hz, C-5), 120.6 (d, J = 22 Hz, C-4), 134.5 (d, J = 5 Hz,
C-6), 159.7−161.6 (d, J = 238 Hz, C-3), 168.0 (d, J = 32 Hz, C-2).
3-Fluoropyridine-2-sulfonamide (29). A solution of 3-fluoropyr-
idine-2-thiol (28) (0.9 g) in 12 N HCl (6 mL) and water (1.5 mL) was
cooled at −15 °C and then treated with chlorine gas for 30 min. 3-
Fluoropyridinesulfonyl chloride was extracted from the reaction
mixture by means of diethyl ether (2 × 30 mL). The combined
organic layers were concentrated to dryness by distillation under
reduced pressure. The residue was dissolved in dioxane (20 mL) and
the resulting solution was added dropwise to a 25% w/v aqueous
solution of ammonia (30 mL). The solution was concentrated and the
white solid precipitate was collected by filtration, dried, and used in the
used in the next step without further purification: H NMR (DMSO-
d6) δ 0.53 (m, 2H, CH(CH2)2), 0.76 (m, 2H, CH(CH2)2), 2.80 (m,
1H, CH(CH2)2), 6.53 (s, 1H, NH), 7.64 (bs, 2H, SO2NH2), 7.86 (d, J
= 2.5 Hz, 1H, 4-H), 8.35 (d, J = 2.5 Hz, 1H, 6-H); 13C NMR (DMSO-
d6) δ 6.8 (CH(CH2)2), 24.4 (CH(CH2)2), 117.2 (C-5), 121.9 (C-3),
135.5 (C-4), 149.8 (C-6), 152.5 (C-2).
7-Chloro-4-cyclopropyl-4H-pyrido[2,3-e]-1,2,4-thiadiazine 1,1-Di-
oxide (21). The residue of 20 was heated in an open vessel at 120 °C
for 1 h. The solvents were removed by distillation under reduced
pressure, and the residue was taken up in methanol, giving a white
precipitate of the title compound (yield 9% over the two last steps),
which was used in the next step without further purification: 1H NMR
(DMSO-d6) δ 1.00 (m, 2H, CH(CH2)2), 1.08 (m, 2H, CH(CH2)2),
3.38 (m, 1H, CH(CH2)2), 8.32 (s, 1H, 3-H), 8.63 (s, 1H, 8-H), 8.92
K
dx.doi.org/10.1021/jm501268r | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
1
next step without further purification: mp 117−119 °C; H NMR
(DMSO-d6) δ 7.77 (m, 1H, 5-H), 7.80 (s, 2H, SO2NH2), 8.00 (ddd, J
= 9.9 Hz/8.5 Hz/1.1 Hz, 1H, 4-H), 8.52 (m, 1H, 6-H); 13C NMR
(DMSO-d6) δ 126.5 (d, J = 19 Hz, C-4), 129.4 (d, J = 5 Hz, C-5),
144.6 (d, J = 5 Hz, C-6), 147.0 (d, J = 15 Hz, C-2), 154.3−156.4 (d, J
= 264 Hz, C-3).
133.1 (d, J = 14 Hz, C-1), 134.5 (C-4), 152.4−154.4 (d, J = 255 Hz,
C-2).
4-Chloro-2-fluorobenzenesulfonamide (33b). White solid; mp
104−106 °C; H NMR (DMSO-d6) δ 7.48 (dd, J = 8.5 Hz/1.3 Hz,
1H, 5-H), 7.72 (dd, J = 10 Hz/1.7 Hz, 1H, 3-H), 7.78 (bs, 2H,
SO2NH2), 7.81 (d, J = 8.3 Hz, 1H, 6-H); 13C NMR (DMSO-d6) δ
117.7 (d, J = 25 Hz, C-3), 125.0 (d, J = 4 Hz, C-6), 129.7 (C-5), 130.7
(d, J = 15 Hz, C-1), 138.0 (d, J = 10 Hz, C-4), 157.0−159.0 (d, J = 256
Hz, C-2).
2-Chloro-6-fluorobenzenesulfonamide (33d). White solid; mp
187−190 °C; 1H NMR (DMSO-d6) δ 7.41 (m, 1H, 5-H), 7.48 (d, J =
7.5 Hz, 1H, 3-H), 7.61 (m, 1H, 4-H), 7.95 (s, 2H, SO2NH2); 13C
NMR (DMSO-d6) δ 116.6 (t, J = 25 Hz, C-5), 127.9 (d, J = 3 Hz, C-
3), 129.8 (d, J = 14 Hz, C-1), 131.9 (C-2), 133.7 (d, J = 11 Hz, C-4),
157.8−159.8 (d, J = 256 Hz, C-6).
General Procedure for the Synthesis of Chloro-Substituted 2-
Cyclopropylaminobenzenesulfonamides (34). A solution of the
appropriate chloro-substituted 2-fluorobenzenesulfonamide (33) (3
g) in dioxane (30 mL) and cyclopropylamine (3 mL) was heated in a
closed vessel at 100−110 °C for 24 h. The solvent and the excess of
amine were removed by distillation under reduced pressure, and the
residue was dissolved in methanol (20 mL). The methanolic solution
was cooled and mixed with water (60 mL) under stirring. The
resulting precipitate was collected by filtration, washed with water,
dried, and used in the next step without further purification (yield 80−
90%).
1
3-Cyclopropylaminopyridine-2-sulfonamide (30). The crude
compound 29 was dissolved in dioxane (15 mL) in a closed vessel
and supplemented with cyclopropylamine (1 mL). The closed vessel
was heated to 90 °C for 48 h. The reaction mixture was evaporated to
dryness and the crude residue containing the title compound was used
1
in the next step without further purification: H NMR (DMSO-d6) δ
0.48 (m, 2H, CH(CH2)2), 0.81 (m, 2H, CH(CH2)2), 2.48 (m, 1H,
CH(CH2)2), 6.46 (s, 1H, NH), 7.26 (bs, 2H, SO2NH2), 7.45 (dd, J =
8.5 Hz/4.3 Hz, 1H, 5-H), 7.57 (dd, J = 8.4 Hz/1 Hz, 1H, 4-H), 7.91
(dd, J = 4.2 Hz/1 Hz, 1H, 6-H); 13C NMR (DMSO-d6) δ 7.2
(CH(CH2)2), 24.1 (CH(CH2)2), 121.1 (C-4), 127.6 (C-5), 135.8 (C-
6), 141.3 (C-3), 142.0 (C-2).
4-Cyclopropyl-4H-pyrido[3,2-e]-1,2,4-thiadiazine 1,1-Dioxide
(31). The crude compound 30 was dissolved in triethyl orthoformate
(20 mL) and heated at 130 °C for 24 h. After cooling, diethyl ether
(10 mL) was added and the precipitate was isolated, dissolved in hot
acetone, treated with charcoal, and filtered, and the filtrate was
concentrated to dryness. The residue was crystallized in methanol to
give a white solid (yield 8% over the three last steps): mp 240−242
1
°C; H NMR (DMSO-d6) δ 1.04 (m, 2H, CH(CH2)2), 1.15 (m, 2H,
The following chloro-substituted 2-cyclopropylaminobenzenesulfo-
namides were obtained.
CH(CH2)2), 3.36 (m, 1H, CH(CH2)2), 7.86 (dd, J = 8.6 Hz/4.4 Hz,
1H, 6-H), 8.11 (s, 1H, 3-H), 8.28 (d, J = 8.6 Hz, 1H, 5-H), 8.71 (d, J =
3.6 Hz, 1H, 7-H); 13C NMR (DMSO-d6) δ 7.2 (CH(CH2)2), 31.6
(CH(CH2)2), 125.8 (d, J = 4 Hz, C-5), 128.2 (C-6), 134.4 (C-4a),
138.7 (C-8a), 147.9 (C-7), 150.4 (C-3).
3-Chloro-2-cyclopropylaminobenzenesulfonamide (34a). 1H
NMR (DMSO-d6) δ 0.52 (m, 2H, CH(CH2)2), 0.64 (m, 2H,
CH(CH2)2), 3.08 (m, 1H, CH(CH2)2), 5.61 (s, 1H, NH), 6.93 (t, J =
7.9 Hz, 1H, 5-H), 7.56 (d, J = 7.8 Hz, 1H, 4-H), 7.63 (bs, 2H,
SO2NH2), 7.66 (dd, J = 7.9 Hz/1.5 Hz, 1H, 6-H); 13C NMR (DMSO-
d6) δ 8.6 (CH(CH2)2), 29.3 (CH(CH2)2), 119.6 (C-5), 124.0 (C-1),
126.8 (C-6), 132.8 (C-3), 135.0 (C-4), 143.5 (C-2).
4-Cyclopropyl-3,4-dihydro-2H-pyrido[3,2-e]-1,2,4-thiadiazine
1,1-Dioxide (32). The title compound was obtained as described for
16 starting from 4-cyclopropyl-4H-pyrido[3,2-e]-1,2,4-thiadiazine 1,1-
dioxide (31) (0.2 g) to give a white solid (yield 84%): mp 198−200
4-Chloro-2-cyclopropylaminobenzenesulfonamide (34b). 1H
NMR (DMSO-d6) δ 0.53 (m, 2H, CH(CH2)2), 0.82 (m, 2H,
CH(CH2)2), 2.50 (m, 1H, CH(CH2)2), 6.23 (s, 1H, NH), 6.76 (dd, J
= 8.4 Hz/1.9 Hz, 1H, 5-H), 7.10 (d, J = 1.8 Hz, 1H, 3-H), 7.42 (s, 2H,
SO2NH2), 7.60 (d, J = 8.4 Hz, 1H, 6-H); 13C NMR (DMSO-d6) δ 7.2
(CH(CH2)2), 24.5 (CH(CH2)2), 112.0 (C-3), 115.1 (C-5), 124.0 (C-
1), 129.9 (C-6), 137.8 (C-4), 146.5 (C-2).
1
°C; H NMR (DMSO-d6) δ 0.65 (m, 2H, CH(CH2)2), 0.91 (m, 2H,
CH(CH2)2), 2.52 (m, 1H, CH(CH2)2), 4.66 (s, 2H, 3-CH2), 7.49 (dd,
J = 8.6 Hz/4.3 Hz, 1H, 6-H), 7.68 (dd, J = 8.6 Hz/1 Hz, 1H, 5-H),
8.00 (bs, 1H, NH), 8.02 (dd, J = 4.2 Hz/1 Hz, 1H, 7-H); 13C NMR
(DMSO-d6) δ 8.3 (CH(CH2)2), 29.0 (CH(CH2)2), 60.6 (C-3), 122.8
(C-5), 127.8 (C-6), 137.9 (C-7), 139.7 (C-4a), 141.9 (C-8a).
General Procedure for the Synthesis of Chloro-Substituted 2-
Fluorobenzenesulfonamides (33). Glacial acetic acid (30 mL) was
introduced in a round-bottom flask (500 mL) and saturated for 30 min
with gaseous sulfur dioxide. To this solution was added an aqueous
solution of cupric chloride (1.5 g in 10 mL), giving a suspension of
cuprous chloride (suspension A). In another flask, the appropriate
chloro-substituted 2-fluoroaniline (5 g) was dissolved in a mixture of
glacial acetic acid (30 mL) and 12 N HCl (15 mL), and the resulting
solution (solution B) was cooled on an ice/salt bath (−5 °C). A
solution of sodium nitrite (2.5 g) in water (10 mL) was added
dropwise to solution B, and the resulting mixture was slowly added to
suspension A and stirred on an ice bath for 15 min. The reaction
mixture was then poured on a mixture of water (200 mL) and diethyl
ether (200 mL). The organic layer was separated, washed with water
(100 mL), and concentrated to dryness under reduced pressure. The
residue of chloro-substituted 2-fluorobenzenesulfonyl chloride was
dissolved in dioxane (25 mL), and this solution was poured with
stirring onto a cooled mixture of concentrated ammonia (25 mL) and
water (10 mL). After 30 min, the reaction mixture was concentrated
under reduced pressure and the resulting precipitate of the final
compound was collected by filtration, washed with water, and
recrystallized in methanol−water (yield 60−70%).
6-Chloro-2-cyclopropylaminobenzenesulfonamide (34d). 1H
NMR (DMSO-d6) δ 0.43 (m, 2H, CH(CH2)2), 0.81 (m, 2H,
CH(CH2)2), 2.43 (m, 1H, CH(CH2)2), 6.79 (d, J = 7.7 Hz, 1H, 5-H),
7.13 (d, J = 8.6 Hz, 1H, 3-H), 7.33 (t, J = 8.1 Hz, 1H, 4-H), 7.47 (s,
1H, NH), 7.63 (bs, 2H, SO2NH2); 13C NMR (DMSO-d6) δ 7.4
(CH(CH2)2), 24.7 (CH(CH2)2), 112.5 (C-5), 118.3 (C-3), 121.8 (C-
1), 132.6 (C-2), 132.7 (C-4), 148.2 (C-6).
General Procedure for the Synthesis of Chloro-Substituted 4-
Cyclopropyl-4H-1,2,4-benzothiadiazine 1,1-Dioxides (35). In an
open vessel, a mixture of the appropriate chloro-substituted 2-
cyclopropylaminobenzenesulfonamide (34) (2 g) and triethyl
orthoformate (20 mL) was heated at 130 °C for 2−24 h. The
resulting suspension was cooled on an ice bath and the insoluble
material was collected by filtration, washed with diethyl ether, dried,
and crystallized in methanol (yield 70−80%).
The following chloro-substituted 4-cyclopropyl-4H-1,2,4-benzothia-
diazine 1,1-dioxides were obtained.
5-Chloro-4-cyclopropyl-4H-1,2,4-benzothiadiazine 1,1-dioxide
1
(35a). White solid; mp 159−160 °C; H NMR (DMSO-d6) δ 0.80
(m, 2H, CH(CH2)2), 1.13 (m, 2H, CH(CH2)2), 4.02 (m, 1H,
CH(CH2)2), 7.55 (t, J = 7.9 Hz, 1H, 7-H), 7.86 (dd, J = 7.9 Hz/1.4
Hz, 1H, 8-H), 7.91 (dd, J = 8 Hz/1.4 Hz, 1H, 6-H), 8.35 (s, 1H, 3-H);
13C NMR (DMSO-d6) δ 10.1 (CH(CH2)2), 35.7 (CH(CH2)2), 121.9
The following chloro-substituted 2-fluorobenzenesulfonamides were
obtained.
3-Chloro-2-fluorobenzenesulfonamide (33a). White solid; mp
149−153 °C; 1H NMR (DMSO-d6) δ 7.42 (t, J = 8 Hz, 1H, 5-H), 7.78
(m, 1H, 4-H), 7.85 (s, 2H, SO2NH2), 7.88 (m, 1H, 6-H); 13C NMR
(DMSO-d6) δ 121.1 (d, J = 17 Hz, C-3), 125.6 (C-6), 127.3 (C-5),
(C-8a), 123.5 (C-8), 126.7 (C-5), 128.0 (C-7), 133.9 (C-4a), 136.6
(C-6), 156.2 (C-3).
6-Chloro-4-cyclopropyl-4H-1,2,4-benzothiadiazine 1,1-dioxide
(35b). White solid; mp 238.5−241.5 °C; 1H NMR (DMSO-d6) δ
1.03 (m, 2H, CH(CH2)2), 1.18 (m, 2H, CH(CH2)2), 3.39 (m, 1H,
L
dx.doi.org/10.1021/jm501268r | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
CH(CH2)2), 7.62 (dd, J = 8.5 Hz/1.6 Hz, 1H, 7-H), 7.85 (d, J = 1.5
Hz, 1H, 5-H), 7.92 (d, J = 8.5 Hz, 1H, 8-H), 8.17 (s, 1H, 3-H); 13C
NMR (DMSO-d6) δ 7.3 (CH(CH2)2), 32.3 (CH(CH2)2), 116.6 (C-
5), 120.9 (C-8a), 126.3 (C-8), 126.9 (C-7), 137.8−137.9 (C-4a/C-6),
151.8 (C-3).
Effect on Excitatory Postsynaptic Response Evoked in CA1 Area
on Rat Hippocampal Slices in Vitro. This assay was performed
following our previously published procedure.31
Effect on Long-Term Potentiation (LTP) of the Postsynaptic
Response Evoked in the Dentate Gyrus in Anesthetized Rats.
Extracellular excitatory postsynaptic field potentials (EPSfP) were
recorded in the dentate gyrus using our previously published
procedure.31
8-Chloro-4-cyclopropyl-4H-1,2,4-benzothiadiazine 1,1-dioxide
1
(35d). White solid; mp 207−209 °C; H NMR (DMSO-d6) δ 1.05
(m, 2H, CH(CH2)2), 1.16 (m, 2H, CH(CH2)2), 3.37 (m, 1H,
CH(CH2)2), 7.63 (d, J = 7.9 Hz, 1H, 5-H), 7.79 (t, J = 8.3 Hz, 1H, 6-
H), 7.86 (d, J = 8.6 Hz, 1H, 7-H), 8.13 (s, 1H, 3-H); 13C NMR
(DMSO-d6) δ 7.6 (CH(CH2)2), 32.9 (CH(CH2)2), 116.3 (C-5), 120.8
(C-8a), 128.4 (C-7), 130.0 (C-8), 133.4 (C-6), 138.3 (C-4a), 149.7
(C-3).
Effect on Object Recognition Test in Mice. The one-trial object
recognition paradigm measuring a form of episodic memory in the
mouse was achieved following our previously described procedure.31,33
ASSOCIATED CONTENT
* Supporting Information
■
General Procedure for the Synthesis of Chloro-Substituted 4-
Cyclopropyl-3,4-dihydro-2H-1,2,4-benzothiadiazine 1,1-Dioxide
(36a). A mixture of the appropriate chloro-substituted 4-cyclo-
propyl-4H-1,2,4-benzothiadiazine 1,1-dioxide (35) (1.9 g) and 2-
propanol (50 mL) was supplemented with finely divided sodium
borohydride (1 g) and then heated at 50−55 °C for 5−10 min. The
solvent was removed by distillation under reduced pressure and the
residue was taken up with water (50 mL). The resulting suspension
was slightly acidified by the addition of 6 N HCl. The title compound
was extracted three times with dichloromethane (3 × 30 mL). The
combined organic layers were dried over magnesium sulfate, and the
filtrate was concentrated under reduced pressure. The residue was
crystallized in methanol−water (yield 80−85%).
S
Evolution of the UV spectrum of compound 6 from pH 3.5 to
7.5 and elemental analysis of the target compounds (16, 22, 26,
32, 36a, 36b, 36d). This material is available free of charge via
the Internet at http://pubs.acs.org.
Accession Codes
The structure coordinates and corresponding structure factor
file of GluA2-LBD-L483Y-N754S with 5 and 6 have been
deposited in the Protein Data Bank under the accession codes
4U4X and 4U4S.
The following chloro-substituted 4-cyclopropyl-3,4-dihydro-2H-
1,2,4-benzothiadiazine 1,1-dioxides were obtained.
AUTHOR INFORMATION
Corresponding Author
*Phone: + 32 4 366 43 65. Fax: + 32 4 366 43 62. E-mail: b.
pirotte@ulg.ac.be.
■
5-Chloro-4-cyclopropyl-3,4-dihydro-2H-1,2,4-benzothiadiazine
1
1,1-Dioxide (36a). White solid; mp 142−143 °C; H NMR (DMSO-
d6) δ 0.75 (m, 2H, CH(CH2)2), 0.79 (m, 2H, CH(CH2)2), 3.09 (m,
1H, CH(CH2)2), 4.74 (s, 2H, 3-CH2), 7.06 (m, 1H, 7-H), 7.60 (m,
2H, 6-H/8-H), 8.14 (s, 1H, NH); 13C NMR (DMSO-d6) δ 9.6
(CH(CH2)2), 36.5 (CH(CH2)2), 62.2 (C-3), 121.6 (C-7), 122.8 (C-
8), 124.7 (C-8a), 130.4 (C-5), 134.9 (C-6), 141.2 (C-4a).
Author Contributions
∥P.F. and A.B.N. equally contributed to the work. J.S.K. and
B.P. equally supervised the work.
Notes
6-Chloro-4-cyclopropyl-3,4-dihydro-2H-1,2,4-benzothiadiazine
1,1-dioxide (36b). White solid; mp 174−175.5 °C; 1H NMR (DMSO-
d6) δ 0.67 (m, 2H, CH(CH2)2), 0.94 (m, 2H, CH(CH2)2), 2.55 (m,
1H, CH(CH2)2), 4.69 (s, 2H, 3-CH2), 6.89 (m, 1H, 7-H), 7.24 (s, 1H,
5-H), 7.56 (m, 1H, 8-H), 7.97 (s, 1H, NH); 13C NMR (DMSO-d6) δ
8.4 (CH(CH2)2), 29.7 (CH(CH2)2), 61.0 (C-3), 113.7 (C-5), 117.0
(C-7), 121.9 (C-8a), 126.2 (C-8), 137.4 (C-6), 145.3 (C-4a).
8-Chloro-4-cyclopropyl-3,4-dihydro-2H-1,2,4-benzothiadiazine
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This study was supported in part by a grant from Servier and
the National Fund for Scientific Research (F.N.R.S., Belgium)
of which P.d.T. is a Senior Research Associate. The technical
assistance of Stephane Counerotte is gratefully acknowledged.
We thank MAX-lab (Lund, Sweden) for providing beamtime,
and The Danish Council for Independent Research (Mobility
PhD grant) (A.B.N., L.O., K.F., J.S.K.) and GluTarget and
Danscatt (A.B.N., T.D., L.O., K.F., J.S.K.) for financial support.
1
1,1-dioxide (36d). White solid; mp 182−184 °C; H NMR (DMSO-
d6) δ 0.67 (m, 2H, CH(CH2)2), 0.93 (m, 2H, CH(CH2)2), 2.51 (m,
1H, CH(CH2)2), 4.63 (s, 2H, 3-CH2), 6.92 (d, J = 7.8 Hz, 1H, 5-H),
7.26 (d, J = 8.7 Hz, 1H, 7-H), 7.40 (t, J = 8.6 Hz, 1H, 6-H), 8.09 (s,
1H, NH); 13C NMR (DMSO-d6) δ 8.8 (CH(CH2)2), 30.3
(CH(CH2)2), 60.4 (C-3), 113.8 (C-5), 119.2 (C-7), 121.7 (C-8a),
130.6 (C-8), 132.8 (C-6), 146.7 (C-4a).
ABBREVIATIONS USED
■
Determination of pKa Values. Determination of pKa values was
carried out at 25 °C using the UV method (PerkinElmer UV−visible
554 spectrophotometer) with buffers ranging from pH 3.0 to 11.0. The
AMPA, α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid;
AMPARs, AMPA receptors; AMPApams, AMPA receptors
positive allosteric modulators; BDNF, brain derived neuro-
trophic factor; BTDs, benzothiadiazine dioxides; CNS, central
nervous system; CTZ, cyclothiazide; EPSfPs, excitatory
postsynaptic field potentials; GluA2-LBD, ligand binding
domain of the AMPAR subunit GluA2; iGluRs, ionotropic
glutamate receptors; ITC, isothermal titration calorimetry; KA,
kainic acid; LBD, ligand binding domain; mGluRs, metabo-
tropic glutamate receptors; NA, noradrenaline; NMDA, N-
methyl-D-aspartic acid; PTDs, pyridothiadiazine dioxides;
TARPs, transmembrane AMPA receptor regulatory proteins;
TMS, tetramethylsilane.
pK values were calculated by the Debye−Huckel equation at the
̈
a
wavelength giving the maximum absorbance.59
Biological Assays. Effect on AMPA-Evoked Membrane Depola-
rization (in Vitro Fluorescence Assay). This assay investigating
AMPA-evoked membrane depolarization was performed on rat
primary brain cultures using fluorescent membrane potential dyes
and an imaging-based plate reader. The assay was performed following
our previously published procedure.31
Effect on AMPA-Mediated Release of Noradrenaline on Rat
Hippocampal Slices. Potentiation of noradrenaline release on rat
hippocampal slices was measured according to our previously reported
procedure.33
Determination of Cerebral and Plasma Concentration of Drug
after Oral Administration. Twenty minutes after oral administration
of the compound, NMRI mice were sacrificed by decapitation. Brain
and blood samples were rapidly removed and frozen. Cerebral and
plasmatic concentrations of the compounds were determined by using
LC−MS/MS detection.
REFERENCES
■
(1) Francotte, P.; de Tullio, P.; Fraikin, P.; Counerotte, S.; Goffin, E.;
Pirotte, B. In search of novel AMPA potentiators. Recent Pat. CNS
Drug Discovery 2006, 1, 239−246.
M
dx.doi.org/10.1021/jm501268r | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(2) Staubli, U.; Rogers, G.; Lynch, G. Facilitation of glutamate
receptors enhances memory. Proc. Natl. Acad. Sci. U. S. A. 1994, 91,
777−781.
(3) Palmer, C. L.; Cotton, L.; Henley, J. M. The molecular
pharmacology and cell biology of R-amino-3-hydroxy-5-methyl-4-
isoxazolepropionic acid receptors. Pharmacol. Rev. 2005, 57, 253−277.
(4) Mansour, M.; Nagarajan, N.; Nehring, R. B.; Clements, J. D.;
Rosenmund, C. Heteromeric AMPA receptors assemble with a
preferred subunit stoichiometry and spatial arrangement. Neuron
2001, 32, 841−853.
(23) Yamada, K. A.; Rothman, S. M. Diazoxide blocks glutamate
desensitization and prolongs excitatory postsynaptic currents in rat
hippocampal neurons. J. Physiol. 1992, 458, 409−423.
(24) Yamada, K. A.; Tang, C. M. Benzothiadiazides inhibit rapid
glutamate receptor desensitization and enhance glutamatergic synaptic
currents. J. Neurosci. 1993, 13, 3904−3915.
(25) Bertolino, M.; Baraldi, M.; Parenti, C.; Braghiroli, D.; DiBella,
M.; Vicini, S.; Costa, E. Modulation of AMPA/kainate receptors by
analogues of diazoxide and cyclothiazide in thin slices of rat
hippocampus. Recept. Channels 1993, 1, 267−278.
(26) Zivkovic, I.; Thompson, D. M.; Bertolino, M.; Uzunov, D.;
DiBella, M.; Costa, E.; Guidotti, A. 7-Chloro-3-methyl-3,4-dihydro-
2H-1,2,4-benzothiadiazine S,S-dioxide (IDRA 21): A benzothiadiazine
derivative that enhances cognition by attenuating DL-alpha-amino-2,3-
dihydro-5-methyl-3-oxo-4-isoxazolepropanoic acid (AMPA) receptor
desensitization. J. Pharmacol. Exp. Ther. 1995, 272, 300−309.
(27) Thompson, D. M.; Guidotti, A.; DiBella, M.; Costa, E. 7-
Chloro-3-methyl-3,4-dihydro-2H-1,2,4-benzothiadiazine S,S-dioxide
(IDRA 21), a congener of aniracetam, potently abates pharmacolog-
ically induced cognitive impairments in patas monkeys. Proc. Natl.
Acad. Sci. U. S. A. 1995, 92, 7667−7671.
(28) Pirotte, B.; Podona, T.; Diouf, O.; de Tullio, P.; Lebrun, P.;
Dupont, L.; Somers, F.; Delarge, J.; Morain, P.; Lestage, P.; Lepagnol,
J.; Spedding, M. 4H-1,2,4-Pyridothiadiazine 1,1-dioxides and 2,3-
dihydro-4H-1,2,4-pyridothiadiazine 1,1-dioxides chemically related to
diazoxide and cyclothiazide as powerful positive allosteric modulators
of (R/S)-2-amino-3-(3-hydroxy-5-methylisoxazol-4-yl)propionic acid
receptors: Design, synthesis, pharmacology, and structure−activity
relationships. J. Med. Chem. 1998, 41, 2946−2959.
(29) Francotte, P.; de Tullio, P.; Goffin, E.; Dintilhac, G.;
Graindorge, E.; Fraikin, P.; Lestage, P.; Danober, L.; Thomas, J.-Y.;
Caignard, D.-H.; Pirotte, B. Design, synthesis, and pharmacology of
novel 7-substituted 3,4-dihydro-2H-1,2,4-benzothiadiazine 1,1-dioxides
as positive allosteric modulators of AMPA receptors. J. Med. Chem.
2007, 50, 3153−3157.
(30) Francotte, P.; de Tullio, P.; Podona, T.; Diouf, O.; Fraikin, P.;
Lestage, P.; Danober, L.; Thomas, J.-Y.; Caignard, D.-H.; Pirotte, B.
Synthesis and pharmacological evaluation of a second generation of
pyridothiadiazine 1,1-dioxides acting as AMPA potentiators. Bioorg.
Med. Chem. 2008, 16, 9948−9956.
(31) Francotte, P.; Goffin, E.; Fraikin, P.; Lestage, P.; Van Heugen, J.-
C.; Gillotin, F.; Danober, L.; Thomas, J.-Y.; Chiap, P.; Caignard, D.-H.;
Pirotte, B.; de Tullio, P. New fluorinated 1,2,4-benzothiadiazine 1,1-
dioxides: Discovery of an orally active cognitive enhancer acting
through potentiation of the 2-amino-3-(3-hydroxy-5-methylisoxazol-4-
yl)propionic acid receptors. J. Med. Chem. 2010, 53, 1700−1711.
(32) Nørholm, A. B.; Francotte, P.; Olsen, L.; Krintel, C.;
Frydenvang, K.; Goffin, E.; Challal, S.; Danober, L.; Botez-Pop, I.;
Lestage, P.; Pirotte, B.; Kastrup, J. S. Synthesis, pharmacological and
structural characterization, and thermodynamic aspects of GluA2-
positive allosteric modulators with a 3,4-dihydro-2H-1,2,4-benzothia-
diazine 1,1-dioxide scaffold. J. Med. Chem. 2013, 56, 8736−8745.
(33) Francotte, P.; Goffin, E.; Fraikin, P.; Graindorge, E.; Lestage, P.;
Danober, L.; Challal, S.; Rogez, N.; Nosjean, O.; Caignard, D.-H.;
Pirotte, B.; de Tullio, P. Development of thiophenic analogues of
benzothiadiazine dioxides as new powerful potentiators of 2-amino-3-
(3-hydroxy-5-methylisoxazol-4-yl)propionic acid (AMPA) receptors. J.
Med. Chem. 2013, 56, 7838−7850.
(5) Robert, A.; Irizarry, S. N.; Hughes, T. E.; Howe, J. R. Subunit
interactions and AMPA receptor desensitization. J. Neurosci. 2001, 21,
5574−5586.
(6) Greger, I. H.; Esteban, J. A. AMPA receptor biogenesis and
trafficking. Curr. Opin. Neurobiol. 2007, 17, 289−297.
(7) Armstrong, N.; Gouaux, E. Mechanisms for activation and
antagonism of an AMPA-sensitive glutamate receptor: Crystal
structures of the GluR2 ligand binding core. Neuron 2000, 28, 165−
181.
(8) Gouaux, E. Structure and function of AMPA receptors. J. Physiol.
2004, 554, 249−253.
(9) Collingridge, G. L.; Isaac, J. T.; Wang, Y. T. Receptor trafficking
and synaptic plasticity. Nature Rev. Neurosci. 2004, 5, 952−962.
(10) Pirotte, B.; Francotte, P.; Goffin, E.; de Tullio, P. AMPA
receptor positive allosteric modulators: A patent review. Expert Opin.
Ther. Pat. 2013, 23, 615−628.
(11) Pirotte, B.; Francotte, P.; Goffin, E.; Fraikin, P.; Danober, L.;
Lesur, B.; Botez, I.; Caignard, D.-H.; Lestage, P.; de Tullio, P. Ring-
fused thiadiazines as core structures for the development of potent
AMPA receptor potentiators. Curr. Med. Chem. 2010, 17, 3575−3582.
(12) Dicou, E.; Rangon, C. M.; Guimiot, F.; Spedding, M.; Gressens,
P. Positive allosteric modulators of AMPA receptors are neuro-
protective against lesions induced by an NMDA agonist in neonatal
mouse brain. Brain Res. 2003, 970, 221−225.
(13) Zarate, C. A., Jr.; Manji, H. K. The role of AMPA receptor
modulation in the treatment of neuropsychiatric diseases. Exp. Neurol.
2008, 211, 7−10.
(14) Alt, A.; Nisenbaum, E. S.; Bleakman, D.; Witkin, J. M. A role for
AMPA receptors in mood disorders. Biochem. Pharmacol. 2006, 71,
1273−1288.
(15) Goff, D. C.; Leahy, L.; Berman, I.; Posever, T.; Herz, L.; Leon,
A. C.; Johnson, S. A.; Lynch, G. A placebo-controlled pilot study of the
ampakine CX516 added to clozapine in schizophrenia. J. Clin.
Psychopharmacol. 2001, 21, 484−487.
(16) Lynch, G. AMPA receptor modulators as cognitive enhancers.
Curr. Opin. Pharmacol. 2004, 4, 4−11.
(17) Sanacora, G.; Zarate, C. A.; Krystal, J. H.; Manji, H. K. Targeting
the glutamatergic system to develop novel, improved therapeutics for
mood disorders. Nat. Rev. Drug Discovery 2008, 7, 426−437.
(18) Adler, L. A.; Kroon, R. A.; Stein, M.; Shahid, M.; Tarazi, F. I.;
Szegedi, A.; Schipper, J.; Cazorla, P. A translational approach to
evaluate the efficacy and safety of the novel AMPA receptor positive
allosteric modulator Org 26576 in adult attention-deficit/hyperactivity
disorder. Biol. Psychiatry 2012, 72, 971−977.
(19) Siegel, G. J.; Chauhan, N. B. Neurotrophic factors in Alzheimer’s
and Parkinson’s disease brain. Brain Res. Rev. 2000, 33, 199−227.
(20) Lauterborn, J. C.; Lynch, G.; Vanderklish, P.; Arai, A.; Gall, C.
M. Positive modulation of AMPA receptors increases neurotrophin
expression by hippocampal and cortical neurons. J. Neurosci. 2000, 20,
8−21.
(21) Lauterborn, J. C.; Truong, G. S.; Baudry, M.; Bi, X.; Lynch, G.;
Gall, C. M. Chronic elevation of brain-derived neurotrophic factor by
ampakines. J. Pharmacol. Exp. Ther. 2003, 307, 297−305.
(22) Lauterborn, J. C.; Pineda, E.; Chen, L. Y.; Ramirez, E. A.; Lynch,
G.; Gall, C. M. Ampakines cause sustained increases in brain-derived
neurotrophic factor signaling at excitatory synapses without changes in
AMPA receptor subunit expression. Neuroscience 2009, 159, 283−295.
(34) Krintel, C.; Frydenvang, K.; Olsen, L.; Kristensen, M. T.; de
Barrios, O.; Naur, P.; Francotte, P.; Pirotte, B.; Gajhede, M.; Kastrup,
J. S. Thermodynamics and structural analysis of positive allosteric
modulation of the ionotropic glutamate receptor GluA2. Biochem. J.
2012, 441, 173−178.
(35) Ptak, C. P.; Ahmed, A. H.; Oswald, R. E. Probing the allosteric
modulator binding site of GluR2 with thiazide derivatives. Biochemistry
2009, 48, 8594−8602.
(36) Weeks, A. M.; Harms, J. E.; Partin, K. M.; Benveniste, M.
Functional insight into development of positive allosteric modulators
of AMPA receptors. Neuropharmacology 2014, 85, 57−66.
N
dx.doi.org/10.1021/jm501268r | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(37) Greger, I. H.; Akamine, P.; Khatri, L.; Ziff, E. B.
Developmentally regulated, combinatorial RNA processing modulates
AMPA receptor biogenesis. Neuron 2006, 51, 85−97.
(38) Meerwein, H.; Dittmar, G.; Gollner, R.; Hafner, K.; Mensch, F.;
Steinfort, O. Aromatic diazo compounds. II. Preparation of aromatic
sulfonyl chlorides, a new modification of the Sandmeyer reaction.
Chem. Ber. 1957, 90, 841−852.
(56) Hayward, S.; Lee, R. A. Improvements in the analysis of domain
motions in proteins from conformational change: DynDom version
1.50. J. Mol. Graph. Model. 2002, 21, 181−183.
(57) Morisawa, Y.; Kataoka, M.; Nagahori, H.; Sakamoto, T.; Kitano,
N.; Kusano, K.; Sato, K. Studies on anticoccidial agents. 13. Synthesis
and anticoccidial activity of nitropyridine-2- and -3-sulfonamides and
derivatives. J. Med. Chem. 1980, 23, 1376−1380.
(58) Pirotte, B.; de Tullio, P.; Lebrun, P.; Antoine, M.-H.; Fontaine,
J.; Masereel, B.; Schynts, M.; Dupont, L.; Herchuelz, A.; Delarge, J. 3-
(Alkylamino)-4H-pyrido[4,3-e]-1,2,4-thiadiazine 1,1-dioxides as
powerful inhibitors of insulin release from rat pancreatic B-cells: A
new class of potassium channel openers? J. Med. Chem. 1993, 36,
3211−3213.
(39) Cowen, M. S.; Beart, P. M. Cyclothiazide and AMPA receptor
desensitization: Analyses from studies of AMPA-induced release of
[3H]-noradrenaline from hippocampal slices. Br. J. Pharmacol. 1998,
123, 473−480.
(40) Lockhart, B.; Iop, F.; Closier, M.; Lestage, P. (S)-2,3-Dihydro-
[3,4]cyclopentano-1,2,4-benzothiadiazine-1,1-dioxide: (S18986-1), a
positive modulator of AMPA receptors enhances (S)-AMPA-mediated
[3H]noradrenaline release from rat hippocampal and frontal cortex
slices. Eur. J. Pharmacol. 2000, 401, 145−153.
(59) Albert, A.; Serjeant, E. P. The Determination of Ionization
Constants; Chapman & Hall: London, 1971; pp 44−64.
(60) Laskowski, R. A.; Macarthur, M. W.; Moss, D. S.; Thornton, J.
M. PROCHECKA program to check the stereochemical quality of
protein structures. J. Appl. Crystallogr. 1993, 26, 283−291.
́
(41) Roux, S.; Sable, E.; Porsolt, R. D. Primary observation (Irwin)
test in rodents for assessing acute toxicity of a test agent and its effects
on behavior and physiological function. Curr. Protoc. Pharmacol. 2005,
Chapter 10, Unit 10.10.
(42) Congreve, M.; Carr, R.; Murray, C.; Jhoti, H. A ‘rule of three’ for
fragment-based lead discovery? Drug Discovery Today 2003, 8, 876−
877.
(43) Qi, J.; Wang, Y.; Jiang, M.; Warren, P.; Chen, G. Cyclothiazide
induces robust epileptiform activity in rat hippocampal neurons both
in vitro and in vivo. J. Physiol. 2006, 571, 605−618.
(44) Bliss, T. V.; Collingridge, G. L. A synaptic model of memory:
Long-term potentiation in the hippocampus. Nature 1993, 361, 31−
39.
(45) Whitlock, J. R.; Heynen, A. J.; Shuler, M. G.; Bear, M. F.
Learning induces long-term potentiation in the hippocampus. Science
2006, 313, 1093−1097.
(46) Ennaceur, A.; Delacour, J. A new one-trial test for neuro-
biological studies of memory in rats. 1: Behavioral data. Behav. Brain
Res. 1988, 31, 47−59.
(47) Kabsch, W. XDS. Acta Crystallogr., Sect. D: Biol. Crystallogr.
2010, 66, 125−132.
(48) Winn, M. D.; Ballard, C. C.; Cowtan, K. D.; Dodson, E. J.;
Emsley, P.; Evans, P. R.; Keegan, R. M.; Krissinel, E. B.; Leslie, A. G.
W.; McCoy, A.; McNicholas, S. J.; Murshudov, G. N.; Pannu, N. S.;
Potterton, E. A.; Powell, H. R.; Read, R. J.; Vagin, A.; Wilson, K. S.
Overview of the CCP4 suite and current developments. Acta
Crystallogr., Sect. D: Biol. Crystallogr. 2011, 67, 235−242.
(49) McCoy, A. J.; Grosse-Kunstleve, R. W.; Adams, P. D.; Winn, M.
D.; Storoni, L. C.; Read, R. J. Phaser crystallographic software. J. Appl.
Crystallogr. 2007, 40, 658−674.
(50) Sun, Y.; Olson, R.; Horning, M.; Armstrong, N.; Mayer, M.;
Gouaux, E. Mechanism of glutamate receptor desensitization. Nature
2002, 417, 245−253.
(51) Langer, G.; Cohen, S. X.; Lamzin, V. S.; Perrakis, A. Automated
macromolecular model building for X-ray crystallography using ARP/
wARP version 7. Nat. Protoc. 2008, 3, 1171−1179.
(52) Emsley, P.; Lohkamp, B.; Scott, W. G.; Cowtan, K. Features and
development of Coot. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2010,
66, 486−501.
(53) Schuttelkopf, A. W.; van Aalten, D. M. F. PRODRG: A tool for
high-throughput crystallography of protein-ligand complexes. Acta
Crystallogr., Sect. D: Biol. Crystallogr. 2004, 60, 1355−1363.
(54) Moriarty, N. W.; Grosse-Kunstleve, R. W.; Adams, P. D.
electronic Ligand Builder and Optimization Workbench (eLBOW): A
tool for ligand coordinate and restraint generation. Acta Crystallogr.,
Sect. D: Biol. Crystallogr. 2009, 65, 1074−1080.
́
(55) Adams, P. D.; Afonine, P. V.; Bunkoczi, G.; Chen, V. B.; Davis,
I. W.; Echols, N.; Headd, J. J.; Hung, L.-W.; Kapral, G. J.; Grosse-
Kunstleve, R. W.; McCoy, A. J.; Moriarty, N. W.; Oeffner, R.; Read, R.
J.; Richardson, D. C.; Richardson, J. S.; Terwilliger, T. C.; Zwart, P. H.
PHENIX: A comprehensive Python-based system for macromolecular
structure solution. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2010, 66,
213−221.
O
dx.doi.org/10.1021/jm501268r | J. Med. Chem. XXXX, XXX, XXX−XXX