Discovery of a new series of 5-HT1A receptor agonists

Article abstract of DOI:10.1016/j.bmcl.2010.01.030

Starting from compounds previously identified as α₁-adrenoceptor antagonists that were also found to bind to the 5-HT_1A receptor, in an attempt to separate the two activities, a new series of 5-HT_1A receptor agonists was identified and shown to have high potency and/or high selectivity. Of these, compound 13, which combines high selectivity (5-HT_1A/α₁ = 151) and good agonist potency (pD₂ = 7.82; E_max = 76), was found to be the most interesting.

Full text of DOI:10.1016/j.bmcl.2010.01.030

Bioorganic & Medicinal Chemistry Letters 20 (2010) 2017–2020
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of a new series of 5-HT1A receptor agonists
Silvia Franchini a, Adolfo Prandi a, Claudia Sorbi a, Annalisa Tait a, Annamaria Baraldi a, Piero Angeli b,
Michela Buccioni b, Antonio Cilia c, Elena Poggesi c, Paola Fossa d, Livio Brasili a,
*
a Dipartimento di Scienze Farmaceutiche, Università degli Studi di Modena e Reggio Emilia, Via Campi 183, 41100 Modena, Italy
b Dipartimento di Scienze Chimiche, Università degli Studi di Camerino, Via S. Agostino 1, 62032 Camerino, Italy
c Divisione Ricerca e Sviluppo, Recordati S.p.A., Via Civitali 1, 20148 Milano, Italy
d Dipartimento di Scienze Farmaceutiche, Università degli Studi di Genova, Viale Benedetto XV 3, 16132 Genova, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 4 July 2009
Revised 14 January 2010
Accepted 14 January 2010
Available online 4 February 2010
Starting from compounds previously identified as
bind to the 5-HT1A receptor, in an attempt to separate the two activities, a new series of 5-HT1A receptor
agonists was identified and shown to have high potency and/or high selectivity. Of these, compound 13,
a1-adrenoceptor antagonists that were also found to
which combines high selectivity (5-HT1A
/
a1 = 151) and good agonist potency (pD2 = 7.82; Emax = 76), was
found to be the most interesting.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
5-HT1A receptor
Agonists
1,3-Dioxolane
Selectivity
5-HT1A agonists and partial agonists have been seen to be effec-
tive in the treatment of anxiety and depression.1–5 In addition to
therapeutic applications in the psychiatric field, recent preclinical
studies have suggested that 5-HT1A receptor agonists also have
important neuroprotective properties.6
More recently, in animal models, it was observed that 5-HT1A
receptor activation is a new molecular mechanism of pain relief
and although we are still waiting for proof-of-concept evidence
in humans, 5-HT1A receptor agonists may rival the opioids in pain
relief therapy.7
showed that compound 1 behaves in the same way as partial ago-
nists (pD2 = 8.80, %Emax = 24). This observation prompted us to con-
duct further research on this new class of compound with the aim
of separating the two activities and very recently we reported on a
first structure–activity relationship study.10 Here, as a continuation
of that study, we report on the synthesis of a new set of derivatives
(3–6 and 8–17) and pharmacological evaluation together with pre-
viously synthesised compounds (2 and 7).
The compounds under investigation were synthesised
(Scheme 1) using standard procedures and characterised by 1H
nuclear magnetic resonance (NMR) spectroscopy and elemental
analysis. The chloro-derivatives, prepared as described previ-
ously,9,11 were aminated in 2-methoxyethanol in the presence
of a catalytic amount of KI, with the appropriate amine either
commercially available or prepared in-house by obtaining a reac-
tion between chloroacetamide and the appropriate phenol, fol-
lowed by reduction with diborane.12 The free bases were then
transformed into the corresponding oxalate salts.
The 5-HT1A receptor belongs to the class of G-protein coupled
receptors (GPCRs), whose members share a number of characteris-
tic amino acid patterns. In particular, the transmembrane amino
acid sequence of the 5-HT1A subtype is worthy of note for its high
degree of homology to the a1-adrenergic receptor (approximately
45%).8 Thus, a great number of ligands show high affinity for recep-
tor systems and poor selectivity.
We reported on a new series of 1,3-dioxolane-based a1-adreno-
ceptor antagonists,9 of which compound 1 showed the highest
affinity and selectivity for the a1D subtype. Given the high degree
The pharmacological profile of compounds 1–17, and BMY-
3748 and 8-OH-DPAT as reference compounds, was determined
of homology of the amino acid sequence between the
a
1-adrener-
at the
activity was assessed by the antagonism of (À)-noradrenaline-in-
duced contraction of rat prostatic vas deferens ( 1A) or thoracic
aorta (
1D) and by the antagonism of (À)-phenylephrine-induced
contraction of rat spleen ( 1B). Radioligand binding assay using
a1-adrenoceptors on different isolated tissues. Blocking
gic and 5-HT1A receptors, further pharmacological investigation of
compound 1 was undertaken and unsurprisingly it was found that
compound 1 binds to human cloned 5-HT1A receptors with a sim-
ilarly high affinity (pKi = 8.45). Moreover, functional experiments
a
a
a
[3H]prazosin to label cloned human
a1-adrenoceptors expressed
in CHO cells, and [3H]8-OH-DPAT to label cloned human 5-YN1F
* Corresponding author. Tel.: +39 59 205 5139; fax: +39 59 205 5131.
E-mail addresses: livio.brasili@unimore.it, brasili.livio@unimore.it (L. Brasili).
receptor expressed in HeLa cells was also used. Functional charac-
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.01.030

2018
S. Franchini et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2017–2020
R
of the methoxy group is primarily responsible for the increased
O
affinity and potency between 1 and 2. Moreover, the oxygen atom
of the phenoxyethyl chain seems to play a similar role. In fact, the
O/CH2 isosteric substitution of compound 2 to obtain compound 4
gives very similar results in terms of both affinity and activity at
Cl
HO
+
H2N
i
R
O
both the
iments at the
a
1 and 5-HT1A receptors. Particularly, in functional exper-
1-adrenergic receptors the decrease in potency
O
a
H2N
+
reaches 100-fold as in the case of the a1D subtype. At the 5-HT1A
receptors, the 13-fold decrease in affinity is accompanied by a
18-fold decrease in potency (pD2 = 5.92 vs 7.36 for compound 1)
and efficacy (Emax) is halved.
By moving the methoxy group to 3-(5) and 4-position (6) a gen-
eral reduction in affinity and potency is observed, indicating its
crucial role when in 2-position. The dimethoxy substitution (7–
10) also seems to determine a general decrease in affinity of the
same order of magnitude and of the four disubstituted derivatives
the one with the highest affinity, at least at the 5-HT1A receptors, is
compound 8, which maintains the same selectivity for the 5-HT1A
receptors as reference compound 2.
ii
R
X
Z
Cl
H2N
Y
iii
X
R
H
N
Y
Z
To investigate other substituents at 2-position, we prepared
compounds 11, 12 and 13. Ethoxy (11) and propoxy (12) deriv-
atives show a small decrease in 5-HT1A receptor affinity and an
iv
3-17
increased affinity at the
resulting in a significant decrease in selectivity. In terms of
activity, while at the 1-adrenergic receptor subtypes it is
a1-adrenergic receptor subtypes, thus
a
Comp
X
O
O
O
O
O
O
O
O
O
O
Y
Z
R
H
Comp
11
X
O
O
O
O
O
S
Y
Z
R
slightly decreased, with the largest variation of about 10-fold
observed at the a1D subtype, at the 5-HT1A receptors the agonist
potency increases about 50-fold (pD2 = 9.08 vs 7.36). When a
phenyl ring (13) replaces the methoxy group, a reduction of
affinity is seen at both the receptor systems. At 5-HT1A this
1
2
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
S
O
O
O
O
O
O
O
2-OC2H5
2-iOC3H7
2-C6H5
3-C6H5
4-C6H5
2-OCH3
2-OCH3
2-OCH3
2-CH3
12
3
O
13
4
CH2
O
2-OCH3
14
reduction is of about threefold, whereas at the
a1-adrenergic
5
3-OCH3
15
receptors it ranges 10–35-fold, resulting therefore in a strong
enhancement of selectivity (151). The agonist potency at the
5-HT1A receptors remains unchanged whilst efficacy doubles.
Therefore, by replacing the methoxy group with a phenyl ring,
a positive effect on selectivity and efficacy of stimulation at
the 5-HT1A receptor is observed.
6
O
4-OCH3
16
7
O
2,6-diOCH3
2,3-diOCH3
3,4-diOCH3
3,5-diOCH3
17
S
8
O
9
O
10
O
Finally, by replacing the oxygen with a sulfur atom at 3-posi-
Scheme 1. Reagents: (i) Na/C2H5OH; (ii) BH3, diglyme; (iii) KI, 2-methoxyethanol;
(iv) C2O4H2, Et2O.
tion to give the 1,3-oxathiolane 16, the affinity at the
a1- and 5-
HT1A receptors is barely affected whereas from a functional point
of view a 10-fold decrease at the a1A subtype and 10-fold in-
crease at the a1B subtype are observed. At the 5-HT1A receptors,
potency increases 38-fold (pD2 = 8.94) and efficacy doubles
(Emax = 76%).
terisation of certain selected compounds at the 5-HT1A receptor
was performed according to the method of Stanton and Beer, using
[
35S]GTP
cS binding, in the cell membranes of HeLa cells transfec-
ted with human cloned 5-HT1A receptor. For experimental details
see Ref. 10.
When both oxygens are replaced by sulfur atoms to give 1,3-
dithiolane 17, the affinity at the 5-HT1A increases (fivefold)
Preliminary results with enantiomers of compound 1 showed
an eudismic ratio of about 2–4 (R/S), therefore all the compounds
were tested as racemates.
whereas at the a1-adrenoceptors it decreases up to 13-fold, as in
the case of the a1D subtype, thus raising the selectivity ratio to
158. The antagonist potency at the a1 subtypes is decreased,
whereas at the 5-HT1A potency increases threefold.
Table 1 lists the pharmacological results. As reported previously,
in functional studies, compound 1 showed selectivity to the
a
1D sub-
These latter results parallel those recently reported10,13 and
seem to confirm that going from 1,3-dioxolane to 1,3-oxathiolane
or 1,3-dithiolane one or more pharmacological parameters that fa-
vour 5-HT1A receptor activity are clearly observed. In the case of
1,3-oxathiolane, the enhanced parameter is potency whereas in
the case of 1,3-dithiolane, selectivity is positively effected. A far
larger series of derivatives will have to be studied to ascertain
whether or not this is a general trend.
In order to better rationalise the results obtained, a pharmaco-
phoric model was derived, from five of the most potent and selec-
tive 5HT1A receptor agonists described in the literature (Chart
1),1,14–17 and compound 13, the most interesting of the series.
Starting from the best geometries obtained by conformational
analysis, a common alignment was derived using the MOE phar-
macophore search module (MOE, Chemical Computing Group
Inc., Montreal, H3A 2R7 Canada, http://www.chemcomp.com),
type 160- and 324-fold that for the a1A and a1B subtypes, respec-
tively. This selectivity was confirmed, albeit to a lesser extent, in
binding studies. Its 2-methoxy derivative 2 shows lower selectivity
as a result of an increase in activity/affinity to the
types. The two compounds bind to the 5-HT1A receptor with an even
higher affinity than the 1-adrenergic receptors. Again in this case,
a1A and a1B sub-
a
compound 2 binds better than compound 1, indicating the positive
role played by the methoxy group in the binding process. However,
the agonist potency is negatively affected, since the pD2 value of 7.36
is about 28 times lower than that of the parent compound 1. There-
fore, as in the case of the
a1-adrenergic receptors, the 2-methoxy
group increases binding affinity whilst the potency at the 5-HT1A
receptor is reduced more than 10-fold.
The 2-methyl derivative 3 decreases affinity and activity at both
receptor systems (a1 and 5-HT1A) suggesting that the oxygen atom

S. Franchini et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2017–2020
2019
Table 1
a
1-Adrenoceptor antagonist potency (pKb)a, affinity constants (pKi)b and selectivitiesc for human recombinant
a
1-adrenoceptor subtypes and 5-HT1A receptors and agonist
d,e
activity (pD2, %Emax
)
at the human recombinant 5-HT1A receptors of compounds 1–17 and BMY-7378 and 8-OH-DPTA
35S]GTP
c
Sd
%Emax
e
Compd
pKba1A
pKba1B
pKba1D
pKia1a
pKia1b
pKia1d
pKi 5-HT1A
Selectivity 5-HT1A
/
a1
[
pD2
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
6.16
7.53
6.02
5.92
5.75
<5
5.73
5.86
5.09
5.41
7.23
7.25
<5
5.86
7.36
5.73
5.97
5.79
6.25
5.89
5.85
6.01
5.49
6.92
7.25
5.47
6.21
5.84
8.44
5.87
7.48
8.37
8.65
6.77
6.86
6.61
6.94
6.93
6.64
6.90
6.77
7.91
7.59
6.10
<5
7.43
7.71
6.93
7.31
7.04
nt
7.00
7.07
<6
6.52
8.18
8.68
6.31
<6
7.20
7.33
6.62
7.38
7.23
nt
7.12
6.76
6.86
6.35
7.27
7.70
<6
<6
<6
7.00
6.93
6.15
<6
7.94
8.03
7.59
7.3
7.24
nt
6.74
6.45
<6
8.45
9.22
8.08
8.10
8.58
nt
7.27
8.12
7.01
<6
9.05
8.77
8.66
7.54
<6
8.99
9.89
8.90
8.43
3
15
3
6
22
—
14
11
1.4
0.2
7
8.8
7.36
—
5.92
5.97
—
—
—
—
—
24
32
—
14
14
—
—
—
—
—
<6
8.17
8.20
6.48
<6
9.08
9.08
7.82
—
36
25
73
—
1
151
>35
—
6
158
1
5.19
5.65
6.71
5.80
7.01
5.08
8.68
7.50
8.40
<6
<6
—
—
7.68
7.45
6.42
6.82
8.18
7.69
8.89
<6
8.94
7.76
9.27
7.83
76
20
26
100
17
BMY-7378
8-OH DPAT
>270
nt: not tested.
a
pKb value agreed within 2%.
b
c
Ki values were calculated at one or two concentrations and agreed within 10%.
Antilog of the difference between the pKi values for the a1a (or a1b, or 1d)-adrenoceptors and the 5-HT1A receptor.
Potency value in the agonist-induced[35S] GTP
S binding-assay.
Maximal stimulation expressed as a percentage of the maximal 5-HT response.
a
d
e
c
Chart 1.
setting tolerance to 1.20 and threshold to 97%. Only those features
showing a score >90% were retained.
Pi ring normal; PiN), F4 (Aro/Hyd/Acc) and F5 (H-bond Acceptor
and H-bond Acceptor projection; Acc/Acc2) are not possessed by
all the molecules in the dataset. Thus, these additional features
are probably important for increasing the agonist potency but they
appear to be less essential than F1 and F2 for anchoring the mole-
cule into the receptor binding site.
However, docking studies will be the further development of
this investigation, both to elucidate at a molecular level the inter-
action between the newly synthesised compounds and the 5HT1A
For the studied compounds, two common pharmacophoric ele-
ments were highlighted, F1 (hydrophobic centre with an H-bond
acceptor or donor function; Hyd/Acc/Don) and F2 (aromatic or
more generally hydrophobic core structure with an H-bond
acceptor function Aro/Hyd/Acc). They were found to be spatially
positioned one in front of one another, at a distance of 3.89 Å
(Fig. 1). The remaining pharmacophoric features, F3 (Aromatic or

2020
S. Franchini et al. / Bioorg. Med. Chem. Lett. 20 (2010) 2017–2020
Acknowledgments
This work was supported by a grant from the Italian Ministry
for University and Research. The authors wish to thank Ms. Rossel-
la Gallesi for performing the elemental analysis.
References and notes
1. Caliendo, C.; Santagada, V.; Perissutti, E.; Fiorino, F. Curr. Med. Chem. 2005, 12,
1721.
2. De Vry, J.; Schreiber, R.; Glaser, T.; Traber, J. In Serotonin1A Receptors in
Depression and Anxiety; Stahl, S. M., Gastpar, M., Keppel-Hesselink, J. M., Traber,
J., Eds.; Raven Press: New York, 1992; p 55.
3. De Vry, J. Psychopharmacology 1995, 121.
4. Gonzalez, L. E.; File, S. E.; Overstreet, D. H. Pharmacol. Biochem. Behav. 1998, 59,
787.
5. Gurwitz, D. Drug Discuss. Today 1999, 4, 142.
6. Berends, A. C.; Luiten, P. G. M.; Nyakas, C. CNS Drug Rev. 2005, 11, 379.
7. Colpaert, F. C. Curr. Opin. Invest. Drugs 2006, 7, 40.
8. Trumpp-Kallmeyer, S.; Hoflack, J.; Bruinvels, A.; Hibert, M. J. Med. Chem. 1992,
35, 3448.
9. Brasili, L.; Sorbi, C.; Franchini, S.; Manicardi, M.; Angeli, P.; Marucci, G.;
Leonardi, A.; Poggesi, E. J. Med. Chem. 2003, 46, 1504.
10. Franchini, S.; Tait, A.; Prandi, A.; Sorbi, C.; Gallesi, R.; Buccioni, M.; Marucci, G.;
De Stefani, C.; Cilia, A.; Brasili, L. ChemMedChem 2009, 4, 196.
11. Angeli, P.; Giannella, M.; Pigini, M.; Gualtieri, F.; Teodori, E.; Valsecchi, B.;
Gaviraghi, G. Eur. J. Med. Chem. 1985, 20, 517.
Figure 1. Pharmacophore features shared by compounds 13 and 18–22: Hydro-
phobic centres with an H-bond acceptor or donor function (Hyd/Acc/Don) coloured
by blue spheres, Aromatic/Hydrophobic/ centres with an H-bond acceptor function
(Aro/Hyd/Acc) coloured by purple spheres, Aromatic or Pi ring normal (PiN)
coloured by orange spheres, H-bond Acceptor and H-bond Acceptor projection (Acc/
Acc2) coloured by cyan spheres. Distances among the key features identified by the
pharmacophore analysis are reported in Angstrom.
12. Pigini, M.; Brasili, L.; Giannella, M.; Giardina, D.; Gulini, U.; Quaglia, W.;
Melchiorre, C. J. Med. Chem. 1988, 31, 2300.
13. Sorbi, C.; Franchini, S.; Tait, A.; Prandi, A.; Gallesi, R.; Angeli, P.; Marucci, G.;
Pirona, L.; Poggesi, E.; Brasili, L. ChemMedChem 2009, 4, 393.
14. Cuisiat, S.; Bourdiol, N.; Lacharme, V.; Newman-Tancredi, A.; Colpaert, F.;
Vacher, B. J. Med. Chem. 2007, 50, 865.
15. Becker, O. M.; Dhanoa, D. S.; Marantz, Y.; Chen, D.; Shacham, S.; Cheruku, S.;
Heifetz, A.; Mohanty, P.; Fichman, M.; Sharadendu, A.; Nudelman, R.;
Kauffman, M.; Noiman, S. J. Med. Chem. 2006, 49, 3116.
16. Kumar, J. S. D.; Majo, V. J.; Hsiung, S.; Millak, M. S.; Liu, K.; Tamir, H.;
Prabhakaran, J.; Simpson, N. R.; Van Heertum, R. L.; Mann, J. J.; Parsey, R. V. J.
Med. Chem. 2006, 49, 125.
receptor, and to define on the receptor surface which amino acidic
residues match the different pharmacophoric features identified in
this preliminary study.
In conclusion, starting from compounds previously identified as
a1-adrenoceptor antagonists that were also found to bind to the
5-HT1A receptor, in an attempt to separate the two activities, a
new series of 5-HT1A receptor agonists was identified. Of these,
some were outstanding for their potency (11, 12 and 16) and oth-
ers for their 5-HT1A/a1 selectivity (13 and 17). Compound 13, how-
ever, which combined high selectivity and good agonist potency,
17. Maurel, J. L.; Autin, J.; Funes, P.; Newman-Tancredi, A.; Colpaert, F.; Vacher, B. J.
Med. Chem. 2007, 50, 5024.
proved to be the most interesting of the series.