Identification of tris-(phenylalkyl)amines as new selective h5-HT2B receptor antagonists

Article abstract of DOI:10.1039/c4md00418c

A series of tris-(phenylalkyl)amines was synthesized and evaluated for affinity to human 5-HT₂ receptors. In general, the compounds displayed high affinity (4 of 11 analogs had K_i values < 10 nM) and good selectivity for the 5-HT_2B receptor vs. other 5-HT₂ receptors. Functional assays revealed that the compounds are 5-HT_2B antagonists. This journal is

Full text of DOI:10.1039/c4md00418c

MedChemComm
View Article Online
View Journal
CONCISE ARTICLE
Identification of tris-IJphenylalkyl)amines as new
selective h5-HT2B receptor antagonists†
Cite this: DOI: 10.1039/c4md00418c
ab
Shashikanth Ponnala,a Nirav Kapadiaab and Wayne Wesley Harding
*
Received 18th September 2014,
Accepted 15th December 2014
A series of tris-IJphenylalkyl)amines was synthesized and evaluated for affinity to human 5-HT2 receptors. In
general, the compounds displayed high affinity (4 of 11 analogs had Ki values < 10 nM) and good selectivity
for the 5-HT2B receptor vs. other 5-HT2 receptors. Functional assays revealed that the compounds are
5-HT2B antagonists.
DOI: 10.1039/c4md00418c
www.rsc.org/medchemcomm
The 5-HT2B receptor is involved in regulation of the CNS,
gastric and intestinal motility and cardiovascular function.
5-HT2B antagonists have been explored as potential
pharmacotherapies for migraine,1 irritable bowel syndrome,2–4
pulmonary hypertension5 and heart failure.6 5-HT2B receptor agonists
display antidepressant activity and 5-HT2B receptor activation is
required for antidepressant actions of selective serotonin
reuptake inhibitors (SSRI's).7 However, 5-HT2B agonism is
known to be associated with the development of valvular
heart disease (VHD) and as such is regarded as an anti-target
in most drug discovery programs.8–10
Despite the promise of 5-HT2B antagonists as useful thera-
peutics, there are no 5-HT2B antagonists that are clinically
approved for the clinical indications mentioned previously.
This is partly because many known ligands are not truly
5-HT2B selective IJ5-HT2B ligands often also have affinity for
the related 5-HT2A and 5-HT2C receptors) and even when
selective there are issues related to ADME properties of the
compounds that prohibit clinical translational studies. Fig. 1
shows some selective 5-HT2B antagonists that are commer-
cially available; these compounds are predominantly used as
biological tools.11–14 The identification of new 5-HT2B pre-
ferring scaffolds is critical in the pursuit of novel chemical
entities that may be developed as useful 5-HT2B antagonist
therapeutics. We describe herein the serendipitous discovery
of a new series of ligands bearing a tris-IJphenylalkyl)amine
scaffold with high affinity and selectivity for the 5-HT2B
receptor. The ease of synthesis of this scaffold makes it
particularly attractive for further structure-activity work to
optimize 5-HT2B affinity, selectivity and antagonist activity
in the quest for 5-HT2B antagonist drugs.
Our research team has been investigating aporphines
based on the natural product nantenine (see inset, Scheme 1)
as ligands for the 5-HT2A receptor and this program has
resulted in the identification of a number of new aporphine-
based 5-HT2A antagonists.15–17 As part of those efforts, we
decided to investigate the importance of molecular rigidity of
the aporphine template on 5-HT2A antagonism. In that
regard, we decided to explore whether the replacement of the
N-methyl group of nantenine with an N-phenylalkyl moiety
and concomitant increase in flexibility would affect 5-HT2A
antagonist activity. We considered that this approach might
allow the ligands multiple possibilities for interaction of the
receptor with N-phenylalkyl groups which seem to be impor-
tant pharmacophoric recognition elements in 5-HT2A ligands,
thus leading to increase in 5-HT2A receptor affinity. Addition-
ally, we reasoned that this approach could lead to more
diverse series of analogs and a much shorter synthetic route
to the compounds, precluding laborious synthesis of the
aporphine template. Thus we engaged the synthesis of com-
pounds 6a–6k as shown in Scheme 1.
The preparation of analogs 6a–6k was readily accom-
plished in 3 steps. In the first step, the commercially avail-
able amine 1 was coupled to acids 2a and 2b to furnish com-
pounds 3a and 3b. Reduction of amides 3a and 3b with
LiAlH4 gave the secondary amines 4a and 4b. Reductive
amination of secondary amines 4a and 4b with various alde-
hydes (5) provided the target molecules 6a–6k (see ESI† for
experimental procedures).
a Chemistry Dept., Hunter College, CUNY, 695 Park Avenue, NY 10065, USA.
E-mail: whardi@hunter.cuny.edu; Fax: +1 212 772 5332; Tel: +1 212 772 5359
b The Graduate Center, City University of New York, 365 5th Ave., NY 10016, USA
† Electronic supplementary information (ESI) available: Typical binding curve
for 5-HT2A, 5-HT2B and 5-HT2C binding experiments (shown for compound 6h
only) and 5-HT2B antagonist functional data. Typical synthetic procedures and
NMR spectra for analogs 6a–6k. See DOI: 10.1039/c4md00418c
Fig. 1 Selective 5-HT2B antagonists.
This journal is © The Royal Society of Chemistry 2015
Med. Chem. Commun.

View Article Online
Concise Article
MedChemComm
Scheme 1 Reagents and conditions: (i) 2, CDI, THF, 0 °C–rt, 12 h; (ii) LiAIH4, THF, 0 °C–rt, 12 h; (iii) 5, NaBHIJOAc)3, DCM, rt, 12 h.
Analogs 6a–6k were submitted to the Psychoactive Drug
Screening Program (PDSP)18 for evaluation of their affinity
to 5-HT2 receptors. Here, the submitted compounds were
first screened in a primary radioligand binding assay (in qua-
druplicate) at a concentration of 10 μM at the three human
5-HT2 receptor sites. Compounds which displayed a mini-
mum of 50% inhibition for a particular receptor in this pre-
liminary assay were then evaluated in secondary radioligand
binding assays (11 concentrations; each in triplicate) to deter-
mine Ki values. These Ki values are compiled in Table 1.
Complete details of the assays performed may be found in
the PDSP assay protocol book (http://pdsp.med.unc.edu/
PDSP%20Protocols%20II%202013-03-28.pdf).
As mentioned before, the motivation behind the design
and synthesis of this set of compounds was due to our inter-
est in identification of 5-HT2A receptor ligands and so we
were a bit surprised at the outcome of the assays. In general,
this series of compounds displays high affinity for the 5-HT2B
receptor and a range of selectivity (from 2 to almost 90-fold)
vs. the 5-HT2A and 5-HT2C subtypes. Most of the analogs had
5-HT2B affinities that were similar or superior to the standard
ligand used – SB206553, which had 5-HT2B affinity of 21 nM
(see ESI† for typical binding curve).
Compound 6a showed good affinity (59 nM, see Table 1)
for the 5-HT2B receptor. This affinity improved upon addition
of one or two methylene groups between the nitrogen atom
Table 1 Binding affinities and 5-HT2B selectivities of compounds 6a–6k at h5-HT2 receptors
Ki (nM)a
Selectivity
Cmpd.
R
n
m
5-HT2A
5-HT2B
5-HT2C
5-HT2A/5-HT2B
5-HT2C/5-HT2B
6a
6b
6c
6d
6e
6f
6g
6h
Phenyl
Phenyl
Phenyl
0
1
2
2
2
2
2
2
2
2
0
1
1
1
1
1
1
1
1
0
0
0
3531 460
1472 190
165 25
140 15
200 22
267 34
919 120
146 19
1507 190
2234 290
226 29
15
59 8.8
17 2.5
26 2.3
5.8 0.6
4.6 0.5
6.8 0.7
36 4.6
4.1 0.5
59 7.6
231 25
21 2.3
1091 140
690 100
399 75
123 16
108 14
206 27
273 35
194 25
103 19
nab
60
87
6
19
41
15
21
24
30
8
47
2
—
2-Methoxyphenyl
3-Methoxyphenyl
4-Methoxyphenyl
2,5-Dimethoxyphenyl
3,4,5-Trimethoxyphenyl
3-Methoxyphenyl
3,4,5-Trimethoxyphenyl
(Z)-Styryl
24
43
39
26
36
26
10
11
6i
6j
6k
241 45
12
Clozapine
SB206553
Ritanserin
21
1.8
a
b
Radioligands are [3H]ketanserin, [3H]LSD and [3H]mesulergine for 5-HT2A, 5-HT2B and 5-HT2C respectively. na – not active defined as:
% inhibition at 10 μM < 50% in primary assay.
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2015

View Article Online
MedChemComm
Concise Article
and the benzene ring (i.e. compounds 6b and 6c; 17 and 26 nM
respectively). In the case of compound 6b, as compared to
compound 6a, the increase in 5-HT2B affinity was accompa-
nied by increases in 5-HT2A and 5-HT2C affinities as well.
However, the selectivity for 5-HT2B vs. 5-HT2A and 5-HT2C
receptors improved (from 60 and 19-fold respectively for 6a
to 87 and 41-fold for 6b). For compound 6c, there was also
an increase in 5-HT2A and 5-HT2C affinities as compared to
6a. However, the selectivity for 5-HT2B was lower than both
6a and 6b (6 and 15-fold respectively for 5-HT2A and 5-HT2C
selectivities). Thus it appears that a 2 carbon chain between
the nitrogen atom and the unsubstituted aryl ring is well
tolerated for 5-HT2B selectivity. As compared to compound
6c, the 2-, 3- and 4-methoxy derivatives 6d–6f showed higher
affinity for the 5-HT2B receptor (5.8, 4.6 and 6.8 nM respec-
tively), indicating excellent tolerance for these substituents on
the scaffold. In general it appears that the position of the
methoxy group on the aromatic ring does not impact 5-HT2B
affinity among this subset of compounds given the similar
affinities observed. Among 6d–6f, the highest 5-HT2B selectiv-
ity vs. 5-HT2A was seen for the 3-methoxy derivative, 6e
(43-fold). The 2-methoxy derivative 6d had the lowest 5-HT2B
selectivities (24 and 21-fold for 5-HT2A and 5-HT2C respec-
tively) in the 6d–6f mono-methoxy series. A 2,5-dimethoxy
substitution pattern did not improve affinity as is evident
from the comparison of 6c (26 nM) and 6g (36 nM). Further-
more, 6g had reduced 5-HT2B affinity when compared to the
2-methoxy derivative 6d (36 vs. 5.8 nM) indicating that a
2-methoxy substitution is preferred to 2,5-dimethoxy substitu-
tion for affinity. Low 5-HT2B selectivities were also seen for
compound 6g (26 and 8-fold for 5-HT2A and 5-HT2C). When
compared to the unsubstituted benzene derivative 6c, a 3,4,5-
trimethoxy substitution pattern (i.e. 6h) gave higher 5-HT2B
affinity (4.1 nM) – comparable to that seen in the mono-
methoxy derivatives 6d–6f. 5-HT2B selectivity for 6h vs. the
5-HT2A receptor was comparable to that seen for 6e and 6f
and selectivity vs. 5-HT2C was improved. In fact, 6h had the
highest 5-HT2B vs. 5-HT2C selectivity (47-fold) of all the
compounds tested.
For compounds 6i–6j in which the nitrogen atom is sepa-
rated from the methylendioxyphenyl moiety by only one
methylene group, the highest 5-HT2B affinity was seen for
compound 6i. Unlike the case where the 3,4,5-trimethoxy-
phenyl analog 6h and 3-methoxyphenyl derivative 6e
displayed similar 5-HT2B affinities, significantly lower 5-HT2B
affinity was seen for the 3,4,5-trimethoxyphenyl derivative 6j
when compared to 3-methoxyphenyl derivative 6i. A compari-
son of 6i with its methylene homologue 6e, shows a reduc-
tion in 5-HT2B affinity for 6i (59 vs. 4.6 nM). Comparison of
5-HT2B affinities for 6j and its homologue 6h also shows a
similar trend (231 vs. 4.1 nM). These pieces of data taken
together indicate that the presence of an ethyl linker between
the nitrogen atom and the methylenedioxyphenyl unit is
more desirable for 5-HT2B affinity. Interestingly, the styryl
derivative 6k maintained very good 5-HT2B affinity despite
the absence of an ethyl linker unit as seen in 6c–6h. Indeed,
the 5-HT2B affinity for 6k was similar to 6c which is tending
to suggest that the presence of a cis double bond locks the
phenylpropyl unit into a favorable conformation for binding
to the 5-HT2B receptor. However, even though good 5-HT2B
affinity was retained in 6k, this was not accompanied by any
improvement in selectivity vs. the other 5-HT2 receptors. Thus
the styryl moiety is not preferred for 5-HT2B selectivity.
To further characterize the pharmacological properties of
the analogs, selected compounds were evaluated for functional
activity at the 5-HT2B receptor in calcium mobilization assays.
Here, the analogs were first tested in a primary assay for
agonist and antagonist activity at a single concentration
(10 μM). For each compound, a secondary assay was performed
if the compound was active in the primary assay. For agonists
identified in the primary assay, concentration–response curves
were run to determine EC50 values in a secondary assay. In
the case of antagonists, concentration–response curves were
performed in the presence of the agonist 5-HT at a concentra-
tion of 3 nM to determine IC50 values.
No significant agonist activity was detected for the com-
pounds in the primary assay. Compound 6i did not display
antagonist activity in the primary assay and so was not tested
in the secondary functional assay. The other compounds
examined were all found to be 5-HT2B receptor antagonists in
the primary assay with pIC50 values ranging from 4.9 to
6.1 in the subsequent secondary assays (Table 2).
In order to gauge the selectivity of the scaffold against other
CNS targets and to determine the mode of antagonist action,
compound 6c (as the compound with the highest 5-HT2B antag-
onist activity and as a representative of the set of analogues),
was submitted for further pharmacological characterization.
The following nanomolar affinities for 6c were returned
from the PDSP broad panel screening: ij5-HT1A (821); 5-HT1D
(451); 5-HT7 (700); α1A (333); α1D (467); α2A (102); α2B (29);
α2C (429); β1 (1885); D1 (1682); D2 (1729); D3 (498); D4 (853);
DAT (498); H1 (1297); kappa opioid receptor (363); mu opioid
receptor (341); NET (11); SERT (1001); sigma1 (176); sigma2
(242)]. No appreciable affinity was seen for the following
sites: 5-HT1B, 5-HT1e, 5-HT3, 5-HT5A, 5-HT6, α1B, β2, β3, BZP;
D5, delta opioid receptor, GABAA, H3, M1–M5 and PBR.
Further functional assays on 6c revealed that it is also an
antagonist at the other 5-HT2 receptor subtypes with pIC50
Table 2 pIC50 data for 5-HT2B antagonist assays
Compound
5-HT2B
6b
6c
6d
6e
6f
6g
6h
6i
5.0
6.1
5.0
5.9
5.4
5.1
5.9
nda
4.9
5.2
6j
6k
a
Not determined – inactive in primary assay.
This journal is © The Royal Society of Chemistry 2015
Med. Chem. Commun.

View Article Online
Concise Article
MedChemComm
receptor, while compound 6h shows the highest selectivity
vs. 5-HT2C. Compound 6c showed moderate (>100 nM) or
no appreciable affinity for a number of other receptor sites
in a broad panel screening (excepting for α2B and NET
where affinities of <30 nM were obtained). We anticipate
that the other analogs will display a similar profile but this
needs to be confirmed in future. The affinity data reveals that
various alkyl chain lengths (between
N and the aro-
matic rings), as well as a variety of methoxylated aromatic
ring substitution patterns can be tolerated for good 5-HT2B
affinity. However, the best 5-HT2B affinities are seen for
compounds that feature a propyl linker between the nitrogen
atom and one aromatic moiety and an ethyl unit between the
nitrogen atom and a methylenedioxyphenyl moiety. Func-
tional activity testing revealed that most of these compounds
are h5-HT2B receptor antagonists. Schild analysis revealed
that compound 6c is a non-competitive 5-HT2B antagonist; it
is possible that the other analogues also display a similar
mode of antagonism given the data obtained and the struc-
tural similarities among the series.
The synthetic tractability of this newly identified tris-
IJphenylalkyl)amine template (only 3, high-yielding synthetic
steps from commercially available materials) provides this
scaffold with a significant advantage for the synthesis of
larger libraries of analogs and promise for optimization of
5-HT2B affinity and selectivity. Additional exploration of the
scaffold should provide new tool compounds that will be
useful for mapping the binding surfaces of the 5-HT2B recep-
tor. Further in vitro as well as in vivo pharmacological char-
acterization of these compounds is an exciting dimension
for future work. We are continuing with these synthetic and
biological investigations and will furnish our findings in this
regard in due course.
Fig. 2 a) Schild analysis on compound 6c (PDSP compound code
26793) b) Schild slope regression.
values of 5.3 and 4.9 nM for 5-HT2A and 5-HT2C receptors
respectively. No appreciable agonist activity was observed
at these receptors.
The pIC50 values obtained for the compounds did not
seem to be in line with the affinities (assuming that the com-
pounds are competitive antagonists). We considered that one
possibility for this apparent discrepancy was that the com-
pounds are non-competitive antagonists. To shed some light
on this issue, compound 6c was submitted for a Schild analy-
sis to clarify the mode of antagonism. The result of this anal-
ysis is presented in Fig. 2. As shown in Fig. 2a, increasing
concentrations of compound 6c (PDSP code 26793), caused a
dextral shift in the dose–response curve with a depression in
the maximum response observed in the absence of antago-
nist. The slope of the Schild plot was significantly different
from unity. This indicates that compound 6c is a non-
competitive 5-HT2B antagonist. The pA2 value as determined
by modified Schild analysis19 was 6.26.
Acknowledgements
This publication was made possible by grant numbers
1SC1GM092282 and G12RR003037 from the National Insti-
tutes of Health. Its contents are solely the responsibility of
the authors and do not necessarily represent the official
views of the NIH or its divisions. Ki determinations, receptor
binding profiles and antagonist functional data, was gener-
ously provided by the National Institute of Mental Health's
Psychoactive Drug Screening Program, contract # HHSN-271-
2008-00025-C (NIMH PDSP). The NIMH PDSP is directed by
Bryan L. Roth MD, PhD at the University of North Carolina
at Chapel Hill and Project Officer Jamie Driscol at NIMH,
Bethesda MD, USA.
For experimental details please refer to the PDSP web site
http://pdsp.med.unc.edu and click on “Binding Assay” or
“Functional Assay” on the menu bar.
Conclusions
In summary, we have identified
a new series of tris-
Notes and references
IJphenylalkyl)amine ligands with high affinity and good selec-
tivity for the h5-HT2B receptor. Of the analogs tested,
compound 6b displayed the highest selectivity vs. the 5-HT2A
1 B. Schaerlinger, P. Hickel, N. Etienne, L. Guesnier and
L. Maroteaux, Br. J. Pharmacol., 2003, 140, 277.
Med. Chem. Commun.
This journal is © The Royal Society of Chemistry 2015

View Article Online
MedChemComm
Concise Article
2 R. A. Borman, N. S. Tilford, D. W. Harmer, N. Day,
E. S. Ellis, R. L. Sheldrick, J. Carey, R. A. Coleman and
G. S. Baxter, Br. J. Pharmacol., 2002, 135, 1144.
3 S. M. O'Mahony, D. C. Bulmer, A. M. Coelho,
P. Fitzgerald, C. Bongiovanni, K. Lee, W. Winchester,
T. G. Dinan and J. F. Cryan, Neurogastroenterol. Motil., 2010,
22, 573, e124.
4 G. J. Sanger, Trends Pharmacol. Sci., 2008, 29, 465.
5 C. J. Rhodes, A. Davidson, J. S. Gibbs, J. Wharton and
M. R. Wilkins, Pharmacol. Ther., 2009, 121, 69.
6 F. Jaffre, J. Callebert, A. Sarre, N. Etienne, C. G. Nebigil,
J. M. Launay, L. Maroteaux and L. Monassier, Circulation,
2004, 110, 969.
7 S. L. Diaz, S. Doly, N. Narboux-Neme, S. Fernandez,
P. Mazot, S. M. Banas, K. Boutourlinsky, I. Moutkine,
A. Belmer, A. Roumier and L. Maroteaux, Mol. Psychiatry,
2012, 17, 154.
8 R. B. Rothman and M. H. Baumann, Expert Opin. Drug Saf.,
2009, 8, 317.
11 D. W. Bonhaus, C. Bach, A. DeSouza, F. H. Salazar,
B. D. Matsuoka, P. Zuppan, H. W. Chan and R. M. Eglen, Br.
J. Pharmacol., 1995, 115, 622.
12 G. S. Baxter, Behav. Brain Res., 1996, 73, 149.
13 J. E. Audia, D. A. Evrard, G. R. Murdoch, J. J. Droste,
J. S. Nissen, K. W. Schenck, P. Fludzinski, V. L. Lucaites,
D. L. Nelson and M. L. Cohen, J. Med. Chem., 1996, 39, 2773.
14 D. W. Bonhaus, L. A. Flippin, R. J. Greenhouse, S. Jaime,
C. Rocha, M. Dawson, K. Van Natta, L. K. Chang, T. Pulido-
Rios, A. Webber, E. Leung, R. M. Eglen and G. R. Martin, Br.
J. Pharmacol., 1999, 127, 1075.
15 S. Chaudhary, S. Ponnala, O. Legendre, J. A. Gonzales,
H. A. Navarro and W. W. Harding, Bioorg. Med. Chem.,
2011, 19, 5861.
16 S. Pecic, P. Makkar, S. Chaudhary, B. V. Reddy, H. A. Navarro
and W. W. Harding, Bioorg. Med. Chem., 2010, 18, 5562.
17 S. Chaudhary, S. Pecic, O. Legendre, H. A. Navarro and
W. W. Harding, Bioorg. Med. Chem. Lett., 2009, 19, 2530.
18 N. H. Jensen and B. L. Roth, Comb. Chem. High Throughput
Screening, 2008, 11, 420.
9 B. L. Roth, N. Engl. J. Med., 2007, 356, 6.
10 J. D. Hutcheson, V. Setola, B. L. Roth and W. D. Merryman,
Pharmacol. Ther., 2011, 132, 146.
19 A. Christopoulos, A. M. Parsons, M. J. Lew and E. E. El-Fakahany,
Eur. J. Pharmacol., 1999, 382, 217.
This journal is © The Royal Society of Chemistry 2015
Med. Chem. Commun.