Molecular basis of peripheral vs central benzodiazepine receptor selectivity in a new class of peripheral benzodiazepine receptor ligands related to alpidem

Article abstract of DOI:10.1021/jm960325j

Alpidem (1), the anxiolytic imidazopyridine, has nanomolar binding affinity for both the central benzodiazepine receptor (cbr) and the peripheral benzodiazepine receptor (pbr). A novel class of pbr ligands related to alpidem has been designed by comparing

Full text of DOI:10.1021/jm960325j

J . Med. Chem. 1996, 39, 4275-4284
4275
Molecu la r Ba sis of P er ip h er a l vs Cen tr a l Ben zod ia zep in e Recep tor Selectivity
in a New Cla ss of P er ip h er a l Ben zod ia zep in e Recep tor Liga n d s Rela ted to
Alp id em
Maurizio Anzini,‡ Andrea Cappelli,*,‡ Salvatore Vomero,‡ Gianluca Giorgi,† Thierry Langer,§ Giancarlo Bruni,
Maria R. Romeo, and Anthony S. Basile|
Dipartimento Farmaco Chimico Tecnologico, Universita` di Siena, Via Banchi di Sotto 55, 53100 Siena, Italy, Centro
Interdipartimentale di Analisi e Determinazioni Strutturali, Universita` di Siena, via P.A. Mattioli 10, 53100 Siena, Italy,
Institute of Pharmaceutical Chemistry, University of Innsbruck, Innrain 52A, A-6020 Innsbruck, Austria, Istituto di
Farmacologia, Universita` di Siena, Le Scotte 6, 53100 Siena, Italy, and Laboratory of Neuroscience, NIDDKD, National
Institutes of Health, Bethesda, Maryland 20892
Received April 30, 1996X
Alpidem (1), the anxiolytic imidazopyridine, has nanomolar binding affinity for both the central
benzodiazepine receptor (CBR) and the peripheral benzodiazepine receptor (PBR). A novel
class of PBR ligands related to alpidem has been designed by comparing the interaction models
of alpidem with PBR and CBR. Several compounds in this class have shown high selectivity
for PBR vs CBR, and the selectivity has been discussed in terms of interaction models. The
binding behavior of the three selected compounds was extensively studied by competition and
saturation assays, and the results suggest that they are capable of recognizing two sites labeled
by [3H]PK11195. The molecular structure of one of the most active compounds (4e) has been
determined by X-ray diffraction and compared with that of alpidem. Molecular modeling studies
suggest that the bioactive conformation of 4e is likely to be very similar to the conformation
found in the crystal.
In tr od u ction
Alpidem (1) binds with nanomolar affinity both PBR
(Ki ) 0.5-7 nM) and CBR (Ki ) 1-28 nM)6 and
stimulates pregnenolone formation from mitochondria
of C6-2B glioma cells.6b The anxiolytic and anticonvul-
sant properties of 1 appear to be elicited both with a
direct mechanism through the interaction with CBR and
with an indirect mechanism by the interaction with PBR
and neurosteroid production.5,6b,c Compound 2, a close
derivative of alpidem, also stimulates pregnenolone
formation, but it appears to be more selective for the
PBR than alpidem, as it shows nanomolar affinity for
the PBR (Ki ) 3.9 nM) and no significant affinity for
the CBR (IC50 > 1000 nM).5 Related compounds, such
as zopiclone (3)7 and 3a ,8 show significant affinity for
the CBR (IC50 ) 29 and 380 nM, respectively) with little
or no affinity for the PBR. We have been interested for
Since their discovery, benzodiazepines have attracted
the attention of researchers owing to their potential and
actual clinical uses and because of their low toxicity.
The efforts made in this field have led to the develop-
ment of many clinical important anxiolytics, anticon-
vulsants, hypnotics, and muscle relaxants and to the
discovery of the benzodiazepine receptor itself.
Soon after its description, it was found that a number
of structurally different classes of substances could bind
to the benzodiazepine receptor.1 Moreover, benzodiaz-
epine receptors have been found both in the central
nervous system (CNS) and in peripheral tissues2 and
have subsequently been divided into two classes: the
central and peripheral benzodiazepine receptors. The
central benzodiazepine receptor (CBR) is localized only
on the cell membrane of neurons, where it is part of the
γ-aminobutyric acid (GABAA) receptor complex, and is
involved in the regulation of GABA-gated chloride ion
currents. In contrast, the peripheral benzodiazepine
receptor (PBR) is located in both CNS and peripheral
tissues. Although PBR are found primarily on the
mitochondrial outer membrane, a nonmitochondrial
localization in some cells has also been suggested.3
While the physiological function of the PBR is not
fully understood (for recent reviews, see refs 3 and 4),
it is clearly not involved in the regulation of chloride
currents. The PBR do appear to be involved in ste-
roidogenesis, the regulation of which is a potential
clinical application of PBR selective compounds.5
‡ Dipartimento Farmaco Chimico Tecnologico.
† Centro Interdipartimentale di Analisi e Determinazioni Strut-
turali.
§ University of Innsbruck.
Istituto di Farmacologia.
| NIDDKD, National Institutes of Health.
X Abstract published in Advance ACS Abstracts, August 15, 1996.
S0022-2623(96)00325-1 CCC: $12.00 © 1996 American Chemical Society

4276 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 21
Anzini et al.
The lack of CBR affinity of compound 2 confirms the
importance of the imidazole nitrogen atom of alpidem
for the interaction with CBR and suggests at the same
time a limited role of the acetamide side chains in CBR
interaction, while the carbonyl dipole and amide lipo-
philic side chains are important for interaction with the
PBR.5 Thus, the acetamide moiety became the target
for further structural modifications aimed at enhancing
PBR selectivity.
Compounds 4-7 and 13 were designed as structural
hybrids of alpidem and zopiclone, maintaining the
minimum requisites for the interaction with both the
PBR and CBR while varying the geometry and lipophi-
licity of the side chains. In addition, the comparative
ability of the PBR and CBR to accommodate large
heteroaromatic portions was explored by the bioisosteric
replacement of 5-chlorine substituent of alpidem with
a condensed benzo ring.
F igu r e 1. Comparison between the schematic representation
of the interaction models of alpidem with CBR and PBR.
several years in chemically modifying nonselective
ligands to produce more selective compounds and have
successfully applied our efforts to different classes of
CNS agents.9 In this paper we describe the results
obtained when alpidem was chosen as the lead com-
pound. The design, synthesis, biological evaluation, and
molecular modeling studies of a new class of potent and
selective PBR ligands are reported. These compounds
should be of particular interest in the light of the fact
that PBR would appear to offer an indirect pharmaco-
logical approach to modulation of GABAA receptor
complex.6d
Dr u g Design
The rational design of selective ligands from a non-
selective mother compound implies the knowledge of key
pharmacophoric elements in the interaction with the
different receptors. These elements can be deduced
from the available structure-activity relationship (SAR)
studies. In the present study, a survey of the literature
has revealed the existence of at least three models for
the interactions of benzodiazepines with the CBR
(Fryer,1 Wermuth,7 and Borea10) and two interaction
models for PBR.6,11 The three CBR interaction models
analyzed are very similar and readily integrated into
one comprehensive model. Figure 1 shows an inte-
grated model for the interaction of alpidem with the
CBR, which takes into account the agonist properties
of alpidem. Furthermore, this model was obtained
following similar topological and electronic guiding
criteria previously used (see refs 1, 7, and 10) in the
model development. The different zones were termed
using the nomenclature proposed by Wermuth: FRA,
freely rotating aromatic ring region, corresponding to
A in Fryer’s model and to AG1 in Borea’s model; δ1,
electron-rich zone, corresponding to π1 in Fryer’s model
and to B1 in Borea’s model; OPR, out-plane region,
corresponding to π3 in Fryer’s model and to AG2 in
Borea’s model; and PAR, planar aromatic region.
Bourguignon has conceived an interaction model for
the PBR very similar to that for CBR6 with some
important differences in the spatial location of the main
pharmacophoric points. Comparison of the interaction
models for the PBR and CBR (Figure 1) predicts that
prevention of the imidazole nitrogen from behaving as
the H-bond acceptor in alpidem should decrease the
compound’s affinity for CBR. This concept served as
the basis for the design of the PBR selective compound
2,5 where the geometry of alpidem was conserved, but
the H-bond acceptor imidazole nitrogen was converted
into a H-bond donor indole nitrogen.
Ch em istr y
The synthesis of the title compounds was performed
from pyrrolo[3,4-b]quinoline 11 or pyrrolo[3,4-b]pyridine
18 through the attack of their active methylene groups
with different electrophilic alkylating agents as the key
steps.
In Scheme 1 the synthesis of ylideneacetic derivatives
4a -f starting from aniline is reported. Aniline (8) was
condensed with ethyl 4-chloroacetoacetate and ethyl
anilinobutenoate intermediate 9 was subjected to Vils-
meier formilation conditions12 to give quinoline 10.
Compound 10 was then reacted with p-chloroaniline in
ethanol to reflux to give key intermediate 11 in 92%
yield, which was used without further purification.
Condensation between 11 and glyoxylic acid mono-
hydrate in glacial acetic acid and acetic anhydride (1:
1) afforded unsaturated acid 12 in 59% yields. The
E-geometry of the exocyclic double bond of 12 was
assigned in the first instance on the basis of low-field
resonance of the carboxylic proton (16.13 ppm) observed
1
in H-NMR spectrum of 12 in deuterated chloroform. It
was assumed that an intramolecular hydrogen bond
between the carboxylic hydrogen and quinoline nitrogen
atom stabilizes the E-geometry of acid 12.
The unsaturated acid 12 was converted into the
corresponding amides 4a -f via mixed anhydrides.
When the conversion of unsaturated acid 12 into the
corresponding N,N-dipropyl amide was carried out via
acid chloride, a partial inversion of configuration at the
exocyclic double bond was observed. The major product
(yield 55%) of this reaction showed Z-geometry (13),

New Class of PBR Ligands Related to Alpidem
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 21 4277
Sch em e 1a
Sch em e 2a
a
Reagents: (i) NaH, MeOCOOMe, DMF; (ii) DBU, DMF; (iii)
BrCH2COOEt/DMF; (iv) NaOH, H2O, EtOH.
ester groups and spontaneous loss of the carboxyl group
attached to the pyrrolidone ring of compound 15 gave
the acetic acid derivative 16 (in this behavior, compound
15 is reminiscent of a â-keto ester) which was converted
into the corresponding amides 5a ,b via mixed anhy-
drides.
The chemistry developed for the synthesis of com-
pounds 4 and 5 was the same used to obtain compounds
6 and 7 as depicted in Scheme 3. Briefly, commercially
available nicotinate 17 was transformed in two steps
into the required intermediate 18 which was used in
glyoxylic condensation to obtain unsaturated acid 19 or
alternatively in a multistep sequence similar to that
used for the synthesis of compound 16 to give saturated
acid 21. Both acids 19 and 21 were transformed into
the corresponding amides 6a ,b and 7a ,b via mixed
anhydrides.
a
Reagents: (i) ClCH2COCH2COOEt, AcOH, C6H6; (ii) DMF,
POCl3, CHCl3; (iii) p-ClC6H4NH2, EtOH; (iv) CHOCOOH‚H2O,
Ac2O, AcOH; (v) i-BuOCOCl, Et3N, RNHR′ (method A or B); (vi)
SOCl2, n-Pr2NH, MeC6H5; (vii) H2, Pd/C, EtAc.
while the E-isomer (4a ) was isolated only in 14% yield.
E-Z-Geometry of the amides 4 and 13 was assigned
on the basis of their H-NMR spectra in comparison with
that of starting acid 12 for which E-geometry was
assumed. Finally, single-crystal X-ray diffraction stud-
ies performed on compound 4e confirmed unequivocally
the E-geometry for this compound and the basic as-
sumption mentioned above.
1
Resu lts a n d Discu ssion
Catalytic reduction of compound 4a gave only trace
amount of expected 5a ; thus, a different approach to the
synthesis of saturated amides 5a ,b was investigated
(Scheme 2). Sodium hydride has been reported to be
effective in deprotonating a pyrrolo[3,4-b]quinoline sys-
tem very similar to 11 to give a resonance-delocalized
anion which could be alkylated with butyl iodide.13
Unfortunately, the attempts at the direct alkylation
of the anion generated from 11 using NaH in DMF with
ethyl bromoacetate were unsuccessful; however, the
anion could be acylated using dimethyl carbonate to give
ester 14 in 92% yield. Compound 14 was deprotonated
in anhydrous DMF by DBU to give a dark blue-colored
anion which was alkylated with ethyl bromoacetate to
Bin d in g Stu d ies. Initially, the affinities of com-
pounds 4-7, 13, and 18 for the PBR and CBR in rat
cerebral cortex were tested by competition assays
against [3H]PK11195 and [3H]flunitrazepam, respec-
tively (see In Vitro Binding Assays. Method A). This
study revealed that compounds 4a ,b,e,f and 6a bound
with high affinity and selectivity to the PBR (Table 2).
In addition, the slopes of the competition curves for
compounds 4a ,e,f were significantly less than 1, sug-
gesting an interaction with multiple PBR-type binding
sites. Further analysis of these compounds using
competition assays against renal and cortical PBR by
means of the selective ligands [3H]Ro5-4864 and [3H]-
PK11195 (see In Vitro Binding Assays. Method B)
revealed that the competition curves for all three
1
diester 15, characterized by its H-NMR spectrum and
used without further purification. Hydrolysis of the

4278 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 21
Anzini et al.
Sch em e 3a
CBR affinity. This suggests that this structural feature
contributes to some extent to the binding process with
CBR, but unfortunately the starting CBR affinity of 7a
is not high enough to appreciate the actual importance
of this side chain.
The corresponding unsaturated derivative 6a did not
show a CBR affinity, showing at the same time a
significant affinity for PBR. The selectivity showed by
6a seems to be related to the geometry of the side chain.
In fact, it is possible to assume for the side chain of 6a
an arrangement near to the mean plane of the molecule
as that showed by the X-ray diffraction studies of 4e
(Figure 4), while in compound 7a the acetamide moiety
appeared to be projected out of the mean plane of the
molecule by sp3 hybridization of the pyrrolinone ring
methylene.
These findings suggested the existence of a further
difference between the interaction model of alpidem
with PBR and CBR: The alpidem side chain probably
interacts (by its carbonyl dipole) with PBR in a position
near to the mean plane, while the interaction with CBR
might take place out of the mean plane of the molecule.
This would imply that alpidem interacts with PBR and
CBR in two different conformations. This hypothesis
appears to be very likely owing to the fairly great
conformational freedom of alpidem. In this respect
compounds 4 and 6 represent conformationally con-
strained analogues of alpidem lacking a degree of
conformational freedom. The high affinity showed by
4a ,e,f suggests that in these compounds the side chain
shows a geometry favorable to the interaction with PBR.
When the E-geometry at the exocyclic double bond of
compound 4a was inverted to give 13 (Z-isomer), all
significant PBR affinity was lost confirming the impor-
tance of the correct geometry of the acetamide moiety
in the interaction with PBR.
a
Reagents: (i) NBS, dibenzoyl peroxide, CCl4; (ii) p-ClC6H4NH2,
EtOH; (iii) CHOCOOH‚H2O, Ac2O, AcOH; (iv) NaH, MeOCOOMe,
DMF; (v) DBU, DMF; (vi) BrCH2COOEt, DMF; (vii) NaOH, H2O,
EtOH.
compounds against [3H]PK11195 were biphasic, regard-
less of the tissue source of PBR (Figure 2), while
competition curves against [3H]Ro5-4864 were monopha-
sic. Saturation assays of [3H]PK11195 binding to PBR
in renal homogenates were performed in the presence
and absence of 4f (Figure 3). These assays confirmed
the existence of two binding sites for [3H]PK11195, in
agreement with the competition studies.
While the above studies suggest that [3H]PK11195
binds with equal affinity to two PBR subtypes in the
cortex and kidney that can be discriminated by com-
pounds 4a ,e,f, it is possible that the second site is a
nonspecific binding site because of the high level of
nonspecific binding associated with [3H]PK11195. Thus,
an attempt was made to establish if the second subtype
was in fact a nonproteinaceous, non-PBR binding site
by measuring specific [3H]PK11195 binding to renal
homogenates boiled for 15 min (Figure 2). Under these
conditions, all specific binding was eliminated, suggest-
ing that the second site does not reflect nonspecific
binding of [3H]PK11195 to membrane lipids.
Further structure-affinity relationships concern the
positive influence (7.2×) of benzocondensation at the
b-edge of the pyrrolo[3,4-b]pyridine nucleus of com-
pounds 6 on PBR affinity. Most probably in this case
the benzene ring plays a role similar to that of the
5-chloro substituent in alpidem and compound 2.
Kozikowski5 and co-workers reported that 5-chloro
substitution in compound 2 improved the PBR affinity
by about 1 order of magnitude (17×), which indeed
appears to be an effect very similar to that induced by
benzofusion in the compounds which are the topic of this
paper.
Conversely, benzofusion at the b-edge of the pyrrolo-
[3,4-b]pyridine nucleus of compound 7a appears not to
be well tolerated by CBR, since compound 5a showed
no significant affinity for these receptors. As far as the
lipophilic amide substituents are concerned, the findings
obtained in this study suggest a limited tolerance to the
steric hindrance; in fact, relatively small lipophilic
groups such as n-propyl, p-chlorophenyl, and p-meth-
oxyphenyl are well accommodated by the receptor, while
the bulkier n-hexyl groups are not so well tolerated, and
finally (quite surprisingly) benzyl groups are not toler-
ated at all. This minor disagreement with the findings
reported by Kozikowski5 could be related to the greater
rigidity of compounds 4 when compared with 2.
Str u ctu r e-Affin ity Rela tion sh ip s. From the re-
sults summarized in Table 2, it appears that only
compound 7a displays a significant affinity for CBR.
This appears to be about 1 order of magnitude lower
than that reported for zopiclone, but it is very similar
to that reported for compound 3a .
Molecu la r Mod elin g. The aim of this part of the
work is to assess whether the conclusions offered in the
previous section on the basis of the classical medicinal
Interestingly, deletion of acetamide moiety in com-
pound 7a to give 18 leads to a significant decrease in

New Class of PBR Ligands Related to Alpidem
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 21 4279
Ta ble 1. Preparative and Analytical Data of Compounds 4-7
compd
Heta
R
R′
method
yield, %
mp, °C (recryst solv)b
formula
anal.c
4a
4b
4c
4d
4e
4f
5a
5b
6a
6b
7a
7b
Q
Q
Q
Q
Q
Q
Q
Q
P
n-Pr
n-He
Me
Et
Me
n-Pr
n-He
A
A
B
B
B
B
A
A
B
B
A
A
67
62
68
69
60
56
71
66
57
75
69
56
215-216 (EA)
C25H24N3O2Cl
C31H36N3O2Cl
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
CHN
178-179 (H-EA)
195-196 (H-EA)
186-187 (H-EA)
159-160 (H-EA)
149-151 (H-EA)
187-188 (H-EA)
152-153 (H-EA)
116-117 (H-C)
111-112 (H)
CH2-C6H5
CH2-C6H5
p-MeO-C6H4
p-Cl-C6H4
n-Pr
n-He
n-Pr
n-He
n-Pr
C27H20N3O2Cl
C28H22N3O2Cl‚1/6H2O
C27H20N3O3Cl
Me
C26H17N3O2Cl2‚1/3H2O
n-Pr
n-He
n-Pr
n-He
n-Pr
n-He
C
C
C
C
C
C
25H26N3O2Cl
31H38N3O2Cl
21H22N3O2Cl
27H34N3O2Cl
21H24N3O2Cl
27H36N3O2Cl
P
P
P
120-121 (H-EA)
81-82 (H)
n-He
a
b
Q, quinoline condensed at b-edge; P, pyridine condensed at b-edge. Recrystallization solvents: EA, ethyl acetate; H, n-hexane; C,
cyclohexane. c Analyses for the elements indicated were within (0.4% of theoretical values.
Ta ble 2. Affinities of Compounds 4-7, 13, and 18 for PBR and
CBR (IC50 (nM) ( SEM)a
compd
PBR
CBR
4a
4b
4c
4d
4e
4f
5a
5b
6a
6b
7a
7b
13
18
40 ( 13b
>1000 (0%)
>1000 (0%)
NTc
NT
NT
150 ( 38
>1000 (6%)
>1000 (23%)
1.9 ( 0.2b
32 ( 4.7b
>1000 (0%)
>1000 (0%)
290 ( 72
33000 ( 2000
>1000
1600 ( 490
>1000 (16%)
>1000 (0%)
1.8 ( 0.54
NT
>1000 (0%)
>1000 (0%)
>1000 (0%)
>1000 (0%)
860 ( 230
>1000 (0%)
NT
>1000 (0%)
PK11195
diazepam
5.6 ( 1.3
a
The IC50 values are the means of three experiments performed
in duplicate; values in parentheses are the inhibitions at the
b
maximal concentration tested (1000 nM). Compounds 4a ,e,f
displaced [3H]PK11195 in
a biphasic manner; linear fitting
analysis (single-site model) gave the IC50 values reported in the
table. c NT, not tested.
chemistry approach can also be substantiated by mo-
lecular mechanics calculations. For this purpose we
started from the crystal structures as determined by
single-crystal X-ray diffraction studies of alpidem14 and
compound 4e (see X-ray Crystallography section). By
taking into account the existing literature on the
description of the pharmacophoric elements of the PBR
ligands, we chose to adopt the same distance parameters
already used by other authors: d, distance between the
electronegative carboxamide oxygen atom and the cen-
troid of the freely rotating aromatic ring, and h, eleva-
tion of the carboxamide oxygen atom above the mean
plane of the molecule.
As it is possible to observe by comparing Figures 4
and 5, the crystal structure of compound 4e shows a
side chain geometry different from that of alpidem (also
in its crystal structure). In fact the values of the
distance parameters d and h in compound 4e are 6.3
and 1.6 Å, respectively, while the same parameter
values of alpidem are 4.6 and 2.3 Å, respectively. The
F igu r e 2. Competition of 4f with [3H]PK11195 (b) or [3H]-
Ro5-4864 (O) for PBR in the brain (panel A) or kidney (panel
B). Both radioligand concentrations were approximately 1 nM.
Each point represents the summary (mean ( SEM) of data
from 8-12 separate observations. Competition assays between
4f and [3H]Ro5-4864 yielded monophasic curves with IC50
values of 2.7 and 2.0 nM (brain and kidney, respectively). In
contrast, competition assays between 4f and [3H]PK11195
revealed the presence of two binding sites, with IC50 values in
the brain of 4.1 and 400 nM and of 0.8 and 180 nM in the
kidney. Competition assays performed between 4f and [3H]-
PK11195 using kidney homogenates boiled for 15 min (4)
showed no significant radioligand binding.
values of the distance parameters d and h of the crystal
structure of compound 4e appear to be considerably
different from the respective values found in the crystal
structure of alpidem, but they correlate well with the

4280 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 21
Anzini et al.
Con clu sion s
Starting from the high affinity of alpidem for both
PBR and CBR, a novel class of potent and selective PBR
ligands was designed by comparing the interaction
models of alpidem with PBR and CBR. Three members
of this class show high affinity and selectivity for PBR
and appear to be capable of recognizing two sites labeled
by [3H]PK11195. Clearly, the development of these
subtype selective ligands based on the alpidem structure
will afford investigators new opportunities to establish
the range of PBR functions and localization. Finally,
the structure-affinity relationships described herein
along with the molecular modeling efforts allow the
refinement of the original interaction model of alpidem
with PBR which can be used to design novel ligands
with improved features.
Exp er im en ta l Section
Melting points were determined in open capillaries on a
Bu¨chi 510 apparatus and are uncorrected. Microanalyses were
carried out using a Perkin-Elmer 240C elemental analyzer.
Merck silica gel 60, 70-230 mesh, was used for column
chromatography, and Riedel-de Haen DC-Mikrokarten SI F
37341 were used for TLC. 1H-NMR spectra were recorded with
a Bruker AC 200 spectrometer in the solvents indicated (TMS
as internal standard); the values of the chemical shifts are
expressed in ppm and coupling constants (J ) in hertz (Hz).
Mass spectra (EI, 70 eV) were recorded on a VG 70-250S
spectrometer. IR, NMR spectra, and elemental analyses were
performed by the Dipartimento Farmaco Chimico Tecnologico,
Universita` di Siena. Mass spectra were performed by Centro
di Analisi e Determinazioni Strutturali, Universita` di Siena.
Gen er a l P r oced u r es for th e Syn th esis of Com p ou n d s
4-7. Meth od A. A mixture of the appropriate carboxylic acid
(12, 16, 21) (0.4 mmol) in anhydrous THF (20 mL) with
triethylamine (0.19 mL, 1.4 mmol) was cooled at -10 °C, and
isobutyl chloroformate (0.057 mL, 0.44 mmol) was added with
stirring under an argon atmosphere. After the mixture stirred
for 30 min at -10 °C, the appropriate amine (0.44 mmol) was
added and the reaction mixture was allowed to reach room
temperature and stirred for 1 h. Then, the reaction mixture
was diluted with dichloromethane (20 mL), washed with brine,
dried over sodium sulfate, and concentrated under reduced
pressure. Recrystallization of the residue from the suitable
solvent gave analytically pure amides 4a ,b, 5a ,b, and 7a ,b.
(E)-2-(4-Ch lor op h en yl)-2,3-d ih yd r o-3-[[(N,N-d i-n -p r o-
p yla m in o)ca r bon yl]m eth ylen e]-1H-p yr r olo[3,4-b]qu in o-
lin -1-on e (4a ): 1H-NMR (CDCl3) 0.78 (t, J ) 7.4, 3H), 1.11 (t,
J ) 7.4, 3H), 1.42-1.58 (m, 2H), 1.86-2.03 (m, 2H), 3.32 (t, J
) 7.4, 2H), 3.53 (t, J ) 7.7, 2H), 5.75 (s, 1H), 7.36-7.42 (m,
2H), 7.51-7.57 (m, 2H), 7.65 (t, J ) 7.9, 1H), 7.81-7.89 (m,
1H), 8.01 (d, J ) 8.0, 1H), 8.17 (d, J ) 8.5, 1H), 8.70 (s, 1H).
(E)-2-(4-Ch lor op h en yl)-2,3-d ih yd r o-3-[[(N,N-d i-n -h exy-
la m in o)ca r bon yl]m eth ylen e]-1H-p yr r olo[3,4-b]qu in olin -
1-on e (4b): 1H-NMR (CDCl3) 0.77 (t, J ) 6.7, 3H), 0.95 (t, J )
6.8, 3H), 1.12-1.61 (m, 14H), 1.87-2.03 (m, 2H), 3.33 (t, J )
7.4, 2H), 3.54 (t, J ) 7.8, 2H), 5.75 (s, 1H), 7.36-7.41 (m, 2H),
7.52-7.57 (m, 2H), 7.65 (t, J ) 7.4, 1H), 7.85 (t, J ) 8.2, 1H),
8.02 (d, J ) 7.6, 1H), 8.18 (d, J ) 8.4, 1H), 8.70 (s, 1H); MS
m/ z 517 (M+, 29).
F igu r e 3. Saturation assays of [3H]PK11195 binding to PBR
in kidney homogenates in the presence (O) and absence (b) of
100 nM 4f. Each point represents the summary (mean ( SEM)
of data from 4 (control) and 12 (4f) separate observations.
Panel A shows the results of the saturation assay presented
as a binding isotherm; panel B presents the same data as a
Scatchard plot. Lines for the two binding sites observed in the
presence of 100 nM 4f (panel B) were drawn using the values
for Kd and Bmax determined by nonlinear regression. [3H]-
PK11195 binds to a single apparent site (Kd ) 1.6 ( 0.1 nM,
Bmax ) 4.2 ( 0.2 pmol/mg of protein) on renal homogenates in
the absence of 4f. Addition of 100 nM 4f to the saturation
assays results in a shift in radioligand affinity consistent with
a competitive inhibition, as well as revealing two binding sites
(Kd ) 2.2 ( 0.8 nM, Bmax ) 1.4 ( 0.2 pmol/mg of protein; Kd
1
1
2
) 29 ( 4.5 nM, Bmax ) 5.1 ( 0.5 pmol/mg of protein).
2
same distances as determined in the assumed bioactive
conformation of PK11195, the typical PBR ligand (d )
6.8 Å, h ) 1.1 Å).11 Furthermore, Lentini and co-
workers suggested that the comparison of the most
favorable conformational families for some different
PBR ligands allowed for a common overlapping zone
characterized by d and h values of 6.1 ( 0.2 and 1.4 (
0.6 Å, respectively.15 This, also in view of the low
flexibility of 4e, strongly suggests that the crystal
conformation of 4e is probably very similar to the
bioactive conformation at PBR sites.
Thus, taking into account the high conformational
flexibility of alpidem,11,14 it can exist in a low-energy
conformer close to the crystal conformation of 4e as
showed in Figure 6. This figure also explains the
apparent bioisosterism of the replacement of the 5-chlo-
rine atom of compound 2 with a condensed benzo ring
(vide infra in Results and Discussion section). On the
other hand, it is interesting to note that the crystal
geometry of alpidem appears to be comparable with that
of a low-energy conformer of 7a which is active at the
CBR sites only (Figure 7).
(()-2-(4-Ch lor op h en yl)-2,3-d ih yd r o-3-[[(N,N-d i-n -p r o-
p yla m in o)ca r bon yl]m et h yl]-1H-p yr r olo[3,4-b]q u in olin -
1-on e (5a ): 1H-NMR (CDCl3) 0.72 (t, J ) 7.4, 3H), 0.81 (t, J )
7.4, 3H), 1.25-1.51 (m, 4H), 2.90-3.22 (m, 6H), 5.91 (t, J )
4.9, 1H), 7.40-7.45 (m, 2H), 7.60-7.68 (m, 3H), 7.79-7.87 (m,
1H), 8.02 (d, J ) 7.4, 1H), 8.16 (d, J ) 8.6, 1H), 8.70 (s, 1H).
(()-2-(4-Ch lor op h en yl)-2,3-d ih yd r o-3-[[(N,N-d i-n -h exy-
la m in o)ca r b on yl]m et h yl]-1H -p yr r olo[3,4-b]q u in olin -1-
on e (5b): 1H-NMR (CDCl3) 0.77-0.90 (m, 6H), 1.11-1.33 (m,
16H), 2.89-3.03 (m, 4H), 3.17 (t, J ) 7.4, 2H), 5.90 (t, J ) 4.8,
1H), 7.39-7.44 (m, 2H), 7.58-7.68 (m, 3H), 7.79-7.87 (m, 1H),
8.02 (d, J ) 8.9, 1H), 8.16 (d, J ) 8.5, 1H), 8.70 (s, 1H).

New Class of PBR Ligands Related to Alpidem
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 21 4281
Ta ble 3. Characterization of Binding Behavior of Compounds 4a ,e,f at PBR by Competition Assays
compd
radioligand
tissue
IC50 , nM
density, receptor1 (% of total)
IC50 , nM
density, receptor2 (% of total)
1
2
4a
[3H]Ro5-4864
[3H]PK11195
[3H]Ro5-4864
[3H]PK11195
[3H]Ro5-4864
[3H]PK11195
[3H]Ro5-4864
[3H]PK11195
[3H]Ro5-4864
[3H]PK11195
[3H]Ro5-4864
[3H]PK11195
cerebral cortex
120 ( 4.7
4.9 ( 1.9
150 ( 18
2.7 ( 0.4
2.3 ( 0.3
9.8 ( 0.3
1.7 ( 0.2
0.7 ( 0.3
4.5 ( 0.2
2.4 ( 0.3
2.1 ( 0.1
1.1 ( 0.3
82
82
76
46
80
71
210 ( 26
160 ( 27
320 ( 18
14 ( 2.3
610 ( 160
140 ( 19
18
18
24
54
20
29
kidney
4e
4f
cerebral cortex
kidney
cerebral cortex
kidney
(()-6-(4-Ch lor op h en yl)-6,7-d ih yd r o-7-[[(N,N-d i-n -h exy-
la m in o)ca r b on yl]m e t h yl]-5H -p yr r olo[3,4-b]p yr id in -5-
on e (7b): 1H-NMR (CDCl3) 0.80-0.92 (m, 6H), 1.15-1.40 (m,
16H), 2.74-3.33 (m, 6H), 5.85 (t, J ) 5.0, 1H), 7.37-7.47 (m,
3H), 7.57-7.62 (m, 2H), 8.15-8.20 (m, 1H), 8.75-8.79 (m, 1H).
Meth od B. To a mixture of the appropriate carboxylic acid
(12 or 19) (0.4 mmol) in dichloromethane (10 mL) with
triethylamine (0.19 mL, 1.4 mmol) cooled at -10 °C under an
argon atmosphere was added isobutyl chloroformate (0.057
mL, 0.44 mmol). The reaction mixture was stirred at -10 °C
for 30 min, and then the appropriate amine (0.44 mmol) was
added. After the mixture stirred at room temperature for 1 h
(48 h for compound 4f), water (20 mL) was added. The organic
layer was washed with water (2 × 20 mL), dried over sodium
sulfate, and concentrated under reduced pressure to give
amides 4c-f and 6a ,b. The amides were recrystallized from
the suitable solvent or purified by column chromatography
eluting with chloroform-ethyl acetate (8:2) (for compounds
4e,f) and then recrystallized.
F igu r e 4. Crystal structure of compound 4e obtained from
X-ray diffraction experiments.
(E)-3-[[(N-Ben zylm et h yla m in o)ca r bon yl]m et h ylen e]-
2-(4-ch lor op h en yl)-2,3-d ih yd r o-1H-p yr r olo[3,4-b]qu in o-
lin -1-on e (4c):16 1H-NMR (CDCl3) 2.99 (s), 3.17 (s), 4.61 (s),
4.87 (s), 5.80 (s), 5.82 (s), 7.22-8.07 (m), 8.25 (d, J ) 8.6), 8.72
(s).
(E)-3-[[(N-Ben zylet h yla m in o)ca r b on yl]m et h ylen e]-2-
(4-ch lor op h en yl)-2,3-d ih yd r o-1H-p yr r olo[3,4-b]qu in olin -
1-on e (4d ):16 1H-NMR (CDCl3) 1.03 (t, J ) 7.3), 1.47 (t, J )
7.3), 3.39 (q, J ) 7.2), 3.67 (q, J ) 7.2), 4.62 (s), 4.88 (s), 5.71
(s), 5.84 (s), 7.16-7.73 (m), 7.81-7.93 (m), 8.00-8.12 (m), 8.25
(d, J ) 8.5), 8.72 (s).
F igu r e 5. Crystal structure of alpidem built from the crystal-
lographic coordinates obtained from ref 14.
(E)-2-(4-Ch lor op h en yl)-2,3-d ih yd r o-3-[[[N-(4-m et h ox-
yp h en yl)m eth yla m in o]ca r bon yl]m eth ylen e]-1H-p yr r olo-
[3,4-b]qu in olin -1-on e (4e):16 1H-NMR (CDCl3) 3.41 (s), 3.55
(s), 3.72 (s), 3.88 (s), 5.60 (s), 5.92 (s), 6.64-6.69 (m), 6.99-
7.04 (m), 7.39-7.47 (m), 7.55-7.72 (m), 7.88-8.04 (m), 8.22-
8.35 (m), 8.64 (s), 8.74 (s).
(E)-2-(4-Ch lor op h en yl)-3-[[[N-(4-ch lor op h en yl)m eth y-
la m in o]ca r bon yl]m eth ylen e]-2,3-d ih ydr o-1H-p yr r olo[3,4-
b]qu in olin -1-on e (4f):16 1H-NMR (CDCl3) 3.42 (s), 3.57 (s),
5.58 (s), 5.90 (s), 6.64-6.69 (m), 7.02-7.15 (m), 7.45-7.73 (m),
7.85-8.07 (m), 8.30 (d, J ) 8.6), 8.65 (s), 8.74 (s).
F igu r e 6. Superposition of a low-energy conformer of alpidem
(light) with the solid state conformation of 4e (dark).
(E)-6-(4-Ch lor op h en yl)-6,7-d ih yd r o-7-[[N,N-d i-n -p r op y-
la m in o)ca r bon yl]m eth ylen e]-5H-p yr r olo[3,4-b]p yr id in -
5-on e (6a ): 1H-NMR (CDCl3) 0.79 (t, J ) 7.4, 3H), 1.02 (t, J )
7.4, 3H), 1.39-1.56 (m, 2H), 1.66-1.82 (m, 2H), 3.26 (t, J )
7.5, 2H), 3.42-3.50 (m, 2H), 5.78 (s, 1H), 7.32-7.55 (m, 5H),
8.18 (d, J ) 8.3, 1H), 8.80-8.83 (m, 1H).
(E)-6-(4-Ch lor op h en yl)-6,7-d ih yd r o-7-[[N,N-d i-n -h exy-
la m in o)ca r bon yl]m eth ylen e]-5H-p yr r olo[3,4-b]p yr id in -
5-on e (6b): 1H-NMR (CDCl3) 0.81 (t, J ) 6.3, 3H), 0.91 (t, J )
6.5, 3H), 1.12-1.54 (m, 14H), 1.67-1.82 (m, 2H), 3.27 (t, J )
7.4, 2H), 3.49 (t, J ) 7.5, 2H), 5.77 (s, 1H), 7.31-7.54 (m, 5H),
8.16-8.20 (m, 1H), 8.80-8.83 (m, 1H).
Eth yl 4-Ch lor o-3-a n ilin ocr oton a te (9). In a 100 mL
flask, attached to a Dean-Stark constant water separator
which was connected to a reflux condenser, 4.66 g (50.0 mmol)
of aniline (8), 6.82 mL (50.0 mmol) of ethyl 4-chloroacetoac-
etate, 50 mL of benzene, and 5 mL of glacial acetic acid were
placed. The flask was heated to reflux for 2 h. The solvent
was then distilled under reduced pressure; toluene (10 mL)
F igu r e 7. Superposition of the crystal structure of alpidem
(light) with a low-energy conformer of 7a (dark).
(()-6-(4-Ch lor op h en yl)-6,7-d ih yd r o-7-[[(N,N-d i-n -p r o-
p yla m in o)ca r bon yl]m eth yl]-5H-p yr r olo[3,4-b]p yr id in -5-
on e (7a ): 1H-NMR (CDCl3) 0.70-0.86 (m, 6H), 1.26-1.47 (m,
4H), 2.76-3.30 (m, 6H), 5.85 (t, J ) 5.0, 1H), 7.36-7.47 (m,
3H), 7.57-7.62 (m, 2H), 8.15-8.20 (m, 1H), 8.75-8.79 (m, 1H).

4282 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 21
Anzini et al.
was added and distilled under reduced pressure (the operation
was repeated twice). The residue was diluted with chloroform-
light petroleum ether (1:1, 50 mL), and the solid was filtered
off. The filtrate was concentrated under reduced pressure.
Purification by culumn chromatography of the residue eluting
with chloroform-light petroleum ether (1:1) afforded 9 (4.8 g,
yield 40%) as an oil which was used in the next step without
further handling: 1H-NMR (CDCl3) 1.30 (t, J ) 6.9, 3H), 4.09
(s, 2H), 4.18 (q, J ) 6.9, 2H), 5.01 (s, 1H), 7.16-7.42 (m, 5H),
10.07 (br s, 1H).
and then 0.065 g (yield 55%) of pure 13 as a white solid.
Recrystallization of 13 from n-hexane-ethyl acetate gave an
analytical sample melting at 172-173 °C: 1H-NMR (CDCl3)
0.80 (t, J ) 7.3, 3H), 0.94 (t, J ) 7.4, 3H), 1.12-1.24 (m, 2H),
1.51-1.67 (m, 2H), 2.91-2.99 (m, 2H), 3.24 (t, J ) 7.5, 2H),
6.83 (s, 1H), 7.28-7.34 (m, 2H), 7.41-7.46 (m, 2H), 7.67 (t, J
) 8.0, 1H), 7.85-7.93 (m, 1H), 8.05 (d, J ) 9.0, 1H), 8.26 (d,
J ) 8.5, 1H), 8.72 (s, 1H). Anal. (C25H24N3O2Cl) C, H, N.
(()-2-(4-Ch lor op h en yl)-2,3-d ih yd r o-3-(m eth oxyca r bo-
n yl)-1H-p yr r olo[3,4-b]qu in olin -1-on e (14). A mixture of 11
(1.0 g, 3.4 mmol) with sodium hydride (0.24 g, 10.0 mmol) in
dimethyl carbonate (12 mL, 142.4 mmol) and anhydrous DMF
(4 mL) was gently refluxed for 30 min under argon atmosphere.
Then, the dark violet mixture was poured into ice-water and
acidified with 0.6 N HCl. The precipitate was collected by
filtration, washed with water and then with diethyl ether, and
dried under reduced pressure to give 1.1 g of pure 14 as a white
solid (yield 92%). An analytical sample recrystallized from
ethyl acetate melted at 243-244 °C: 1H-NMR (CDCl3) 3.78
(s, 3H), 5.90 (s, 1H), 7.40-7.45 (m, 2H), 7.65-7.76 (m, 3H),
7.90 (t, J ) 8.1, 1H), 8.05 (d, J ) 8.0, 1H), 8.27 (d, J ) 8.5,
1H), 8.74 (s, 1H). Anal. (C19H13N2O3Cl) C, H, N.
Eth yl 2-(Ch lor om eth yl)-3-qu in olin eca r boxyla te (10).
To 1.7 mL (22.0 mmol) of anhydrous DMF cooled at 0-5 °C
was slowly added phosphorus oxychloride (6.0 mL, 65.5 mmol).
The resultant reagent was stirred at room temperature for 30
min and then cooled to 0-5 °C. A solution of 9 (4.8 g, 20.0
mmol) in chloroform (40 mL) was added, and the reaction
mixture was heated to reflux for 45 min. Then, the cooled
reaction mixture was poured into saturated sodium bicarbon-
ate solution and extracted with chloroform (3 × 20 mL). The
combined organic extracts were washed with saturated sodium
bicarbonate solution (2 × 20 mL), dried over sodium sulfate,
and concentrated under reduced pressure. The residue was
recrystallized from ethanol to give 10 (3.2 g, yield 64%) as
white prisms. An analytical sample melted at 120-121 °C:
1H-NMR (CDCl3) 1.48 (t, J ) 7.1, 3H), 4.50 (q, J ) 7.2, 2H),
5.28 (s, 2H), 7.62 (t, J ) 7.4, 1H), 7.79-7.94 (m, 2H), 8.13 (d,
J ) 8.4, 1H), 8.82 (s, 1H). Anal. (C13H12NO2Cl) C, H, N.
2-(4-Ch lor op h en yl)-2,3-d ih yd r o-1H-p yr r olo[3,4-b]qu in -
olin -1-on e (11). A mixture of 10 (2.3 g, 9.2 mmol) in ethanol
(60 mL) with 4-chloroaniline (3.5 g, 27.4 mmol) was refluxed
for 53 h. The reaction mixture was allowed to reach room
temperature, and the crystals formed were collected by filtra-
tion, washed with cold ethanol and then with light petroleum
ether, and finally dried under reduced pressure. Thus, 2.5 g
(yield 92%) of pure 11 was obtained as white needles. An
analytical sample recrystallized from ethanol-DMF melted at
268-270 °C: 1H-NMR (CDCl3) 5.03 (s, 2H), 7.41-7.46 (m, 2H),
7.66 (t, J ) 7.3, 1H), 7.83-7.92 (m, 3H), 8.04 (d, J ) 8.3, 1H),
8.19 (d, J ) 8.5, 1H), 8.71 (s, 1H). Anal. (C17H11N2OCl) C, H,
N.
(E)-3-(Ca r boxym eth ylen e)-2-(4-ch lor op h en yl)-2,3-d ih y-
d r o-1H-p yr r olo[3,4-b]qu in olin -1-on e (12). A mixture of 11
(0.3 g, 1.02 mmol) in glacial acetic acid (5 mL) and acetic
anhydride (5 mL) with glyoxylic acid monohydrate (0.3 g, 3.26
mmol) was heated at 110 °C for 5 h under nitrogen atmo-
sphere. Then 0.3 g of glyoxylic acid monohydrate was added,
and the mixture was stirred at 110 °C for an additional 1 h.
Afterward the cooled reaction mixture was poured into ice-
water and allowed to stand at room temperature overnight.
The precipitate was extracted with chloroform (2 × 50 mL),
and the organic layer was extracted with diluted NaOH (pH
9-10) (2 × 60 mL). The organic layer was discarded, and the
acqueous phase was acidified (pH 4-5) with 3 N HCl and
extracted with chloroform (3 × 20 mL). The combined extracts
were dried over sodium sulfate and concentrated under
reduced pressure to give crude 12 which was purified by
column chromatography eluting with chloroform-ethyl acetate
(8:2) to give 0.21 g of pure 12 as white crystals (yield 59%).
An analytical sample recrystallized from ethanol-chloroform
melted at 272-273 °C: 1H-NMR (CDCl3) 5.89 (s, 1H), 7.27-
7.35 (m, 2H), 7.54-7.61 (m, 2H), 7.86 (t, J ) 8.0, 1H), 8.03-
8.11 (m, 1H), 8.19 (d, J ) 8.4, 1H), 8.32 (d, J ) 8.5, 1H), 8.96
(s, 1H), 16.13 (s, 1H). Anal. (C19H11N2O3Cl‚1/3H2O) C, H, N.
(Z)-2-(4-Ch lor op h en yl)-2,3-d ih yd r o-3-[[(N,N-d i-n -p r o-
p yla m in o)ca r bon yl]m eth ylen e]-1H-p yr r olo[3,4-b]qu in o-
lin -1-on e (13). A mixture of 12 (0.094 g, 0.27 mmol) in thionyl
chloride (3 mL, 41 mmol) was stirred at room temperature for
1.5 h. The excess of thionyl chloride was removed under
reduced pressure, and toluene (15 mL) was added. To the
resulting suspension, cooled at 0-5 °C, was added N,N-di-n-
propylamine (1.0 mL, 7.29 mmol). After stirring at room
temperature for 15 min, the mixture was diluted with dichlo-
romethane (30 mL), washed with water (3 × 20 mL), dried,
and finally concentrated under reduced pressure. Purification
of the residue by chromatography, eluting with chloroform-
ethyl acetate (8:2), gave first 0.017 g (yield 14%) of pure 4a
(()-2-(4-Ch lor op h en yl)-2,3-d ih yd r o-3-[(eth oxyca r bon -
yl)m eth yl]-3-(m eth oxyca r bon yl)-1H-p yr r olo[3,4-b]qu in o-
lin -1-on e (15). To a solution of 14 (0.1 g, 0.28 mmol) in
anhydrous DMF (15 mL) under an argon atmosphere was
slowly added DBU (0.45 mL, 3.0 mmol) under stirring at room
temperature. After 10 min at room temperature to the dark
blue-colored solution was slowly added ethyl bromoacetate (0.4
mL, 3.6 mmol) up to the complete decoloration of the solution.
The mixture was stirred for 10 min at room temperature and
then poured into ice-water and extracted with dichlo-
romethane (4 × 20 mL). The combined extracts were washed
with water (5 × 50 mL), dried over sodium sulfate, and
concentrated under reduced pressure. Purification of the
residue by column chromatography, eluting with chloroform-
ethyl acetate (8:2), gave 15 (0.11 g, yield 90%) as a thick oil
which was characterized only by its 1H-NMR spectrum: 1H-
NMR (CDCl3) 0.92 (t, J ) 7.4, 3H), 3.29 (d, J ) 17.7, 1H), 3.72
(s, 3H), 3.76-3.96 (m, 3H), 7.28-7.34 (m, 2H), 7.39-7.46 (m,
2H), 7.66 (t, J ) 7.2, 1H), 7.82-7.90 (m, 1H), 8.05 (d, J ) 7.7,
1H), 8.21 (d, J ) 8.5, 1H), 8.74 (s, 1H).
(()-3-(Ca r boxym eth yl)-2-(4-ch lor op h en yl)-2,3-d ih yd r o-
1H-p yr r olo[3,4-b]qu in olin -1-on e (16). To a solution of 15
(0.55 g, 1.25 mmol) in ethanol (35 mL) was added 3 N sodium
hydroxide solution (15 mL, 45.0 mmol), and the resulting
mixture was stirred at room temperature for 2 h. Then, the
organic solvent was removed under reduced pressure, and
water (50 mL) was added. The basic solution was extracted
with dichloromethane (2 × 15 mL), and the organic extracts
were discarded, while the aqueous solution was acidified with
3 N HCl. The precipitate was collected by filtration and dried
to give crude 16 which was purified by recrystallization from
ethanol-chloroform (0.36 g, yield 82%). An analytical sample
melted at 280-282 °C dec: 1H-NMR (DMSO-d6) 2.86-3.18 (m,
2H), 5.84 (t, J ) 4.1, 1H), 7.56 (d, J ) 8.7, 2H), 7.69-7.74 (m,
3H), 7.93 (t, J ) 7.5, 1H), 8.15 (d, J ) 8.5, 1H), 8.25 (d, J )
8.2, 1H), 8.88 (s, 1H), 12.17 (s, 1H). Anal. (C19H13N2O3Cl) C,
H, N.
6-(4-Ch lor op h en yl)-6,7-d ih yd r o-5H-p yr r olo[3,4-b]p yr i-
d in -5-on e (18). A mixture of ethyl 2-methylnicotinate (17)
(1.6 mL, 10.4 mmol) in carbon tetrachloride (50 mL) with NBS
(1.7 g, 9.6 mmol) and 75% dibenzoyl peroxide (0.2 g, 0.6 mmol)
was refluxed for 17 h. Then, the cooled reaction mixture was
filtered, and the filtrate was concentrated under reduced
pressure. The residue was dissolved in ethanol (45 mL), and
4-chloroaniline (4.0 g, 31.3 mmol) was added. After refluxing
for 3 days, the reaction mixture was allowed to reach room
temperature and a solid crystallized on standing. The crystals
were collected by filtration, washed with cold ethanol and then
with light petroleum ether, and finally dried to give pure 18
(0.9 g, yield 35%). An analytical sample recrystallized from
ethanol melted at 198-199 °C: 1H-NMR (CDCl3) 4.90 (s, 2H),
7.38-7.49 (m, 3H), 7.79-7.86 (m, 2H), 8.17-8.21 (m, 1H),
8.78-8.81 (m, 1H). Anal. (C13H9N2OCl) C, H, N.

New Class of PBR Ligands Related to Alpidem
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 21 4283
(E)-7-(Ca r boxym eth ylen e)-6-(4-ch lor op h en yl)-6,7-d ih y-
d r o-5H-p yr r olo[3,4-b]p yr id in -5-on e (19). A mixture of 18
(0.3 g, 1.23 mmol) in glacial acetic acid (5 mL) and acetic
anhydride (5 mL) with glyoxylic acid monohydrate (0.3 g, 3.26
mmol) was refluxed for 3 h under a nitrogen atmosphere.
Then the cooled reaction mixture was poured into ice-water
and allowed to stand at room temperature ovenight. The dark
precipitate was collected by filtration, washed with water,
dried, and purified by column chromatography eluting with
chloroform-ethyl acetate (8:2) to give pure 19 (0.22 g, yield
59%). An analytical sample recrystallized from ethanol-
chloroform melted at 278-279 °C dec: 1H-NMR (CDCl3) 5.86
(s, 1H), 7.25-7.29 (m, 2H), 7.53-7.58 (m, 2H), 7.76-7.83 (m,
1H), 8.46 (d, J ) 7.9, 1H), 8.89-8.93 (m, 1H), 15.51 (s, 1H).
Anal. (C15H9N2O3Cl) C, H, N.
(()-6-(4-Ch lor op h en yl)-6,7-d ih yd r o-7-(m eth oxyca r bo-
n yl)-5H-p yr r olo[3,4-b]p yr id in -5-on e (20). Starting from 18
(0.8 g, 3.27 mmol) the title compound 20 was prepared
following the procedure described above for 14. Pure 20 was
obtained as a white solid (0.82 g, yield 83%). An analytical
sample was obtained from ethyl acetate as white needles
melting at 223-224 °C dec: 1H-NMR (CDCl3) 3.77 (s, 3H), 5.78
(s, 1H), 7.38-7.43 (m, 2H), 7.49-7.56 (m, 1H), 7.64-7.69 (m,
2H), 8.20-8.25 (m, 1H), 8.80-8.83 (m, 1H). Anal. (C15H11N2O3-
Cl) C, H, N.
(()-7-(Ca r boxym eth yl)-6-(4-ch lor op h en yl)-6,7-d ih yd r o-
5H-p yr r olo[3,4-b]p yr id in -5-on e (21). To a solution of 20
(0.5 g, 1.65 mmol) in anhydrous DMF (35 mL) under an argon
atmosphere was added DBU (2.4 mL, 16.05 mmol) at room
temperature. After stirring at the same temperature for 10
min, to the resulting red solution was slowly added ethyl
bromoacetate (1.6 mL, 14.43 mmol) up to the complete decol-
oration of the solution. The mixture was stirred for 10 min
at room temperature and then poured into ice-water and
extracted with dichloromethane (4 × 25 mL). The organic
layer was thoroughly washed with water (5 × 60 mL), dried
over sodium sulfate, and concentrated under reduced pressure.
The residue was dissolved in ethanol (35 mL), and a 3 N
sodium hydroxide solution (15 mL, 45.0 mmol) was added
under vigorous stirring at room temperature. After 2 h at the
same temperature, the organic solvent was removed under
reduced pressure and water (50 mL) was added. The basic
solution was extracted with dichloromethane (2 × 15 mL); the
combined organic extracts were discarded, and the aqueous
solution was acidified with 3 N HCl. The precipitate was
collected by filtration, washed with water, and dried to give
pure 21 (0.29 g, yield 58%). An analytical sample obtained
from ethyl acetate melted at 196-198 °C: 1H-NMR (DMSO-
d6) 2.71-3.01 (m, 2H), 5.70 (t, J ) 4.5, 1H), 7.50-7.67 (m, 5H),
8.17 (d, J ) 8.1, 1H), 8.80 (d, J ) 5.0, 1H), 12.15 (s, 1H). Anal.
(C15H11N2O3Cl) C, H, N.
geometry-optimized fragments (Tripos fragment library as
implemented in Sybyl 6.1) and subsequent minimization of the
potential energy to the closest local minimum using the
standard Tripos force field.19 Charges were calculate using
the method of Gasteiger and Marsili.20 Molecular dynamics
simulation runs were performed using the following protocol:
10 ps of thermal equilibration at 1000 K, 200 ps simulation
at 300 K; time step, 1 fs; temperature-coupling factor, 10 fs;
nonbonded reset frequency, 25 fs; momentum removal fre-
quency, 25 fs; starting velocity, Boltzman.
Conformers were saved every 100 fs from the simulation
period at 300 K resulting in a total number of 2000 structures.
For conformer classification, two geometric features were
defined and evaluated as follows: (i) a centroid (dummy atom
located in the rms distance of all carbon atoms of the phenyl
substituent) and (ii) a plane (defined by rms of all atoms of
the fused aromatic system). The distance between the centroid
and the amide oxygen atom and the height of the latter above
the plane were measured in all conformers recorded.
In Vitr o Bin d in g Assa ys. Meth od A. Male CRL:CD(SD)-
BR-COBS rats were killed by decapitation. Their brains were
rapidly dissected into the various areas and stored at -80 °C
until the day of assay.
CBR. The synaptosomal membrane of rat cerebral cortex
was prepared according to Mohler and Okada.21 Rat cerebral
cortices were homogenized (Polytron PTA 10TS) in 22 vol of
0.32 M sucrose at 0 °C (3 × 20 s). The homogenate was
centrifuged for 10 min at 1000g at 4 °C, and the supernatant
was recentrifuged for 10 min at 20000g (Beckman model J 2-
21 refrigerated centrifuge). The pellet was resuspended in 22
vol of phosphate buffer (25 mM, pH 7.4) and centrifuged as
before. The final pellet was resuspended in the incubation
buffer (phosphate, 25 mM, pH 7.4) just before the binding
assay.
[3H]Flunitrazepam (specific activity 79.6 Ci/mmol; NEN)
binding was assayed in a final incubation volume of 1.0 mL,
consisting of 0.5 mL of membrane suspension (2.5 mg of tissue,
wet weight), 0.1 mL of [3H]flunitrazepam (final concentration
) 1 nM), 0.1 mL of displacing agent (dissolved in 5% DMSO)
or solvent, and 0.3 mL of the incubation buffer.
Incubations (20 min at room temperature) were terminated
by rapid filtration under vacuum through GF/B filters which
were then washed with 13.5 mL (3 × 4.5 mL) of phosphate
buffer (25 mM, pH 7.4) using a Brandel M-24R cell harvester.
Nonspecific binding was defined as nondisplaceable binding
in the presence of diazepam (10 µM). Blank experiments were
carried out to determine the effect of the solvent (5% DMSO)
on the binding.
P BR. The binding assays were performed as described in
ref 22. The frozen cortices were homogenized in 50 vol of ice-
cold phosphate buffer, 50 mM, pH 7.4, using a Polytron PTA
10TS (3 × 20 s). The homogenate was centrifuged for 10 min
at 1000g at 4 °C, and the supernatant was recentrifuged at
50000g for 10 min (Beckman model J 2-21 refrigerated centri-
fuge). Each pellet was washed another three times by resus-
pension in the same volume of fresh buffer and centrifuged
again at 50000g for 10 min without intermediate incubations.
The pellet obtained was finally resuspended in phosphate (50
mM, pH 7.4) incubation buffer just before the binding assay.
[3H]PK11195 (specific activity 85.5 Ci/mmol; NEN) binding
was assayed in a final incubation volume of 1.0 mL, consisting
of 0.5 mL of membrane suspension, 0.1 mL of [3H]ligand, and
0.1 mL of displacing agent or solvent. Tissue concentration
and [3H]PK11195 final concentration were 2.3 mg of tissue/
sample and 1 nM, respectively.
X-r a y Cr ysta llogr a p h y. A single crystal of 4e was sub-
mitted to X-ray data collection on a Siemens P4 four-circle
diffractometer with graphite-monochromated Mo KR radiation.
The ω/2θ scan technique was used. The structure was solved
by direct methods. The refinement was carried out by full-
matrix anisotropic least-squares on F for all non-H atoms by
minimizing the function ∑w(|Fo| - |Fc|)2. Atomic scattering
factors including f ′ and f′′ were taken from ref 17. Structure
solution, analysis, and refinement were carried out by using
the SHELXTL PC package.17
4e: C27H20N3O3Cl (MW 469.9), a pale-yellow single crystal,
dimensions 0.3 × 0.3 × 0.08 mm, was used for data collection;
orthorhombic; space group Pbca; a ) 19.613(4) Å, b ) 10.456-
(2) Å, c ) 22.118(4) Å, V ) 4536(2) Å3, Z ) 8, Dc ) 1.38 g/cm3;
2961 independent reflections (Rint ) 0.016) were collected at
22 °C, of which 1985 were observed with F > 3σ(F). No
absorption correction was applied. The final refinement
converged to R ) 0.047 and Rw ) 0.036. Minimum and
maximum heights in last ∆F map of -0.23 and 0.21 eÅ-3. Full
crystallographic details will be given elsewhere.
Incubations (120 min at 4 °C) were stopped by rapid
filtration under vacuum through GF/B filters which were then
washed with 12 mL (3 × 4 mL) of ice-cold phosphate buffer
(50 mM, pH 7.4) using a Brandel M-24R cell harvester.
Nonspecific binding was assayed in the presence of PK11195
(10 µM).
Molecu la r Mod elin g a n d Com p u ta tion a l P r oced u r es.
The entire molecular modeling study was performed using the
a SGI Indy 4400
workstation. Molecules were built up either using X-ray
Dried filters were immersed in vials containing 10 mL of
Filter Count (Packard) liquid scintillation cocktail for the
measurement of trapped radioactivity with a Packard Tri-Carb
300C liquid scintillation spectrometer at a counting efficiency
of about 60%. The concentration of the test compounds that
Sybyl18 software package installed on
coordinates (alpidem, 4e) or de novo (7a ) starting from

4284 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 21
Anzini et al.
inhibited [3H]ligand binding by 50% (IC50) was determined by
log-probit analysis with six concentrations of the displacers,
each performed in duplicate.
Fundation Symposium Series; Costa, E., Paul, S. M., Eds.;
Thieme Medical Publisher, Inc.: New York, 1991; pp 171-176.
(c) Romeo, E.; Auta, J .; Kozikowski, A. P.; Ma, D.; Papadopoulos,
V.; Puia, G.; Costa, E.; Guidotti, A. 2-Aryl-3-Indoleacetamides
(FGIN-1): A New Class of Potent and Specific Ligands for the
Mitochondrial DBI Receptor (MDR). J . Pharmacol. Exp. Ther.
1992, 262, 971-978. (d) Kozikowski, A. P.; Ma, D.; Romeo, E.;
Auta, J .; Papadopoulos, V.; Puia, G.; Costa, E.; Guidotti, A.
Synthesis of (2-Arylindol-3-yl)acetamides as Probes of Mitochon-
drial Steroidogenesis - A New Mechanism for GABAA Receptor
Modulation. Angew. Chem., Int. Ed. Engl. 1992, 31, 1060-1062.
(7) Tebib, S.; Bourguignon, J . J .; Wermuth, C. G. The Active
Analogue Approach Applied to the Pharmachophore Identifica-
tion of Benzodiazepine Receptor Ligands. J . Comput.-Aided Mol.
Des. 1987, 1, 153-170 and references cited therein.
(8) Tomczuk, B. E.; Taylor, C. R.; Moses, L. M.; Sutherland, D. B.;
Lo, Y. S.; J ohnson, D. N.; Kinnier, W. B.; Kilpatrick, B. F.
2-Phenyl-3H- imidazo[4,5-b]pyridine-3-acetamides as Non-Ben-
zodiazepine Anticonvulsants and Anxiolytics. J . Med. Chem.
1991, 34, 2993-3006.
(9) (a) Anzini, M.; Cappelli, A.; Vomero, S.; Giorgi, G.; Langer, T.;
Hamon, M.; Nacera, M.; Emerit, B. M.; Cagnotto, A.; Skorupska,
M.; Mennini, T.; Pinto, J . C. Novel Potent and Selective 5-HT3
Receptor Antagonists Based on Arylpiperazine Skeleton: Syn-
thesis, Structure, Biological Activity and CoMFA Studies. J .
Med. Chem. 1995, 38, 2692-2704. (b) Cappelli, A.; Anzini, M.;
Vomero, S.; Menziani, M. C.; De Benedetti, P. G.; Sbacchi, M.;
Mennuni, L. Synthesis, Biological Evaluation, and Quantitative
Receptor Docking Simulation of 2-Acylaminoethyl-1,4-benzodi-
azepines as Novel Tifluadom-Like Ligands with High Affinity
and Selectivity for κ-Opioid Receptors. J . Med. Chem. 1996, 39,
860-872.
(10) Borea, P. A.; Gilli, G.; Bertolasi, V.; Ferretti, V. Stereochemical
Features Controlling Binding and Intrinsic Activity Properties
of Benzodiazepine-Receptor Ligands. Mol. Pharmacol. 1987, 31,
334-344.
(11) Georges, G.; Vercauteren, D. P.; Vanderveken, D. J .; Horion, R.;
Evrard, G.; Fripiat, J . G.; Andre, J . M.; Durant, F. Structural
and Electronic Analysis of Peripheral Benzodiazepine Ligands:
Description of the Pharmacophoric Elements for Their Receptor.
Int. J . Quant. Chem. Quant. Biol. Symp. 1990, 17, 1-25.
(12) Adams, D. R.; Dominguez, J . N.; Perez, J . A. Synthesis of
Quinolines by Reaction of Anilinobutenoates with Vilsmeier
Reagent. Tetrahedron Lett. 1983, 24, 517-518.
(13) Anzini, M.; Cappelli A.; Vomero, S. Synthesis of 2-Substituted
2,3-dihydro-9-phenyl-1H-pyrrolo[3,4-b]quinolin-3-ones as Poten-
tial Peripheral Benzodiazepine Receptor Ligands. Heterocycles
1994, 38, 103-111.
(14) Georges, G. J .; Vercauteren, D. P.; Evrard, G. H.; Durant, F. V.;
George, P. G.; Wick, A. E. Characterization of the Physico-
Chemical Properties of the Imidazopyridine derivative Alpidem.
Comparison with Zolpidem. Eur. J . Med. Chem. 1993, 28, 323-
335.
Meth od B. [3H]Ro5-4864 or [3H]PK11195 binding to PBR
in rat brain and kidney was performed as previously de-
scribed.23 Rats were sacrificed according to AAALAC guide-
lines and the brain and a kidney removed and placed into ice-
cold 0.32 M sucrose. The brain (1.5-2 g) and a section of
kidney (100-150 mg) were homogenized in 50 vol of 50 mM
Tris-HCl (pH 7.4, 0-4 °C) using a Polytron (Brinkmann
Instruments, Westbury, NY). The homogenates were centri-
fuged once at 20000g for 20 min, and the pellet was retained
and resuspended in 100 (brain) or 400 (kidney) vol of 50 mM
Tris-HCl buffer. An aliquot of tissue suspension (250 µL of
brain ≈ 200 µg of protein; 50 µL of kidney ≈ 20 µg of protein)
was added to each assay tube in addition to 50 µL of [3H]Ro5-
4864 (specific activity 86.9 Ci/mmol) or [3H]PK11195 (specific
activity 83.5 Ci/mmol; DuPont/New England Nuclear, Boston,
MA). The final concentrations used of both radioligands were
1 nM for competition assays and 0.25-10 nM for saturation
assays.
Nonspecific binding was determined using either unlabeled
PK11195 or Ro5-4864 (50 µM, final concentration 10 µM) for
[3H]Ro5-4864 or [3H]PK11195 binding, respectively. Sufficient
buffer was added to bring the final volume to 500 µL. All
assays were performed in duplicate. The assay was initiated
by the addition of tissue and terminated after 60 min of
incubation at 0-4 °C ([3H]Ro5-4864) or at room temperature
([3H]PK11195) by rapid filtration over no. 32 glass fiber filters
(Schleicher & Schuell, Keene, NH) using a Brandel M-24R
filtering manifold (Brandel Instruments, Gaithersburg, MD).
The samples were washed with two 5 mL aliquots of 50 mM
Tris-HCl buffer (pH 7.4, 0-4 °C). The radioactivity retained
by the filters was measured by an LS 5801 liquid scintillation
spectrometer (Beckman Instruments, Fullerton, CA) using 4
mL of scintillation cocktail (Cytoscint, ICN Biomedical, Aurora,
OH). Protein concentrations were determined using the
bicinchoninic acid technique (Pierce Chemicals, Rockland, IL).
Sigmoidal competition and hyperbolic saturation curves
were fit to the appropriate data using nonlinear regression
binding techniques. The goodness-of-fit of curves modeling
single and multiple binding sites to the data was assessed
using the F-test. All analyses were performed using Prism2
(GraphPad Softaware, San Diego, CA).
(15) Lentini, G.; Bourguignon, J . J .; Wermuth, C. G. Ligands of the
Peripheral-Type Benzodiazepine Binding Sites (PBS): Structure-
Activity Relationships and Computer-Aided Conformational
Analysis. In QSAR: Rational Approaches to the Design of
Bioactive Compounds; Silipo, C., Vittoria, A., Eds.; Elsevier
Science Publishers B.V.: Amsterdam, 1991; pp 257-260.
(16) Since the amide nitrogen of this compound bears two different
substituents, its 1H-NMR spectrum shows the presence of two
different rotamers in equilibrium. For the sake of simplification
the integral intensities are not given.
(17) SHELXTL PC, Siemens Analytical X-Ray Instruments, Inc.,
Madison, WI; Rel. 4.1, 1990.
(18) SYBYL 6.1, Tripos Assoc., 1699 S. Hanley Rd., Suite 303, St.
Louis, MO 63144.
(19) Clark, M.; Cramer, R. D., III; Van Opdenbosch, N. Validation
of the General Purpose TRIPOS 5.2 Force Field. J . Comput.
Chem. 1989, 10, 982-1012.
(20) Gasteiger, J .; Marsili, M. Iterative Partial Equalization of Orbital
Electronegativity - A Rapid Access to Atomic Charges. Tetrahe-
dron 1980, 36, 3219-3288.
(21) Mohler, H.; Okada, T. Benzodiazepine Receptor: Demonstration
in the Central Nervous System. Science 1977, 198, 849-851.
(22) Anzini, M.; Cappelli, A.; Vomero, S.; Cagnotto, A.; Skorupska,
M. Synthesis and Benzodiazepine Receptor Affinity of 2,3-
Dihydro-9-phenyl-1H-pyrrolo[3,4-b]quinolin-1-one and 3-Carbe-
thoxy-4-phenylquinoline Derivatives. Farmaco 1992, 47, 191-
202.
(23) (a) Basile, A. S.; Klein, D. C.; Skolnick, P. Characterization of
Benzodiazepine Receptors in the Bovine Pineal Gland: Evidence
for the Presence of an Atypical Binding Site. Mol. Brain Res.
1986, 1, 127-135. (b) Basile, A. S.; Ostrowsky, N.; Skolnick, P.
Aldosterone Reversible Decrease in the Density of Renal Pe-
ripheral Benzodiazepine Receptors in the Rat After Adrenalec-
tomy. J . Pharmacol. Exp. Ther. 1987, 240, 1006-1013.
Ack n ow led gm en t. The authors are grateful to Dr.
J ean J acques Bourguignon for his collaboration. This
work was financially supported by Italian MURST (60%
and 40% funds).
Refer en ces
(1) Fryer, R. I. Ligand Interaction at the Benzodiazepine Receptor.
In Comprehensive Medicinal Chemistry; Hansch, C., Sammes,
P. G., Taylor, J . B., Eds.; C. A. Ramsden Pergamon Press: New
York, 1990; Vol. 3, Chapter 12.8, pp 539-566 and references
cited therein.
(2) (a) Braestrup, C.; Squires, R. F. Specific Benzodiazepine Recep-
tors in Rat Brain Characterized By High Affinity [3H]Diazepam
Binding. Proc. Natl. Acad. Sci. U.S.A. 1977, 74, 3805-3809. (b)
Taniguchi, T.; Wang, J . K. T.; Spector, S. [3H]Diazepam Binding
Sites on Rat Heart and Kidney. Biochem. Pharmacol. 1982, 31,
589-590.
(3) Gavish, M.; Katz, Y.; Bar-Ami, S.; Weizman, R. Biochemical,
Physiological, and Pathological Aspects of the Peripheral Ben-
zodiazepine Receptor. J . Neurochem. 1992, 58, 1589-1601.
(4) Peripheral Benzodiazepine Receptors; Giesen-Crouse, E., Ed.;
Academic Press: London, 1993.
(5) Kozikowski, A. P.; Ma, D.; Brewer, J .; Sun, S.; Costa, E.; Romeo,
E.; Guidotti, A. Chemistry, Binding Affinities, and Behavioral
Properties of a New Class of “Antineophobic” Mitochondrial DBI
Receptor Complex (mDRC) Ligands. J . Med. Chem. 1993, 36,
2908-2920.
(6) (a) Bourguignon, J . J . Endogenous and Synthetic Ligand of
Mitochondrial Benzodiazepine Receptors: Structure-Affinity
Relationships. In Peripheral Benzodiazepine Receptors; Giesen-
Crouse, E., Ed.; Academic Press: London, 1993; pp 59-85. (b)
Costa, E.; Romeo, E.; Auta, J .; Papadopoulos, V.; Kozikowski,
A. P.; Guidotti, A. Is There a Pharmacology of Brain Steroido-
genesis? In Neurosteroid and Brain Function, Fidia Research
J M960325J