Synthesis, biological activity, and SARs of pyrrolobenzoxazepine derivatives, a new class of specific 'peripheral-type' benzodiazepine receptor ligands

Article abstract of DOI:10.1021/jm960251b

The 'peripheral-type' benzodiazepine receptor (PBR) has been reported to play a role in many biological processes. We have synthesized and tested a novel series of PBR ligands based on a pyrrolobenzoxazepine skeleton, in order to provide new receptor ligands. Several of these new compounds proved to be high affinity and selective ligands for PBR, and benzoxazepines 17f and 17j were found to be the most potent ligands for this receptor to have been identified to date. The SAR and the molecular modeling studies detailed herein delineated a number of structural features required for improving affinity. Some of the ligands were employed as 'molecular yardsticks' to probe the spatial dimensions of the lipophilic pockets L1 and L3 in the PBR cleft and to determine the effect of occupation of L1 and L3 with respect to affinity, while other C-7 modified analogues provided information specifically on the hydrogen bonding with a putative receptor site H1. The new pyrrolobenzoxazepines were tested in rat cortex, a tissue expressing high density of mitochondrial PBR, and exhibited IC₅₀ and K(i) values in the low nanomolar or subnanomolar range, as measured by the displacement of [³H]PK 11195 binding. A subset of the highest affinity ligands was also found to have high affinities for [³H]PK 11195 and [³H]Ro 5-4864 binding in rat adrenal mitochondria. All the ligands in this subset are stimulators of steroidogenesis having similar potency and extent of stimulation as PK 11195 and Ro 5-4864 of steroidogenesis in the mouse Y-1 adrenocortical cell line.

Full text of DOI:10.1021/jm960251b

J . Med. Chem. 1996, 39, 3435-3450
3435
Articles
Syn th esis, Biologica l Activity, a n d SARs of P yr r oloben zoxa zep in e Der iva tives, a
New Cla ss of Sp ecific “P er ip h er a l-Typ e” Ben zod ia zep in e Recep tor Liga n d s1
Giuseppe Campiani,† Vito Nacci,*,† Isabella Fiorini,† Maria P. De Filippis,† Antonio Garofalo,† Silvia M. Ciani,†
Giovanni Greco,‡ Ettore Novellino,‡ D. Clive Williams,§ Daniela M. Zisterer,§ Margaret J . Woods,§
Camelia Mihai,§ Cristina Manzoni,| and Tiziana Mennini|
Dipartimento Farmaco Chimico Tecnologico, Universita’ di Siena, Banchi di Sotto 55, 53100 Siena, Italy,
Istituto di Ricerche Farmacologiche “Mario Negri”, via Eritrea 62, 40126 Milano, Italy, Department of Biochemistry,
Trinity College, Dublin 2, Ireland, and Dipartimento di Chimica Farmaceutica e Tossicologica, Universita’ di Napoli
“Federico II”, via D. Montesano 49, 80131 Napoli, Italy
Received April 1, 1996X
The “peripheral-type” benzodiazepine receptor (PBR) has been reported to play a role in many
biological processes. We have synthesized and tested a novel series of PBR ligands based on
a pyrrolobenzoxazepine skeleton, in order to provide new receptor ligands. Several of these
new compounds proved to be high affinity and selective ligands for PBR, and benzoxazepines
17f and 17j were found to be the most potent ligands for this receptor to have been identified
to date. The SAR and the molecular modeling studies detailed herein delineated a number of
structural features required for improving affinity. Some of the ligands were employed as
“molecular yardsticks” to probe the spatial dimensions of the lipophilic pockets L1 and L3 in
the PBR cleft and to determine the effect of occupation of L1 and L3 with respect to affinity,
while other C-7 modified analogues provided information specifically on the hydrogen bonding
with a putative receptor site H1. The new pyrrolobenzoxazepines were tested in rat cortex, a
tissue expressing high density of mitochondrial PBR, and exhibited IC50 and Ki values in the
low nanomolar or subnanomolar range, as measured by the displacement of [3H]PK 11195
binding. A subset of the highest affinity ligands was also found to have high affinities for
[3H]PK 11195 and [3H]Ro 5-4864 binding in rat adrenal mitochondria. All the ligands in this
subset are stimulators of steroidogenesis having similar potency and extent of stimulation as
PK 11195 and Ro 5-4864 of steroidogenesis in the mouse Y-1 adrenocortical cell line.
In tr od u ction
sion,25 but the involvement of PBR in the regulation of
steroid biosynthesis is probably the most intriguing
aspect of its biological role with respect to possible
therapeutic applications.26 Steroid hormones, such as
pregnenolone, 3R-hydroxy-5R-pregnan-20-one (3R-OH-
DHP), dehydroepiandrosterone, and 3R,21-dihydroxy-
5R-pregnan-20-one (THDOC) are mainly produced pe-
ripherally by the adrenal gland. It has been recently
demostrated that the synthesis of these steroids may
also occur within the CNS in glial cells.27 Recent reports
suggest that the PBR, located in the outer mitochondrial
membrane in glial and peripheral endocrine cells, plays
a crucial role in steroid biosynthesis28-32 by regulating
the rate-limiting translocation of cholesterol from the
outer to the inner mitochondrial membrane before its
transformation by cytochrome P450scc into pregnenolone
which is further metabolized to other bioactive hor-
mones. DBI, an endogenous PBR ligand abundant in
steroidogenic cells, and synthetic ligands, such as PK
11195, Ro 5-4864, and FGIN-1-27, facilitate the trans-
port of cholesterol inside the mitochondria, and are
found to increase the rate of pregnenolone synthesis in
mitochondrial fractions in Leydig and adrenocortical cell
lines.33-35 Adrenally derived or centrally produced
steroid hormones (e.g. 3R-OH-DHP, pregnenolone, and
THDOC), can influence neuronal excitability via differ-
ent targets. These steroids can bind with high affinity
a recognition site coupled to the GABAA-benzodiaz-
The “peripheral-type” benzodiazepine receptor (PBR)
is pharmacologically distinct from the central-type
benzodiazepine receptor (CBR) associated with GABA-
regulated chloride channels in the central nervous
system (CNS).2-6 Although the PBR has been found
within the CNS,4,7,8 mainly in astrocytes, it is abundant
in peripheral tissues, including kidney,9,10 heart,11,12
erythrocytes,13,14 and particularly in endocrine tissues
(testis and adrenal cortex).15 This receptor is character-
ized by high affinity binding of isoquinolines, such as
PK 11195 (1),16 of benzodiazepines, such as Ro 5-4864
(2),7 of imidazopyridines (alpidem 3),17 of indole deriva-
tives (4),18 and of benzothiazepines (5).19 The PBR has
been characterized as being predominantly associated
with mitochondria, specifically on the outer mitochon-
drial membrane.20 Physiologically, the PBR may be
involved in regulatory mechanisms that vary among the
different tissues in which it is present. PBR-mediated
effects include calcium homeostasis,21 lipid metabo-
lism,22 mitochondrial oxidation,23 inhibition of cell pro-
liferation,24 and alteration of protooncogene expres-
* To whom correspondence should be addressed.
† Dipartimento Farmaco Chimico Tecnologico, Universita’ di Siena.
‡ Dipartimento di Chimica Farmaceutica, Universita’ di Napoli
“Federico II”.
§ Department of Biochemistry, Trinity College, Dublin.
| Istituto di Ricerche Farmacologiche “Mario Negri”, Milano.
X Abstract published in Advance ACS Abstracts, August 1, 1996.
S0022-2623(96)00251-8 CCC: $12.00 © 1996 American Chemical Society

3436 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 18
Campiani et al.
Ch a r t 1
Sch em e 1
epine receptor complex and can modulate the action of
glutamate on NMDA receptors.36-38 Their behavioral
properties are probably the consequence of these inter-
actions.39 Thus PBR could represent an alternative site
for an indirect modulation of GABAergic and glutamater-
gic neurotransmission by increasing the level of periph-
erally or centrally derived steroid hormones, followed
by their GABAA and NMDA receptor modulation. In
this article we present a further development in our
research on novel PBR ligands. We detail herein the
synthesis of pyrrolobenzoxazepine derivatives 6 (Chart
1) and their structure-activity relationships (SARs) for
PBR affinity associated with variation of the substitu-
ents on the heterocyclic system. Several of these new
compounds bind the PBR with both high affinity and
specificity, and benzoxazepines 17f and 17j were found
to be the most potent ligands for this receptor to have
been identified to date. A subset of ligands with high
affinity in brain was further tested for their ability to
displace [3H]PK 11195 and [3H]Ro 5-4864 from the rat
adrenal receptor and for their potencies to modulate
steroid biosynthesis in a Y-1 adrenocortical cell line, and
all were found to stimulate pregnenolone formation with
potencies equivalent to PK 11195 and Ro 5-4864. The
SARs developed in the present study are discussed in
the light of a previously reported pharmacophore model,40
schematically represented in Chart 2. According to this
model, a hydrogen bond-donating function (H1) and two
lipophilic pockets (L1 and L3) within the receptor
binding site are available to our ligands. The presence
of an additional lipophilic region (L4) has been deduced
from analysis of the SARs relative to the newly inves-
tigated pyrrolobenzoxazepines.
Ch em istr y
The new pyrrolobenzoxazepinones 16a -c were syn-
thesized as shown in Scheme 1, according to the
procedure utilized to prepare already known 16d -g,41
starting from the appropriate 1-arylpyrrole. 1-(3-Chloro-
2-hydroxyphenyl)pyrrole (9) was synthesized following
the procedure illustrated in eq 1,42 while 13 was
prepared according to reference 43. The esters 14a -c
were obtained in good yield starting from 9 and 13 by
Ch a r t 2. H1 is a Hydrogen Bond Donor Site on the Receptor Protein. L1, L3, and L4 are Lipophilic Pockets in the
Protein at the Binding Site

Pyrrolobenzoxazepines as PBR Ligands
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 18 3437
Ta ble 1. Physical Data for Compounds 8-22
compd
A
X
R
R′
R′′
% yielda
mp (°C)
recryst solvent
EtOAc
hexanes
EtOAc/hexanes
formula
analysisb
8c
9
45
76
74
40
53
53
93
56
94
95
98
74
40
60
-
70-71
64-65
102-104
-
C6H4ClNO3
10H8ClNO
C8H7N3O2
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C
11a
12a
12b
14a
14b
14c
15a
15b
15c
16a
16b
16c
16d d
16ed
16fd
16gd
17a
17b
17c
17d
17e
17f
17g
17h
17i
17j
17k
17l
17m
17n
17o
17p
19
-
C
C
C
C
C
C
C
C
C
10H10BrN3O2
11H13BrO2
21H21NO3
20H17N4O3
20H18ClNO3
19H17NO3
18H14N4O3
18H14ClNO3
19H15NO2
C18H12N4O2
18H12ClNO2
-
-
-
-
-
-
-
-
140-142
130 (d)
220-221
79-81
113-115
122-123
-
hexanes
EtOAc/hexanes
hexanes
EtOH
benzo
benzo
benzo
benzo
benzo
[2,3]naphtho
benzo
benzo
benzo
benzo
benzo
benzo
benzo
benzo
benzo
benzo
benzo
H
H
Cl
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Cl
H
H
p-tolyl
p-N3-phenyl
phenyl
phenyl
2-naphthyl
phenyl
ethyl
ethyl
ethyl
phenyl
phenyl
phenyl
phenyl
phenyl
p-tolyl
p-tolyl
p-tolyl
EtOAc/hexanes
hexanes
-
C
-
-
-
-
-
-
-
-
-
-
-
-
-
O
O
O
O
O
O
S
O
O
O
O
O
O
O
O
O
Me
N(Me)2
Me
67
73
78
83
66
73
59
60
56
72
69
65
67
65
58
53
63
58
90
83
-
-
C
16H15NO3
C17H18N2O3
C20H15NO3
C25H23NO3
C21H18N2O3
C23H22N2O3
C23H22N2O2S
C21H17NO3
C22H20N2O3
C24H24N2O3
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
104-106
118-119
101-102
151-152
61-63
150-151
129-130
154-155
130-132
157-158
173-175
-
hexanes
EtOH
C6H11
N(Me)2
N(Et)2
N(Et)2
Me
N(Me)2
N(Et)2
Me
N(Me)2
N(Me)2
N(Me)2
Me
hexanes
EtOH
hexanes
hexanes
EtOH
hexanes/ethyl ether
hexanes/EtOAc
EtOH
EtOH
-
hexanes
EtOH
hexanes
hexanes/EtOAc
EtOH
hexanes
hexanes
benzo
benzo
benzo
benzo
[2,3]naphtho
[2,3]naphtho
2-naphthyl
2-naphthyl
p-N3-phenyl
phenyl
phenyl
phenyl
C24H17NO3
C25H20N2O3
C21H17N5O3
C21H17ClN2O3
24H17NO3
C25H20N2O3
20H15NO3
20H15NO2
19H15NO2
21H17NO3
175-177
164-165
189-191
119-120
108-110
88-89
69-71
C
N(Me)2
C
C
C
C
20
21
22
a
b
c
Yields refer to isolated and purified materials. All the compounds were analyzed within (0.4% of the theoretical values. Reference
d
42 (mp 66-70 °C, 20% yield). Reference 41.
Sch em e 2
treatment with sodium hydride and the appropriate
ethyl R-bromoarylacetate (12), 12a ,b in turn prepared
as shown in eq 2.44 Hydrolysis of esters 14a -c with
sodium hydroxide (15a -c) followed by intramolecular
cyclization using phosphorus pentachloride gave the
benzoxazepinones 16a -c. Treatment of the corre-
sponding potassium enolates of 16a -g with selected
acyl chlorides finally yielded the desired oxazepines
17a -p (see Table 1). The C-7 modified oxazepines 19,
20, and 22 were synthesized starting from 16d (see
Scheme 2) which was transformed to the ester deriva-
tive 19 by exposure of its triflate intermediate 18 to
carbon monoxide and methanol in presence of tetrakis-
(triphenylphosphine)palladium.45 The ketone 20 was

3438 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 18
Ta ble 2. Receptor Binding Affinity of Compounds 17a -p
Campiani et al.
compd
17a
17b
17c
17d
17e
17f
17g
17h
17i
17j
17k
17l
A
Y
R
R′
R′′
Me
N(Me)2
Me
C6H11
N(Me)2
N(Et)2
N(Et)2
Me
N(Me)2
N(Et)2
Me
IC50 (nM)a ((SEM)
Ki (nM)a ((SEM)
benzo
benzo
benzo
benzo
benzo
benzo
benzo
benzo
benzo
benzo
benzo
benzo
benzo
benzo
[2,3]naphtho
[2,3]naphtho
O
O
O
O
O
O
S
O
O
O
O
O
O
O
O
O
H
H
H
H
H
H
H
H
H
H
H
H
H
Cl
H
H
ethyl
ethyl
phenyl
phenyl
phenyl
phenyl
phenyl
p-tolyl
p-tolyl
>5000
589 ( 81
12.1 ( 1
81.9 ( 20
5.1 ( 0.3
0.68 ( 0.18
3.8 ( 1.3
2.3 ( 0.5
1.3 ( 0.2
0.94 ( 0.06
76.6 ( 7
6.0 ( 1.7
3.6 ( 0.5
1.3 ( 0.1
278 ( 19
37.3 ( 9.4
4.0 ( 1
230 ( 32
4.7 ( 0.4
32.0 ( 8
2.0 ( 0.12
0.26 ( 0.07
1.5 ( 0.5
0.91 ( 0.2
0.52 ( 0.08
0.36 ( 0.02
30.0 ( 2.7
2.4 ( 0.7
p-tolyl
2-naphthyl
2-naphthyl
p-N3-phenyl
phenyl
phenyl
phenyl
N(Me)2
N(Me)2
N(Me)2
Me
17m
17n
17o
17p
5a b
1.4 ( 0.2
0.51 ( 0.04
108.0 ( 7
14.5 ( 3.6
1.8 ( 0.4
N(Me)2
PK 11195
2.0 ( 0.1
0.78 ( 0.04
a
The concentration of the tested compounds that inhibited [3H]PK 11195 binding to rat cortex homogenate by 50% (IC50) was determined
by log-probit analysis with 6 concentrations of the displacers, each performed in triplicate. Values are the mean ( SE of at least three
b
separate experiments, performed in triplicate. Reference 19b.
Ta ble 3. Comparison of PBR Binding Affinity of Compounds
Modified at C-7
receptor protein. Binding data of PK 11195 is also
included. Benzoxazepine 17i, one of the most potent
PBR ligands, was found to be inactive on the chloride
channel associated to GABAA receptor and on CBR, as
indicated by its lack of inhibition of [35S]TBPS and [3H]-
flunitrazepam binding, respectively, to rat cerebral
membranes, even at the highest concentration tested
(10-5 M).19a This result suggests that the chemical
structures under study are selective for PBR and do not
compd
17c
17e
17f
17g
19
20
R
IC50 (nM)a ((SEM) Ki (nM)a ((SEM)
interact with CBR. Binding studies using [3H]PK 11195
and [3H]Ro 5-4864 were also carried out using rat
adrenal mitochondria for compounds 5a , 17e, 17h -j,
and 17l,m . Rat adrenals were chosen as an additional
source of tissue since they have previously been reported
to contain a very high density of high affinity PBR
binding sites15 with identical binding properties to the
brain receptor. The compounds tested in adrenal mi-
tochondria were found to exhibit significant binding
affinity (Ki values range from 0.8-4.8 nM) for the PBR.
In fact these Ki values are similar to the Kd values
obtained for [3H]PK 11195 and [3H]Ro 5-4864 binding
(1.5 and 2.2 nM, respectively) (Table 5). A subset of
PBR ligands showing high affinity for the PBR have
been examined for their ability to stimulate preg-
nenolone formation in Y-1 cells.
OCOMe
12.1 ( 1.0
5.1 ( 0.3
0.68 ( 1.8
3.8 ( 1.3
148 ( 39
64.6 ( 39
1930 ( 610
2.0 ( 0.1
4.7 ( 0.4
OCON(Me)2
OCON(Et)2
OCSN(Et)2
CO2Me
COMe
CH2OCOMe
2.0 ( 0.12
0.26 ( 0.07
1.5 ( 0.5
57.8 ( 15
25.2 ( 15
754 ( 237
0.78 ( 0.04
22
PK 11195
a
The concentration of the tested compounds that inhibited
[3H]PK 11195 binding to rat cortex homogenate by 50% (IC50) was
determined by log-probit analysis with 6 concentrations of the
displacers, each performed in triplicate. Values are the mean (
SE of at least three separate experiments performed in triplicate.
obtained by treatment of 19 with methyllithium at 0
°C46 (in these conditions from the reaction mixture was
recovered only 10% of the corresponding tertiary alco-
hol), while the ester 22 was synthesized by esterification
with acetyl bromide of 21, in turn prepared by lithium
aluminum hydride reduction of 19 (see Table 1).
Both PK 11195 and Ro 5-4864 have previously been
shown to maximally stimulate steroid production by
2-fold in Y-1 cells33,35 and to elicit similar effects in
MA10 Leydig cells,47 C6-2B glioma cells48 and in freshly
isolated rat adrenal cells,33 albeit at much higher
concentrations than theoretically required to saturate
the PBR. All the compounds tested stimulated ste-
roidogenesis in a concentration-dependent manner with
similar concentration dependencies to PK 11195 and Ro
5-4864. In order to make a direct comparison of
efficacies, all the compounds were tested in the same
experiment at a single concentration (40 µM). At the
same time clonazepam, a ligand selective for the central-
Biologica l Stu d ies
The affinities of the new pyrrolobenzoxazepine de-
rivatives for the PBR in rat cortex homogenate are
illustrated in Tables 2 and 3. For sake of comparison
the affinities of a set of previously described benzothi-
azepines are reported in Table 4. The binding data
presented in Tables 2 and 3 represent the ability of the
tested compounds to displace [3H]PK 11195 from the

Pyrrolobenzoxazepines as PBR Ligands
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 18 3439
Ta ble 4. Receptor Binding Affinity of Compounds 5a -g
compd
5a
5b
5c
5d
5e
5f
5g
R
R′
R′′
R′′′
phenyl
phenyl
phenyl
4′-Cl-phenyl
phenyl
phenyl
Riv
IC50 (nM) ((SEM)
Ki (nM) ((SEM)
ref
H
H
H
H
Me
H
H
H
H
H
H
H
Me
H
Cl
H
H
H
H
H
H
CON(Me)2
COMe
CO(NMe)2
COMe
COMe
COMe
4.0 ( 1
20 ( 2
9.0 ( 1
30 ( 3
1.8 ( 0.4
7.6 ( 0.7
3.5 ( 0.4
11.7 ( 1.1
263 ( 29
44 ( 8
19b
19a
19a
19b
19b
19b
19a
675 ( 75
112 ( 21
>105
4′-MeO-phenyl
CO-3′-pyridyl
PK 11195
2.0 ( 0.1
0.78 ( 0.04
Ta ble 5. Ki Values for Tested Compounds for Displacing
described in the Experimental Section and Results and
Discussion, respectively.
[3H]Ro 5-4864 and [3H]PK 11195 from the Rat Adrenal PBR
Resu lts a n d Discu ssion
A comparison of the IC50 values relative to the three
pairs of isosteric pyrrolobenzothiazepines (Table 4) and
pyrrolobenzoxazepines listed in Tables 2 and 3 (17c vs
5b, 17e vs 5c and 17n vs 5a ) reveals that replacement
of the endocyclic sulfur (S5) with an oxygen (O5)
increases affinities by 2-3 fold. This order of magnitude
is not large enough to propose an involvement of the
O5 atom in a hydrogen bond with the receptor. It is
well known that oxygen is a much better hydrogen bond
acceptor than sulfur in the sp3 hybridization state,50,51
and hence differences in potency should be more evident
if O5 was actually engaged in this type of interaction.
It is possible that the extra affinity of the pyrroloben-
zoxazepines originates from an improved stereoelec-
tronic complementarity between the O5 and a specific
subsite of the receptor. Alternatively, modifications of
C-S(O) distances and C-S(O)-C angle would introduce
changes in the overall 3D arrangement of the tricyclic
system and thus indirectly favor interactions involving
other parts of the ligand.
Figure 1 shows the hypothetical active conformations
of Ro 5-4864, 17c, and 5d aligned about the L1, L3, and
H1 pharmacophoric elements (compounds 17c and 5d
were chosen as representatives of the two isosteric
series). Table 7 summarizes the pharmacophore-
consistent conformations of all the modeled structures.
According to the pharmacophoric distances listed in
Table 7 and as illustrated in Figure 1, the pyrroloben-
zoxazepines are more superimposable on the L1, L3, and
H1 points of Ro 5-4864 than are their sulfurated
isosteres. One may speculate that the beneficial effect
of the S5/O5 replacement originates from a more precise
orientation of the anchoring functions into the receptor
cavity. Despite these geometric differences, it seems
reasonable to assume that pyrrolobenzoxazepines and
pyrrolobenzothiazepines share a similar binding mode
at the PBR. This assumption is supported by parallel
structure-affinity relationships in the two series. Specif-
ically, affinity is enhanced by conversion of the 7-acetoxy
into a 7-(dimethylcarbamoyl)oxy group (17c vs 17e and
5b vs 5c) or by the presence of a chlorine at the
4-position (17e vs 17n and 5c vs 5a ). The binding to
the receptor is weakened by substitutions at the 2- and
3-positions (17e vs 17p and 5b vs 5e and 5f).
Ki (nM) ((SEM)a
[3H]Ro
5-4864
[3H]PK
11195
compd
A
R
R′
X
5a
phenyl
phenyl
p-tolyl
p-tolyl
p-tolyl
2-naphthyl
p-N3-phenyl
Cl OCON(Me)2
S
2.6 ( 0.5 2.5 ( 0.6
0.8 ( 0.3 1.0 ( 0.2
1.7 ( 3.2 2.8 ( 1.0
1.3 ( 1.0 1.1 ( 0.1
1.9 ( 0.5 1.9 ( 0.4
1.8 ( 0.3 1.9 ( 0.7
3.6 ( 0.8 4.8 ( 1.2
17e
17h
17i
17j
17l
17m
H
H
H
H
H
H
OCON(Me)2
OCOMe
OCON(Me)2
OCON(Et)2
OCON(Me)2
OCON(Me)2
O
O
O
O
O
O
a
Ki values represent the mean ( SEM of three determinations.
The Ki values and errors for each compound were generated by
the use of the computer programmes EBDA and LIGAND.
type GABAA/benzodiazepine receptor, did not elicit any
stimulatory effect at this concentration. This suggests
that the stimulatory effect of the tested compounds on
steroid biosynthesis may be mediated through the PBR
(Table 6).
Preliminary studies have been carried out in order
to assess the potential of 17m as a photoaffinity ligand
for the PBR. Whereas light-dependent irreversible
inhibition of the binding of [3H]PK 11195 and [3H]Ro
5-4864 to rat adrenal mitochondria was observed, the
inhibition was not blocked by the presence of Ro 5-4864
and PK 11195. Further work is necessary to expand
these findings and to assess the receptor site involved
in the binding of the presented compounds.
Com p u ta tion a l Ch em istr y
Molecular modeling studies were performed to assist
the design of the investigated compounds and the
development of structure-affinity relationships.
Semiempirical quantum-mechanics and molecular me-
chanics methods were employed in combination to
identify pharmacophore-consistent conformations of
some structurally representative pyrrolobenzoxazepines.
Crystal structures determined by us40 or available from
the Cambridge Structural Database (CSD)49 were used
to compare experimental versus theoretical geometries.
Protocols and outcomes of the computational studies are
Compounds 17a and 17b, both bearing an ethyl in

3440 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 18
Campiani et al.
Ta ble 6. Stimulatory Effect on Steroid Biosynthesis of Y-1
Cells
is actually needed for nanomolar affinity as previously
postulated by us.40 The hydrophobic constants π of the
ethyl and phenyl substituents are 1.02 and 1.96, re-
spectively, within the classical reference system of
benzene.52 From the IC50 values of the pairs 17a /17c
and 17b/17e it can be concluded that halving the
lipophilicity of the 6-substituent reduces affinity by
more than a 100 fold. These data confirm the key role
of the L3 site in the binding to the PBR. By introducing
lipophilic substituents at the 4′-position of the 6-phenyl
ring we attempted to exploit the lipophilic interaction
at the L3 receptor region. Indeed, methylation at the
4′-position of 17c and 17e afforded 17h and 17j,
respectively, which turned out to be 4-5 times more
potent. The higher affinity of the 4′-azido derivative
17m , compared with 17e, might similarly be the con-
sequence of a more favorable hydrophobic interaction
at the L3 site (the π values52 of the methyl and azido
groups are 0.56 and 0.46, respectively). When the same
4′-methyl group was introduced in the structure of 17f,
the affinity of the resulting compound (17j) was com-
parable to that of 17f in rat brain. Clearly, in this case
we cannot make use of a simple additive model of
ligand-receptor interaction to explain why parallel
structural modifications do not produce parallel changes
in affinity. Replacement of the 6-phenyl with the 6-(2-
naphthyl) group (17k ,l vs 17c,e), and of the benzo-fused
ring with a [2,3]naphtho-fused ring (17o,p vs 17c,e) was
realized to probe the dimensions of the L1 and L3 sites.
While 17k and 17l retained a nanomolar affinity,
although lower than that of the parent compounds
17c,e, the compounds 17o,p showed an affinity 4- to
5-fold lower, indicating that 2- and 3- positions of the
benzo-fused ring of 17c,e are located in close proximity
of the steric boundaries of the L1 region, and that the
latter region cannot tolerate an additional large benzo-
fused ring which extends to an area of negative interac-
tion (17o,p ) (Chart 2).
Com p a r ison of a C-7 Ar oyl (5g) w ith a C-7
Cycloa lk a n oyl (17d ) Gr ou p . According to recently
published crystallographic and theoretical studies19,40
the inactivity of several 7-aroyloxy benzothiazepines
was thought to depend on steric and/or electrostatic
repulsive interactions involving their aryl moieties and
a specific site of the PBR. Such aryl groups might be
in fact too bulky and/or too electronegative to fit into
the receptor cavity. Addressing this issue might be
important for delineating the physicochemical nature
of the receptor region complementary to the 7-substitu-
ent of either pyrrolobenzothiazepines or pyrrolobenzox-
azepines. Compound 17d , featuring the encumbering
nonaromatic 7-cyclohexylcarbonyloxy side chain, was
therefore synthesized and tested. The affinity of 17d
is about 7 times lower than its reference analogue 17c
whereas, in the pyrrolobenzothiazepine series, the
insertion of a 7-aroyloxy group led to a total loss of
affinity (see 5g in Table 4). Figure 2 shows 17d and
5g in their pharmacophore-consistent conformation
(geometries are described in Table 7) along with their
electrostatic potentials contoured at -5 kcal/mol. It
should be noted that the side chains of the two molecules
are rather similar in shape but quite different in their
electrostatic properties. Taken together, these observa-
tions suggest that the disruptive effect of a 7-aroyloxy
group should be mainly ascribed to an electrostatic
repulsive interaction occurring between the π-conjugat-
a
The values represent the mean ( SEM of three experiments.
For comparison purpose the effect of PK 11195, Ro 5-4864, and
Clonazepam on steroid production at 40 µM concentration is also
included.
place of an aryl group at the 6-position, were synthe-
sized to verify whether the interaction about the L3 site

Pyrrolobenzoxazepines as PBR Ligands
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 18 3441
Ta ble 7. Geometries of Pharmacophore-Consistent Conformations
a
5d is substituted with a chlorine atom at the 4′-position.
ed system of aromatic residues and an electron-rich site
of the PBR. The lower potency of 17d with respect to
17c might result, to a lesser extent, from a nonoptimal
steric complementarity between the relatively rigid
cyclohexane moiety and the receptor.
P yr r oloben zoxa zep in es 17c,e,f, 19, 20, a n d 22:
Effect of Mod ifica tion s a t P osition 7 on th e Hy-
d r ogen Bon d in g w ith th e Recep tor Bin d in g Site
(H1). In a recent paper40 we have proposed the relative
3D location of a hypothetical receptor protic function
(H1 site) donating a vital hydrogen bond to a carbonyl
or sulfonyl oxygen of the 7-substituent (see Chart 2).
This H1 locus was placed at about 2 Å from the acceptor
oxygen, 6.0-6.5 Å from the L1 site and 7.8-8.2 Å from
the L3 site. To assess the predictive ability of our 3D
pharmacophore model, we prepared compounds 19, 20,
and 22. While all the so far reported19,40,53 active
pyrrolobenzothiazepines and pyrrolobenzoxazepines fea-
ture a 7-OCOR or 7-OSO2R side chain, 19, 20, and 22
bear the 7-CO2Me, 7-COMe, and 7-CH2OCOMe groups,
respectively. These new 7-substituents were expected
to broaden the spectrum of steric and conformational
properties in our data set so as to permit a more
accurate definition of the 3D pharmacophoric require-
ments. Figure 3 shows the overlapping active confor-
mations of 19, 20, 22, and 17c.

3442 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 18
Campiani et al.
F igu r e 1. Stereopair picture of the active conformations of 17c (black) and 5d (blue) overlapped on the template Ro 5-4864 (red)
about the H1, L1, and L3 pharmacophoric points.
F igu r e 2. Stereopair picture of the conformations of 17d (black) and 5g (red) consistent with the pharmacophore model.
Electrostatic potentials (blue) are contoured at the -5 kcal/mol level.
F igu r e 3. Stereopair picture of the superimposed pharmacophore-consistent conformations of 17c (green), 19 (blue), 20 (red),
and 22 (black). The yellow contour highlights the extra-volume of 22 hypothesized to interfere sterically with the binding to the
PBR.
The H1-L1 and H1-L3 distances in 19 and 20 do
not exactly match those of 17c (see also Table 7).
However, an average distance deviation of about 0.7 Å
was considered still compatible with appreciable affin-
ity. The moderate potency of these two compounds
confirmed our predictions. In contrast, 22 fits exactly
the pharmacophore geometry. Subsequent testing of 22
showed that this compound exhibits only micromolar
affinity. Probably, although the pharmacophoric ele-
ments in 22 are correctly oriented in the space, a portion
of this ligand experiences steric repulsion within the
binding cavity. To test this hypothesis, molecular
volume manipulations were performed on the structures
of 22 and of the active ligands Ro 5-4864, 17c, 17e, 17f,
19, and 20 superimposed about the H1, L1, and L3
points. By subtracting the union volume of the above
listed active compounds from the volume of 22, we
generated the chickenwire contours depicted in Figure
3 (for sake of clarity, Ro 5-4864, 17e, and 17f are not
shown). The highlighted extra-volume, surrounding

Pyrrolobenzoxazepines as PBR Ligands
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 18 3443
F igu r e 4. Stereopair picture showing crystal structures of compounds containing the thiocarbamoyloxy fragment. Dashed lines
correspond to intermolecular hydrogen bonds accepted by the sp2-sulfur of the considered fragment.
Ta ble 8. Intermolecular Hydrogen Bond Geometries in Crystal
mostly the terminal methyl group of 22, corresponds to
Structures Featuring the OsC(dS)N Substructure
the locus hypothesized to interfere sterically with the
binding. The difference between the IC50 values of 19
and 20 can be understood in the light of the work of
Raevsky et al.51 who parametrized the hydrogen bonding
abilities for a large number of molecules. According to
the Raevsky’s scale of “proton acceptor free energy
factors” (Ca descriptor) the carbonyl oxygen of methyl
acetate is a worse hydrogen bond acceptor than that of
refcodea
d (Å)
θ (deg)
τ(N,C,S,H) (deg)
acetone (Ca values of 1.60 and 2.03, respectively).
Replacing the 7-COMe substituent of 20 with a 7-COOMe
group so as to give 19 might be considered analogous
to converting acetone into methyl acetate. Hence, the
lower affinity of 19 with respect to 20 would be the
consequence of a weaker hydrogen bond formed between
19 and the H1 receptor site. In the same compilation,
the large Ca value assigned to N,N-dimethylacetamide
(2.99) seems also consistent with the hypothesis that
the 7-(dimethylcarbamoyl)oxy is more effective than the
7-acetoxy group in accepting a hydrogen bond from the
same H1 receptor hydrogen (17c vs 17e and 17h vs 17j).
Of course, the use of the Ca descriptor in structure-
affinity relationships is limited to strictly congeneric
compounds and requires that other potentially relevant
factors (steric, electrostatic, hydrophobic, etc.) are prop-
erly taken into account. Following this concept, it is
difficult to explain the subnanomolar affinity of the
7-(N,N-diethylcarbamoyl)oxy derivative 17f as simply
due to a much stronger hydrogen bond interaction. In
fact, the Ca value of N,N-diethylacetamide (3.10) is only
0.11 units above that of N,N-dimethylacetamide while
17f is 8 times more potent than 17e. A possible reason
for the remarkably high potency of 17f is that at least
one of the distal methyl groups of the N,N-diethyl
moiety fits optimally a receptor lipophilic region not
probed by the ligands tested so far. For simplicity we
will designate such a hypothetical lipophilic region as
L4 site (see Chart 2). However, since the (CO)N-C
bonds in 17f are free to rotate, we are unable to
establish the precise relative 3D location of the L4 site.
bummeb
confamb
2.62
2.68
2.82
2.67
2.48
2.61
2.66
2.89
2.27
2.48
2.30
2.47
2.47
2.76
2.56
2.50
2.54
2.49
109.9
105.5
247.6
111.1
108.5
110.8
111.4
157.5
103.0
102.4
105.0
103.2
106.5
111.4
108.9
113.8
110.4
106.2
-5.41
10.07
-165.81
10.56
18.12
11.27
-8.51
40.15
-6.79
6.35
-4.41
10.84
20.56
45.87
-12.08
-2.28
-4.36
25.99
dixpab
dojvaz
dojvedc
fezrad
filhaj
gajroy
keplumc
kepmat
mescza
pinpilc
vovwei
a
Refcode is the reference code of the entry stored in the
Cambridge Structural Database (version 5.09); in all the entries
b
c
X ) N. The sulfur accepts two hydrogen bonds. More than one
molecule per cell.
found with intramolecular S‚‚‚H hydrogen bonds yield-
ing a mean distance of 2.57 Å. In the Tripos force
field,54 the sp2-S and H atoms begin to repell each other
at distances shorter than 2.94 Å. In one entry (confam)
the same sulfur is engaged in two hydrogen bonds.
Interestingly, the retrieved crystal structures show a
hydrogen bond directionality similar to that reported
by Murray-Rust and Glusker for >CdO‚‚‚H interac-
tions.55 Figure 4 shows the thirteen structures aligned
on the common OsC(dS)N< fragment with the S‚‚‚H
hydrogen bonds drawn as dashed lines.
P yr r oloben zoxa zep in e 17g: Effect of Rep la ce-
m en t of th e C-7 Ca r bon yl w ith a C-7 Th ioca r bon yl
Gr ou p . Compound 17g, formally derived by replace-
ment of the sp2-oxygen of 17f with a sp2-sulfur was
prepared to further elucidate the energetics of the
interactions at the H1 and L4 sites. The high binding
affinity of compound 17g raised the question of whether
the sulfur atom in compounds containing the OsC-
(dS)N< fragment can effectively accept hydrogen bonds.
To address this issue, we performed a substructure-
based search in the Cambridge Structural Database
(CSD).49 As summarized in Table 8, thirteen hits were
The above structural data confirm that the sp2-sulfur
atom of 17g can accept a hydrogen bond from the H1
receptor site. The particular behaviour of this sulfur
probably should be ascribed to a mesomeric effect played
the thioamidic nitrogen.
In the Raevsky’s compilation51 the Ca values of N,N-
dimethylthioacetamide and acetone are very close (1.91
and 2.03, respectively).56 Thus, the free energy involved
in the formation of a hydrogen bond with the receptor
should be roughly the same for 17g and 17c. However,
if this was the only relevant factor influencing the
affinity, the two compounds should be nearly equipotent.

3444 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 18
Campiani et al.
EtOAc. The organic layers were washed with 10% sodium
hydrogen carbonate solution and brine, dried, and concen-
trated. The residue was purified by column chromatography
(chloroform) to give 9 (0.3 g) as colorless prisms: IR (CHCl3)
The IC50 value of 17g (3.8 nM), falling between those
of 17c (12.1 nM) and 17f (0.7 nM), seems consistent with
the hypothesis that one of the N-ethyl groups of 17g,
as was previously postulated for 17f, contributes favor-
ably to the ligand-receptor binding through a direct
hydrophobic interaction.
3420 cm-1 1H NMR (CDCl3) δ 5.47 (s, 1 H), 6.37 (m, 2 H),
;
6.89-7.00 (m, 3 H), 7.20-7.34 (m, 2 H); 13C NMR (CDCl3) δ
109.9, 120.9, 121.4, 121.7, 124.5, 127.5, 129.4, 145.6.
p-Azid op h en yla cetic Acid (11a ). To a solution of 4-ami-
nophenylacetic acid 10 (1.5 g, 10 mmol) in 6 N HCl (20 mL)
and concentrated H2SO4 (4 mL), cooled at 0 °C, was added
sodium nitrite (0.81 g, 12 mmol). After stirring at 0 °C for 30
min, 60 mg of urea were added and the mixture was stirred
for an additional 15 min. A solution of sodium azide (1.3 g,
20 mmol) in water (4 mL) was slowly added, and the reaction
mixture was stirred at 0 °C for 5 h and then filtered. The
solid residue was washed with cold water, dried, and crystal-
lized to give 11a (1.32 g) as a pale yellow solid: IR (Nujol)
2121, 1695 cm-1; 1H NMR (DMSO-d6) δ 3.63 (s, 2 H), 7.00 (m,
2 H), 7.25 (m, 2 H).
Con clu sion s
In this paper we have reported the synthesis of a
novel class of PBR ligands based on a benzoxazepine
skeleton. These compounds showed high affinity and
specificity for rat brain PBR and compounds 17f and
17j were found to be the most potent PBR ligands to
have been identified to date. Some of the most potent
compounds were also found to inhibit both [3H]PK 11195
and [3H]Ro 5-4864 binding to the rat adrenal PBR with
high affinity (Ki values ∼ 1 nM). In order to elucidate
the usefulness of such compounds they were screened
against one of the proposed roles for the mitochondrial
PBR, namely, regulation of steroidogenesis. These
highest affinity compounds stimulated steroidogenesis
in a mouse Y-1 adrenocortical cell line model with
similar potencies and efficacies to each other and to PK
11195 and Ro 5-4864. In view of the high affinity and
selectivity of these pyrrolobenzoxazepines for PBR,
these compounds represent novel probes for the mito-
chondrial PBR. They may also be useful in probing the
rate-limiting intramembrane transport of cholesterol for
steroidogenesis.
(()-r-Br om o-p -a zid op h en yla cet ic Acid E t h yl E st er
(12a ). p-Azidophenylacetic acid 11a (0.65 g, 3.7 mmol) was
dissolved in anhydrous ethanol (4.6 mL) and a catalytic
amount of concentrated H2SO4 (30 µL) was added. The
reaction mixture was refluxed for 5 h and then concentrated
in vacuo. The residue was taken up in dichloromethane, and
the organic phase was washed with 10% Na2CO3 solution and
brine, dried, and evaporated. The oily residue was chromato-
graphed (30% EtOAc in hexanes) to give 0.42 g (56%) of ethyl
ester as a solid: mp 40-41 °C (hexanes); IR (CHCl3) 2221,
1
1720 cm-1; H NMR (CDCl3) δ 1.25 (t, 3 H, J ) 6.9 Hz), 3.59
(s, 2 H), 4.15 (q, 2 H, J ) 6.9 Hz), 7.0-7.25 (m, 4 H). Anal.
(C10H11N3O2) C, H, N.
A mixture of the above ester (0.42 g, 2.05 mmol) and
N-bromosuccinimide (0.36 g, 20.5 mmol) in carbon tetrachlo-
ride (9.5 mL) was refluxed 20 h. The succinimide was filtered
off, and the filtrate was concentrated. The residue was directly
purified by flash chromatography (toluene) to give the bromo
ester 12a (0.23 g) as a colorless oil: IR (neat) 2220, 1739 cm
-1; 1H NMR (CDCl3) δ 1.28 (t, 3 H, J ) 7.0 Hz), 4.23 (m, 2 H),
5.31 (s, 1 H), 7.00-7.55 (m, 4 H).
Exp er im en ta l Section
Melting points were determined using an Electrothermal
8103 apparatus and are uncorrected. IR spectra were taken
with Perkin-Elmer 398 and FT 1600 spectrophotometers. 1H-
NMR spectra were recorded on a Bruker 200 MHz spectrom-
eter with TMS as internal standard; the values of chemical
shifts (δ) are given in ppm and coupling constants (J ) in Hz.
All reactions were carried out in an argon atmosphere.
Progress of the reaction was monitored by TLC on silica gel
plates (Riedel-de-Haen, Art. 37341). Merck silica gel (Kieselgel
60) was used for chromatography (70-230 mesh) and flash
chromatography (230-400 mesh) columns. Extracts were
dried over MgSO4, and solvents were removed under reduced
pressure. Elemental analyses were performed on a Perkin-
Elmer 240C elemental analyzer, and the results are within
(0.4% of the theoretical values, unless otherwise noted. Yields
refer to purified products and are not optimized. Physical data
for compounds 8, 9, 11, 12, 14-17, and 19-22 are reported
in Table 1.
(()-r-Br om o-p-tolylacetic Acid Eth yl Ester (12b). Simi-
larly to the procedure as described for 12a , the p-tolylacetic
acid ethyl ester was prepared starting from p-tolylacetic acid
11b (3.0 g, 20 mmol). After chromatography (dichloromethane)
the ester (2.95 g, 83%) was obtained as a colorless oil: IR
1
(CHCl3) 1730 cm -1; H NMR (CDCl3) δ 1.23 (t, 3 H, J ) 7.0
Hz), 2.32 (s, 3 H), 3.56 (s, 2 H), 4.14 (q, 2 H, J ) 7.1 Hz), 7.0-
7.25 (m, 4 H). Anal. (C11H14O2) C, H, N.
The ethyl ester (1.0 g, 5.6 mmol) was used in the subsequent
bromination step to obtain 12b, as described for 12a . The
bromo ester 12b was obtained as a colorless oil: IR (neat) 1735
1
cm -1; H NMR (CDCl3) δ 1.27 (t, 3 H, J ) 7.5 Hz), 2.34 (s, 3
H), 4.23 (m, 2 H), 5.34 (s, 1 H), 7.14-7.55 (m, 4 H).
(()-r-[[2-(1H-P yr r ol-1-yl)ph en yl]oxy]-p-tolylacetic Acid
Eth yl Ester (14a ). Sodium hydride (90 mg, 3.9 mmol) was
added to a solution of 13 (0.41 g, 2.6 mmol) in anhydrous THF
(5 mL) at rt. The reaction mixture was stirred for 1 h at rt,
and then a solution of ethyl R-bromo-p-tolylacetate (12b) (1.0
g, 3.9 mmol) in anhydrous THF (5 mL) was added dropwise.
After 12 h at rt, the solvent was removed in vacuo, and the
residue was taken up in dichloromethane. The organic layers
were washed with brine, dried, and evaporated. The residue
6-Ch lor o-2-n itr op h en ol (8). The title compound has been
obtained, starting from 2-chlorophenol, following the procedure
reported in reference 41. 1H NMR (CDCl3) δ 6.97 (t, 1 H, J )
7.9 Hz), 7.71 (dd, 1 H, J ) 7.9, 2.0 Hz), 8.07 (t, 1 H, J ) 1.7
Hz), 11.05 (s, 1 H).
1-(3-Ch lor o-2-h yd r oxyp h en yl)p yr r ole (9). To a suspen-
sion of 8 (5.0 g, 20 mmol) in ethanol (350 mL) was added a
solution of sodium hydrosulfite (20.2 g, 116.3 mmol) in water
(93 mL) within 1 h. The resulting mixture was refluxed for 2
h and stirred at rt for 12 h. Then the mixture was filtered,
and the filtrate was concentrated. The residue was taken up
in EtOAc, dried, and evaporated. Removal of the solvent
afforded a residue which was chromatographed (chloroform)
to give the corresponding aniline (2.9 g) as a white solid:41 1H
NMR (CDCl3) δ 3.84 (br s, 2 H), 5.45 (br s, 1 H), 6.60-6.75
(m, 3 H). A pure sample was then tranformed in its hydro-
chloride salt by mean of 5% methanolic HCl and used in the
next step without further purification.
was purified by chromatography (toluene) to give 0.46 g of 14a
-1
as a colorless oil: IR (neat) 1734 cm
;
1H NMR (CDCl3) δ
1.16 (t, 3 H, J ) 7.0 Hz), 2.33 (s, 3 H), 4.13 (q, 2 H, J ) 7.0
Hz), 5.48 (s, 1 H), 6.31 (m, 2 H), 6.95-7.40 (m, 10 H).
(()-r-[[2-(1H -P yr r ol-1-yl)p h en yl]oxy]-p -a zid op h en y-
la cetic Acid Eth yl Ester (14b). Starting from 13 (0.37 g,
2.34 mmol), the title compound was obtained (reaction time
18 h) following the procedure described for 14a . After chro-
matography (toluene), 14b was obtained as a colorless oil: IR
1
(neat) 2121, 1751 cm-1; H NMR (CDCl3) δ 1.16 (t, 3 H, J )
7.0 Hz), 4.14 (q, 2 H, J ) 7.0 Hz), 5,47 (s, 1 H), 6.32 (m, 2 H),
The above hydrochloride salt (0.2 g, 1.1 mmol) was dissolved
in boiling water (2 mL) and successively were added sodium
acetate trihydrate (0.15 g), boiling glacial acetic acid (4 mL),
and 2,5-dimethoxytetrahydrofuran (0.11 mL). After refluxing
for 5 min the mixture was rapidly cooled and extracted with
6.95-7.44 (m, 10 H).
(()-r-[[6-Ch lor o-2-(1H-p yr r ol-1-yl)p h en yl]oxy]p h en y-
la cetic Acid Eth yl Ester (14c). Similarly to 14a , the ester
14c was prepared starting from 0.5 g (2.6 mmol) of 9 (reaction

Pyrrolobenzoxazepines as PBR Ligands
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 18 3445
time 12 h). 14c was obtained as colorless oil: IR (neat) 1730
7-Acet oxy-6-p h en ylp yr r olo[2,1-d ][1,5]b en zoxa zep in e
(17c). Starting from 16d (0.2 g, 0.73 mmol), the title com-
pound was obtained according to the procedure as for 17a and
1
cm-1; H NMR (CDCl3) δ 1.13 (t, 3 H, J ) 6.9 Hz), 4.07 (m, 2
H), 4.91 (s, 1 H), 6.32 (m, 2 H), 6.88 (m, 2 H), 6.90-7.30 (m, 8
H).
1
recrystallized as colorless prisms: IR (CHCl3) 1780 cm-1; H
NMR (CDCl3) δ 2.19 (s, 3 H), 6.40 (m, 1 H), 7.15-7.39 (m, 10
(()-r-[[2-(1H-P yr r ol-1-yl)ph en yl]oxy]-p-tolylacetic Acid
(15a ). The ester 14a (0.78 g, 2.3 mmol) was dissolved in 26
mL of EtOH/THF mixture (1:1), and 5% aqueous NaOH (22.4
mL) was slowly added. The reaction mixture was stirred at
rt for 3 h, concentrated, and acidified with 4 N HCl until pH
3-4. The suspension was extracted with EtOAc, and the
organic phase was washed with brine, dried, and concentrated.
The residue was crystallized to give the acid 15a (0.67 g) as
colorless prisms: IR (KBr) 1730 cm-1; 1H NMR (CDCl3) δ 2.32
(s, 3 H), 5.45 (s, 1 H), 6.33 (m, 2 H), 6.90-7.35 (m, 10 H).
(()-r-[[2-(1H -P yr r ol-1-yl)p h en yl]oxy]-p -a zid op h en y-
la cetic Acid (15b). Starting from 14b (0.6 g, 1.65 mmol),
the title compound was obtained (reaction time 3 h) following
the procedure as for 15a . After purification by flash chroma-
tography (5% MeOH in EtOAc), 15b was crystallized as
colorless prisms: IR (KBr) 2220, 1730 cm-1; 1H NMR (CDCl3)
δ 5.48 (s, 1 H), 6.30 (m, 2 H), 6.90-7.50 (m, 10 H).
H), 7.75 (m, 1 H).
7-[(Cycloh exylca r b on yl)oxy]-6-p h en ylp yr r olo[2,1-d ]-
[1,5]ben zoxa zep in e (17d ). Starting from 16d (0.3 g, 1.08
mmol), the title compound was obtained following the proce-
dure as for 17a , using cyclohexanecarbonyl chloride. 17d was
recrystallized as pale yellow prisms: IR (Nujol) 1750 cm-1
;
1H NMR (CDCl3) δ 1.11-1.98 (m, 10 H), 2.45 (m, 1 H), 6.37
(m, 1 H), 7.08-7.42 (m, 10 H), 7.56-7.75 (m, 1 H).
7-[(Dim et h ylca r b a m oyl)oxy]-6-p h en ylp yr r olo[2,1-d ]-
[1,5]ben zoxa zep in e (17e). Similarly to 17a , the oxazepine
17e was prepared starting from 0.2 g (0.73 mmol) of 16d
(reaction time 8 h at 30 °C), using dimethylcarbamoyl chloride.
17e was obtained as white prisms: IR (CHCl3) 1740 cm-1; 1H
NMR (CDCl3) δ 2.93 (s, 3 H), 3.09 (s, 3 H), 6.41 (m, 1 H),
7,10-7.42 (m, 10 H), 7.71 (m, 1 H).
7-[(Dieth ylcar bam oyl)oxy]-6-ph en ylpyr r olo[2,1-d][1,5]-
ben zoxa zep in e (17f). Similarly to 17a , the oxazepine 17f
was prepared starting from 0.3 g (1.32 mmol) of 16d (reaction
time 6 h at 30 °C), using diethylcarbamoyl chloride. 17f was
(()-r-[[6-Ch lor o-2-(1H-p yr r ol-1-yl)p h en yl]oxy]p h en y-
la cetic Acid (15c). Similarly to 15a the acid 15c was
prepared starting from 100 mg (0.3 mmol) of 14c (reaction time
3 h). 15c was obtained as colorless prisms: IR (neat) 3420,
obtained as white prisms: IR (CHCl3) 1726 cm-1 1H NMR
;
1
1710 cm -1; H NMR (CDCl3) δ 4.70 (s, 1 H), 5.39 (br s, 1 H),
(CDCl3) δ 1.02-1.29 (m, 6 H), 3.27-3.49 (m, 4 H), 6.40 (m, 1
H), 7.06-7.44 (m, 10 H), 7.78 (m, 1 H).
6.15 (m, 2 H), 6.74-7.26 (m, 10 H).
7-[(Dieth ylth ioca r ba m oyl)oxy]-6-p h en ylp yr r olo[2,1-d ]-
[1,5]ben zoxa zep in e (17g). Similarly to 17a , the oxazepine
17g was prepared starting from 0.14 g (0.51 mmol) of 16d
(reaction time 12 h at rt), using diethylthiocarbamoyl chloride.
(()-6-p -Tolylp yr r olo[2,1-d ][1,5]b en zoxa zep in -7(6H )-
on e (16a ). Phosphorus pentachloride (0.4 g, 1.92 mmol) was
added to a solution of acid 15a (0.58 g, 1.89 mmol) in dry 1,2-
dichloroethane (8.5 mL) within 20 min. The reaction mixture
was stirred at rt for 5 h and then was poured into crushed
ice, basified with 10% NaOH solution, and extracted with
chloroform. The organic layers were washed with brine, dried,
and evaporated. The residue was chromatographed (dichlo-
romethane and hexanes 2:1) and recrystallized to yield 300
mg of oxazepinone 16a as colorless prisms: IR (CHCl3) 1640
17g was obtained as pale pink prisms: IR (CHCl3) 1150 cm-1
;
1H NMR (CDCl3) δ 1.25 (m, 6 H), 3.80 (m, 4 H), 6.32 (m, 1 H),
6.40 (m, 1 H), 7.18-7.43 (m, 9 H), 7.79 (m, 1 H); 13C NMR
(CDCl3) δ 11.7, 13.7, 44.0, 48.2, 109.5, 110.6, 121.0, 122.0,
122.2, 125.8, 126.7, 126.9, 128.1, 128.3, 128.7, 133.2, 133.6,
135.9, 145.1, 151.8, 185.6; MS m/z 390 (M+, 20), 313, 281, 246
(100).
1
cm-1; H NMR (CDCl3) δ 2.31 (s, 3 H), 5.50 (s, 1 H), 6.48 (m,
1 H), 7.0-7.40 (m, 10 H).
7-Acet oxy-6-p -t olylp yr r olo[2,1-d ][1,5]b en zoxa zep in e
(17h ). Starting from 16a (0.2 g, 0.69 mmol) the title compound
was obtained as colorless prisms, following the procedure as
(()-6-(p-Azid op h en yl)p yr r olo[2,1-d ][1,5]ben zoxa zep in -
7(6H)-on e (16b). Similarly to 16a , the oxazepinone 16b was
prepared starting from 0.52 g (1.57 mmol) of 15b (reaction time
8 h at 30 °C). 16b was recrystallized as white prisms: IR
(CHCl3) 2107, 1648 cm-1; 1H NMR (CDCl3) δ 5.51 (s, 1 H), 6.51
(m, 1 H), 6.95-7.44 (m, 10 H).
1
for 17a : IR (CHCl3) 1760 cm-1; H NMR (CDCl3) δ 2.19 (s, 3
H), 2.37 (s, 3 H), 6.38 (m, 1 H), 7.14-7.39 (m, 9 H), 7.65 (m, 1
H).
7-[(Dim eth ylcar bam oyl)oxy]-6-p-tolylpyr r olo[2,1-d][1,5]-
ben zoxa zep in e (17i). Similarly to 17a , the oxazepine 17i
was prepared starting from 0.2 g (0.69 mmol) of 16a (reaction
time 12 h at rt), using dimethylcarbamoyl chloride. 17i was
(()-4-Ch lor o-6-p h en ylp yr r olo[2,1-d ][1,5]ben zoxa zep in -
7(6H)-on e (16c). Similarly to 16a , the oxazepinone 16c was
prepared starting from 0.74 g (2.38 mmol) of 15c (reaction time
10 h at rt). 16c was obtained as white prisms: IR (CHCl3)
1
recrystallized as colorless prisms: IR (CHCl3) 1730 cm-1; H
1
1640 cm -1; H NMR (CDCl3) δ 5.57 (s, 1 H), 6.50 (m, 1 H),
NMR (CDCl3) δ 2.37 (s, 3 H), 2.93 (s, 3 H), 3.09 (s, 3 H), 6.32
(m, 1 H), 6.39 (m, 1 H), 7.13-7.40 (m, 8 H), 7.67 (d, 1 H, J )
8.0 Hz).
7.09-7.44 (m, 10 H); 13C NMR (CDCl3) δ 90.5, 112.2, 120.6,
121.1, 125.9, 126.3, 128.1, 128.3, 128.6, 129.0, 129.6, 133.5,
135.5, 189.5.
7-[(Dieth ylca r ba m oyl)oxy]-6-p-tolylp yr r olo[2,1-d ][1,5]-
ben zoxa zep in e (17j). Similarly to 17a , the oxazepine 17j
was prepared starting from 0.32 g (1.1 mmol) of 16a (reaction
time 8 h at rt), using diethylcarbamoyl chloride and was
Gen er a l P r oced u r e for P r ep a r a t ion of Com p ou n d s
17a -p . This procedure is illustrated for the preparation of
7-acetoxy-6-ethylpyrrolo[2,1-d][1,5]benzoxazepine (17a ). To a
suspension of potassium hydride (0.13 g, 1.1 mmol, 35% in oil)
in anhydrous THF (1.5 mL) was added the ketone 16g (0.25
g, 1.1 mmol) dissolved in anhydrous THF (1 mL). The reaction
mixture was stirred at rt for 2 h, and then acetyl chloride (78
µL, 1.1 mmol) was slowly added. After stirring for 12 h at rt,
the solvent was removed in vacuo and the residue was taken
up in EtOAc. The organic layer was washed with brine, dried,
and concentrated. The residue was chromatographed (dichlo-
romethane and hexanes 1:1) to give 196 mg of 17a as a
1
recrystallized as colorless prisms: IR (CHCl3) 1722 cm-1; H
NMR (CDCl3) δ 1.08-1.52 (m, 6 H), 2.37 (s, 3 H), 3.25-3.48
(m, 4 H), 6.32 (m, 1 H), 6.39 (d, 1 H, J ) 1.9 Hz), 7.10-7.40
(m, 8 H), 7.65 (d, 1 H, J ) 8.25 Hz).
7 -Ac e t o x y -6 -( 2 -n a p h t h y l ) p y r r o l o [ 2 , 1 -d ] [ 1 , 5 ] -
ben zoxa zep in e (17k ). Starting from 0.15 g (0.46 mmol) of
16e, the title compound was obtained as colorless prisms,
following the procedure described for 17a : IR (Nujol) 1760
cm-1; 1H NMR (CDCl3) δ 1.83 (s, 3 H), 6.40 (m, 1 H), 6.86 (m,
1 H), 7.10-8.11 (m, 12 H).
1
colorless oil: IR (CHCl3) 1765 cm-1; H NMR (CDCl3) δ 1.21
(t, 3 H, J ) 7.2 Hz), 2.22 (s, 3 H), 2.36 (q, 2 H, J ) 7.2 Hz),
6.25 (m, 1 H), 6.33 (m, 1 H), 7.06 (m, 1 H), 7.15-7.34 (m, 4
H).
7-[(Dim et h ylca r b a m oyl)oxy]-6-(2-n a p h t h yl)p yr r olo-
[2,1-d ][1,5]ben zoxa zep in e (17l). Similarly to 17a , the ox-
azepine 17l was prepared starting from 90 mg (0.27 mmol) of
16e (reaction time 18 h at 30 °C), using dimethylcarbamoyl
chloride. 17l was obtained as colorless prisms: IR (CHCl3)
1725 cm-1; 1H NMR (CDCl3) δ 2.61 (s, 3 H), 2.71 (s, 3 H), 6.41
(m, 1 H), 6.88 (m, 1 H), 7.13-8.16 (m, 12 H).
7-[(Dim eth ylcar bam oyl)oxy]-6-(p-azidoph en yl)pyr r olo-
[2,1-d ][1,5]ben zoxa zep in e (17m ). Starting from 0.16 g (0.51
mmol) of 16b, the title compound was obtained as a colorless
tick oil, using dimethylcarbamoyl chloride, and following the
7-[(Dim eth ylcar bam oyl)oxy]-6-eth ylpyr r olo[2,1-d][1,5]-
ben zoxa zep in e (17b). Starting from 16g (0.25 g, 1.1 mmol)
the title compound was obtained according to the procedure
described for 17a , by mean of dimethylcarbamoyl chloride. 17b
was recrystallized as colorless prisms: IR (CHCl3) 1730 cm-1
;
1H NMR (CDCl3) δ 1.23 (t, 3 H, J ) 7.0 Hz), 2.40 (q, 2 H, J )
7.0 Hz), 2.93 (s, 3 H), 3.08 (s, 3 H), 6.24 (m, 1 H), 6.32 (m, 1
H), 7.05 (m, 1 H), 7.13-7.35 (m, 4 H).

3446 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 18
procedure described for 17a : IR (neat) 2220, 1725 cm-1 1H
NMR (CDCl3) δ 2.93 (s, 3 H), 3.11 (s, 3 H), 6.40 (m, 1 H), 7.00-
7.40 (m, 9H), 7.77 (m, 1 H).
4-Ch lor o-7-[(dim eth ylcar bam oyl)oxy]-6-ph en ylpyr r olo-
[2,1-d ][1,5]ben zoxa zep in e (17n ). Starting from 0.16 g (0.51
mmol) of 16c, the benzoxazepine 17n was obtained as colorless
prisms following the procedure as described for 17a , using
dimethylcarbamoyl chloride: IR (CHCl3) 1725 cm-1; 1H NMR
(CDCl3) δ 2.94 (s, 3 H), 3.04 (s, 3 H), 6.49 (m, 1 H), 7.05-7.40
(m, 9 H), 7.65-7.76 (m, 1 H); MS m/z 380 (M+, 20), 308, 280,
72 (100).
Campiani et al.
;
with brine, dried, and concentrated. The residue was chro-
matographed (hexanes and dichloromethane 1:1) and recrys-
tallized to give the ester 22 (14 mg) as colorless prisms: IR
(CHCl3) 1753 cm-1; 1H NMR (CDCl3) δ 2.05 (s, 3 H), 4.94 (s, 2
H), 6.42 (m, 1 H), 6.49 (m, 1 H), 7.05-7.60 (m, 10H); 13C NMR
(CDCl3) δ 21.0, 63.7, 110.5, 110.7, 114.7, 121.4, 122.1, 122.8,
125.7, 127.0, 128.3, 128.7, 129.1, 129.9, 133.6, 134.9, 151.7,
154.6, 170.9.
Mitoch on d r ia l Ben zod ia zep in e Recep tor Bin d in g As-
sa y. Male CRL:CD(SD)BR rats (Charles River Italia, Calco,
CO, Italy) weighing about 150 g, were used in this experiment.
The rats were housed in groups of five in plastic cages, kept
under standard conditions (rt 21 ( 1 °C, relative humidity 55
( 10%, 12-12 h light, dark cycle) and given tap water and
food ad libitum. They were decapitated unanesthetized, and
the brains were rapidly removed and dissected into anatomi-
cally recognizable areas. Cortices were homogenized in about
50 vol of ice-cold phosphate-buffered saline, 50 mM, pH 7.4,
using an Ultra Turrax TP 1810 (2 × 20 s) instrument and
centrifuged at 50000g for 10 min. The pellet was washed three
more times by resuspension in fresh buffer and centrifuged
as before. The last pellet was resuspended just before the
binding assay. For mitochondrial benzodiazepine binding,57
10 mg of original wet tissue weight was incubated with 1 nM
[3H]PK 11195 (specific activity 85.8 Ci/mmol; NEN) in 1 mL
final volume for 120 min at 4 °C in the presence of 8-12
increasing concentrations of drugs. Nonspecific binding was
determined using 1 µM PK 11195. Incubation was stopped
by rapid filtration under vacuum through glass fiber filters
(Printed Filtermat B, Wallac) which were then washed with
12 mL of ice cold buffer, using a Brandel M48 RP harvester.
Filters were put into sample bags with 25 mL of Betaplate
Scint (LKB) and counted in a 1204 BS Betaplate liquid
scintillation counter, with a counting efficiency of about 45%.
IC50s were determined by nonlinear58 fitting of binding inhibi-
tion curves, using the Allifit program running on an IBM AT
personal computer. Each point was the mean of triplicate
samples.
Ad r en a l Mitoch on d r ia l P r ep a r a tion a n d [3H]P K 11195
a n d [3H]Ro 5-4864 Bin d in g. For binding studies, mitochon-
dria were prepared from adrenals of Wistar rats (200-250 g)
killed by cervical dislocation. Adrenal glands were removed,
homogenized in buffer A (50 mM Tris/Cl, pH 7.4, containing
0.25 M sucrose and 1 mM EDTA) with a Teflon pestle in a
glass homogenizer. The homogenate was centrifuged at 600g
for 10 min, and the resulting supernatant was then centrifuged
at 10000g for 10 min. The mitochondrial pellet was then
suspended in buffer A and centrifuged at 10000g for a further
10 min. The resulting washed mitochondrial pellet was
resuspended in 50 mM Tris/Cl, pH 7.4, at a concentration of 2
mg protein/mL as determined by the method of Markwell et
al.,59 for use in binding assays. All procedures were performed
at 4 °C. Adrenal mitochondria (20 µg of protein) were
incubated with either [3H]PK 11195 (2 nM) or [3H]Ro 5-4864
(2 nM) (NEN), in 50 mM Tris/Cl, pH 7.4, with a range of
concentrations of the tested compounds (0.1 nM-1 µM) dis-
solved in this buffer containing 0.1% ethanol, in a total volume
of 0.5 mL at 4 °C. Total and nonspecific binding in each case
was determined in the absence and presence respectively of
unlabeled PK 11195 or Ro 5-4864 (1 µM). All samples were
incubated in triplicate for 60 min. Assays were terminated
by filtration through Whatman glass fiber filters (GF/B, 2.5
cm) using a Brandel cell harvester. Radioactivity trapped on
the filters was determined by liquid scintillation counting. The
rat adrenal mitochondria express a large number of [3H]PK
11195 and [3H]Ro 5-4864 binding sites (77 and 50 pmol/mg),
of high affinity (Kd of 1.5 and 2.2 nM, respectively). The IC50
and subsequent Ki values for each compound were generated
by the use of the computer programs EBDA and LIGAND.
Photoaffinity labeling of rat adrenal mitochondria with 17m
was performed as previously described.60
4-Acet oxy-5-p h en yln a p h t h o[2,3-b]p yr r olo[1,2-d ][1,4]-
oxa zep in e (17o). Starting from 0.1 g (0.31 mmol) of 16f, the
title compound was obtained as colorless prisms, following the
1
procedure described for 17a : IR (Nujol) 1755 cm-1; H NMR
(CDCl3) δ 2.20 (s, 3 H), 6.46 (m, 1 H), 7.26 (m, 1 H), 7.38-8.20
(m, 12 H).
4-[(Dim eth ylca r ba m oyl)oxy]-5-p h en yln a p h th o[2,3-b]-
p yr r olo[1,2-d ][1,4]oxa zep in e (17p ). Starting from 0.3 g
(0.92 mmol) of 16f, 17p was obtained as colorless prisms
according to the procedure described for 17a , using dimeth-
ylcarbamoyl chloride: IR (CHCl3) 1725 cm-1; 1H NMR (CDCl3)
δ 2.92 (s, 3 H), 3.09 (s, 3 H), 6.46 (m, 1 H), 7.25-7.84 (m, 13
H).
7-(Me t h o x y c a r b o n y l)-6-p h e n y lp y r r o lo [2,1-d ][1,5]-
ben zoxa zep in e (19). To a suspension of potassium hydride
(0.41 g, 3.5 mmol, 35% in oil) in anhydrous DME (5 mL) was
slowly added a solution of the benzoxazepinone 16d (1.0 g, 3.63
mmol) in anhydrous DME (2 mL) under argon. After stirring
for 30 min at rt the reaction mixture was cooled at -78 °C
and triflic anhydride (0.61 mL, 3.63 mmol) was added. After
1 h at -78 °C, the mixture was allowed to warm to rt. The
solvent was removed under vacuum, and the residue, consist-
ing of the O-triflate 18, was dissolved in freshly distilled DMF
(15 mL), and palladium diacetate (22 mg, 0.11 mmol), triph-
enylphosphine (58 mg, 0.21 mmol), and anhydrous MeOH (6.5
mL) were added. The reaction mixture was purged with
carbon monoxide and was heated at 50 °C for 2 h under a
carbon monoxide atmosphere. After cooling, the solvent was
removed under vacuum and the residue was purified by
chromatography (30% hexanes in dichloromethane) to give 19
(0.72 g) as pale yellow prisms: IR (CHCl3) 1723 cm-1; 1H NMR
(CDCl3) δ 3.64 (s, 3 H), 6.43 (m, 1 H), 6.53 (m, 1 H), 7.16-7.63
(m, 10 H); 13C NMR (CDCl3) δ 52.4, 111.0, 112.4, 115.4, 121.6,
122.1, 122.9, 126.0, 126.9, 127.1, 127.8, 128.2, 129.5, 133.3,
135.0, 151.4, 155.2, 168.1.
7-Acetyl-6-ph en ylpyr r olo[2,1-d][1,5]ben zoxazepin e (20).
To a solution of ester 19 (100 mg, 0.31 mmol) in anhydrous
DME (0.3 mL) cooled at -20 °C was added methyllithium (0.43
mL, 0.688 mmol, 1.6 M solution in diethyl ether). After
stirring for 4 h at 0 °C under argon, the reaction mixture was
quenched with water (1 mL) and then extracted with diethyl
ether. The organic layers were washed with brine, dried, and
concentrated. The residue was chromatographed (hexanes and
dichloromethane 1:1) to give the ketone 20 (55 mg) as a white
solid: IR (CHCl3) 1695 cm-1; 1H NMR (CDCl3) δ 2.12 (s, 3 H),
6.38 (m, 1 H), 7.13-7.59 (m, 11 H); 13C NMR (CDCl3) δ 31.7,
111.0, 112.3, 121.8, 122.0, 122.9, 123.7, 125.9, 127.1, 127.7,
128.5, 128.5, 129.4, 129.8, 133.3, 134.7, 151.5, 153.0, 202.6.
7-(H y d r o x y m e t h y l)-6-p h e n y lp y r r o lo [2,1-d ][1,5]-
ben zoxa zep in e (21). To a suspension of LAH (6.6 mg, 0.17
mmol) in anhydrous THF (1 mL) was added the ester 19 (50
mg, 0.17 mmol), and the mixture was stirred for 2 h at rt under
argon. After quenching with 10% ammonium chloride solution
(1 mL), the mixture was extracted with EtOAc, and the organic
layers were washed with brine, dried, and concentrated. The
residue was purified by chromatography (dichloromethane)
and recrystallized to afford 21 (44 mg) as colorless prisms: IR
(CHCl3) 3422 cm-1; 1H NMR (CDCl3) δ 1.63 (s, 1 H), 4.54 (s, 2
H), 6.46 (m, 1 H), 6.59 (m, 1 H), 7.05-7.67 (m, 10H).
7-(Ac e t o x y m e t h y l)-6-p h e n y lp y r r o lo [2,1-d ][1,5]-
ben zoxa zep in e (22). To a solution of alcohol 21 (15 mg, 0.05
mmol) and dry triethylamine (43 µL, 0.31 mmol) in anhydrous
THF (0.3 mL) cooled at 0 °C was added acetyl bromide (15.2
µL, 0.21 mmol). The reaction mixture was stirred at rt for 3
h under argon. After quenching with water (1 mL) the mixture
was extracted with EtOAc and the organic phase was washed
Ster oid Biosyn th esis. Y1 adrenocortical cells were ob-
tained from the European Collection of Animal Cell Culture
(Salisbury, UK). They were cultured in Dulbecco’s Modified
Eagles Medium supplemented with 2 mM glutamine, genta-
mycin (100 mg/mL), and 10%FCS. Cultures were maintained
in a humidified atmosphere of 5% CO2/95% air at 37 °C. Y1

Pyrrolobenzoxazepines as PBR Ligands
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 18 3447
cells were seeded in 24-well plates at a density of 3 × 105 cells/
well in a final volume of 1 mL. Prior to measurement of
pregnenolone production, the cells were washed three times
with a simple salts medium consisting of 140 mM NaCl, 5 mM
KCl, 1.8 mM CaCl2, 1 mM MgSO4, 10 mM glucose, 10 mM
HEPES/NaOH, pH 7.4, plus 0.1% BSA. During experiments,
cells were incubated with this simple salts medium in an air
incubator at 37 °C. In order to measure pregnenolone secreted
into the medium, its further metabolism was blocked by the
addition of trilostane (5 µM) and SU 10603 (20 µM) (inhibitors
of 3â-hydroxysteroid dehydrogenase and 17R-hydroxylase,
respectively) to the simple salts medium. The addition of the
tested compounds, PK 11195, Ro 5-4864, or clonazepam to the
Y1 cells was made by the complete change of the simple salts
medium to medium containing the appropriate concentration
(40 µM) of compound. The final concentration of ethanol was
constant for all the wells within each experiment and did not
exceed 0.5% (v/v), a concentration which on its own had no
effect on steroid production. At the end of the incubation
period (2 h) the cell medium was saved and centrifuged at
1500g for 10 min. The amount of pregnenolone secreted into
the medium was quantified by radioimmunoassay (RIA) using
an antibody obtained from ICN Biochemical Inc. (California),
under the conditions described by the manufacturer. Analysis
of the RIA data was performed using the Apple Macintosh
ASSAYZAP program obtained from Biosoft.
Com p u ta tion a l Ch em istr y. All molecular modeling was
performed with use of the software package SYBYL61 running
on a Silicon Graphics Indigo XS24 workstation. Geometry
optimizations were realized with the semiempirical quantum-
mechanics methods MNDO,62 AM1,63 and PM364 available in
the MOPAC program65 as well as with the molecular-mechan-
ics SYBYL/Tripos force field.54 MOPAC was run using default
settings and the keyword “MMOK” for compounds containing
an amide bond. Tripos energy minimizations were carried out
under conditions detailed in a previous article.40
Conformational analyses were carried out on PM3-optimized
structures of pyrrolobenzoxazepines to identify global mini-
mum conformers and receptor-recognized conformations. The
difference in energy between the two types of conformations,
calculated with the Tripos force field, provided an estimate of
the strain energy introduced in the binding to the receptor.
Whenever this energy exceeded the threshold value of 5 kcal/
mol the conformation was rejected.
Ta ble 9. Geometric Descriptors of Crystallographic and
Calculated Structures of Derivatives of Oxepine, 1,3-Oxazepine,
and 1,4-Oxazepine
stern bow
C-O(S) C-O(S)- angle angle
refcodea
cofpoc10
bebsoq
cufjiw10
donvor
foltop
jasgal
peplol
sagdot
sixpuk
tpboxz
cofpoc10 (Tripos)
cofpoc10 (MNDO)
cofpoc10 (PM3)
cofpoc10 (AM1)
17c (PM3)
17e (PM3)
(Å)b
C (deg) (deg)c (deg)d rmsd (Å)e
1.41
1.43
1.40
1.40
1.39
1.41
1.39
1.38
1.38
1.42
1.34
1.38
1.40
1.40
1.40
1.40
108.94
111.14
105.10
109.92
104.72
105.31
105.62
115.99
110.80
106.88
136.60
120.49
111.41
112.66
113.52
113.38
60.88 24.42
62.84 28.56
56.84 40.21
56.74 25.92
57.49 39.27
59.00 35.54
57.10 36.03
51.56 22.85
45.84 27.22
59.59 30.99
0.000
0.057
0.098
0.022
0.101
0.079
0.077
0.055
0.088
0.061
0.292
0.104
0.031
0.038
0.053
0.050
2.13
0.28
45.52 18.00
55.62 24.33
55.30 22.49
54.94 28.53
53.65 28.59
a
Refcode, except for 17c and 17e, is the reference code of the
entry retrieved from the Cambridge Structural Database (version
5.09); for theoretically derived structures, the geometry-optimiza-
b
tion method is reported in parentheses. Mean of two distances.
c
Stern angle is the dihedral angle formed by the plane passing
through atoms 2,3,6,7 and the plane passing through atoms 1,2,7
d
of the seven-membered ring. Bow angle is the dihedral angle
formed by the plane passing through atoms 2,3,6,7 and the plane
e
passing through atoms 3,4,5,6 of the seven-membered ring. The
rmsd is the root mean square distance resulting from the fitting
of each structure on cofpoc10 about the non-hydrogen atoms of
the seven-membered ring.
its structural simplicity and high resolution (R ) 0.056). Our
calculations showed that the PM3 method is the best param-
etrized for reproducing the geometries of the oxepine ring (see
results in Table 9) and hence, we inferred, the most reliable
for generating molecular models of pyrrolobenzoxazepines.
This inference is supported by the observation that the recently
reported40 crystal structures of the pyrrolobenzothiazepine
derivatives 5c, 5d , and 5g (see Table 4) and of 2,7-di-t-
butylthiaphenine (reference code bikset in the CSD) could be
satisfactorily overlapped about the non-hydrogen atoms of
their sulfurated seven-membered rings with rmsd values
within 0.102 Å. In other words, the hypothesis that oxepine
and 1,4-oxazepine rings are geometrically similar relies on the
observation that the corresponding isosteric thiaphenine and
1,4-thiazepine rings are indeed similar. Incidentally, the PM3
method performed well in reproducing the geometry of the
pyrrolobenzothiazepine system of 5c, 5d , and 5g. The fitting
of experimental and PM3 geometries about the fourteen non-
hydrogen atoms of the tricyclic nucleus gave rmsd values
within 0.091 Å.
Mod elin g of 17c a n d 17e. As a first step, conformers of
17c and 17e were constructed by PM3 geometry-optimization
of input coordinates derived from the crystal structures of the
benzothiazepines 5b and 5c.40 The oxazepine rings of the
resulting output structures were found to be nicely superim-
posable with the oxepine ring of cofpoc10 (see data in Table
9). However, a comparison of the experimental and PM3
optimized structures of 5c revealed that this semiempirical
method did not adequately reproduce the geometry of the
7-OCON(Me)2 substituent. Specifically, in the PM3 geometry
the amidic nitrogen is 0.28 Å removed from the plane defined
by the three neighboring carbons whereas in the crystal
structure this distance is only 0.03 Å. Thus, we decided to
mutate 5c into 17e by a PM3 partial geometry optimization
The April 1995 release (3D graphics 5.09 version for UNIX
platforms) of the Cambridge Structural Database (CSD)49 was
searched through substructure queries (metallorganics were
not considered). Hits from these searches were used to
compare experimental versus theoretical geometries of specific
fragments.
Molecules were superimposed by minimizing the root mean
square distance (rmsd) relative to selected atom pairs using
the SYBYL/FIT command. Three pseudoatoms (H1, L1, and
L3) were positioned on the structures of the modeled com-
pounds: L1 and L3 are, respectively, centroids of the fused
benzene and the pendant phenyl ring whereas H1 is located 2
Å away from the carbonyl oxygen in the plane of the >CdO
system. Generation of the structure of Ro 5-4864 is reported
in our previous study.40
Ch oice of th e Geom etr y-Op tim iza tion Meth od for
P yr r oloben zoxa zep in e Der iva tives. A substructure search
in the CSD was performed to detect compounds featuring an
oxepine ring or any corresponding isosteric ring resulting from
CH/N replacements. For this purpose we used the CSD/
QUEST graphical routine. Table 9 lists the resulting eleven
hits’ reference codes and some geometric parameters of the
seven-membered rings. Only the tbpoxz entry is a 1,3-
oxazepine derivative whereas the remaining compounds con-
tain an oxepine nucleus. All the retrieved crystal structures
show a high degree of similarity about the oxygenated cycle.
This latter assumes a boat conformation with stern and bow
angles falling within intervals of 46-63° and 24-40°, respec-
tively (stern and bow angles are defined in the legend of Table
9).
The Tripos, MNDO, AM1, and PM3 methods were applied
to energy-minimize the input crystal structure of cofpoc10 (2,7-
diphenyloxepine), selected among the entries in Table 9 for

3448 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 18
Campiani et al.
which left invariant the geometry of the 7-OCON(Me)2 group.
The obtained structures of 17c and 17e were assumed to be
reasonably close to their global minimum conformers. To find
the receptor-recognized conformations of 17c and 17e, sys-
tematic conformational searches were carried out using the
SYBYL/SEARCH routine. Starting from the PM3 global
minimum conformer, rotatable bonds about the 7-substituent
and the CdO‚‚‚H1 angle were scanned according to a proce-
dure described in ref 40. Distances H1-L1 and H1-L3 and
Tripos energies were recorded in each search run (the L1-L3
distance being invariant). Conformations whose energy was
above 5 kcal/mol from the global minimum were rejected. The
selected conformations were those which best matched the
pharmacophore distances in Ro 5-4864. Both 17c and 17e
were superimposed on Ro 5-4864 about the H1, L1, and L3
points. The active conformations of Ro 5-4864, 17c and 5d
are described in Table 7 and shown in Figure 1. Compound
5d (see Table 4), whose active conformation was directly
available from our previous study,40 was selected as represen-
tative of the pyrrolobenzothiazepine series.
successively fully optimized with PM3 to yield the hypothetical
global minimum conformers of 19, 20, and 22.
Each of the three PM3 global minima was aligned on the
potent analogue 17c about the non-hydrogen atoms of the
common pyrrolobenzoxazepine system and then submitted to
a systematic conformational search using the SYBYL/SEARCH
routine. In this procedure, the above listed torsion angles and
the CdO‚‚‚H1 angle were scanned through 5° and 10° steps.
For each compound we selected a conformation within 5 kcal/
mol from the global minimum in which the intermolecular
distance between the H1 of the analyzed molecule and the H1
of 17c was lowest. Such
a procedure ensured that the
putatively active conformations of 19, 20, and 22 could overlap
not only about the H1, L1, and L3 points of the template Ro
5-4864 but also about the pyrrolobenzoxazepine skeleton of
17c and other similar analogues. Results are reported in Table
7 and Figure 3. Molecular volume manipulations were per-
formed using the SYBYL/MVOLUME option to highlight the
“extra-volume” of the inactive compound 22 with respect to
the active compounds Ro 5-4864, 17c, 17e, 17d , 19, and 20
(see Figure 3).
Sea r ch in th e CSD for Hyd r ogen Bon d s In volvin g th e
Su lfu r of a Th ioca r ba m oyloxy Moiety. A substructure-
based search in the CSD was performed to investigate the
hydrogen bond capability of the sp2-sulfur in compounds
containing the OsC(dS)N< fragment. The graphical interface
of the CSD/QUEST routine was employed for this purpose.
Table 8 lists the 3D query based on the following geometric
constraints: S‚‚‚H distance comprised between 1.5 and 3.0 Å,
CdS‚‚‚H angle comprised between 90° and 270°. Results are
listed in Table 8 and shown in Figure 4.
Mod elin g of 17d . A trial model of 17d was built by
replacing the methyl group of the 17c PM3 global minimum
with a cyclohexyl group in a chair conformation. This moiety
was retrieved from the SYBYL fragment library.
Two separate conformational analyses on 17d were carried
out with SYBYL/GRIDSEARCH depending on whether the
carboxy group was equatorial or axial with respect to the
cyclohexane ring. This latter was always in the chair confor-
mation. The rotatable bonds C7-O1′′, C2′′-C3′′, and C6-C1′
(see labels in Table 7) were scanned with 30° steps. The
intervals of variation were 0-330° for the first two bonds and
0-150° for the latter bond. In all the minimizations the
torsional angle C7-O1′′-C2′′-C3′′ started from a value of
180°. The lowest energy conformation from each search run
was energy-minimized with PM3. According to the PM3
calculations, the “equatorial” conformer was about 2 kcal/mol
more stable than the “axial” conformer. This result is in
agreement with searches performed in the CSD which showed
that several crystal structures of carboxycyclohexane deriva-
tives display equatorial arrangements.
We propose for 17d an active conformation, obtained by
manual adjustment of the C7-O1′′ and C2′′-C3′′ rotatable
bonds, in which the 7-carboxy moiety overlaps that of 17c and
the cyclohexane ring is oriented equatorially. In such a
geometry, the pharmacophoric distances in 17d are practically
coincident with those of 17c and 17e, and the cyclohexane
shares much of the volume occupied by the 3′-pyridine ring of
5g. The pharmacophore-consistent conformations of 17d and
5g are described in Table 7 and shown in Figure 2.
The hypothesized active conformation of 17d is consistent
with modeling studies recently conducted on a potent pyr-
rolobenzothiazepine analogue bearing a 7-(morpholin-4′-yl)-
carboxy group.53 Owing to the amidic conjugation, in this
compound the morpholine moiety is somewhat coplanar with
the carbonyl system and thus oriented in the binding site
similarly to the cyclohexane ring of 17d and the 3’-pyridine
ring of 5g.
Molecular electrostatic potentials for 17d and 5g, based on
PM3 partial atomic charges (single point calculations), were
contoured at -5 kcal/mol through the SYBYL/POTENTIAL
routine (see Figure 2).
Ack n ow led gm en t. This work was supported by
grant from MURST (60%) and CNR, Roma.
Refer en ces
(1) Part of this work has been presented in a preliminary form at
the “II Congresso Congiunto Italiano-Spagnolo di Chimica
Farmaceutica”, (ML19), Aug 30-Sept 3, 1995, Ferrara, Italy.
(2) Braestrup, C.; Squiref, R. F. Specific Benzodiazepine Receptors
in Rat Brain Characterized by High-Affinity [3H]Diazepam
Binding. Proc. Natl. Acad. Sci. U.S.A. 1977, 74, 3805-3809.
(3) Regan, J . W.; Yamamura, H. I.; Yamada, S.; Roeske, W. R. High
Affinity Renal [3H]Flunitrazepam Binding: Characterization,
Localization, and Alteration in Hypertension. Life Sci. 1981, 28,
991-997.
(4) Schoemaker, H.; Bliss, M.; Yamamura, H. I. Specific High
Affinity Saturable Binding of [3H]Ro 5-4864 to Benzodiazepine
Binding Site in Rat Cerebral Cortex. Eur. J . Pharmacol. 1981,
71, 473-475.
(5) Costa, E.; Guidotti, A. Molecular Mechanisms in the Receptor
Actions of Benzodiazepines. Annu. Rev. Pharmacol. Toxicol.
1979, 19, 531-545.
(6) Knapp, R. J .; Malatynska, E.; Yamamura, H. I. From Binding
Studies to the Molecular Biology of GABA Receptors. Neurochem.
Res. 1990, 15, 105-112.
(7) Marangos, P. J .; Pate, J .; Boulanger, J . P.; Clark-Rosenberg, R.
Characterization of Peripheral-type Benzodiazepine Binding Site
in Brain Using [3H]Ro 5-4864. Mol. Pharmacol. 1982, 22, 26-
32.
(8) Benavides, J .; Quartaronet, D.; Imbault, F.; Malgouris, C.; Uzan,
A.; Renault, C.; Dubroeucq, M. C.; Gueremy, C.; Le Fur, G.
Labeling of “Peripheral-type” Benzodiazepine Binding Site in the
Rat Brain by Using [3H]PK 11195, an Isoquinoline Carboxamide
Derivative: Kinetic Studies and Autoradiographic Localization.
J . Neurochem. 1983, 41, 1744-1750.
(9) Schoemaker, H.; Boles, R. G.; Horst, W. D.; Yamamura, H. I.
Specific High Affinity Binding Site for [3H]Ro 5-4864 in Rat
Brain and Kidney. J . Pharmacol. Exp. Ther. 1983, 225, 61-69.
(10) Saano, V. Affinity of Various for Benzodiazepine Binding Sites
in Rat Brain, Heart, and Kidneys in Vitro. Acta Pharmacol.
Toxicol. 1986, 58, 333-338.
Mod elin g of 19, 20, a n d 22. Starting molecular models
for 19, 20, and 22 were built by appropriate modifications of
the 17c PM3 global minimum using Tripos standard bond
lengths and bond angles. Systematic conformational searches
aimed at identifying global minimum energy conformers for
19, 20, and 22 were performed using the SYBYL/GRID-
SEARCH routine. Energy minimizations were conducted with
the Tripos force field. The rotatable bonds C7-C1′′, C1′-O2′′,
and C6-C1′ (see labels in Table 7) were scanned with 30°
increments within 0-330° for the first two bonds and within
0-150° for the latter bond. The CO2Me side chain in 19 was
kept fixed with the carbonyl oxygen and the methyl group in
a cis-like arrangement. Similarly, the methylene fragment
and the carbonyl oxygen in 22 were always in a cis-disposition.
The lowest energy conformation from each search run was
(11) Davies, L. P.; Huston, V. Peripheral Benzodiazepine Binding
Sites in Heart and Their Interaction with Dipyridamole. Eur.
J . Pharmacol. 1981, 73, 209-211.
(12) Le Fur, G.; Guillox, F.; Rufat, P.; Benavides, J .; Uzan, A.;
Renault, C.; Dubroeucq, M. C.; Gueremy, C. Peripheral Benzo-
diazepine Binding Sites: Effect of PK 11195, 1-(2-Chlorophenyl)-
N-methyl-(1-methylpropyl)-3-isoquinoline Carboxamide. II. In
Vivo Studies. Life Sci. 1983, 32, 1849-1856.
(13) Kendall, D. A.; Nahorsky, S. R. Dihydropyridine Calcium Chan-
nels Activators and Antagonists Influence Depolarization-
Evoked Inositol Phospholipid Hydrolysis in Brain. Eur. J .
Pharmacol. 1985, 115, 31-36.

Pyrrolobenzoxazepines as PBR Ligands
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 18 3449
(14) Olson, J . M.; Ciliax, B. J .; Mancini, W. R.; Young, A. B. Presence
of Peripheral-Type Benzodiazepine Binding Sites on Human
Erythrocyte Membranes. Eur. J . Pharmacol. 1988, 142, 47-53.
(15) De Souza, E. B.; Anholt, R. R. H.; Murphy, K. M.; Snyder, S.
M.; Kuhar, M. J . Peripheral-Type Benzodiazepine Receptors in
Endocrine Organs: Autoradiographic Localization in Rat Pitu-
itary, Adrenal, and Testis. Endocrinology 1985, 116, 567-573.
(16) Le Fur, G.; Perrier, J .; Uzan, A.; Renault, C.; Dubroeucq, M. C.;
Gueremy, C. Peripheral Benzodiazepine Binding Sites: Effect
of PK 11195, 1-(2-Chlorophenyl)-N-methyl-(1-methylpropyl)-3-
isoquinoline Carboxamide. I. In Vitro Studies. Life Sci. 1983,
32, 1839-1847.
(17) Almirante, L.; Mugnaini, A.; Rugarli, P.; Gamba, A.; Zefilippo,
E.; De Toma, N.; Murmann, W. Derivatives of Imidazole. III.
Synthesis and Pharmacological Activities of Nitriles, Amides,
and Carboxylic Acid Derivatives of Imidazo[1,2-a]pyridine. J .
Med. Chem. 1969, 12, 122-126.
(31) McEnery, M. W.; Snowman, A. N.; Trifletti, R. R.; Snyder, S. H.
Isolation of the Mitochondrial Benzodiazepine Receptor Associ-
ated with the Voltage Dependent Anion Channel and the
Adenine Nucleotide Carrier. Proc. Natl. Acad. Sci. U.S.A. 1992,
89, 3170-3174.
(32) McCauley, L. D.; Park, C. H.; Lan, M. C.; Tonich, J . M.; Shively,
J . E.; Gee, K.W. Benzodiazepines and Peptides Stimulate
Pregnenolone Synthesis in Brain Mitochondria. Eur. J . Phar-
macol. 1995, 276, 145-143.
(33) Mukhin, A. G.; Papadopoulos, V.; Costa, E.; Krueger, K. E.
Mitochondrial Benzodiazepine Receptors Regulate Steroid Bio-
synthesis. Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 9813-9816.
(34) Besman, M. J .; Yanagibashi, K.; Lee, T. D.; Kawamura, M.; Hall,
P. F.; Shively, J . E. Identification of des-(Gly-Ile)-Endozepine
as an Effector of Corticotropin-Dependent Adrenal Steroido-
genesis: Stimulation of Cholesterol Delivery is Mediated by the
Peripheral Benzodiazepine Receptors. Proc. Natl. Acad. Sci.
U.S.A. 1989, 86, 4897-4901.
(35) Krueger, K. E.; Papadopoulos, V. Peripheral-Type Benzodiaz-
epine Receptors Mediate Translocation of Cholesterol from the
Outer to the Inner Membranes in Adrenocortical Cells. J . Biol.
Chem. 1990, 265, 15015-15022.
(18) 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.
(19) (a) Fiorini, I.; Nacci, V.; Ciani, S. M.; Garofalo, A.; Campiani,
G.; Savini, L.; Novellino, E.; Greco, G.; Bernasconi, P.; Mennini,
T. Novel Ligands for Mitochondrial Benzodiazepine Receptors:
6-Arylpyrrolo[2,1-d][1,5]benzothiazepine Derivatives. Synthesis,
Structure-Activity Relationships, and Molecular Modeling Stud-
ies. J . Med. Chem. 1994, 37, 1427-1438. (b) Greco, G.; Novellino,
E.; Fiorini, I.; Nacci, V.; Campiani, G.; Ciani, S. M.; Garofalo,
A.; Bernasconi, P.; Mennini, T. A Comparative Molecular Field
Analysis Model for 6-Arylpyrrolo[2,1-d][1,5]benzothiazepines
Binding Selectively to the Mitochondrial Benzodiazepine Recep-
tor. J . Med. Chem. 1994, 37, 4100-4108.
(36) Purdy, R. H.; Moore, P. H.; Morrow, A. L.; Paul, S. M. The 3a-
Hydroxyring A-Reduced Metabolities of Progesterone and
Deoxycorticosterone: Natural Ligands of GABAA Receptors. In
Neurosteroids and Brain Function; Costa, E., Paul, S. M., Eds.;
Thieme Medical Publishers: New York, 1991; Vol. 8, pp 95-
101.
(37) Majewska, M. D. Neurosteroids: GABA-Agonists and GABA-
Antagonists Modulators of the GABAA Receptors. In Neuroster-
oids and Brain Function; Costa, E., Paul, S. M., Eds.; Thieme
Medical Publishers: New York, 1991; Vol. 8, pp 109-117.
(38) Wu, F.; Gibbs, T. T.; Farb, D. H. Pregnenolone Sulfate:
A
(20) (a) Anholt, R. R. H.; Pedersen, P. L.; De Souza, E. B.; Snyder, S.
H. The Peripheral-Type Benzodiazepine Receptor: Localization
to the Mitochondrial Outer Membrane. J . Biol. Chem. 1986, 261,
576-583. (b) Basile, A. S.; Skolnick, P. Subcellular Localization
of “Peripheral-Type” Binding Sites for Benzodiazepines in Rat
Brain. J . Neurochem. 1986, 46, 305-308. (c) Hirsch, J . D.; Beyer,
C. F.; Malkowitz, L.; Louillis, C. C.; Blume, A. J . Characteriza-
tion of Ligand Binding to Mitochondrial Benzodiazepine Recep-
tors. Mol. Pharmacol. 1989, 35, 164-172.
(21) (a) Mestre, M.; Carriot, T.; Belin, C.; Uzan, A.; Renault, C.;
Dubroeucq, M.C.; Gueremy, C.; Le Fur, G. Electrophysiological
and Pharmacological Characterization of Peripheral Benzodi-
azepine Receptors in Guinea Pig Heart Preparation. Life Sci.
1984, 35, 953-962. (b) Grupp, I. L.; French, J . F.; Matlib, M. A.
Benzodiazepine RO 5-4864 Increases Coronary Flow. Eur. J .
Pharmacol. 1987, 143, 143-147.
(22) (a) Newman, A. H.; Hogue, B. A.; Basile, A. S.; Hansford, R. G.;
Chiang, P. K.; Moreno-Sanchez, R. Inhibition of Oxidative
Phosphorylation by Peripheral-Type Benzodiazepines. FASEB
J . 1989, 3, A703. (b) Moreno-Sanchez, R.; Hogue, B. A.; Bravo,
C.; Newman, A. H.; Basile, A.S.; Chiang, P. K. Inhibition of
Substrate Oxidation in Mitochondria by Peripheral-Type Ben-
zodiazepine Receptor Ligand AHN 086. Biochem. Pharmacol.
1990, 41, 1479-1484. (c) Larcher, J .-C.; Vayssiere, J .-L.; LeMar-
quer, F. J .; Cordeau, L. R.; Keane, P. E.; Bachy, A.; Grus, F.;
Croizat, B. P. Effect of Peripheral Benzodiazepines Upon the
Positive Allosteric Modulator at the N-Methyl-D-aspartate
Receptor. Mol. Pharmacol. 1991, 40, 333-336.
(39) Romeo, E.; Auta, J .; Kozikowski, A. P.; Ma, D.; Papadopoulos,
V.; Puia, G.; Costa, E.; Guidotti, A. 2-Aryl-3-indoloacetamides
(FGIN-1) a New Class of Potent and Specific Ligands for the
Mitochondrial DBI Receptor. J . Pharmacol. Exp. Ther. 1993,
262, 971-978.
(40) Dalpiaz, A.; Bertolasi, V.; Borea, P. A.; Nacci, V.; Fiorini, I.;
Campiani, G.; Mennini, T.; Manzoni, C.; Novellino, E.; Greco,
G. A Concerted Study Using Binding Measurements, X-ray
Structural Data, and Molecular Modeling on the Stereochemical
Features Responsible for the Affinity of 6-Arylpyrrolo[2,1-d][1,5]-
benzothiazepines toward Mitochondrial Benzodiazepine Recep-
tors. J . Med. Chem. 1995, 38, 4730-4738.
(41) Campiani, G.; Nacci, V.; Fiorini, I.; De Filippis, M. P.; Garofalo,
A.; Greco, G.; Novellino, E.; Altamura, S.; Di Renzo, L. Pyr-
rolobenzothiazepinones and Pyrrolobenzoxazepinones: Novel
and Specific Non-Nucleoside HIV-1 Reverse Transcriptase In-
hibitors with Antiviral Activity. J . Med. Chem. 1996, 39, 2672-
2680.
(42) Katz, L.; Cohen, M. S. Benzoxazole Derivatives. I. 2-Mercapto-
benzoxazoles. J . Org. Chem. 1954, 19, 758-766.
(43) Artico, M.; Porretta, G. C.; De Martino, G. Synthesis of 4H-
Pyrrolo[2,1-c][1,4]benzoxazine. J . Heterocycl. Chem. 1971, 8,
283-287.
(44) Ethyl p-Tolyl-R-bromoacetate (12b) has been already prepared
following a different synthetic route than that described in eq
2. See (a) Hoch, J .; Choisy, J . M. Action of Sodamide on
R-Bromoarylacetates. Compt. Rend. 1959, 248, 3314-3316. (b)
Salageanu, N.; Sileanu, I. Action of Some Chemical Substances
on the Growth of the Roots of Beans. An. univ. Bucuresti, Ser.
Stiint. nat. 1956, 135-141. (c) For the preparation of 12c see
reference 19a.
(45) Cacchi, S.; Morera, E.; Ortar, G. Palladium-Catalyzed Carbo-
nylation of Enol Triflates. A Novel Method for One-Carbon
Homologation of Ketones to R,â-Unsaturated Carboxylic Acids
Derivatives. Tetrahedron Lett. 1985, 26, 1109-1112.
(46) Meinwald, J .; J ensen, C. B.; Lewis, A.; Swithenbank, C. Highly
Strained Bicyclic Systems. X. The Chemistry of 6-Substituted
exo-5-Chlorobicyclo[2.1.1]hexanes. J . Org. Chem. 1964, 29,
3469-3475.
(47) Papadopoulos, V.; Mukhin, A. G.; Costa, E.; Krueger, K. E. The
Peripheral-Type Benzodiazepine Receptor is Functionally Linked
to Leydig Cell Steroidogenesis. J . Biol. Chem. 1990, 265, 3772-
3779.
(48) Papadopoulos, V.; Guarneri, P.; Krueger, K. E.; Guidotti, A.;
Costa, E. Pregnenolone Biosynthesis in C6-2B Glioma Cell
Mitochondria: Regulation by a Mitochondrial Diazepam Binding
Inhibitor Receptor. Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 5113-
5117.
O2
Consumption of Neuroblastoma Cells. Eur. J . Pharmacol.
1989, 161, 197-202.
(23) Hirsch, J . D.; Beyer, C. F.; Malkowitz, L.; Beer, B.; Blume, A. J .
Mitochondrial Benzodiazepine Receptors Mediate Inhibition of
Mitochondrial Respiratory Control. Mol. Pharmacol. 1988, 34,
157-163.
(24) Wang, J . K. T.; Morgan, J . L.; Spector, S. Benzodiazepines that
Bind at Peripheral Sites Inhibit Cell Proliferation. Proc. Natl.
Acad. Sci. U.S.A. 1984, 81, 753-756.
(25) Curran, T.; Morgan, J . I. Superinduction of c-fos by Nerve
Growth Factor in the Presence of Peripherally Active Benzodi-
azepines. Bioassays 1987, 7, 255-258.
(26) (a) Baulieu, E.-E.; Robel, P. J . Neurosteroids: a New Brain
Function? J . Steroid Biochem. Mol. Biol. 1990, 37, 395-403. (b)
Anholt, R. R. H.; De Souza, E. B.; Kuhar, M. J .; Snyder, S. H.
Depletion of Peripheral-Type Benzodiazepine Receptors after
Hypophysectomy in Rat Adrenal Gland and Testis. Eur. J .
Pharmacol. 1985, 110, 41-46.
(27) J ung-Testas, I.; Hu, Z.; Baulieu, E. E. Biosynthesis of Preg-
nenolone and Progesterone in Primary Cultures of Rat Glial
Cells. Endocrinology 1989, 125, 2083-2091.
(28) Costa, E.; Guidotti, A. Diazepam Binding Inhibitor (DBI). A
Peptide with Multiple Biological Actions. Life Sci. 1981, 49, 325-
344.
(29) Massotti, M.; Slobodiansky, E.; Konkel, D.; Costa, E.; Guidotti,
A. Regulation of Diazepam Binding Inhibitor in Rat Adrenal
Gland by Adrenocorticotropin. Endocrinology 1991, 129, 591-
596.
(30) Sprengel, R.; Werner, P.; Seeburg, P. Molecular Cloning and
Expression of cDNA Encoding a Peripheral-Type Benzodiazepine
Receptor. J . Biol. Chem. 1989, 264, 20415-20421.
(49) Allen, F. H.; Bellard, S.; Brice, M. D.; Cartwright, B. A.;
Doubleday, A.; Higgs, H.; Hummelink, T.; Hummelink-Peters,
B. G.; Kennard, O.; Motherwell, W. D. S.; Rodgers, J . R.; Watson,
D. G. The Cambridge Crystallographic Data Centre: Computer-
Based Search, Retrieval, Analysis and Display of Information.
Acta Crystallogr. 1979, B35, 2331-2339.

3450 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 18
Campiani et al.
(50) Abraham, M. H.; Duce, P. P.; Prior, D. V.; Barratt, D. G.; Morris,
J . J .; Taylor, P. J . Hydrogen Bonding. Part 9. Solute Proton
Donor and Proton Acceptor Scales for Use in Drug Design. J
Chem. Soc. Perkin Trans. 2 1989, 1355-1375.
(51) Raevsky, O. A.; Grigorev, V. Y.; Kireev, D. B.; Zefirov, N. S.
Complete Thermodynamic Description of H-Bonding in the
Framework of Multiplicative Approach. Quant. Struct.-Act.
Relat. 1992, 11, 49-63.
(52) Hansch, C.; Leo, A.; Unger, S.; Kim, K. H.; Nikaitani, D.; Lien,
E. “Aromatic” Substituent Constants for Structure-Activity
Correlations. J . Med. Chem. 1973, 11, 1207-1216.
(53) Nacci V. et al. Unpublished results.
(58) Munson, P. G.; Roadbard, D. Computers in Endocrinology; Raven
Press: New York, 1984; pp 117-145.
(59) Markwell, M.; Haas, S. M.; Bieker, L.; Tolbert, N. A. A Modifica-
tion of the Lowry Procedure to Simplify Protein Determination
in Membrane and Lipoprotein Samples. Anal. Biochem. 1980,
87, 206-210.
(60) Moynagh, P. N.; Bailey, C. J .; Boyce, S. J .; Williams, D. C.
Immunological Studies on the Rat Peripheral-Type Benzodiaz-
epine Acceptor. Biochem. J . 1991, 275, 419-425.
(61) SYBYL Molecular Modeling System (version 6.01), TRIPOS Ass.
St. Louis, MO.
(54) Clark, M.; Cramer, R. D., III; Van Opdenbosch, N. Validation
of the General Purpose Tripos 5.2 Force Field. J . Comput. Chem.
1989, 8, 892-1012.
(62) Dewar, M. J . S.; Thiel, W. Ground States of Molecules. 38. The
MNDO Method. Approximations and Parameters. J . Am. Chem.
Soc. 1977, 99, 4899-4907.
(63) Dewar, M. J . S.; Zoebish, E. G.; Healy, E. F.; Stewart, J . J . P.
AM1: a New General Purpose Mechanical Molecular Model. J .
Am. Chem. Soc. 1985, 107, 3902-3909.
(64) Stewart, J . J . P. Optimization of Parameters for Semi-Empirical
Methods. 1. The PM3 method. J . Comput. Chem. 1989, 10, 209-
220.
(65) MOPAC (version 5.00), Quantum Chemistry Program Exchange,
No. 455, 1989.
(55) Murray-Rust, P.; Glusker, J . P. Directional Hydrogen Bonding
to sp2- and sp3-Hybridized Oxygen Atoms and its Relevance to
Ligand-Macromolecule Interactions. J . Am. Chem. Soc. 1984,
106, 1018-1025.
(56) The Ca values for N,N-diethylthioacetamide is not reported in
ref 50.
(57) Cantoni, L.; Rizzardini, M.; Skorupska, M.; Cagnotto, A.; Cat-
egoni, A.; Pecora, N.; Frigo, L.; Ferrarese, C.; Mennini, T. Hepatic
Protoporphyria is Associated with a Decrease in Ligand Binding
for Mitochondrial Benzodiazepine Receptors in Liver. Biochem.
Pharmacol. 1992, 44, 1159-1164.
J M960251B