Piperazine imidazo[1,5-a]quinoxaline ureas as high-affinity GABA(A) ligands of dual functionality

Article abstract of DOI:10.1021/jm9801307

A series of imidazo[1,5-a]quinoxaline piperazine ureas appended with a tert-butyl ester side chain at the 3-position was developed. Analogues within this series have high affinity for the γ-aminobutyric acid A (GABA(A))/benzodiazepine receptor complex with efficacies ranging from inverse agonists to full agonists. Many analogues were found to be partial agonists as indicated by [³⁵S]TBPS and Cl^- current ratios. Uniquely, a number of these analogues were found to have a bell-shaped dose-response profile in the α₁β₂γ₂ subtype as determined by whole cell patch-clamp technique, where in vitro efficacy was found to decrease with increasing drug concentration. Many of the compounds from this series were effective in antagonizing metrazole-induced seizures, consistent with anticonvulsant and possibly anxiolytic activity. Additionally, several analogues were also effective in lowering cGMP levels (to control values) after applied stress, also consistent with anxiolytic-like properties. The most effective compounds in these screens were also active in animal models of anxiety such as the Vogel and Geller assays. The use of the piperazine substituent allowed for excellent drug levels and a long duration of action in the central nervous system for many of the quinoxalines, as determined by ex vivo assay. Pharmacokinetic analysis of several compounds indicated excellent oral bioavailability and a reasonable half-life in rats. From this series emerged two partial agonists (55, 91) which had good activity in anxiolytic models, acceptable pharmacokinetics, and minimal benzodiazepine-type side effects.

Full text of DOI:10.1021/jm9801307

J . Med. Chem. 1999, 42, 1123-1144
1123
P ip er a zin e Im id a zo[1,5-a ]qu in oxa lin e Ur ea s a s High -Affin ity GABAA Liga n d s of
Du a l F u n ction a lity†
E. J on J acobsen,*,‡ Lindsay S. Stelzer,‡ Ruth E. TenBrink,‡ Kenneth L. Belonga,‡ Donald B. Carter,§
Haesook K. Im,§ Wha Bin Im,§ Vimala H. Sethy,§ Andy H. Tang,§ Philip F. VonVoigtlander,§ J ames D. Petke,|
Wei-Zhu Zhong, and J ohn W. Mickelson∇
Departments of Structural, Analytical and Medicinal Chemistry, Central Nervous System Diseases Research, Computer Aided
Drug Discovery, and Pharmacokinetics and Bioanalysis Research, Upjohn Laboratories, Pharmacia & Upjohn, Inc.,
Kalamazoo, Michigan 49001
Received February 27, 1998
A series of imidazo[1,5-a]quinoxaline piperazine ureas appended with a tert-butyl ester side
chain at the 3-position was developed. Analogues within this series have high affinity for the
γ-aminobutyric acid A (GABAA)/benzodiazepine receptor complex with efficacies ranging from
inverse agonists to full agonists. Many analogues were found to be partial agonists as indicated
by [35S]TBPS and Cl- current ratios. Uniquely, a number of these analogues were found to
have a bell-shaped dose-response profile in the R1â2γ2 subtype as determined by whole cell
patch-clamp technique, where in vitro efficacy was found to decrease with increasing drug
concentration. Many of the compounds from this series were effective in antagonizing metrazole-
induced seizures, consistent with anticonvulsant and possibly anxiolytic activity. Additionally,
several analogues were also effective in lowering cGMP levels (to control values) after applied
stress, also consistent with anxiolytic-like properties. The most effective compounds in these
screens were also active in animal models of anxiety such as the Vogel and Geller assays. The
use of the piperazine substituent allowed for excellent drug levels and a long duration of action
in the central nervous system for many of the quinoxalines, as determined by ex vivo assay.
Pharmacokinetic analysis of several compounds indicated excellent oral bioavailability and a
reasonable half-life in rats. From this series emerged two partial agonists (55, 91) which had
good activity in anxiolytic models, acceptable pharmacokinetics, and minimal benzodiazepine-
type side effects.
The action of γ-aminobutyric acid (GABA) on the
GABAA chloride ion channel complex controls the excit-
ability of many central nervous system (CNS) path-
ways.1 Numerous chemical classes have binding sites
on this macromolecular ionophore, including the ben-
zodiazepines, neurosteroids, and barbiturates. Com-
pounds which interact with the benzodiazepine receptor
(BzR) and allosterically modulate the action of GABA
on neuronal chloride flux have a continuum of intrinsic
activity,2 ranging from full agonists (anxiolytic, hyp-
notic, and anticonvulsant agents) through antagonists
to inverse agonists (proconvulsant and anxiogenic
agents). Partial agonists lie within this continuum and
may have reduced benzodiazepine-mediated side effects
such as physical dependence, amnesia, oversedation
(anxiolytics), and muscle relaxation.3 The discovery and
characterization of partial agonist ligands have been
reported by several research groups.4-10 Furthermore,
recent molecular biology studies have demonstrated that
several different receptor subunits (R, â, γ, δ) combine
to form the GABAA receptor complex,11 with at least
three to four native receptor subtypes identified thus
far.12 Subtype selective ligands may also allow for the
discrimination between useful anxiolytic or hypnotic
activity and overt side effects, and as such they are a
topic of current research.
We recently reported on an imidazo[1,5-a]quinoxaline
urea 1 (PNU-91571, Figure 1) which was a partial
agonist in vitro (R1â2γ2 and R3â2γ2 subtypes) and
displayed useful activity in anxiolytic models, yet lacked
typical benzodiazepine-type side effects such as physical
dependence and ethanol potentiation.13 Unfortunately
this analogue contained a 5′-cyclopropyl-1′,2′,4′-oxadia-
zole group at the 3-position, which in the case of 2
(pandiplon, PNU-78875)9a is metabolized to release
cyclopropanecarboxylic acid, leading to an increase in
serum triglycerides.14 Although 1 had a different meta-
bolic profile than 2, concerns over possible degradation
of the oxadiazole group to release cyclopropane carboxy-
lic acid precluded further development of this compound.
Efforts to replace the oxadiazole with other aromatic
groups such as phenyl,15 oxazole, or isoxazole were
successful with respect to maintaining excellent binding
affinity; however, unacceptable benzodiazepine (BzD)
side effects were typically observed. While the use of
an ester substituent at the 3-position also maintained
affinity, intrinsic activity was notably decreased. In our
exploration of the 3-phenyl series,15 the use of a cis-2,6-
dimethylpiperazine as the urea substituent was found
to dramatically improve CNS penetration and duration
of action (by ex vivo assay) as compared to other ureas
†A portion of this material was presented at the Winter Conference
on Medicinal and Bioorganic Chemistry, Steamboat Springs, CO,
J anuary 28, 1995.
‡ Department of Structural, Analytical and Medicinal Chemistry.
§ Department of Central Nervous Diseases System Research.
| Department of Computer Aided Drug Discovery.
Department of Pharmacokinetics and Bioanalysis Research.
∇ Present address: 3M Pharmaceuticals, 3M Center Bldg., St. Paul,
MN 55144-1000.
10.1021/jm9801307 CCC: $18.00 © 1999 American Chemical Society
Published on Web 03/20/1999

1124 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7
J acobsen et al.
Sch em e 1a
F igu r e 1. Quinoxalines 1-4.
lacking a basic nitrogen. More importantly, compounds
such as 3 (PNU-97035, Figure 1) were found to display
biphasic efficacy by in vitro measurement, becoming
increasingly antagonistic with increasing drug concen-
tration. While the observed biphasic efficacy did not
serve to minimize side effects such as physical depen-
dence, which in this series was severe, this unique
property was worthy of further evaluation. With proper
adjustment of intrinsic activity, a therapeutic agent with
biphasic efficacy similar to that observed for 3 would
serve to limit its own agonistic effects over a wide dose
range, potentially minimizing its abuse potential. In this
paper we report on the use of a tert-butyl ester sub-
stituent at the 3-position which, in combination with a
C-methyl piperazine urea (i.e., 4, Figure 1), provides for
partial agonist activity. Furthermore, several analogues
were identified which have excellent activity in anx-
iolytic models, minimal BzD-type side effects, and
acceptable pharmacokinetic properties.
a
Reagents: (a) tBuOH, THF; diethyl chlorophosphate; tBuOK,
CNCH2CO2 Bu; (b) phosgene or triphosgene, EtNiPr2, CH2Cl2; (c)
t
8a -8c, EtNiPr2, CH2Cl2; (d) TFA, CH2Cl2, 0 °C; (e) R3CHO,
NaCNBH3, MeOH; (f) R3Cl, EtNiPr2, CH2Cl2.
elaborated piperazines, a t-Boc protecting group (8a , 9)
was utilized which was easily removed using TFA to
give 10, with no cleavage of the tert-butyl ester observed.
The introduction of an N-alkyl group in 10 to give
piperazine 11 (R3 ) alkyl) was generally carried out by
the reductive amination procedure of Borch.17 Alterna-
tively, the N-alkyl group was incorporated in piperazine
8c (R3 ) alkyl) to give 11 directly (7 f 11). Simple
acylation of 10 provided analogues 71 and 72.
Ch em istr y
An alternative sequence, particularly useful for the
synthesis of gem-4,4-dimethylimidazo[1,5-a]quinoxaline
14, is shown in Scheme 2. Conversion of 5 to piperazine
urea 13 via the carbamoyl chloride proceeded unevent-
fully. As described above, reaction of the enol phospho-
nate of 13 with tert-butyl isocyanoacetate provided 10
or 11 in good yield. Elaboration of 10 to the desired
quinoxaline 11, if desired, was carried out as detailed
in Scheme 1. This sequence was critical for the synthesis
of 57 and 58, which contained gem-dimethyl groups (i.e.,
14) at the 4-position.
The synthesis of the quinoxalin-2-one starting mate-
rials 5a -5g was carried out as previously described.13,16
Incorporation of a trifluoromethyl group into this tem-
plate at the 7-position was accomplished as shown in
Scheme 3. Reaction of 2-chloro-5-trifluoromethyl-
nitrobenzene 15 with dibenzylamine provided 16 in
excellent yield. Reduction of 16 using Raney nickel
provided aniline 17, which was alkylated with ethyl
bromoacetate in neat Hunig’s base to give 18 in a
reasonable 70% yield. Hydrogenation using Pearlman’s
catalyst provided a 97% yield of the desired quinoxaline,
which was isolated in its enol form 19 due to the
The general synthesis of the tert-butyl imidazo[1,5-
a]quinoxaline carboxylates was carried out in a fashion
similar to that of the oxadiazoles reported previously13,16
and is shown in Scheme 1. Imidazo-annulation of
quinoxalin-2-one (5),13,16 via the intermediate enol phos-
phonate using tert-butyl isocyanoacetate, provided imi-
dazo[1,5-a]quinoxaline 6 in 60-80% yield. The carbam-
oyl chloride 7 was formed by reaction of 6 with phosgene
or triphosgene in the presence of Hunig’s base. This
intermediate (7) could either be isolated or, more
frequently, reacted in situ with the desired piperazine
8a -8c to give ureas 9, 10, and 11, respectively. Protec-
tion of the piperazine (8a vs 8b) was usually not
necessary for the cis-2,6-dimethylpiperazine reagent,
with excellent yields usually achieved. For less bulky
piperazines, regioisomeric or bis-addition to 7 often
occurred. These side reactions could be avoided (par-
tially) through the use of a large excess of the desired
piperazine. For example, in the case of 76 the use of
1.0 equiv of 2-methylpiperazine (8b, R1 ) Me) provided
only a 19% yield of the urea. The yield could be
significantly improved (64%) by the use of 10 equiv of
the piperazine reagent. Nonetheless, for the more

Piperazine Imidazo[1,5-a]quinoxaline Ureas
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7 1125
Sch em e 2a
Sch em e 3a
a
Reagents: (a) Bn2NH, EtNiPr2, 140 °C; (b) H2NHNH2‚H2O,
Raney nickel, EtOH; (c) BrCH2CO2Et, EtNiPr2, 120 °C; (d)
Pd(OH)2, H2, EtOH; (e) tBuOK, THF; diethyl chlorophosphate;
CNCH2CO2 Bu, BuOK.
t
t
a
Reagents: (a) phosgene, EtNiPr2, CH2Cl2; (b) 8a -8c, EtNiPr2,
t
CH2Cl2; (c) tBuOK, THF; diethyl chlorophosphate; CNCH2CO2 Bu,
tBuOK.
electron-withdrawing effects of the trifluoromethyl group.
Cyclization of 19 under standard conditions provided
the desired imidazo[1,5-a]quinoxaline 20.
A number of noncommercial piperazines (Figure 2)
were synthesized according to literature precedent, with
minor protecting group manipulation, if necessary.
These piperazines and bicyclo-derived analogues 21,18
22,19 23,20 24,21 25,22 26,22 27,23 and 2824 were prepared
to fully explore the structure-activity relationship
(SAR) of the urea substituent. Piperazine 32 (1-isopro-
pyl-cis-2,6-dimethylpiperazine) was synthesized by
straightforward chemistry as shown in Scheme 4.
The chiral 3,5-methylated piperazine urea analogues
(89-100) were synthesized by one of two methods. As
illustrated in Scheme 1, reaction of carbamoyl chloride
7 with the individual enantiomers of 2-methylpipera-
zine25 or trans-2,6-dimethylpiperazine25 provided the
desired quinoxaline ureas 10 directly. An alternative
approach using an intramolecular Mitsunobu reaction26
was also developed as shown in Scheme 5. Following
our previously outlined chemistry,25 but without pro-
tecting groups, the desired antipode of Boc-alanine 33
was converted to amides 34A and 34B by reaction of
the CDI-intermediate with ethanolamine or 1-amino-
2-propanol ((R)-shown). Reduction of 34A or 34B with
borane-methyl sulfide complex and hydrolysis of the
intermediate borane adduct with aqueous HCl followed
by aqueous potassium hydroxide provided amines 35A
and 35B in good yield. Reaction of either of these amines
F igu r e 2. Selected piperazine intermediates.
with carbamoyl chloride 7 provided amides 36A and
36B, which were deprotected with TFA to give amino
alcohols 37A and 37B. Intramolecular cyclization of 37A
and 37B using Mitsunobu conditions provided the
desired piperazines 38A and 38B. This latter strategy,
which avoided protecting group manipulation, was quite
efficient and provided the 3-methyl- and trans-3,5-di-
methylpiperazines in >98% ee as determined by HPLC

1126 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7
J acobsen et al.
Sch em e 4a
benzodiazepine receptor subtypes to assess binding
selectivity.29,30 Classical benzodiazepines interact with
the R1â2γ2, R2â2γ2, R3â2γ2, and R5â2γ2 subtypes with
nearly equal affinity,31 while atypical BzR ligands such
as Cl 218872 and Zolpidem have moderate selectivity
for the R1â2γ2 subtype.4a,31 In contrast, classical benzo-
diazepines do not interact with the R6â2γ2 subtype
located in cerebellar granule cells, whereas ligands such
as Ro 15-4513 and (RS)-1-(5-cyclopropyl-1,2,4-oxadiazol-
3-yl)-12,12a-dihydroimidazo[1,5-a]pyrrolo[2,1-c]quinoxa-
lin-10(11H)-one (PNU-89267) have high affinity for this
subtype.16,32 The search for selective ligands for these
subtypes, as well as the elucidation of their function,
remains an important goal in this area.
a
Reagents: (a)di-tert-butyldicarbonate,CH2Cl2;(b)(CH3)2CHCH2I,
K2CO3, CH3CN; (c) TFA, CH2Cl2, 0 °C.
Two different methods were utilized to evaluate the
in vitro efficacy of these compounds. The TBPS shift
ratio33 was determined for each analogue by measuring
its effect on tert-butylbicyclophosphorothionate (TBPS)
binding to the picrotoxin convulsant site on the GABAA
chloride complex. Differences in TBPS binding presum-
ably occur due to conformational changes in the chloride
ionophore as it is allosterically modulated by the binding
of the test compound to the benzodiazepine receptor. As
reported previously,13 the resultant value (expressed as
a ratio of that for the test drug to that of diazepam) is
1 for a full agonist and 0 for an antagonist, with
negative values for inverse agonists. Partial agonists
have values intermediate between 0 and 1. A more
direct measure of in vitro efficacy was determined by a
Cl- current assay.33-35 The synaptic chloride conduc-
tance resulting from the activation of the GABAA
receptor complex by GABA is modulated by ligands
acting at the benzodiazepine receptor. Full agonists
increase Cl- current and antagonists have no effect,
while inverse agonists decrease ion flow. The test
compounds are compared to diazepam and thus have a
numerical scale like the TBPS assay. In general, the
chloride current assay provides a more direct measure
of efficacy as it is run in a single subtype (R1â2γ2),
whereas the TBPS shift ratio is determined across the
native population. Results from these assays are often
different due to the heterogeneity of the native GABAA
receptor population. Furthermore, as the Cl- current
and TBPS assays are also run at different drug concen-
trations, biphasic piperazine ureas15 such as 3 (Figure
1) often display dramatically different values. Many of
the urea analogues were evaluated at two different drug
concentrations in the Cl- current assay in order to
assess possible biphasic effects. In addition, as benzo-
diazepine ligands are known to display efficacy selectiv-
ity in different receptor subtypes,36 selected analogues
were screened for intrinsic efficacy in the R1â2γ2 and
R3â2γ2 subtypes.
Sch em e 5a
a
Reagents: (a) CDI, CH2Cl2; ethanolamine or (R)-1-amino-2-
propanol; (b) BH3‚DMS, THF; 10% HCl; KOH, THF, H2O, reflux;
(c) 7, EtNiPr2, CH2Cl2; (d) TFA, CH2Cl2, 0 °C; (e) DEAD, PPh3,
THF.
1
or H NMR analysis.27 Table 1 highlights the physical
data and methods used for the synthesis of the imidazo-
[1,5-a]quinoxaline piperazines as well those for a few
simple urea analogues.
To provide a quick measure of in vivo efficacy, most
analogues were evaluated for their ability to antagonize
metrazole-induced seizures (clonic and tonic).37 While
a direct measure of anticonvulsant activity, this assay
is also predictive of anxiolytic properties, as standard
full agonist benzodiazepine agents such as diazepam,
alprazolam, and zolpidem (and partial agonists 2, breta-
zenil, and abecarnil) are extremely effective (ED50’s
0.17-4.5 mg/kg). The results of the flunitrazepam,
Resu lts a n d Discu ssion
The piperazine quinoxalines were initially evaluated
for binding affinity and in vitro efficacy. Binding affinity
for the benzodiazepine receptor in rat cortical mem-
branes was determined by competition experiments
with radiolabeled [3H]flunitrazepam.28 Selected ana-
logues were also evaluated for affinity to individual

Piperazine Imidazo[1,5-a]quinoxaline Ureas
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7 1127
Ta ble 1. Physical Data for Imidazo[1,5-a]quinoxaline Ureas, Carbamates, and Amides
yield
(%)
compd
R1
R2
R3
R6
R7
mp (°C)
method
formula
anal.
C, H, N
C, H, N
C, H, N
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57b
58b
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
H
H
H
H
Me
H
Me
H
Me
H
Me
H
Me
H
H
Me
H
Me
H
Me
H
Me
H
H
H
H
H
H
H
H
H
Cl
Cl
H
H
H
Me
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
F
260-261
186-187
190-192
166-168
154-156
187-188
169-171
200.5-201
172-173
A
A
B
E
B
A
E
A
B
A
B
A
E
E
A
B
A
B
E
E
E
E
E
E
E
A
A
A
C
B
A
C
70
C
C
C
C
C
20H25N5O3‚(HCl)‚(H2O)3
22H29N5O3
cis-3,5-dimethyl
cis-3,5-dimethyl
cis-3,5-dimethyl
cis-3,5-dimethyl
H
87
78
68
53
59
48
80
92
86
88
59
69
73
74
82
82
81
47
30
53
66
62
48
47
79
77
74
84
64
96
95
87
62
60
64
31
23H31N5O3
22H28N5O3F
23H30N5O3F
C, H, N
F
C, H, Na
H
H
Cl
Cl
Cl
Cl
H
C20H24N5O3Cl
21H26N5O3Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N
C, H, N
C, H, N
C, H, N
C, H, N
H
C
C
C
C
C
C
C
C
C
C
cis-3,5-dimethyl
cis-3,5-dimethyl
cis-3,5-dimethyl
cis-3,5-dimethyl
H
cis-3,5-dimethyl
cis-3,5-dimethyl
cis-3,5-dimethyl
cis-3,5-dimethyl
cis-3,5-dimethyl
cis-3,5-dimethyl
cis-3,5-dimethyl
cis-3,5-dimethyl
22H28N5O3Cl‚(C4H8O2)1/4
23H30N5O3Cl‚(H2O)1/2
22H28N5O3Cl‚(H2O)1/4
23H30N5O3Cl‚(H2O)3/4
21H27N5O3‚(H2O)1/8
23H31N5O3‚(H2O)1/5
24H33N5O3
>300
H
221-222
214-214.5
188-190
200-201
212-213.5
208-210
184-186
177-181
171-175
200-201
184-186
167-167.5
205-206
210-211.5
194-196
187.5-189
222-224
233-235
225-226
127-130
186-187
211-213
154-154.5
156-156.5
136-137.5
152-153
160-161
169-172
128-130
186-187
168-169
164-165
161-162
206-207
140-143
217-220
270-272
H
Me
Me
Me
H
23H31N5O3‚(H2O)1/4
H
24H33N5O3
CF3
CF3
Cl
Cl
H
H
H
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
F
C23H28N5O3F3
C24H30N5O3F3‚(H2O)2/3
C24H32N5O3Cl
C25H34N5O3Cl
C18H22N4O3
C, H, N
C, H, N
C, H, N, Clc
C, H, N, Cl
C, H, N
NR2 ) NMe2
NR2 ) piperidine
NR2 ) morpholine
NR2 ) NMe2
C21H26N4O3
C20H24N4O4
C, H, N
C, H, N
C18H21N4O3Cl
C20H23N4O3Cl‚(C4H8O2)1/4
C24H32N5O3Cl
C23H28N5O3Cl
25H27N6O3Cl
C22H25N5O3ClF3
C24H32N5O3Cl‚(C4H8O2)1/2‚(H2O)1.33 C, H, N, Cl
C25H34N5O3Cl
C24H29N5O3ClF3
C24H30N5O4Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
NR2 ) pyrrolidine
H
H
H
H
H
H
H
tBu
cPr
2-pyridyl
C
H
CH2CF3
cis-3,5-dimethyl
cis-3,5-dimethyl
cis-3,5-dimethyl
cis-3,5-dimethyl
cis-3,5-dimethyl
cis-2,6-dimethyl
cis-2,6-dimethyl
cis-2,6-dimethyl
Et
iPr
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N,d Cl, S
C, H, N
CH2CF3
COCH3
SO2Me
H
Me
Me
H
H
H
H
H
H
Me
H
H
H
C23H30N5O5ClS‚(CH2Cl2)1/10
D
B
E
C
C
C
22H28N5O3F
23H30N5O3F
23H30N5O3Cl
F
C, H, N
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
3-CH3
H
A
C
A
D
C
C
B
A
C
D
19 (64)e C21H26N5O3Cl
trans-3,5-dimethyl
trans-2,5-dimethyl
cis-2,3-dimethyl
trans-2,3-dimethyl
3,3-dimethyl
36
56
41
30
56
82
71
65
63
C
C
C
C
C
C
C
C
22H28N5O3Cl‚(H2O)4/5
22H28N5O3Cl‚(C3H4O4)‚(H2O)2/3
22H28N5O3Cl
22H28N5O3Cl
22H28N5O3Cl‚(H2O)1/4
23H30N5O3Cl
24H32N5O3Cl‚(H2O)1/4
30H40N5O3Cl‚(H2O)1/4
C21H24N5O3Cl‚(H2O)
3,3-dimethyl
3,3,5,5-tetramethyl
bis-spiro-cyclohexyl
86
87
Me
H
H
H
Cl
Cl
174-176
207-210
C
D
48
79
C
22H26N5O3Cl‚(H2O)1/4
C, H, N, Cl
C, H, N, Cl
C22H26N5O3Cl‚(H2O)1/4
88
89
90
91
92
93
94
95
96
97
98
99
100
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Cl
Cl
Cl
Cl
213-216
158-160
157-159
121-125
125-127
161-165
148-152
140-144
127-130
187-188
176.5-180.5
151-154
157-158
B
F
D
F
80
59
76
71
55
59
71
60
67
64
64
52
65
C
C
C
C
C
C
C
C
C
23H28N5O3Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N, Cl
C, H, N
C, H, N
C, H, N
C, H, N
3(S)-Me
3(R)-Me
3(S)-Me
3(R)-Me
3(S)-Me
3(R)-Me
3(S)-Me
3(R)-Me
3(S)-Me
3(R)-Me
3(S)-Me
3(R)-Me
H
H
21H26N5O3Cl‚(H2O)4/5
21H26N5O3Cl‚(H2O)1/2
22H28N5O3Cl‚(H2O)1/3
22H28N5O3Cl‚(H2O)1/3
22H29N5O3‚(H2O)1/2
22H29N5O3‚(H2O)3/5
23H31N5O3‚(H2O)1/2
23H31N5O3‚(H2O)1/2
5(S)-Me
5(R)-Me
H
Cl
F
Me
Me
Me
Me
CF3
CF3
CF3
CF3
D
D
C
C
D
D
F
H
5(S)-Me
5(R)-Me
H
C22H26N5O3F3
C, H, N
H
C22H26N5O3F3‚(H2O)1/5
C23H28N5O3F3‚(H2O)1/2
C23H28N5O3F3
C, H, Nc
C, H, N
5(S)-Me
3(R)-Me
F
C, H, N
a
b
d
N: calcd, 15.79; found, 15.29. gem-4,4-Dimethyl. c Cl: calcd, 7.48; found, 7.99. N: calcd, 13.15; found, 12.55. e Yield of 76 when 10
equiv of 8b was used.

1128 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7
J acobsen et al.
Ta ble 2. [3H]Fnz Binding, TBPS Shift, Cl- Current Changes, and Metrazole Antagonism Data for Imidazo[1,5-a]quinoxaline
Piperazine Ureas with Varying A-Ring Substitution
[
35S]TBPS
shiftb,c
∆Cl- currentb,c
R1â2γ2 5.0 µM/0.5 µM
∆Cl- currentb,c
R3â2γ2 5.0 µM
metrazoled ED50
(95% confidence interval) mg/kg, ip
compd
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
R1
R3
R6,7
Ki (nM)a
H
Me
H
H
H
H
H
7-F
4.11
3.04
6.67
1.04
2.69
3.99
10.87
3.27
4.48
28.16
56.26
14.7
4.66
9.50
250
0.04
/0.35
0.0/0.50
-0.1/0.50
0.0/0.50
-0.1/1.0
0.0/0.40
0.98/
0.3/0.85
-0.2/1.0
0.29/
>50
-0.018
-0.82
-0.14
0.44
-0.02
0.57
0.48
-0.06
0.05
0.22
0.30
0.38
0.11
0.54
0.48
0.31
0.26
>50
Me Me
Me
Me Me 7-F
H
H
>50
H
35.3 (21.8-57.1)
35.3 (21.8-57.1)
H
7-Cl
Me 7-Cl
7-Cl
>50
0.44
0.54
1.00
7.4
3.7
Me
H
Me Me 7-Cl
Me 6-Cl
Me Me 6-Cl
H
Me
Me Me 7-Me
Me 6-Me
Me Me 6-Me
Me 7-CF3
Me Me 7-CF3
Me 7-Cl
Me Me 7-Cl
14.8
>50
>50
H
0.71/
H
H
7-Me
7-Me
0.40/0.55
0.30/0.84
1.37/
0.22/
0.33/
0.50/
0.81/
0.96/
2.60/
21.0 (15.0-29.5)
0.1
0.56
1.6
>50
>50
21.0 (14.9-29.5)
6.2
0.40
0.60
H
54
55
56
470
H
32.38
31.42
19.93
39.75
4.9
0.1
57e
58e
diazepam
1
H
1.23
1.43
1.0 ( 0.2
0.69
1.08
0.68
1.0 ( 0.2
1.0 ( 0.2
0.68/
0.50 (0.38-1.2)
21
1.0
2
1.6
0.06
0.04
1.6
a
Mean binding affinity against [3H]flunitrazepam; see ref 28 and the Experimental Section for methods. The standard error was <(10%
b
of the mean. Diazepam is defined as a full agonist which gives a value of 1. Antagonists are defined as having a shift value of 0; partial
agonists are intermediate. c See the Experimental Section. The standard error was <(10% of the mean. Antagonism of metrazole-
d
induced clonic convulsions in the rat after ip injection; see the Experimental Section. e gem-4,4-Dimethyl.
TBPS shift ratio, Cl- current, and metrazole assays are
provided in Tables 2-7.
of these analogues were active in the metrazole assay.
The 7-fluoro analogues had greater efficacy with 43
being a full agonist at 0.5 µM in the chloride current
screen. Both of these derivatives displayed biphasic
properties and were weak metrazole antagonists. Sub-
stitution at the 7-position with a larger group (Cl, CH3,
CF3) provided partial to full agonist analogues, many
of which displayed potent metrazole antagonism activ-
ity. In the TBPS assay only 44 and 47 were antagonists,
with the rest of the ureas partial agonists. Biphasic
effects were clearly present with 47 displaying full
agonist properties at 0.5 µM. This derivative, along
with 45, 46, 52, and 56, was an effective metrazole
antagonist. In comparing the simple piperazine ureas,
an N-methyl group is important for activity, with 45
quite effective in the metrazole screen as compared to
44. In contrast, the N-methyl group is not necessary for
metrazole activity for the cis-3,5-dimethylpiperazine
analogues. The requirement of a bulky 7-substituent for
potent metrazole activity is unique to this tert-butyl
ester series, as the corresponding 3-oxadiazole13 and
3-phenyl15 analogues were generally effective with un-
substituted A-rings. Substitution at the 4-position with
gem-dimethyl groups was also tolerated, with 57 and
58 suffering only a 6-8-fold loss in binding affinity as
compared to 46 and 47. As with the corresponding
3-oxadiazole and 3-phenyl series, this substitution
provided for enhanced efficacy, with 58 having a 2.6-
fold greater Cl- current than diazepam. Several ana-
Our initial screening efforts were directed toward the
evaluation of the cis-3,5-dimethyl- and cis-3,5,4-tri-
methylpiperazine ureas with a variety of A-ring sub-
stituents. Most of these analogues had excellent affinity
for the benzodiazepine receptor with Ki’s typically
between 1.0 and 15 nM as shown in Table 2. The major
exceptions proved to be the 6-Cl (49) and 6-methyl (54)
analogues with 54 having a Ki of 470 nM. The decreased
affinity of the 6-substituted analogues is consistent with
the SAR of the 3-phenyl analogues reported earlier15
and is in contrast to the typical benzodiazepine SAR.38
Of the substituents examined at the 7-position, only the
trifluoromethyl analogues (55, 56) suffered a decrease
in affinity (32 nM). Substitution on the A-ring was found
to correlate quite well with intrinsic activity and potency
in the metrazole antagonism assay. The unsubstituted
analogues proved to be antagonists by TBPS shift and
Cl- current assays, with 41 a potent inverse agonist
(TBPS shift ) -0.82). Consistent with this potent
inverse agonist activity, 41 did promote seizures at high
doses (>100 mg/kg). This property is known for benzo-
diazepine inverse agonists.1,2 However, biphasic effects
were observed in the Cl- current screen with 39-41
behaving as partial agonists at 0.5 µM (Cl- current
0.35-0.50). The 6-chloro and 6-methyl analogues were
quite similar, with TBPS shift ratios and Cl- current
consistent with partial agonist intrinsic efficacy. None

Piperazine Imidazo[1,5-a]quinoxaline Ureas
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7 1129
F igu r e 3. TBPS binding dose-response curve of 46 and 47.
Full agonist ) -1.0, antagonist ) 0.0, full inverse agonist )
+1.0. Each point represents the average of three measure-
ments with experimental errors less than 15%. Typically,
diazepam at 1 µM reduced TBPS binding by 34 ( 4%, while
DMCM at 0.1 µM increased TBPS binding by 48%, under the
same experimental conditions in this study.
F igu r e 4. Cl- current dose-response curve for 47 in several
GABAA receptor subtypes. Full agonist ) 200, antagonist )
100, full inverse agonist ) 0. Diazepam at 1 µM increased
GABA currents by 143 ( 44% in the R1â2γ2 subtype and by
312 ( 110% in R3â2γ2.
also evaluated with the results provided in Table 3. All
of these analogues, regardless of the amine (NR2) or R7
substituents, had high binding affinity. The unsubsti-
tuted quinoxalines 59-61 (R7 ) H) were antagonists
by the TBPS and Cl- current assays, while the 7-chloro
analogues had slightly enhanced intrinsic efficacy and
were partial agonists. The TBPS shift ratio and Cl-
current values were nearly identical for each analogue
with no biphasic effects observed. Only the 7-chloro
analogues, 62 and 63, were active in the metrazole
screen, albeit to a moderate degree.
The SAR of the distal nitrogen substituent (R3) was
evaluated for both piperazine and cis-3,5-dimethylpip-
erazine moieties while maintaining a 7-chloro group on
the A-ring. As shown in Table 4, the R3 substituent had
little effect on affinity with only 70 having a Ki of >25
nM. Most analogues within this series were partial to
full agonists based on the TBPS assay. The two excep-
tions, 64 (antagonist) and 69 (inverse agonist), con-
tained bulky R3 groups. Biphasic properties were evi-
dent for three analogues (68-70), with isopropyl-69
displaying (like 47) a particularly large change in
chloride current between the different drug concentra-
tions. Biphasic effects were absent for 71 and 72 which
lack a basic nitrogen. Interestingly, 65-67 displayed
enhanced Cl- current with increased drug concentra-
tion, a property typically observed for BzR ligands.
Clearly, the biphasic properties observed for compounds
such as 47 and 69 are quite susceptible to modification
of the R3 substituent, in contrast to changes in the
A-ring or 3-position where biphasic effects are main-
tained, indicating that the distal piperazine nitrogen
may be involved with this unique property. Within this
series only 65, 68, and 71 were effective metrazole
antagonists. Interestingly, 67, which was nearly a full
agonist in vitro, was inactive in this assay. Of these
logues within this subseries were evaluated for intrinsic
efficacy in the R3â2γ2 subtype. Three analogues (45, 52,
58) had significantly less efficacy in the R3â2γ2 subtype
as compared to R1â2γ2, while the rest of the derivatives
evaluated showed no selectivity. Three analogues (46,
51, 55) were partial agonists in both subtypes.
The biphasic properties of several analogues within
this series were explored in depth. The dose-response
curves for 46 and 47 in the TBPS shift assay are shown
in Figure 3. Both compounds at low concentrations
reduced, and at high concentrations enhanced, [35S]-
TBPS binding. As previously reported for 47,39 the
agonist properties of this compound are blocked by Ro
15-1788, an antagonist at the benzodiazepine site. The
inverse agonist properties of 47 were not blocked by this
BzD antagonist, indicating that this and other related
quinoxaline piperazines also interact with a second site
on the GABAA receptor which is distinct from the
benzodiazepine site. Analysis of the binding data for 47
in the presence of Ro 15-1788 indicates a single low-
affinity site with an estimated Kd of 407 nM. Appar-
ently, when 47 occupies this low-affinity site it allo-
sterically influences the GABAA receptor, leading to a
reversal of agonist activity on the BzD site and inhibi-
tion of GABA-induced Cl- current, effectively limiting
its own agonistic actions. The bell-shaped dose-
response profile of 47, as monitored with GABA-induced
Cl- current, was also evident in the R3â2γ2 subtype as
shown in Figure 4. In the R6â2γ2 subtype, 47 had no
effect on Cl- current at low concentration. However, at
high concentrations in this subtype, the inverse agonist
properties manifested by this compound at the low-
affinity site were evident. Neither 46 nor 47 induced
seizures at high doses (>150 mg/kg).
To provide a comparison for the quinoxaline pipera-
zines, a small number of simple urea analogues were

1130 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7
J acobsen et al.
Ta ble 3. [3H]Fnz Binding, TBPS Shift, Cl- Current Changes, and Metrazole Antagonism Data for Imidazo[1,5-a]quinoxaline
Piperazine Ureas with Varying A-Ring Substitution
[
35S]TBPS
shiftb,c
∆Cl- currentb,c
R1â2γ2 5.0 µM
metrazoled
ED50 (95% confidence limits) mg/kg, ip
compd
NR2
NMe2
piperidine
morpholine
NMe2
R7
Ki (nM)a
59
60
61
62
63
H
H
H
Cl
Cl
5.25
1.35
2.47
5.56
2.56
0.15
0.08
0.00
0.51
0.53
0.09
0.00
0.14
0.35
0.38
>50
>50
>50
10.5 (5.2-21.4)
pyrrolidine
29.7
a
Mean binding affinity against [3H]flunitrazepam; see ref 28 and the Experimental Section for methods. The standard error was <(10%
b
of the mean. Diazepam is defined as a full agonist which gives a value of 1. Antagonists are defined as having a shift value of 0; partial
agonists are intermediate. c See the Experimental Section. The standard error was <(10% of the mean. Antagonism of metrazole-
d
induced clonic convulsions in the rat after ip injection; see the Experimental Section.
Ta ble 4. [3H]Fnz Binding, TBPS Shift, Cl- Current Changes, and Metrazole Antagonism Data for Imidazo[1,5-a]quinoxaline
Piperazine Ureas
Ki
[
35S]TBPS
shiftb,c
∆Cl- currentb,c
R1â2γ2 5.0 µM/0.5 µM
metrazoled
ED50 (95% confidence limits) mg/kg, ip
compd
R1
R3
(nM)a
64
65
66
67
68
69
70
71
72
H
H
H
H
Me
Me
Me
Me
Me
tBu
cPr
20.91
5.39
15.1
10.86
10.47
5.16
37.84
4.66
-0.03
0.76
0.32
0.81
0.15
-0.97
0.67
0.94
0.76
0.12/
50
0.75/0.42
0.83/0.35
1.00/0.80
0.50/0.94
-0.32/0.9
0.00/0.5
0.79/
12.5 (6.0-26.0)
>50
>50
3.7 (2.6-5.2)
35.3 (21.9-57.1)
42.0 (29.9-59.0)
14.8
2-pyridyl
CH2CF3
Et
iPr
CH2CF3
COCH3
SO2Me
4.18
0.90/
>50
a
Mean binding affinity against [3H]flunitrazepam; see ref 28 and the Experimental Section for methods. The standard error was <(10%
b
of the mean. Diazepam is defined as a full agonist which gives a value of 1. Antagonists are defined as having a shift value of 0; partial
agonists are intermediate. c See the Experimental Section. The standard error was <(10% of the mean. Antagonism of metrazole-
d
induced clonic convulsions in the rat after ip injection; see the Experimental Section.
analogues, only N-ethyl 68 was more effective than 45
in the metrazole assay.
nM. In the TBPS shift assay, these analogues (76-84)
ranged from partial inverse agonists (78 and 84) to
nearly full agonists (83). Results from the Cl- current
screen reflected a similar efficacy range with biphasic
properties observed for most analogues. Most of the
C-methylpiperazine derivatives were effective in the
metrazole screen, in contrast to unsubstituted 44, with
partial agonists 76, 77, and 79 particularly effective.
The stereochemistry of the appended C-methyl groups
often had significant effects on metrazole activity. In
particular with the 2,3-dimethylpiperazine analogues,
cis-analogue 79 was 25-fold more potent than the trans-
regioisomer 80, even though in vitro efficacy was nearly
identical for the two analogues.
Two rigid bicyclic piperazine ring systems were
evaluated (Table 6). All four analogues had reasonable
binding affinity (5-30 nM) and were partial agonists
as indicated by their TBPS shift ratios. Both N-methyl
analogues had greater efficacy than diazepam in the
chloride current screen, while the unsubstituted coun-
terparts were partial agonists. Biphasic properties were
noted for 86 and 87, indicating that the additional cyclic
As the C-methyl groups of 46 and related analogues
have a profound effect on in vitro efficacy and metrazole
antagonism as compared to unsubstituted 44, a thor-
ough SAR exploration of the regiochemistry of the
appended methyl groups was undertaken. Reversal of
the cis-dimethylpiperazine group provided reasonable
affinity for the 7-fluoro analogues 73 and 74 (Ki’s of 3.5
and 13 nM), while the 7-chloro analogue, 75, had
notably decreased binding affinity (Table 5). In vitro
efficacy increased slightly for 73 and 74 as compared
to the corresponding regioisomers (42 and 43), while the
7-chloro analogue, 75, had significantly increased in-
trinsic efficacy as compared to 47. The enhanced efficacy
for this substitution pattern was also evident from the
metrazole screening results, where all three analogues
were quite potent. All of the other urea analogues, which
had varying regiochemistry and number of C-methyl
groups, had excellent affinity. The only exception proved
to be 84 which, given the bulky bis-spirocyclohexylpip-
erazine, was reasonably well tolerated with a Ki of 79

Piperazine Imidazo[1,5-a]quinoxaline Ureas
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7 1131
Ta ble 5. [3H]Fnz Binding, TBPS Shift, Cl- Current Changes, and Metrazole Antagonism Data for C-Methyl
Imidazo[1,5-a]quinoxaline Ureas
[
35S]TBPS
shiftb,c
∆Cl- currentb,c
∆Cl- currentb,c
metrazoled ED50
compd
R1
R3 R7 Ki (nM)a
R1â2γ2 5.0 µM/0.5 µM R3â2γ2 5.0 µM (95% confidence limits) mg/kg, ip
73
74
75
76
77
78
79
80
81
82
83
84
cis-2,6-di-Me
cis-2,6-di-Me
cis-2,6-di-Me
3-Me
trans-3,5-di-Me
trans-2,5-di-Me
cis-2,3-di-Me
trans-2,3-di-Me
3,3-di-Me
H
Me
Me Cl
H
H
H
H
H
H
F
F
3.45
13.3
90.2
4.48
2.35
7.80
2.34
4.03
3.03
4.35
3.20
79.47
0.02
0.63
1.02
0.18
0.41
-0.63
0.27
0.07
-0.01
0.21
0.77
-0.38
0.0/0.72
0.75/
1.71/
10.5 (5.1-21.3)
8.8 (5.4-14.3)
6.2
0.69
Cl
Cl
Cl
Cl
Cl
Cl
0.17/0.5
0.8/0.60
-0.05/0.3
0.0/0.70
0.48/0.68
0.0/0.33
0.50/1.1
1.14/
8.8 (5.5-20.2)
10.5 (5.5-20.2)
17.6 (17.6-17.6)
0.70 (0.4-1.2)
17.7 (10.9-28.6)
>50
3,3-di-Me
3,3,5,5-tetra-Me
bis-spiro-cyclohexyl
Me Cl
H
H
35.3 (21.8-57.10
3.70 (2.2-6.3)
Cl
Cl
-0.70/0.1
>50
a
Mean binding affinity against [3H]flunitrazepam; see ref 28 and the Experimental Section for methods. The standard error was <(10%
b
of the mean. Diazepam is defined as a full agonist which gives a value of 1. Antagonists are defined as having a shift value of 0; partial
agonists are intermediate. The standard error was <(10% of the mean. c See the Experimental Section. Antagonism of metrazole-
d
induced clonic convulsions in the rat after ip injection; see the Experimental Section.
Ta ble 6. [3H]Fnz Binding, TBPS Shift, Cl- Current Changes, and Metrazole Antagonism Data for Imidazo[1,5-a]quinoxaline Ureas
[
35S]TBPS
shiftb,c
∆Cl- currentb,c
R1â2γ2 5.0 µM/0.5 µM
metrazoled ED50
(95% confidence limits) mg/kg, ip
compd
R3
Ki (nM)a
85
86
87
88
H
CH3
H
30.1
4.89
5.45
9.15
0.33
0.68
0.20
0.68
0.50/0.30
1.0/1.6
0.2/0.73
2.0/2.1
>50
7.4 (5.3-10.4)
>50
CH3
29.7 (17.7-49.9)
a
Mean binding affinity against [3H]flunitrazepam; see ref 28 and the Experimental Section for methods. The standard error was <(10%
b
of the mean. Diazepam is defined as a full agonist which gives a value of 1. Antagonists are defined as having a shift value of 0; partial
agonists are intermediate. c See the Experimental Section. The standard error was <(10% of the mean. Antagonism of metrazole-
d
induced clonic convulsions in the rat after ip injection; see the Experimental Section.
structure and the boat conformation of piperazine 86
are tolerated. Only 86 was an effective metrazole
antagonist.
enhanced affinity over the (S)-enantiomers. Greater
differences were noted for the trans-3,5-dimethylpip-
erazine ureas, with the (R,R)-analogues having 3-7-
fold enhanced affinity over the (S,S)-antipodes. The
TBPS shift ratios varied from antagonists to partial
agonists, with most enantiomeric pairs having similar
efficacy. The major exceptions proved to be the 7-methyl
analogues where both the 3-(S)-Me and 3,5-(S,S)-di-
methylpiperazines (93 and 95) had significantly en-
hanced intrinsic efficacy. Chloride current measure-
ments for these chiral analogues varied from antago-
nists to full agonists. While most enantiomers had
roughly equal efficacy, several exceptions were noted.
The 3,5-(S,S)-dimethylpiperazine analogues in both the
7-Cl (91) and 7-CH3 (95) series had 2-fold greater
efficacy (at 0.5 µM) than their antipodes. Biphasic
efficacy was also noted for most analogues. The major
exceptions proved to be 95 and the 7-CF3 analogues
(97-100). These compounds lacked biphasic effects and
were typically full agonists. Quinoxaline 91 had notably
Two piperazines from the C-methyl urea series (Table
5) were selected for further synthesis and pharmacologi-
cal evaluation in chiral form. The targeting of the
individual enantiomers of 76 and 77 was undertaken
because of their partial agonist in vitro properties and
anti-metrazole activity, as well as an interest to fully
evaluate the SAR of the methyl groups at the 3- and
5-positions (as in 46) of the piperazine ring system. The
A-ring substituent was varied to include 7-Cl, 7-Me, and
7-CF3, as several quinoxalines with these groups (i.e.,
46, 51, and 55) had excellent activity in the screening
assays. As shown in Table 7, analogues from the 7-Cl
and 7-Me series generally had good binding affinity (<25
nM) independent of the C-methyl stereochemistry. In
contrast, the 7-CF3 derivatives had decreased binding
affinity, with 99 having a Ki of 277 nM. In the 3-meth-
ylpiperazine series, the 3-(R)-analogues had equal or

1132 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7
J acobsen et al.
Ta ble 7. [3H]Fnz Binding, TBPS Shift, Cl- Current Changes, and Metrazole Antagonism Data for Chiral Imidazo[1,5-a]quinoxaline
Ureas
[
35S]TBPS
shiftb,c
∆Cl- currentb,c
metrazoled ED50
(95% confidence limits) mg/kg, ip
compd
R3
R5
R7
Cl
Cl
Cl
Ki (nM)a
R1â2γ2 5.0 µM/0.5 µM
89
90
91
92
93
94
95
96
97
98
99
100
3(S)-Me
3(R)-Me
3(S)-Me
3(R)-Me
3(S)-Me
3(R)-Me
3(S)-Me
3(R)-Me
3(S)-Me
3(R)-Me
3(S)-Me
3(R)-Me
H
H
6.71
5.96
16.7
5.13
17.9
5.47
25.6
5.17
42.82
23.21
276.8
37.13
0.12
0.01
0.37
0.56
0.59
0.19
0.77
0.10
0.54
0.60
0.41
0.68
0.10/0.8
0.0/0.6
0.4/1.05
0.05/0.5
0.4/1.1
0.10/0.8
1.3/1.3
0.05/0.7
0.70/0.8
1.0/1.0
1.2 (0.9-4.0)
3.7 (2.1-6.7)
7.4 (3.5-15.9)
1.3 (0.6-2.8)
25.0 (16.9-37.0)
2.7 (1.4-5.1)
7.4 (3.5-15.9)
6.3 (3.0-13.1)
25.0
5(S)-Me
5(R)-Me
H
Cl
Me
Me
Me
Me
CF3
CF3
CF3
CF3
H
5(S)-Me
5(R)-Me
H
H
8.9 (3.9-20.4)
39.7 (8.4-186.3)
4.4 (2.7-7.2)
5(S)-Me
5(R)-Me
1.00/0.5
0.8/0.7
a
Mean binding affinity against [3H]flunitrazepam; see ref 28 and the Experimental Section for methods. The standard error was <(10%
b
of the mean. Diazepam is defined as a full agonist which gives a value of 1. Antagonists are defined as having a shift value of 0; partial
agonists are intermediate. c See the Experimental Section. The standard error was <(10% of the mean. Antagonism of metrazole-
d
induced clonic convulsions in the rat after ip injection; see the Experimental Section.
decreased intrinsic activity in the R3â2γ2 subtype (0.00)
as compared to R1â2γ2 (0.40). Most of the chiral pipera-
zine ureas were effective metrazole antagonists with
only the low-affinity analogues displaying correspond-
ingly higher ED50’s. Enantiomeric pairs often differed
slightly in potency in this assay. However, this dis-
similarity can be easily be rationalized by differences
in binding affinity and intrinsic activity as detailed
above.
On the basis of the results of the in vitro (binding,
TBPS, chloride current) and metrazole assays, a number
of the most promising compounds were evaluated
further for in vivo efficacy, BzD antagonism, and typical
benzodiazepine-type side effects (Table 8). Analogues
were also evaluated in an ex vivo binding assay to
determine CNS bioavailability and the duration of
action after oral dosing. This ex vivo assay has been
shown to be useful in evaluating these parameters for
a number of BzD ligands.40a,b In this screen, specific
binding of the test drug, as determined through a (3H)-
FNZ binding assay of brain homogenate after various
periods of drug administration, is compared to the in
vitro binding of the desired analogue. This ratio is
directly related to the concentration of drug and its
active metabolites in the brain. Within this series, the
ex vivo assay proved to be predictive (with the caveat
of including active metabolites) in identifying com-
pounds with good oral activity.
Quinoxalines with gem-4,4-dimethyl substitution (57,
58) or N4′-substitution with a methyl, ethyl, or isopropyl
group (47, 68, 69) also had excellent activity in the ex
vivo assay. However, trifluoroethyl, acetyl, or sulfonyl
substitution (70-72) in the dimethylpiperazine series
led to dramatically decreased apparent oral bioavail-
ability. Reversal of the cis-dimethylpiperazine (73-75)
resulted in rapid clearance from the CNS, although 73
and 74 had acceptable ex vivo binding. All of the simple
piperazines (R1 ) R2 ) H) containing various NR3
substituents had poor ex vivo activity (or were cleared
rapidly) further highlighting the unique properties of
the 3,5-methylated piperazine group. Of the C-meth-
ylpiperazine regioisomers evaluated, the 3-methyl- (76),
trans-3,5-dimethyl- (77), and 2,3-dimethylpiperazine
isomers (79, 80) had reasonable ex vivo binding with a
long duration of action. Like the cis-2,6-dimethylpip-
erazines, the gem-3,3-dimethylpiperazines 81-83 were
cleared rapidly. None of the bicyclic analogues were
particularly effective in this assay. Evaluation of the
chiral 3-methyl and trans-3,5-dimethyl analogues in the
ex vivo assay resulted in a similar SAR to that observed
for 46 and related analogues. In particular, the 7-chloro
analogues had excellent CNS bioavailability while the
7-CH3 and 7-CF3 analogues were notably inferior.
Analogues having reasonable oral activity and dura-
tion of action in the ex vivo screen were also evaluated
in a stress-induced cyclic 3′,5′-guanosine monophos-
phate (cGMP) assay. Acute stress by electroconvulsive
shock, electric footshock, immobilization, or forced
swimming in ice-cold water has been shown to increase
cGMP levels.41,42 Numerous BzD anxiolytic agents are
effective in lowering the levels of cGMP after electric
footshock stress.41b,42 This biochemical model of stress
has proven to be sensitive and useful in predicting anti-
stress and anxiolytic activity of many standard anx-
iolytic agents. For example, diazepam and 2 were both
potent, lowering the cGMP to control levels (80/182 and
The observed CNS availability of the piperazine ureas
was superior to similar urea analogues (dimethylamino,
piperidine, morpholine) which are lacking in a basic
nitrogen. For instance, piperidine 60 had only 44% ex
vivo binding as compared to 82% for 40. In general, the
cis-3,5-dimethylpiperazine substituent provided 70-
95% ex vivo binding, with or without an N-methyl
substituent. Interestingly, the 7-Cl, 7-F, and unsubsti-
tuted analogues had superior ex vivo binding as com-
pared to those with 6-Cl, 7-Me, or 7-CF3 groups.

Piperazine Imidazo[1,5-a]quinoxaline Ureas
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7 1133
Ta ble 8. Benzodiazepine Antagonism, ex Vivo Binding, cGMP, and Acute Physical Dependence Data for Imidazo[1,5-a]quinoxalines
acute electroshock
oral ex vivo binding assayb
max % inhibition
(time in min for peak effect)
cerebellar GMP
assayc % control
((drug + stress)/stress)
BZD antag. tractiond ED50
(95% confidence interval)
mg/kg
physical dependence activitya
compd
40
(drug MA50/V122 MA50
)
82 (60)
95 (60)
94 (60)
29 (30)
71 (240)
86 (120)
44 (30)e
49 (60)
148/157
126/171
131/171
152/162
131/205
70/195
>30
42
43
45
46
47
48
49
50
51
52
55
56
57
58
60
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
85
86
88
89
90
91
92
93
94
95
96
97
98
99
100
diazepam
1
>30
21.8 (10.6-44.8)
A (14.4/20.5)
A (18.0/21.3)
IA (18.1/17.7)
11 (5.0-22.0)
>30
7.0 (3.0-18.0)
162/171
176/171
>30
IA (18.1/19.2)
A (18.2/25.8)
A (16.5/19.7)
A (15.9/21.3)
IA (19.2/19.2)
A (17.3/21.0)
A (13.9/18.8)
A (11.5/17.7)
IA (19.2/21.2)
IA (20.8/21.2)
A (20.9/24.0)
IA (21.9/23.8)
A (17.9/24.0)
IA (25.2/25.2)
IA (19.2/19.9)
IA (19.2/18.4)
IA (24.9/23.8)
A (18.9/21.9)
IA (18.9/19.7)
>30
>30
55 (240)
64 (30)
52 (120)
17 (240)
94 (60)
84 (30)
44 (60)
115/157
93/162
73/205
>30
>30
>30
21.8 (10.6-44.8)
63/206
79/206
178/195
174/166
>30
4 (2-9)
14 (7-28)
>30
62 (30)
24 (240)
3.4 (120)
10 (240)
84 (30)
85 (30)
19 (240)
22 (60)
8 (120)
69 (30)e
68 (30)e
32 (30)e
68 (240)
63 (60)
71 (60)
78 (60)
55 (30)
13 (30)e
11 (30)e
47 (30)e
4 (60)
11 (60)
26 (30)
74 (30)
75 (30)
45 (60)
62 (120)
32 (60)
51 (240)
17 (120)
22 (30)
21 (120)
30 (240)
>30
>30
>30
180/205
114/220
>30
>30
>30
>30
>30
173/171
141/171
204/205
>30
>30
IA (18.5/17.7)
A (15.6/23.8)
>30
>30
17.3 (7.9-37.9)
>30
A (17.7/19.7)
A (15.4/18.4)
A (21.9/25.8)
IA (24.0/24.9)
IA (25.6/23.8)
A (25.8/27.7)
IA (21.6/24.0)
154/220
67/165
>30
35 (19.4-61.6)
>30
>30
>30
>30
A (18.5/24.9)
A (19.4/25.6)
A (16.1/23.6)
IA (24.6/25.2)
A (19.2/25.2)
A (19.9/26.9)
A (20.7/23.5)
A (20.7/24.0)
>30
100/177
104/177
124/179
>30
27.5 (15.4-48.9)
>30
>30
102/183
108/183
152/193
86/173
162/198
159/180
>30
17.3 (7.9-37.9)
>30
IA (23.0/23.6)
A (17.9/26.9)
A (22.6/26.9)
>30
>30
>30
24 (30)
167/180
80/182
119/195
137/158
>30
Af (18.2/25.4)
IA (23.2/24.9)
IA (21.2/22.2)
>30
1.0
2
1.7 (0.8-3.8)
a
b
A, p < 0.05; IA, no significant activity; see the Experimental Section. See the Experimental Section. The standard error was <( 5%
of the mean. c See the Experimental Section. The standard error was <( 10% of the mean. See the Experimental Section. e Low levels
d
(<10%) of the test compound detected at 60 min. f Tested at 15 mg/kg.
119/195 of (drug + stress)/stress response, respectively).
In this assay, the levels of cGMP (baseline ) 100%) were
measured in mice receiving drug after electric foot shock
stress. These levels were compared to a control group
which received vehicle prior to the shock treatment. The
resultant values are expressed as percent of control for
(drug + stress)/stress animals, respectively.
A significant number of the quinoxaline piperazines
were effective in lowering cGMP levels after applied
stress. This is in sharp contrast to the piperidine and
pyrrolidine analogues (60 and 63, respectively) which
were inactive. Within the cis-3,5-dimethylpiperazine
series, a bulky substituent at the 7-position of the A-ring
(Cl, Me, CF3) was required, with 46 and 55 particularly
effective. The unsubstituted and 6-chloro analogues
were inactive, while the 7-fluoro quinoxalines were
weakly active, paralleling their limited anti-metrazole
activity. An N-alkyl substituent was also tolerated in
this ring system, with 47, 52, and 69 having excellent
anti-stress activity, although the N-ethyl derivative 68
was inactive. The two gem-4,4-dimethyl full agonist
analogues (57 and 58) were also extremely effective.
Both the cis-2,6-dimethylpiperazines and the only simple
piperazine tested, 45 (which lacked a C-methyl group),

1134 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7
J acobsen et al.
Ta ble 9. Vogel’s Punished Licking Assay
were ineffective. Regioisomeric methyl groups were also
tolerated on the piperazine ring with cis-2,3-dimeth-
ylpiperazine (79) and 3,3,5,5-tetramethylpiperazine (83)
analogues effective. In the chiral piperazine series, all
analogues tested with a 7-Cl or 7-Me group were active
(except 95), whereas the 7-CF3 derivatives were inef-
fective. This latter result was quite surprising given the
potent anti-stress properties of 55. A good correlation
between anti-stress activity in the cGMP assay to
metrazole antagonism properties and ex vivo binding
was noted for most analogues. In particular, those
compounds which were potent in the metrazole assay
and which also had reasonable CNS availability by ex
vivo assay were generally effective in the cGMP screen.
Exceptions were 69, which was a weak metrazole
antagonist yet was potent in the cGMP assay, and the
7-methyl analogues (93-96), which typically performed
poorly by ex vivo assay but were effective in the cGMP
screen. Consistent with the above, compounds which
had weak metrazole activity or poor ex vivo binding
were typically ineffective in the cGMP assay. Several
apparent exceptions (68, 73-75) most likely result from
rapid clearance from the CNS (as determined by the ex
vivo assay).
mean no.
of shocksa
compd
dose (mg/kg)
n
46
0
9
3.8 ( 0.3
5.3 ( 0.8
3.1 ( 0.3
5.7 ( 0.8
12 ( 2.6**
19 ( 4.1**
8.4 ( 3
0.3
1.0
3.0
10
10
10
10
10
10
10
10
9
10.0
12 mg/kg CDPZ
0
0.3
1.0
3.0
47
55
69
91
8 ( 2
7 ( 3
18 ( 4*
10.0
6
10
8
8
8
9
9
9
8
9
10
9
9
8
18 ( 5
3 mg/kg diazepam
0
0.3
1.0
3.0
10.0
12 mg/kg CDPZ
0
0.3
1.0
3.0
10.0
12 mg/kg CDPZ
0
13 ( 3
4.9 ( 1.2
4.6 ( 0.9
4.9 ( 0.8
6.4 ( 0.9
7.9 ( 1.2*
23 ( 4**
6.6 ( 0.7
6.3 ( 1.8
5.3 ( 0.6
5 ( 1
3.4 ( 0.4**
26 ( 4.8
3.3 ( 0.4
3.9 ( 1
5.2 ( 1.2
5.6 ( 1.2
8.8 ( 2**
14.8 ( 3.9**
9
0.3
1.0
3.0
10.0
10
10
10
10
10
Most of the compounds were also evaluated for their
ability to antagonize the muscle relaxation effects of a
benzodiazepine full agonist, triazolam. Traction was
assessed 30 min after separate administration of tri-
azolam and the test compound. The antagonist fluma-
zenil, at 2.0 mg/kg, was effective in blocking the muscle-
relaxing effects of triazolam, as were partial agonists 1
(1.0 mg/kg) and 2 (1.7 mg/kg). Traditional full agonist
benzodiazepine ligands are not effective in blocking
muscle relaxation in this assay. In general, the qui-
noxaline piperazines were not benzodiazepine antago-
nists, with only 45 and 47 having reasonably low ED50’s
(11.0 and 7.0 mg/kg, respectively). The two nonpipera-
zine analogues (60, 63) evaluated were also BzD an-
tagonists. Derivatives from this series, like those re-
ported in the 3-phenyl classes,15 display little correlation
between intrinsic efficacy and BzD antagonism. This
discrepancy apparently is not due to poor CNS penetra-
tion or extensive metabolism to inactive metabolites, as
many analogues displayed excellent activity by ex vivo
assay.
12 mg/kg CDPZ
a
Single asterisk (*) indicates p < 0.05 and double asterisk (**)
p < 0.01 compared to “0” group in that experiment using Wilcoxon
rank sum test.
assessed for electroshock seizure thresholds (MA50
compared to vehicle MA50), which typically were lower
for mice undergoing flumazenil-precipitated withdrawal.
Standard benzodiazepine ligands such as diazepam
were active in this paradigm at doses as low as 15 mg/
kg/day, while partial agonists 1 and 2 were not.
Many of the compounds evaluated in the acute
electroshock physical dependence assay were active,
regardless of their in vitro efficacy. Quinoxalines 57 and
58 were particularly active. The two nonpiperazine
analogues (60, 63) were inactive. Not surprisingly,
analogues which had poor ex vivo binding properties
were usually inactive, although exceptions such as 66
were noted. Nonetheless, the correlation was quite good,
especially considering the different dosing regimes.
Importantly, several compounds (47, 55, 69, 91) were
identified which had reasonable properties in the me-
trazole, cGMP, and ex vivo assays, yet lacked physical
dependence. Another extremely interesting result from
the physical dependence assay was the enantiomeric
pair 91 and 92. While 91 had slightly decreased affinity
and ex vivo binding as compared to its enantiomer (92),
the lack of physical dependence for this analogue is
quite surprising and indicates that minor structural
changes can have an impact on overt side effects.
On the basis of their efficacy and side effect profile,
several analogues (47, 55, 69, and 91) were selected for
evaluation in Vogel’s punished licking assay.9a Even
though it produced a moderate degree of physical
dependence, 46 was included due to its close relation-
ship to some of the lead compounds. As shown in Table
9, 47 was the most effective compound, significantly
active at 3 mg/kg (and nearly so at 10 mg/kg). In this
Physical dependence is one of the more undesirable
side effects typically promoted by benzodiazepine full
agonists. Clinically, manifestations of physical depen-
dence are undesirable, as they are similar to the
disorder being treated.43 The degree of acute physical
dependence was assessed through an electroshock sei-
zure assay.43 Withdrawal from chronic treatment with
high doses of benzodiazepines may in some cases result
in signs and symptoms related to hyperexcitability of
the CNS. In animals, such changes can be quantified
by determining the current threshold (MA50) to elicit
electroshock-induced seizures. The lowering of the elec-
troshock threshold precipitated by a benzodiazepine
antagonist after an acute regimen of test compound was
used to quantify the development of physical depen-
dence. Mice received the test drug for 3 days (150 mg/
kg/day). Twenty-four h after the last administration, the
mice received an iv injection of a benzodiazepine an-
tagonist (flumazenil). Five minutes later, the mice were

Piperazine Imidazo[1,5-a]quinoxaline Ureas
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7 1135
Ta ble 10. Geller-Cook Activity of 46 and 47
first-pass metabolism was minimized through the 3,5-
dimethylpiperazine substitution.
treatment dose (mg/kg)
n
punished ratea nonpunished ratea
Molecu la r Con for m a tion s. The low-energy con-
formers of a number of compounds were determined by
molecular mechanics methods. Calculations were per-
formed using both the AMBER* and MM2* force fields,
which were developed for the MacroModel44 system of
programs as extensions of the original AMBER45 and
MM246 parameter sets. Initial structures to be mini-
mized were generated by means of an internal coordi-
nate Monte Carlo procedure47 in which torsional angles
around all rotatable bonds were allowed to vary. In a
typical calculation, 100-200 initial structures were
minimized, and duplicate minimized structures were
eliminated to give a final set of conformers. Energy-
minimized structures were also obtained from semi-
empirical quantum mechanical calculations using the
MOPAC program48 with the AM149 parameter set. The
latter method was applied selectively, using the four or
five lowest-energy conformers obtained from the molec-
ular mechanics studies as initial structures.
Figure 5 shows face and edge views of two low-energy
structures for 46, obtained with the MM2* force field.
Structures A and B are identical except for the orienta-
tions of the 5-substituent, which differ by a 180° rotation
about the N5-C bond. Both structures contain a planar
region composed of the imidazoquinoxaline ring and
3-carboxylate moiety, with the tert-butyl and piperazine
functional groups occupying different out-of-plane re-
gions. Calculations using both the AMBER* force field
and the MOPAC/AM1 method yielded structures with
no significant geometrical differences from those ob-
tained with MM2*. It should be noted that for each
structure shown there exists a pair of mirror image
isomers of identical energy which differ geometrically
due to different puckering of the imidazoquinoxaline
ring.
In both the present and previous computational
studies of imidazoquinoxaline ureas,13,15 the identity of
the lowest-energy conformer could not be unequivocally
determined. In the case of 46, MM2*-based calculations
predicted structure A to be most stable, with structure
B higher in energy by 3.25 kcal/mol. Moreover, three
other conformers not shown in Figure 5 were found with
energies between those of structures A and B. Specifi-
cally, structures differing from A by (a) a 180° rotation
of the ester group about the C3-C bond, (b) a 180°
rotation of the carbonyl-N-piperazine bond (N1′-CO),
and (c) rotations a and b combined were obtained with
respective energies 0.37, 2.16, and 2.51 kcal/mol higher
than that of structure A.
In contrast, both the AMBER*- and AM1-based
calculations predicted structure B to be more stable than
A by 0.93 and 2.09 kcal/mol, respectively. Furthermore,
according to the AMBER*-based studies, a 180° rotation
of the ester group as in rotation “a” above did not result
in any energy increase, while the piperazine rotamer
equivalent to rotation “b” above was higher in energy
relative to structure B by approximately 0.84 kcal/mol.
control
46
46
control
47
47
0
12
6
6
17
6
6
23 ( 2
35 ( 2
47 ( 12
43 ( 11
34 ( 4
38 ( 5
56 ( 19
36 ( 6
1.0
3.0
0
0.3
1.0
3.0
46 ( 16*
27 ( 9
18 ( 4
22 ( 12
53 ( 20*
65 ( 16*
47
6
a
Asterisk (*) indicates p < 0.05 compared to own control using
Wilcoxon signed rank test.
particular screen 47 was more effective than 3.0 mg/kg
of diazepam. Three of the other analogues (46, 55, 91)
were also active (10 mg/kg), although they were less
effective than 12 mg/kg of chlordiazepoxide. Interest-
ingly, 69 significantly decreased the number of shocks
taken at 10 mg/kg (inactive at lower doses), possibly
consistent with its potent inverse agonist properties
noted at 5 µM in the chloride current assay. Two of
these analogues (46, 47) were also evaluated in the
Geller-Cook conflict assay,9a with data provided in
Table 10. As shown, both of these quinoxalines were
active on punished schedule components at 1 mg/kg
with 47 also effective at 3 mg/kg. Consistent with their
partial agonist profile, neither analogue was active on
nonpunished components.
The pharmacokinetics and oral bioavailability of the
four compounds active in the Vogel assay were evalu-
ated in rats, with data shown in Table 11. Among these
compounds, 47 had the highest total plasma clearance
(2.71 L/h/kg) and the shortest terminal elimination half-
life, averaging 0.57 h after iv administration. Serum
concentrations of 47 in rats after oral administration
were all below the limit of quantitation. A major
metabolite with much higher serum concentration than
the parent was observed after oral administration of 47
to rats and was identified as the N-demethylated
product of 47, namely 46. These data suggest that 47
is subject to metabolic clearance and is not orally
bioavailable due to extensive first-pass metabolism in
rats. Quinoxaline 46 was much more metabolically
stable in rats, with a 2-fold lower total plasma clearance
and 2-fold longer half-life as compared to 47. The oral
bioavailability of 46 in rats was nearly 100%, showing
limited first-pass metabolism of this compound after
oral dosing. The trans isomer of 46 (91) had lower
plasma clearance and a longer half-life in rats than 46.
However, the oral bioavailability of 91 was only 54%,
suggesting a more significant first-pass effect of 91 after
oral administration. Quinoxaline 55, with a 7-trifluo-
romethyl substituent replacing the chloro of 46, had the
lowest plasma clearance (0.73 L/h/kg) and longest half-
life (2.4 h) in rats. The oral bioavailability of 55 averaged
84%. These data suggest that the trifluoromethyl re-
placement further improved the in vivo metabolic
stability. The evaluation of the pharmacokinetics of
pyrrolidine urea 63 was also undertaken (po dosing
only) to provide a comparison to the piperazine ana-
logues. As indicated in the table, this analogue had a
significantly lower Cmax (0.33 mg/mL) as compared to
46 (1.17 mg/mL), although it had a longer t1/2 (7.6 h). A
major metabolite was observed for 63. The concentration
ratio (metabolite/parent) was approximately 0.5, indi-
cating significant first-pass metabolism for 63. This
The structures and relative energies of the majority
of the other piperazine imidazoquinoxaline ureas were
expected to be very similar to those described above for
46. This was found to be the case in selected studies on
39, 47, and piperidine analogue 60. With the 6-methyl

1136 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7
J acobsen et al.
Ta ble 11. Selected Pharmacokinetic Parameters following Intravenous (iv) and Oral (po) Administration of 46, 47, 55, 63, or 91 in a
Solution Formulation to the Male Sprague-Dawley Rats
a
compd
route
n
dose (mg/kg)
Cmax (µg/mL)
tmax (h)
AUC (µg‚h/mL)
t1/2 (h)
Clp (L/h/kg)
Fb (%)
46
iv
po
iv
po
iv
po
po
iv
4
4
3
3
4
4
3
4
4
5.10 ( 0.21
5.22 ( 0.15
4.74 ( 0.29
4.75 ( 0.05
4.97 ( 0.03
5.03 ( 0.07
5.46 ( 0.09
5.06 ( 0.06
5.01 ( 0.11
3.54 ( 0.81
3.57 ( 0.99
3.46 ( 3.30
c
7.47 ( 2.72
6.43 ( 5.34
3.68 ( 0.54
5.01 ( 1.21
2.62 ( 1.08
1.2 ( 0.3
1.8 ( 0.3
0.57 ( 0.22
c
2.4 ( 0.3
2.7 ( 0.6
7.6 ( 1.4
1.5 ( 0.2
2.4 ( 0.7
1.51 ( 0.39
1.17 ( 0.49
0.5-4
101
c
47
55
2.71 ( 1.85
c
0.73 ( 0.28
c
c
1.16 ( 0.83
0.33 ( 0.07
1-2
1-2
84
63
91
1.07 ( 0.32
po
0.66 ( 0.14
0.25-0.5
54
a
b
Data presented are the value range. Calculated using average AUC. c Compound 47 serum concentrations below the lower limit of
quantitation (50 ng/mL).
F igu r e 6. Elements of the Wermuth and Gardner pharma-
cophoric models: (a) aromatic region, (b and c) hydrogen-
acceptor site points δ1 and δ2, (d) freely rotating substituent,
and (e) out-of-plane substituent.
Region d has been characterized as containing a
“freely-rotating aromatic group” by Wermuth50 or an
ester group according to Gardner,4a and the present tert-
F igu r e 5. Molecular structures of relevant low-energy con-
formers of 46. Hydrogen atoms are not shown.
butyl ester corresponds to the latter case. In previous
studies of 3-phenyl imidazoquinoxaline analogues,15 it
analogue, 53, calculated structures and energetics were
was shown that BzR binding affinity could be substan-
tially reduced by forcing the 3-phenyl substituent into
an out-of-plane orientation relative to the imidazoqui-
noxaline ring. In the present analogues, the 3-carboxy-
late group was found to be in-plane in all of the
calculated structures, which is consistent with the high
affinity generally observed. However, the receptor will
tolerate some out-of-plane character in this region, as
exemplified by the tert-butyl group shown in Figure 5.
As described above in detail for 46, the piperazine
urea substituent in the present analogues is predicted
to assume a number of possible orientations, all of which
may be characterized as out-of-plane. An out-of-plane
substituent in region e of Figure 6 has been associated
with full agonist activity according to Wermuth et al.,50
which may explain how minor changes with the pip-
erazine substituent can dramatically modulate efficacy.
The aryl substituents, R6 or R7, which have only a
minimal effect on the urea conformation, presumably
modulate efficacy through electronic effects. In sum-
mary, the quinoxaline piperazine ureas described herein
are consistent with a model in which an aromatic region
and one or perhaps two hydrogen-acceptor sites allow
for receptor recognition (affinity), while efficacy is
largely controlled by the cooperative effects of the
aromatic R6 or R7 substituent and the out-of-plane R5
urea.
also similar to those described for 46 although the
piperazine substituent in 53 was positioned somewhat
more out-of-plane than that in 46, due to increased
puckering in the imidazoquinoxaline ring. For the gem-
dimethyl analogue, 57, the calculated structures re-
mained similar to those of 46. However, in contrast to
46, all three computational methods predicted structure
A to be more stable than structure B, with energy
differences of 6.67, 3.35, and 0.77 kcal/mol from the
MM2*-, AMBER*-, and AM1-based calculations, respec-
tively. This preference of structure A over B has been
found in other gem-dimethyl imidazoquinoxalines as
well.13
P h a r m a cop h or e Mod els. As discussed in previous
publications,13,15 the imidazoquinoxaline series of ben-
zodiazepine receptor ligands (BzR) has been shown to
possess a number of features consistent with general
models of BzR recognition and activation given by
Wermuth et al.50 and Gardner.4a These models hold for
the present analogues as well. Referring to Figure 6,
the fused phenyl ring and divalent nitrogen (N2) of the
imidazoquinoxaline ring system correspond, respec-
tively, to the aromatic group and hydrogen-acceptor site
δ1, considered to be essential for receptor recognition.
A second H-acceptor site δ2 may correspond to the R5
carbonyl group, provided that it is oriented as in
structure A in Figure 5. In the Wermuth model, the
distance between δ1 and δ2 was given as 3.2 Å. However,
in an alternative model, Cook51 has suggested that this
distance be extended to as large as 6.5 Å, a value
consistent with the present structures.
Con clu sion
Compounds within the imidazo[1,5-a]quinoxaline pip-
erazine series are high-affinity ligands for the GABAA/

Piperazine Imidazo[1,5-a]quinoxaline Ureas
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7 1137
times, dry the organic layers with the indicated drying agent,
filter off the drying agent, and remove the solvent using a
rotary evaporator under reduced pressure. Tetrahydrofuran
(THF) and ether were distilled from sodium and benzophenone.
Dichloromethane was distilled from calcium hydride, and DMF
was dried over 3 Å molecular sieves. All other solvents were
EM Science HPLC grade, distilled in glass. Diethyl chloro-
phosphate and potassium tert-butoxide (1.0 M in THF) were
purchased from the Aldrich Chemical Co., Milwaukee, WI.
Phosgene in toluene (CAUTION: phosgene is highly toxic and
should be used with extreme care) was purchased from Fluka
Chemie AG or Columbia. All reactions were run under
nitrogen or argon.
benzodiazepine receptor. Through manipulation of the
substituents at the 5-, 6-, and 7-positions, analogues of
widely varying intrinsic efficacy were identified. Typi-
cally, the combination of a 3,5-dimethylpiperazine urea
with a bulky 7-substituent provided analogues with
partial agonist in vitro activity. Furthermore, this
substitution pattern generally resulted in excellent
activity in a number of in vivo models measuring
anxiolytic and anticonvulsant activity, including the
metrazole and cGMP assays. Additionally, several ana-
logues (46, 47, 55, 91) were also found to be active in
the Vogel punished licking assay. Most derivatives from
this series were evaluated in an electroshock physical
dependence screen with a number of the quinoxalines
identified which did not produce this undesired BzD side
effect. The degree of CNS penetration, as well as the
duration of action in brain, was determined for a
number of these derivatives by ex vivo assay, with the
cis-3,5-dimethylpiperazine ureas typically providing for
superior activity. Pharmacokinetic evaluation of five of
the more interesting quinoxalines confirmed the results
of the ex vivo assay, with 46 and 55 having 100% and
84% bioavailability, respectively. From the combination
of these assays, 55 and 91 were identified as partial
agonists which were effective in models of anxiety, had
excellent pharmacokinetics, and yet were consistent
with the partial agonist hypothesis by producing no
major BzD side effects.
The biphasic nature expressed by many of the qui-
noxaline piperazine ureas, as determined by in vitro
analysis, sets this series apart from other BzD ligands.
The dual functionality observed for 46, 47, and 91 is,
to our knowledge, unprecedented among BzD ligands.
Therapeutically, compounds within this class could be
unique due to their ability to limit their own agonistic
actions. This property should effectively control the
agonistic activity of these and related piperazine ureas
over a wide range of doses and may serve to limit their
abuse potential. Further pharmacological evaluation of
selected analogues from this series will be reported in
due course.
7-Ch lor o-4,5-d ih yd r oim id a zo[1,5-a ]qu in oxa lin e-3-ca r -
boxylic Acid 1,1-Dim eth yleth yl Ester (6c). Potassium tert-
butoxide (60.0 mL, 60.0 mmol, 1.0 M in THF) was added to a
solution of 6-chloro-1,2,3,4-tetrahydroquinoxalin-2-one13 (5c;
10.0 g, 54.8 mmol) and THF (180 mL) at -20 °C. The solution
was allowed to warm to 0 °C over 30 min and was cooled to
-40 °C. Diethyl chlorophosphate (9.00 mL, 62.3 mmol) was
added and the solution allowed to warm to room temperature
over 30 min. The solution was cooled to -78 °C, and a solution
of the tert-butyl isocyanoacetate (9.00 g, 63.8 mmol) and THF
(15.0 mL) was added. Potassium tert-butoxide (60.0 mL, 60.0
mmol) was then added dropwise over several minutes. The
solution was stirred at -78 °C for 30 min and then allowed to
warm to room temperature over 2 h. After stirring at room
temperature for 3 h, aqueous workup (ethyl acetate, MgSO4),
concentration, and trituration of the residue with hexane/ethyl
acetate gave 9.42 g of 6c as a mauve powder. Purification of
the filtrate by flash chromatography (1:1 ethyl acetate:hexane)
and trituration (ether/hexane) gave an additional 1.27 g (10.7
g total, 64%) of product (mp 261-263 °C): IR (mineral oil)
1683, 1522, 1367, 1323, 1302, 1254, 1154, 1057, 968 cm-1; 1H
NMR (300 MHz, CDCl3) δ 7.93 (s, ArH), 7.30 (d, J ) 8.4 Hz,
ArH), 6.75-6.85 (m, ArH), 6.79 (s, ArH), 4.81 (s, NCH2), 4.13
t
(br s, NH), 1.61 (s, Bu); MS (EI) m/e 305, 249, 231, 203.
Exa m p le P r oced u r e for th e Syn th esis of Ur ea 10 fr om
Am in e 6. Meth od A: 7-Ch lor o-5-[(1-(cis-3,5-d im eth yl)-
p ip er a zin o)ca r bon yl]-4,5-d ih yd r oim id a zo[1,5-a ]qu in oxa -
lin e-3-ca r boxylic Acid 1,1-Dim eth yleth yl Ester (46). Tri-
phosgene (643 mg, 2.17 mmol) was added to a mixture of 6c
(1.20 g, 3.92 mmol), CH2Cl2 (42 mL), and diisopropylethyl-
amine (0.80 mL, 4.6 mmol) at 0 °C. The mixture was stirred
for 1 h at 0 °C and then allowed to warm to room temperature.
After stirring for 3 h, the resultant solution was cooled to 0
°C. Diisopropylethylamine (0.80 mL, 4.6 mmol) and cis-2,6-
dimethylpiperazine (766 mg, 6.71 mmol) were added succes-
sively. The solution was stirred at 0 °C for 1 h and at room
temperature for 16 h. Basic workup (CH2Cl2, NaHCO3, MgSO4)
and purification by flash chromatography (5:1 ethyl acetate:
MeOH) gave 1.59 g of the urea as a foam. Crystallization from
ether/hexane provided 1.39 g (80%) of 46 as a tan powder (mp
200.5-201 °C): IR (mineral oil) 1689, 1643, 1512, 1424, 1316,
1306, 1159, 1154, 967 cm-1; 1H NMR (300 MHz, CDCl3) δ 7.99
(s, ArH), 7.45 (d, J ) 8.6 Hz, ArH), 7.10-7.20 (m, ArH, 2 H),
4.94 (s, ArNCH2), 3.67 (d, J ) 13.8 Hz, 2 H), 2.80-2.95 (m, 2
Exp er im en ta l Section
Ch em istr y. Thin-layer and flash chromatography utilized
E. Merck silica gel (230-400 mesh). Melting points were taken
on a Thomas-Hoover capillary melting point apparatus and
are uncorrected. Mass spectra, infrared spectra, and combus-
tion analysis were obtained by the Physical and Analytical
Chemistry Department of the Upjohn Co. 1H NMR spectra
were recorded at 300 MHz with a Bruker Model AM-300
spectrometer.
In cases where synthetic intermediates or products were
isolated by “aqueous workup (organic solvent, drying agent)”,
the procedure was to quench the reaction with H2O, dilute with
the indicated organic solvent, separate the organic layer,
extract the aqueous layer several times with the organic
solvent, dry the combined organic layers with the indicated
drying agent, filter off the drying agent, and remove the
solvent using a rotary evaporator at reduced pressure. When
“basic workup (organic solvent, aqueous basic solvent, drying
agent)” is indicated, the procedure was similar to aqueous
workup, except the indicated aqueous base was used instead
of H2O. When “acidic workup (organic solvent, organic solvent,
drying agent)” is indicated, the procedure was to dilute the
reaction mixture with the first indicated solvent, extract the
organic solution several times with 10% HCl, basify the
combined acidic layers with solid KOH, extract the basic
solution with the second indicated organic solvent several
t
H), 2.48 (apparent t, J ) 11.5 Hz, 2 H), 1.63 (s, Bu), 1.03 (d,
J ) 6.3 Hz, CHCH3, 6 H); MS (EI) m/e 445, 389, 318, 306,
276, 248, 230, 141. Anal. (C22H28N5O3Cl‚(C4H8O2)1/4) C, H, N,
Cl.
Gen er a l R ed u ct ive Am in a t ion P r oced u r e for 11.
Meth od B: 7-Ch lor o-4,5-d ih yd r o-5-[(1-(cis-3,5,4-tr im eth -
yl)p ip er a zin o)ca r bon yl]im id a zo[1,5-a ]qu in oxa lin e-3-ca r -
boxylic Acid 1,1-Dim eth yleth yl Ester (47). Sodium cy-
anoborohydride (351 mg, 5.59 mmol) was added to a mixture
of 46 (0.859 g, 1.93 mmol), MeOH (12 mL), and formalin (2.8
mL). The solution was stirred for 16 h at room temperature
and was concentrated. Basic workup (CH2Cl2, NaHCO3, Mg-
SO4) gave a foam which was crystallized from ether/hexane
to give 0.813 g (two crops, 92%) of 47 as a tan powder (mp
172-173 °C): IR (mineral oil) 1694, 1688, 1669, 1505, 1274,
1
1153, 965 cm-1; H NMR (300 MHz, CDCl3) δ 7.99 (s, ArH),
7.44 (d, J ) 8.6 Hz, ArH), 7.19 (s, ArH), 7.13 (dd, J ) 7.7, 1.5

1138 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7
J acobsen et al.
Hz, ArH), 4.94 (s, ArNCH2), 3.60 (d, J ) 13.0 Hz, 2 H), 2.73
(apparent t, J ) 12.3 Hz, 2 H), 2.27 (s, NCH3), 2.15-2.35 (m,
7-Ch lor o-4,5-d ih yd r o-5-[(1-(4-m eth a n esu lfon yl-cis-3,5-
dim eth yl)piper azin o)car bon yl]im idazo[1,5-a ]qu in oxalin e-
3-ca r boxylic Acid 1,1-Dim eth yleth yl Ester (72). Methane-
sulfonyl chloride (0.15 mL, 1.9 mmol) was added to a mixture
of 46 (0.600 g, 1.35 mmol), diisopropylethylamine (0.61 mL,
3.5 mmol), and CH2Cl2 (15.0 mL) at 0 °C. The mixture was
allowed to warm to room temperature and stir for 80 h. Basic
workup (CH2Cl2, 1 N KHCO3, Na2SO4) and purification by
flash chromatography (1% CH3OH/CH2Cl2) gave 0.439 g (62%)
of 72 as a brown foam (mp 156-157 °C): IR (mineral oil) 1724,
1667, 1508, 1423, 1325, 1305, 1281, 1248, 1152, 1126, 964
cm-1; 1H NMR (300 MHz, CDCl3) δ 8.00 (s, ArH), 7.48 (d, J )
8.6 Hz, ArH), 7.17 (dd, J ) 8.6, 1.5 Hz, ArH), 7.10 (narrow m,
ArH), 4.99 (s, ArNCH2), 4.00-4.20 (m, 2 H), 3.61 (d, J ) 13.3
Hz, 2 H), 3.17 (dd, J ) 14.6, 6.1 Hz, 2 H), 2.88 (s, SO2CH3),
1.62 (s, C(CH3)3), 1.37 (d, J ) 7.0 Hz, CHCH3, 6 H); MS (EI)
m/e 523, 467, 450, 248, 232, 231, 230, 219, 136. Anal.
(C23H30N5ClO5S‚(H2O)1/3) C, H, N, Cl.
t
2 H), 1.63 (s, Bu), 1.05 (d, J ) 6.2 Hz, CHCH3, 6 H); MS (EI)
m/e 459, 403, 248, 230, 155, 141, 127, 113, 98, 86. Anal.
(C23H30N5O3Cl‚(H2O)1/2) C, H, N, Cl.
7-Ch lor o-5-(ch lor oca r bon yl)-4,5-d ih yd r oim id a zo[1,5-a ]-
qu in oxa lin e-3-ca r boxylic Acid 1,1-Dim eth yleth yl Ester
(7c). Triphosgene (1.60 g, 5.40 mmol) was added to a solution
of 6c (3.00 g, 9.81 mmol) and diisopropylethylamine (2.1 mL,
12 mmol) in CH2Cl2 (81 mL) at 0 °C. The solution was stirred
at 0 °C for 1 h and then at room temperature for 1 h. Water
(5 mL) was added and the mixture concentrated. Basic workup
(EtOAc, dilute NaHCO3, MgSO4) gave an oil which was
triturated in 10% EtOAc/hexane. The resulting solids were
filtered, washed with hexane, and dried to give 3.32 g (92%)
of the desired compound as an off-white solid: 1H NMR (300
MHz, CDCl3) δ 8.04 (s, ArH), 7.87 (br s, ArH), 7.53 (d, J ) 8.6
Hz, ArH), 7.35-7.45 (m, ArH), 5.46 (narrow s, ArNCH2), 1.64
t
(s, Bu).
6-Ch lor o-3,3-d im eth yl-1,2,3,4-tetr a h yd r o-4-[(1-(cis-3,5-
d im eth yl)p ip er a zin o)ca r bon yl]qu in oxa lin -2-on e (13g; R 1
) R2 ) cis-3,5-d im eth yl, R ) H). A solution of carbamyl
chloride 12g13 (14.2 mmol), CH2Cl2 (40 mL), and diisopropyl-
ethylamine (2.7 mL, 16 mmol) was cooled to 0 °C. To this was
added cis-2,6-dimethylpiperazine (2.28 g, 20.0 mmol) as a solid.
The solution was stirred at 0 °C for 1 h and for 16 h at room
temperature. Basic workup (CH2Cl2, NaHCO3, MgSO4) and
purification by flash chromatography (6:1 ethyl acetate:MeOH)
gave 3.86 g of the urea as a foam. Crystallization from ethyl
acetate/hexane gave 2.68 g (54%) of the product as a pale
yellow powder (mp 151-156 °C): IR (mineral oil) 1697, 1684,
1672, 1648, 1502, 1451, 1235 cm-1; 1H NMR (300 MHz, CDCl3)
δ 9.27 (s, NH, rotamer), 9.19 (s, NH, rotamer), 6.93 (dd, J )
8.3, 2.1 Hz, ArH), 6.75-6.90 (m, ArH), 6.55 (d, J ) 2.0 Hz,
ArH), 4.41 (d, J ) 13.0 Hz, 1 H), 3.48 (d, J ) 13.0 Hz, 1 H),
2.60-3.00 (m, 2 H), 2.25-2.55 (m, 2 H), 1.73 (s, CCH3), 1.30-
1.85 (m, NH), 1.44 (s, CCH3), 1.14 (d, J ) 6.2 Hz, CHCH3),
0.95 (d, J ) 6.2 Hz, CHCH3); MS (EI) m/e 350, 141, 84, 72.
Exa m p le P r oced u r e for Cycliza tion of 13 w ith
ter t-Bu tyl Isocya n oa ceta te to Give 10. Meth od E: 7-
Ch lor o-4,4-d im eth yl-4,5-d ih yd r o-5-[(1-(cis-3,5-d im eth yl)-
p ip er a zin o)ca r b on yl]im id a zo[1,5-a ]q u in oxa lin e-3-ca r -
boxylic Acid 1,1-Dim eth yleth yl Ester (57). Potassium tert-
butoxide (3.00 mL, 3.00 mmol, 1.0 M in THF) was added to a
solution of 13g (R1 ) R2 ) cis-3,5-dimethyl, R ) H; 1.00 g,
2.85 mmol) and THF (10 mL) at 0 °C. The solution was stirred
at 0 °C for 30 min, and diethyl chlorophosphate (0.44 mL, 3.0
mmol) was added. The solution was allowed to warm to room
temperature and stir for 30 min. The solution was cooled to
-7 °C, and a solution of the tert-butyl isocyanoacetate (466
mg, 3.30 mmol) and THF (1.0 mL) was added. Potassium tert-
butoxide (3.0 mL, 3.0 mmol) was added dropwise over several
min. The solution was stirred at 0 °C for 1.5 h and allowed to
warm to room temperature and stir for 2 h. Aqueous workup
(ethyl acetate, MgSO4) and purification by flash chromatog-
raphy (10:1 ethyl acetate:MeOH) gave 1.10 g of an oil which
was crystallized from ether/hexane to give 631 mg (47%) of
57 as a white powder (mp 171-175 °C): IR (mineral oil) 1673,
1663, 1510, 1421, 1363, 1254, 1233, 1151, 1142 cm-1; 1H NMR
(300 MHz, CDCl3) δ 7.96 (s, ArH), 7.39 (dd, J ) 8.4, 1.5 Hz,
ArH), 6.95-7.05 (m, ArH), 6.63 (narrow m, ArH), 4.40-4.55
(m, 1 H), 3.35-3.50 (m, 1 H), 2.60-2.95 (m, 2 H), 2.20-2.55
(m, 2 H), 2.08 (s, CCH3), 1.90-2.10 (m, 1 H), 1.70 (s, CCH3),
1.63 (s, tBu), 1.14 (d, J ) 6.1 Hz, CHCH3), 0.93 (d, J ) 6.9 Hz,
CHCH3); MS (EI) m/e 474, 473, 458, 402, 345, 244, 141. Anal.
(C24H32N5O3Cl) C, H, N, Cl.
Exa m p le P r oced u r e for th e Rea ction of Ca r ba m oyl
Ch lor id e 7 w ith a n Am in e. Meth od C: (3R)-7-Ch lor o-4,5-
d ih yd r o-5-[(1-(3-m et h yl-4-ter t-b u t yloxyca r b on yl)p ip er -
a zin o)ca r b on yl]im id a zo[1,5-a ]q u in oxa lin e-3-ca r b oxyl-
ic Acid 1,1-Dim eth yleth yl Ester (9c; R1 ) 3(R)-Me, R2
)
H). The carbamoyl chloride 7c (1.03 g, 2.80 mmol) was added
to a solution of tert-butyl (R)-2-methyl-1-piperazinecarboxylate25a
(600 mg, 3.00 mmol), diisopropylethylamine (0.70 mL, 4.0
mmol), and CH2Cl2 (25 mL) at 0 °C. The solution was stirred
at 0 °C for 1 h and at room temperature for 16 h. Basic workup
(CH2Cl2, NaHCO3, MgSO4) and purification by flash chroma-
tography (2:1 EtOAc:hexane) gave 1.32 g (88%) of the desired
product as a foam which was carried on crude: IR (mineral
oil) 1727, 1695, 1670, 1508, 1409, 1392, 1367, 1167, 1152 cm-1
;
1H NMR (300 MHz, CDCl3) δ 8.02 (s, ArH), 7.47 (d, J ) 8.6
Hz, ArH), 7.23 (s, ArH), 7.16 (dd, J ) 8.4, 1.5 Hz, ArH), 4.97
(ABq, J AB ) 16.9 Hz, ∆γ ) 109.9 Hz, ArNCH2), 4.20-4.35 (m,
1 H), 3.78 (d, J ) 13.8 Hz, 1 H), 3.55-3.70 (m, 2 H), 3.00-
t
3.20 (m, 2 H), 2.85-3.00 (m, 1 H), 1.62 (s, ArCO2 Bu), 1.45 (s,
t
NCO2 Bu), 1.16 (d, J ) 6.8 Hz, CHCH3); MS (EI) m/e 531, 475,
419, 402, 248, 230, 171, 127.
Exa m p le P r oced u r e for Dep r otection of ter t-Bu tyl-
oxyca r bon yl 9 to P r ovid e 10. Meth od D: (3R)-7-Ch lor o-
4,5-dih ydr o-5-[(1-(3-m eth yl)piper azin o)car bon yl]im idazo-
[1,5-a ]qu in oxa lin e-3-ca r boxylic Acid 1,1-Dim eth yleth yl
Ester (90). A solution of 1.29 g (2.42 mmol) of 9c (R1 ) 3(R)-
Me, R2 ) H), CH2Cl2 (20 mL), and TFA (12 mL) was stirred at
0 °C for 1 h and concentrated. Basic workup (CH2Cl2, NaHCO3,
MgSO4) gave 902 mg of the product as a foam which was
crystallized from EtOAc/ether/hexane to give 793 mg (76%) of
90 as a white powder (mp 157-159 °C), >98% ee by HPLC
analysis:27 [R]25 +3° (c ) 0.75, CHCl3); IR (mineral oil) 1691,
D
1641, 1512, 1159, 1152 cm-1
;
1H NMR (300 MHz, CDCl3) δ
8.00 (s, ArH), 7.46 (d, J ) 8.4 Hz, ArH), 7.10-7.35 (m, ArH, 2
H), 4.95 (s, ArNCH2), 3.60-3.80 (m, 2 H), 2.75-3.10 (m, 4 H),
t
2.62 (t, J ) 10.7 Hz, 1 H), 1.63 (s, BuO), 1.07 (d, J ) 6.1 Hz,
CHCH3); MS (EI) m/e 431, 375, 318, 248, 230, 203, 179, 127.
Anal. (C21H26N5O3Cl‚(H2O)1/2) C, H, N, Cl.
7-Ch lor o-5-[(1-(4-a cet yl-cis-3,5-d im et h yl)p ip er a zin o)-
ca r b on yl]-4,5-d ih yd r oim id a zo[1,5-a ]q u in oxa lin e-3-ca r -
boxylic Acid 1,1-Dim eth yleth yl Ester (71). Acetyl chloride
(0.10 mL, 1.4 mmol) was added to a mixture of 46 (0.600 g,
1.35 mmol), diisopropylethylamine (0.41 mL, 2.4 mmol), and
CH2Cl2 (15.0 mL). The mixture was allowed to warm to room
temperature and stir for 16 h. Basic workup (CH2Cl2, 1 N
KHCO3, Na2SO4) and purification by crystallization (ether)
gave 0.57 g (87%) of 71 as an off-white powder (mp 154-154.5
°C): IR (mineral oil) 1687, 1678, 1668, 1632, 1511, 1420, 1413,
1409, 1391, 1366, 1348, 1335, 1149 cm-1; 1H NMR (300 MHz,
CDCl3) δ 8.00 (s, ArH), 7.48 (d, J ) 8.5 Hz, ArH), 7.17 (dd, J
) 10.6, 2.0 Hz, ArH), 7.11 (narrow m, ArH), 5.00 (s, ArNCH2),
3.55-3.80 (br s, 2 H), 3.00-3.20 (m, 2 H), 2.11 (s, COCH3),
1.63 (s, C(CH3)3), 1.15-1.40 (m, CHCH3, 6 H); MS (EI)
m/e 487, 431, 248, 231, 230, 141. Anal. (C24H30N5ClO4) C, H,
N, Cl.
N,N-Diben zyl-2-n itr o-4-tr iflu or om eth yla n ilin e (16). A
solution of 4-chloro-3-nitrobenzotrifluoride (15, 10.0 mL, 67.0
mmol), dibenzylamine (25.0 mL, 130 mmol), and diisopropy-
lethylamine (15.0 mL, 86.1 mmol) was heated at 140 °C for 5
h. The solution was allowed to cool and stir overnight at room
temperature. The reaction mixture was partitioned between
ether (500 mL) and water (250 mL). The organic layer was
separated, washed with water (250 mL), and partitioned with
water (300 mL). Concentrated HCl was added until the

Piperazine Imidazo[1,5-a]quinoxaline Ureas
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7 1139
aqueous layer was pH 1 (pH paper). The dibenzylamine
hydrochloride byproduct was removed by filtration, and the
two layers were separated. The organic layer was washed with
water (200 mL) and brine (2 × 100 mL), dried (MgSO4), and
concentrated to give 25.2 g (97%) of 16 as an orange solid. This
material was used in the next step without further purifica-
and CH2Cl2 (120 mL) at 0 °C. The solution was stirred for 1 h
at 0 °C and 16 h at room temperature. Aqueous workup (CH2-
Cl2, K2CO3) gave 10.9 g (93%) of 30 as a viscous oil, homoge-
neous by TLC analysis: 1H NMR (300 MHz, CDCl3) δ 3.7-
4.2 (m, 2 H), 2.7-2.85 (m, 2 H), 2.2-2.45 (m, 2 H), 1.46 (s,
tBuO), 1.06 (d, J ) 6.3 Hz, CHCH3, 6 H).
tion: IR (mineral oil) 1621, 1532, 1323, 1268, 1164, 1134 cm-1
;
ter t-Bu tyl 4-Isop r op yl-cis-3,5-d im eth yl-1-p ip er a zin e-
ca r boxyla te (31). A mixture of 30 (10.9 g, 50.9 mmol),
2-iodopropane (50 mL), acetonitrile (100 mL), and K2CO3 (15.0
g, 109 mmol) was heated at reflux for 5 days. Additional
2-iodopropane (25 mL), K2CO3 (15 g, 109 mmol), and acetoni-
trile (25 mL) were added after 2 days. After cooling to room
temperature, aqueous workup (CH2Cl2, MgSO4) and purifica-
tion by flash chromatography (10:1 ethyl acetate:MeOH) gave
5.39 g (41%) of 31 as an oil: 1H NMR (300 MHz, CDCl3) δ
3.35-3.50 (m, 2 H), 3.10-3.30 (m, 3 H), 2.85-3.05 (m, 2 H),
1H NMR (300 MHz, CDCl3) δ 8.07 (narrow m, ArH), 7.56 (dd,
J ) 8.8, 2.1 Hz, ArH), 7.05-7.40 (m, ArH, 11 H), 4.30 (s, NCH2,
4 H); MS (EI) m/e 386, 369, 295, 263, 248, 91.
2-Diben zyla m in o-5-tr iflu or om eth yla n ilin e (17). Hydra-
zine monohydrate (15.8 mL, 326 mmol) was added dropwise
to a stirred mixture of 16 (25.2 g, 65.2 mmol), EtOH (250 mL),
and Raney nickel (Aldrich, ca. 6.3 g) at 0 °C. The mixture was
stirred at 0 °C for 1 h and then at room temperature overnight.
Aqueous workup (ether, brine wash, MgSO4) gave 23.0 g (99%)
of a brown oil which was used without further purification in
the next step: 1H NMR (300 MHz, CDCl3) δ 7.15-7.40 (m,
ArH, 10 H), 6.95 (s, ArH), 6.86 (d, J ) 1.1 Hz, ArH, 2 H), 4.24
(s, NH2), 4.08 (s, NCH2, 4 H).
t
1.46 (s, BuO), 1.05 (d, J ) 6.4 Hz, CHCH3, 6 H), 1.04 (d, J )
6.7 Hz, CHCH3, 6 H).
1-Isop r op yl-cis-2,6-d im eth ylp ip er a zin e (32). A solution
of 31 (780 mg, 3.04 mmol), CH2Cl2 (10.0 mL), and trifluoro-
acetic acid (10.0 mL) was stirred for 1 h at 0 °C and
concentrated. Basic workup (CH2Cl2, 20% NaOH, K2CO3) gave
427 mg (90%) of 32 as an oil: 1H NMR (300 MHz, CDCl3) δ
3.15-3.30 (m, CH(CH3)2), 2.80-2.90 (m, 2 H), 2.50-2.70 (m,
4 H), 1.30-1.70 (m, NH), 1.12 (d, J ) 7.0 Hz, 6 H), 1.06 (d, J
) 5.9 Hz, 6 H).
N -(2-Dib e n zyla m in o-5-t r iflu or om e t h ylp h e n yl)gly-
cin e Eth yl Ester (18). A solution of 17 (22.3 g, 62.6 mmol),
diisopropylethylamine (24.0 mL, 138 mmol), and ethyl bro-
moacetate (20.9 mL, 188 mmol) was heated at 120 °C for 1 h.
The mixture was allowed to cool to room temperature. Aqueous
workup (EtOAc, brine wash, MgSO4) and purification by flash
chromatography (1f10% EtOAc/hexane) gave a solid which
was triturated in hexane to provide 19.5 g (70%) of 18 as a
white solid (mp 85-87 °C): IR (mineral oil) 1740, 1211, 1178,
(S)-[2-[(2-Hyd r oxyeth yl)a m in o]-1-m eth yl-2-oxoeth yl]-
ca r ba m ic Acid 1,1-Dim eth yleth yl Ester (34A). CDI (3.44
g, 21.2 mmol) was added to a solution of Boc-L-alanine 33 (4.00
g, 21.1 mmol) and CH2Cl2 (64 mL). The solution was stirred
for 1 h at room temperature. Ethanolamine (1.30 mL, 21.5
mmol) was then added. The solution was stirred for 70 h at
room temperature and then concentrated. Purification by flash
chromatography (9:1 EtOAc:MeOH) gave 4.13 g (84%) of 34A
1
1170, 1102, 1089 cm-1; H NMR (300 MHz, CDCl3) δ 7.20-
7.35 (m, ArH, 10 H), 6.80-6.95 (m, ArH, 2 H), 6.67 (s, ArH),
5.72 (br s, NH), 4.30 (q, J ) 7.1 Hz, CH2CH3), 4.06 (s, NCH2-
Ar, 4 H), 3.93 (s, CH2CO2Et), 1.32 (t, J ) 7.2 Hz, CH2CH3);
MS (EI) m/e 442, 351, 277, 187, 91.
3,4-Dih yd r o-2-h yd r oxy-6-t r iflu or om e t h ylq u in oxa -
lin e (19). A mixture of 18 (14.6 g, 33.0 mmol), ethanol (600
mL), and palladium hydroxide on carbon (3.65 g) was hydro-
genated at 40 psi for 3 days. The mixture was filtered and
concentrated. Basic workup (EtOAc, NaHCO3, MgSO4) of the
residue gave 6.94 g (97%) of 19 as a white solid (mp 97-98
°C): IR (mineral oil) 1688, 1672, 1424, 1324, 1315, 1252, 1132,
1111 cm-1; 1H NMR (300 MHz, CDCl3) δ 9.31 (br s, OH), 7.01
(d, J ) 7.3 Hz, ArH), 6.89 (s, ArH), 6.83 (d, J ) 8.1 Hz, ArH),
4.06 (s, NCH2); MS (EI) m/e 216, 187, 167, 140, 91. Anal.
(C9H7N2OF3‚(H2O)1/2) C, H, N.
as an oil: IR (neat) 3308, 1694, 1659, 1531, 1368, 1169 cm-1
;
1H NMR (300 MHz, CDCl3) δ 6.55-6.70 (m, 1 H), 4.90-5.10
(m, 1 H), 4.12 (q, J ) 6.9 Hz, NCHCH3), 3.72 (t, J ) 5.0 Hz,
t
NCH2CH2OH), 3.35-3.55 (m, NCH2CH2), 1.45 (s, BuO), 1.38
(d, J ) 7.1 Hz, CHCH3); MS (EI) m/e 232, 202, 176, 159, 144,
113, 102, 88.
(S)-[2-[(2-Hydr oxyeth yl)am in o]-1-m eth yleth yl]car bam -
ic Acid 1,1-Dim eth yleth yl Ester (35A). Borane-methyl
sulfide complex (3.00 mL, 30.0 mmol, 10.0 M) was added to a
solution of 34A (2.77 g, 11.9 mmol) and THF (70 mL). The
solution was stirred for 16 h at room temperature, and the
reaction was quenched slowly with 10% HCl. THF (100 mL),
water (25 mL), and KOH (4.00 g) were added, and the mixture
was heated at reflux for 24 h. After the mixture cooled to room
temperature, the organic layer was removed under reduced
pressure. The aqueous layer was saturated with sodium
chloride and extracted several times with CH2Cl2. The organic
layers were dried (K2CO3), filtered, and concentrated to give
1.72 g (66%) of 35A as a white solid (mp 54-58 °C) sufficiently
4,5-Dih yd r o-7-tr iflu or om eth ylim id a zo[1,5-a ]qu in oxa -
lin e-3-ca r boxylic Acid 1,1-Dim eth yleth yl Ester (20). Po-
tassium tert-butoxide (35.0 mL, 35.0 mmol, 1.0 M in THF) was
added to a mixture of 19 (6.30 g, 29.1 mmol) and THF (100
mL) at -40 °C. The solution was stirred at -40 °C for 10 min
and then at room temperature for 30 min. The solution was
cooled to -50 °C and diethyl chlorophosphate (5.5 mL, 38
mmol) added. The solution was stirred at -50 °C for 10 min
and then at room temperature for 40 min. The solution was
cooled to -78 °C, and tert-butyl isocyanoacetate (5.1 mL, 35
mmol) was added. Potassium tert-butoxide (35.0 mL, 35.0
mmol) was added dropwise over several minutes. The solution
was allowed to warm slowly and was stirred at room temper-
ature for 4 days. Water (10 mL) was added and the mixture
concentrated. Aqueous workup (EtOAc, MgSO4) and tritura-
tion in 20% EtOAc/hexane gave 4.13 g of the desired material
as a white solid (mp 242-244 °C). Concentration of the mother
liquor and purification by flash chromatography (20% EtOAc/
CH2Cl2) followed by trituration in 20% EtOAc/hexane gave an
additional 619 mg of 20. The total yield of 20 was 4.75 g
(48%): IR (mineral oil) 1704, 1329, 1295, 1247, 1167, 1157,
pure to be carried on crude: [R]25 -3° (c ) 1.0, CHCl3); IR
D
(mineral oil) 3355, 1681, 1519, 1366, 1165, 1073 cm-1; 1H NMR
(300 MHz, CDCl3) δ 4.50-4.65 (m, 1 H), 3.60-3.90 (m, 1 H),
3.63 (t, J ) 5.4 Hz, CH2O), 2.75-2.90 (m, 2 H), 2.55-2.70 (m,
t
2 H), 1.50-2.10 (m, OH, NH), 1.45 (s, Bu), 1.14 (d, J ) 6.7
Hz, CHCH3); MS (EI) m/e 200, 187, 145, 131, 101, 74.
(S)-7-Ch lor o-5-[[[2-[[(1,1-dim eth yleth oxy)car bon yl]am i-
n o]p r op yl](2-h yd r oxyet h yl)a m in o]ca r b on yl]-4,5-d ih y-
d r oim id a zo[1,5-a ]q u in oxa lin e-3-ca r b oxylic Acid 1,1-
Dim eth yleth yl Ester (36A; R7 ) Cl). The carbamoyl chloride
7c (1.32 g, 3.58 mmol) was added to a solution of 35A (858
mg, 3.93 mmol), diisopropylethylamine (0.90 mL, 5.2 mmol),
and CH2Cl2 (25 mL) at 0 °C. The solution was stirred at 0 °C
for 1 h and at room temperature for 16 h. Basic workup (CH2-
Cl2, NaHCO3, MgSO4) and purification by flash chromatogra-
phy (EtOAc) gave 1.58 g (80%) of 36A as a tan foam which
was carried on crude (mp 118-122 °C): IR (mineral oil) 1723,
1
1115, 1049 cm-1; H NMR (300 MHz, CDCl3) δ 8.03 (s, ArH),
7.48 (d, J ) 8.3 Hz, ArH), 7.09 (d, J ) 8.3 Hz, ArH), 7.03
t
(narrow m, ArH), 4.86 (s, NCH2), 1.62 (s, Bu); MS (EI) m/e
339, 283, 266, 237, 186. Anal. (C16H16N3O2F3) C, H, N.
ter t-Bu t yl cis-3,5-Dim et h yl-1-p ip er a zin eca r b oxyla t e
(30). A solution of di-tert-butyl dicarbonate (11.8 g, 54.1 mmol)
and CH2Cl2 (30 mL) was added dropwise over 15 min to a
solution of cis-2,6-dimethylpiperazine 29 (6.25 g, 54.7 mmol)
1710, 1510, 1250, 1157 cm-1 1H NMR (300 MHz, CDCl3) δ
;
7.97 (s, ArH), 7.43 (d, J ) 8.6 Hz, ArH), 7.39 (s, ArH), 7.12
(dd, J ) 6.9, 1.5 Hz, ArH), 4.92 (ABq, J AB ) 15.3 Hz, ∆γ )
28.0 Hz, ArNCH2), 4.55-4.70 (m, 1 H), 3.70-4.00 (m, 3 H),

1140 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7
3.30-3.60 (m, 3 H), 3.11 (dd, J ) 14.6, 6.1 Hz, 1 H), 2.65-
2.85 (m, 1 H), 1.61 (s, ArCO2 Bu), 1.43 (s, NCO2 Bu), 1.05 (d,
J ) 6.6 Hz, CHCH3); MS (EI) m/e 549, 493, 437, 350, 248, 230,
203, 189, 145.
J acobsen et al.
4,5-d ih yd r oim id a zo[1,5-a ]qu in oxa lin e-3-ca r boxylic Acid
1,1-Dim eth yleth yl Ester (36B, R7 ) Cl). The carbamoyl
chloride 7c (1.51 g, 4.10 mmol) was added to a solution of 35B
(1.05 g, 4.52 mmol), diisopropylethylamine (1.03 mL, 5.91
mmol), and CH2Cl2 (33 mL) at 0 °C. The solution was stirred
at 0 °C for 1 h and at room temperature for 5 h. Basic workup
(CH2Cl2, NaHCO3, MgSO4) and purification by flash chroma-
tography (EtOAc) gave 1.79 g (77%) of 36B (R7 ) Cl) as a pale
yellow powder which was carried on crude: IR (mineral oil)
t
t
(S)-5-[[(2-a m in op r op yl)(2-h yd r oxyeth yl)a m in o]ca r bo-
n yl]-7-ch lor o-4,5-d ih yd r oim id a zo[1,5-a ]q u in oxa lin e-3-
ca r boxylic Acid 1,1-Dim eth yleth yl Ester (37A; R7 ) Cl).
A solution of 36A (R7 ) Cl; 1.54 g, 2.80 mmol), CH2Cl2 (22
mL), and TFA (14 mL) was stirred at 0 °C for 1 h and then
concentrated. Basic workup (CH2Cl2, NaHCO3, MgSO4) gave
956 mg (76%) of 37A (R7 ) Cl) as a foam which was carried
on crude: 1H NMR (300 MHz, CDCl3) δ 7.99 (s, ArH), 7.44 (d,
J ) 8.6 Hz, ArH), 7.21 (d, J ) 2.3 Hz, ArH), 7.10-7.20 (m,
ArH), 4.93 (ABq, J AB ) 15.3 Hz, ∆γ ) 83.6 Hz, ArNCH2), 3.65-
1724, 1710, 1657, 1510, 1416, 1251, 1159, 1151, 1127 cm-1
;
1H NMR (300 MHz, CDCl3) δ 7.99 (s, ArH), 7.52 (narrow m,
ArH), 7.43 (d, J ) 8.6 Hz, ArH), 7.11 (dd, J ) 9.2, 1.5 Hz,
ArH), 4.92 (s, ArNCH2), 4.45-4.60 (m, 1 H), 4.05-4.25 (m, 1
H), 3.80-4.00 (m, 1 H), 3.05-3.50 (m, 4 H), 1.61 (s, Ar-
t
t
t
CO2 Bu), 1.43 (s, NCO2 Bu), 1.18 (d, J ) 6.2 Hz, CHCH3), 1.02
(d, J ) 6.7 Hz, CHCH3); MS (EI) m/e 563, 451, 434, 364, 276,
248, 230, 203, 159, 141.
3.75 (m, 2 H), 3.25-3.55 (m, 5 H), 1.60 (s, BuO), 1.17 (d, J )
6.4 Hz, CHCH3).
Exa m p le P r oced u r e for In tr a m olecu la r Mitsu n obu
Cycliza t ion of a n Am in o Alcoh ol t o P ip er a zin e-10.
Meth od F : (3S)-7-Ch lor o-4,5-d ih yd r o-5-[(1-(3-m eth yl)-
p ip er a zin o)ca r b on yl]im id a zo[1,5-a ]q u in oxa lin e-3-ca r -
boxylic Acid 1,1-Dim et h ylet h yl E st er (89). Diethyl azo-
dicarboxylate (DEAD, 0.43 mL, 2.7 mmol) was added to a
solution of 37A (R7 ) Cl; 951 mg, 2.11 mmol), triphenylphos-
phine (745 mg, 2.84 mmol), and THF (18.6 mL). The solution
was stirred at room temperature for 16 h and then concen-
trated. Purification by flash chromatography (4:1 EtOAc:
MeOH) gave the product as a foam which was crystallized from
EtOAc/ether/hexane to give 542 mg (59%, two crops) of 89 as
a white powder (mp 158-160 °C), >98% ee by HPLC ana-
lysis:27 [R]25D -1° (c ) 0.92, EtOH); IR (mineral oil) 1687, 1646,
1611, 1511, 1435, 1423, 1390, 1359, 1311, 1295, 1282, 1154,
[R-(R*,S*)]-5-[[(2-Am in op r op yl)(2-h yd r oxyp r op yl)a m i-
n o]ca r bon yl]-7-ch lor o-4,5-d ih yd r oim id a zo[1,5-a ]qu in ox-
a lin e-3-ca r boxylic Acid 1,1-Dim eth yleth yl Ester (37B, R7
) Cl). A solution of 36B (R7 ) Cl; 1.77 g, 3.14 mmol), CH2Cl2
(25 mL), and TFA (16 mL) was stirred at 0 °C for 1 h and
then concentrated. Basic workup (CH2Cl2, NaHCO3, MgSO4)
gave 1.07 g (73%) of 37B (R7 ) Cl) as a pale yellow foam which
was carried on crude: IR (mineral oil) 1722, 1695, 1661, 1510,
1417, 1294, 1282, 1216, 1152, 1129 cm-1; 1H NMR (300 MHz,
CDCl3) δ 7.99 (s, ArH), 7.44 (d, J ) 8.6 Hz, ArH), 7.23 (narrow
m, ArH), 7.13 (dd, J ) 8.4, 2.3 Hz, ArH), 4.92 (ABq, J AB ) 15.3
Hz, ∆γ ) 66.7 Hz, ArNCH2), 4.00-4.15 (m, 1 H), 3.30-3.50
(m, 3 H), 3.10-3.30 (m, 1 H), 3.07 (dd, J ) 14.6, 10.0 Hz, 1
t
H), 1.62 (s, BuO), 1.09 (d, J ) 6.1 Hz, CHCH3), 1.05 (d, J )
1
1133, 968 cm-1; H NMR (300 MHz, CDCl3) δ 7.99 (s, ArH),
6.2 Hz, CHCH3); MS (EI) m/e 463, 407, 364, 248, 230, 203,
179, 159, 141.
7.45 (d, J ) 8.6 Hz, ArH), 7.21 (s, ArH), 7.13 (dd, J ) 8.5, 2.1
Hz, ArH), 4.95 (s, ArNCH2), 3.60-3.80 (m, 2 H), 2.75-3.00
Meth od F . (3S,5S)-7-Ch lor o-5-[(1-(3,5-tr a n s-d im eth yl)-
p ip er a zin o)ca r bon yl]-4,5-d ih yd r oim id a zo[1,5-a ]qu in oxa -
lin e-3-car boxylic Acid 1,1-Dim eth yleth yl Ester (91). DEAD
(0.46 mL, 2.9 mmol) was added to a solution of 37B (R7 ) Cl;
1.05 g, 2.26 mmol), triphenylphosphine (800 mg, 3.05 mmol),
and THF (20 mL). The solution was stirred at room temper-
ature for 16 h and then concentrated. Purification by flash
chromatography (4:1 EtOAc:MeOH) gave 835 mg of 91 as an
oil which was crystallized from ether/hexane to give 711 mg
(71%, two crops) of the amine as a white powder (mp
121-125 °C), >98% ee by 1H NMR analysis27 of the Mosher
t
(m, 4 H), 2.55 (dd, J ) 12.6, 10.6 Hz, 1 H), 1.63 (s, Bu), 1.04
(d, J ) 6.3 Hz, CHCH3); MS (EI) m/e 431, 375, 358, 317, 248,
230, 203, 179, 127. Anal. (C21H26N5O3Cl‚(H2O)4/5) C, H, N, Cl.
(S)-[2-[(2(R)-H yd r oxyp r op yl)a m in o]-1-m et h yl-2-oxo-
eth yl]ca r ba m ic Acid 1,1-Dim eth yleth yl Ester (34B). CDI
(3.44 g, 21.2 mmol) was added to a solution of Boc-L-alanine
(4.00 g, 21.1 mmol) and CH2Cl2 (64 mL). The solution was
stirred for 1 h at room temperature and then (R)-1-amino-2-
propanol (2.50 mL, 31.8 mmol) was added. The solution was
stirred for 16 h at room temperature and then concentrated.
Purification by flash chromatography (EtOAc) gave 4.60 g
(89%) of 34B as a clear oil: IR (neat) 3308, 2978, 1694, 1659,
amide:52 [R]25 +47° (c ) 0.92, CHCl3); IR (mineral oil) 1724,
D
1695, 1662, 1508, 1419, 1295, 1279, 1266, 1243, 1150, 1127
cm-1; 1H NMR (300 MHz, CDCl3) δ 7.99 (s, ArH), 7.45 (d, J )
8.5 Hz, ArH), 7.10-7.20 (m, ArH, 2 H), 4.95 (ABq, J AB ) 16.0
Hz, ∆γ ) 77.9 Hz, ArNCH2), 3.30-3.45 (m, 2 H), 3.15-3.30
1530, 1368, 1250, 1170 cm-1 1H NMR (300 MHz, CDCl3) δ
;
6.45-6.60 (m, 1 H), 4.85-5.00 (m, 1 H), 4.13 (q, J ) 6.9 Hz,
NCHCH3), 3.90-4.00 (m, 1 H), 3.35-3.50 (m, 1 H), 3.05-3.20
t
t
(m, 2 H), 3.02 (dd, J ) 13.0, 6.1 Hz, 2 H), 1.62 (s, BuO), 1.09
(m, 1 H), 1.45 (s, BuO), 1.38 (d, J ) 7.1 Hz, CHCH3), 1.19 (d,
(d, J ) 6.5 Hz, CHCH3, 6 H); MS (EI) m/e 445, 389, 318, 306,
248, 230, 141. Anal. (C22H28N5O3Cl‚(H2O)1/3) C, H, N, Cl.
J ) 6.3 Hz, CHCH3); MS (EI) m/e 202, 173, 144, 102, 88.
(S)-[2-[(2(R)-Hydr oxypr opyl)am in o]-1-m eth yleth yl]car -
ba m ic Acid 1,1-Dim eth yleth yl Ester (35B). Borane-meth-
yl sulfide complex (4.70 mL, 47.0 mmol, 10.0 M) was added to
a solution of 34B (4.54 g, 18.4 mmol) and THF (110 mL). The
solution was stirred for 16 h at room temperature and then
quenched slowly with 10% HCl. Water (40 mL) and KOH (10.0
g) were added, and the mixture heated at reflux for 26 h. After
cooling to room temperature, the organics were removed under
reduced pressure. The aqueous layer was saturated with
sodium chloride which was extracted several times with CH2-
Cl2. The organic layers were dried (MgSO4), filtered, and
concentrated to give 3.22 g (75%) of 35B as a white solid,
sufficiently pure to be carried on crude. A small portion was
recrystallized from hot ether/hexane to give a white crystalline
GABAA Recep tor Exp r ession a n d Mem br a n e P r ep a r a -
tion . DNA manipulations and general baculovirus methods
(Sf-9 cell cultivation, infection, and isolation; purification of
recombinant viruses) were performed as described elsewhere.29
The Sf-9 cells were infected at a multiplicity of infection of 3
PFU of the following viruses: AcNPV-R1, -R3, or -R6, AcNPV-
â2, and AcNPV-γ2. Infected cells were used for electrophysi-
ological measurements at 48 h postinfection or for membrane
preparations at 60 h postinfection. The stable cell lines
expressing R1, R3, or R6, â2, and γ2 subunits of GABAA were
derived by transfection of plasmids containing cDNA and a
plasmid encoding G418 resistance into human kidney cells
(A293 cells) as described elsewhere.30 After 2 weeks of selection
in 1 mg/mL G418, cells positive for all three GABAA receptor
mRNAs by Northern blotting were used for electrophysiology
to measure GABA-induced Cl- currents.
material (mp 77-78 °C): [R]25 -28° (c ) 0.98, CHCl3); IR
D
(mineral oil) 3374, 1683, 1525, 1177, 1160, 1093 cm-1; 1H NMR
(300 MHz, CDCl3) δ 4.45-4.65 (m, 1 H), 3.65-3.90 (m, 2 H),
2.50-2.85 (m, 3 H), 2.40 (dd, J ) 12.0, 9.4 Hz, 1 H), 2.05-
2.50 (m, NH, OH), 1.45 (s, tBuO), 1.15 (d, J ) 6.2 Hz, CHCH3),
1.14 (d, J ) 6.7 Hz, CHCH3); MS (EI) m/e 214, 187, 159, 141,
131, 115, 88.
For equilibrium binding measurements, Sf-9 cells infected
with baculovirus-carrying cDNAs for R1, R3, or R6, â2, and γ2
subunits were harvested in 2 L batches at 60 h postinfection.
The membranes were prepared following the procedure de-
scribed previously.29 Briefly, the membranes were prepared
in normal saline after homogenization with a Polytron PT 3000
[R-(R*,S*)]-7-Ch lor o-5-[[[2-[[(1,1-d im eth yleth oxy)ca r -
bon yl]a m in o]p r op yl](2-h ydr oxyp r op yl)a m in o]ca r bon yl]-

Piperazine Imidazo[1,5-a]quinoxaline Ureas
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7 1141
homogenizer (Brinkman) for 4 min. Unbroken cells and large
nuclei aggregates were removed by centrifugation at 1000g
for 10 min. Then the membranes were recovered with the
second centrifugation of the supernatant at 40000g for 50 min.
The membranes were resuspended to a final concentration of
5 mg/mL in a solution containing 300 mM sucrose, 5 mM Tris/
HCl, pH 7.5, and glycerol to a final concentration of 20% and
stored at -80 °C. Equilibrium binding of [3H]flunitrazepam
or [3H]Ro-4513 to the cloned GABAA receptors was measured
in a 500 µL volume of normal saline containing 6 nM [3H]-
flunitrazepam or [3H]Ro-4513, varying concentrations of test
ligands, and 50 µg of membrane protein. The mixture was
incubated for 60 min at 4 °C, filtered over a Whatman glass
fiber filter, and washed four times with cold normal saline.
The filter was then counted for radioactivity in the presence
of a scintillation cocktail (Insta Gel).
Control mice received an equal volume of the vehicle (1 mL/
100 g). Animals were sacrificed at 30, 60, 120, and 240 min
after administration of drug. Whole brain minus cerebellum
was quickly removed and stored at -78 °C until used for the
binding assay. Brains were thawed and then homogenized in
50 volumes of cold (4 °C) 50 mM Tris-HCl buffer. [3H]-
Flunitrazepam binding was determined by incubating 1.0 mL
aliquots of the homogenate with 0.1 mL of [3H]flunitrazepam
to give a final concentration of 0.68 nM, 0.1 mL of distilled
water or flurazepam (10 µM), and 0.8 mL of 50 mM Tris-HCl
buffer, pH 7.4, to give a final volume of 2.0 mL. The mixture
was incubated for 30 min at 25 °C and then filtered under
vacuum through Whatman GF/B filters. The filter and incuba-
tion tube were rinsed four times 5 mL aliquots of buffer. The
filter paper was placed in a scintillation vial to which 15 mL
of ACS (Amersham) cocktail was added. Radioactivity was
counted by a liquid scintillation spectrometer. Specific binding
was defined as total binding minus binding in the presence of
10 µM flurazepam. Specific binding represented over 95% of
total binding. The results are expressed as percent inhibition
of (3H)-FNZ binding, with the ratio calculated as previously
described.40a
Cer eb ella r cGMP Assa y. This assay was performed as
described previously.42 Thus in brief, the test compounds were
dissolved in vehicle 122 and administered orally to male CF-1
mice. Control mice received an equal volume of vehicle (1 mL/
100 g). Thirty minutes after drug or vehicle treatment, animals
were sacrificed and the levels of the cerebellar cGMP deter-
mined by the procedure described by Burkard.53 The cerebel-
lum was quickly removed and homogenized in 10 volumes of
1% perchloric acid. The samples were kept on ice for 30 min,
boiled for 5 min, and then centrifuged at 17000g for 20 min.
The content of cGMP in the supernatant was measured by a
radioimmunoassay. Animals in the stressed group were sub-
jected to electric footshock (1.0 mA for 10 s) stress 30 min after
drug or vehicle administration. Mice were immediately sac-
rificed with the levels of cGMP estimated as described above.
The results are expressed as percent of control for animals
treated with (drug + stress)/stress with statistical analysis
done using one-way analysis of variance and subsequently by
Student’s t-test.
GABAA Recep tor Bin d in g. The in vitro binding affinity
of the imidazo[1,5-a]quinoxalines for GABAA was determined
as previously described with minor modification.28 Freshly
prepared rat cerebellar membranes were suspended in 300 mM
sucrose and 10 nM Hepes/Tris, pH 7.4. Typically, the reaction
medium contained 6 mM [3H]flunitrazepam, 50 µg of mem-
brane protein, test drugs at various concentrations or vehicle
in 200 µL, 118 nM NaCl, 10 mM Hepes/Tris, pH 7.4, and 1
mM MgCl2. The mixtures were incubated for 60 min at 4 °C.
The amount of binding was determined with rapid filtration
techniques using Whatman GF/B filters.
[
35S]-ter t-Bu tyl Bicyclop h osp h or oth ion a te ([35S]TBP S)
Bin d in g. Binding of [35S]TBPS in the rat brain membranes
was measured in the medium containing 2 nM [35S]TBPS,
unless specified otherwise, 50 µg of membrane proteins, 1 M
NaCl, and 10 mM Tris/HCl, pH 7.4, in a total volume of 500
µL. Drugs were added in concentrated methanolic solutions,
and the level of methanol did not exceed 0.2% and was
maintained constant in all tubes. The mixtures were incubated
for 120 min at 24 °C. The reaction mixtures were filtered over
a Whatman GF/B filter under vacuum. The filters were washed
three times with 4 mL of the reaction buffer without radio-
isotope and counted for radioactivity. Nonspecific binding was
estimated in the presence of 2 µM diazepam or unlabeled
TBPS and subtracted to compute specific binding.33
Acu te P h ysica l Dep en d en ce.43 Male CF-1 mice (10/group)
were dosed sc twice daily with the test compound, once at 0800
and again at 1600 (with 2 times the 0800 dose) for 3 days.
Twenty-four h after the last dose, they received an intravenous
injection of the benzodiazepine antagonist flumazenil (2.5 mg/
kg in 5% aqueous N,N-dimethylacetamide). Five minutes later,
the electroshock seizure threshold was estimated by the up-
down method in which the stimulus current was lowered or
elevated by a 0.05 log interval if the preceding animal did or
did not convulse, respectively. From the data thus generated,
a threshold (effective current of 50) was calculated. The other
parameters of the transpinnal (e.g., delivered across the ears
via saline-soaked ear clip electrodes) square wave stimulation
were held constant (0.6 ms pulses at 100 Hz for 0.2 s). Test
compounds, the precipitated withdrawal from which signifi-
cantly lowered (below 95% confidence interval of parallel
control group treated for 3 days with saline and injected
intravenously with flumazenil) the seizure threshold, were
considered to have caused physical dependence.
Electr op h ysiology. The whole cell configuration of the
patch clamp technique was used to record the GABA-mediated
Cl- currents in the A293 cells expressing the R1â2γ2 subtype,
as described earlier.35 Briefly, patch pipets made of borosilicate
glass tubes were fire-polished and showed a tip resistance of
0.5-2 MΩ when filled with a solution containing 140 mM
CsCl, 11 mM EGTA, 4 mM MgCl2, 2 mM ATP, and 10 mM
Hepes, pH 7.3. The cell-bathing external solution contained
135 mM NaCl, 5 mM KCl, 1 mM MgCl2, 1.8 mM CaCl2, and 5
mM Hepes, pH 7.2 (normal saline). GABA at the concentration
of 5 µM in the external solution with or without indicated
drugs was applied through a U-tube placed within 100 µM of
the cells for 10 s, unless indicated otherwise. The current was
recorded with an Axopatch 1D amplifier and a CV-4 headstage
(Axon Instrument Co.). A Bh-1 bath headstage was used to
compensate for changes in bath potentials. The currents were
recorded with a Gould recorder 220. GABA currents were
measured at the holding potential of -60 mV at room tem-
perature (21-24 °C).
Ben zod ia zep in e An ta gon ism . Male CF-1 Charles River
mice were injected ip simultaneously with 1 mg/kg triazolam
and test compound starting at 100 mg/kg. A dose-response
was run on a 0.5 log scale. All drugs were made up in No. 122
sterile vehicle (0.25% aqueous methyl cellulose). Fifteen
minutes later, they were tested for loss of traction response
(muscle relaxation) as indicated above. If the expected loss of
traction in response to triazolam was not observed, the test
compound was assumed to be an antagonist. Lower doses (at
0.5 log intervals) of the test compound were then evaluated to
establish the antagonist dose50 (ED50).
Metr a zole An ta gon ism . Compounds were tested for their
ability to antagonize metrazole-induced convulsions in rats
after ip injection as described elsewhere.37 Briefly, male CF-1
mice were injected with metrazole (85 mg/kg, sc) and convul-
sive seizure was elicited 15 min later with an auditory
stimulation (5 dB for 10 s). Drugs tested for metrazole
antagonism were injected ip 30 min before the metrazole
challenge, 4 mice/dose at a 0.3 log dose interval. ED50’s for
protection against tonic seizure were calculated by the method
of Spearman-Karber (Finney, D. J . Statistical Methods in
Biological Assay).
Ex Vivo Bin d in g Assa y. This assay was carried out as
previously described with minor modification.40a Briefly, the
test drug in vehicle 122 was administered orally to CF-1 mice.
Vogel’s P u n ish ed Lick in g Assa y. Following a previously
reported procedure,9a naive male Fischer 344 rats (except for
the 42 experiment which used less responsive Sprague-

1142 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7
J acobsen et al.
Dawley rats) were deprived of water for 48 h prior to lick
training in the morning of the test day. Those drinking
appreciably in the morning were tested in the afternoon 30
min after ip injection of the test compound, with at least 8
animals/dose group. Drugs were dissolved or suspended in a
0.25% methyl cellulose and saline solution (Upjohn vehicle
122). Training consisted of placement in the chamber for up
to 5 min or until rats licked the drinking spout for 1 min.
Testing consisted of a 3 min session in which each twentieth
lick produced a 0.5 s duration application of distributed electric
shock to the grid floor and drinking spout. Data presented are
the means and standard errors of the mean of the number of
shocks produced in each group by licking. Statistical signifi-
cance was determined by using a one-way analysis of varience
followed by the nonparametric Wilcoxon rank sum test com-
paring the test dose to that of the contemporaneous vehicle
group.
weighting was used to calculate the unknown sample concen-
trations by inverse prediction against the calibration curve.
P h a r m a cok in etic An a lysis. Selected pharmacokinetic
parameters were calculated using model-independent meth-
ods.55 The maximum serum concentration (Cmax) and the time
at which Cmax was reached (tmax) were obtained from the
observed concentration-time data. The area under the serum
concentration time curve (AUC) extrapolated to infinity was
calculated using the equation AUC ) AUC(0,tlast) + Ctlast/k,
where AUC(0,tlast) is the area under the curve from time zero
to the last time point with a measurable serum concentration
(Ctlast), and k, the terminal disposition rate constant, is
determined from the linear regression slope of the terminal
portion of the log serum concentration-time curve. The
terminal disposition half-life (t1/2) was obtained from 0.693/k.
Total plasma clearance, Clp, was calculated from Doseiv/AUCiv
following iv administration. The percent absolute bioavail-
ability (F) was determined by F ) (AUCpo ‚ Doseiv ‚ 100%)/
(AUCiv ‚ Dosepo).
Geller -Cook Con flict Assa y. Following a previously
9a
reported procedure,
Sprague-Dawley rats (male) were
Refer en ces
trained for many weeks to press a lever for food on a MULT
VI60/FR10 reinforcement schedule in which a 0.5 s duration
footshock was administered with each food pellet during the
FR10 components of the schedule. Sessions consisted of four
5-min VI60 components alternating with four 2-min FR10
components. Shock intensity was adjusted to depress respond-
ing under the FR10 components to approximately the rate on
VI60 components. This generally constituted a halving of FR10
response rate. After performance stabilized, the test drug or
vehicle was injected ip 30 min before the session, with at least
6 animals/dose group. A drug effect for each rat was measured
by an increase in responding compared to the immediately
preceding vehicle session. Statistical significance for each dose
was estimated by the Wilcoxon signed-rank test.
(1) (a) Squires, R., Ed. GABA and Benzodiazepine Receptors; CRC
Press: Boca Raton, FL, 1988; Vol. I, II. (b) Haefely, W. In
Benzodiazepines; Beaumont, G., Brandon, S., Leonard, B. E.,
Eds.; J ohn Wiley and Sons: New York, 1990; pp 1-18. (c)
Haefely, W. Partial Agonists of the Benzodiazepine Receptor:
From Animal Data to Results in Patients. In Chloride Channels
and Their Modulation by Neurotransmitters and Drugs; Biggio,
G., Costa, E., Eds.; Raven Press: New York, 1988; pp 275-292.
(2) (a) Gardner, C. R. Functional In Vivo Correlates of the Benzo-
diazepine Agonist-Inverse Agonist Continuum. Prog. Neurobiol.
1988, 31, 425-476. (b) Braestrup, C.; Nielsen, M. In Receptor
Biochemistry and Methodology; Olsen, R. W., Venter, J . C., Eds.;
A. R. Liss, Inc.: New York, 1986; Vol. 5, p 167. (c) Bare, T. M.;
Resch, J . F.; Patel, J . B. Anxiolytics and Sedative-Hypnotics. In
Annual Reports in Medicinal Chemistry; Bailey, D. M., Ed.;
Academic Press: New York, 1987; Vol. 22, pp 11-20. (d) Haefely,
W.; Kyburz, E.; Gerecke, M.; Mohler, H. Recent Advances in the
Molecular Pharmacology of Benzodiazepine Receptors and in the
Structure-Activity Relationships of Their Agonists and Antago-
nists. In Advances in Drug Research; Testa, B., Ed.; Academic
Press: New York, 1985; Vol. 14, pp 165-322. (e) Braestrup, C.;
Honore, T.; Nielsen, M.; Petersen, E. N.; J ensen, L. H. Ligands
for Benzodiazepine Receptors With Positive and Negative Ef-
ficacy. Biochem. Pharmacol. 1984, 33, 859-862.
(3) (a) Petersen, E. N.; J ensen, L. H.; Diejer, J .; Honore, T. New
Perspectives in Benzodiazepine Receptor Pharmacology. Phar-
macopsychiatry 1986, 19, 4-6. (b) Haefely, W. Psychopharma-
cology of Anxiety. Eur. Neuropsychopharmacol. 1991, 1, 89-95.
(4) For recent reviews, see: (a) Gardner, C. R. A Review of Recently-
Developed Ligands for Neuronal Benzodiazepine Receptors and
Their Pharmacological Activities. Prog. Neuro-Psychopharmacol.
Biol. Psychiatry 1992, 16, 755-781. (b) Gammill, R. B.; Carter,
D. B. Neuronal BZD Receptors: New Ligands, Clones and
Pharmacology. In Annual Reports Medicinal Chemistry; Bristol,
J . A., McCall, J . M., Eds.; Academic Press: New York, 1993;
Vol. 28, pp 19-27.
(5) (a) Stephens, D. N.; Schneider, H. H.; Kehr, W.; Andrews, J . S.;
Rettig, K.-J .; Turski, L.; Schmiechen, R.; Turner, J . D.; J ensen,
L. H.; Petersen, E. N.; Honore, T.; Bondo Hansen, J . Abecarnil,
a Metabolically Stable, Anxioselective â-Carboline Acting at
Benzodiazepine Receptors. J . Pharmacol. Exp. Ther. 1990, 253,
334-343. (b) Turski, L.; Stephens, D. N.; J ensen, L. H.; Petersen,
E. N.; Meldrum, B. S.; Patel, S.; Bondo Hansen, J .; Loscher, W.;
Schneider, H. H.; Schmiechen, R. Anticonvulsant Action of the
â-Carboline Abecarnil: Studies in Rodents and Baboon, Papio
papio. J . Pharmacol. Exp. Ther. 1990, 253, 344-352. (c) Bal-
lenger, J . C.; McDonald, S.; Noyes, R.; Rickels, K.; Sussman, N.;
Woods, S.; Patin, J .; Singer, J . The First Double-Blind, Placebo-
Controlled Trial of a Partial Benzodiazepine Agonist Abecarnil
(ZK 112-119) in Generalized Anxiety Disorder. Psychopharmacol.
Bull. 1991, 27, 171-179.
P h a r m a cok in etic Eva lu a tion . An im a l Stu d ies. Male
Sprague-Dawley rats, surgically prepared with indwelling
superior vena cava (SVC) cannulas54 were dosed the drug of
choice at a nominal dose of 5 mg/kg as an intravenous (iv) bolus
via the SVC or as an oral gavage (po). The following week,
the rats were administered the same drug at a nominal dose
of 5 mg/kg by the opposite route in a crossover fashion. The
volume of the dose was based on the weight of the rat on the
day of dosing and the measured concentration of the solution
formulation. The actual dose for each rat was calculated by
weighing the dosing syringe prior to and after the dosing. The
rats were fasted overnight prior to and 4 h post to drug
administration. Water was available ad libitum. Blood samples
(300 µL) were collected via the SVC before dosing (0 h) and at
2, 5, 10, and 30 min and at 1, 2, 4, 8, and 24 h after iv dosing,
and before dosing (0 h) and at 15 and 30 min and 1, 2, 4, 8,
and 24 h after po dosing. The blood was allowed to clot at room
temperature before centrifugation. The serum was transferred
to a clean labeled vial and stored at -15 to -25 °C until assay.
An a lytica l Meth od s. Unknown serum samples (100 µL)
were extracted using phenyl solid-phase extraction (SPE)
columns (100 mg/0.1 mL Varian Sample Preparation Products,
Harbor City, CA) with 1 mL of acetone, followed by 1 mL of
acetonitrile as elution solvents. The elution solvents were
taken to dryness in a TurboVap LV Evaporator (Zymark
Corporation, Hopkinton, MA) set at 40 °C. The drug and IS
were redissolved in 300 µL of acetonitrile/water (30/70, v/v).
The redissolved eluates were analyzed by reversed phase high-
performance liquid chromatography using a Zorbax SB-CN
column (250 mm × 4.6 mm o.d., 5 µM particle size, Dupont,
Wilmington, DE). The mobile phase was acetonitrile/water (28/
72, v/v, for 46, 47, and 91; 25/75, v/v, for 63; and 35/15, v/v,
for 55) containing 0.1% trifluoroacetic acid by volume. The
column effluent was monitored by a Spectra Physics Spectra
FOCUS UV detector at 230 nm. Quantitation was accom-
plished by peak height ratio of the drug to the internal
standard using a Harris NightHawk Computer System. The
calibration curve, along with a statistical evaluation of stan-
dards linear fit, were computed by a linear regression program.
The equation Y ‚ IS ) aX (forced through the origin) with no
(6) (a) Pieri, L.; Hunkeler, W.; J auch, R.; Merz, W. A.; Roncari, G.;
Timm, U. Ro 16-6028. Drugs Future 1988, 13, 730-735. (b)
Haefely, W.; Martin, J . R.; Schoch, P. Novel Anxiolytics That
Act as Partial Agonists at Benzodiazepine Receptors. Trends
Pharmacol. Sci. 1990, 11, 452-456.
(7) Morimoto, Y.; Fukuda, T.; Yakushiji, T.; Takehara, S.; Anami,
K.; Yamamoto, Y.; Yasumatsu, H.; Nakao, T.; Setoguchi, M.;
Tahar, T. Anxiolytic Properties of Y-23684, a Benzodiazepine
Receptor Partial Agonist in Comparison with Diazepam. Proc.
17th CINP Congr., Kyoto, 1990; Abstr. P-12-2-9, p 169.
(8) George, P.; Rossey, G.; Depoortere, H.; Mompon, B.; Allen, J .;
Wick, A. Imidazopyridines: Towards Novel Hypnotic and Anxi-
olytic Drugs. Farmaco 1991, 46 (suppl.), 277-288.

Piperazine Imidazo[1,5-a]quinoxaline Ureas
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7 1143
(9) (a) Tang, A. H.; Franklin, S. R.; Himes, C. S.; Ho, P. M.
Behavioral Effects of U-78875, a Quinoxalinone Anxiolytic with
Potent Benzodiazepine Antagonist Activity. J . Pharmacol. Exp.
Ther. 1991, 259, 248-254. (b) Petke, J . D.; Im, H. K.; Im, W. B.;
Blakeman, D. P.; Pregenzer, J . F.; J acobsen, E. J .; Hamilton, B.
J .; Carter, D. B. Characterization of Functional Interactions of
Imidazoquinoxaline Derivatives with Benzodiazepine-γ-Ami-
nobutyric AcidA Receptors. Mol. Pharmacol. 1992, 42, 294-301.
(10) Maryanoff, B. E.; Ho, W.; McComsey, D. F.; Reitz, A. B.; Grous,
P. P.; Nortey, S. O.; Shank, R. P.; Dubinsky, B.; Taylor, R. J .;
Gardocki, J . F. Potential Anxiolytic Agents. Pyrido[1,2-a]benz-
imidazoles: A New Structural Class of Ligands for the Benzo-
diazepine Binding Site on GABA-A Receptors. J . Med. Chem.
1995, 38, 16-20.
(11) (a) Pritchett, D. B.; Sontheimer, H.; Shivers, B. D.; Ymer, S.;
Kettenmann, H.; Schonfield, P. R.; Seeburg, P. H. Importance
of a Novel GABAA Receptor Subunit for Benzodiazepine Phar-
macology. Nature 1989, 338, 582-585. (b) Pritchett, D. B.;
Seeburg, P. H. γ-Aminobutyric Acid Type A Receptor Point
Mutation Increases the Affinity of Compounds for the Benzodi-
azepine Site. Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 1421-1425.
(12) (a) Skolnick, P.; Wong, G. In Imidazopyridines in Anxiety
Disorders: A Novel Experimental and Therapeutic Approach;
Zivkovic, B., Langer, S., Bartholini, G., Eds.; Raven Press: New
York, 1992. For a recent review, see: Yeh, H. H.; Grigorenko,
E. V. Deciphering the Native GABAA Receptor: Is There Hope?
J . Neurosci. Res. 1995, 41, 567-571.
(13) J acobsen, E. J .; TenBrink, R. E.; Stelzer, L. S.; Belonga, K. L.;
Carter, D. B.; Im, H. K.; Im, W. B.; Sethy, V. H.; Tang, A. H.;
VonVoigtlander, P. F.; Petke, J . D. High Affinity Partial Agonist
Imidazo[1,5-a]quinoxaline Amides, Carbamates and Ureas at the
GABAA/Benzodiazepine Receptor Complex. J . Med. Chem. 1996,
39, 158-175.
(14) Ulrich, R. G.; Cramer, C. T.; Bacon, J . A.; Funk, G. M.; Hunt,
C. E.; Kholmukhamedov, E.; Petrella, D. K.; Sun, E. L. Interfer-
ence with Hepatocelluar Metabolism by a Quinoxalinone Anxi-
olytic and Its Carboxylic Acid Metabolite: What Factors Con-
tribute Towards Cell Death? 1994 Annual Society of Toxicology
Meeting, Dallas, TX, Mar. 14-17, 1994.
(25) (a) Mickelson, J . W.; Belonga, K. L.; J acobsen, E. J . Asymmetric
Synthesis of 2,6-Methylated Piperazines. J . Org. Chem. 1995,
60, 4177-4183. (b) Mickelson, J . W.; J acobsen, E. J . Asymmetric
Synthesis of (2R,6R) and (2S, 6S)-2,6-Dimethylpiperazine. Tet-
rahedron: Asymmetry 1995, 6, 19-22.
(26) Mitsunobu, O. The Use of Diethyl Azodicarboxylate and Triph-
enylphosphine in Synthesis and Transformation of Natural
Products. Synthesis 1981, 1-28.
(27) The ratio of enantiomers was determined by HPLC analysis of
either the amine or Mosher52 amide using a Chiralpak AD
column (Daicel Chem. Ind., no. 53-45-20622) eluting with 30%
iPrOH/hexane or by 1H NMR analysis of the Mosher amide.
(28) Sethy, V. H.; Harris, D. W. Determination of Biological Activity
of Alprazolam, Triazolam and Their Metabolites. J . Pharm.
Pharmacol. 1982, 34, 115-116.
(29) Carter, D. B.; Thomsen, D. R.; Im, W. B.; Lennon, D. J .; Ngo, D.
M.; Gale, W.; Im, H. K.; Seeburg, P. H.; Smith, M. W. Functional
Expression of GABAA Chloride Channels and Benzodiazepine
Binding Sites in Baculovirus Infected Insect Cells. Bio/ Technol-
ogy 1992, 10, 679-681.
(30) Hamilton, B. J .; Lennon, D. J .; Im, H. K.; Im, W. B.; Seeburg,
P. H.; Carter, D. B. Stable Expression of Cloned Rat GABAA
Receptor Subunits in a Human Kidney Cell Line. Neurosci. Lett.
1993, 153, 206-209.
(31) (a) Pritchett, D. B.; Seeburg, P. H. γ-Aminobutyric AcidA
Receptor R5-Subunit Creates Novel Type II Benzodiazepine
Receptor Pharmacology. J . Neurochem. 1990, 54, 1802-1804.
(b) Pritchett, D. B.; Luddens, H.; Seeburg, P. H. Type I and Type
II GABAA-Benzodiazepine Receptors Produced in Transfected
Cells. Science 1989, 245, 1389-1392.
(32) Luddens, H.; Pritchett, D. B.; Kohler, M.; Killisch, I.; Keinanen,
K.; Monyer, H.; Sprengel, R.; Seeburg, P. H. Cerebellar GABAA
Receptor Selective for a Behavioral Alcohol Antagonist. Nature
1990, 346, 648-651.
(33) Im, W. B.; Blakeman, D. P. Correlation Between γ-Aminobutyric
AcidA Receptor Ligand-Induced Changes in tert-Butylbicyclo-
phosphoro[35S]thionate Binding and 36Cl- Uptake in Rat Cere-
brocortical Membranes. Mol. Pharmacol. 1991, 39, 394-398.
(34) Schwartz, R. D.; Suzdak, P. D.; Paul, S. M. γ-Aminobutyric Acid
(GABA)- and Barbiturate-Mediated 36Cl- Uptake in Rat Brain
Synaptoneurosomes: Evidence for Rapid Desensitization of the
GABA Receptor-Coupled Chloride Ion Channel. Mol. Pharmacol.
1986, 30, 419-426.
(15) J acobsen, E. J .; Stelzer, L. S.; Belonga, K. L.; Carter, D. B.; Im,
W. B.; Sethy, V. H.; Tang, A. H.; VonVoigtlander, P. F.; Petke,
J . D. 3-Phenyl Substituted Imidazo[1,5-a]quinoxalin-4-ones and
Imidazo[1,5-a]quinoxaline Ureas Which Have High Affinity at
the GABAA/Benzodiazepine Receptor Complex. J . Med. Chem.
1996, 39, 3820-3836.
(16) TenBrink, R. E.; Im, W. B.; Sethy, V. H.; Tang, A. H.; Carter,
D. B. Antagonist, Partial Agonist, and Full Agonist Imidazo-
[1,5-a]quinoxaline Amides and Carbamates Acting through the
GABAA/Benzodiazepine Receptor. J . Med. Chem. 1994, 37, 758-
768.
(35) Im, H. K.; Im, W. B.; Hamilton, B. J .; Carter, D. B.; VonVoigt-
lander, P. F. Potentiation of GABA-induced Chloride Currents
by Various Benzodiazepine Site Agonists with R1γ2, â2γ2 and
R1â2γ2 Subtypes of Cloned GABAA Receptors. Mol. Pharmacol.
1993, 44, 866-870.
(36) Puia, G.; Vicini, S.; Seeburg, P. H.; Costa, E. Influence of
Recombinant γ-Aminobutyric AcidA Receptor Subunit Composi-
tion on the Action of Allosteric Modulators of γ-Aminobutyric
AcidA-Gated Cl- Currents. Mol. Pharmacol. 1991, 39, 691-696.
(37) Rudzik, A. D.; Hester, J . B.; Tang, A. H.; Straw, R. N.; Friis, W.
(17) Borch, R. F.; Bernstein, M. D.; Durst, H. D. The Cyanoborohy-
dride Anion as a Selective Reducing Agent. J . Am. Chem. Soc.
1971, 93, 2897-2904.
Triazolobenzodiazepines,
A New Class of Central Nervous
System-Depressant Compounds. In The Benzodiazepines; Ga-
ratini, S., Mussini, E., Randall, L. O., Eds.; Raven Press: New
York, 1973; pp 285-297.
(18) (a) Imbach, J . L.; Katritzky, A. R.; Kolinski, R. A. The Confor-
mational Analysis of Heterocyclic Systems. Part VIII. Kinetics
of Quaternisation of N-Alkylpiperidines and N-Alkylpiperazines
in Acetonitrile. J . Chem. Soc. B 1966, 556-562. (b) Cook, M. J .;
J ones, R. A. Y.; Katritzky, A. R.; Moreno Manas, M.; Richards,
A. C.; Sparrow, A. J .; Trepanier, D. L. The Conformational
Analysis of Saturated Heterocycles. Part XLIX. The Conforma-
tion of Ring NH-Groups in Piperazines, Hexahydropyrimidines,
Tetrahydro-1,3-oxazines, and Tetrahydro-1,3-thiazines. J . Chem.
Soc., Perkin II 1973, 325-331.
(38) Kyburz, E. Serendipity in Drug Discovery. The Case of BzR
Ligands. Il Farmaco, 1989, 44, 345-382.
(39) Im, H. K.; Im, W. B.; Pregenzer, J . F.; Carter, D. B.; J acobsen,
E. J .; Hamilton, B. J . Characterization of U-97775 as a GABAA
Receptor Ligand of Dual Functionality in Cloned Rat GABAA
Receptor Subtypes. Brit. J . Pharmacol. 1995, 115, 19-24.
(40) (a) Sethy, V. H.; Francis, J . W.; Elfring, G. Onset and Duration
of Action of Benzodiazepines as Determined by Inhibition of (3H)-
Flunitrazepam Binding. Drug Dev. Res. 1987, 10, 117-121. (b)
Sethy, V. H.; Daenzer, C. L.; Russell, R. R. Interaction of
N-Alkylaminobenzophenones with Benzodiazepine Receptors. J .
Pharm. Pharmacol. 1983, 35, 194-195.
(19) Braish, T. F.; Fox, D. E. Synthesis of (S,S)- and (R,R)-2-Alkyl-
2,5-diazabicyclo[2.2.1]heptanes. J . Org. Chem. 1990, 55, 1684-
1687.
(20) (a) Cignarella, G.; Nathansohn, G. Bicyclic Homologues of
Piperazine. Synthesis of 8-Methyl-3,8-diazabicyclooctanes. I. J .
Org. Chem. 1961, 26, 1500-1504. (b) Cignarella, G.; Nathansohn,
G.; Occelli, E. Bicyclic Homologues of Piperazine. II. Synthesis
of 3,8-Diazabicyclo[3.2.1]octane. New Synthesis of 8-Methyl-3,8-
diazabicyclo[3.2.1]octane. J . Org. Chem. 1961, 26, 2747-2751.
(c) Blackman, S. W.; Baltzly, R. The Synthesis of 3,8-Diazabicyclo-
[3.2.1]octane and Some of Its N-Substituted Derivatives. J . Org.
Chem. 1961, 26, 2750-2755.
(21) Aspinall, S. R. A Synthesis of Monoketopiperazines. J . Am.
Chem. Soc. 1940, 62, 1202-1204.
(22) Ishiguro, T.; Matsumura, M. Synthesis of Piperazines. VIII.
Synthesis of R(trans)- and â(cis)-2,3-Dimethylpiperazines.
Yakugaku Zasshi 1958, 78, 229-231.
(23) Yoshioka, T.; Mori, E.; Murayama, K. Studies on Stable Free
Radicals. XIII. Synthesis and ESR Spectral Properties of Hin-
dered Piperazine N-Oxyls. Bull. Chem. Soc. J pn. 1972, 45,
1855-1860.
(24) Sudo, R.; Ichihara, S. Reactions of Alicyclic Aminonitrile. Bull.
Chem. Soc. J pn. 1963, 36, 34-37.
(41) (a) Dinnendahl, V. Effect of Stress on Mouse Brain Cyclic
Nucleotide Levels In Vivo. Brain Res. 1975, 100, 716-719. (b)
Dinnendahl, V.; Gumulka, S. W. Stress-induced Alterations of
Cyclic Nucleotide Levels in Brain: Effect of Centrally Acting
Drugs. Psychopharmacology 1977, 52, 243-249. (c) Goldberg,
N. D.; Haddox, M. K.; Hartle, D. K.; Hadden, J . W. The Biological
Role of Cyclic 3′,5′-Guanosine Monophosphate. In Pharmacology
and the Future of Man; Maxwell, R. A., Acheson, G. H., Eds.;
Karger: Basel, 1972; pp 146-169.
(42) Sethy, V. H.; Oien, T. T. Role of cGMP in the Mechanism of
Anxiolytic Activity of U-78875. Pharmacol. Biochem. Behavior.
1991, 39, 379-382.
(43) (a) VonVoigtlander, P. F.; Lewis, R. A. A Rapid Screening Method
for the Assessment of Benzodiazepine Receptor-Related Physical
Dependence in Mice. J . Pharmacol. Methods 1991, 26, 1-5. (b)
Fialip, J .; Aumaitre, O.; Eschalier, A.; Maradeix, B.; Dordain,
G.; Lavarenne, J . Benzodiazepine Withdrawal Seizures: Analy-
sis of 48 Case Reports. Clin. Neuropharmacol. 1987, 10, 538-
544.

1144 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 7
J acobsen et al.
(44) Mohamadi, F.; Richards, N. G. J .; Guida, W. C.; Liskamp, R.;
Lipton, M.; Caufield, C.; Chang, G.; Hendrickson, T.; Still, W.
C. MacroModel-An Integrated Software System for Modeling
Organic and Bioorganic Molecules Using Molecular Mechanics.
J . Comput. Chem. 1990, 11, 440-467.
(51) Zhang, W.; Koehler, K. F.; Zhang, P.; Cook, J . M. Development
of a Comprehensive Pharmacophore Model for the Benzodiaz-
epine Receptor. Drug Des. Discovery 1995, 12, 193-248.
(52) Dale, J . A.; Mosher, H. S. Nuclear Magnetic Resonance Enan-
tiomer Reagents. Configurational Correlations via Nuclear
Magnetic Resonance Chemical Shifts of Diastereomeric Man-
delate, O-Methylmandelate, and R-Methoxy-R-trifluorometh-
ylphenylacetate (MTPA) Esters. J . Am. Chem. Soc. 1973, 95,
512-519.
(53) Burkard, W. P.; Bonetti, E. P.; Haefely, W. The Benzodiazepine
Antagonist Ro 15-1788 Reverses the Effect of Methyl-â-carboline-
3-carboxylate but not of Harmaline on Cerebellar cGMP and
Motor Performance in Mice. Eur. J . Pharmacol. 1985, 109, 241-
247.
(45) Weiner, S. J .; Kollman, P. A.; Nguyen, D. T.; Case, D. A. An All
Atom Force Field for Simulations of Proteins and Nucleic Acids.
J . Comput. Chem. 1986, 7, 230-252.
(46) Allinger, N. L. Conformational Analysis. 130. MM2. A Hydro-
carbon Force Field Utilizing V1 and V2 Torsional Terms. J . Am.
Chem. Soc. 1977, 99, 8127-8134.
(47) Chang, G.; Guida, W. C.; Still, W. C. An Internal Coordinate
Monte Carlo Method for Searching Conformational Space. J . Am.
Chem. Soc. 1989, 111, 4379-4386.
(48) Stewart, J . J . P. MOPAC, A Semiempirical Molecular Orbital
Program. J . Comput.-Aided Mol. Des. 1990, 4, 1-105.
(49) Dewar, M. J . S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J . J . P.
AM1: A New General Purpose Quantum Mechanical Molecular
Model. J . Am. Chem. Soc. 1985, 107, 3902-3909.
(50) Tebib, S.; Bourguignon, J .-J .; Wermuth, C.-G.; The Active
Analogue Approach Applied to the Pharmacophore Identification
of Benzodiazepine Receptor Ligands. J . Comput.-Aided Mol. Des.
1987, 1, 153-170.
(54) Waynforth, H. B. In Experimental and Surgical Technique in
the Rat; Academic Press: London, 1980; pp 50-57, 72-74.
(55) Gibaldi, M.; Perrier, D. In Pharmacokinetics, 2nd ed.; Swarbrick,
J ., Ed.; Marcel Dekker: New York, 1982; pp 5, 411.
J M9801307