Design, synthesis and binding properties of novel and selective 5-HT3 and 5-HT4 receptor ligands

Article abstract of DOI:10.1016/S0223-5234(00)01187-9

This work reports the synthesis and the binding tests on the 5-HT₃ and 5-HT₄ receptors of new thienopyrimidopiperazine and piperazinylacylaminodimethylthiophene derivatives, in order to identify potent and selective ligands for each receptor. The compound with higher affinity and selectivity for the 5-HT₃ over the 5-HT₄ receptor was the 3-amino-2-(4-benzyl-l-piperazinyl)-5,6-dimethyl-thieno[2,3-d]pyrimidin-4(3H)- one 28 (5-HT₃ K_i = 3.92 nM, 5-HT₄ not active), the compound with higher affinity and selectivity for the 5-HT₄ over the 5-HT₃ receptor was the 2-[4-[4-(2-pyrimidinyl)-l-piperazinyl]butanoylamino]-4,5-dimethyl-3-thiophene carboxylic acid ethyl ester 41 (5-HT₄K_i= 81.3 nM, 5-HT₃ not active). Conformational analyses were carried out on the compounds of the piperazinylacylaminodimethylthiophene series (39-42) taking compound 41 as the template.

Full text of DOI:10.1016/S0223-5234(00)01187-9

Eur. J. Med. Chem. 35 (2000) 1065–1079
´
© 2000 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved
PII: S0223-5234(00)01187-9/FLA
Original article
Design, synthesis and binding properties of novel and selective 5-HT3
and 5-HT4 receptor ligandsꢀ
Maria Modicaa, Maria Santagatia*, Salvatore Guccionea, Filippo Russoa, Alfredo Cagnottob,
Mara Goeganb, Tiziana Menninib
aDipartimento di Scienze Farmaceutiche, Universita` di Catania, viale A. Doria 6, 95125 Catania, Italy
bIstituto di Ricerche Farmacologiche ‘Mario Negri’, via Eritrea 62, 20157 Milan, Italy
Received 27 March 2000; revised 8 August 2000; accepted 8 August 2000
Abstract – This work reports the synthesis and the binding tests on the 5-HT3 and 5-HT4 receptors of new thienopyrimidopiperazine
and piperazinylacylaminodimethylthiophene derivatives, in order to identify potent and selective ligands for each receptor. The
compound with higher affinity and selectivity for the 5-HT3 over the 5-HT4 receptor was the 3-amino-2-(4-benzyl-1-piperazinyl)-5,6-
dimethyl-thieno[2,3-d]pyrimidin-4(3H)-one 28 (5-HT3 Ki=3.92 nM, 5-HT4 not active), the compound with higher affinity and
selectivity for the 5-HT4 over the 5-HT3 receptor was the 2-[4-[4-(2-pyrimidinyl)-1-piperazinyl]butanoylamino]-4,5-dimethyl-3-thio-
phenecarboxylic acid ethyl ester 41 (5-HT4 Ki=81.3 nM, 5-HT3 not active). Conformational analyses were carried out on the
´
compounds of the piperazinylacylaminodimethylthiophene series (39–42) taking compound 41 as the template. © 2000 Editions
scientifiques et me´dicales Elsevier SAS
5-HT3 and 5-HT4 receptors / ligands / arylpiperazines / conformational analysis
1. Introduction
Of the serotonin receptors, 5-HT3 is the only one
belonging to the ligand-gated ion channel receptor
family; 5-HT3 antagonists are used as antiemetic
agents to prevent the vomiting associated with
chemotherapy or radiation-induced emesis, but there
are numerous potential therapeutic applications such
as pain, psychosis, memory impairment, depression,
anxiety, schizophrenia and drug abuse. Less is known
about selective and potent 5-HT3R agonists and their
therapeutic potential. The key pharmacophoric ele-
ments of 5-HT3R antagonists generally include an
aromatic moiety, a carbonyl dipole (or a bioisosteric
equivalent function) and a basic amino group, as in
the case of drugs such as ondansetron, granisetron or
tropisetron [7].
Serotonin modulates the activity of both the central
nervous system and peripheral tissues; as far as we
know it acts on 14 receptor subtypes and plays a role
in a wide range of physiological and pathophysiologi-
cal processes. Considerable attention has been paid to
the identification of agents which act selectively on
each of these receptor subtypes.
For many years we have been studying high-affinity
selective 5-HT1A receptor subtype ligands [1–3] and
molecular modelling studies have recently been re-
ported [4–6].
At present, our interest is focused on 5-HT3 and
5-HT4 receptors too.
ꢀ Accounts of this paper were presented at the Italian–Hun-
garian–Polish Joint Meeting on Medicinal Chemistry. Giar-
dini Naxos-Taormina, Italy, 28 September–1 October 1999
and the EUCO-CC3, Third European Conference on Compu-
tational Chemistry, Budapest, Hungary, 4–8 September 2000.
* Correspondence and reprints.
The 5-HT4 receptor, positively coupled with adeny-
late cyclase, first identified in 1988 and recently cloned
[8], is localized in gastrointestinal, atrial and urinary
bladder tissue, as well as in the central nervous system
(CNS). Therapeutic applications, such as the treat-
ment of irritable bowel syndrome, atrium arrhythmia
E-mail address: msantaga@mbox.unict.it (M. Santagati).

1066
M. Modica et al. / European Journal of Medicinal Chemistry 35 (2000) 1065–1079
This work reports the synthesis, binding tests and
molecular modelling of new derivatives A and B
(figure 1), bearing the three key pharmacophoric ele-
ments, in order to identify potent and selective ligands
for each receptor.
The thienopyrimidothioalkylpiperazine system was
chosen because the compound A, with R1=R2 =
CH3; R3 =2-pyrimidinyl, X=NNH2, Y=S(CH2)3,
shown in our earlier work on high-affinity and selec-
tive 5-HT1A receptor ligands [1], has a good affinity
for the 5-HT3 receptor (IC50 =20 nM).
and urinary incontinence are proposed for the antag-
onists to the 5-HT4 receptor, while the use of agonists
(as prokinetic drugs) is suggested in the treatment of
gastrointestinal motility disorders such as gastro-
oesophageal reflux, functional dyspepsia and
constipation.
Despite being so different, the same structural ele-
ments for the 5-HT3 receptor characterize the 5-HT4
pharmacophore, as shown by QSAR studies on 5-
HT4 ligands like, for example, SB 204070, GR 113808
(figure 1) and others [7]. This explains the poor selec-
tivity for the 5-HT3 and 5-HT4 receptors of many
ligands, which often show affinity for other serotonin
and monoamine receptors, too.
The structural modifications referred to this com-
pound chosen as the lead were, the introduction in the
R1 and R2 positions of a tetramethylene chain instead
of two methyls and in the R3 position of the
Figure 1. Ligands for the 5-HT3 and 5-HT4 receptors and the new synthesized compounds A and B.

M. Modica et al. / European Journal of Medicinal Chemistry 35 (2000) 1065–1079
1067
orthomethoxyphenyl and the benzyl groups instead of
the 2-pyrimidinyl, the replacement of the NNH2
group with the NH or S groups, together with the
shortening and the elimination of the thioalkyl chain.
The compounds with the piperazine nucleus di-
rectly bound to the heterocyclic system, are related to
quipazine (figure 1) and its derivatives, potent ligands
for the 5-HT3 receptor [9].
The compounds of type B (figure 1) were prepared
and tested because the flexibility of structure might
influence the affinity and selectivity for the 5-HT4
receptor over the 5-HT3 receptor [8]. On the most
active compounds for the 5-HT4 receptor (see Section
4), we carried out studies of conformational analysis.
A conformational analysis using the Monte Carlo
method as implemented in Macromodel (version 6.5)
[10, 11] was used to obtain the main conformational
and pharmacophoric requirements for receptor bind-
ing taking compound 41 as the template, therefore,
assuming its conformation as the bioactive one.
2. Chemistry
Derivatives 8–17 were prepared by heating at
reflux in chloroform, the appropriate isothiocyanates
1–4 [2, 12] and piperazines 5–7 that were commer-
cially available, (figure 2).
The piperazinethienothiazines 18 and 19 were ob-
tained from compounds 10 or 12 in sulfuric acid at
room temperature.
The piperazinethienopyrimidinones 26–31 were
prepared by boiling at reflux derivatives 8–13 with
hydrazine monohydrate. Only the hydrazine deriva-
tives 24 and 25 were isolated by reaction of deriva-
tives 14–17 with hydrazine monohydrate in the same
Figure 2. Synthetic procedure to compounds 18, 19, 24–33. Reagents and conditions; (a) CHCl3, reflux; (b) H2SO4 conc., r.t.; (c)
N2H4.H2O, EtOH, reflux; (d) HCl, NaNO2, r.t.; (e) CH3I, H2O, r.t.; (f) N2H4.H2O, 2-propanol, reflux.

1068
M. Modica et al. / European Journal of Medicinal Chemistry 35 (2000) 1065–1079
or slightly modified conditions. Analytical and spec-
tral data of compounds 24 and 25 were like those of
the derivatives obtained by reaction of hydrazine
monohydrate and the methylthio derivatives 22 and
23, respectively, which were in turn prepared from the
monopotassium salts 20 [13] and 21 [2] and methyl
iodide (analytical and spectral data were reported in
section figure 2).
In the compounds 32 and 33, the N3 amino group
of the derivatives 27 and 28 was removed with an
aqueous sodium nitrite solution and hydrochloric
acid.
The ethyl ester of 2-amino-4,5-dimethyl-3-thio-
phenecarboxylic acid 34 [14] heated at reflux with
chloroacylchlorides in chloroform gave the
chloroderivatives 35 [15], 36, 37 and 38 [2], which with
1-(2-pyrimidinyl)piperazine dihydrochloride in boiling
dimethylformamide under reflux gave compounds
39–42.
Compound 43 was obtained by boiling at reflux the
derivative 39 and hydrazine monohydrate in 1-
propanol. Attempts to isolate other superior homo-
logues using the same or slightly modified conditions
were unsuccessful (figure 3).
The
pyrimidinylpiperazine-thioalkylthieno[2,3-
d]pyrimidinone 47 was prepared from the monopotas-
sium salt 46 [1] and chloroethylpyrimidinylpiperazine
45, which was obtained from 1-(2-pyrim-
idinyl)piperazine dihydrochloride 44 and 1-bromo-2-
chloroethane (figure 4).
The proposed structures of compounds 8–19, 26–
33, 39–43, and 47 were confirmed by elemental analy-
ses, spectroscopic IR data and 1H-NMR spectra
(tables I, II and III).
Figure 3. Synthetic procedure to compounds 35–43. Reagents and conditions: (a) ClCO(CH2)nCl, CHCl3, reflux; (b) 1-(2-pyrim-
idinyl)piperazine dihydrochloride, K2CO3, dimethylformamide, reflux; (c) N2H4.H2O, 1-propanol, reflux.
Figure 4. Synthetic procedure to compound 47. Reagents and conditions; (a) Cl(CH2)2Br, K2CO3, dimethylformamide, r.t.; (b)
EtOH, reflux.

M. Modica et al. / European Journal of Medicinal Chemistry 35 (2000) 1065–1079
Table I. Physical and chemical properties of compounds 8–17, 39–42.
1069
Compound R1
R2
Q
R3
Melting
point (°C) solvent yield (%)
Recrystallized
I.R.
Formula
(KBr) (cm−1
)
n(NH) n(CꢀO)
8
9
CH3 CH3
CH3 CH3
CS
CS
C6H4OCH3(o) 165–167
EtOH/dioxane, 65
EtOH/dioxane, 41
/
/
1645
1650
C21H27N3O3S2
C18H23N5O2S2
a
2-C4H3N2
207–208
decomp.
111–113
10
11
12
13
14
15
16
CH3 CH3
ꢁ(CH2)4ꢁ
ꢁ(CH2)4ꢁ
ꢁ(CH2)4ꢁ
CS
CS
CS
CS
CS
CS
CH2C6H5
EtOH, 92
EtOH/dioxane, 68
DMF, 28
EtOH, 33
DMF, 30
DMF, 95
EtOH/dioxane/
DMF, 95
EtOH/dioxane/
DMF, 51
EtOH/dioxane, 40 3225
/
/
/
/
/
/
/
1660
1650
1650
1650
1660
1660
C21H27N3O2S2
C23H29N3O3S2
C20H25N5O2S2
C23H29N3O2S2
C25H27N3O3S2
C22H23N5O2S2
C6H4OCH3(o) 145–146
a
2-C4H3N2
CH2C6H5
210–211
144–146
H
H
C6H5
C6H5
C6H4OCH3(o) 187–189
a
2-C4H3N2
213–215
CH3 COOC2H5 CS
C6H4OCH3(o) 180–182
1705, 1660 C23H29N3O5S2
a
17
CH3 COOC2H5 CS
2-C4H3N2
206–208
/
1710, 1655 C20H25N5O4S2
a
39
40
41
42
CH3 CH3
CH3 CH3
CH3 CH3
CH3 CH3
COCH2
2-C4H3N2a
CO(CH2)2 2-C4H3N2a
CO(CH2)3 2-C4H3N2a
CO(CH2)4 2-C4H3N2
191
1680, 1665 C19H25N5O3S
1700, 1650 C20H27N5O3S
1675, 1660 C21H29N5O3S
1700, 1660 C22H31N5O3S
135–137
88–90
86–88
EtOH, 69
3245
3240
3270
b, 30
b, 25
a
2-Pyrimidinyl ring.
Purified by liquid chromatography.
b
3. Pharmacology
5-HT3 receptor; the other tested compounds do not have
any affinity for either receptor. Such a trend is compli-
ant with what we already know, that is, those com-
pounds where the basic nitrogen connected to the
aromatic acyl group is constrained within rigid struc-
tures, like for example, the piperidine into a tropane
ring system, show higher affinity for the 5-HT3 receptor
rather than the 5-HT4 receptor; on the contrary, flexible
structures where the aromatic acyl group is connected to
the piperidine ring through an amide linkage have a
higher affinity for the 5-HT4 receptor rather than the
5-HT3 receptor [8].
The thioxomethylamino derivatives 8–17, the
thienothiazinone derivatives 18 and 19, the piper-
azinylacylamino derivatives 39–42, the derivatives
with the thienopyrimidinone system either directly
bound to the piperazine (compounds 26–33) or
through
a
methylene (compound 43) or
a
thiomethylene chain (compound 47), were evaluated
for their affinity on the 5-HT3 and 5-HT4 receptors by
radioligand binding assays. The results are shown in
table IV as Ki values.
The compound with highest affinity for the 5-HT3
receptor is the 28 (Ki=3.92 nM), followed with a much
lower affinity by the derivatives 33 (Ki=265 nM) and
18 (Ki=377 nM).
4. Results and discussion
The result of the compound 28 shows that besides the
three pharmacophoric elements, what is important in
this series for the selectivity for the 5-HT3 over the
5-HT4 receptors is the presence of (i) the two methyls in
position 5 and 6 of the thiophene nucleus; in fact, the
substitution of the two methyls with a cyclic te-
tramethylene chain in compound 31 (Ki=2035 nM)
4.1. Structure–affinity relationships
The thienothiazinone derivative 18 and some com-
pounds of the 26–33 series show affinity with a great
selectivity for the 5-HT3 over the 5-HT4 receptor; the
piperazinylacylamino derivatives 39–42 show affinity
for the 5-HT4 receptor with a great selectivity over the

1070
M. Modica et al. / European Journal of Medicinal Chemistry 35 (2000) 1065–1079

M. Modica et al. / European Journal of Medicinal Chemistry 35 (2000) 1065–1079
1071
Table III. 1H-NMR spectra of compounds 8–19, 26–33, 39–43 and 47.
Compound l (DMSO-d6)
8
9
1.33 (t, J=7 Hz, 3H, CH3CH2), 2.21 (s, 6H, 2CH3), 3.08 (m, 4H, piperazine H), 3.81 (s, 3H, OCH3), 4.03 (m, 4H, piperazine
H), 4.32 (q, J=7 Hz, 2H, CH3CH2), 6.87–7.01 (m, 4H, ArH), 12.16 (s, 1H, NH).
1.33 (t, J=7 Hz, 3H, CH3CH2), 2.21 (s, 3H, CH3), 2.23 (s, 3H, CH3), 3.90 (m, 4H, piperazine H), 4.06 (m, 4H, piperazine
H), 4.33 (q, J=7 Hz, 2H, CH3CH2), 6.69 (t, J=4.6 Hz, 1H, ArH), 8.41 (d, J=4.6 Hz, 2H, ArH), 12.06 (s, 1H, NH).
1.31 (t, J=7 Hz, 3H, CH3CH2), 2.20 (s, 6H, 2CH3), 2.50 (m, 4H, piperazine H), 3.54 (s, 2H, CH2), 3.90 (m, 4H, piperazine
H), 4.30 (q, J=7 Hz, 2H, CH3CH2), 7.31 (m, 5H, ArH), 12.08 (s, 1H, NH).
1.32 (t, J=7 Hz, 3H, CH3CH2), 1.72 (m, 4H, H-6,7), 2.59 (m, 2H, H-5), 2.73 (m, 2H, H-8), 3.09 (m, 4H, piperazine H), 3.82
(s, 3H, OCH3), 4.04 (m, 4H, piperazine H), 4.31 (q, J=7 Hz, 2H, CH3CH2), 6.91–6.98 (m, 4H, ArH), 12.20 (s, 1H, NH).
1.32 (t, J=7 Hz, 3H, CH3CH2), 1.72 (m, 4H, H-6,7), 2.60 (m, 2H, H-5), 2.73 (m, 2H, H-8), 3.90 (m, 4H, piperazine H), 4.05
(m, 4H, piperazine H), 4.31 (q, J=7 Hz, 2H, CH3CH2), 6.69 (t, J=4.6 Hz, 1H, ArH), 8.41 (d, J=4.6 Hz, 2H, ArH), 12.10
(s, 1H, NH).
10
11
12
13
14
15
1.30 (t, J=7 Hz, 3H, CH3CH2), 1.71 (m, 4H, H-6,7), 2.57 (m, 2H, H-5), 2.70 (m, 2H, H-8), 3.32 (m, 4H, piperazine H), 3.54
(s, 2H, CH2), 3.90 (m, 4H, piperazine H), 4.28 (q, J=7 Hz, 2H, CH3CH2), 7.33–7.35 (m, 5H, ArH), 12.10 (s, 1H, NH).
1.36 (t, J=7 Hz, 3H, CH3CH2), 3.12 (m, 4H, piperazine H), 3.82 (s, 3H, OCH3), 4.08 (m, 4H, piperazine H), 4.37 (q, J=7
Hz, 2H, CH3CH2), 6.88–7.68 (m, 6H, thiophene H and ArH), 11.93 (s, 1H, NH).
1.18 (t, J=7.4 Hz, 3H, CH3CH2), 3.12 (m, 4H, piperazine H), 3.94 (m, 4H, piperazine H), 4.40 (q, J=7.4 Hz, 2H,
CH3CH2), 6.71 (t, J=4.8 Hz, 1H, ArH), 7.30–7.68 (m, 6H, thiophene H and ArH), 8.43 (d, J=4.8 Hz, 2H, ArH), 11.84 (s,
1H, NH).
16
17
1.29 (t, J=7 Hz, 3H, CH3CH2), 1.35 (t, J=7 Hz, 3H, CH3CH2), 2.70 (s, 3H, CH3), 3.12 (m, 4H, piperazine H), 3.82 (s, 3H,
OCH3), 4.07 (m, 4H, piperazine H), 4.26 (q, J=7 Hz, 2H, CH3CH2), 4.37 (q, J=7 Hz, 2H, CH3CH2), 6.88–6.98 (m, 4H,
ArH), 12.43 (s, 1H, NH).
1.29 (t, J=7 Hz, 3H, CH3CH2), 1.35 (t, J=7 Hz, 3H, CH3CH2), 2.70 (s, 3H, CH3), 3.93 (m, 4H, piperazine H), 3.82 (s, 3H,
OCH3), 4.08 (m, 4H, piperazine H), 4.26 (q, J=7 Hz, 2H, CH3CH2), 4.37 (q, J=7 Hz, 2H, CH3CH2), 6.70 (t, J=4.8 Hz,
1H, ArH), 8.42 (t, J=4.8 Hz, 2H, ArH), 12.34 (s, 1H, NH).
18
19
2.24 (s, 6H, 2CH3), 2.49 (m, 4H, piperazine H), 3.52 (s, 2H, CH2), 3.65 (m, 4H, piperazine H), 7.32–7.34 (m, 5H, ArH).
1.76 (m, 4H, H-6,7), 2.65 (m, 2, H-5), 2.77 (m, 2H, H-8), 3.81 (m, 4H, piperazine H), 3.89 (m, 4H, piperazine H), 6.72 (t,
J=4.6 Hz, 1H, ArH), 8.43 (d, J=4.6 Hz, 2H, ArH).
26
27
28
29
30
31
32
33
39
2.33 (s, 3H, CH3), 2.38 (s, 3H, CH3), 3.12 (m, 4H, piperazine H), 3.61 (m, 4H, piperazine H), 3.82 (s, 2H, OCH3), 5.68 (s, 2H,
NH2), 6.95 (m, 4H, ArH).
2.30 (s, 3H, CH3), 2.35 (s, 3H, CH3), 3.53 (m, 4H, piperazine H), 3.86 (m, 4H, piperazine H), 5.68 (s, 2H, NH2), 6.67 (t,
J=4.6 Hz, 1H, ArH), 8.39 (d, J=4.6 Hz, 2H, ArH).
2.31 (s, 3H, CH3), 2.36 (s, 3H, CH3), 2.54 (m, 4H, piperazine H), 3.46 (m, 4H, piperazine H), 3.55 (s, 2H, CH2), 5.62 (s, 2H,
NH2), 7.34 (m, 5H, ArH).
1.77 (m, 4H, H-6,7), 2.67 (m, 2H, H-5), 2.83 (m, 2H, H-8), 3.10 (m, 4H, piperazine H), 3.59 (m, 4H, piperazine H), 3.80 (s,
3H, OCH3), 5.66 (s, 2H, NH2), 6.89–6.96 (m, 4H, ArH).
1.77 (m, 4H, H-6,7), 2.67 (m, 2H, H-5), 2.83 (m, 2H, H-8), 3.55 (m, 4H, piperazine H), 3.87 (m, 4H, piperazine H), 5.69 (s,
2H, NH2), 6.67 (t, J=4.8 Hz, 1H, ArH), 8.40 (d, J=4.6 Hz, 2H, ArH).
1.76 (m, 4H, H-6,7), 2.50 (m, 4H, piperazine H), 2.65 (m, 2H, H-5), 2.82 (m, 2H, H-8), 3.44 (m, 4H, piperazine H), 3.52 (s,
2H, CH2), 5.60 (s, 2H, NH2), 7.29–7.34 (m, 5H, ArH).
2.25 (s, 3H, CH3), 2.30 (s, 3H, CH3), 3.67 (m, 4H, piperazine H), 3.81 (m, 4H, piperazine H), 6.67 (t, J=4.6 Hz, 1H, ArH),
8.39 (d, J=4.6 Hz, 2H, ArH).
2.26 (s, 3H, CH3), 2.30 (s, 3H, CH3), 3.10 (m, 4H, piperazine H), 4.33 (m, 6H, piperazine H and CH2), 7.47–7.59 (m, 5H,
ArH).
1.30 (t, J=7 Hz, 3H, CH3CH2), 2.21 (s, 3H, CH3), 2.24 (s, 3H, CH3), 2.59 (m, 4H, piperazine H), 3.31 (s, 2H, CH2), 3.83 (m,
4H, piperazine H), 4.30 (q, J=7 Hz, 2H, CH3CH2), 6.65 (t, J=4.6 Hz, 1H, ArH), 8.37 (d, J=4.6 Hz, 2H, ArH), 12.14 (s,
1H, NH).
40
41
42
1.31 (t, J=7 Hz, 3H, CH3CH2), 2.21 (s, 3H, CH3), 2.24 (s, 3H, CH3), 2.53 (m, 4H, piperazine H), 2.69 (m, 4H, CH2CH2),
3.84 (m, 4H, piperazine H), 4.27 (q, J=7 Hz, 2H, CH3CH2), 6.63 (t, J=4.6 Hz, 1H, ArH), 8.37 (d, J=4.6 Hz, 2H, ArH),
11.75 (s, 1H, NH).
1.31 (t, J=7.2 Hz, 3H, CH3CH2), 1.80 (m, 2H, CH2CH2CH2), 2.18 (s, 3H, CH3), 2.20 (s, 3H, CH3), 2.36 (m, 6H, CH2N and
piperazine H), 2.52 (t, J=6.8, 2H, COCH2), 3.65 (m, 4H, piperazine H), 4.29 (q, J=7.2 Hz, 2H, CH3CH2), 6.60 (t, J=4.6
Hz, 1H, ArH), 8.33 (d, J=4.6 Hz, 2H, ArH), 10.97 (s, 1H, NH, exchanges with D2O).
1.31 (t, J=7 Hz, 3H, CH3CH2), 1.52 (m, 2H, CH2), 1.64 (m, 2H, CH2), 2.19 (s, 3H, CH3), 2.22 (s, 3H, CH3), 2.28–2.56 (m,
8H, CH2N, piperazine H and COCH2), 3.69 (m, 4H, piperazine H), 4.28 (q, J=7 Hz, 2H, CH3CH2), 6.61 (t, J=4.6 Hz, 1H,
ArH), 8.34 (d, J=4.6 Hz, 2H, ArH), 10.96 (s, 1H, NH).
43
47
2.36 (s, 3H, CH3), 2.41 (s, 3H, CH3), 2.57 (m, 4H, piperazine H), 3.75 (m, 6H, CH2N and piperazine H), 6.18 (s, 2H, NH2,
exchanges with D2O), 6.62 (t, J=4.6 Hz, 1H, ArH), 8.35 (d, J=4.6 Hz, 2H, ArH).
2.28 (s, 3H, CH3), 2.32 (s, 3H, CH3), 2.45 (m, 4H, piperazine H), 2.60 (t, J=6.8 Hz, 2H, CH2N), 3.13 (t, J=6.8 Hz, SCH2),
3.68 (m, 4H, piperazine H), 5.64 (s, 2H, NH2), 6.56 (t, J=4.6 Hz, 1H, ArH), 8.30 (d, J=4.6 Hz, 2H, ArH).

1072
M. Modica et al. / European Journal of Medicinal Chemistry 35 (2000) 1065–1079
Table IV. Affinity of compounds 8–19, 26–33, 39–43, 47 and
of two reference compounds on 5-HT3 and 5-HT4 serotonin
receptors.
research on ligands with affinity and selectivity towards
the 5-HT3 receptor (besides being studied in terms of
3-D-QSAR).
Ki (nM) (9S.D.)a
The structural features which differentiate the com-
pounds 39–42 among them, the most active and selec-
tive towards the 5-HT4 receptor over the 5-HT3
receptor, is the length of the amidomethylenic chain
which links the thiophene to the piperazinylpyrimidine
system. The one with higher affinity for the 5-HT4
receptors is the 2-[4-[4-(2-pyrimidinyl)-1-piperazinyl]bu-
tanoylamino]-4,5-dimethyl-3-thiophenecarboxylic acid
ethyl ester 41 (Ki=81.3 nM), followed by the pentanoy-
lamino derivative 42 (Ki=118 nM), ethanoylamino 39
(Ki=332 nM) and propanoylamino 40 (Ki=564 nM).
Compound 41 has also been shown to have some selec-
tivity since it has been found not to have affinity either
for 5-HT1A or a1A receptors.
Compound 5-HT3 rat
cortex
5-HT4 guinea-pig striatum
[3H] Zacopride [3H] GR 113808
8
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
N.A.b
332927
564936
81.399
118930
N.A.b
N.A.b
9
N.A.b
10
N.A.b
11
N.A.b
12
N.A.b
13
N.A.b
14
N.A.b
15
N.A.b
16
N.A.b
17
N.A.b
18
377960
Due to the important result we obtained, further
research is currently in progress for improving the
affinity of these structures for the 5-HT4 receptors.
19
N.A.b
26
N.A.b
27
N.A.b
28
3.9290.88
29
N.A.b
4.2. Conformational analysis
30
N.A.b
31
20359240
N.A.b
The relative spatial disposition of the above listed
functionalities as adopted by compound 41 (n=3),
showing the highest level of receptor affinity was first
investigated and then considered to characterize the
‘active conformation(s)’, in which the molecule might
interact at the receptor site both in its neutral (figure 5)
and charged states (figure 6). The latter, which most
likely is the bioactive form, is due to the protonation of
the piperazine N1. Charged forms as bioactive are well
documented for biomolecules with ionizable basic
groups. Evidence coming out from the computational
studies reported below further support this view. As-
suming the piperazine ring of the molecule as charged
on the N1 atom at the pH of the biological experiment,
a planar geometry was identified which allows for the
formation of a seven terms pseudo-cycle ‘head-tail’ in
the most active compounds 41 and 42, which includes
the protonated piperazine nitrogen and the opposite
amide oxygen atom in the side chain. This pseudo-cycle
32
33
265989
N.A.b
39
40
N.A.b
41
N.A.b
42
N.A.b
43
N.A.b
47
31069261
0.8090.1
354997
Quipazine
Serotonin
79.3910
a
Ki values were derived from the IC50 values, determined by
dose–inhibition curves with seven concentrations of the dis-
placers each performed in triplicate, according to the method
of Cheng and Prusoff [20].
b
B50% inhibition at 10−5 M.
leads to a remarkable decrease in affinity; (ii) the amino
group bound to the N3 of the pyrimidinone, as may be
seen from the results of the thienothiazinone and the
thienopyrimidinone derivatives 18 and 33; (iii) the ben-
zyl bound to the N4 of the piperazine, analogously to
the well known N4 benzyl derivatives showing affinity
and selectivity for the 5-HT3 receptor [9]. The N4 or-
thomethoxyphenyl and the N4 pyrimidine derivatives 26
and 27 do not show affinity for either receptor.
may act as
a determinant of activity ‘selecting’
favourable/unfavourable conformations for binding of
the molecule to the receptor.
All the most active 5-HT4R ligands (41 and 42) show
one energetically favourable planar low energy confor-
mation (15 conformers within 1 kcal mol−1) with 2.49
0.5 as the dihedral angle (~), which defined by the
S1-C2-Nam-Cam atom sequence. The less active deriva-
Finally, compound 28 acting as a full agonist at the
5-HT3 receptor in the Bezold–Jarisch reflex (unpub-
lished data) has been chosen by us as a lead for further

M. Modica et al. / European Journal of Medicinal Chemistry 35 (2000) 1065–1079
1073
Figure 5. U-shape conformation of compound 41 in vacuo (see text for explanation). The piperazine ring is in a twisted boat
conformation.
tives 39 and 40 show a less planar geometry as measured
by a dihedral angle (~) of 8.0090.6.
The above planar conformation allows the anti C4
carbonyl–amide NH groups in the side chain of com-
pounds 39–42 to form a strong hydrogen bond
X-ray studies are currently in progress to complement
the computational in vacuo and aqueous results on a
possible lowest energy boat conformation as detected
for the protonated form of compound 42 (n=4). Pro-
vided that false minima might come out from the calcu-
lation itself, it may be further speculated that compound
42 should reach the receptor in an unfavourable boat
conformation which precludes it from interacting at all
with the receptor binding site.
The additional pseudo-cycle, formed by six terms
instead of seven as for compounds 41, negatively affects
the activity of compound 40 (n=2) making its structure
too rigid, therefore, (figure 7) unfavourably locating the
pyrimidine ring and precluding the ligand from key
interactions in the ligand–receptor binding.
The better activity of compound 39 (n=1) when
compared with compound 40 (n=2) might be due to a
more pronounced rotational freedom of the side chain
lacking the additional hydrogen bond. It may allow the
compound to reach a bioactive conformation close to
that of compound 41 (n=3), however, the shorter chain
might preclude this ligand from substantial key
interactions.
,
(2.00090.1 A), thus forming a six atoms pseudo-cycle,
which contributes to make the molecule planar. A sec-
,
ond hydrogen bond (1.7290.06 A) between the proto-
nated piperazine nitrogen and the opposite amide
oxygen atom in the side chain of compounds 41 (n=3)
constrains the molecule to form a seven terms pseudo-
cycle, which favours the chair conformation of the
piperazine ring (figure 6). This fact might contribute to
the optimal spatial arrangement of the pyrimidine ring,
so emphasizing the possible role of heteroaromatic inter-
actions in the ligand-binding process.
Compound 42 (n=4) showed the additional seven
terms hydrogen bond only when the calculations were
carried out in vacuo, so suggesting that the steric hin-
drance might mainly affect the activity (exceeding
molecular included volume) [16] depicting the ligand
receptor interaction in a hydrophobic environment al-
lowing the protonated piperazine nitrogen and the
amide oxygen to interact, without the ‘breaking off’ due
to the solvation effects.
Considering the molecules in their neutral form (in
accordance with a suggestion during the review process),

1074
M. Modica et al. / European Journal of Medicinal Chemistry 35 (2000) 1065–1079
Figure 6. Ball-and-stick representation of compound 41 (n=3) lowest energy conformation in the charged (protonated) form.
Hydrogen bonds are displayed by dotted lines. Arrow points to the additional hydrogen bond (see text for explanation).
the overall view does not change. Likewise, their pro-
tonated counterparts — all the examined compounds
39–42, are characterized by a six atoms pseudo-cycle
formed by the hydrogen bonding between the sp3 oxy-
gen of the ester oxygen and the amide NH in the side
rings, collapses to an unfolded planar geometry under
the solvation energy effects, when the conformational
search is carried out in aqueous conditions.
Comparing the two states — neutral versus proto-
nated, it can be supposed that the hydrogen bond
between the protonated piperazine nitrogen (N1) and
the opposite amide oxygen, therefrom the resulting ad-
ditional virtual cycle, stabilizes the piperazine ring in a
chair conformation, which affects the shape of the pen-
dant moiety of the molecules and strongly conditions
the biological activity, whereas, the y clouds interac-
tions between the aromatic rings (thiophene-phenyl)
are responsible for the stabilization of the neutral
form.
The phenyl substituted nitrogen of the piperazine
ring in the neutral form, likewise the positively charged
form, gives an angle of 120° with the phenyl group, so
showing a strong sp2 character (electron delocalization
with the phenyl ring) and further supporting the N1
piperazine nitrogen as the putative site of protonation
at a physiological pH of 7.4 as in the biological assay.
,
chain (2.00090.1 A), as reported above. The most
active compounds 41, 42 and the ‘less inactive’ com-
pound 39 show a common ‘U-shape’ conformation (15
conformers within 2 kcal mol−1) with a differently
distorted piperazine ring and stabilized by thiophene-
phenyl y clouds interactions. This conformation is sta-
ble in aqueous conditions and might allow the proper
location of the potential pharmacophoric groups. The
strong intramolecular interaction between the two aro-
matic rings should preclude the molecules to establish
electronic interactions with the receptor.
Compound 40 shows only in vacuo the above re-
ported ‘U-shape’ geometry, with a strongly distorted
piperazine ring. The resulting high energy unstable ‘U-
shape conformer’, visual inspection of which clearly
shows a decreased overlapping of the two aromatic

M. Modica et al. / European Journal of Medicinal Chemistry 35 (2000) 1065–1079
1075
Figure 7. Superposition of compounds 40 (n=2) (tube), 41 (n=3) (ball-and-spoke) and 42 (n=4) (ball-and-wire) lowest energy
conformers in the charged (protonated) form, showing the different spatial arrangement of the piperazine ring.
5. Conclusions
Based on the computational result, it might be
speculated on a dynamic process of interaction by a
change in the intramolecular stabilizing interaction
from the electronic as in the uncharged form to that
electrostatic (H bond) as in the protonated form,
therefore, a conformational change from a distorted
boat-twisted conformation to an active chair
conformation.
The protonation might act as a switching on mech-
anism from the silent to the active form. Compound 40
is unable to reach the optimal conformational require-
ments either in the neutral or in the charged form,
whereas compound 39 shows the putative optimal
prerequisites when modelled as uncharged/charged but
its action is probably limited because of the lack of key
interactions.
The synthesis and binding tests on the 5-HT3 and
5-HT4 receptors of new thienopyrimidopiperazine and
piperazinylacylaminodimethylthiophene derivatives
have led to selective ligands for each receptor. The
compound with higher affinity and selectivity for the
5-HT3 over the 5-HT4 receptor was the 3-amino-2-(4-
benzyl-1-piperazinyl)-5,6-dimethyl-thieno[2,3-d]pyri-
midin-4(3H)-one 28, on the other hand, the compound
with higher affinity and selectivity for the 5-HT4 over
the 5-HT3 receptor was the 2-[4-[4-(2-pyrimidinyl)-
1 - piperazinyl]butanoylamino] - 4,5 - dimethyl - 3 -
thiophenecarboxylic acid ethyl ester 41. Conforma-
tional analyses were carried out on compounds of the
piperazinylacylaminodimethylthiophene series (39–
42), acting as ligands at the 5-HT4 receptor.

1076
M. Modica et al. / European Journal of Medicinal Chemistry 35 (2000) 1065–1079
6. Experimental protocols
6.1.3.1. 3-Amino-2-(methylthio)-6-phenyl-
thieno[2,3-d]pyrimidin-4(3H)-one 22
Recrystallized from ethanol/dioxane, yield: 2 g, 72%;
6.1. Chemistry
m.p. 213–215 °C; IR (KBr) 3310 and 3205 (NH), 1685
Melting points were determined in open capillary
tubes on a Gallenkamp Melting point apparatus and are
uncorrected. Elemental analyses for C, H, N and S were
obtained on an EA1108 Elemental Analyzer Fisons-
Carlo Erba instrument and were within 0.4% of the
theoretical values. The IR spectra were recorded with
FT-IR Perkin–Elmer 1600 spectrometer in KBr disks.
1H-NMR spectra were obtained at 200 MHz on a
Varian Inova-Unity 200 spectrometer in DMSO-d6 solu-
tion and are expressed as l units (ppm) relative to TMS
as the internal standard; coupling constants (J) are in
Hertz. Signal multiplicities are presented by s (singlet), d
(doublet), t (triplet), q (quartet), br s (broad singlet),
and m (multiplet). The purity of the compounds was
checked by thin-layer chromatography (TLC) on Merck
silica gel 60 F-254 plates.
(CꢀO) cm−1 1H-NMR (DMSO-d6) l 2.43 (s, 3H,
.
SCH3), 5.83 (s, 2H, NH2), 7.36–7.77 (m, 6H, ArH and
thiophene H). Anal. (C13H11N3OS2) C, H, N, S.
6.1.3.2. 5-Amino-3-methyl-6-(methylthio)-5H-
thieno[2,3-d]pyrimidin-4-one-2-carboxylic
acid methyl ester 23
Recrystallized from ethanol/dioxane, yield: 1.5 g,
52%; m.p. 217–219 °C; IR (KBr) 3315 and 3215 (NH),
1725 and 1650 (CꢀO) cm−1 1H-NMR (DMSO-d6) l
.
1.30 (t, J=7 Hz, 3H, CH3CH2), 2.42 (s, 3H, CH3), 2.81
(s, 3H, SCH3), 4.29 (q, J=7, 2H, CH3CH2), 5.77 (s, 2H,
NH2). Anal. (C11H13N3O3S2) C, H, N, S.
6.1.4. General procedures for compounds 24 and 25
A mixture of the appropriate derivative (14–17) (1.73
mmol) and hydrazine monohydrate (3.4 mL, 68 mmol)
in ethanol (5 mL) was boiled under reflux for 12 h. After
cooling, the solid was collected, washed with ethanol,
dried and recrystallized. From derivatives 14 and 15,
compound 24 was isolated, while 16 and 17 produced
compound 25. The same products were also isolated
when the refluxing time was reduced (2 h), or a lower
amount of hydrazine monohydrate was used.
Compounds 24 and 25 were like those obtained from
methylthio derivatives 22 and 23 (3.46 mmol), respec-
tively, and hydrazine monohydrate (8 mL) boiled under
reflux for 6 h in 2-propanol (15 mL), but in this prepara-
tion the yields were higher.
6.1.1. General procedure for compounds 8–17
To a solution of the appropriate isothiocyanate (1–4)
(4.15 mmol) in chloroform (15 mL), substituted piper-
azine 5 or 6 (5.3 mmol) was added. A suspension of the
1-(2-pyrimidinyl) piperazine dihydrochloride 7 (1.25 g,
5.3 mmol) and triethylamine (2 mL) in chloroform (15
mL) was added for derivatives 9, 12, 15 and 17. The
mixture was heated under reflux for 6–8 h. After cool-
ing, the solvent was removed under reduced pressure
and the sticky residue was collected, washed with etha-
nol, dried and recrystallized by a suitable solvent. Com-
pounds 12 and 15, after cooling, were obtained as a
solid (table I).
6.1.4.1. 3-Amino-2-hydrazino-6-phenyl-(1H,3H)-
thieno[2,3-d]pyrimidine-2,4-dione 24
6.1.2. General procedure for compounds 18 and 19
Compounds 10 and 12 (2 mmol), were stirred for 10
min in concentrated sulfuric acid (14 mL) and kept at
room temperature for 2 and 6 days, respectively. The
solution was then poured into cold water and neutral-
ized with a 10% NaOH solution. The solid was col-
lected, washed with water, dried and recrystallized by a
suitable solvent (table II).
Recrystallized from ethanol/dioxane, yield: 0.15 g,
31%; m.p. 258–260 °C; IR (KBr) 3310 and 3205 (NH),
1
1675 (CꢀO) cm−1. H-NMR (DMSO-d6) l 4.44 (s, 2H,
NH2), 5.40 (s, 2H, NH2), 7.24–7.68 (m, 6H, ArH and
thiophene H), 8.47 (s, 1H, NH). Anal. (C12H11N5OS) C,
H, N, S.
6.1.3. General procedure for compounds 22 and 23
To a suspension of the monopotassium salt 20 or 21
(9.6 mmol) in water (100 mL), methyl iodide (1.8 mL,
28.9 mmol) was added and the mixture was stirred at
room temperature for 0.5–1 h. The solid was collected,
washed with water, dried and recrystallized by a suitable
solvent.
6.1.4.2. 5-Amino-6-hydrazino-3-methyl-5H,7H-
thieno[2,3-d]pyrimidin-4,6-dione-2-carboxyhydrazide 25
Recrystallized from dioxane, yield: 0.2 g, 43%; m.p.
284–286 °C dec; IR (KBr) 3295 and 3205 (NH), 1685
1
(CꢀO) cm−1. H-NMR (DMSO-d6) l 2.63 (s, 3H, CH3),
4.45 (s, 4H, NH2), 5.32 (s, 2H, NH2), 8.46 (s, 1H, NH),
9.22 (s, 1H, NH). Anal. (C8H11N7O2S) C, H, N, S.

M. Modica et al. / European Journal of Medicinal Chemistry 35 (2000) 1065–1079
1077
6.1.5. General procedure for compounds 26–31
6.1.8. General procedure for compounds 39–42
The procedure used was the same of compounds 24
and 25 from derivative 14–17. A mixture of the ap-
propriate derivative (8–13) (1.73 mmol) and hydrazine
monohydrate (3.4 mL, 68 mmol) in ethanol (5 mL)
was boiled under reflux for 12 h. Compounds 10 and
13 were boiled under reflux for 2 h. After cooling, the
solid was collected, washed with ethanol, dried and
recrystallized by a suitable solvent (table II).
A mixture of the appropriate chloroderivative (35–
38) (6.6 mmol), 1-(2-pyrimidinyl) piperazine dihy-
drochloride (1.6 g, 6.9 mmol) and potassium carbonate
(1.9 g, 13.8 mmol) was boiled under reflux for 2 h in
dimethylformamide (20 mL). After cooling, the solid
was separated and the solvent was removed under re-
duced pressure. The solid obtained (compounds 39 and
40) was collected, washed with water, dried and recrys-
tallized by a suitable solvent. The sticky product ob-
tained (compounds 41 and 42) was purified by column
chromatography (1:1, cyclohexane, ethyl acetate) (table
II).
6.1.6. General procedure for compounds 32 and 33
To a suspension of derivatives 27 and 28 (2.1
mmol), respectively, in 6 N hydrochloric acid (15 mL),
an aqueous sodium nitrite solution (0.73 g, 10.58
mmol in 22 mL of water) was added dropwise with
stirring at 0°C. The mixture was stirred at room tem-
perature for 15 min, the solid was collected, resus-
pended in water, to the suspension were added some
drops of a 10% NaOH solution, the solid was then
collected, washed with water, dried and recrystallized
by a suitable solvent (table II).
6.1.9. 3-Amino-2-[1-[4-(2-pyrimidinyl)-1-
piperazinyl]methyl]-5,6-dimethyl-thieno[2,3-d]-
pyrimidin-4(3H)-one 43
A mixture of compound 39 (0.75 g, 1.86 mmol) and
hydrazine monohydrate (1.8 mL, 36 mmol) was boiled
under reflux for 24 h in 1-propanol (4.5 mL). After
cooling, the solid was collected, washed with ethanol,
dried and recrystallized (table II).
6.1.7. General procedure for compounds 36 and 37
A mixture of the amino ester 34 (20 mmol) and
3-chloropropionyl chloride or 4-chlorobutyryl chloride
(20 mmol) was heated under reflux in chloroform (40
mL) for 2 h. After cooling, the solution was washed
with water, the organic layer was separated, dried over
anhydrous sodium sulfate and evaporated under re-
duced pressure to give a solid, that was used for the
next step without further purification.
6.1.10. 1-(2-Chloroethyl)-4-(2-pyrimidinyl) piperazine 45
A
mixture of 1-(2-pyrimidinyl)piperazine dihy-
drochloride 44 (1.5 g, 6.3 mmol), 1-bromo-2-
chloroethane (0.58 mL, 6.97 mmol) and K2CO3 (2.6 g,
18.9 mmol) was stirred at room temperature in
dimethylformamide (8 ml) for 22 h. The solid was
separated and from the solution the solvent was re-
moved under reduced pressure. The sticky product was
purified by column chromatography (7.5:2.5, ethyl ace-
tate, MeOH) to give compound 45 (0.3 g, 20%); m.p.
6.1.7.1. 2-[(Chloropropionyl)amino]-4,5-dimethyl-3-
thiophenecarboxylic acid ethyl ester 36
Yield: (3.3 g, 57%); m.p. 80–82 °C; IR (KBr) 3245
1
(NH), 1705 and 1655 (CꢀO) cm−1. H-NMR (DMSO-
1
61–63 °C; H-NMR (DMSO-d6) l 2.47 (m, 4H, piper-
d6) l 1.32 (t, J=7 Hz, 3H, CH3CH2), 2.19 (s, 3H,
CH3), 2.23 (s, 3H, CH3), 3.03 (t, J=6.2 Hz, 2H,
COCH2), 3.88 (t, J=6.2 Hz, 2H, CH2Cl), 4.30 (q,
J=7 Hz, 2H, CH3CH2), 10.99 (s, 1H, NHCO). Anal.
(C12H16ClNO3S) C, H, N, S.
azine H), 2.68 (t, J=6.6 Hz, 3H, CH2N), 3.74 (m, 6H,
CH2Cl and piperazine H), 6.62 (t, J=4.8 Hz, 1H,
ArH), 8.35 (d, J=4.8 Hz, 2H, ArH). Anal.
(C10H15ClN4.1/2H2O) C, H, N, S.
6.1.7.2. 2-[(Chlorobutyryl)amino]-4,5-dimethyl-3-
thiophenecarboxylic acid ethyl ester 37
6.1.11. 3-Amino-2-[[2-[4-(2-pyrimidinyl)-1-
piperazinyl]ethyl]thio]-5,6-dimethyl-thieno[2,3-d]-
pyrimidin-4(3H)-one 47
A mixture of potassium salt 46 (1.16 g, 3.6 mmol)
and 1-(2-chloroethyl)-4-(2-pyrimidinyl)piperazine 45
(1.16 g, 4.32 mmol) was boiled under reflux for 6 h in
ethanol (15 mL). The suspension was cooled, the solid
was collected, washed with ethanol, dried and recrys-
tallized by a suitable solvent (table II).
Yield: (1.8 g, 29%); m.p. 45–47 °C; IR (KBr) 3300
and 3220(NH), 1655 broad (CꢀO) cm−1
.
1H-NMR
(DMSO-d6) l 1.32 (t, J=7.2 Hz, 3H, CH3CH2), 2.07
(m, 2H, CH2), 2.17 (s, 3H, CH3), 2.20 (s, 3H, CH3),
2.64 (t, J=6.6 Hz, 2H, COCH2), 3.69 (t, J=6.6 Hz,
2H, CH2Cl), 4.28 (q, J=7 Hz, 2H, CH3CH2), 10.96
(s, 1H, NHCO). Anal. (C13H18ClNO3S) C, H, N, S.

1078
M. Modica et al. / European Journal of Medicinal Chemistry 35 (2000) 1065–1079
6.2. Pharmacology
binding, Ki values were derived from the IC50 values
[20]. 5-HT1A and a1A receptors binding assays were
performed as earlier described [1].
6.2.1. In 6itro binding assays
Male CRL:CD(SD)BR-COBS rats (about 150 g,
Charles river, Italy) and male CRL:(HA) BR albino
guinea-pigs (about 300 g, Charles river, Italy) were
killed by decapitation; their brains were rapidly dis-
sected into the various areas (rat cortex for 5-HT3 and
guinea-pig striatum for 5-HT4) and stored at −80 °C
until the day of assay.
Tissues were homogenized in 50 vol. of ice-cold Tris–
HCl, 25 mM, pH 7.4 for 5-HT3 or Hepes HCl, 50 mM,
pH 7.4, for 5-HT4, using an Ultra Turrax TP-1810
homogenizer (2×20 s), and homogenates were cen-
trifuged at 50 000×g for 10 min (Beckman Avanti J-25
refrigerated centrifuge). Each pellet was resuspended in
the same volume of fresh buffer, incubated at 37 °C for
10 min and centrifuged again at 50 000×g for 10 min.
The pellet was then washed once by resuspension in
fresh buffer and centrifuged as before.
6.3. Molecular modelling
6.3.1. Conformational analysis
Starting structures were built using the fragment li-
brary within MacroModel 6.5 [10, 11]. All calculations
were performed on a Silicon Graphics O2 R/5000 work-
station running IRIX 6.5.6.
Conformational analyses were performed with
MacroModel version 6.5 using the Monte Carlo Multi-
ple Minimum Search protocol. All the active ligands
were considered both in their neutral and protonated
positively charged forms. Since quaternization can be
tolerated, it can be assumed that the basic piperazine
center is binding in its protonated form to 5-HT4 recep-
tor, although within reason, the basicity of the basic
centre is not critical.
The pellet obtained was finally resuspended in the
appropriate incubation buffer (Hepes, 50 mM, pH 7.4,
containing 10 mM pargyline for 5-HT4 and Tris–HCl,
25 mM, pH 7.4, containing 10 mM pargyline for 5-HT3)
just before the binding assay.
All minimizations were performed using the AM-
BER* force-field as implemented in Macromodel 6.5
using the GB/SA continuum water model to check for
the hydrogen bond stability.
Prior to submitting the ligands to the Monte Carlo
protocol, a minimization was carried out using the
AMBER* force field and the GB/SA continuum water
model. Default options were used with the Full Matrix–
Newton-Raphson (FMNR) minimizer, allowing for a
[3H] Zacopride (figure 1) [17] (SA 85.0 Ci mmol−1
Amersham, for 5-HT3) binding was assayed in a final
incubation volume of 0.5 mL, consisting of 0.25 mL of
tissue (10 mg per sample), 0.25 mL of the [3H] ligand
(0.4 nM) and 0.01 mL of displacing agent or solvent,
nonspecific binding was measured in presence of 1 mM
quipazine. [3H] GR 113 808 (figure 1) [18] (SA 84.0 Ci
mmol−1 Amersham, for 5-HT4), binding was assayed in
a final incubation volume of 1.0 mL, consisting of 0.5
mL of tissue (20 mg per sample), 0.5 mL of the [3H]
ligand (0.1 nM) and 0.02 mL of displacing agent or
solvent, nonspecific binding was measured in presence of
10 mM serotonin. Incubations (30 min at 25 °C) were
stopped by rapid filtration under vacuum through GF/B
filters which were then washed with 12 mL (4×3 times)
of ice-cold Tris–HCl, 25 mM, pH 7.4, or Hepes HCl, 50
mM, pH 7.4, using a Brandel M-48R cell harvester.
Dried filters were immersed in vials containing 4 mL of
Ultima Gold MV (Packard) and counted in a LKB1214
RACKBETA liquid scintillation spectrometer with a
counting efficiency of about 60%. Drugs were tested in
triplicate at different concentrations (from 10−5 to
10−10 M) and dose–inhibition curves were analysed by
the allfit [19] program to obtain the concentration of
unlabelled drug that caused 50% inhibition of ligand
maximum of 500 iterations per structure, until a gradi-
−1
,
ent of 0.05 kcal A
was reached.
To search the conformational space, 5000 MC steps
were performed on each starting conformation.
Least squares superposition of all the non-hydrogen
atoms was used to eliminate duplicate conformations.
In consideration of the high flexibility of the ligands
under investigation, an energy cut-off of 5.0 kcal mol−1
,
high enough to map the conformational space including
the bioactive conformation, was applied to the search
results.
Acknowledgements
Q. McDonald (Schrodinger Inc.) and the Schrodinger
Inc.’s technical support are gratefully acknowledged for
the helpful discussion and technical support on
MACROMODEL 6.5. Dr A. Orjales (PhD, Director of
Research), L. Labeaga (Head of Pharmacology) and R.
Corco´stegui (Technician of the Section of Pharma-

M. Modica et al. / European Journal of Medicinal Chemistry 35 (2000) 1065–1079
1079
[9] Rault S., Lancelot J., Prunier H., Robba M., Renard P.,
Delagrange P., Pfeiffer B., Caignard D., Guardiola-Lemaitre
B., Hamon M., J. Med. Chem. 39 (1996) 2068–2080.
cologi) (FAES, S.A. Bilbao, Espan˜a) are gratefully ac-
knowledged for the functional assays.
[10] Schro¨dinger Inc. 1500 S.W. First Avenue, Suite 1180 Portland
OR 97201/One Exchange Place, Suite 604, Jersey City, NJ
07302.
References
[11] MACROMODEL 6.5: F. Mohamadi, N.G.J. Richards, W.C.
Guida, R. Liskamp, M. Lipton, C. Caufield, G. Chang, T.
Hendrickson, W.C. Still, J. Comput. Chem. 11 (1990) 440–467.
[1] Modica M., Santagati M., Russo F., Parotti L., De Gioia L.,
Selvaggini C., Salmona M., Mennini M., J. Med. Chem. 40
(1997) 574–585.
[12] Kienze F., Kaiser A., Minder R.E., Helv. Chim. Acta 66 (1983)
148–157.
[2] Modica M., Santagati M., Russo F., Selvaggini C., Cagnotto
A., Mennini T., Eur. J. Med. Chem. 35 (2000) 677–689.
[13] Santagati A., Modica M., Santagati M., Caruso A., Cutuli V.,
Pharmazie 49 (1994) 64–65.
[3] Modica M., Santagati M., Santagati A., Russo F., Cagnotto
A., Goegan M., Mennini T., Bioorg. Med. Chem. Lett. 10
(2000) 1089–1092.
[14] Gewald K., Schinke E., Bo¨ttcher P., Chem. Ber. 99 (1966)
94–100.
[15] Sauter F., Reich G., Stanetty, Arch. Pharm. 309 (1976) 908–
913.
[4] Santagati M., Doweyko A., Santagati A., Modica M., Guc-
cione S., Chen H.M., Uccello Barrettam G., Balzano F., Pro-
ceedings of the 12th European Symposium on QSARs, 1998,
pp. 183–194.
[16] Lo´pez-Rodr´ıguez M.L., Morcillo M.J., Benhamu´ B., Rosado
M.L., J. Comput.-Aided Mol. Des. 11 (1997) 589–599.
[5] Santagati M., Chen H.M., Santagati A., Modica M., Guccione
S., Russo F., Uccello Barretta G., Balzano F., Proceedings of
the 12th European Symposium on QSARs, 1998, pp. 433–439.
[17] Bolanos F.J., Schechter L.E., Miquel M.C., Emerit M.B., Ru-
migny J.F., Hamon M., Gozlan H., Biochem. Pharmacol. 40
(1990) 1541–1550.
[6] Guccione S., Doweyko A., Chen H.M., Uccello Barretta G.,
Balzano F., J. Comput.-Aided Mol. Des. (2000) 647–657.
[18] Grossman C.J., Kilpatrick G.J., Bunce K.T., Br. J. Pharmacol.
109 (1993) 618–624.
[7] Gaster L.M., King F.D., Med. Res. Rev. 17 (1997) 163–214.
[19] De Lean K.W., Munson P.J., Rodbard D., Am. J. Physiol. 235
(1978) E97–E102.
[8] Schaus M.J., Thopson D.C., Bloomquist W.E., Susemichel
A.D., Calligaro D.O., Cohen M.L., J. Med. Chem. 41 (1998)
1943–1955.
[20] Cheng Y., Prusoff W.H., Biochem. Pharmacol. 22 (1973) 3099–
3108.