Novel arylsulfonamide derivatives with 5-HT6/5-HT7 receptor antagonism targeting behavioral and psychological symptoms of dementia

Article abstract of DOI:10.1021/jm401895u

In order to target behavioral and psychological symptoms of dementia (BPSD), we used molecular modeling-assisted design to obtain novel multifunctional arylsulfonamide derivatives that potently antagonize 5-HT _6/7/2A and D₂ receptors, without interacting with M ₁ receptors and hERG channels. In vitro studies confirmed their antagonism of 5-HT_7/2A and D₂ receptors and weak interactions with key antitargets (M₁R and hERG) associated with side effects. Marked 5-HT₆ receptor affinities were also observed, notably for 6-fluoro-3-(piperidin-4-yl)-1,2-benzoxazole derivatives connected by a 3-4 unit alkyl linker with mono- or bicyclic, lipophilic arylsulfonamide moieties. N-[4-[4-(6-Fluoro-1,2-benzoxazol-3-yl)piperidin-1-yl]butyl] benzothiophene-2-sulfonamide (72) was characterized in vitro on 14 targets and antitargets. It displayed dual blockade of 5-HT₆ and D₂ receptors and negligible interactions at hERG and M₁ receptors. Unlike reference antipsychotics, 72 displayed marked antipsychotic and antidepressant activity in rats after oral administration, in the absence of cognitive or motor impairment. This profile is particularly attractive when targeting a fragile, elderly BPSD patient population.

Full text of DOI:10.1021/jm401895u

Article
pubs.acs.org/jmc
Novel Arylsulfonamide Derivatives with 5‑HT6/5-HT7 Receptor
Antagonism Targeting Behavioral and Psychological Symptoms of
Dementia
Marcin Kołaczkowski,*,†,‡ Monika Marcinkowska,‡ Adam Bucki,‡ Maciej Pawłowski,‡ Katarzyna Mitka,§
Jolanta Jaskowska,§ Piotr Kowalski,§ Grzegorz Kazek,‡ Agata Siwek,‡ Anna Wasik,‡ Anna Wesołowska,‡
́
Paweł Mierzejewski,∥ and Przemyslaw Bienkowski∥
†Adamed Ltd., Pien
́ ́ ́
kow 149, 05-152 Czosnow, Poland
‡Faculty of Pharmacy, Jagiellonian University Collegium Medicum, 9 Medyczna Street, 30-688 Cracow, Poland
§Cracow University of Technology, 24 Warszawska Street, 31-155 Cracow, Poland
∥Institute of Psychiatry and Neurology, 9 Sobieskiego Street, 02-957 Warsaw, Poland
S
* Supporting Information
ABSTRACT: In order to target behavioral and psychological
symptoms of dementia (BPSD), we used molecular modeling-
assisted design to obtain novel multifunctional arylsulfonamide
derivatives that potently antagonize 5-HT6/7/2A and D2
receptors, without interacting with M1 receptors and hERG
channels. In vitro studies confirmed their antagonism of 5-
HT7/2A and D2 receptors and weak interactions with key
antitargets (M1R and hERG) associated with side effects. Marked 5-HT6 receptor affinities were also observed, notably for 6-
fluoro-3-(piperidin-4-yl)-1,2-benzoxazole derivatives connected by a 3−4 unit alkyl linker with mono- or bicyclic, lipophilic
arylsulfonamide moieties. N-[4-[4-(6-Fluoro-1,2-benzoxazol-3-yl)piperidin-1-yl]butyl]benzothiophene-2-sulfonamide (72) was
characterized in vitro on 14 targets and antitargets. It displayed dual blockade of 5-HT6 and D2 receptors and negligible
interactions at hERG and M1 receptors. Unlike reference antipsychotics, 72 displayed marked antipsychotic and antidepressant
activity in rats after oral administration, in the absence of cognitive or motor impairment. This profile is particularly attractive
when targeting a fragile, elderly BPSD patient population.
do not induce cognitive deficits would permit improved
treatment of BPSD.
INTRODUCTION
■
Over 50% of patients with Alzheimer’s disease (AD) or with
other dementias experience psychotic symptoms,1 and, together
with other behavioral and psychological symptoms, their
treatment constitutes a substantial unmet medical need. Indeed,
effective pharmacological treatment of psychosis in elderly
patients with cognitive deficits can be particularly intractable
because psychotropic medications used in patients with
dementia may themselves be a cause of further cognitive
impairment.2 Moreover, the use of currently available
antipsychotics in elderly patients with dementia has also been
associated with other side effects, including motor, metabolic,
or cardiac disturbances and, consequently, increased mortality.3
Despite “boxed warnings” issued by the Food and Drug
Administration (FDA) for antipsychotics and their lack of
approval for treatment of Behavioral and Psychological
Symptoms of Dementia (BPSD),4,5 such drugs have been
commonly used off-label.6 This is probably due to the fact that
psychosis and physical aggression by dementia patients (rather
than their cognitive impairment) are a major challenge for
caregivers and are the most common cause of patient
institutionalization.1,2 Therefore, the discovery of novel
antipsychotic drugs with an improved safety profile and that
At a pharmacological level, converging lines of evidence
indicate that blockade of serotonin 5-HT6 receptors (5-HT6Rs)
may be implicated in pro-cognitive effects due to the facilitation
of cholinergic transmission,7,8 in antidepressant activity due to
the increase in noradrenergic and dopaminergic tones, as well
as in an anxiolytic effect, mediated by interaction with GABA-
ergic transmission.9,10 These findings are further supported by
the exclusive localization of 5-HT6 receptors in the central
nervous system (CNS), especially in limbic and cortical brain
areas involved in the control of mood and cognition.11
Recently, the selective 5-HT6R antagonists 3-phenylsulfonyl-
8-(piperazin-1-yl)quinoline (SB-742457) and [2-(6-fluoro-1H-
indol-3-yl)-ethyl]-[3-(2,2,3,3-tetrafluoropropoxy)-benzyl]-
amine (Lu AE58054) showed pro-cognitive effects in phase II
clinical trials with AD patients,12,13 confirming the therapeutic
potential and relevance of 5-HT6R antagonists for dementia
patients.
Received: December 10, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm401895u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 1. Design of multifunctional ligands. Proposed binding modes presented for (1R)-3,N-dimethyl-N-[1-methyl-3-(4-methylpiperidin-1-
yl)propyl]benzenesulfonamide (SB-258719)24 in 5-HT7R (A) and compound 2 in the 5-HT6R (C) and hERG channel (D). Amino acid residues
engaged in ligand binding (within 4 Å from the ligand atoms) are shown as thick sticks together with their van der Waals surfaces (wire frame).
Distances between key amino acid moieties and ligand structural elements are shown in Supporting Information, Tables S1 and S2. The numbers in
brackets next to the residues in the hERG channel indicate the subunit number. Docking of 2 in M1R resulted in no valid poses (E). Dotted yellow
lines represent H-bonds with polar residues. For the sake of clarity, ECL2 was hidden (A and C). TMH, transmembrane helix; ECL, extracellular
loop.
Another serotonin receptor subtype, 5-HT7, may play a role
in the control of circadian rhythms, sleep, thermoregulation,
pain, and migraine, as well as cognitive processes. Indeed, the
high affinity and antagonistic activity of several antipsychotic
and antidepressant drugs at 5-HT7 receptors suggest a potential
role of these receptors in the pathophysiology of many
neuropsychiatric disorders. Accordingly, selective 5-HT7
receptor antagonists produce antidepressant and anxiolytic
activity in rats and mice.14 Moreover, Galici et al. and Waters et
al. showed that a selective 5-HT7 receptor antagonist (2R)-1-
[(3-hydroxyphenyl)sulfonyl]-2-(2-(4-methyl-1-piperidinyl)-
ethyl)pyrrolidine may also evoke antipsychotic-like effects.15,16
The above considerations suggest that targeting 5-HT6 and/
or 5-HT7 receptors with antagonist drugs could constitute a
promising strategy for treating symptoms of BPSD while
avoiding some of the side effects of current antipsychotic drugs.
Nevertheless, due to the complex pathology of dementia and
accompanying behavioral and psychological symptoms, it seems
unlikely that focusing on a single therapeutic target would be
sufficient to provide adequate clinical benefit, and it is likely
that successful development of novel anti-BPSD agents should
involve a “designed” multifactorial approach.17,18 Indeed,
although some of the currently available antipsychotics block
5-HT6 or 5-HT7 receptors, they also interact with many other
targets, including those that may elicit some side effects and/or
limit (mask) their therapeutic efficacy. For example, these drugs
also bind potently to muscarinic and histamine receptors, which
are associated with cholinergic interference and sedation,
B
dx.doi.org/10.1021/jm401895u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 2. Binding mode of compound 2 in (A) 5-HT7R, (B) 5-HT2AR, (C) D2R, and (D) 5-HT6R. Residues in the binding site (within 4 Å from the
ligand atoms) are presented as thick sticks together with their van der Waals surfaces (wire frame). Distances between key amino acid moieties and
ligand structural elements are shown in Supporting Information, Table S2. Dotted yellow lines represent H-bonds with polar residues. Where
possible, ECL2 was partially or entirely hidden for the sake of clarity.
respectively. In addition, they act as potent 5-HT2C receptor
antagonists, a property which is a risk factor for metabolic
dysfunction.19,20 Some authors proposed that an improved
antipsychotic profile may be achieved by combining 5-HT6
antagonism with an absence of antimuscarinic activity, as is
observed for sertindole, but not clozapine or olanzapine.21
However, clinical evaluation of sertindole was hampered by its
arrhythmogenic potential, caused by potent hERG channel
inhibition.
On the basis of the above considerations, we designed a
series of novel arylsulfonamide derivatives displaying high
affinity for serotonin 5-HT6, 5-HT7, and 5-HT2A receptor
subtypes, as well as dopamine D2 receptors, and devoid of
significant interaction with muscarinic receptors or hERG
channels.22 The present study describes computer-aided design
and the synthesis, as well as the in vitro receptor profile of this
new series of multimodal molecules. Moreover, the extensive
pharmacological characterization of one of the most promising
compounds, compound 72, is presented and discussed in
relation to comparative data for 8 currently used reference
antipsychotics.23
preferentially in the cavity between transmembrane helices
(TMHs) 7−3 (Figure 1A). We also hypothesized that 5-HT7R
ligands capable of forming strong interactions with this site may
not be dependent on interactions with the cavity between
TMHs 4−6, which is much more structurally conserved
between monoaminergic GPCRs. Therefore, such ligands are
likely to be more selective than those requiring specific
interactions with the more conserved cavity between TMHs 4−
6. Considering the fact that our goal was to design
multifunctional ligands acting on 5-HT7R, we found that the
N-aminoalkylarylsulfonamide fragment is a suitable basis for
structural modifications. Indeed, it provides strong 5-HT7R
antagonism by anchoring between TMHs 7−3 (particularly
Phe3.28 and Arg7.36) and allows for the introduction of
additional moieties providing activity at other targets, such as
the dopamine D2 receptor (D2R), serotonin 5-HT2A receptor
(5-HT2AR), or serotonin transporter (SERT) (Figure 1B).22 In
the present study, we focused on the moieties that block D2 and
5-HT2A receptors and exert potential antipsychotic-like activity.
In view of this objective, we searched for commercially available
fragments that could provide antagonism of D2 and 5-HT2A
receptors and would be synthetically suitable for combination
with the arylsulfonamide moiety by an alkyl linker. To this end,
we analyzed structures of the D2/5-HT2A ligands available in
the ChEMBL database (Ki < 10 nM for both receptors) and
identified the arylamine moieties that could bind in the cavity
between TMHs 4−6, which had been identified as a tolerant
region for chemical expansion. Benzazolepiperazine/piperidine
moieties were found to meet the above-mentioned criteria, and
RESULTS AND DISCUSSION
■
Design of Multifunctional Ligands. In the course of our
research into the structure of monoaminergic G protein-
coupled receptors (GPCRs), we proposed a model of the
binding mode of selective 5-HT7 receptor (5-HT7R)
antagonists.24 On the basis of molecular modeling studies, we
concluded that the most selective 5-HT7 antagonists bind
C
dx.doi.org/10.1021/jm401895u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 3. General structures of hybrid molecules, benzazolepiperazines/piperidines I−III and arylsulfonyl fragments Ar (1−15).
their drug-likeness was further supported by the fact that they
could be found in some clinically used drugs.25 As a next step, a
prototype hybrid molecule (compound 2) was designed by
combining the 3-methyl-N-propylbenzenesulfonamide frag-
ment with the 3-(piperazin-1-yl)-1,2-benzothiazole moiety
(Figure 1C).
The molecule was then docked to the models of 5-HT7, 5-
HT2A, and D2 receptors, showing good fit and favorable
interactions that confirmed the assumed ligand binding mode at
those targets (Figure 2A, B, and C, respectively). Because 5-
HT6R antagonism was considered a highly desirable feature of
these ligands, the prototype, hybrid compound 2, was also
docked to the 5-HT6R model, in order to assess the possibility
of its interactions with this target or potential modifications
required for this activity to occur. Surprisingly, compound 2
was very well accommodated by the 5-HT6 receptor, suggesting
that it possessed marked affinity for this site, without particular
modifications. The general orientation of the ligand in the
binding site was similar to that observed for 5-HT7, 5-HT2A, or
D2 receptors (Figures 1C and 2D).
(piperazin-1-yl)-1,2-benzothiazole moiety of compound 2 was
found to form charge-reinforced H-bond interactions between
the protonated nitrogen atom of piperazine and Asp3.32, as
well as hydrophobic/aromatic interactions between the
benzisothiazole moiety and the aromatic cluster of TMH6,
showing very similar poses in this conserved pocket in all 4
receptors. The arylsulfonamide moiety occupies the second
pocket, forming interactions between the aryl ring and
particular aromatic residue (Phe/Trp3.28). Because of the
relatively higher sequence diversity in the second pocket, there
are some differences in the arrangement of the arylsulfonamide
moiety between the receptors. For example, the aromatic
residue in the 3.28 position determines the conformation of the
latter. In the case of 5-HT6R and 5-HT2AR containing the Trp
residue, the arylsulfonamide moiety expands between TMH1
and TMH2, while in 5-HT7R and D2R with the Phe residue, it
kinks in between TMH2 and TMH3. Both of those geometries
are additionally stabilized by H-bonds, formed between the
sulfonamide group and Arg181, Arg7.36, or Ser226 of 5-HT6R,
5-HT7R, and 5-HT2AR, respectively.
As mentioned above for 5-HT7R, the binding sites of 5-
HT2A, D2, and 5-HT6R are also divided into two pockets that
extend to both sides of the main ligand-anchoring residue in
monoaminergic receptors, Asp3.32. The first pocket is situated
between TMHs 4−6, and its most important role in ligand
binding lies in the formation of hydrophobic/aromatic
interactions with aromatic clusters consisting of Phe6.51,
Phe6.52, and Trp6.48, which are the conserved amino acid
residues in these receptors. The second pocket is formed
between TMHs 7−3, with the contribution of residues that
ensure both aromatic (Phe3.28 in 5-HT7R and D2R or Trp3.28
in 5-HT6R and 5-HT2AR) and H-bond interactions (Arg7.36 in
5-HT7R, Asn7.36 in 5-HT2AR, and Thr7.39 in D2R). The 3-
Encouraged by the promising in silico properties, which
suggested high affinity for the most important therapeutic
targets (5-HT6, 5-HT7, 5-HT2A, and D2 receptors), possible
interactions with the undesired off-targets (called “anti-
targets”), hERG channels or M1 receptors, were also examined
because blockade of these sites could negatively affect the
compounds’ safety profile and mask potential therapeutic
activity. In contrast to 5-HT6, 5-HT7, 5-HT2A, and D2
receptors, compound 2 was not optimally accommodated
within the binding sites of the antitargets. In the case of the M1
receptor, the relatively long molecule could not freely expand
throughout the binding site and form favorable interactions,
similar to those observed in the targeted receptors, due to the
D
dx.doi.org/10.1021/jm401895u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Table 1. Structure and Receptor Binding Profile of Compounds 1−81
a
Data expressed as the mean SD of two independent experiments in duplicate. Affinities of reference compounds are as follows: 5-HT6 Ki = 4.1
0.7 nM and 5-HT7 Ki = 0.9 0.1 nM (methiothepin), 5-HT2A Ki = 1.0 0.2 nM (risperidone), D2 Ki = 4.0 0.3 nM (haloperidol), M1 Ki = 15
3.1 nM (pirenzepine), hERG IC50 = 89 18 nM (E-4031).
relatively limited volume of binding sites in muscarinic
receptors (Figure 1E and Figure S1 in Supporting Informa-
tion).26,27 In general, amino acid sequence homology in the
region of the binding site between M1R and the receptors
described above is low. There are numerous differences in
sequence, which contribute to various dissimilarities in shape
and properties. The binding site of M1R is substantially
restricted spatially with higher contribution of hydrophilic
residues. The reason for that is basically the presence of
tyrosine residues, which replace several important amino acids
present in previously discussed receptors. The first binding
pocket (between TMHs 4−6) is confined by the presence of
Tyr3.33 (instead of Val in the above-mentioned receptors) and
Tyr6.51 (instead of Phe). Nevertheless, crucial for selectivity is
in fact the absence of the second binding pocket (between
TMHs 7−3). It is caused by the presence of Tyr7.39, Tyr2.61,
and Tyr2.64, which respectively substitute Thr7.39, Ala2.61,
and Asn2.64 in 5-HT6R, thus making the second binding cavity
inaccessible for the ligand. Docking studies suggested the
potentially poor blocking activity of compound 2 at the hERG
channel as well. Although it could be accommodated in the
hERG channel model, its binding mode was suboptimal, due to
the lack of the hydrophobic/aromatic interactions in regions of
four phenylalanine (Phe656) residues, which are considered
important for channel blockade (Figure 1D and Figure S2 in
Supporting Information).28,29 The prototype molecule binds to
the model by a H-bond between the protonated nitrogen of
piperazine and Ser624, although it seems to be insufficiently
stabilized by aromatic interactions. In particular, there is no
additional, required interaction in the region between
Phe656(1) and Phe656(4) or in the vicinity of Phe656(3),
where the sulfonamide moiety is placed.
All those findings prompted us to synthesize a diverse set of
multifunctional ligands that explored various parts of the
prototype compound 2, including arylsulfonamide fragment,
alkyl linker, and arylamine moiety (Figure 1F). The selection of
particular building blocks was based primarily on their best fit
to the binding site of 5-HT6R, which was considered the most
important receptor target. Analysis of the composition of the
binding site of 5-HT6R as well as the binding mode of
compound 2 and its close analogues revealed that the aryl
moiety connected with sulfonamide fragment penetrates the
hydrophobic regions near TMHs 7 and 2. It forms van der
Waals interactions with the hydrophobic residues of TMH2
(Ala2.61, Ala2.65), TMH3 (Trp3.28), or TMH7 (Trp7.40),
and therefore, the arylsulfonyl moieties selected for the
designed series were predominantly lipophilic, with a few
exceptions to confirm this hypothesis (Figure 3).
Moreover, the space available in the regions accommodating
the arylsulfonamide moiety suggested that mono- or bicyclic
aromatic fragments should be considered in the diverse set
(Figure 2A). On the basis of the visual inspection of ligand−
receptor complexes, the optimal length of the alkyl linker was
assumed to be 3 or 4 units, and those lengths were selected to
be thoroughly explored in the designed series. The longer
linkers were not considered due to a high molecular weight of
the resulting hybrids. However, considering their lower
molecular weight, several combinations with a 2-unit alkyl
linker were also obtained for comparison. All of the designed
molecules were docked to 5-HT6, 5-HT7, 5-HT2A, D2, and M1
E
dx.doi.org/10.1021/jm401895u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 1. General Synthesis of Hybrid Molecules
a
Reagents and conditions: (a) K2CO3, KI, DMF, rt, 48 h or DIPEA, KI, DMF/CH3CN 1:1, 70 °C, 3 days; (b) 40% aq MeNH2, rt, 76 h; (c) Et3N,
DCM, 3 h, 10 °C then rt.
Figure 4. Binding mode of the compounds (A) 51 and (B) 72 in the 5-HT6 receptor. Residues in the binding site (within 4 Å from the ligand
atoms) are presented as thick sticks together with their van der Waals surfaces (wire frame). Distances between key amino acid moieties and ligand
structural elements are shown in Supporting Information, Tables S3 and S4. Dotted yellow lines represent H-bonds with polar residues. ECL2 was
partially hidden for the sake of clarity. TMH, transmembrane helix; ECL, extracellular loop.
receptors as well as hERG channels, and their binding mode
was analyzed, confirming in general the observations made for
compound 2 and its close analogues. Consequently, the series
of 185 combinations was selected for chemical synthesis.22 In
this article, for the sake of clarity, we present 81 representative
examples out of a total of 185 compounds synthesized and
tested (Table 1).
Synthesis. A series of hybrid molecules (1−81) was
prepared in the three-step synthesis as presented in Scheme
1. In the first step, the alkylation of benzazolepiperazines/
piperidines I−III with 2-(bromoalkyl)-1H-isoindoline-1,3(2H)-
diones IV−VI afforded corresponding derivatives VII−XIV.
Next, hydrolysis of the 1H-isoindoline-1,3(2H)-dione group
with 40% methylamine aqueous solution afforded key
intermediates XV−XXII, which were reacted with various
commercially available sulfonyl chlorides (1−15) to give final
arylsulfonamide derivatives (1−81).
Structure−Affinity Relationships. The series of hybrid
molecules that we had generated was evaluated in radioligand
binding assays measuring their affinity for 5-HT6, 5-HT7, 5-
HT2A, D2, and M1 receptors as well as in the automated patch-
clamp assay for hERG channel blockade (Table 1).
The majority of the compounds displayed marked affinity (Ki
< 100 nM) for the targeted receptors and much lower activity
at the antitargets (<50% inhibition of ligand binding at 1 μM).
Moreover, numerous compounds possessed high affinity at the
main targets (Ki < 10 nM), with some of them binding in the
subnanomolar range, thus confirming the validity of the applied
rational design approach. In general, the compounds displayed
the highest affinity for 5-HT7 and 5-HT2A receptors as well as
strong binding to D2 receptors, i.e., with the sites that were
primarily targeted during the design process of the hybrid
molecules. According to the molecular modeling studies, high
affinity for 5-HT7 receptors is provided by the N-amino-
alkylarylsulfonamide moiety, which strongly interacts with the
part of the binding site between TMHs 7−3. This makes it
relatively less sensitive to the influence of the arylamine part,
accommodated in the region of TMHs 4−6. These properties
of the N-aminoalkylarylsulfonamide moiety are complemented
by those of the benzazolepiperazine/piperidine moiety, which
provides high affinity for 5-HT2A and D2 receptors, the former
being favored in this respect. However, 5-HT6R affinity varied
widely, and some clear structure−activity relationships could be
observed.
First, the analogues containing the 6-fluoro-3-(piperidin-4-
yl)-1,2-benzoxazole III were somewhat more active at 5-HT6
sites than those containing 3-(piperazin-1-yl)-1,2-benzothiazole
I, but both of these groups were much more active than the
corresponding 3-(piperazin-1-yl)-1,2-benzoxazole II derivatives.
Similar preferences were observed also for the other main
F
dx.doi.org/10.1021/jm401895u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
targets, although to a less marked extent in the case of 5-HT7R
and 5-HT2AR between 6-fluoro-3-(piperidin-4-yl)-1,2-benzox-
azole and 3-(piperazin-1-yl)-1,2-benzothiazole derivatives.
Second, the arylsulfonamide moieties containing nitrogen or
oxygen atoms in the aryl part were less potent at 5-HT6R than
those which are more hydrophobic, especially those containing
a benzene ring in the distal part of this fragment (e.g.,
compounds 68 and 77 vs 79 and 81). These findings are in line
with our docking experiments, which positioned the aromatic
moieties of arylsulfonamide groups in the hydrophobic pocket
between TMHs 7 and 2 of 5-HT6R (Figure 4).
Table 2. Functional Data for the Selected Compounds with
Highest Affinity for Main Targets
a
Kb [nM]
compd
5-HT6
5-HT7
5-HT2A
D2
14
38
51
53
55
63
66
68
72
74
76
77
7.6 0.2
11 1.9
2.5 0.0
12 3.3
7.4 2.1
17 1.8
3.8 0.2
8.3 1.7
9.9 5.1
21 4.0
14 1.0
22 3.5
5.8 0.6
44 9.0
9.2 3.9
0.5 0.3
2.1 0.7
0.8 0.5
3.0 0.9
8.8 1.2
1.0 0.1
7.8 3.3
2.8 0.6
2.2 1.2
5.6 2.1
110 2.0
11 1.4
15 1.0
5.5 1.9
11 6.0
1.8 0.6
14 1.5
1.1 0.1
8.4 0.6
2.4 0.4
7.3 2.8
17 3.5
15 1.0
3.4 0.8
11 2.2
23
7
7.0 0.6
76 25
14 2.5
6.5 0.6
8,8 1.5
55 5.0
5.0 1.7
9.5 3.5
Third, the analogues with the ethyl linker were substantially
less active than those with longer spacers. Interestingly,
differing structure−activity relationships were observed for
the compounds with 3 or 4 unit linkers. In the groups of 6-
fluoro-3-(piperidin-4-yl)-1,2-benzoxazole III and 3-(piperazin-
1-yl)-1,2-benzothiazole I, the propyl analogues formed the
optimal combinations with the monocyclic arylsulfonamides,
while the butyl derivatives were more active in combination
with the bicyclic ones. These observations could be explained
by the somewhat different binding modes of ligands with
mono- and bicyclic arylsulfonamides in the 5-HT6R binding site
(Figure 4). It is hypothesized that monocyclic substituents
attached to arylpiperazine by shorter propyl linkers reach a
lower part of the pocket, interacting with the hydrophobic
surface of Trp3.28, Ala2.61, and possibly Trp7.40 and Tyr7.43.
At the same time, the sulfonamide moiety creates an H-bond
with polar Arg181. However, bicyclic derivatives with a butyl
linker tend to explore the pocket even further, interacting with
Ala2.65 and Trp7.40. In this case, the sulfonamide moiety
establishes an H-bond with Asp7.36, rather than with Arg181.
The middle position occupied by the other derivative
configurations are unfavorable, probably due to the fact that
the aryl substituent takes position in the vicinity of polar
residues, i.e., Arg181, Asp7.36, or Asn2.64. The same trend,
although less pronounced, could be noticed regarding the 5-
HT7R affinity. On the contrary, in the group of 3-(piperazin-1-
yl)-1,2-benzoxazole II, the 4-unit alkyl linker derivatives were
significantly more active at all receptors, especially 5-HT6,
regardless of the arylsulfonamide structure. The reason for this
discrepancy, as well as for the generally low activity of propyl
derivatives of 3-(piperazin-1-yl)-1,2-benzoxazole, is currently
unclear and could not be explained by the present molecular
modeling experiments.
Intrinsic Activity Studies. On the basis of the binding
affinity results, a series of 12 compounds possessing affinity
below 20 nM for all four targeted receptors was selected for
functional profile characterization using in vitro measures of
receptor activation. All of the tested compounds confirmed
their high activity at the targets of interest and showed
significant antagonist properties (Table 2).
No activation of any of the receptors was observed, and the
compounds showed potent antagonist properties at all the
receptors, especially the 5-HT7. The compounds with
monocyclic arylsulfonamides and propyl linker (e.g., com-
pounds 51 and 55) showed low subnanomolar Kb values at 5-
HT7 sites, which is in line with previous findings on optimal
structures for 5-HT7R antagonism.24,30 No clear impact of
linker length or arlylsulfonamide fragment composition on 5-
HT6R antagonism was observed in this group of high affinity
compounds. Potent 5-HT6R antagonists were identified among
both the monocyclic arylsulfonamides with propyl linker and
bicyclic arylsulfonamides with butyl linker, achieving single-
SB-271046
SB-269970
ketanserin
1.2 0.2
2.9 0.2
chlorpromazine
1.3 0.5
a
Data are expressed as the mean
experiments in duplicate.
SD of two independent
digit nanomolar Kb values (compounds 14, 51, 55, 66, 68, and
72). Seven compounds (38, 51, 55, 66, 68, 72, and 76)
displayed potent antagonistic properties at all four main targets,
with Kb values <20 nM. Recently, we have published the
pharmacological profile of a representative example of this
series, compound 51.31 Here, we present the extended in vitro
and in vivo characterization of compound 72 possessing the
most balanced receptor affinity profile as well as improved
activity after oral administration.
Extended in Vitro Profile of Compound 72. Compound
72, tested in the independent laboratory of Cerep (France),
displayed very high affinity and antagonism at all four main
targets (Ki and Kb values <20 nM for 5-HT6, 5-HT7, 5-HT2A,
and D2 receptors). A notable feature of the compound is its
concurrent antagonism of 5-HT6 and D2 receptors (Kb = 16
and 13 nM, respectively), which corresponds to our goal of
identifying compounds that combine activity at targets that are
associated with both antipsychotic-like properties and favorable
cognitive properties. In comparison, compound 72 did not bind
significantly to muscarinic receptors (Ki >1000 nM) and only
weakly blocked hERG channels (IC50 ∼4 μM, ChanTest,
USA), thus achieving a large separation (several orders of
magnitude) between the activity at the principle therapeutic
targets and that at some key antitargets. These observations are
in sharp contrast to some existing drugs that act at both D2 and
5-HT6 receptors, e.g., olanzapine and clozapine, which also
strongly antagonize muscarinic receptors (Ki < 10 nM),32 or
sertindole, which is a potent hERG channel blocker (IC50 = 2.7
nM).33
In addition to the main targets and antitargets, compound 72
was also tested for affinity and intrinsic activity at a battery of
dopaminergic (D3, D4, D1), adrenergic (α1A, α2C), histaminergic
(H1), muscarinic (M2−M5), and serotonergic (5-HT1A, 5-
HT2C) receptors as well as the serotonin transporter (SERT)
(Tables 3 and 4).
Compound 72 displayed high affinity (Ki < 10 nM) for D1,
D3, α1A, and α2C receptors and showed potent antagonism (Kb
< 10 nM) at D3 and α1A receptors. Antagonism of D1 and α2C
receptors was somewhat weaker but also observed (Kb = 37 and
G
dx.doi.org/10.1021/jm401895u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Table 3. Extended Receptor Binding Profile of Compound 72
receptor
cell line
radioligand or read-out
reference ligand (Ki, nM)
Ki
SEM (nM)
4.1 1.0
h 5-HT6
h 5-HT7
h 5-HT2A
h 5-HT1A
h 5-HT2C
h D2S
CHO cells
CHO cells
HEK-293 cells
HEK-293 cells
HEK-293 cells
HEK-293 cells
CHO cells
CHO cells
CHO cells
CHO cells
CHO cells
HEK-293 cells
CHO cells
CHO cells
CHO cells
CHO cells
CHO cells
CHO cells
CHO cells
[3H]LSD
[3H]LSD
serotonin (33)
serotonin (0.8)
ketanserin (0.3)
8-OH-DPAT (0.4)
RS102221 (0.6)
(+)butaclamol (0.04)
(+)butaclamol (0.04)
clozapine (17)
1.9 0.2
0.5 0.09
110 53
540 139
1.8 0.31
1.1 0.24
14 0.82
3.3 0.23
0.8 0.09
1.9 1.1
69 19
[3H]ketanserin
[3H]8-OH-DPAT
[3H]mesulergine
[3H]methyl-spiperone
[3H]methyl-spiperone
[3H]methyl-spiperone
[3H]SCH23 390
[3H]prazosin
h D3
h D4
h D1
SCH23390 (0.1)
prazosin (0.04)
yohimbine (0.5)
pyrilamine (1.7)
pirenzepine (16)
methoctramine (27)
4-DAMP (0.44)
4-DAMP (0.88)
4-DAMP (0.48)
imipramine (2.1)
E-4031 (89)
h α1A
h α2C
[3H]RX 821 002
[3H]pyrilamine
[3H]pirenzepine
[3H]AF-DX 384
[3H]4-DAMP
h H1
b
h M1
>1000
b
h M2
>1000
b
h M3
>1000
b
h M4
[3H]4-DAMP
>1000
b
h M5
[3H]4-DAMP
>1000
c
h SERT
hERG
[3H]imipramine
>1000
c
inhibition of K+ current
4000 900
a
Data are expressed as the mean SEM of three independent experiments performed in duplicate. Each cell line was transfected to stably express
b
c
the indicated human recombinant receptor. Less than 40% inhibition of binding observed at a concentration of 1 μM (n = 2). IC50 value. h: human
recombinant.
a
Table 4. Extended Functional Profile of Compound 72
receptor
functional test
agonist (conc, nM)
reference ligand (IC50, nM)
Kb SEM [nM]
h 5-HT6
h 5-HT7
h 5-HT2A
h 5-HT1A
h 5-HT2C
h D2S
cAMP
cAMP
IP1
serotonin (300)
serotonin (300)
serotonin (100)
8-OH-DPAT (100)
serotonin (10)
methiothepin (11)
methiothepin (1.5)
ketanserin (14)
WAY100635 (2.1)
SB-206553 (9.4)
butaclamol (5.0)
butaclamol (16)
clozapine (168)
SCH23390 (1.7)
WB4101 (1.0)
16 4.7
4.5 1.3
4.8 0.87
>1000
cellular dielectric spectroscopy
IP1
N.C.
cAMP
dopamine (300)
dopamine (10)
dopamine (100)
dopamine (300)
epinephrine (3.0)
epinephrine (30)
histamine (300)
13 8.2
2.3 0.47
23 6.3
37 18
1.8 0.17
25 4.5
140 20
h D3
cAMP
h D4
cAMP
h D1
cAMP
intracellular Ca2+ release
h α1A
h α2C
cAMP
intracellular Ca2+ release
rauwolscine (16)
pyrilamine (3.0)
h H1
a
Data are expressed as the mean SEM of three independent experiments performed in duplicate. cAMP, cyclic adenylyl monophosphate; IP1,
inositol phosphate; h, human recombinant; N.C., noncalculable.
25 nM, respectively). Significant binding affinity for D4 and H1
receptors (Ki = 14 and 69 nM, respectively) was in line with the
antagonistic effects at those sites (Kb = 23 and 136 nM,
respectively). However, compound 72 only weakly bound to 5-
HT1A and 5-HT2C receptors or to the serotonin transporter,
resulting in very weak antagonist properties (Kb or IC50 > 1000
nM).
Of the above properties of compound 72, strong blockade of
D3 receptors may be considered as a desirable attribute since it
elicits procognitive effects via the facilitation of acetylcholine
release.34 Moreover, antagonism of α2C adrenergic receptors
could also positively influence the mood-modulating properties
of compound 72 due to the elevation of prefrontal cortical
monoamine outflow.35
without significant sedation or hypotension.36 Furthermore, in
patients with Alzheimer’s disease, α1 receptor deregulation has
been reported in the frontal cortex of the post-mortem brain,37
suggesting that targeting this receptor may correct a neuro-
biological imbalance associated with the disorder. However,
antagonism of peripheral α1 receptors is associated with
hypotensive activity, which would need to be addressed in
elderly patients with potentially fragile cardiac function.
In spite of significant activity at H1 receptors, the separation
between the antagonist properties at those sites and the main
therapeutic targets is an order of magnitude (Kb = 140 nM vs
4.5−16 nM). In contrast, antipsychotic compounds with
known, strong antihistaminic properties, like olanzapine or
clozapine, show the opposite trend, with higher affinity at H1
receptors than at the principal therapeutic targets.32 Since the
blockade of H1 receptors has been associated with sedative
effects,38 the relatively weak antihistaminic activity of
compound 72 may contribute to its low propensity to elicit
sedation (see the next section). Moreover, the low affinity and
Antagonism of central α1 receptors may also be beneficial for
a compound targeting behavioral and psychological symptoms
of dementia, especially agitation and aggression. Indeed, the α1
receptor antagonist, prazosin, elicited pronounced antiagitation
and antiaggressive effects in patients with Alzheimer’s disease
H
dx.doi.org/10.1021/jm401895u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
absence of antagonist properties of compound 72 at 5-HT2C
receptors avoids concerns associated with this site related to
increased appetite and body weight or metabolic disturban-
ces.39
enabled us to generate a series of compounds that display the
desired in vitro receptor profile. Namely, we managed to obtain
multimodal ligands that exert simultaneous blockade of 5-HT6
and D2 receptors and without deleterious interactions with key
antitargets (muscarinic receptors or hERG channels) and
distinct from currently available antipsychotics. Such a
molecular interaction pattern, enriched with antagonism of
other important targets, including 5-HT7 and 5-HT2A receptors,
should elicit potentially beneficial psychotropic effects with an
improved side effect profile.
In addition, in vivo characterization of a representative
molecule from the series, compound 72, revealed its excep-
tional activity in behavioral tests. Unlike reference antipsy-
chotics, compound 72 achieved marked therapeutic-like activity
in rat models of psychosis and depression without interfering
with cognitive function or motor control and indicates that
further investigation of this chemical series is warranted.
In Vivo Activity of Compound 72. In view of the
encouraging in vitro profile described above, compound 72 was
subjected to pharmacological experiments in rats. Specifically,
the antipsychotic-like and antidepressant-like activity of
compound 72 was compared with its influence on cognition.
Psychotic disturbances in dementia may have different
neurobiological substrates than in schizophrenia, so non-
dopaminergic models of psychosis were employed. Indeed,
MK-801-induced hyperlocomotion and DOI-induced head
twitches represent experimental paradigms based on the
disruption of glutamate and serotonin transmission, respec-
tively. Antidepressant-like activity was measured using Porsolt’s
forced swim test (FST), a well validated procedure for primary
screening, while the influence of compound 72 on cognition
was tested using the passive avoidance test. Moreover, given
that potential motor impairment is an important consideration
in elderly patients, induction of catalepsy and changes in
spontaneous locomotor activity were also quantified. The
results showed that compound 72 dose-dependently reversed
hyperlocomotion and head-twitches induced by MK-801 and
DOI, with minimum effective doses (MED) of 10 and 3 mg/kg
p.o. respectively, thus demonstrating a pronounced antipsy-
chotic-like effect. Furthermore, it decreased immobility time in
the FST by 20% over a relatively broad dose range, from 0.3 to
3 mg/kg p.o., consistent with a prominent antidepressant-like
effect. The effect size of compound 72 in the FST was
substantial and similar to that of the tricyclic antidepressant,
imipramine, tested under the same conditions. In contrast,
among the 8 clinically used antipsychotic drugs tested as
references, only clozapine was able to show activity in the FST
at more than one dose, with a maximal effect size >11% of
control value, whereas the other drugs had little or no activity in
the FST (Supporting Information, Table S5).23 In addition to a
robust, oral activity in the therapeutic-like models, compound
72 did not disrupt memory in the passive avoidance test even at
the highest dose tested (100 mg/kg p.o.). This is a striking
observation because dementia patients with BPSD also suffer
from core, cognitive deficits that should not be further
exacerbated by pharmacotherapy. In contrast, the reference
antipsychotic drugs dose-dependently disrupted cognition in
the passive avoidance test at antipsychotic-like doses, in
accordance with clinical findings on the cognitive impairments
elicited by these medications (Supporting Information, Table
S5).2,23 The only exception was aripiprazole, which did not
disrupt passive avoidance response but also failed to exhibit
robust antipsychotic-like effects. This observation is in line with
clinical studies on aripiprazole, showing a lack of satisfactory
antipsychotic efficacy in dementia patients.40 Finally, com-
pound 72 did not significantly modify spontaneous locomotor
activity or induce catalepsy up to a dose of 100 mg/kg p.o.,
suggesting that it does not interfere with normal motor control.
In contrast, most of the reference antipsychotics tested
produced dose-dependent catalepsy, and all of them produced
significant sedation at doses in the therapeutic-like range
(Supporting Information, Table S5).23
EXPERIMENTAL SECTION
■
Molecular Modeling. The homology models of the human D2
dopamine and 5-HT7 serotonin receptors used herein were the same
as those generated and described in our previously published
papers.41,42 The additional homology models of other GPCRs
presented in the current article (i.e., serotonin 5-HT6, 5-HT2A and
muscarinic M1 receptors) were derived in accordance with the same,
well validated method.
It should be noted that the models used in the present work differ
from most of previously published models of the GPCRs in
question,43−47 in at least one of the two main aspects. First, they
were prepared using templates of recently characterized high-
resolution crystal structures of human β2 adrenergic and D3
dopaminergic receptors, displaying much higher structural identity
with the modeled receptors than the previously used templates, e.g.,
bovine rhodopsin. Second, the optimization of the binding sites of the
new models was performed using the recently introduced method-
ology, based on the ligand-induced fit, which has been shown to result
in significantly more robust models.41 The specific characteristics of
the receptor models used in the present study are described in detail
below.
The homology model of the hERG potassium channel and its
induced fit structure applied for docking studies were acquired from
http://www.schrodinger.com.48 The novel models used for docking
experiments were built on the template of β2 adrenergic receptor
(β2R) crystal structure (Protein Data Bank ID: 2RH1).49 Reasonable
sequence similarity between modeled receptors and the adrenergic β2
receptor justified the choice of the latter structure as a template for
homology modeling. The crystal structure containing the heptahelical
bundle and all of the originally occurring loops was used, with the
exception of artificial fragments replacing the third intracellular loop
(ICL3) in the crystallized receptor, which were removed. Short loops
were created then to avoid modeling of the long ICL3, which is distant
from the binding site and therefore beyond the scope of our
experiments. Sequence alignment was performed via GeneSilico
Metaserver,50 which is a gateway to multiple fold recognition servers.
The results proposed by a profile−profile comparison tool, hhsearch,
were selected as the starting alignments between the sequences of 5-
HT6, 5-HT2A, and M1 receptors (UniProt accession numbers P50406,
P28223, and P11229, respectively)51 and the template. The alignments
were manually adjusted in Swiss Pdb Viewer and submitted to the
SwissModel server. Using the SwissModel algorithm, energy-
minimized (GROMOS96) homology models were obtained.52−54
Further improvements of the models were applied using software
implemented in Schrodinger Suite 2009. All of the operations
̈
engaging Schrodinger’s tools were executed with default settings,
̈
unless stated otherwise. The homology models were validated by
processing in Protein Preparation Wizard, which consisted of
optimizing the hydrogen-bonding network and restrained minimiza-
tion of the whole system with the OPLS_2005 force field. The least
CONCLUSIONS
Taken together, the present study indicates that the applied
drug design and structure−activity relationship (SAR) strategy
■
I
dx.doi.org/10.1021/jm401895u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
conserved fragments of the models, i.e., extracellular loops (ECLs),
among which ECL2 is known to be involved in ligand binding, were
additionally refined using the Prime Refinement tool.
General Procedure for the Deprotection of the Phthalimide
Group (XV−XXII). A mixture of 6 mmol of an appropriate compound
VII−XIV and 30 mL of 40% aqueous solution of methylamine was
stirred at room temperature for 3 days. To the resulting solution, 30
mL of 20% aqueous solution of sodium hydroxide was added, and the
resulted mixture was stirred for 1.5 h. Then 4 g of sodium chloride was
added, and the solution was extracted with methylene chloride (30
mL). The organic layer was washed with water (30 mL) and then
dried over anhydrous magnesium sulfate. The product was obtained by
evaporation of methylene chloride from dry solution. If necessary, the
product was purified using column chromatography on silica gel using
chloroform/methanol 95:5 as eluent. The structure was confirmed by
mass spectrometry.
General Procedure for the Preparation of Sulfonamide
Derivatives (1−81). Triethylamine (3.6 mmol (7.2 equiv)) and 0.5
mmol (1 equiv) of suitable arylsulfonyl chloride (1−15) were added to
the solution of 0.5 mmol (1 equiv) of amine, XV−XXII, in 10 mL of
methylene chloride at 10 °C. The reaction mixture was left at room
temperature for 3 h, then the solvent and excess of triethylamine were
evaporated. The resulting precipitate was then dissolved in 10 mL of
methylene chloride and washed subsequently with 5% solution of
sodium hydrogen carbonate (10 mL) and water (10 mL). The organic
layer was dried over anhydrous magnesium sulfate, and the solvent was
evaporated. Crude sulfonamides were purified by crystallization from
methanol and some of them were by using column chromatography
on silica gel using chloroform/methanol 9:1 as eluent. The structure
and purity of prepared compounds was confirmed by LCMS data, and
for selected compounds, structure identification was confirmed by
NMR analysis.
Binding pockets of the resulting homology models were optimized
using modified induced fit docking (IFD) workflow.55,56 This method
of ligand-based structure optimization proved to be useful in
homology modeling of GPCRs.41 It combines flexible ligand docking
(Glide algorithm, which uses the GlideScore scoring function)57,58
with protein structure prediction and side chain refinement (Prime
protein prediction program).59,60 Modification of the IFD procedure
consisted of appending another cycle of optimization of the receptor
structure and docking the ligand using Glide XP (extra precision). The
modified procedure allowed the models to improve in terms of
assessing the proper ligand binding mode.
For each optimized receptor, a group of several bioactive
compounds were selected for the purpose of ligand-steered binding
site optimization. Two-dimensional structures of the ligands were
sketched or sourced from the ChEMBL database61 and then converted
to 3D and optimized using LigPrep (OPLS_2001 force field was
applied). If available, ring conformations for cyclic ligands were taken
from their crystal structures deposited in the Cambridge Crystalo-
graphic Data Centre (CCDC) database and retained in the process of
flexible docking.62 The top-scoring ligand−receptor complexes
obtained by way of IFD were inspected visually to check the
compliance with the common binding mode for monoaminergic
receptor ligands. That procedure resulted in models that served as
molecular targets in further docking studies.
Parameters of Glide docking procedure were set to defaults, with
the exception of docking precision set to XP (extra precision) and the
flexible docking option retaining original conformations of amide
bonds. H-bond constraints were applied on Asp3.32 since its
interaction with the protonated nitrogen atom of the ligand is
considered crucial for the monoaminergic GPCRs.63 The centroid of a
grid box (22 × 22 × 22 Å) was located on Asp3.32.
Chemical information for the 9 structurally most representative
examples is presented below, while the data for all the other
compounds are available in Supporting Information.
N-(2-(4-(1,2-Benzothiazol-3-yl)piperazin-1-yl)ethyl)-3-methylben-
zenesulfonamide (1). The title compound was prepared starting from
amine (XV, R = H, X = N, Y = S, n = 0; 0.5 mmol, 131.19 mg) and 3-
methylbenzenesulfonyl chloride (1, 0.5 mmol, 95.32 mg). Yield: 49%
Glide, induced fit docking, LigPrep, Prime, and Protein Preparation
Wizard were implemented in Schrodinger Suite 2009, which was
̈
1
(102 mg), yellowish oil. H NMR (300 MHz, CDCl3): δ 7. 79−7. 86
licensed for Adamed Ltd.
(m, 2H), 7.66−7.72 (m, 2H), 7.31−7.49 (m, 4H), 5.28 (s, 1H), 3.43−
3.48 (m, 4H), 3.01−3.08 (m, 2H), 2.46−2.53 (m, 6H), 2.43 (s, 3H).
13C NMR (75 MHz, CDCl3): δ 163.59, 152.76, 139.37, 139.30,
Chemistry. General Methods. All the reagents were purchased
from Aldrich, Alfa Aesar, and Fluorochem.
1HNMR and 13CNMR spectra were obtained with a Varian BB 200
spectrometer using TMS (0.00 ppm) in chloroform-d1 and DMSO-d6
and were recorded at 300 and 75 MHz, respectively. J values are in
hertz (Hz), and splitting patterns are designated as follows: s (singlet),
bs (broad singlet), d (doublet), t (triplet), and m (multiplet).
The UPLC/MS system consisted of a Waters Aquity UPLC
coupled to a Waters SQD mass spectrometer. Chromatographic
separations were carried out using the Acquity UPLC BEH C18
column, 2.1 × 100 mm and 1.7 μm particle size. The column was
maintained at 60 °C and eluted under gradient conditions: 0−4 min, a
linear gradient from 80−0.1% of eluent A, at a flow rate of 0.5 mL/
min. Eluent A, water/formic acid (0.02%, v/v); and eluent B,
methanol−acidic gradient. Eluent A, water/formic acid/ammonia
solution (0.01%:0.1%, v/v/v); and eluent B, methanol−alkaline
gradient. The UPLC/MS purity of all compounds except 58 and 60
was confirmed to be >95%. The table with the specific purity values for
all compounds in 2 gradients is available in Supporting Information.
Synthetic Procedures. General Procedure for the Synthesis of
Phthalimide Derivatives VII−XIV. A mixture of 10 mmol of
appropriate benzazole/I−III and 10 mmol of appropriate 2-
(bromoalkyl)-1H-isoindoline-1,3(2H)-dione IV−VI, 30 mmol of
potassium carbonate, catalytic amount of potassium iodide, and 20
mL of N,N-dimethylformamide was stirred at room temperature until
the disappearance of starting materials (TLC monitoring). The
reaction mixture was subsequently poured into 50 mL of water, and
the precipitate thus formed was isolated by filtration. The crude
product was suspended in 20 mL of methanol, and then the solid was
filtered off and dried to constant weight. Alternatively, the reaction was
carried out in the mixture of acetonitrile and N,N-dimethylformamide
stirring at 70 °C for 3 days in the presence of 30 mmol of DIPEA and
a catalytic amount of potassium iodide.
133.48, 129.00, 127.88, 127.61, 127.50, 124.16, 123.96, 123.72, 120.61,
55.81, 52.28, 49.81, 39.17, 21.39. MS: 417 [M + H+].
N-(3-(4-(1,2-Benzothiazol-3-yl)piperazin-1-yl)propyl)-3-methyl-
benzenesulfonamide (2). The title compound was prepared starting
from the amine (XVI, R = H, X = N, Y = S, n = 1; 0.5 mmol, 138.20
mg) and 3-methylbenzenesulfonyl chloride (1, 0.5 mmol, 95.32 mg).
1
Yield: 49% (105 mg), yellowish oil. H NMR (300 MHz, DMSO-d6)
δ: 8.03 (t, 2H, J = 7.2 Hz), 7.60−7.39 (m, 6H), 3.35−3.39 (m, 4H),
2.82−2.94 (m, 2H), 2.46−2.50 (m, 4H), 2.38 (s, 3H), 2.30 (t, 2H, J =
7.2 Hz), 1.54 (quintet, 2H, J = 6.9 Hz). 13C NMR (75 MHz, CDCl3):
δ 163.78, 152.18, 139.96, 139.85, 132.42, 128.89, 127.63, 127.57,
127.51, 124.04, 123.98, 123.94, 120.61, 58.18, 52.89, 49.91, 43.96,
23.96, 20.73. MS: 431 [M + H+].
N-(4-(4-(1,2-Benzothiazol-3-yl)piperazin-1-yl)butyl)-3-methylben-
zenesulfonamide (3). The title compound was prepared starting from
the amine (XVII, R = H, X = N, Y = S, n = 2, 0.5 mmol, 145.21 mg)
and 3-methylbenzenesulfonyl chloride (1, 0.5 mmol, 95.32 mg). Yield:
51% (113 mg), yellowish oil. 1H NMR (300 MHz, DMSO-d6) δ: 8.05
(d, 2H, J = 7.2 Hz), 7.41−7.60 (m, 6H), 3.35−3.40 (m, 4H), 2.72−
2.79 (m, 2H), 2.48−2.50 (m, 4H), 2.38 (s, 3H), 2.26−2.28 (m, 2H),
1.44−1.36 (m, 4H). 13C NMR (75 MHz, CDCl3): δ 163.70, 152.61,
139.81, 139.75, 133.21, 128.89, 127.51, 127.31, 126.91, 124.20, 124.07,
123.67, 120.61, 58.21, 48.76, 42.77, 21.34. MS: 445 [M + H+].
N-(3-(4-(Benzo[d]isoxazol-3-yl)piperazin-1-yl)propyl)-3-methyl-
benzenesulfonamide (27). The title compound was prepared starting
from amine (XIX, R = H, X = N, Y = O, n = 1, 0.5 mmol, 130.17 mg)
and 3-methylbenzenesulfonyl chloride (1, 0.5 mmol, 95.32 mg). Yield:
1
47% (120 mg), yellowish oil. H NMR (300 MHz, CDCl3): δ 7.63−
7.68 (m, 2H), 7.32−7.41 (m, 2H), 7.24−7.27 (m, 2H), 7.13−7.18 (m,
1H), 6.99−7.08 (m, 1H), 3.58−3.76 (m, 4H), 3.08−3.14 (m, 2H),
J
dx.doi.org/10.1021/jm401895u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 5. Radioligand Binding Assay Conditions for 5 HT6, 5 HT7, 5 HT2A, and D2 Receptors
radioligand (final
concentration/Kd)
nonspecific
binding
incubation
condition
receptor (source)
assay buffer
5-HT6 (human recombinant, HEK- [3H]LSDm (2.5/2.0 nM)
293 cells)
10 μM
methiothepin
50 mM Tris, 10 mM MgCl2,0.5 mM EDTA, pH 7.4 60 min, 37 °C
5-HT7 (human recombinant, CHO- [3H]LSD (3.0/2.4 nM)
K1 cells)
10 μM
methiothepin
50 mM Tris, 10 mM MgSO4, 0.5 mM EDTA, pH 7.4 120 min, 27 °C
5-HT2A (human recombinant,
CHO-K1 cells)
[3H]ketanserin (1.0/0.5 nM) 100 μM serotonin 50 mM Tris, 4 mM CaCl2, 0.1% ascorbic acid, pH 7.4 60 min, 27 °C
D2 (human recombinant, CHO-K1 [3H]N-methylspiperone
10 μM
(+)-butaclamol
50 mM HEPES, 50 mM NaCl, 5 mM MgCl2, 0.5 mM 60 min, 37 °C
EDTA, pH 7.4
cells)
(0.4/0.2 nM)
2.40−2.55 (m, 6H), 1.64−1.72 (m, 2H). 13C NMR (75 MHz, CDCl3):
δ 163.28, 152.70, 143.28, 139.96, 139.28, 133.28, 130.58, 127.28,
124.03, 118.42, 116.36, 108.78. MS: 515 [M + H+].
was performed using cryopreserved membranes from cells stably
expressing the relevant human receptor under conditions as indicated
in Table 5. Fifty microliters of working solution of the tested
compounds, 50 μL radioligand solution, and 150 μL of diluted
membranes prepared in assay buffer were transferred to 96-well
microplates. These were covered with sealing tape, mixed, and
incubated. The reaction was terminated by rapid filtration through a
UniFilter 96 GF/B filter microplate, and 10 rapid washes with 200 μL
of 50 mM Tris buffer (4 °C, pH 7.4) were performed using a vacuum
manifold and 96-well pipettor. The UniFilter microplates were dried
overnight at 37 °C in a dry incubator. The UniFilter bottoms were
sealed, and 30 μL of liquid scintillator Betaplate Scint (PerkinElmer)
was added to each well. The plates were allowed to equilibrate for 1 h,
and then radioactivity was counted in a MicroBeta TriLux 1450
scintillation counter (PerkinElmer) at approximately 30% efficiency.
Data were fitted to a one-site curve-fitting equation with Prism 5
(GraphPad Software), and Ki values were calculated using the Cheng−
Prusoff equation.
A large number of data points were generated in the screening
experiments: in total, 185 compounds were tested on 4 receptors; of
these, 81 representative compounds, tested on 4 receptors, are shown
in the present article. The Ki values shown in Table 1 are the means of
two independent experiments. Each compound was tested in 10
concentrations from 1 × 10−06 M to 1 × 10−10 M (final
concentration). All of the assays were carried out in duplicate (n = 2).
The assays presented in Table 1 were performed at the Faculty of
Pharmacy, Jagiellonian University Collegium Medicum, except for
M1R and hERG, which were tested by Cerep and ChanTest,
respectively (see below).
Functional Assays for 5 HT6, 5 HT7, 5 HT2A, and D2
Receptors. One millimolar stock solutions of the tested compounds
were prepared in DMSO. Serial dilutions were prepared in 96-well
microplate in assay buffers using an automated pipetting system
epMotion 5070 (Eppendorf). Two independent experiments in
duplicate were performed, and 6 to 10 concentrations were tested.
A cellular aequorin-based functional assay was performed with γ-
irradiated recombinant CHO-K1 cells expressing mitochondrially
targeted Aequorin, human GPCR, and the promiscuous G protein α16
for 5-HT6 and 5-HT2A receptors, Gαqi/5 for D2 receptor
(PerkinElemer). The assay was performed according to the standard
protocol provided by the manufacturer. After thawing, cells were
transferred to assay buffer (DMEM/HAM’s F12 with 0.1% protease-
free BSA) and centrifuged. The cell pellet was resuspended in assay
buffer and coelenterazine h was added at final concentrations of 5 μM.
The cell suspension was incubated at 21 °C, protected from light with
constant agitation for 4 h, and then diluted with assay buffer to a
concentration of 250,000 cells/mL. After 1 h of incubation, 50 μL of
the cell suspension was dispensed using automatic injectors built into
the radiometric and luminescence plate counter MicroBeta2 LumiJET
(PerkinElmer, USA) into white opaque 96-well microplates preloaded
with test compounds. Immediate light emission generated following
calcium mobilization was recorded for 30−60 s. In antagonist mode,
after 15−30 min of incubation the reference agonist was added to the
above assay mix, and light emission was recorded again. Final
concentration of the reference agonist was equal to EC80: 40 nM
serotonin for the 5-HT6 receptor, 30 nM α-methylserotonin for the 5-
HT2A receptor, and 30 nM apomorphine for the D2 receptor.
N-(4-(4-(Benzo[d]isoxazol-3-yl)piperazin-1-yl)butyl)-3-methyl-
benzenesulfonamide (28). The title compound was prepared starting
from the amine (XVIII, R = H, X = N, Y = O, n = 2, 0.5 mmol, 137.18
mg) and 3-methylbenzenesulfonyl chloride (1, 0.5 mmol, 95.32 mg).
Yield: 53% (117 mg), pale yellow solid. 1H NMR (300 MHz, DMSO-
d6): δ 10.61 (bs, 1H), 8.21−8.00 (m, 1H), 7.55−7.65 (m, 4H), 7.42−
7.50 (m, 2H), 7.30−7.36 (m, 1H), 4.08−4.31 (m, 2H), 3.42−3.58 (m,
4H), 3.00−3.25 (m, 4H), 2.70−2.78 (m, 2H), 2.30 (s, 3H), 1.60−1.72
(m, 2H), 1.35−1.48 (m, 2H). 13C NMR (75 MHz, DMSO-d6) δ
163.77, 161.10, 140.30, 139.38, 133.47, 130.91, 129.58, 127.12, 124.05,
117.25, 110.70, 55.82, 50.42, 45.18, 40.77, 21.33, 20.97, 26.30. MS:
445 [M + H+].
N-[2-[4-(6-Fluoro-1,2-benzoxazol-3-yl)piperidin-1-yl]ethyl]-3-
methylbenzenesulfonamide (50). The title compound was prepared
starting from the amine (XXII, R = F, X = C, Y = O, n = 0, 0.5 mmol,
131.66 mg) and 3-methylbenzenesulfonyl chloride (1, 0.5 mmol, 95.32
mg). Yield: 41% (85 mg), yellowish oil. 1H NMR (300 MHz, CDCl3)
δ: 7.66−7.71 (m, 3H), 7.38−7.64 (m, 2H), 7.26−7.36 (m, 1H), 7.08
(t, 1H, J = 9 Hz), 5.27 (s, 1H), 2.98−3.08 (m, 3H), 2.73−2.77 (m,
2H), 2.42−2.48 (m, 5H), 2.07−2.16 (m, 2H), 1.96−2.03 (m, 4H). 13C
NMR (75 MHz, CDCl3): δ 165.62, 163.91, 162.40, 160.27, 139.25,
139.92 132.90, 128.71, 127.20, 123.90, 122.12 (d, J = 11 Hz), 112.62
(d, J = 25.4 Hz), 97.30 (d, J = 26.5 Hz), 55.23, 52.81, 39.53, 34.12,
30.41, 21.30. MS: 418 [M + H+].
N-[3-[4-(6-Fluoro-1,2-benzoxazol-3-yl)piperidin-1-yl]propyl]-3-
methylbenzenesulfonamide (51). The title compound was prepared
starting from the amine (XXI, R = F, X = C, Y = O, n = 1, 0.5 mmol,
138.67 mg) and 3-methylbenzenesulfonyl chloride (1, 0.5 mmol, 95.32
1
mg). Yield: 55% (118 mg), white solid. H NMR (300 MHz, DMSO-
d6): δ 7.80−7.85 (m, 1H), 7.63−7.69 (m, 2H), 7.34−7.41 (m, 2H),
7.20−7.25 (m, 1H), 7.04−7.12 (m, 1H), 3.08−3.12 (m, 3H), 2.96−
3.04 (m, 2H), 2.44−2.50 (m, 2H), 2.40 (s, 3H), 2.04−2.18 (m, 6H),
1.62−1.70 (m, 2H). 13C NMR (75 MHz, CDCl3): δ 165.77, 164,14 (d,
J = 13.8 Hz), 162.45, 160.71, 139.90, 139.19, 133.15, 128.90, 127.31,
124.02, 123.10 (d, J = 11 Hz), 112.78 (d, J = 24.8 Hz), 97.53 (d, J =
26.5 Hz), 58.23, 53.19, 44.28, 34. 25, 30.41, 23.95, 21.36. MS: 432 [M
+ H+].
N-(4-(4-(6-Fluorobenzo[d]isoxazol-3-yl)piperidin-1-yl)butyl)-
benzo[b]thiophene-2-sulfonamide (72). The title compound was
prepared starting from the amine (XX, R = F, X = C, Y = O, n = 2, 0.5
mmol, 145.68 mg) and benzo[b]thiophene-2-sulfonyl chloride (10, 0.5
mmol, 116.35 mg). Yield: 87% (211 mg), yellowish oil. 1H NMR (300
MHz, CDOD3) δ: 7.90−7.98 (m, 4H), 7.45−7.51 (m, 2H), 7.38−7.44
(m, 1H), 7.15−7.23 (m, 1H), 3.46−3.70 (m, 3H), 3.04−3.28 (m, 6H),
2.22−2.42 (m, 4H), 1.82−1.90 (m, 2H), 1.68−1.56 (m, 2H). 19F-
NMR (300 MHz, CDOD3): δ −109.30−109.45 (1F, m). 13C NMR
(75 MHz, CDOD3): δ 166.15, 163.86 (d, J = 13.8 Hz), 162.84, 159.43,
141.59 (d, J = 15 Hz), 137.82, 128.5, 126.9, 125.3, 125.25, 122.9 (d, J
= 11 Hz), 122.3, 116.7, 112.41 (d, J = 25 Hz), 96.81 (d, J = 27 Hz).
MS: 488 [M + H+].
Radioligand Binding Assays for 5-HT6, 5-HT7, 5-HT2A, and D2
Receptors. One millimolar stock solutions of the compounds to be
tested were prepared in DMSO. Serial dilutions of compounds were
prepared in 96-well microplates in assay buffers using an automated
pipetting system (epMotion 5070; Eppendorf). Radioligand binding
K
dx.doi.org/10.1021/jm401895u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
For the 5-HT7 receptor, adenylyl cyclase activity was monitored
using cryopreserved CHO-K1 cells, expressing the human serotonin 5-
HT7 receptor. Thawed cells were resuspended in stimulation buffer
(HBSS, 5 mM HEPES, 0.5 IBMX, and 0.1% BSA at pH 7.4) at
200,000 cells/mL. Ten microliters of cell suspension was added to 10
μL of tested compounds loaded onto a white opaque half area 96-well
microplate. An antagonist response experiment was performed with 10
nM serotonin as the reference agonist, and the agonist and antagonist
were added simultaneously. Cell stimulation was performed for 60 min
at room temperature. After incubation, cAMP measurements were
performed with homogeneous TR-FRET immunoassay using the
LANCE Ultra cAMP kit (PerkinElmer, USA). Ten microliters of Eu-
cAMP Tracer Working Solution and 10 μL of ULight-anti-cAMP
Tracer Working Solution were added, mixed, and incubated for 1 h.
The TR-FRET signal was read on an EnVision microplate reader
(PerkinElmer, USA).
IC50 and EC50 were determined by nonlinear regression analysis
using GraphPad Prism 6.0 software. The log IC50 was used to obtain
the Kb by applying the Cheng−Prusoff approximation. The assays
presented in Table 2 were performed at Faculty of Pharmacy,
Jagiellonian University Collegium Medicum.
Other in Vitro Studies. The extended pharmacological profile in
vitro for compound 72, with respect to GPCRs and SERT, was
outsourced to Cerep (Le Bois l’Eveque, Poitiers, France). An outline
̂
dibenzo[a,d]cyclohepten-5,10-imine (MK-801).64 Briefly, groups of
drug-naive rats (n = 7−8) were transferred in their home cages to the
experimental room 24 h prior to testing and allowed to habituate for
60 min and returned to the colony room. The next day, locomotor
activity was assessed in black octagonal open fields (80 cm in diameter,
30 cm high) under dim light and continuous white noise (65 dB).
Each animal was placed in the center of the open field and allowed to
explore the whole area for 30 min. Subjects did not have visual contact
with other rats. Forward locomotion (cm/30 min) was recorded and
analyzed with the aid of a computerized video tracking system
(Videomot, TSE, Bad Homburg, Germany).
Rats were preadministered the drug or vehicle p.o. 60 min before
the start of the locomotor activity test. Fifteen minutes before the start
of the test, rats were administered MK-801 (0.3 mg/kg i.p.). When
assessing spontaneous locomotor activity, rats were administered
saline 15 min prior to the test.
DOI-Induced Head Twitches. All tests were carried out in a sound-
attenuated experimental room between 10:00 a.m. and 4:00 p.m. DOI-
induced head twitches were scored as described by Schreiber et al.65
Rats were injected with 2,5-dimethoxy-4-iodoamphetamine (DOI 2.5
mg/kg, i.p.) and placed in glass observation cages (25 × 25 × 40 cm3,
W × H × L) with wood chip bedding on the floor. Five minutes later,
head twitches were counted for 5 min by a trained observer. The test
compound or its vehicle was injected p.o. 60 min before the start of
the observation period.
Forced Swimming Test. The procedure used to determine
antidepressant-like activity was based on the technique described
previously.66 Briefly, rats were individually placed in glass cylinders (40
cm in height, 17 cm in diameter) filled with water (temperature: 23
1 °C) at a height that made it impossible to reach the bottom with
hind paws (25 cm) There were two swimming sessions separated by
24 h: an initial 15 min pretest and a 5 min test. The duration of
immobility (s) in the test session was recorded by a blind observer
located in an adjacent room with the aid of a video camera. A rat was
considered immobile when it floated without moving except to keep
its head above the water surface. The test compound or its vehicle was
administered p.o. 60 min before the experiment.
Step-through Passive Avoidance Test. Effects of antipsychotics on
memory function were evaluated as previously described.23,31 Briefly,
the passive avoidance apparatus (PACS-30, Columbus Instruments,
Columbus, OH, USA) comprised four identical stainless-steel cages
with black Plexiglas covers. Each cage consisted of lighted and dark
compartments (23 × 23 × 23 cm3) and a stainless-steel grid floor. The
compartments were separated by an automated sliding door. In the
training (acquisition) session, animals were individually placed in the
lighted compartment and allowed to explore it for 10 s. The sliding
door was then opened, and the step-through latency for animals to
enter the dark compartment was measured (300-s cutoff time). As
soon as the animals entered the dark compartment, the door was
closed. An inescapable foot-shock (0.5 mA pulse constant current for 3
s) was delivered 3 s later through the grid floor with a constant current
shock generator (EACS-30 Columbus Instruments). The test
compound or its vehicle was administered p.o. 60 min before the
start of the training session. All vehicle-treated animals entered the
dark compartment during the training session and received a foot-
shock. Drug-treated animals that did not enter the dark compartment
in the training session were not subjected to the test session. In the
present study, all of the tested animals entered the dark compartment.
The test session was performed 24 h after the training session using
the same paradigm but without the foot-shock. Step-through latencies
for animals to enter the dark compartment were measured with a 300-s
cutoff time. Drug-induced decreases in step-through latencies to enter
the dark compartment in the test session were taken as a measure of
the drug’s “amnesic” effects.
of methodologies is shown in Tables 3 and 4. Further methodological
details are available on the company Web site (www.cerep.fr). Data
from all experiments were analyzed using nonlinear curve fitting
programs, and the results are given as Ki values for binding affinity or
Kb values for antagonist potency.
Blockade of hERG-mediated potassium currents was carried out by
ChanTest (Cleveland, Ohio) and expressed as the mean % of
inhibition at 1.0 × 10−06 M (Table 1) or IC50 value (Table 3). Ability
to block hERG potassium channels was determined using the
electrophysiological method and cloned hERG potassium channels
(KCNH2 gene, expressed in CHO cells) as biological material. The
effects were evaluated using an IonWorks Quattro system (Molecular
Devices Corporation, Union City CA). hERG current was elicited
using a pulse pattern with fixed amplitudes (conditioning prepulse,
−80 mV for 25 ms; test pulse, +40 mV for 80 ms) from a holding
potential of 0 mV. hERG current was measured as a difference
between the peak current at 1 ms after the test step to +40 mV and the
steady-state current at the end of the step to +40 mV. Data acquisition
and analyses were performed using the IonWorks Quattro system
operation software (version 2.0.2; Molecular Devices Corporation,
Union City, CA). Data were corrected for leak current. The hERG
block was calculated as % block = (1 − I TA/IControl) × 100%, where
IControl and ITA were the currents elicited by the test pulse in control
and in the presence of a test compound, respectively.
All assays for the lead compound 72 (shown in Tables 3 and 4)
were performed in 3−4 independent experiments. The Ki and Kb
values were determined from 6 concentrations from 1 × 10−06 M to 1
× 10−10 M. All of the assays were carried out in duplicate (n = 2).
In Vivo Studies. Animals. Drug-naive male Wistar rats (Charles
River, Sulzfeld, Germany) weighing 200−225 g on arrival were used
for behavioral experiments. Animals were supplied by the breeder 2−3
weeks before experimentation. Rats were housed four per plastic cage
and kept in a room with constant environmental conditions (22
1
°C, relative humidity 60%, a 12:12 light-dark cycle with lights on at
7:00 a.m.). During this time, the rats were weighed and handled
several times. Tap water and standard laboratory chow (Labofeed H,
WPIK, Kcynia, Poland) were available ad libitum. All tests were carried
out in a sound-attenuated experimental room between 09:00 a.m. and
3:00 p.m. All procedures for the treatment of rats in the present study
were as humane as possible and were reviewed and approved by the
institutional ethics committee. Procedures conformed to animal care
standards laid down in Polish regulations and complied with
international European guidelines (Directive no. 86/609/EEC).
MK-801-Induced Hyperlocomotion. Antipsychotic-like activity was
assessed by the inhibition of hyperactivity elicited by the NMDA
receptor antagonist, [5R,10S]-[+]-5-methyl-10,11-dihydro-5H-
Catalepsy Bar Test. Catalepsy was assessed using the bar test. Each
rat was placed on a clean, smooth table with a wooden bar (2 × 3 × 25
cm3, H × W × L) suspended 10 cm above the working surface. The
animal’s hindlimbs were freely placed on the table, the tail laid out to
the back, and the forelimbs gently placed over the bar. The length of
time the animal touched the bar with both front paws was measured
L
dx.doi.org/10.1021/jm401895u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
up to a preset cutoff time of 180 s. Results of each trial were scored as
follows: 0 for holding the position for <15 s, 1 for holding it for 15−
29.9 s, 2 for holding it for 30−59.9 s, and a maximum score of 3 for
staying on the bar for >60 s. The minimum cataleptogenic dose was
defined as the lowest dose inducing a mean catalepsy score of >1.67
Catalepsy was scored 30, 60, and 120 min after administration of
vehicle or a test drug (p.o.).
Acetylcholine Release in the Cortex: Pharmacological Characterization
of 4-(2-Bromo-6-pyrrolidin-1-ylpyridine-4-sulfonyl)phenylamine, a
Potent and Selective 5-HT6 Receptor Antagonist. J. Med. Chem.
2003, 46, 1273−1276.
(8) Liu, K. G.; Robichaud, A. J. 5-HT6 Antagonists as Potential
Treatment for Cognitive Dysfunction. Drug Dev. Res. 2009, 70, 145−
168.
(9) Wesołowska, A. Potential Role of the 5-HT6 Receptor in
Depression and Anxiety: An Overview of Preclinical Data. Pharmacol.
Rep. 2010, 62, 564−577.
(10) Wesołowska, A.; Nikiforuk, A. Effects of the Brain-Penetrant
and Selective 5-HT6 Receptor Antagonist SB-399885 in Animal
Models of Anxiety and Depression. Neuropharmacology 2007, 52,
1274−1283.
(11) Woolley, M. L.; Marsden, C. A.; Fone, K. C. 5-ht6 Receptors.
Curr. Drug Targets 2004, 3, 59−79.
(12) Maher-Edwards, G.; Zvartau-Hind, M.; Hunter, A. J.; Gold, M.;
Hopton, G.; Jacobs, G.; Davy, M.; Williams, P. Double-Blind,
Controlled Phase II Study of a 5-HT6 Receptor Antagonist, SB-
742457, in Alzheimer’s Disease. Curr. Alzheimer Res. 2010, 7, 374−
385.
(13) H. Lundbeck A/S. Lundbeck’s Lu AE58054 Meets Primary
Endpoint in Large Placebo-Controlled Clinical Proof of Concept
Study in People with Alzheimer’s Disease. Corp. Release 472, 1−4.
Published Online: May 29, 2012. http://investor.lundbeck.com/
releasedetail.cfm?ReleaseID=677436 (accessed Oct 23, 2012).
(14) Wesołowska, A.; Nikiforuk, A.; Stachowicz, K. Potential
Anxiolytic and Antidepressant Effects of the Selective 5-HT7 Receptor
Antagonist SB 269970 after Intrahippocampal Administration to Rats.
Eur. J. Pharmacol. 2006, 553, 185−190.
(15) Galici, R.; Boggs, J. D.; Miller, K. L.; Bonaventure, P.; Atack, J.
R. Effects of SB-269970, a 5-HT7 Receptor Antagonist, in Mouse
Models Predictive of Antipsychotic-Like Activity. Behav. Pharmacol.
2008, 19, 153−159.
(16) Waters, K. A.; Stean, T. O.; Hammond, B.; Virley, D. J.; Upton,
N.; Kew, J. N. C.; Hussain, I. Effects of the Selective 5-HT(7)
Receptor Antagonist SB-269970 in Animal Models of Psychosis and
Cognition. Behav. Brain Res. 2012, 228, 211−218.
(17) Kim, D. H.; Maneen, M. J.; Stahl, S. M. Building a Better
Antipsychotic: Receptor Targets for the Treatment of Multiple
Symptom Dimensions of Schizophrenia. Neurotherapeutics 2009, 6,
78−85.
(18) Morphy, R.; Rankovic, Z. Designed Multiple Ligands. An
Emerging Drug Discovery Paradigm. J. Med. Chem. 2005, 48, 6523−
6543.
(19) Reynolds, G. P.; Kirk, S. L. Metabolic Side Effects of
Antipsychotic Drug Treatment: Pharmacological Mechanisms. Phar-
macol. Ther. 2010, 125, 169−179.
(20) Bai, Y. M.; Chen, T.-T.; Liou, Y.-J.; Hong, C.-J.; Tsai, S.-J.
Association between HTR2C Polymorphisms and Metabolic Syn-
drome in Patients with Schizophrenia Treated with Atypical
Antipsychotics. Schizophr. Res. 2011, 125, 179−186.
ASSOCIATED CONTENT
* Supporting Information
Additional binding site and ligand binding mode descriptions,
extended chemical information and compound characterization,
and detailed purity data for all compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
*Phone: 48-697064058. Fax: 48-126205458. E-mail: marcin.
kolaczkowski@adamed.com.pl; marcin.kolaczkowski@uj.edu.pl.
Notes
The authors declare the following competing financial
interest(s): Marcin Kolaczkowski is an employee of Adamed
Ltd.
■
ACKNOWLEDGMENTS
■
This study was financially supported by Adamed Ltd. and the
National Centre for Research and Development (NCBR)
Grant No. KB/88/12655/IT1-C/U/08.
ABBREVIATIONS USED
■
BPSD, behavioral and psychological symptoms of dementia; 5-
HT2AR, 5-HT2A serotonin receptor; 5-HT6R, 5-HT6 serotonin
receptor; 5-HT7R, 5-HT7 serotonin receptor; D2R, D2
dopamine receptor; M1R, M1 muscarinic receptor; β2R, β2
adrenergic receptor; TMH, transmembrane helix; ECL,
extracellular loop; ICL, intracellular loop; IFD, induced fit
docking
REFERENCES
■
(1) Liperoti, R.; Pedone, C.; Corsonello, A. Antipsychotics for the
Treatment of Behavioral and Psychological Symptoms of Dementia
(BPSD). Curr. Neuropharmacol. 2008, 6, 117−124.
(2) Jeste, D. V.; Blazer, D.; Casey, D.; Meeks, T.; Salzman, C.;
Schneider, L.; Tariot, P.; Yaffe, K. ACNP White Paper: Update on Use
of Antipsychotic Drugs in Elderly Persons with Dementia. Neuro-
psychopharmacology 2008, 33, 957−970.
(3) Carson, S.; McDonagh, M. S.; Peterson, K. A Systematic Review
of the Efficacy and Safety of Atypical Antipsychotics in Patients with
Psychological and Behavioral Symptoms of Dementia. J. Am. Geriatr.
Soc. 2006, 54, 354−361.
(4) U.S. Food and Drug Administration. Public Health Advisory:
Deaths with Antipsychotics in Elderly Patients with Behavioral
Disturbances; Center for Drug Evaluation and Research: Rockville,
MD, Apr 11, 2005. www.fda.gov/Drugs/DrugSafety (accessed Nov 28,
2013).
(5) U.S. Food and Drug Administration. Information on Conventional
Antipsychotics. Center for Drug Evaluation and Research: Rockville,
MD, Jun 16, 2008. www.fda.gov/Drugs/DrugSafety (accessed Nov 28,
2013).
(6) Schulze, J.; van den Bussche, H.; Glaeske, G.; Kaduszkiewicz, H.;
Wiese, B.; Hoffmann, F. Impact of Safety Warnings on Antipsychotic
Prescriptions in Dementia: Nothing Has Changed but the Years and
the Substances. Eur. Neuropsychopharmacol. 2013, 23, 1034−1042.
(7) Riemer, C.; Borroni, E.; Levet-Trafit, B.; Martin, J. R.; Poli, S.;
(21) Rodefer, J. S.; Nguyen, T. N.; Karlsson, J.-J.; Arnt, J. Reversal of
Subchronic PCP-Induced Deficits in Attentional Set Shifting in Rats
by Sertindole and a 5-HT6 Receptor Antagonist: Comparison among
Antipsychotics. Neuropsychopharmacology 2008, 33, 2657−2666.
́
(22) Kołaczkowski, M.; Kowalski, P.; Jaskowska, J.; Marcinkowska,
M.; Mitka, K.; Bucki, A.; Wesołowska, A.; Pawłowski, M.
Arylsulfonamides for the Treatment of CNS Diseases. PCT Int.
Appl. WO 2012035123 (A1), Mar 22, 2012.
(23) Kołaczkowski, M.; Mierzejewski, P.; Bienkowski, P.;
Wesołowska, A.; Newman-Tancredi, A. Antipsychotic, Antidepressant,
and Cognitive-Impairment Properties of Antipsychotics: Rat Profile
and Implications for Behavioral and Psychological Symptoms of
Dementia. Naunyn−Schmiedeberg’s Arch. Pharmacol. 2014, 387, 545−
557.
(24) Kołaczkowski, M.; Nowak, M.; Pawłowski, M.; Bojarski, A. J.
Receptor-Based Pharmacophores for Serotonin 5-HT7R Antagonists-
Implications to Selectivity. J. Med. Chem. 2006, 49, 6732−6741.
Porter, R. H. P.; Bos, M. Influence of the 5-HT6 Receptor on
̈
M
dx.doi.org/10.1021/jm401895u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(25) Lemke, T. L.; Williams, D. A. Foye’s Principles of Medicinal
Chemistry; Lippincott Williams & Wilkins: Philadelphia, PA, 2008.
(26) Haga, K.; Kruse, A. C.; Asada, H.; Yurugi-Kobayashi, T.;
Shiroishi, M.; Zhang, C.; Weis, W. I.; Okada, T.; Kobilka, B. K.; Haga,
T.; Kobayashi, T. Structure of the Human M2Muscarinic Acetylcho-
line Receptor Bound to an Antagonist. Nature 2012, 482, 547−551.
(27) Kruse, A. C.; Hu, J.; Pan, A. C.; Arlow, D. H.; Rosenbaum, D.
M.; Rosemond, E.; Green, H. F.; Liu, T.; Chae, P. S.; Dror, R. O.;
Shaw, D. E.; Weis, W. I.; Wess, J.; Kobilka, B. K. Structure and
Dynamics of the M3Muscarinic Acetylcholine Receptor. Nature 2012,
482, 552−556.
(28) Pearlstein, R. A.; Vaz, R. J.; Kang, J.; Chen, X.-L.;
Preobrazhenskaya, M.; Shchekotikhin, A. E.; Korolev, A. M.;
Lysenkova, L. N.; Miroshnikova, O. V.; Hendrix, J.; Rampe, D.
Characterization of HERG Potassium Channel Inhibition Using
CoMSiA 3D QSAR and Homology Modeling Approaches. Bioorg.
Med. Chem. Lett. 2003, 13, 1829−1835.
(39) Kirk, S. L.; Glazebrook, J.; Grayson, B.; Neill, J. C.; Reynolds, G.
P. Olanzapine-Induced Weight Gain in the Rat: Role of 5-HT2C and
Histamine H1 Receptors. Psychopharmacology 2009, 207, 119−125.
(40) De Deyn, P.; Jeste, D. V.; Swanink, R.; Kostic, D.; Breder, C.;
Carson, W. H.; Iwamoto, T. Aripiprazole for the Treatment of
Psychosis in Patients with Alzheimer’s Disease: A Randomized,
Placebo-Controlled Study. J. Clin. Psychopharmacol. 2005, 25, 463−
467.
(41) Kołaczkowski, M.; Bucki, A.; Feder, M.; Pawłowski, M. Ligand-
Optimized Homology Models of D1 and D2 Dopamine Receptors:
Application for Virtual Screening. J. Chem. Inf. Model. 2013, 53, 638−
648.
́ ́
(42) Kowalski, P.; Jaskowska, J.; Bojarski, A. J.; Duszynska, B.; Bucki,
A.; Kołaczkowski, M. Evaluation of 1-Arylpiperazine Derivative of
Hydroxybenzamides as 5-HT 1A and 5-HT7 Serotonin Receptor
Ligands: An Experimental and Molecular Modeling Approach. J.
Heterocycl. Chem. 2011, 48, 192−198.
(43) Aranda, R.; Villalba, K.; Ravina, E.; Masaguer, C. F.; Brea, J.;
̃
(29) Osterberg, F.; Aqvist, J. Exploring Blocker Binding to a
Homology Model of the Open hERG K+ Channel Using Docking and
Molecular Dynamics Methods. FEBS Lett. 2005, 579, 2939−2944.
(30) Forbes, I. T.; Dabbs, S.; Duckworth, D. M.; Jennings, A. J.; King,
F. D.; Lovell, P. J.; Brown, A. M.; Collin, L.; Hagan, J. J.; Middlemiss,
D. N.; Riley, G. J.; Thomas, D. R.; Upton, N. (R)-3,N-Dimethyl-N-[1-
methyl-3-(4-methyl-piperidin-1-Yl) propyl]benzenesulfonamide: The
First Selective 5-HT7 Receptor Antagonist. J. Med. Chem. 1998, 41,
655−657.
́
Areias, F.; Domínguez, E.; Selent, J.; Lopez, L.; Sanz, F.; Pastor, M.;
Loza, M. I. Synthesis, Binding Affinity, and Molecular Docking
Analysis of New Benzofuranone Derivatives as Potential Antipsy-
chotics. J. Med. Chem. 2008, 51, 6085−6094.
(44) Dukat, M.; Mosier, P. D.; Kolanos, R.; Roth, B. L.; Glennon, R.
A. Binding of Serotonin and N1-Benzenesulfonyltryptamine-Related
Analogs at Human 5-HT6 Serotonin Receptors: Receptor Modeling
Studies. J. Med. Chem. 2008, 51, 603−611.
(45) De la Fuente, T.; Martín-Fontecha, M.; Sallander, J.; Benhamu,
́
́
(31) Kołaczkowski, M.; Mierzejewski, P.; Bienkowski, P.;
B.; Campillo, M.; Medina, R. A.; Pellissier, L. P.; Claeysen, S.; Dumuis,
Wesołowska, A.; Newman-Tancredi, A. ADN-1184 a Monoaminergic
Ligand with 5-HT6/7 Receptor Antagonist Activity: Pharmacological
Profile and Potential Therapeutic Utility. Br. J. Pharmacol. 2014, 171,
973−984.
́
A.; Pardo, L.; Lopez-Rodríguez, M. L. Benzimidazole Derivatives as
New Serotonin 5-HT6 Receptor Antagonists. Molecular Mechanisms
of Receptor Inactivation. J. Med. Chem. 2010, 53, 1357−1369.
(46) Van Loevezijn, A.; Venhorst, J.; Iwema Bakker, W. I.; de Korte,
C. G.; de Looff, W.; Verhoog, S.; van Wees, J.-W.; van Hoeve, M.; van
de Woestijne, R. P.; van der Neut, M. A. W.; Borst, A. J. M.; van
Dongen, M. J. P.; de Bruin, N. M. W. J.; Keizer, H. G.; Kruse, C. G.
N′-(Arylsulfonyl)pyrazoline-1-carboxamidines as Novel, Neutral 5-
Hydroxytryptamine 6 Receptor (5-HT6R) Antagonists with Unique
Structural Features. J. Med. Chem. 2011, 54, 7030−7054.
(32) Millan, M. J.; Brocco, M.; Rivet, J. M.; Audinot, V.; Newman-
Tancredi, A.; Maiofiss, L.; Queriaux, S.; Despaux, N.; Peglion, J. L.;
Dekeyne, A. S18327 (1-[2-[4-(6-Fluoro-1, 2-benzisoxazol-3-Yl)-
piperid-1-yl]ethyl]3-phenyl imidazolin-2-one), a Novel, Potential
Antipsychotic Displaying Marked Antagonist Properties at Alpha(1)-
and Alpha(2)-Adrenergic Receptors: II. Functional Profile and a
Multiparametric Comparison with Haloperidol, Clozapine, and 11
Other Antipsychotic Agents. J. Pharmacol. Exp. Ther. 2000, 292, 54−
66.
(33) Kongsamut, S.; Kang, J.; Chen, X.-L.; Roehr, J.; Rampe, D. A
Comparison of the Receptor Binding and HERG Channel Affinities
for a Series of Antipsychotic Drugs. Eur. J. Pharmacol. 2002, 450, 37−
41.
(47) Medina, R. A.; Sallander, J.; Benhamu, B.; Porras, E.; Campillo,
́
́
M.; Pardo, L.; Lopez-Rodríguez, M. L. Synthesis of New Serotonin 5-
HT7 Receptor Ligands. Determinants of 5-HT7/5-HT1A Receptor
Selectivity. J. Med. Chem. 2009, 52, 2384−2392.
(48) Farid, R.; Day, T.; Friesner, R. A.; Pearlstein, R. A. New Insights
about HERG Blockade Obtained from Protein Modeling, Potential
Energy Mapping, and Docking Studies. Bioorg. Med. Chem. 2006, 14,
3160−3173.
(34) Millan, M. J.; Di Cara, B.; Dekeyne, A.; Panayi, F.; De Groote,
L.; Sicard, D.; Cistarelli, L.; Billiras, R.; Gobert, A. Selective Blockade
of Dopamine D(3) versus D(2) Receptors Enhances Frontocortical
Cholinergic Transmission and Social Memory in Rats: A Parallel
Neurochemical and Behavioral Analysis. J. Neurochem. 2007, 100,
1047−1061.
(49) Cherezov, V.; Rosenbaum, D. M.; Hanson, M. A.; Rasmussen, S.
G. F.; Thian, F. S.; Kobilka, T. S.; Choi, H.-J.; Kuhn, P.; Weis, W. I.;
Kobilka, B. K.; Stevens, R. C. High-Resolution Crystal Structure of an
Engineered Human beta2-Adrenergic G Protein-Coupled Receptor.
Science 2007, 318, 1258−1265.
(50) Kurowski, M. A.; Bujnicki, J. M. GeneSilico Protein Structure
Prediction Meta-Server. Nucleic Acids Res. 2003, 31, 3305−3307.
(51) Jain, E.; Bairoch, A.; Duvaud, S.; Phan, I.; Redaschi, N.; Suzek,
B. E.; Martin, M. J.; McGarvey, P.; Gasteiger, E. Infrastructure for the
Life Sciences: Design and Implementation of the UniProt Website.
BMC Bioinformatics 2009, 10, 136.
(52) Arnold, K.; Bordoli, L.; Kopp, J.; Schwede, T. The SWISS-
MODEL Workspace: A Web-Based Environment for Protein Structure
Homology Modelling. Bioinformatics 2006, 22, 195−201.
(53) Guex, N.; Peitsch, M. C. SWISS-MODEL and the Swiss-
PdbViewer: An Environment for Comparative Protein Modeling.
Electrophoresis 1997, 18, 2714−2723.
(54) Schwede, T.; Kopp, J.; Guex, N.; Peitsch, M. C. SWISS-
MODEL: An Automated Protein Homology-Modeling Server. Nucleic
Acids Res. 2003, 31, 3381−3385.
(35) Franberg, O.; Marcus, M. M.; Svensson, T. H. Involvement of 5-
̊
HT2A Receptor and α2-Adrenoceptor Blockade in the Asenapine-
Induced Elevation of Prefrontal Cortical Monoamine Outflow. Synapse
2012, 66, 650−660.
(36) Wang, L. Y.; Shofer, J. B.; Rohde, K.; Hart, K. L.; Hoff, D. J.;
McFall, Y. H.; Raskind, M. A.; Peskind, E. R. Prazosin for the
Treatment of Behavioral Symptoms in Patients with Alzheimer
Disease with Agitation and Aggression. Am. J. Geriatr. Psychiatry
2009, 17, 744−751.
(37) Szot, P.; White, S. S.; Greenup, J. L.; Leverenz, J. B.; Peskind, E.
R.; Raskind, M. A. Compensatory Changes in the Noradrenergic
Nervous System in the Locus Ceruleus and Hippocampus of
Postmortem Subjects with Alzheimer’s Disease and Dementia with
Lewy Bodies. J. Neurosci. 2006, 26, 467−478.
(38) Reiner, P. B.; Kamondi, A. Mechanisms of Antihistamine-
Induced Sedation in the Human Brain: H1 Receptor Activation
Reduces a Background Leakage Potassium Current. Neuroscience 1994,
59, 579−588.
(55) Sherman, W.; Beard, H. S.; Farid, R. Use of an Induced Fit
Receptor Structure in Virtual Screening. Chem. Biol. Drug Des. 2006,
67, 83−84.
N
dx.doi.org/10.1021/jm401895u | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(56) Sherman, W.; Day, T.; Jacobson, M. P.; Friesner, R. A.; Farid, R.
Novel Procedure for Modeling Ligand/receptor Induced Fit Effects. J.
Med. Chem. 2006, 49, 534−553.
(57) Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.;
Klicic, J. J.; Mainz, D. T.; Repasky, M. P.; Knoll, E. H.; Shelley, M.;
Perry, J. K.; Shaw, D. E.; Francis, P.; Shenkin, P. S. Glide: A New
Approach for Rapid, Accurate Docking and Scoring. 1. Method and
Assessment of Docking Accuracy. J. Med. Chem. 2004, 47, 1739−1749.
(58) Friesner, R. A.; Murphy, R. B.; Repasky, M. P.; Frye, L. L.;
Greenwood, J. R.; Halgren, T. A.; Sanschagrin, P. C.; Mainz, D. T.
Extra Precision Glide: Docking and Scoring Incorporating a Model of
Hydrophobic Enclosure for Protein-Ligand Complexes. J. Med. Chem.
2006, 49, 6177−6196.
(59) Jacobson, M. P.; Friesner, R. A.; Xiang, Z.; Honig, B. On the
Role of the Crystal Environment in Determining Protein Side-Chain
Conformations. J. Mol. Biol. 2002, 320, 597−608.
(60) Jacobson, M. P.; Pincus, D. L.; Rapp, C. S.; Day, T. J. F.; Honig,
B.; Shaw, D. E.; Friesner, R. A. A Hierarchical Approach to All-Atom
Protein Loop Prediction. Proteins 2004, 55, 351−367.
(61) Gaulton, A.; Bellis, L. J.; Bento, A. P.; Chambers, J.; Davies, M.;
Hersey, A.; Light, Y.; McGlinchey, S.; Michalovich, D.; Al-Lazikani, B.;
Overington, J. P. ChEMBL: A Large-Scale Bioactivity Database for
Drug Discovery. Nucleic Acids Res. 2012, 40, D1100−1107.
(62) Allen, F. H. The Cambridge Structural Database: A Quarter of a
Million Crystal Structures and Rising. Acta Crystallogr., Sect. B 2002,
58, 380−388.
(63) Strader, C. D.; Sigal, I. S.; Candelore, M. R.; Rands, E.; Hill, W.
S.; Dixon, R. A. Conserved Aspartic Acid Residues 79 and 113 of the
Beta-Adrenergic Receptor Have Different Roles in Receptor Function.
J. Biol. Chem. 1988, 263, 10267−10271.
(64) Andine,
́
P.; Widermark, N.; Axelsson, R.; Nyberg, G.; Olofsson,
U.; Martensson, E.; Sandberg, M. Characterization of MK-801-
̊
Induced Behavior as a Putative Rat Model of Psychosis. J. Pharmacol.
Exp. Ther. 1999, 290, 1393−1408.
(65) Schreiber, R.; Brocco, M.; Audinot, V.; Gobert, A.; Veiga, S.;
Millan, M. J. (1-(2,5-Dimethoxy-4 iodophenyl)-2-aminopropane)-
Induced Head- Twitches in the Rat Are Mediated by 5-Hydroxytrypt-
amine (5-HT) 2A Receptors: Modulation by Novel 5-HT2A/2C
Antagonists, D1 Antagonists and 5-HT1A Agonists. J. Pharmacol. Exp.
Ther. 1995, 273, 101−112.
(66) Porsolt, R. D.; Anton, G.; Blavet, N.; Jalfre, M. Behavioral
Despair in Rats: A New Model Sensitive to Antidepressant
Treatments. Eur. J. Pharmacol. 1978, 47, 379−391.
(67) Ogren, S. O.; Hall, H.; Kohler, C.; Magnusson, O.; Sjostrand, S.
̈
̈
E. The Selective Dopamine D2 Receptor Antagonist Raclopride
Discriminates between Dopamine-Mediated Motor Functions. Psycho-
pharmacology 1986, 90, 287−294.
O
dx.doi.org/10.1021/jm401895u | J. Med. Chem. XXXX, XXX, XXX−XXX