Synthesis and pharmacological characterisation of a conformationally restrained series of indole-2-carboxylates as in vivo potent glycine antagonists

Article abstract of DOI:10.1016/S0014-827X(01)01141-7

After the identification of GV150526, the indole-2-carboxylate template was further explored in order to identify novel potential anti-stroke agents. In particular, the SAR of the side chain present at the C-3 position of the indole nucleus was widely stu

Full text of DOI:10.1016/S0014-827X(01)01141-7

www.elsevier.com/locate/farmac
Il Farmaco 56 (2001) 791–798
Synthesis and pharmacological characterisation of a
conformationally restrained series of indole-2-carboxylates as in
vivo potent glycine antagonists
Romano Di Fabio *, Gianluca Araldi, Davide Baraldi, Alfredo Cugola,
Daniele Donati, Paola Gastaldi, Simone A. Giacobbe, Fabrizio Micheli,
Giorgio Pentassuglia
GlaxoSmithKline Group, GlaxoWellcome SpA, Medicines Research Centre, Via Fleming 4, 37135, Verona, Italy
Abstract
After the identification of GV150526, the indole-2-carboxylate template was further explored in order to identify novel potential
anti-stroke agents. In particular, the SAR of the side chain present at the C-3 position of the indole nucleus was widely studied.
In this paper, the synthesis and the pharmacological profile of a further class of conformationally restricted analogues of
GV150526 as in vitro and in vivo potent glycine antagonists is reported. In particular, a pyrazolidinone derivative was identified
as a potent neuroprotective agent in animal models of cerebral ischaemia. © 2001 Elsevier Science S.A. All rights reserved.
Keywords: Stroke; Neuroprotection; Glycine antagonists; Indole-2-carboxylates
1. Introduction
ing hypothesis, some years ago we became interested in
the glycine antagonists field. In particular, the explo-
ration of indole-2-carboxylates led to the identification
of the indole-2-carboxylate (1) (GV150526), shown in
Fig. 1, as a selective glycine antagonist endowed with
an outstanding neuroprotective activity in animal mod-
els of cerebral ischaemia. [20,21]. To further explore the
SAR of the C-3 side chain belonging to this indole
template, several classes of modified indole-2-carboxy-
lates derivatives were prepared [22–25] and two exam-
ples of particular relevance to this paper are reported in
Fig. 1 (2 and 3). In the present study, the synthesis and
the biological and pharmacological characterisation of
some compounds of the general structure 4 [Fig. 1:
X=(CH2)n, n=1, 2; Y=N, O] were performed in
order to evaluate the effect of the introduction of
different heteroatoms within cyclised and conforma-
tionally restricted structures.
Since the discovery of the neurotoxic properties of
glutamate [1], a significant body of evidence has been
gained on the instrumental link between neurotoxicity
induced by glutamate and stroke [2–7]. This dramatic
pathological event, caused by sudden reduction of cere-
bral blood flow, is currently associated to a high unmet
need due to the lack of effective neuroprotective thera-
pies [8] able to protect cerebral tissues, avoiding irre-
versible neurological impairments. Neuronal damage
results from an abnormal overload of Ca2+ into the
post-synaptic neurons, through the overactivation, by
glutamate, of the ionotropic NMDA receptors [9]. The
crucial role of glycine as co-agonist of glutamate, in the
activation of the ion channel associated to the NMDA
receptor, has been well documented [10,11].
In the last decade the strychnine-insensitive glycine
binding site has been hypothesised to be a key target
for the identification of suitable antagonists [12–19] as
potential neuroprotective agents. Following this work-
2. Chemistry
Compounds 2 and 3 were previously synthesised
* Corresponding author.
E-mail address: rdf26781@glaxowellcome.co.uk (R. Di Fabio).
following the general sequential aldol-type condensa-
0014-827X/01/$ - see front matter © 2001 Elsevier Science S.A. All rights reserved.
PII: S0014-827X(01)01141-7

792
R. Di Fabio et al. / Il Farmaco 56 (2001) 791–798
bly, the presence of compound 9 confirmed the
condensation–lactonisation–elimination mechanism
previously proposed [26]. Moreover, it is worth under-
lining that all isolated products were devoid of the Boc
protecting group on the indole nitrogen, suggesting a
more complex reaction mechanism with respect to the
previous case [26].
As far as the regioselectivity of the reaction is con-
cerned, the presence of the anti-lactone 9 in the reaction
mixture suggests that this compound, probably for
steric reasons, cannot reach the necessary antiperipla-
nar conformational arrangement to give the elimination
reaction. The corresponding syn-analogue instead,
should evolve rapidly affording the desired E-olefin
derivative 7.
Due the presence of ethyl ester 8 in the reaction
mixture, at the moment, it is difficult to explain the
different mechanisms controlling this condensation re-
action. However, the liability of the Boc protecting
group on the indole nitrogen in basic medium (release
of 1 eq. LiOEt after lactonisation) should play an
essential role in the formation of 8. Further studies are
currently going on to clarify these particular aspects
and will be published in due time. To obtain the target
compound 10, the free carboxyl derivative 7 was crys-
tallised and submitted to the following deprotection
reaction of the NꢀBoc group to give 10 in high yield.
After the optimisation of the reaction conditions, the
same synthetic route was successfully applied to the
preparation of the different compounds shown in Table
1 (entries 15–17) (Scheme 3) [29].
Fig. 1. GV150526 and conformationally restricted indole-2-carboxy-
late derivatives. X=(CH2)n, n=1, 2; Y=O, NH, NꢀCH3, NꢀBoc.
tion–lactonisation–elimination reaction as reported in
Ref. [26].
All the new indole-2-carboxylate derivatives were
characterised in terms of in vitro affinity at the glycine-
binding site, assessed by displacement of [3H]-glycine
according to Kishimoto [27].
To better define the structure–activity relationship
(SAR) of the conformationally restricted analogues of
compound 1, and to test the effect of the introduction
on these cyclic scaffolds of different heteroatoms as
previously attempted [20,21], a selected number of ana-
logues of derivatives 2 and 3 were prepared [Fig. 1:
X=(CH2)n, n=1, 2; Y=O, NH, NꢀCH3, NꢀBoc].
To synthesise this second series of compounds a
modified version of the previously described [26] aldol-
type condensation–lactonisation–elimination reaction
was set up. This time, the main problem to be tackled
was related to the potential chemical instability of these
new compounds in the strong acid and/or basic condi-
tions necessary to remove the NꢀSEM and the ethyl
ester protecting groups, respectively. In this event, the
NꢀBoc derivative of intermediate 5 was prepared. Actu-
ally, the NꢀBoc protecting group was easier to remove
using milder conditions with respect to the NꢀSEM
one. However, when 6 was submitted to the previous
aldol-type condensation reaction, a different reactivity
with respect to the analogous SEM-derivative was
observed.
3. Results and discussion
3.1. In 6itro affinity studies
All the prepared derivatives were characterised in
terms of in vitro affinity to the glycine-binding site.
The Ki values were measured from at least six-point
inhibition curves and they are the geometric means of
at least three independent experiments. The standard
error of the mean was less than 0.05 for all the deriva-
tives tested. All the derivatives showed complete selec-
tivity (\1000 fold) for other ionotropic receptors,
namely NMDA, AMPA and KA. From the results
achieved by this limited series of compounds, shown in
Table 1, the reduction of both the lipophilic character
(10 versus 2) and the steric bulk seem to maximise the
affinity at the receptor in the five-membered ring series
(10 versus 15 and 7a, respectively). In the six-membered
ring series (16 versus 3) the reduction in lipophilicity
seems to have poor effect, possibly because of an
already exceeded steric hindrance within the receptor
pocket (excluded volumes). In particular, the N-phenyl
pyrazolidinone derivative 10 was identified as the most
As an example (Scheme 1), when aldehyde derivative
6 was treated with the Li enolate of pyrazolidinone 14a,
prepared in three steps from phenylhydrazine 11
(Scheme 2), at −50 °C in THF and then at room
temperature, a mixture of free carboxyl derivative 7, the
corresponding ethyl ester 8 and the chemically stable
anti-lactone 9 (ratio 4.6:2.7:1) was isolated [28]. Nota-

R. Di Fabio et al. / Il Farmaco 56 (2001) 791–798
793
Scheme 1. Synthesis of the pyrazolidinone derivatives.
Scheme 2. Synthesis of the N-phenyl pyrazolidinone intermediates.
potent compound of this series of indole-2-carboxylate
(pKi=8.0 versus 8.5 for compound 1).
This indole derivative was found to be as effective as
compound 1 in blocking convulsions induced by prior
i.c.v. administration (1 min) of the exogenous agonist
3.2. In 6i6o characterisation in animal model of stroke
N-methyl-D-aspartate (ED50=0.1 and 0.06 mg/kg, i.v.,
for compound 10 and 1, respectively). In particular,
animals were observed for the occurrence of generalised
seizures during the first 30 min after the treatment with
NMDA and were considered protected if convulsions
did not occur within this period. In view of this result,
compound 10 evaluated in the middle cerebral artery
occlusion model (MCAo) [31] in rats, exhibiting a high
neuroprotective activity when administered i.v. both
In view of the promising in vitro activity to the
glycine-binding site associated to the NMDA receptor,
compound 10 was tested in vivo in the NMDA-induced
convulsion model in mice [30], a surrogate model of
stroke, and the research complied with national legisla-
tion, with the company policy on the Care and Use of
Animals and with the related codes of practice.

794
R. Di Fabio et al. / Il Farmaco 56 (2001) 791–798
4. Conclusions
pre- and post-ischaemia. In particular, in the latter
protocol, this novel indole derivative, as in the case of
compound 1, was able to stop the progression of the
brain damage [20] when given, as a single 3 mg/kg i.v.
bolus, up to 6 h after the occlusion of the middle
cerebral artery (MCA).
The further exploration of indole-2-carboxylate al-
lowed to identify a novel series of glycine antagonists
exhibiting high in vitro affinity to the glycine-binding
site and outstanding neuroprotective profile in the ani-
mal model of cerebral ischemia. In particular, the pyra-
zolidinone derivative compound 10, the most potent
glycine antagonist belonging to this selected series of
indole-2-carboxylates, in view of the observed activity
in the MCAo model in rats can be considered as one of
the most in vivo potent glycine antagonists identified to
date.
Table 1
In vitro affinity to the glycine-binding site
5. Experimental
5.1. Chemistry
Nuclear magnetic resonance spectra were recorded
with a Varian VXR5000S spectrometer (300 MHz);
chemical shifts are expressed as l units (part per mil-
lion) downfield from TMS. Infrared spectra were
recorded in a Bruker IFS48 spectrometer as Nujol
emulsion. The FAB-mass spectra were measured in a
Fisons VG-4 instrument. Elemental analysis (C, H, N)
were performed in a CHNS-O EA-1108 Elemental
Analyser and were within 90.4% of the theoretical
values. Melting points are uncorrected and were deter-
mined with a capillary melting point apparatus. TLC
was performed in Merck silica gel 60, F254 plates. Flash
column chromatography was carried out in silica gel
(230–400 mesh, G60 Merck). All reagents were of
commercial quality from freshly opened containers.
Compounds 2 and 3 have been prepared as previously
reported [24].
Displacement of [3H]-glycine. Ki values were measured from at least
six-point inhibition curves and they are the geometric means of at
least three independent experiments. The standard error of the mean
was less than 0.05 for all compounds.
5.2. Ethyl 1-tert-butoxycarbonyl-3-formyl-4,6-
dichloroindole-2-carboxylate (6)
To a stirred suspension of the ester 5 (2.2 g, 7.7
mmol) in CH2Cl2 (70 ml) was added 4-dimethylamino
pyridine (0.01 g, 0.08 mmol) and a solution of di-
tert-butyl dicarbonate (2 g, 9.16 mmol) in CH2Cl2
(20 ml) and the reaction mixture was stirred for 1 h.
The solvent was evaporated and the crude residue
was triturated in Et2O/petroleum (10:10 ml) to yield
compound 6 as a white solid (2.7 g, 91%): m.p.
142–144 °C. IR (Nujol): wmax 1768, 1742, 1692 cm−1
.
MS; m/e: 386 (M+1). 1H NMR (CDCl3): l 10.77
(s, 1H), 8.24 (d, 1H), 7.42 (d, 1H), 4.50 (q, 2H), 1.65
(s, 9H), 1.43 (t, 3H). Anal. (C17H17Cl2NO5): C,
H, N.
Scheme 3. Synthesis of conformationally restrained indole derivatives.

R. Di Fabio et al. / Il Farmaco 56 (2001) 791–798
795
5.3. General procedure A for the preparation of
compounds 7a, 7b, 15 and 17
(t, 1H), 7.60 (dd, 2H), 7.46 (d, 1H), 7.37 (t, 2H), 7.19
(d, 1H), 7.19 (tt, 1H), 3.84 (bs, 2H), 2.56 (bs, 2H), 1.28
(s, 9H). Anal. (C25H23Cl2N3O5): C, H, N.
To a stirred solution of the intermediates 14a–d (1.2
eq.) in THF was added a 1 M solution of lithium
bis(trimethylsilyl)amide in THF (1.4 eq.) at −50 °C,
then the solution was stirred for 30 min at −20 °C.
After cooling at −50 °C, a solution of the aldehyde 6
(1.0 eq.) in THF was added dropwise, then the reaction
mixture was slowly warmed at room temperature (r.t.)
and stirred for 2 h. The solution was diluted with
AcOEt and washed with a 0.1 N solution of HCl and
water, then the organic layer was separated, dried over
Na2SO4, and evaporated in vacuo. The obtained com-
pounds were purified by crystallisation or by flash
column chromatography.
5.3.3. (E)-4,6-Dichloro-3-(1-methyl-3-oxo-2-phenyl-
pyrazolidin-4-ylidenemethyl)-1H-indole-2-carboxylic
acid (15)
Compound 15 was prepared from 6 (0.37 g, 0.96
mmol) and 14c according to general procedure A. The
crude compound was purified by trituration in Et2O to
obtain compound 15 as a pale-yellow solid (0.06 g,
15%): m.p. \220 °C. IR (Nujol): wmax 3244, 1676,
1
1653 cm−1. MS; m/e: 416 (M+1). H NMR (DMSO-
d6): l 13.75 (bs, 1H), 12.48 (s, 1H), 7.84 (d, 2H), 7.83 (s,
1H), 7.45 (d, 1H), 7.42 (t, 2H), 7.27 (d, 1H), 7.16 (t,
1H), 4.08 (bs, 1H), 3.63 (bs, 1H), 3.31 (s, 3H). Anal.
(C20H15Cl2N3O3): C, H, N.
5.3.1. (E)-3-(1-tert-Butoxycarbonyl-3-oxo-2-phenyl-
pyrazolidin-4-ylidenemethyl)-4,6-dichloro-1H-indole-2-
carboxylic acid (7a)
5.3.4. (E)-4,6-Dichloro-3-(3-oxo-2-phenyl-isoxazolidin-
4-ylidenemethyl)-1H-indole-2-carboxylic acid (17)
Compound 17 was prepared from 6 (0.95 g, 2.5
mmol) and 14d according to general procedure A. The
crude residue was purified by flash chromatography
using CH2Cl2/MeOH (9:1) as elution solvent to yield
compound 17 as a pale-yellow solid (0.16 g, 16%): m.p.
Compound 7a was prepared from 6 (0.46 g, 1.75
mmol) and 14a according to general procedure A. The
crude compound was crystallised from AcOEt/hexane
to yield compound 7a as a pale-yellow solid (0.26 g,
27%): m.p. \240 °C. IR (Nujol): wmax 3400–3200,
1715, 1688, 1659 cm−1. MS; m/e: 502 (M+1). 1H
NMR (DMSO-d6): l 12.6 (bs, 1H), 7.81 (t, 1H), 7.58
(d, 2H), 7.48 (d, 1H), 7.41 (t, 2H), 7.30 (d, 2H), 7.17 (t,
1H), 4.53 (d, 2H), 1.20 (s, 9H). Anal. (C24H21Cl2N3O5):
C, H, N.
\220 °C. IR (Nujol): wmax 3304, 1672, 1645 cm−1
.
1
MS; m/e: 403 (M+1). H NMR (DMSO-d6): l 13.89
(bs, 1H), 12.59 (s, 1H), 7.85 (t, 1H), 7.74 (dd, 2H), 7.47
(d, 1H), 7.45 (t, 2H), 7.30 (d, 1H), 7.20 (t, 1H), 5.07 (d,
2H). Anal. (C19H12Cl2N2O4): C, H, N.
The liquor mother was evaporated and purified by
flash chromatography using cyclohexane/EtOAc as elu-
tion solvent to yield compounds 8 and 9.
5.4. General procedure B for the preparation of
compounds 14a and 14b
8: Yellowish solid. M.p.=190 °C. IR (Nujol): wmax
3321, 1728, 1709, 1674 cm−1. MS; m/e: 530 (M+1).
1H NMR (CDCl3): l 9.27 (bs, 1H), 7.96 (t, 1H), 7.68
(d, 2H), 7.38 (t, 2H), 7.35 (d, 1H), 7.20 (d, 1H), 7.17 (m,
1H), 6.59 (d, 2H), 3.93 (s, 3H), 1.34 (q, 2H), 1.32 (t,
3H), 1.24 (s, 9H). Anal. (C26H25Cl2N3O5): C, H, N.
9: White solid. M.p.=220 °C. IR (Nujol): wmax 3279,
1792, 1728, 1680 cm−1. MS; m/e: 502 (M+1). 1H
NMR (DMSO-d6): l 13.00 (bs, 1H), 7.59 (d, 1H), 7.52
(dd, 2H), 7.44 (d, 1H), 7.40 (t, 2H), 7.17 (tt, 1H), 6.27
(d, 1H), 3.95 (ddd, 1H), 3.91 (dd, 1H), 3.48 (dd, 1H),
1.19 (s, 9H). Anal. (C24H21Cl2N3O5): C, H, N.
To a stirred mixture of 1-tert-butoxycarbonyl-2-
phenylhydrazine (12) (1 eq.) and K2CO3 (1.7 eq.) in
DMF was added dropwise the appropriate acyloyl
chloride (1.4 eq.) and the reaction mixture was stirred
at reflux for 9 h and at r.t. overnight. The mixture was
diluted with Et2O and washed with water; then the
organic layer was separated, dried over Na2SO4, and
evaporated in vacuo. The crude residue was purified by
flash chromatography using cyclohexane/EtOAc (8:2)
as elution solvent. Notably, when the previous reaction
was run in two steps isolating intermediates 13a and
13b, compounds 14a and 14b were obtained in higher
yields.
5.3.2. (E)-3-(1-tert-Butoxycarbonyl-3-oxo-2-phenyl-
tetrahydro-pyridazin-4-ylidenemethyl)-4,6-dichloro-1H-
indole-2-carboxylic acid (7b)
5.4.1. 3-Oxo-2-phenyl-pyrazolidine-1-carboxylic acid
tert-butyl ester (14a)
Compound 7b was prepared from 6 (0.43 g, 1.1
mmol) and 14b according to general procedure A. The
crude compound was crystallised from AcOEt/hexane
at 0 °C to yield compound 7b as a pale-yellow solid
(0.2 g, 35%): m.p. \240 °C. IR (Nujol): wmax 1717,
1672, 1657 cm−1. MS; m/e: 516 (M+1). 1H NMR
(DMSO-d6, 60 °C): l 13.5 (bs, 1H), 12.8 (bs, 1H), 8.0
Compound 14a was prepared from 12 (5.6 g, 27
mmol) and 3-chloropropionyl chloride according to
general procedure B and obtained as a white solid (5.4
g, 75%): m.p.=150 °C. IR (Nujol): wmax 1773, 1692
1
cm−1. MS; m/e: 263 (M+1). H NMR (DMSO-d6): l
7.45 (m, 2H), 7.35 (m, 2H), 7.12 (m, 1H), 4.07 (t, 2H),
2.74 (t, 2H), 1.26 (s, 9H). Anal. (C14H18N2O3): C, H, N.

796
R. Di Fabio et al. / Il Farmaco 56 (2001) 791–798
5.4.2. 3-Oxo-2-phenyl-tetrahydro-pyridazine-1-
carboxylic acid tert-butyl ester (14b)
tion solvent to yield compound 14d as an oil (2.5 g,
82%): IR (Nujol): wmax 1697 cm−1. MS; m/e: 164
(M+1). H NMR (CDCl3): l 7.69 (m, 2H), 7.37 (m,
2H), 7.14 (tt, 1H), 4.52 (t, 2H), 3.00 (t, 2H). Anal.
(C9H9NO2): C, H, N.
1
Compound 14b was prepared from 12 (1 g, 4.8
mmol) and 4-bromobutyryl chloride according to gen-
eral procedure B and obtained as a white solid (0.951 g,
70%): m.p.=69–71 °C. IR (Nujol): wmax 1771, 1690
1
cm−1. MS; m/e: 277 (M+1). H NMR (DMSO-d6): l
5.5. General procedure C for the preparation of
compounds 10 and 16
7.55 (d, 2H), 7.38 (t, 2H), 7.19 (t, 1H), 4.5–3.0 (broad,
2H), 2.5–1.0 (broad, 13H). Anal. (C15H20N2O3): C, H,
N.
A solution of compounds 7a and 7b (1 eq.) in tri-
fluoroacetic acid/CH2Cl2 was stirred for 1 h, then the
solvent was evaporated and the crude residue was
purified by trituration in Et2O.
5.4.3. 1-Methyl-2-phenyl-3-pyrazolidinone (14c)
A solution of compound 14a (4.6 g, 17.5 mmol) in
trifluoroacetic acid/CH2Cl2 was stirred for 1 h, then the
solvent was evaporated and the crude residue was
dissolved in AcOEt and washed with a saturated solu-
tion of NaHCO3 and water. The organic layer was
separated, dried over Na2SO4, and evaporated in vacuo
to obtain 2-phenyl-3-pyrazolidinone (2.6 g, 92%). The
latter was dissolved in DMF (35 ml), cooled at
−20 °C, then methyl trifluoromethanesulfonate (2.54
ml, 23.2 mmol) was added dropwise. The reaction
mixture was slowly warmed at r.t. and stirred for 3 h.
The solution was diluted with Et2O and washed with a
saturated solution of NH4Cl and water, then the or-
ganic layer was separated, dried over Na2SO4, and
evaporated in vacuo. The crude residue was purified by
flash chromatography using AcOEt/cyclohexane (1:1)
as elution solvent to yield compound 14c as a white
foam (0.75 g, 25%): IR (Nujol): wmax 1697 cm−1. MS;
5.5.1. 4,6-Dichloro-3-(5-oxo-1-phenyl-pyrazolidin-4-
ylidenemethyl)-1H-indole-2-carboxylic acid (10)
Compound 10 was prepared from 7a (0.255 g, 0.51
mmol) according to general procedure C and obtained
as a solid (0.18 g, 87%): m.p. \250 °C. IR (Nujol):
wmax 3250–3150, 1717, 1684 cm−1. MS; m/e: 402 (M+
1
1). H NMR (DMSO-d6): l 13.70 (bs, 1H), 12.49 (bs,
1H), 7.90 (d, 2H), 7.70 (t, 1H), 7.46 (d, 1H), 7.39 (t,
2H), 7.27 (d, 1H), 7.12 (t, 1H), 6.45 (bs, 1H), 3.83 (bs,
2H). Anal. (C19H13Cl2N3O3): C, H, N.
5.5.2. 4,6-Dichloro-3-(3-oxo-2-phenyl-tetrahydro-
pyridazin-4-ylidenemethyl)-1H-indole-2-carboxylic acid
(16)
Compound 16 was prepared from 14b (0.12 g, 0.24
mmol) according to general procedure C and obtained
as a solid (0.05 g, 50%): m.p. \250 °C. IR (Nujol):
1
m/e: 177 (M+1). H NMR (CDCl3): l 7.78 (d, 2H),
1
7.37 (t, 2H), 7.13 (tt, 1H), 3.8–3.6 (bm, 1H), 3.3–3.0
(bm, 2H), 2.7–2.4 (bm, 1H), 2.65 (s, 3H). Anal.
(C10H12N2O): C, H, N.
wmax 3277, 1672, 1612 cm−1. MS; m/e: 416 (M+1). H
NMR (DMSO-d6): l 13.50 (bs, 1H), 12.36 (s, 1H), 8.0
(t, 1H), 7.63 (d, 2H), 7.43 (d, 1H), 7.33 (t, 2H), 7.22 (d,
1H), 7.12 (t, 1H), 6.02 (t, 1H), 3.35 (m, 2H), 3.10 (m,
2H). Anal. (C20H15Cl2N3O3): C, H, N.
5.4.4. 2-Phenyl-3-isoxazolidinone (14d)
To a suspension of 5% Rh/C (0.13 g) in THF (23 ml)
was added nitrobenzene (5 g, 40.6 mmol). The mixture
was cooled at 0 °C and hydrazine monohydrate (2 g,
40.6 mmol) was added portionwise, maintaining the
temperature below 25 °C throughout the addition. Af-
ter that the reaction mixture was stirred at 25 °C for 2
h, then filtered and concentrated in vacuo. The residue
was dissolved in DMF (30 ml) and cooled at −5 °C,
then K2CO3 (5.6 g) and 3-chloropropionyl chloride (3.8
ml, 39.6 mmol) were added. The reaction mixture was
stirred at 25 °C for 1 h, then diluted with Et2O and
washed with water. The organic layer was separated,
dried over Na2SO4, and evaporated in vacuo. The crude
residue was purified by crystallisation from AcOEt and
cyclohexane to yield N-(3-chloropropionyl)-N-phenyl-
hydrohylamine (4 g, 46%). The latter was dissolved in
acetone (150 ml), then K2CO3 (2.5 g) was added. The
reaction mixture was stirred for 6 h, then concentrated
in vacuo. The crude residue was purified by flash
chromatography using AcOEt/cyclohexane (8:2) as elu-
5.6. Biology
5.6.1. Binding assay
NCEs were dissolved at a 5 mM concentration in
100% DMSO and tested at seven different concentra-
tions in duplicate in the [3H]-glycine displacement
experiments and, in the other binding assays, at five
concentrations (from 10 nM to 100 mM). A reference
compound was always included as internal control.
Affinity to the glycine-binding site was measured by
inhibition of the binding of [3H]-glycine to crude synap-
tic membranes prepared from adult rat cerebral cortex,
as described by Kishimoto et al. [27]. Incubation (20
min at 4 °C) was carried out in 50 mM Tris/citrate (pH
7.10) using 20 nM [3H]-glycine. Data for displacement
experiments, performed to determinate the inhibition
constants (Ki) of the displacer ligands, were analysed
using the nonlinear curve-fitting software LIGAND [32].
The Ki values were measured from at least six-point

R. Di Fabio et al. / Il Farmaco 56 (2001) 791–798
797
inhibition curves and are geometric means of at least
three different experiments.
dent’s t-test for unpaired data using the GBSTAT 5.3
program (Dynamic Microsystem).
5.6.2. Anticon6ulsant acti6ity
NCEs were evaluated in vivo by assessing their anti-
convulsant effect [30] when convulsions were induced in
male CD-1 mice (18–29 g) by i.c.v. injection of NMDA
(1 nmol/mouse), 1 min after i.v. administration of the
NCE. Animals were observed for the occurrence of
generalised seizures during the first 30 min after the
treatment with NMDA and were considered protected
if convulsions did not occur within this period. The
percentage of animals showing anticonvulsant activity
in each treatment group was recorded and ED50 values
were estimated along with their 95% confidence limits.
Acknowledgements
The authors would like to thank the personnel of the
Mass Spectrometry Laboratory, the NMR Spec-
troscopy Laboratory and the Pharmaceutical Develop-
ment Department for analytical data. Finally, thanks
are due to Dr. Mugnaini, Mr. E. Valerio and Mr. S.
Costa for the in vitro and in vivo pharmacological
characterisation.
References
5.6.3. Neuroprotecti6e acti6ity
[1] D.W. Choi, S.M. Rothman, Annu. Rev. Pharmacol. Toxicol. 31
(1991) 171–182.
[2] S.H. Graham, J. Chen, F.R. Sharp, P.R. Simon, J. Cerebr.
Blood Flow Metab. 13 (1993) 88–97.
[3] J. McCullogh, Br. J. Clin. Pharmacol. 34 (1992) 106–114.
[4] D.W. Choi, Cerebrovasc. Brain Metab. Rev. 2 (1990) 105–147.
[5] B. Meldrum, S. Garthwaite, Trends Pharmacol. Sci. 11 (1990)
379–387.
[6] D.W. Choi, J. Neurosci. 10 (1990) 2493–2501.
[7] B. Meldrum, Cerebrovasc. Brain Metab. Rev. 2 (1990) 27–57.
[8] J.A. Zivin, Drugs 3 (1997) 83–89.
[9] G.L. Collingridge, J.C. Watkins (Eds.), The NMDA Receptor,
2nd ed., IRL Press, Oxford, 1994.
[10] J.W. Johnson, P. Ascher, Nature 325 (1987) 529–531.
[11] N.W. Kleckner, R. Dingledine, Science 241 (1988) 835–837.
[12] J.A. Kemp, P.D. Leeson, Trends Pharmacol. Sci. 14 (1993)
20–25.
[13] P.D. Leeson, L.L. Iversen, J. Med. Chem. 37 (1994) 4053–4067.
[14] L.L. Iversen, J.A. Kemp, in: G.L. Collingridge, J.C. Watkins
(Eds.), The NMDA Receptor, 2nd ed., IRL Press, Oxford, 1994,
pp. 469–486 (chap. 20).
5.6.3.1. MCAo (distal MCA occlusion). The experimen-
tal procedure was performed according to Tamura et
al. [31], with minor modifications. Male Sprague–Daw-
ley rats (280–350 g) were anaesthetised with chloral
hydrate (400 mg/kg, i.p.). The animals were maintained
normothermic by means of a heating pad and placed
under an operating microscope. A skin incision was
made, the temporalis muscle was reflected and a
craniectomy was performed by drilling at the junction
between the medial wall and the roof of the infero-tem-
poral fossa. The dura was opened and the MCA was
permanently occluded by electrocoagulation proximal
to the frontal branch. The muscles and the skin were
reported to the original position and the skin incision
was closed. Rats were allowed to recover for 24 h
before the evaluation of the ischaemic damage.
[15] R. Di Fabio, G. Gaviraghi, A. Reggiani, La Chimica e l’Indus-
tria 78 (1996) 283–289.
[16] M.E. Tranquillini, A. Reggiani, Expert Opin. Investig. Drugs 8
(1999) 1837–1848.
[17] F. Micheli, A. Pasquarello, Investig. Drugs 3 (2000) 318–328.
[18] D. Donati, F. Micheli, Exp. Opin. Ther. Patients 10 (2000)
667–673.
[19] M. Antolini, D. Donati, F. Micheli, Curr. Opin. Investig. Drugs
1 (2000) 230–235.
[20] R. Di Fabio, A.M. Capelli, N. Conti, A. Cugola, A. Feriani, P.
Gastaldi, G. Gaviraghi, C.T. Hewkin, F. Micheli, A. Missio, M.
Mugnaini, A. Pecunioso, A.M. Quaglia, E. Ratti, L. Rossi, G.
Tedesco, D.G. Trist, A. Reggiani, J. Med. Chem. 40 (1997)
841–850.
[21] R. Di Fabio, A. Cugola, D. Donati, A. Feriani, G. Gaviraghi, E.
Ratti, D.G. Trist, A. Reggiani, Drugs Fut. 23 (1998) 61–69.
[22] C.T. Hewkin, R. Di Fabio, N. Conti, A. Cugola, P. Gastaldi, F.
Micheli, A.M. Quaglia, Arch. Pharm. Med. Chem. 332 (1999)
55–58.
[23] F. Micheli, R. Di Fabio, A.M. Capelli, A. Cugola, O. Curcu-
ruto, A. Feriani, P. Gastaldi, G. Gaviraghi, C. Marchioro, A.
Orlandi, A. Pozzan, A.M. Quaglia, A. Reggiani, F. van Amster-
dam, Arch. Pharm. Med. Chem. 332 (1999) 73–80.
[24] S.A. Giacobbe, R. Di Fabio, D. Baraldi, A. Cugola, D. Donati,
Synth. Commun. 29 (1999) 3125.
5.6.3.2. Histology. Animals were sacrificed 24 h after
MCAo, the brain removed carefully and sectioned
coronally at 0.5 mm intervals by using a motorised
vibroslice. The brain slices were immersed in a solution
of 2% triphenyltetrazolium chloride (TTC) at 37 °C for
20 min and then stored in 4% neutral buffered forma-
lin. The TTC staining clearly distinguishes between the
normal tissue (stained in red) and the ischaemic area
(white, not stained).
The area of cerebral damage in each coronal section
was assessed by using an imagine analyser (Imaging
Research Inc., Canada). The infarct volumes were cal-
culated by using the trapezoidal rule method.
5.6.3.3. Drugs and statistical analysis. The compounds
were dissolved in DMSO and then diluted to a final 3%
DMSO concentration with distilled water. Solutions
were prepared fresh at the day of the experiment.
Values are expressed as mean9SEM. Statistical dif-
ferences between groups were analysed using the Stu-

798
R. Di Fabio et al. / Il Farmaco 56 (2001) 791–798
[25] R. Di Fabio, N. Conti, E. De Magistris, A. Feriani, S. Provera,
F.M. Sabbatini, A. Reggiani, L. Rovatti, J. Med. Chem. 42
(1999) 3486–3493.
[26] S.A. Giacobbe, D. Baraldi, R. Di Fabio, Biorg. Med. Chem.
Lett. 8 (1998) 1689–1692.
alternative synthetic route. These results will be published in due
time.
[29] All the compounds reported in Table 1 are contained in a Glaxo
Patent: A. Cugola, R. Di Fabio, G. Pentassuglia, WO 9510517,
1995.
[27] H. Kishimoto, J.R. Simon, M.H. Aprison, J. Neurochem. 37
(1981) 1015–1024.
[30] C. Chiamulera, S. Costa, A. Reggiani, Psycopharmacology 112
(1990) 551–552.
[28] The authors would like to thank Dr Andrea Bozzoli, Dr Zadeo
Cimarosti and Dr Paolo Maragni (Chemical Development De-
partment) for the isolation of compound 9. Notably, they were
able to prepare the corresponding syn lactone following an
[31] A. Tamura, D.I. Graham, J. McCullogh, G.M. Teasdale, J.
Cerebr. Flow Metab. 1 (1981) 53–60.
[32] P.D. Munson, D. Rodbard, Anal. Biochem. 107 (1980) 220–
239.