Design, stereoselective synthesis, configurational stability and biological activity of 7-chloro-9-(furan-3-yl)-2,3,3a,4-tetrahydro-1H-benzo[e]pyrrolo[2,1- c][1,2,4]thiadiazine 5,5-dioxide

Article abstract of DOI:10.1016/j.bmc.2014.07.017

Chiral 5-arylbenzothiadiazine derivatives have recently attracted particular attention because they exhibit an interesting pharmacological activity as AMPA receptor (AMPAr) positive modulators. However, investigations on their configurational stability suggest a rapid enantiomerization in physiological conditions. In order to enhance configurational stability, preserving AMPAr activity, we have designed the novel compound (R,S)-7-chloro-9-(furan-3-yl)-2,3,3a,4-tetrahydro-1H-benzo[e]pyrrolo[2,1-c][1,2, 4]thiadiazine 5,5-dioxide bearing a pyrrolo moiety coupled with the 5-(furan-3-yl) substituent on benzothiadiazine core. A stereoselective synthesis was projected to obtain single enantiomer of the latter compound. Absolute configuration was assigned by X-ray crystal structure. Patch clamp experiments evaluating the activity of single enantiomers as AMPAr positive allosteric modulator showed that R stereoisomer is the active component. Molecular modeling studies were performed to explain biological results. An on-column stopped-flow bidimensional recycling HPLC procedure was applied to obtain on a large scale the active enantiomer with enantiomeric enrichment starting from the racemic mixture of the compound.

Full text of DOI:10.1016/j.bmc.2014.07.017

Bioorganic & Medicinal Chemistry 22 (2014) 4667–4676
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Design, stereoselective synthesis, configurational stability and
biological activity of 7-chloro-9-(furan-3-yl)-2,3,3a,4-tetrahydro-1H-
benzo[e]pyrrolo[2,1-c][1,2,4]thiadiazine 5,5-dioxide
b
Marina Maria Carrozzo a, Umberto Maria Battisti b, Giuseppe Cannazza b, , Giulia Puia ,
⇑
Federica Ravazzini b, Aurelia Falchicchio c, Serena Perrone a, Cinzia Citti a, Krzysztof Jozwiak d,
Daniela Braghiroli b, Carlo Parenti b, Luigino Troisi a
a Department of Biological and Environmental Sciences and Technologies, University of Salento, via Prov.le Lecce-Monteroni, 73100 Lecce, Italy
b Department of Life Sciences, University of Modena & Reggio Emilia, via Campi 287, 41125 Modena, Italy
c Institute of Crystallography, CNR, Via Giovanni Amendola, 122/O, 70126 Bari, Italy
d Laboratory of Medicinal Chemistry and Neuroengineering, Department of Chemistry, Medical University of Lublin, ul. W. Chodzki 4a, 20-093 Lublin, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
Chiral 5-arylbenzothiadiazine derivatives have recently attracted particular attention because they exhi-
bit an interesting pharmacological activity as AMPA receptor (AMPAr) positive modulators. However,
investigations on their configurational stability suggest a rapid enantiomerization in physiological condi-
tions. In order to enhance configurational stability, preserving AMPAr activity, we have designed the
novel compound (R,S)-7-chloro-9-(furan-3-yl)-2,3,3a,4-tetrahydro-1H-benzo[e]pyrrolo[2,1-c][1,2,4]
thiadiazine 5,5-dioxide bearing a pyrrolo moiety coupled with the 5-(furan-3-yl) substituent on benzo-
thiadiazine core. A stereoselective synthesis was projected to obtain single enantiomer of the latter com-
pound. Absolute configuration was assigned by X-ray crystal structure. Patch clamp experiments
evaluating the activity of single enantiomers as AMPAr positive allosteric modulator showed that R
stereoisomer is the active component. Molecular modeling studies were performed to explain biological
results. An on-column stopped-flow bidimensional recycling HPLC procedure was applied to obtain on a
large scale the active enantiomer with enantiomeric enrichment starting from the racemic mixture of the
compound.
Received 16 June 2014
Revised 10 July 2014
Accepted 11 July 2014
Available online 18 July 2014
Keywords:
AMPAr
Chiral benzothiadiazines
Enantiomerization
Stopped-flow chromatography
Cognitive enhancers
Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction
excitotoxicity. In this context, there is a rapid growing interest in
positive allosteric modulators of AMPA receptors as potentiators of
L
-Glutamate is the major excitatory neurotransmitter in the
glutamatergic function.1–6 Compared to direct AMPA receptor ago-
nists, allosteric modulators are anticipated to possess finer tuning
to increase glutamatergic function since they have no effects in the
absence of the natural ligand in the synapse. These agents, binding
to an allosteric site, enhance receptor function by decreasing desen-
sitization and/or deactivation. Several chemical classes of positive
modulators of AMPA receptors have been described in the last years
as reported in many review articles.5,7–13 Among the different clas-
ses of positive allosteric modulators of AMPA receptors, benzothi-
adiazines represent one of the most investigated since they proved
to be beneficial in the management of neurological disorder such
as schizophrenia, depression, Alzheimer’s disease, Parkinson’s dis-
ease, attention-
mammalian central nervous system (CNS). Two distinct families of
-Glutamate receptor, widely distributed in the CNS, have been pre-
sented: the ionotropic (ligand-gated ion channel, iGluRs) and
metabotropic (G-protein coupled) receptors.1–4 On the basis of their
sensitivity to selective agonists, iGluRs have been further classified
L
into three subclasses: kainic acid (KA), N-methyl-
(NMDA) and the
D
-aspartic acid
a-amino-3-hydroxy-5-methyl-4-isoxazoleprop-
ionic acid (AMPA) receptors.5 A dysfunction of glutamatergic neuro-
transmission has been implicated in a number of neurological and
psychiatric diseases.6 Many studies have been carried out to develop
compounds able to enhance glutamatergic function without causing
deficit/hyperactivity disorder (ADHD) and respiratory depres-
sion.14–21 Among benzothiadiazine derivatives, IDRA21, a com-
pound resulting from saturation of the nitrogenAcarbon double
⇑
Corresponding author. Tel.: +39 059 2055136; fax: +39 059 2055750.
E-mail address: giuseppe.cannazza@unimore.it (G. Cannazza).
http://dx.doi.org/10.1016/j.bmc.2014.07.017
0968-0896/Ó 2014 Elsevier Ltd. All rights reserved.

4668
M. M. Carrozzo et al. / Bioorg. Med. Chem. 22 (2014) 4667–4676
bond of the thiadiazide ring of the antihypertensive diazoxide, was
demonstrated as a promising clinical candidate.
2. Results and discussion
IDRA21 has attracted particular clinical interest despite its
modest in vitro activity since it is effective in increasing learning
and memory performances in behavioral tests, representing an
important lead compound able to cross the blood–brain barrier
(BBB).22,23 Recent crystallographic studies regarding the binding
mode of benzothiadiazine derivatives on S1S2 GluA2 subunits of
AMPAr suggested that these compounds bind to the dimer
interface, leading to inhibition of the receptor desensitization by
stabilization of the ligand-binding domain within a dimer inter-
face.24,25 In particular detailed analysis of the crystal structure of
IDRA21 in complex with S1S2 GluA2 subunits reveal the presence
of an unfilled hydrophilic pocket that could accommodate an aryl
or heteroaryl substituent at C5 atom of IDRA21 core.26 Molecular
modeling studies have suggested that heteroaromatic substituents,
like thiophenyl or furanyl, and aromatic substituents, on C5 posi-
tion of IDRA21 can fit into the unfilled hydrophilic pocket increas-
ing the affinity of IDRA21derivatives towards AMPA receptor.
Therefore a series of 5-arylbenzothiadiazine derivatives were
recently designed and synthesized by our research group.26 Among
them, 7-chloro-5-(3-furanyl)-3-methyl-3,4-dihydro-2H-1,2,
4-benzothiadiazine 1,1-dioxide (1, Fig. 1) has attracted particular
attention because it exhibited an interesting pharmacological
activity as AMPAr positive allosteric modulator.26 Since we have
recently reported a rapid enantiomerization in aqueous solution
and a rapid hydrolysis in acid medium of IDRA21 and related
compounds, a primary goal of the present work is to develop a
compound with high configurational and chemical stability under
conditions similar to those that they will meet in vivo.27–35 One of
the most relevant IDRA21 derivatives, 2,3,3a,4-tetrahydro-1H-
pyrrolo[2,1-c][1,2,4]benzothiadiazine 5,5-dioxide (2, Fig. 1)
reached clinical trials due to its ability to act as cognitive
enhancing agent in normal young and aged rodents.36–38
2.1. Design and synthesis of ( )-7-chloro-9-(furan-3-yl)-
2,3,3a,4-tetrahydro-1H-benzo[e]pyrrolo[2,1-c][1,2,4]thiadiazine
5,5-dioxide (3)
Recently (S)-2 (S18986) has attracted particular attention since
in vivo experiments demonstrated its cognition enhancing proper-
ties when it is administered orally at low dose.36–38 Subsequent ste-
reo and chemical evaluation studies have suggested that (S)-2
(S18986) is stable to enantiomerization and hydrolysis.33 Thus in
order to enhance stability towards enantiomerization and hydroly-
sis, preserving AMPAr affinity, 7-chloro-9-(furan-3-yl)-2,3,3a,4-tet-
rahydro-1H-benzo[e]pyrrolo[2,1-c][1,2,4]thiadiazine 5,5-dioxide
(3) was designed and synthesized combining several chemical
features of compounds 1 and 2. Hence, the designed 3 was prepared
following the synthetic route reported in Scheme 1. The intermedi-
ate ( )(R,S)-7-chloro-2,3,3a,4-tetrahydro-1H-benzo[e]pyrrolo[2,1-
c][1,2,4]thiadiazine 5,5-dioxide (7) was prepared as previously
described by Cameroni et al. with minor modifications.39 Subse-
quently, 7 was halogenated in acidic conditions to give 8, that was
coupled with 3-furanyl boronic acid by Suzuki Pd-catalyzed reaction
giving the desired compound 3 (Scheme 1). The latter was fully char-
acterized by IR, GC–MS and NMR spectroscopy analysis.
2.2. Configurational and chemical stability of ( )-7-chloro-9-
(furan-3-yl)-2,3,3a,4-tetrahydro-1H-benzo[e]pyrrolo[2,1-
c][1,2,4]thiadiazine 5,5-dioxide (3)
Recent studies conducted by dynamic chromatography on com-
pound ( )-1 underlined its configurational and chemical lability.35
Indeed during the chiral separation on a Chiralcel OD-R column at
temperatures between 25 and 45 °C employing aqueous mobile
phases (water/acetonitrile) at different pH a pronounced plateau
was observed between the peaks corresponding to the two enanti-
omers, indicating a rapid enantiomerization of ( )-1.35 Differently
the same experiments applied to ( )-3 evidenced its stereo stabil-
ity during chromatographic separation in aqueous mobile phases.
No enantiomerization and hydrolysis occurred when ( )-3 was
chromatographed with similar conditions (see Fig. 2).
Further studies have demonstrated that all activity resided in the
S enantiomer of 2; moreover, it showed high chemical and configu-
rational stability.33 Starting from this interestingchemical and phar-
macological profile, we have designed a new compound bearing the
pyrrolo moiety of compound 2 coupled with 5-aryl substituent on
benzothiadiazine core following the approach of combining in a
single molecule two different pharmacophores. 7-Chloro-9-(furan-
3-yl)-2,3,3a,4-tetrahydro-1H-benzo[e]pyrrolo[2,1-c][1,2,4]thiadiazine
5,5-dioxide (3) was designed and synthesized in order to enhance
configurational and chemical stability preserving the relevant
biological activity (Fig. 1).
In order to evaluate stereo and chemical stability of ( )-3 in
absence of chiral stationary phase a recently developed method
was employed.32
Stopped-flow bidimensional recycling HPLC (sf-BD-rHPLC)
method was applied employing immobilized artificial membrane
(IAM) stationary phase as reactor column. No enantiomerization
and/or hydrolysis product peaks appeared in the sf-BD-rHPLC
chromatograms performed in IAM column at both pH 2.2 and 7.4
after 90 min at 37.5 °C. A moderate racemization occurs only at
pH 1 at 45 °C for 90 min suggesting a great increase of the
configurational stability of ( )-3 with respect to compound ( )-1.
Moreover the hydrolysis product of ( )-3 was not observed sug-
gesting a complete chemical stability of the latter compound. A
similar trend was previously observed for compound 2 confirming
the stabilizing effect of the pyrrolo moiety.33
O
O
Cl
S
NH
interesting pharmacological activity
N
H
CH3
O
O
Cl
S
6
9
5
4NH
7
8
3a
O
1
1a
N
3
1
2
2.3. Stereoselective synthesis of (S)-7-chloro-9-(furan-3-yl)-2,3,
3a,4-tetrahydro-1H-benzo[e]pyrrolo[2,1-c][1,2,4]thiadiazine 5,5
-dioxide and (R)-7-chloro-9-(furan-3-yl)-2,3,3a,4-tetrahydro-1H
-benzo[e]pyrrolo[2,1-c][1,2,4]thiadiazine 5,5-dioxide
O
O
O
S
N
3
NH
configurational and chemical stability
Since previous studies suggested that chiral benzothiadiazines
display a stereoselective pharmacological action, it was important
to evaluate the activity of single stereoisomers of ( )-3 as AMPAr
modulators.40,41
2
Figure 1. Design of 7-chloro-9-(furan-3-yl)-2,3,3a,4-tetrahydro-1H-benzo[e]pyr-
rolo[2,1-c][1,2,4]thiadiazine 5,5-dioxide (3).

M. M. Carrozzo et al. / Bioorg. Med. Chem. 22 (2014) 4667–4676
4669
O
O
O
O
O
O
Cl
S
N
NH2
NH2
Cl
S
N
N
Cl
S
Cl
S
NH
(iii)
(ii)
(i)
O
O
NH2
NHCO(CH2)3Cl
7
4
5
6
v)
(i
O
O
O
O
Cl
S
N
Cl
S
N
NH
NH
)
(v
Br
3
8
O
Scheme 1. Synthesis of 3. Reagents and conditions: (i) c-chlorobutyryl chloride (2 equiv), N,N-dimethylacetamide, 0 °C; (ii) NaOH (2%), 100 °C, 15 min; (iii) LiAlH4 (2 equiv),
diethylether, À10 °C; (iv) N-bromosuccinimide (6 equiv), acetic acid, acetonitrile, room temperature; (v) Na2CO3 (3 equiv), 3-furanyl boronic acid (1.2 equiv), tetrakis
palladium (0.5 equiv), H2O/dioxane, 110 °C.
The stereoselective reduction of 9 to obtain (S)-3 was performed
in presence of aminocarbinol-(+)-(2S,3R)-4-dimethylamino-3-
methyl-1,2-diphenyl-2-butanol (ChiraldÒ) as chiral selector and
LiAlH4 as reductive agent (Scheme 2).
A similar protocol was employed by Desos et al. to obtain
S18986 with high optical purity.41 The enantiomeric excess
achieved in this reaction was about 70% as calculated by chiral
HPLC. Since enantiomeric excess obtained by asymmetric synthesis
was insufficient to conduct biological experiments, a semiprepara-
tive chromatographic method was developed in order to purify the
enantiomer of 3. The enriched enantiomeric mixture was
chromatographed on semipreparative Chiralcel OD column with
hexane/2-propanol 95:5 (v/v) as mobile phase. The collected frac-
tions containing the single enantiomer show high enantiomeric
excess values (ee > 99%). Enantiomeric excess of (+)-3 was calcu-
lated employing an OD-R column with water/acetonitrile 50:50
(v/v) as mobile phase. Subsequently, optical rotation values for
the enantiomer obtained was calculated and the specific rotation
in chloroform was [a] +32.4° (26 mg/mL; chloroform, 24 °C). In
Figure 2. On column racemization of ( )-3. Condition: column Chiralcel OD-RH;
mobile phase water/acetonitrile (50:50 v/v), T = 45 °C.
D
order to assign the absolute configuration, single crystal X-ray
diffraction analysis were performed on the isolated dextrorotatory
enantiomer (Fig. 3). The data analysis indicated that the
enantiomer obtained employing ChiraldÒ as chiral auxiliary had S
configuration. Thus the reaction proceeds with the same
stereochemical outcome observed for S18986.41
Moreover, the configurational stability observed for ( )-3
prompted us to develop an asymmetric synthesis. The synthetic
pathway employed is shown in Scheme 2. Once obtained racemic
3, it was oxidized in presence of potassium permanganate in basic
conditions to give 9, which was subjected to asymmetric reduction.
In order to obtain (R)-3 the procedure described by Cohen et al.
and Deeter et al., was applied to 9.42–44
O
O
Cl
S
N
NH
(S)
(ii)
O
O
O
O
Cl
S
N
Cl
S
N
NH
N
O
(i)
3
(S)-
O
O
Cl
S
N
3
9
O
NH
(R)
O
(iii)
O
3
(R)-
Scheme 2. Reagents and conditions: (i) KMnO4, KOH 2%; (ii) Chirald (3 equiv), LiAlH4 (1.5 equiv), diethylether, 0 °C; (iii) (2R,3S)-(À)-4-dimethylamino-1,2-diphenyl-3-
methyl-2-butanol (3 equiv), LiAlH4 (1.5 equiv), toluene 0 °C.

4670
M. M. Carrozzo et al. / Bioorg. Med. Chem. 22 (2014) 4667–4676
They demonstrated that by choosing ChiraldÒ or its enantiomer
(2R,3S)-(À)-4-dimethylamino-1,2-diphenyl-3-methyl-2-butanol in
complex with LiAlH4 for the reduction of acetylenic ketones it was
possible to obtain selectively alcohols with opposite configuration
with good enantioselectivity.43 This strategy applied to compound
9 furnished the desired (R)-3 with good enantiomeric excess (73%);
further purification on semipreparative Chiralcel OD column gave
enantiopure (R)-3 (ee > 99%) with [
form, 24 °C).
a]
À33.8° (34 mg/mL; chloro-
D
2.4. Biological activity
The activity of compound 3 and its enantiomers as allosteric
modulators of AMPA/kainate receptors was evaluated by patch-
clamp technique in primary cultures of cerebellar neurons. Kainate
(KA)-evoked current was mainly mediated by AMPAr activation
because application of GYKI 53655 (100 lM), an AMPA receptors
antagonist, almost completely abolished the current (data not
shown). Since IDRA 21 is one of the most relevant benzothiadiazine
derivatives reported in literature as AMPAr-positive allosteric
modulator, it was selected as reference compound. Data obtained
indicate that compound 3 and IDRA21 potentiate by 30 9%
Figure 3. Molecular structure of (+)-(S)-3 at 293 K. Anisotropic displacements are
depicted at the 50% probability level.
(n = 4) and 8 11% (n = 4) KA-evoked currents at 10 lM (Fig. 4).
In order to highlight possible stereospecific interactions of 3 with
AMPA/kainite receptors, (+)-(R)-3 and (À)-(S)-3 were also tested.
Figure 4. (A) Representative whole cell recording showing KA (100
application of (R)-3 (10 M) (middle trace) and after washing (right trace). (B) Histogram showing the potentiation of (R)-3, (S)-3, 3 and IDRA 21 of 5-Will-evoked current and
of KA-evoked current. Each data point is the mean SEM of at least five cells. Significative differences among (R)-3 and (S)-3 enantiomers were obtained (T-test, p < 0.05).
lM)-evoked current recorded from a cerebellar neuron in culture in control conditions (left traces), after
l

M. M. Carrozzo et al. / Bioorg. Med. Chem. 22 (2014) 4667–4676
4671
Figure 5. (a) (S)-3 (magenta) and (R)-3 (yellow) in GluA2 dimer interface. (b) Binding mode of (S)-3 (magenta) and (R)-3 (yellow). (c) Position of the pyrrole ring of (R)-3
(yellow) in the binding pocket. (d) Position of the pyrrole ring of (S)-3 (magenta) in the binding pocket. The hydrophobic regions are coloured in blue, the hydrophilic regions
in red.
Figure 6. Principal polar interactions of (R)-3 (yellow, panel a and b) and (S)-3 (magenta, panel c and d). Blue dashes indicate H-bond interactions. Red dashes indicate
absence of H-bond interactions.

4672
M. M. Carrozzo et al. / Bioorg. Med. Chem. 22 (2014) 4667–4676
Figure 7. ‘Deracemization’ of ( )-3 to obtain (À)-(R)-3. Conditions: a) Chiralcel OD-R, hexane:2-propanol 95:5 (v/v), 0 °C; b) IAM, IPA (1%TFA), 45 °C.
Data obtained indicate a marked difference in activity: (+)-(R)-3
enhances the current by 74 27% while (À)-(S)-3 by 12 7%. These
results indicate that (+)-(R)-3 is more active than (À)-(S)-3 while
Desos et al. found, for a structurally similar compound (S18986),
that the active component was the S enantiomer.41 Probably these
two allosteric modulators have different orientations in the ligand
binding domain that promote different interactions with the recep-
tor. Subsequently compound 3 and its enantiomers were tested as
modulators of the current evoked by (À)-(S)-5-Fluorowillardiine
(5-Will), a potent and selective AMPAr agonist in order to evaluate
and confirm their AMPAr selectivity.45
KA-evoked currents was confirmed since (À)-(R)-3 was more
active than (+)-(S)-3. The extent of potentiation of the KA-evoked
currents for 3 and its enantiomers was lower than that obtained
on 5-Will-evoked current, suggesting a selectivity towards GluA1
and GluA2 AMPAr subunits.
2.5. Docking studies
Several and various binding modes have been observed for ben-
zothiadiazine type compounds active as AMPA positive allosteric
modulators.25,24 Thus in order to understand the possible stereo-
specific interactions of (À)-(R)-3 and (+)-(S)-3 docking studies
were performed. Taking advantage of crystal structures of the
AMPA GluA2 ligand binding domain co-crystallized with several
benzothiadiazines, Molegro Virtual Docker (MVD) software was
applied to dock (À)-(R)-3 and (+)-(S)-3 within the binding pocket
The racemate potentiate 5-Will evoked current of 170 56%
(mean SE) (n = 7) while (À)-(R)-3 and (+)-(S)-3 enhance the
current of 245 55% (n = 5) and 122 30% (n = 7), respectively.
IDRA 21 potentiation of the current was 10 21% (n = 4) at
10 lM. The stereoselective activity previously observed on

M. M. Carrozzo et al. / Bioorg. Med. Chem. 22 (2014) 4667–4676
4673
of the GluA2 dimer interface. The software MVD was previously
evaluated on several crystal structure of benzothiadiazine.26,34
The average root mean square distance (RMSD) of the best ranked
pose of tested compounds compared to their binding pose in their
respective crystal structures was found to be less than 1.0 Å prov-
ing that MVD is able to accurately dock this class of compounds.
The crystal structure obtained for (+)-(S)-3 was employed as input
file to build the ligands with Spartan Wavefunction 08. Among the
tested crystal structures, the GluA2 dimer in complex with Cyclo-
thiazide and IDRA21 (PDB codes: 1LBC and 3IL1) were selected
for docking studies of both enantiomers 3.25 The docking protocol
template was built using the following chemical properties of
Cyclothiazide and IDRA21: ring atoms, hydrogen-bond acceptors,
hydrogen-bond donors and steric interactions. Minimization was
performed after docking in order to increase the precision of the
analysis.
The docking output clearly demonstrated that (À)-(R)-3 and
(+)-(S)-3 adopt a binding mode close to that of IDRA21 (Fig. 5). This
result is particularly significant since it was obtained for both the
crystal structures employed.
The primary polar interaction partners for both the enantiomers
of 3 are Pro494 and Gly731 (Fig. 6). The carbonyl oxygen atom of
Pro494 forms a polar interaction with the N4 atom of (À)-(R)-3
and (+)-(S)-3, whereas the amidic backbone of Gly731 makes an
hydrogen bond with the oxygen of the sulphone moiety since it’s
located for both the stereoisomer within 3 Å.
The stopped-flow bidimensional recycling HPLC method con-
sists in a 3-step experimental protocol. In step 1, the racemate
was injected and quantitatively separated on chiral column. Subse-
quently, one of the two enantiomers was trapped into a reactor
achiral column, which was filled with appropriate racemization
solvent. Afterwards, the mobile phase flow was reinforced into
the achiral column and the two enantiomers were separated in the
chiral column. The cycle can be repeated by trapping one of the
two eluted enantiomers in the achiral column (Fig. 7).
( )-3 was injected on a semipreparative Chiralcel OD with hex-
ane/2-propanol 95:5 (v/v) as mobile phase. The first eluted stereo-
isomer (À)-(R)-3 of the racemic mixture injected was initially
collected. The second eluted enantiomer, the undesired (+)-(S)-3,
was trapped in the achiral column, that was subsequently filled
with 2-propanol added of 1% (v/v) of trifluoroacetic acid as catalyst.
The trapped enantiomer was left to racemize for 60 min at 45 °C.
Afterwards the original mobile phase (hexane/2-propanol 95:5
(v/v)) was introduced in the achiral column forcing the original
enantiomer (+)-(S)-3 and its interconversion product (À)-(R)-3 to
the semipreparative Chiralcel OD where they was enantioseparat-
ed. Thus by harvesting again the selected stereoisomer it is possi-
ble to collect additional 25% of (À)-(R)-3. The racemization/
separation cycle could be repeated several times in order to
increase the yield of (À)-(R)-3. 200 mg of racemic compound by
consecutive injections afforded approximately 170 mg of the
desired enantiomer.
However, in contrast to what observed in IDRA21 crystal struc-
ture, the N1a hydrogen atom in compound 3 has been substituted
with a pyrrole group; therefore, the hydrogen bond to Ser754 is not
possible. The lack of this interaction could explain the moderate
decrease of activity showed for both the enantiomer of 3 with
respect to 1. The main difference observed for (À)-(R)-3 and (+)-
(S)-3 is the position of the pyrrole. The S configuration of the C3a
atom imposes a spatial orientation on the five membered ring such
that favored steric interactions with hydrophobic cleft defined by
Val750, Leu751, and Leu259 residues (Fig. 5). As described by Ptak
et al. the substituent at C3 and thus the hydrophobic interactions
played an important role in the orientation and in the position of
the benzothiadiazine rings. In compound (+)-(S)-3 the pyrrole moi-
ety is buried deep inside in the lipophilic cavity previously
described preventing the formation of the hydrogen bond between
the heteroatom of 3-furanyl fragment and the amidic backbone of
Ser497 (Figs. 5 and 6). This interaction, as recently reported, is cru-
cial in conferring high activity as AMPA positive allosteric modula-
tor.26,44 On the other hand, (À)-(R)-3 poorly interact with the
hydrophobic region of the ligand binding domain since the pyrrole
is located diametrically opposite to Val750, Leu751, and Leu259
residues (Fig. 5). Therefore the furan moiety could be accommo-
dated within the hydrophilic pocket lined by Lys763, Tyr424,
Ser729, Phe495 and Ser497 residues interacting via hydrogen bond
with the latter amino acid.
This method will assist us to gain access to relevant data on the
pharmacokinetic profile of (À)-(R)-3.
3. Conclusions
Recently, we designed and synthesized compound 1 that has
attracted particular attention because it exhibited an interesting
pharmacological activity as AMPA receptor positive modulator.
Preliminary configurational stability studies suggested a rapid
enantiomerization in condition similar to those it will meet in vivo.
In order to enhance stability towards enantiomerization and
hydrolysis, preserving AMPAr affinity, 7-chloro-9-(furan-3-yl)-
2,3,3a,4-tetrahydro-1H-benzo[e]pyrrolo[2,1-c][1,2,4]thiadiazine
5,5-dioxide ( )-3 was synthesized. Studies on ( )-3 confirmed a
great increase in the configuration stability and a complete sup-
pression of the hydrolysis in physiological conditions. A stereose-
lective synthesis was developed to obtain the single enantiomers
of ( )-3. The absolute configuration of the enantiomers of 3 was
assigned thanks to X-ray diffraction spectroscopy. Biological stud-
ies suggested a stereospecific interaction of (À)-(R)-3 with AMPA
receptor. Molecular modeling experiments performed on AMPA
GluA2 ligand binding domain identified the probable binding mode
of (À)-(R)-3 and the source of the stereospecific recognition with
the receptor. The ‘on line racemization’ technique was applied to
synthesize the eutomer (À)-(R)-3 on preparative scale. In vivo
pharmacokinetic experiments in rats are in development in order
to confirm the stereo-pharmacological activity of 3.
2.6. Synthesis of (À)-(R)-3 and (+)-(S)-3 by on-line racemization
4. Experimental
Notwithstanding the asymmetric synthesis previously
described proved to be successful, the modest enantiomeric excess
obtained with the reaction, the high costs of the chiral auxiliaries
employed and the moderate yields obtained led us to develop a
new strategy to gain access to enantiopure (À)-(R)-3.
4.1. Instrumentation
The chromatographic apparatus was a Shimadzu LC-10AD
Pump (Shimadzu Italia, Milan), a Merck Hitachi L-6200A Pump
(Merck KGaA, Darmstadt, Germany), a Rheodyne 7725 manual
Our goal was to achieve (À)-(R)-3 on a preparative scale in order
to conduct full and detailed pharmacological in vivo experiments.
The good enantioselectivity achieved during the chromatogra-
phy of ( )-3 on a semipreparative Chiralcel OD column and the
moderate racemization observed at pH 1 suggested the possible
application of a recently developed enantiomeric enrichment
process.46
injector equipped with a 20 ll sample loop (Jasco Europe, Italy,
Milan). A Merck Hitachi L-7400UV (Merck KGaA, Darmstadt,
Germany) was used as detector. Chromatograms were recorded
with a Jasco J-700 program (Jasco Europe, Italy, Milan). Two
Rheodyne 7000 valves were used to switch the mobile phase flow

4674
M. M. Carrozzo et al. / Bioorg. Med. Chem. 22 (2014) 4667–4676
(Jasco Europe, Italy, Milan). Column temperature regulation was
obtained with a Jasco CO-2067 column oven (Jasco Europe, Italy,
Milan).
Mp: 214–216 °C, 1H NMR (400 MHz, CDCl3): d = 2.01–2.15 (m,
3H), 2.38–2.45 (m, 1H), 3.17–3.22 (m, 1H), 4.24–4.31 (m, 1H),
4.86 (d, J = 12.4 Hz, 1H), 5.13–5.18 (m, 1H), 7.55 (d, J = 2.3 Hz,
1H), 7.61 (d, J = 2.3 Hz, 1H); 13C NMR (100 MHz, CDCl3): d = 23.8,
31.1, 52.2, 71.8, 114.8, 123.1, 126.4, 129.1, 137.6, 141.5.
GC–MS 70 eV: m/z 338 (100) [M+2], 273 (60), 257 (30), 246
(58), 193 (56), 166 (58), 109 (35).
The columns used were Chiralcel OD-R [cellulose tris (3,5-dim-
ethylphenylcarbamate); 250 Â 4.6 mm I.D.; 10
[cellulose tris (3,5-dimethylphenylcarbamate); 250 Â 10 mm I.D.;
10 m] purchased from Chiral Technologies Europe, Illkirch, France.
Optical rotation ( ) was measured with the P-2000 Digital
lm], Chiralcel OD
l
a
HRMS-ESI: calcd for C10H11BrClN2O2S: 336.9407 found:
336.9412.
Polarimeter (cell-length 100 mm, volume 1 ml) from Jasco Europe,
Italy, Milan.
Melting points were determined with an Electrothermal
Apparatus and they are uncorrected.
4.2.2.3. 7-Chloro-9-(furan-3-yl)-2,3,3a,4-tetrahydro-1H-benzo
[e]pyrrolo[2,1-c][1,2,4]thiadiazine 5,5-dioxide (3).
To a stir-
IR spectra were recorded on a PerkinElmer Model 1600 FT-IR
spectrometer. 1H NMR spectra were recorded with a Bruker DPX
400 spectrometer with CDCl3 as solvent and tetramethylsilane
(TMS) as external standard. Chemical shifts (d) are in part per mil-
lion and coupling constant (J) in hertz. Multiplicities are abbrevi-
ated as follows: s, singlet; d, doublet; dd, double doublet; t,
triplet; m, multiplet. The electrospray ionization (HR-ESI-MS)
experiments were carried out in a hybrid QqTOF mass spectrometer
(PE SCIEX-QSTAR) equipped with an ion spray ionization source. MS
(+) spectra were acquired by direct infusion (5 mL/min) of a solu-
tion containing the appropriate sample (10 pmol/mL), dissolved
in a solution 0.1% acetic acid, methanol/water 50:50 at the
optimum ion voltage of 4800 V.
ring solution of 9-bromo-7-chloro-2,3,3a,4-tetrahydro-1H-
benzo[e]pyrrolo[2,1-c][1,2,4]thiadiazine 5,5-dioxide (8) (1 mmol)
and 3-furanylboronic acid (1.2 mmol) in water/dioxane (1:1 v/v),
tetrakis(triphenylphosphine)palladium (5% mol) and Na2CO3
(3 mmol) were added. The reaction mixture was heated at 100 °C
for 3 h and then cooled down to room temperature. The mixture
was neutralized with HCl 1 M and extracted with ethyl acetate.
Combined organic layers were dried over anhydrous sodium
sulfate and concentrate under vacuum. Column chromatography
(petroleum ether/ethyl acetate 7:3 (v/v)) provides the pure
compound as a yellow solid. Yield 75%.
Mp: 218–220 °C; 1H NMR (400 MHz, CDCl3): d = 1.88–1.99 (m,
3H), 2.33–2.43 (m, 1H), 2.76–2.82 (m, 1H), 3.28–3.34 (m, 1H),
4.86 (d, J = 12.4 Hz, 1H), 5.21–5.27 (m, 1H), 6.61 (d, J = 2.6 Hz,
1H,), 7.32 (d, J = 2.6 Hz, 1H,), 7.47–7.49 (m, 2H), 7.63 (s, 1H); 13C
NMR (100 MHz, CDCl3): d = 23.4, 32.0, 51.8, 71.8, 110.9, 122.5,
123.6, 125.9, 127.4, 127.9, 134.6, 140.6, 141.2, 143.4.
GC–MS 70 eV: m/z 324 (100) [M+], 295 (25), 259 (23), 205 (30),
113 (20).
All pH measurements were made using Orion Research EA940
pH-meter.
HPLC-grade acetonitrile, n-hexane and 2-propanol were
obtained from Sigma–Aldrich (Milan, Italy).
4.2. Synthesis
HRMS-ESI: calcd for C14H14ClN2O3S: 325.0408; found:
325.0415.
4.2.1. 7-Chloro-5-(2-furanyl)-3-methyl-3,4-dihydro-2H-1,2,4-
benzothiadiazine 1,1-dioxide (1)
Compound 1 was obtained as described by Battisti et al.26 Yield
45% (three steps).
4.2.3. Asymmetric synthesis
4.2.3.1. 7-Chloro-9-(3-furanyl)-2,3-dihydro-1H-benzo[e]pyrrol-
Mp: 195–197 °C; 1H NMR (400 MHz, CDCl3): d = 1.62 (d,
J = 6.1 Hz, 3H), 4.51 (d, J = 13.0 Hz, 1H), 5.10–5.13 (m, 1H), 5.87
(s, 1H, broad), 6.56 (dd, J = 1.8 Hz, 3.3 Hz, 1H), 6.65 (d, J = 3.3 Hz,
1H), 7.50 (d, J = 2.4 Hz, 1H), 7.55 (d, J = 1.3 Hz, 1H), 7.59 (d,
J = 2.4 Hz, 1H).
o[2,1-c][1,2,4]thiadiazine 5,5-dioxide (9).
The compound
was synthesized as described by Cameroni et al. starting from
7-chloro-9-(furan-3-yl)-2,3,3a,4-tetrahydro-1H-benzo[e]pyrrolo[2,
1-c][1,2,4]thiadiazine 5,5-dioxide (3).39 Yield 50%.
Mp: 222–224 °C; 1H NMR (400 MHz, CDCl3): d = 2.05 (dd,
J = 7.8 Hz, 7.9 Hz, 2H,), 2.92 (t, J = 7.9 Hz, 2H), 3.78 (t, J = 7.8 Hz,
2H), 6.49 (d, J = 2.3 Hz, 1H), 7.41 (d, J = 2.3 Hz, 1H), 7.51–7.53 (m,
2H), 7.60 (s, 1H); 13C NMR (100 MHz, CDCl3): d = 18.9, 33.5, 54.2,
113.3, 123.6, 124.4, 127.8, 130.4, 132.6, 134.1, 136.9, 140.3,
141.1, 164.5.
GC–MS (70 eV): m/z 298 (100) [M+], 283 (67), 255 (17), 192
(52), 164 (35), 128 (68).
4.2.2. 7-Chloro-9-(furan-3-yl)-2,3,3a,4-tetrahydro-1H-benzo[e]
pyrrolo[2,1-c][1,2,4]thiadiazine 5,5-dioxide (3)
4.2.2.1.
( )(R,S)-7-Chloro-2,3,3a,4-tetrahydro-1H-pyrrolo[2,1-
The compound
GC–MS 70 eV: m/z 322 (84) [M+], 293 (100), 259 (14), 229 (51),
190 (31).
c][1,2,4]benzothiadiazine 5,5-dioxide (7).
was synthesized as previously described by Cameroni et al. starting
from 2-amino-5-chlorobenzensulfonamide.39 Yield 70% (three
steps).
HRMS-ESI: calcd for C14H11ClN2O3S: 322.0179; found:
322.0178.
Mp: 226–228 °C; 1H NMR (400 MHz, CDCl3): d = 1.88–1.99 (m,
1H), 2.00–2.12 (m, 2H), 2.20–2.24 (m, 1H), 2.51–2.57 (m, 1H), 3.33–
3.40 (m, 1H), 3.52–3.55 (m, 1H), 5.15–5.18 (m, 1H), 6.53 (d,
J = 8.4 Hz, 1H,), 7.29 (dd, J = 3.1 Hz, 8.4 Hz, 1H), 7.61 (d, J = 3.1 Hz, 1H).
GC–MS (70 eV): m/z 258 (90) [M+], 193 (100), 165 (89), 138
(40), 75 (50).
4.2.3.2. (S)-(+)-7-Chloro-9-(furan-3-yl)-2,3,3a,4-tetrahydro-1H-
benzo[e]pyrrolo[2,1-c][1,2,4]thiadiazine
(3). The compound was synthesized following the protocol
reported by Cannazza et al.33,41 Yield 85%.
5,5-dioxide
(S)-
Mp: 218–220 °C; 1H NMR (400 MHz, CDCl3): d = 1.88–1.99 (m,
3H), 2.33–2.43 (m, 1H), 2.76–2.82 (m, 1H), 3.28–3.34 (m, 1H),
4.86 (d, J = 12.4 Hz, 1H), 5.21–5.27 (m, 1H), 6.61 (d, J = 2.6 Hz,
1H), 7.32 (d, J = 2.6 Hz, 1H), 7.47–7.49 (m, 2H), 7.63 (s, 1H).
4.2.2.2. 9-Bromo-7-chloro-2,3,3a,4-tetrahydro-1H-benzo[e]pyr-
rolo[2,1-c][1,2,4]thiadiazine 5,5-dioxide (8).
To a stirring
[
a]
D
+32.4° (15 mg/mL; chloroform, 24 °C).
solution of 7 (1 mmol) in ACN (4 ml), a solution of N-bromosuccin-
imide (1.5 mmol) in AcOH (1.2 ml) was added dropwise. The
reaction was stirred at room temperature until no starting material
was detected by TLC. Subsequently water was added and the
resulting precipitate was recovered by filtration. The filtrate pro-
vides pure compound as a white solid. Yield 75%.
4.2.3.3.
(R)-(À)-7-Chloro-9-(furan-3-yl)-2,3,3a,4-tetrahydro-1H-
benzo[e]pyrrolo[2,1-c][1,2,4]thiadiazine 5,5-dioxide (R)-(3).
A
dry toluene solution, containing the chiral amino alcohol (2R,3S)-(À)-
4-dimethylamino-1,2-diphenyl-3-methyl-2-butanol (3.0 mmol) was
added at 0 °C to a magnetically stirred solution of LiA1H4 (1.5 mmol)

M. M. Carrozzo et al. / Bioorg. Med. Chem. 22 (2014) 4667–4676
4675
in dry toluene. The reaction mixture was vigorously stirred at 0 °C dur-
ing 15 min. Compound 9 (1.0 mmol) was added in several portions and
the mixture was stirred for 2 h at 0 °C. The reaction was hydrolysed by
adding dropwise 1 N HCl (vigorous evolution of gas). Then the reaction
mixture was neutralized with1 N NaOH and extracted with ethyl ace-
tate. The organic extracts were dried over Na2SO4 and evaporated under
vacuum to give reduced product as a white solid (ee: 76%). The com-
pound was further purified by chiral HPLC to increase enantiomeric
excess. Yield 69%.
trypsin (0.24 mg/mL; Sigma Aldrich, Milan, Italy) and plated at a
density of 0.8 Â 106 cells/mL on 35 mm Falcon dishes coated with
poly-L-lysine (10 lg/mL, Sigma Aldrich). Cells were plated in basal
Eagle’s Medium (BME; Celbio, Milan, Italy), supplemented with
10% fetal bovine serum (Celbio), 2 mM glutamine, 25 mM KCl
and 100 lg/mL gentamycin (Sigma Aldrich) and maintained at
37 °C in 5% CO2. After 24 h in vitro, the medium was replaced with
1:1 mixture of BME and Neurobasal medium (Celbio, Milan) con-
taining 2% B27 supplement, 1% antibiotic, and 0.25% glutamine
(Invitrogen). At 5 days in vitro (DIV5), cytosine arabinofuranoside
Mp: 217–219 °C; 1H NMR (400 MHz, CDCl3): d = 1.88–1.99 (m,
3H), 2.33–2.43 (m, 1H), 2.76–2.82 (m, 1H), 3.28–3.34 (m, 1H),
4.86 (d, J = 12.4 Hz, 1H), 5.21–5.27 (m, 1H), 6.61 (d, J = 2.6 Hz,
1H), 7.32 (d, J = 2.6 Hz, 1H), 7.47–7.49 (m, 2H), 7.63 (s, 1H).
(Ara-C) was added at a final concentration of 1 lM.
Recordings were performed at room temperature, under volt-
age-clamp in the whole-cell configuration of the patch-clamp tech-
nique on cells. Electrodes pulled from borosilicate glass
(Heidelberg, FRG) on a vertical puller (PB-7, Narishige) and had a
[a]
À33.8° (15 mg/mL; chloroform, 24 °C).
D
4.3. X-ray crystallography
resistance of 5–6 M X. Currents were amplified with an Axopatch
1D amplifier (Axon Instruments, Foster). The recording chamber
was continuously perfused at 5 mL/min with an artificial extracel-
lular solution composed of (mM): 145 NaCl, 5 KCl, 1 CaCl2, 5 Hepes,
5 Glucose, 20 Sucrose, pH 7.4 with NaOH. Electrode intracellular
solution contained (mM): 140 KCl, 3 MgCl2, 5 EGTA, 5 Hepes, 2
ATP-Na, pH 7.3 with KOH. Drugs were applied directly by gravity
through a Y-tube perfusion system.
X-ray diffraction experiments were carried out at room temper-
ature (293°K) by a Bruker-Nonius KappaCCD single crystal diffrac-
tometer, equipped
a with a charge-coupled device (CCD detector),
using monochromatized MoK (k radiation = 0.71073 Å). The auto-
matic data collection strategy was defined by the COLLECT soft-
ware, cell determination and refinement by DIRAX and data
reduction by EVAL.47,48 Absorption effects were corrected by SAD-
ABS, exploiting a semi-empirical approach. Crystal structure was
solved by SIR2011 and refined by SHELXL-97.49–51
4.6. Docking studies
The GluA2-S1S2J crystal structures in complex with Cyclothia-
zide and IDRA21 were retrieved from the Protein Data Bank (PDB
codes:1LBC and 3IL1) and importedinto MVD.25 All watermolecules
and co-factors were deleted. (S)-3 and (R)-3 were built in Spartan
employing the crystal structure of (S)-3 as input file. Then they were
exported as mol2 files and docked in GluA2 by using MVD. We used
the template docking available in MVD and evaluated MolDock
score, Rerank score, and protein–ligand interaction score from Mol-
Dock and MolDock [GRID] options. Cyclothiazide and IDRA21 were
selected as reference compounds for the template. It was used the
default settings, including a grid resolution of 0.30 Å, the MolDock
optimizer as a search algorithm, and the number of runs was set to
10. A population size of 50, maximum iteration of 2000, scaling fac-
tor of 0.50, crossover rate of 0.90. The maximum number of poses to
generate was increased to 10 from a default value of 5.
4.4. Chromatography
The separation factor (a) was calculated as k2/k1 and retention
factors (k1 and k2) as k1 = (t1 À t0)/t0 where t1 and t2 refer to the
retention times of the first and second eluted enantiomers. The res-
olution factor (Rs) was calculated by the formula Rs = 2(t2 À t1)/
(w1 + w2) where w1 and w2 are the peak widths at base for the first
and second eluted enantiomers. The dead time of the columns (t0)
was determined by injection of 1,3,5-tri-tert-butylbenzene.
Pure (+)(S) and (À)(R) enantiomers of 7-chloro-9-(furan-3-yl)-
2,3,3a,4-tetrahydro-1H-benzo[e]pyrrolo[2,1-c][1,2,4]thiadiazine
5,5-dioxide (3) were obtained by semipreparative HPLC on Chiral-
cel OD semipreparative column with fraction collection of the
respective peaks. The mobile phase consisted of n-hexane and
2-propanol 95:5 (v/v). The compound was dissolved in n-hexane.
References and notes
The injection volume was 500 ll. The detector was set at
254 nm. The collected fractions corresponding to the enantiomers
were analyzed by injection on the same column and in the same
chromatographic conditions.
Separation of enantiomers of compound 3 was carried out iso-
cratically at 25–45 °C on Chiralcel OD-RH column using water/ace-
tonitrile (50:50, v/v) as mobile phase. The compound was dissolved
in ethanol and subsequently diluted 1:10 (v/v) with mobile phase
1. Krogsgaard-Larsen, P.; Ebert, B.; Lund, T. M.; Brauner-Osborne, H.; Slk, F. A.;
Johansen, T. N.; Brehm, L.; Madsen, U. Eur. J. Med. Chem. 1996, 31, 515.
2. Excitatory Amino Acid Receptors, Design of Agonists and Antagonists; Krogsgaard-
Larsen, P., Hansen, J. J., Eds.; Ellis Horwood: Chichester, 1992.
3. The NMDA Receptor; Collingridge, G. L., Watkins, J. C., Eds.; Oxford University
Press: Oxford, 1994.
4. Kew, J. N. C.; Kemp, J. A. Psychopharmacology 2005, 179, 4.
5. Francotte, P.; de Tullio, P.; Fraikin, P. Recent Pat. CNS Drug Discov. 2006, 1, 239.
6. Mayer, M. L.; Armstrong, N. Annu. Rev. Physiol. 2004, 66, 161.
7. O’Neill, M. J.; Bleakman, D.; Zimmerman, D. M.; Nisenbaum, E. S. Curr. Drug
Targets CNS Neurol. Disord. 2004, 3, 181.
8. O’Neill, M. J.; Dix, S. IDrugs 2007, 10, 185.
9. Zarate, C. A.; Manji, H. K. Exp. Neurol. 2008, 211, 7.
10. Hashimoto, K. Brain Res. Rev. 2009, 61, 105.
11. Pirotte, B.; Francotte, P.; Goffin, E. Curr. Med. Chem. 2010, 17, 3575.
12. Ward, S. E.; Harries, M. Curr. Med. Chem. 2010, 17, 3503.
13. Pirotte, B.; Francotte, P.; Goffin, E.; de Tullio, P. Expert Opin. Ther. Patents 2013,
23, 615.
14. Grove, S. J. A.; Jamieson, C.; Maclean, J. C. F.; Morrow, J. A.; Rankovic, Z. J. Med.
Chem. 2010, 53, 7271.
15. Ward, S. E.; Bax, B. D.; Harries, W. Br. J. Pharmacol. 2010, 160, 181.
16. Philips, D.; Sonnenberg, J.; Arai, A. C.; Vaswani, R.; Krutzik, P. O.; Kleisli, T.;
Kessler, M.; Granger, R.; Lynch, G.; Chamberline, A. R. Bioorg. Med. Chem. 2002,
10, 1229.
17. Braghiroli, D.; Puia, G.; Cannazza, G.; Tait, A.; Parenti, C.; Losi, G.; Baraldi, M. J.
Med. Chem. 2002, 45, 2355.
18. 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. J. Med. Chem. 1998,
41, 2946.
at final concentration of 100
50 l. The detector was set at 254 nm.
The sf-BD-rHPLC method was previously described.24 The
lg/ml. The injection volume was
l
racemic mixture of 3 was injected on Chiralcel OD. The mobile
phase consisted of n-hexane and 2-propanol 95:5 (v/v). The injec-
tion volume was 500 ll. The detector was set at 254 nm. The race-
mization of the trapped enantiomer was conducted employing an
IAM column as reactor column and isopropanol added of 1% (v/v)
of trifluoroacetic acid as racemization solvent. The reactor column
was left 1 h at 45 °C.
4.5. Electrophysiology
Primary cultures of cerebellar granule neurons were prepared
from 7 days old Sprague–Dawley rats as reported in the
literature.52 Briefly, cells from the cerebellum were dispersed with

4676
M. M. Carrozzo et al. / Bioorg. Med. Chem. 22 (2014) 4667–4676
19. Graindorge, E.; Francotte, P.; Boverie, S.; de Tullio, P.; Pirotte, B. Curr. Med.
Chem. 2004, 4, 95.
35. Cannazza, G.; Battisti, U. M.; Carrozzo, M. M.; Cazzato, A. S.; Braghiroli, D.;
Parenti, C.; Troisi, L. J. Chromatogr.,
A 2014, in press, http://dx.doi.org/
20. Hashimoto, K.; Okamura, N.; Shimizu, E.; Iyo, M. Curr. Med. Chem. 2004, 4, 147.
21. Francotte, P.; de Tullio, P.; Goffin, E.; Dintilhac, G.; Graindgorge, E.; Fraikin, P.;
Lestage, P.; Danober, L.; Thomas, J. Y.; Caignard, D. H.; Pirotte, B. J. Med. Chem.
2007, 50, 3153.
22. Arai, A.; Guidotti, A.; Costa, E.; Lynch, G. NeuroReport 1996, 7, 2211.
23. Thompson, D. M.; Guidotti, A.; DiBella, M.; Costa, E. Proc. Natl. Acad. Sci. U.S.A.
1995, 92, 7667.
doi:10.1016/j.chroma.2014.05.073.
36. Rosi, S.; Giovannini, M. G.; Lestage, P. J.; Munoz, C.; Corte, L. D.; Pepeu, G.
Neurosci. Lett. 2004, 361, 120.
37. Vandesquille, M.; Krazem, A.; Louis, C.; Pierre, L.; Beracochea, D.
Psychopharmacology 2011, 215, 709.
38. Lebrun, C.; Pilliere, E.; Lestage, P. Eur. J. Pharm. 2000, 401, 205.
39. Cameroni, R.; Bernabei, M. T.; Forni, F.; Baggi, G. Il Farmaco Ed. Sci. 1976, 31,
508.
40. Harpsoe, K.; Varming, T.; Gouliiaev, A. H.; Peters, D.; Liljefors, T. J. Mol. Graphics
Modell. 2007, 26, 213.
41. Desos, P.; Serkiz, B.; Morain, P.; Lepagnol, J.; Cordi, A. Bioorg. Med. Chem. Lett.
1996, 6, 3003.
42. Yamaguchi, S.; Mosher, H. S. J. Org. Chem. 1973, 38, 1870.
43. Cohen, N.; Lopresti, R. J.; Neukom, C.; Saucy, G. J. Org. Chem. 1980, 45, 582.
44. Deeter, J.; Frazier, J.; Staten, G.; Staszak, M.; Weigel, L. Tetrahedron Lett. 1990,
131, 7101.
24. Sun, Y.; Olson, R.; Horning, M.; Armstrong, N.; Mayer, M.; Gouaux, E. Nature
2002, 417, 245.
25. Ptak, C. P.; Ahmed, A. H.; Oswald, R. E. Biochemistry 2009, 48, 8594.
26. Battisti, U. M.; Jozwiak, K.; Cannazza, G.; Puia, G.; Stocca, G.; Braghiroli, D.;
Parenti, C.; Brasili, L.; Carrozzo, M. M.; Citti, C. Med. Chem. Lett. 2012, 3, 25.
27. Cannazza, G.; Braghiroli, D.; Tait, A.; Baraldi, M.; Parenti, C.; Lindner, W.
Chirality 2001, 13, 94.
28. Cannazza, G.; Carrozzo, M. M.; Braghiroli, D.; Parenti, C. J. Chromatogr., A 2008,
1212, 41.
29. Carrozzo, M. M.; Cannazza, G.; Battisti, U.; Braghiroli, D.; Parenti, C. Chirality
2010, 22, 389.
45. Jane, D. E.; Hoo, K.; Kamboj, R.; Deverill, M.; Bleakman, D.; Mandelzys, A. J. Med.
Chem. 1997, 40, 3645.
30. Cannazza, G.; Carrozzo, M. M.; Battisti, U.; Braghiroli, D.; Parenti, C.; Troisi, A.;
Troisi, L. Chirality 2010, 22, 789.
46. Cannazza, G.; Carrozzo, M. M.; Battisti, U.; Braghiroli, D.; Parenti, C. J.
Chromatogr., A 2009, 1216, 5655.
31. Carrozzo, M. M.; Cannazza, G.; Battisti, U. M.; Braghiroli, D.; Troisi, L.; Parenti,
C. J. Chromatogr., A 2010, 1217, 8136.
32. Cannazza, G.; Battisti, U. M.; Carrozzo, M. M.; Brasili, L.; Braghiroli, D.; Parenti,
C. Chirality 2011, 23, 851.
33. Cannazza, G.; Carrozzo, M. M.; Braghiroli, D.; Parenti, C. J. Chromatogr., B 2008,
875, 42.
47. Nonius COLLECT and EVAL; Nonius BV: Delft, The Netherlands, 2002.
48. Duisenberg, A. J. J. Appl. Crystallogr. 1992, 25, 92.
49. Sheldrick, G. M. SADABS; University of Göttingen: Germany, 2002.
50. Burla, M. C.; Caliandro, R.; Camalli, M.; Carrozzini, B.; Cascarano, G. L.;
Giacovazzo, C.; Mallamo, M.; Mazzone, A.; Polidori, G.; Spagna, R. J. Appl.
Crystallogr. 2012, 45, 357.
34. Battisti, U. M.; Carrozzo, M. M.; Cannazza, G.; Puia, G.; Troisi, L.; Braghiroli, D.;
Parenti, C.; Jozwiak, K. Bioorg. Med. Chem. 2011, 19, 7111.
51. Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112.
52. Murase, K.; Ryu, P. D.; Randic, M. Neurosci. Lett. 1989, 103, 56.