Synthesis of 11-aryl-5H-imidazo[2,1-c][1,4]benzodiazepines and their benzodiazepine and A1 adenosine binding activity

Article abstract of DOI:10.1016/S0014-827X(01)01133-8

In the context of a research program aimed at elucidating the properties of the 5H-imidazo[2,1-c][1,4]benzodiazepine system, a series of 11-aryl-5H-imidazo[2,1-c][1,4]benzodiazepines (3a-i) and their 10,11-dihydro-derivatives (4a-i) has been synthesized. The synthetic strategy includes the preparation of the aryl-[1-(2-nitrobenzyl)-1H-imidazol-2-yl]methanones (5a-i) followed by their reduction and subsequent cyclization. Affinities of compounds 3a-i and 4a-i for central benzodiazepine as well as for adenosine A₁-receptors were determined by radioligand binding assays. Among the unsaturated analogues, the highest activity at both receptors is displayed by 11-(2-thienyl) derivative 3e. The hydrogenated analogues 4a-i do not exhibit considerable binding affinity either for central benzodiazepine or for adenosine A₁-receptors.

Full text of DOI:10.1016/S0014-827X(01)01133-8

www.elsevier.com/locate/farmac
Il Farmaco 56 (2001) 771–778
Synthesis of 11-aryl-5H-imidazo[2,1-c][1,4]benzodiazepines and
their benzodiazepine and A1 adenosine binding activity
Sabrina Castellano a,*, Laura Zorzin a, Chiara Florio b, Fabiana Frausin b,
Giorgio Stefancich a
a Dipartimento di Scienze Farmaceutiche, P.le Europa 1, 34127 Trieste, Italy
b Dipartimento di Scienze Biomediche, 6ia Giorgieri 7, 34127 Trieste, Italy
Received 16 June 2000; accepted 30 May 2001
Abstract
In the context of a research program aimed at elucidating the properties of the 5H-imidazo[2,1-c][1,4]benzodiazepine system,
a series of 11-aryl-5H-imidazo[2,1-c][1,4]benzodiazepines (3a–i) and their 10,11-dihydro-derivatives (4a–i) has been synthesized.
The synthetic strategy includes the preparation of the aryl-[1-(2-nitrobenzyl)-1H-imidazol-2-yl]methanones (5a–i) followed by
their reduction and subsequent cyclization. Affinities of compounds 3a–i and 4a–i for central benzodiazepine as well as for
adenosine A1-receptors were determined by radioligand binding assays. Among the unsaturated analogues, the highest activity at
both receptors is displayed by 11-(2-thienyl) derivative 3e. The hydrogenated analogues 4a–i do not exhibit considerable binding
affinity either for central benzodiazepine or for adenosine A1-receptors. © 2001 Elsevier Science S.A. All rights reserved.
Keywords: Imidazobenzodiazepines; Benzodiazepine receptors; Adenosine receptors; Radioligand binding
1. Introduction
As a further development on this area, we previously
reported the synthesis of the novel 5H-imidazo[2,1-
c][1,4]benzodiazepine tricyclic ring system [2].
Derivatives of type 1 and 2 (Fig. 1) were subjected to
investigations to evaluate their pharmacological proper-
ties. Some of the tested compounds were comparable to
chlordiazepoxide in sedative and muscle-relaxant activi-
ties, but did not show considerable binding affinity for
central benzodiazepine receptors (BzR) [3].
The effects exerted in vivo by these compounds led us
to proceed with further investigations in this field. In
order to evaluate the effect of an additional aromatic or
heteroaromatic group, a series of 11-aryl-5H-imi-
dazo[2,1-c][1,4]benzodiazepines (3a–i) and their 10,11-
dihydro-derivatives (4a–i) have been designed and
synthesized (Fig. 2).
Classic benzodiazepine derivatives have attracted the
attention of researchers owing to their interesting phar-
macological activities and their low toxicity. They are
widely used as anxiolytics, antidepressants, hypnotics,
anticonvulsants and muscle relaxants. Annelation of
the 1,4-benzodiazepine nucleus, formally obtained by
fusion of a five-membered heterocycle on the diazepine
skeleton, led to a number of compounds sharing attrac-
tive pharmacological properties [1].
As preliminary in vivo assays (data not shown) indi-
cated that these compounds were also able to induce
sedative and muscle relaxant effects on mice, the title
compounds were first studied for their interaction with
BzR.
Fig. 1.
Furthermore, we chose to evaluate the affinity of the
title compounds also towards the adenosine A1-recep-
tor. In fact, besides the GABA receptor system, the
* Correspondence and reprints
E-mail address: facfarm@univ.trieste.it (S. Castellano).
0014-827X/01/$ - see front matter © 2001 Elsevier Science S.A. All rights reserved.
PII: S0014-827X(01)01133-8

772
S. Castellano et al. / Il Farmaco 56 (2001) 771–778
Fig. 2. Structures of 11-aryl-5H-imidazo[2,1-c][1,4]benzodiazepines (3a–i) and 10,11-dihydro-derivatives (4a–i).
purinergic system has also been involved in the central
mechanism of action of benzodiazepines [4].
their potency to displace [3H]flumazenil (at 0.5 nM)
from its binding site at rat brain cortical membranes.
Screening was repeated at six concentrations (1, 3, 10,
30, 100, 300 mM) for compounds 3a,d–f, which pre-
sented at 100 mM a percentage of displacement higher
than 50% and the Ki values were calculated. The
affinity for the benzodiazepine recognition site was then
investigated in the presence of GABA and the GABA
shift (Ki compound/Ki compound+GABA) was deter-
mined (Table 3).
2. Chemistry
The 11-aryl-imidazobenzodiazepines 3a–i and 4a–i
(Table 2) were obtained by reduction and subsequent
cyclization of the corresponding aryl-[1-(2-nitrobenzyl)-
1H-imidazol-2-yl]methanones (5a–i) (Table 1) obtained
as reported below. In particular, compounds 3a–f,h,i
were obtained by reduction of 5a–f,h,i with iron(II)
sulfate and ammonium hydroxide [2]. The 10,11-dihy-
droderivatives 4a–f,h,i were prepared by reducing the
nitro-methanones 5a–f,h,i with iron powder in boiling
glacial acetic acid [5] (Scheme 1).
When this method was applied to 5g, the 11-(4-acet-
ylaminophenyl) - 5H - imidazo[2,1 - c][1,4]benzodiazepine
(6) was obtained, which, by hydrolysis with 20%
NaOH, furnished 3g. Compound 4g was prepared
through the reductive cyclization of 5g with titanium
trichloride [6] (Scheme 2).
The aryl-[1-(2-nitrobenzyl)-1H-imidazol-2-yl]metha-
nones 5a–g were obtained by reacting the 1-(2-ni-
trobenzyl)-1H-imidazole (7) [2] with the suitable aroyl
chlorides in the presence of triethylamine (Scheme 3).
Synthesis of 5h was carried out reacting 5c with imida-
zole in the presence of sodium hydride. Compound 5i
was obtained from 5c by using an excess of
methylpiperazine (Scheme 4).
3.2. Biological e6aluation of compounds 3a–i and 4a–i
as ligands for A1-adenosine binding sites
The title compounds were further investigated at
three concentrations (1, 10, 100 mM) in radioligand
binding assay for affinity at A1-adenosine receptors in
rat brain cortical membrane preparations. The A1-selec-
tive agonist [3H]2-chloro-N6-cyclopentyladenosine
(CCPA) was used at 1 nM as ligand. Screening was
repeated at seven concentrations (10, 30, 50, 100, 300,
500, 750 mM) for compounds 3a,e, which presented at
100 mM a percentage of displacement higher than 40%
and the Ki values were calculated (Table 4).
3. Results
3.1. Biological e6aluation of compounds 3a–i and 4a–i
as ligands for benzodiazepine binding sites
All compounds were screened at four concentrations
(1, 10, 30, 100 mM) in radioligand binding assay for
Scheme 1. Synthesis of compounds 3a–f,h,i and 4a–f,h,i. (i)
FeSO4·7H2O, NH4OH, H2O, EtOH; (ii) Fe, AcOH.

Table 1
Yields, physical, analytical and spectroscopic data of compounds 5a–i
No. Formula (MW)
Yield (%)
M.p. (°C)
crystallization
solvent
MS m/z [M+] IR
(cm−1)w(CꢀO)
1H NMR: l (ppm), (J in Hz)
5a
5b
5c
5d
5e
C17H13N3O3
(307.30)
C18H15N3O4
(337.33)
C17H12N3FO3
(325.29)
C17H12N3FO3
(325.29)
52
68
65
80
70
90 benzene:
petroleum ether
80–82 ethanol
307
337
325
325
313
1641
1625
1638
1648
1608
6.06 (s, 2H, CH2), 6.74–7.78 (m, 1H, ArH), 7.22–7.63 (m, 7H, ArH), 8.14–8.28 (m, 3H,
ArH).
3.83 (s, 3H, CH3), 6.08 (s, 2H, CH2), 6.73–6.81 (m, 1H, ArH), 7.29–7.62 (m, 4H, ArH), 8.19
(dd, J=8.0 and 1.7, 1H, ArH), 8.30 (d, J=8.9, 2H, ArH), 8.45 (d, J=8.9, 2H, ArH).
6.05 (s, 2H, CH2), 6.72–6.79 (m, 1H, ArH), 7.09–7.60 (m, 6H, ArH), 8.19 (dd, J=8.1 and
J=1.5, 1H, ArH), 8.31–8.42 (m, 2H, ArH).
6.10 (s, 2H, CH2), 6.74–6.81 (m, 1H, ArH), 7.08–7.73 (m, 8H, ArH), 8.18 (dd, J=7.9 and
J=1.5, 1H, ArH).
6.09 (s, 2H, CH2), 6.67–6.75 (m, 1H, H benzene), 7.11–7.22 (m, 2H, H thiophene and H
imidazole), 7.35–7.58 (m, 3H, H thiophene and H benzene), 7.71 (dd, J=4.8 and J=1.0,
1H, H thiophene), 8.19 (dd, J=7.6 and J=1.6, 1H, H benzene), 8.46 (dd, J=3.0 and
J=1.0, 1H, H thiophene).
95 ethanol
/
92–93 benzene:
petroleum ether
98–99 benzene:
petroleum ether
C
15H11N3O3S
(313.33)
5
)
5f
C
15H11N3O4
90
103–104 ethanol
297
1631
6.07 (s, 2H, CH2), 6.58–6.62 (m, 1H, H furane), 6.68–6.76 (m, 1H, H benzene), 7.22 (s, 1H,
H imidazole), 7.35 (s, 1H, H imidazole), 7.39–7.58 (m, 2H, H benzene), 7.64–7.68 (m, 1H, H
furane), 8.08–8.20 (m, 2H, H benzene and H furane).
(297.27)
7
8
5g
5h
5i
C17H12N4O5
(352.30)
44
143–144 benzene:
petroleum ether
120–123 ethanol:
water
123–125 benzene:
petroleum ether
352
373
405
1641
1637
1622
6.08 (s, 2H, CH2), 6.73–6.82 (m, 1H, ArH), 7.28–7.62 (m, 4H, ArH), 8.21 (dd, J=8.0 and
J=1.7, 1H, ArH), 8.30 (d, J=8.9, 2H, ArH), 8.44 (d, J=8.9, 2H, ArH).
6.05 (s, 2H, CH2), 6.72–6.78 (m, 1H, ArH), 7.25–7.60 (m, 8H, ArH), 7.94 (s, 1H, ArH), 8.17
(dd, J=7.9 and J=1.5, 1H, ArH), 8.42 (d, J=8.5, 2H, ArH).
2.35 (s, 3H, CH3), 2.51–2.58 (m, 4H, H piperazine), 3.35–3.42 (m, 4H, H piperazine), 6.03
(s, 2H, CH2), 6.72–6.79 (m, 1H, ArH), 6.88 (d, J=9.0, 2H, ArH), 7.12–7.54 (m, 4H, ArH),
8.17 (dd, J=7.8 and 1.5, 1H, ArH), 8.28 (d, J=9.0, 2H, ArH).
C
20H15N5O3
28 a
64 a
(373.37)
C22H23N5O3
(405.45)
a From 5c.

Table 2
Yields, physical, analytical and spectroscopic data of compounds 3a–i and 4a–i
No. Formula (MW) Yield % M.p. (°C)
crystallization solvent
MS m/z [M+] 1H NMR: l (ppm), (J in Hz)
3a
3b
3c
3d
3e
3f
C
17H13N3
95 a
79 b
81 b
81 b
98 a
98 a
99 a,c
52 b
80 b
98 a
98 a
98 a
98 a
66 b
60 b
180 ethanol:water
158 ethanol:water
259
289
5.00 (s, 2H, CH2), 7.10–7.67 (m, 9H, ArH), 8.15–8.40 (m, 2H, ArH).
(259.31)
C18H15N3O
(289.33)
3.85 (s, 3H, CH3), 4.98 (s, 2H, CH2), 6.98 (d, J=8.9, 2H, ArH), 7.07 (s, 1H, H imidazole), 7.15–7.51 (m, 5H
ArH), 8.12 (d, J=8.9, 2H, ArH).
5.02 (s, 2H, CH2), 7.08–7.52 (m, 8H, ArH), 8.10–8.22 (m, 2H, ArH).
C
17H12N3F
144–145 ethanol:water 277
206–208 ethanol:water 277
(277.30)
C17H12N3F
(277.30)
C15H11N3S
(265.34)
5.06 (s, 2H, CH2), 7.04–7.58 (m, 9H, ArH), 7.86–7.98 (m, 1H, ArH).
5.01 (s, 2H, CH2), 7.08–7.56 (m, 8H, ArH), 7.98–8.02 (m, 1H, ArH).
164–165 ethanol
151–152 ethanol
220–222 benzene
265
249
274
325
357
261
291
279
279
267
C
15H11N3O
5.00 (s, 2H, CH2), 6.58 (m, 1H, H furane), 7.10 (s, 1H, H imidazole), 7.20–7.58 (m, 6H, H benzene, H
imidazole and H furane), 7.71 (m, 1H, H furane).
3.97 (s, 2H, disappears with D2O, NH2), 5.00 (s, 2H, CH2), 6.73 (d, J=8.7, 2H, ArH), 7.07–7.5 (m, 6H,
ArH), 7.99 (d, J=8.7, 2H, ArH).
(249.27)
C17H14N4
(274.33)
C20H15N5-
(325.38)
C22H23N5
(357.47)
C17H15N3
(261.33)
C18H17N3O
(291.35)
C17H14N3F
(279.31)
C17H14N3F
(279.31)
3g
3h
3i
231–233 benzene:
petroleum ether
208–210 benzene:
ligroin
5.08 (s, 2H, CH2), 7.12–7.57 (m, 10H, ArH), 7.94 (s, 1H, H imidazole), 8.28 (d, J=8.8, 2H, ArH).
/
2.36 (s, 3H, CH3), 2.54–2.62 (m, 4H, H piperazine), 3.33–3.39 (m, 4H, H piperazine), 5.00 (s, 2H, CH2), 6.96
(d, J=8.9, 2H, ArH), 7.06–7.50 (m, 6H, ArH), 8.07 (d, J=8.9, 2H, ArH).
4.37 (d, J=4.8, 1H, disappears with D2O, NH), 4.77 (d, J=14.8, 1H, CHH), 5.03 (d, J=14.8, 1H, CHH),
5.80 (d, J=4.8, 1H, CH), 6.74–7.37 (m, 11H, ArH).
3.77 (s, 3H, OCH3), 4.30 (d, J=4.1, 1H, disappears with D2O, NH), 4.80 (d, J=14.8, 1H, CHH), 5.03 (d,
J=14.8, 1H, CHH), 5.72 (d, J=4.1, 1H, CH), 6.72–7.29 (m, 10H, ArH).
4.31 (d, J=4.1, 1H, disappears with D2O, NH), 4.82 (d, J=14.7, 1H, CHH), 5.04 (d, J=14.7, 1H, CHH),
5.76 (d, J=4.1, 1H, CH), 6.74–7.36 (m, 10H, ArH).
4.04 (d, 1H, J=3.6, disappears with D2O, NH), 5.08–5.24 (m, 2H CH2), 6.12 (d, J=3.6, 1H, CH), 6.65–7.29
(m, 10H, ArH).
4.50 (d, J=4.6, 1H, disappears with D2O, NH), 4.82 (d, J=14.6, 1H, CHH), 5.14 (d, J=14.6, 1H, CHH),
6.05 (d, J=4.6, 1H, CH), 6.75–7.20 (m, 9H, ArH).
4.35 (d, J=4.3, 1H, disappears with D2O, NH), 5.00–5.12 (m, 2H, CH2), 5.86 (d, J=4.3, 1H, CH), 6.06 (m,
1H, H furane), 6.25 (m, 1H, H furane), 6.74–7.20 (m, 6H, H imidazole and H benzene), 7.31 (m, 1H, H
furane).
4a
4b
4c
4d
4e
4f
225 benzene
223–224 ethanol
141–142 benzene
173–174 ethanol
5
)
7
8
C
15H13N3S
184–185 ethyl
acetate:petroleum ether
167–171 benzene:ligroin 251
(267.34)
C15H13N3O
(251.28)
4g
4h
4i
C17H16N4
(276.34)
C20H17N5
(327.38)
C22H25N5
(359.47)
96 a
60 b
83 b
168–171 toluene:
petroleum ether
220–222 toluene:
petroleum ether
148–150 ethyl acetate:
petroleum ether
276
327
359
3.70 (br s, 2H, disappears with D2O, NH2), 4.30 (br s, 1H disappears with D2O, NH), 4.79 (d, J=14.7, 1H,
CHH), 5.05 (d, J=14.7, 1H, CHH), 5.66 ( br s, 1H, CH), 6.58–7.30 (m, 10H, ArH).
4.32 (d, J=3.8, 1H disappears with D2O, NH), 4.86 (d, J=14.8, 1H, CHH), 5.08 (d, J=14.8, 1H, CHH),
5.81 (d, J=3.8, 1H, CH), 6.65–7.25 (m, 12H, ArH), 7.81 (s, 1H, H imidazole).
2.34 (s, 3H, CH3), 2.54–2.60 (m, 4H, H piperazine), 3.18–3.25 (m, 4H, H piperazine), 4.33 (d, J=4.0, 1H,
disappears with D2O, NH), 4.80 (d, J=14.7, 1H, CHH), 5.06 (d, J=14.7, 1H, CHH), 5.70 (d, J=4.0, 1H,
CH), 6.70–7.61 (m, 10H, ArH).
a Purified by recrystallization.
b Purified by column chromatography on alumina (eluent: chloroform).
c Yield of hydrolysis.

S. Castellano et al. / Il Farmaco 56 (2001) 771–778
775
Scheme 2. Synthesis of compound 3g and compound 4g (i) Fe, AcOH; (ii) NaOH 20%; (iii) TiCl3, HCl, H2O, AcOH.
administration of the tested compounds. However, an
involvement of the purinergic system in the mechanism
of action of these compounds can still be put forward
4. Discussion
The binding results show that the 10,11-dihydro-11-
aryl-5H-imidazo[2,1-c][1,4]benzodiazepines 4a–i do not
exhibit noticeable binding affinity for central benzodi-
azepine receptors. Among the dehydrogenated ana-
logues, derivatives with unsubstituted aryl moiety at
position 11 (3a,e,f) exhibited a benzodiazepine affinity
ranging from 14.0 to 47.7 mM (Table 3). The 11-(o-
fluoro)phenyl derivative 4d displayed a moderate in-
crease in binding activity in comparison to the
unsubstituted parent compound 4a, whereas para sub-
stitution of the phenyl ring is endowed with a drastic
decrease in affinity. In opposition, in compounds lack-
ing the aryl moiety at position 11, only two hydro-
Scheme 3. Synthesis of aryl-[1-(2-nitrobenzyl)-1H-imidazol-2-
yl]methanones 5a–g (i) ArCOCl, N(C2H5)3, acetonitrile.
genated derivatives of type
2
(X=Y=H and
X=Y=OCH3) showed some affinity, with IC50 values
of 22 and 14 mM, respectively (IC50 diazepam=0.05
mM) [3].
Since GABA shift values of the compounds 3a,d–f
are higher than 1, it can be assumed that they may act
at benzodiazepine receptor as agonists. In fact the
GABA shift indicates a different trend in the efficacy of
a benzodiazepine receptor ligand: in this experiment the
affinity for the benzodiazepine recognition site is deter-
mined in the presence and in the absence of GABA.
The addition of GABA increases the affinity of various
benzodiazepine receptor agonists while the potency of
various antagonists remained unaltered [7,8].
Scheme 4. Synthesis of aryl-[1-(2-nitrobenzyl)-1H-imidazol-2-
yl]methanones 5h,i (i) NaH, dimethylacetamide (ii) methylpiperazine,
120 °C.
Table 4 shows that, among all the tested compounds
the highest affinity towards adenosine A1-receptor is
displayed by compound 3e, with a Ki value of 7.07 mM.
The same compound was the most active at the benzo-
diazepine receptor. Compound 3a, bearing a 11-unsub-
stituted phenyl ring, exhibits a Ki value of 15.17
whereas all the other compounds, both saturated and
unsaturated, do not display notable binding activity.
Taken together, the results of the binding studies do
not appear to support the hypothesis of a direct in-
volvement of benzodiazepine- or adenosine A1-recep-
tors in mediating the behavioral changes induced by
Table 3
Binding constants at benzodiazepine receptor for compounds 3a,d–f
Compound
Ki9SEM (mM) a
GS b
3a
3d
3e
3f
47.799.33
24.793.42
14.090.93
33.491.02
0.018590.0025
1.5890.15
1.790.2
1.4390.5
1.4490.18
1.5590.09
Diazepam
a Ki values are means of four determinations.
b GABA-shift=Ki compound/Ki (compound+100 mM GABA)
performed in four independent experiments.

776
S. Castellano et al. / Il Farmaco 56 (2001) 771–778
Table 4
form as eluent. The first eluates were collected and the
solvent was evaporated to afford compounds 5a–g.
Yields, melting points, analytical and spectroscopic
data are reported in Table 1.
Binding constants at A1-adenosine receptor for compounds 3a,e
Compound
Ki (mM) a
Theophylline
3a
3e
2.53
15.17
7.07
5.1.2. [4-(1H-Imidazol-1-yl)phenyl]-[1-(2-nitrobenzyl)-
1H-imidazol-2-yl]methanone (5h)
To a suspension of sodium hydride (1.08 g, 45.0
mmol) in dry dimethylacetamide (50 ml) a solution of
5c (9.75 g, 30.0 mmol) and imidazole (3.06 g, 45.0
mmol) in the same solvent (50 ml) was added dropwise;
the mixture was stirred at 100 °C for 18 h. The reaction
mixture was cooled, poured into water (300 ml), aci-
dified to pH 2.0 with concentrated hydrochloric acid
and then extracted with chloroform (3×50 ml). The
combined organic layers were discarded while the
aqueous phase was made alkaline by adding sodium
hydrogen carbonate and extracted with chloroform
(3×50 ml), washed with a saturated sodium chloride
solution and dried over anhydrous sodium sulfate.
Evaporation of the solvent gave an oil which was
purified by chromatography (alumina) with chloroform
as eluent. The first eluates were collected and solvent
evaporated to give compound 5h. Yield, melting point,
analytical and spectroscopic data are reported in Table
1.
a Ki values are means of two determinations.
on the basis of the ability of benzodiazepine derivatives
to bind to the adenosine equilibrative transport system
[9–11]. As a consequence of the blockade of adenosine
uptake, the extracellular levels of the purine rise, thus
allowing activation of the purinergic system-mediated
responses. Since this mechanism could at least in part
mediate the sedative and muscle relaxant effects, the
title compounds are presently under evaluation for their
putative ability to interfere with the uptake of
adenosine.
5. Experimental
5.1. Chemistry
Melting points were determined on a Bu¨chi 530
apparatus and are uncorrected. Mass spectra data were
determined on a V6-Micromass 7070H mass spectrome-
ter. The infrared spectra were recorded on a Jasco
5.1.3. [4-(4-Methyl-piperazin-1-yl)phenyl]-
[1-(2-nitrobenzyl)-1H-imidazol-2-yl]methanone (5i)
A mixture of 5c (3.25 g, 10.0 mmol) and methylpiper-
azine (5.01 g, 50.0 mmol) was heated at 120 °C for 6 h.
After cooling, the reaction mixture was taken up in
chloroform (300 ml), washed with a saturated sodium
chloride solution and dried over anhydrous sodium
sulfate. Evaporation of the solvent gave an oil which
was purified by chromatography (alumina) with chloro-
form as eluent. The first eluates were collected and
solvent evaporated to give compound 5h. Yield, melting
point, analytical and spectroscopic data are reported in
Table 1.
1
(FT/IR-200) spectrophotometer. The H NMR spectra
were obtained at 200 MHz using a Varian instrument,
using CDCl3 as solvent unless indicated, with TMS as
internal standard. Alumina chromatography was per-
formed using Merck aluminium oxide 90 (0.063–0.200
mm). Elemental analysis were performed by Dr. Emilio
Cebulec at the Chemistry Department of the University
of Trieste and the results of the C, H, N determinations
were within 90.4% of those calculated for the corre-
sponding formulae.
5.1.1. General procedure for preparation of the aryl-
[1-(2-nitrobenzyl)-1H-imidazol-2-yl]methanones (5a–g)
A solution of the appropriate aroyl chloride (33.0
mmol) in dry acetonitrile (30 ml) was added to a
solution of 1-(2-nitrobenzyl)-1H-imidazole (7) (6.1 g,
30.0 mmol) and triethylamine (3.34 g, 33.0 mmol) in the
same solvent (50 ml). The mixture was stirred at 80 °C
overnight and, after cooling at room temperature, the
solid was filtered off. The solution was evaporated to
give an oil which was treated with water (100 ml) and
extracted with chloroform (3×50 ml). The combined
organic layers were washed with a saturated sodium
chloride solution and dried over anhydrous sodium
sulfate. Evaporation of the solvent gave an oil which
was purified by chromatography (alumina) with chloro-
5.1.4. General procedure for preparation of the
5H-imidazo [2,1-c][1,4]benzodiazepines (3a–f,h,i)
A boiling solution of the suitable aryl-[1-(2-nitroben-
zyl)-1H-imidazol-2-yl]methanone
(5a–f,h,i)
(4.31
mmol) in ethanol (20 ml) was added to a suspension of
iron(II) sulfate heptahydrate (12.0 g, 43.1 mmol) in
water (20 ml) and 32% ammonium hydroxide (2 ml).
The reaction mixture was heated under reflux for 4 h
while 32% ammonium hydroxide was dropped (20 ml).
The hot mixture was filtered and the inorganic material
was washed with boiling ethanol. The solution was
partially evaporated and then extracted with chloro-
form (3×100 ml). The combined organic layers were
washed with water followed by a saturated sodium
chloride solution, dried over anhydrous sodium sulfate

S. Castellano et al. / Il Farmaco 56 (2001) 771–778
777
and evaporated to give compounds 3b–d,f,h,i. For
5.1.7. 10,11-Dihydro-11-(4-aminophenyl)-5H-imidazo-
compounds 3a,e the crude residue was dissolved in
toluene (100 ml) and refluxed overnight in the presence
of p-toluensulfonic acid with azeotropic removal of
water by a Dean–Stark trap. After cooling the solution
was washed with a saturated sodium chloride solution,
dried over anhydrous sodium sulfate and evaporated to
give compounds 3a,e. Purification methods, yields,
melting points, analytical and spectroscopic data are
reported in Table 2.
[2,1-c][1,4]benzodiazepine (4g)
A solution of 5g (1.00 g, 2.84 mmol) in acetic acid–
water 1:1 (15 ml) was treated with a 15% titanium
trichloride hydrochloric acid solution (39.0 ml, d=
1.20, 16 mol equiv.) and stirred for 10 min at room
temperature. The reaction mixture was basified with
15% sodium hydroxide aqueous solution and extracted
with dichloromethane–methanol 95:5 (3×70 ml). The
combined organic layers were washed with water, dried
over anhydrous sodium sulfate and evaporated to give
4g. Purification method, yield, melting point, analytical
and spectroscopic data are reported in Table 2.
5.1.5. 11-(4-Aminophenyl)-5H-imidazo[2,1-c]-
[1,4]benzodiazepine (3g)
Iron powder (3.40 g, 60.0 mg-atom) was added
slowly to a solution of 5g (1.76 g, 5.00 mmol) in glacial
acetic acid (100 ml) and the mixture was refluxed
overnight while stirring. After concentration of the
solvent under reduced pressure, the residue was taken
up with ethyl acetate (100 ml) and water (50 ml). The
organic phase was washed twice with water followed by
a saturated sodium chloride solution, dried over anhy-
drous sodium sulfate and evaporated to give the 11-(4-
acetylaminophenyl) - 5H - imidazo[2,1 - c][1,4]benzodia-
zepine (6) (1.17 g, 3.70 mmol, 74% yield), m.p. 242 °C
(ethanol); IR (potassium bromide): w(CꢀO) 1687, NꢁH
5.2. Biochemistry
5.2.1. Benzodiazepine binding assay
Affinity for the benzodiazepine recognition site was
determined in membranes prepared from rat cerebral
cortex according to Lipartiti et al. [12]. In brief, cere-
bral cortices from adult, male Sprague–Dawley rats
were dissected on ice and homogenized in 10 volumes
of ice-cold 0.32 M sucrose in a glass homogenized.
After centrifugation at 1000×g for 20 min at 4 °C, the
pellet was discarded and the supernatant was cen-
trifuged at 20 000×g for 20 min at 4 °C. The resulting
pellet was resuspended in ice-cold distilled water and
centrifuged at 8000×g for 20 min at 4 °C. The super-
natant was collected, centrifuged at 48 000×g for 20
min at 4 °C and the final pellet was stored at −20 °C
for at least 24 h. On the day of the assay, the pellet was
thawed, suspended in 50 mM Tris–HCl, pH 7.4, cen-
trifuged at 20 000×g for 20 min at 4 °C and resus-
pended in Tris–HCl pH 7.4 containing 120 mM NaCl
and 5 mM KCl.
1
3067 cm−1; H NMR (DMSO-d6): l 2.10 (s, 3H, CH3),
5.22 (s, 2H, CH2), 7.15–7.80 (m, 10H, ArH), 10.23 (s,
1H, 1H, disappears with D2O, NH), ms: m/z 316 (M+).
A solution of the acetylderivative 6 (0.50 g, 1.58 mmol)
in ethanol (50 ml) was treated with an aqueous solution
of 20% sodium hydroxide (3 ml) and the reaction
mixture was heated under reflux overnight. After cool-
ing the solution was concentrated under reduced pres-
sure, diluted with water (50 ml) and extracted with
chloroform (3×50 ml). The combined extracts were
washed with water, dried over anhydrous sodium sul-
fate and evaporated to give 3g. Purification method,
yield, melting point, analytical and spectroscopic data
are reported in Table 2.
For binding assay, 0.5 nM [3H]flumazenil (87 Ci/
mmol, New England Nuclear, Boston, MA) was incu-
bated in 96-well MultiScreen-FB microtiter plates
(Millipore) in 250 ml (total volume) of Tris–HCl pH 7.4
containing 120 mM NaCl and 5 mM KCl. Nonspecific
binding was determined in the presence of 3 mM di-
azepam and was 4–7% of total binding. Test com-
pounds were dissolved in dimethyl sulfoxide and diluted
with incubation buffer immediately before use. Final
vehicle concentration in incubation wells was main-
tained constant (1%) in each well. For GABA-shift
measurement, 100 mM GABA was included in the
incubation mixture. Assays were performed in dupli-
cate. The incubation was started by the addition of 50
mg protein/well, carried out for 60 min at 25 °C and
terminated by vacuum filtration on a MultiScreen vac-
uum manifold (Millipore). Wells were rinsed twice with
200 ml of ice-cold Tris–HCl, 50 mM, pH 7.4 and the
flexible plate underdrain was removed. The plate was
placed inside the Wallac cassette for Millipore Multi-
Screen Filtration Plate, SuperMix liquid scintillation
5.1.6. General procedure for preparation of the
10,11-dihydro-5H-imidazo [2,1-c][1,4]benzodiazepines
(4a–f,h,i)
Iron powder (1.70 g, 30.0 mg-atom) was added
slowly to a solution of the suitable aryl-[1-(2-nitroben-
zyl)-1H-imidazol-2-yl]methanone (5) (5.00 mmol) in
glacial acetic acid (100 ml) and the mixture was refluxed
overnight while stirring. After concentration of the
solvent under reduced pressure, the residue was taken
up with ethyl acetate (100 ml) and water (50 ml). The
organic phase was washed twice with water followed by
a saturated sodium chloride solution, dried over anhy-
drous sodium sulfate and evaporated to give 4a–f,h,i.
Purification methods, yields, melting points, analytical
and spectroscopic data are reported in Table 2.

778
S. Castellano et al. / Il Farmaco 56 (2001) 771–778
cocktail (25 ml/well) was added and the plate counted in
a MicroBeta Trilux liquid scintillation counter (Wallac,
Finland).
References
[1] R.I. Fryer, Ligand interaction at the benzodiazepine receptor,
in: Comprehensive Medicinal Chemistry, vol. 3, Pergamon
Press, New York, 1990, pp. 539–566.
[2] G. Stefancich, M. Artico, S. Massa, F. Corelli, Research on
nitrogen heterocyclic compounds; XIII. Synthesis of 5H-imi-
5.2.2. Adenosine binding assay
Affinity for the adenosine A1 receptor was deter-
mined in membranes prepared from mouse cerebral
cortex as previously described [13]. In brief, cerebral
cortex from adult, male CBA mice were dissected on
ice, homogenized in 10 volumes of ice-cold Tris–HCl
50 mM, pH 7.4 with an Ulta-Turrax and centrifuged at
40 000×g for 20 min at 4 °C. The pellet was washed
with 10 volumes of Tris–HCl and again centrifuged at
40 000×g for 20 min at 4 °C. The final pellet was
resuspended in 10 volumes of Tris–HCl and 1 ml
aliquots were stored at −80 °C. On the day of the
assay, membranes were thawed and incubated for 60
min at 25 °C with 2U/ml adenosine deaminase, diluted
and used in the binding assay.
dazo[2,1-c][1,4]benzodiazepine,
Synthesis 4 (1981) 321–322.
a novel tricyclic ring system,
[3] G. Stefancich, M. Artico, F. Correlli, R. Silvestri, G. de Feo,
G. Mazzanti, L. Durando, M. Palmery, New psychotropic
agents. I. Synthesis and pharmacologic activity of derivatives of
5H-imidazo-[2,1-c][1,4]benzodiazepine, Farmaco 40 (1985) 429–
441.
[4] J.W. Phillis, M.H. O’Regan, Benzodiazepine interaction with
adenosine systems explains some anomalies in GABA hypothe-
sis, Trends Pharmacol. Sci. 9 (1988) 153–154.
[5] D.C. Owsley, J.J. Bloomfield, The reduction of nitroarenes with
iron/acetic acid, Synthesis (1977) 118–120.
[6] M. Somei, K. Kato, S. Inoue, Titanium (III) chloride for the
reduction of heteroaromatic and aromatic nitro compounds,
Chem. Pharm. Bull. 28 (1980) 2515–2518.
For binding assay, 1 nM [3H]CCPA (42.8 Ci/mmol,
New England Nuclear, Boston, MA) was incubated in
96-well MultiScreen-FB microtiter plates (Millipore) in
300 ml (total volume) of Tris–HCl pH 7.4. Nonspecific
binding was determined in the presence of 1 mM
theophylline and was 5% of total binding. Test com-
pounds were dissolved in dimethyl sulfoxide and diluted
with Tris–HCl buffer immediately before use. Final
vehicle concentration in incubation wells was main-
tained constant (1%) in each well. Assays were per-
formed in triplicate. The incubation was started by the
addition of 50 mg protein/well, carried out for 180 min
at 25 °C and terminated by vacuum filtration as de-
scribed above.
Protein content was determined by the method of
Lowry [14], using bovine serum albumine as a standard.
Potencies for displacement (IC50 values) were deter-
mined with the aid of the computerized program
PHARM/PCS using at least six different concentrations
for each compound [15].
[7] D. Catarzi, L. Cecchi, V. Colotta, F. Melani, G. Filacchioni, C.
Martini, L. Giusti, A. Lucacchini, Structure–activity relation-
ships of 1,2,4-triazolo[1,5-a] quinoxalines and their 1-deaza
analogues imidazo[1,2-a]quinoxalines at the benzodiazepine re-
ceptor, J. Med. Chem. 37 (1994) 2846–2850.
[8] H. Mohler, J.C. Richards, Agonist and antagonist benzodi-
azepine receptor interaction in vitro, Nature 294 (1981) 763–
765.
[9] J.K. Buolamwini, Nucleoside transport inhibitors: structure ac-
tivity relationships and potential therapeutic applications, Curr.
Med. Chem. 4 (1997) 33–66.
[10] T.W. Stone, Actions of benzodiazepines and the benzodiazepine
antagonist flumazenil may involve adenosine, J. Neurol. Sci.
163 (1999) 199–201.
[11] E. Narimatsu, M. Aoki, Involvement of the adenosine neuro-
modulatory system in the benzodiazepine-induced depression of
excitatory synaptic transmissions in rat hippocampal neurons in
vitro, Neurosci. Res. 33 (1999) 57–64.
[12] M. Lipartiti, R. Arban, E. Fadda, A. Zanotti, P. Giusti, Char-
acterization of [3H]-imidazenil binding to rat brain membranes,
Br. J. Pharmacol. 114 (1995) 1159–1164.
[13] C. Florio, A.M. Rosati, U. Traversa, R. Vertua, Strain-related
differences in adenosine receptor density and in behavioral sen-
sitivity to adenosine, Pharmacol. Biochem. Behav. 49 (1994)
271–276.
[14] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randal, Protein
measurement with the Folin phenol reagent, J. Biol. Chem. 93
(1951) 265–275.
The binding affinity constants of the compounds (Ki)
were calculated according to the following equation
[16]: Ki=IC50/(1+[L]/Kd, where [L] is the concentra-
tion of the radioligand utilized in binding experiments.
[15] R.J. Tallarida, R.B. Murray, Manual of Pharmacologic Calcu-
lation with Computer Programs, Springer (Eds.), New York,
1987.
Acknowledgements
[16] J.P. Bennet, H.I. Yamamura, Neurotransmitter, hormone, or
drug receptor binding models, in: H.I. Yamamura (Ed.), Neu-
rotransmitter Receptor Binding, Raven Press, New York, 1985,
pp. 61–89.
Ministero dell’Universita` e della Ricerca Scientifica e
Tecnologica (MURST) is gratefully acknowledged for
its financial support.