High-affinity partial agonist imidazo[1,5-a]quinoxaline amides, carbamates, and ureas at the γ-aminobutyric acid A/benzodiazepine receptor complex

158

J . Med. Chem. 1996, 39, 158-175

High -Affin ity P a r tia l Agon ist Im id a zo[1,5-a ]qu in oxa lin e Am id es, Ca r ba m a tes,

a n d Ur ea s a t th e γ-Am in obu tyr ic Acid A/Ben zod ia zep in e Recep tor Com p lex^†

E. J on J acobsen,*^,‡Ruth E. TenBrink,^‡Lindsay S. Stelzer,^‡Kenneth L. Belonga,^‡Donald B. Carter,^§

Haesook K. Im,^§Wha Bin Im,^§Vimala H. Sethy,^§Andy H. Tang,^§Philip F. VonVoigtlander,^§and

J ames D. Petke^|

Departments of Medicinal Chemistry, Central Nervous System Diseases Research, and Computational Chemistry,

Upjohn Laboratories, The Upjohn Company, Kalamazoo, Michigan 49001

Received November 14, 1994^X

A series of imidazo[1,5-a]quinoxaline amides, carbamates, and ureas which have high affinity

for the γ-aminobutyric acid A/benzodiazepine receptor complex was developed. Compounds

within this class have varying efficacies ranging from antagonists to full agonists. However,

most analogs were found to be partial agonists as indicated by [³⁵S]TBPS and Cl^-current ratios.

Many of these compounds were also effective in antagonizing metrazole-induced seizures in

accordance with anticonvulsant and possible anxiolytic activity. Selected quinoxalines displayed

limited benzodiazepine-type side effects such as ethanol potentiation and physical dependence

in animal models. Dimethylamino urea 41 emerged as the most interesting analog, having a

partial agonist profile in vitro while possessing useful activity in animal models of anxiety

such as the Vogel and Geller assays. In accordance with its partial agonist profile, 41 was

devoid of typical benzodiazepine side effects.

GABA (γ-aminobutyric acid) is the major inhibitory

neurotransmitter in the brain, controlling the excit-

ability of many central nervous system (CNS) pathways.

The principal mode of action for this neurotransmitter

occurs by modulation of the GABA_Achloride ion channel

complex.¹Numerous chemical classes such as the

benzodiazepines, neurosteroids, and barbiturates have

binding sites on this macromolecular complex and

allosterically modulate the action of GABA on neuronal

chloride flux. Of these, compounds which mediate their

actions at the benzodiazepine receptor (BzR) are the

most widely studied. Ligands which interact at the

benzodiazepine (BZD) site have a continuum of intrinsic

activity,²ranging from full agonists (anxiolytic, hyp-

notic, and anticonvulsant agents) through antagonists

to inverse agonists (proconvulsant and anxiogenic

agents). The possibility of partial agonists existing

within this continuum is intriguing, as they may be

devoid of typical benzodiazepine-mediated side effects

such as physical dependence, amnesia, oversedation

(anxiolytics), and muscle relaxation or, in the case of

inverse agonists, convulsive activity.³Furthermore,

recent molecular biology studies have demonstrated that

several different receptor subunits (R, â, γ, δ) combine

to form the GABA_Areceptor complex,⁴with at least

three to four native receptor subtypes identified thus

far.⁵Subtype selective ligands may also allow for the

discrimination between useful anxiolytic or hypnotic

activity and overt side effects.

of anxiety disorders. Preliminary results indicate that

benzodiazepine-type side effects are lessened for some

of these agents, which is consistent with the animal

pharmacology. As part of a general structure-activity

relationship (SAR) study of U-78875, which was re-

moved from clinical trials due to liver enzyme induction,

a new class of imidazo[1,5-a]quinoxalines was discov-

ered.^12,13Leads from this series, as exemplified by

structures 1 and 2 (Figure 1), have high affinity for the

benzodiazepine receptor complex. More importantly,

compounds from this class had varying efficacy, span-

ning a range from antagonists to full agonists, with

several compounds displaying in vitro properties con-

sistent with that of a partial agonist.

In order to expand the scope of this series, we were

especially interested in modification of the substituent

at the 5-position. As reported previously,^12,13relatively

large amide and carbamate substituents could be in-

corporated into the 5-position of 2. Most interesting was

the possibility of incorporating other substituents at this

position to both modify the electronic density and

rotational direction of the carbonyl group as well as to

further explore the steric requirements of this site.

Additional exploration of this series where the 5-sub-

stituent was varied to also include ureas and thio-

carbamates, and where the A-ring was modified to

include 6- and 7-chloro groups was carried out. The 5′-

oxadiazole substituent was also varied to include other

alkyl groups. Many compounds within this expanded

series were partial agonists as determined by in vitro

analysis. Furthermore, selected urea analogs had good

activity in various animal models of anxiety yet dis-

played limited benzodiazepine-type side effects. Herein

we would like to describe the synthesis, in vitro efficacy,

and pharmacology of this unique class of imidazo[1,5-

a]quinoxalines. Efforts directed toward the further

exploration of the tetracyclic ring system (1), where the

electronic density and orientation of the carbonyl group

is varied systematically, will be reported on shortly.

Several compounds that are reported to be partial

agonists at the benzodiazepine receptor^6-11are shown

in Chart 1. Most of these compounds, abecarnil,⁶

bretazenil,⁷alpidem,⁹panadiplon¹⁰(U-78875), and DN-

2327,¹¹have undergone clinical trials for the treatment

^†A portion of this material was presented at the 207th American

Chemical Society National Meeting, March 13-17, 1994.

^‡Department of Medicinal Chemistry.

^§Department of Central Nervous System Diseases Research.

^|Department of Computational Chemistry.

^XAbstract published in Advance ACS Abstracts, November 15, 1995.

Partial Agonists at GABA_A/ Benzodiazepine Receptor

J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 159

Ch a r t 1

quinoxalines prepared by these general methods. Physi-

cal data for 33, 35, 48, 66, and 71 were reported

previously.¹²

Several of the requisite quinoxaline starting materials

(3a -c,g) were synthesized as previously reported.¹²The

corresponding 5- and 6-chloro analogs were prepared

as shown in Scheme 3. Following the procedure of

Lumma,¹⁵imine 14 was prepared by reaction of 13 with

glyoxylic acid and subsequently reduced with sodium

borohydride to provide the 6-chloro isomer 15. The

5-chloroquinoxaline could not be prepared by this

sequence due to poor regioselectivity in the imine

cyclization step. Instead, addition of tert-butylamine to

2,3-dichloronitrobenzene (16) provided nitro 17, which

was reduced with Raney nickel to provide diamine 18.

Acylation of 18 with chloroacetyl chloride and subse-

quent cyclization provided quinoxaline 19. Deprotection

by treatment of 19 with 2 N sulfuric acid provided a

mixture of imine 20 and amine 21 (1:23). The crude

mixture was treated with sodium borohydride to give

the 5-chloroquinoxaline template (21).

F igu r e 1.

Ch em istr y

The general synthesis of the imidazo[1,5-a]quinox-

alines 7, 8, and 10 is shown in Scheme 1. Two methods

are applicable, depending on the particular substitution

patterns. Reaction of the desired quinoxaline template¹²

3 with potassium tert-butoxide, followed by diethyl

chlorophosphate, provided an intermediate enol phos-

phonate, 4. Without isolation, enol 4 was combined with

the desired isocyanide¹⁴5 and additional potassium tert-

butoxide to provide the imidazo[1,5-a]quinoxaline 6 ring

system. Amides of 6 were formed under standard

conditions to provide 7, whereas selected ureas 8 were

formed by reaction of 6 with the desired isocyanate

(neat) or carbamoyl chloride. A more general synthesis

of urea, carbamate, and thiocarbamate analogs was also

developed. Reaction of amine 6 with phosgene or

triphosgene in the presence of Hunig’s base provided

carbamoyl chloride 9, which was sufficiently stable for

long-term storage. However, 9 was most conveniently

generated and reacted with the desired nucleophile

(amines, alcohols, thiols) in situ to provide the appropri-

ate analogs 10.

The gem-3,3-dimethyl-6-chloro template was synthe-

sized as shown in Scheme 4. While 2-amino-2-methyl-

propionic acid would not add to 2,4-dichloronitrobenzene

(22), 2-amino-2-methyl-1-propanol (neat, 110 °C) under-

went the aromatic nucleophilic substitution in excellent

yield to give 23. J ones oxidation following the procedure

of Djerassi¹⁶provided acid 24, which was esterified (MeI,

K₂CO₃) to give ester 25. Reduction and in situ cycliza-

tion of 25 proceeded well, either by reaction with

titanium trichloride or by catalytic hydrogenation using

sulfur-poisoned Pt/C, to give the quinoxaline template

26.

An alternative route to gem-4-dimethyl analogs of the

imidazo[1,5-a]quinoxalines, which was found to be

particularly useful (but not required), is shown in

Scheme 2. In this sequence amine 3 was functionalized

at the 5-position with the imidazole ring-forming reac-

tion carried out subsequently. Thus, the carbamoyl

chloride of 3 was formed by reaction with phosgene to

provide 11, which could then be reacted directly with

the desired nucleophile to give 12. The imidazole

annulation sequence was then carried out as described

above to provide the target compound 10. Table 1

highlights the physical data of the imidazo[1,5-a]-

A similar cyclization strategy was used for the syn-

thesis of the 5-methyl template 29 as shown in Scheme

5. Alkylation of 2-methyl-6-nitroaniline (27) with ethyl

bromoacetate provided nitro ester 28. Reduction of the

nitro group and in situ cyclization as described above

under catalytic hydrogenation conditions resulted in 29.

The oxadiazole isocyanide reagents were synthesized

following the general procedure of Watjen¹⁴as il-

lustrated in Scheme 6. Acylation of amide oxime 30

with the desired acid halide and cyclization of the

160 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1

J acobsen et al.

Sch em e 1^a

^aReagents: (i) (1) 1 N tBuOK, THF, (2) diethyl chlorophosphate; (ii) 1 N tBuOK, THF; (iii) R⁵COCl or (R⁵CO)₂O, THF or CH₂Cl₂,

EtNiPr₂; (iv) R²NCO or R²R³NCOCl, CH₂Cl₂, EtNiPr₂; (v) phosgene in toluene or triphosgene, CH₂Cl₂or THF, EtNiPr₂; (vi) HNu, EtNiPr₂.

intermediate by heating in water at reflux provided

oxadiazole 31 in good yield. Dehydration of the forma-

mide using phosphorus oxychloride gave the required

isocyanide reagents.

drug to that of diazepam, is 1 for a full agonist, 0 for an

antagonist, and negative values for inverse agonists. A

second and more direct measure of in vitro efficacy was

determined by a ³⁶Cl^-uptake assay.^18-20The synaptic

chloride conductance effected by GABA activating the

GABA_Areceptor complex is modulated by ligands acting

at the benzodiazepine receptor. Full agonists increase

current, and antagonists have no effect, while inverse

agonists decrease ion flow. The test compounds are

compared to diazepam, and thus like the TBPS assay,

a full agonist has a value of 1 or greater, antagonists

having no effect (0), while inverse agonists have nega-

tive values. To provide a quick measure of in vivo

efficacy, most analogs were evaluated for their ability

to antagonize metrazole-induced seizures (clonic and

tonic).²¹While 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 are

extremely effective in this assay, as are partial agonists

Resu lts a n d Discu ssion

The binding affinity of the imidazo[1,5-a]quinoxalines

at the benzodiazepine receptor in rat cortical mem-

branes was determined by competition experiments

with radiolabeled [³H]flunitrazepam.¹⁷The in vitro

efficacy of these compounds was measured by two

different methods. The TBPS ratio¹⁸was determined

for each compound by measuring its effect on tert-butyl

bicyclophosphorothionate (TBPS) binding to the picro-

toxin convulsant site on the GABA_Achloride complex.

Changes in TBPS binding presumably occur due to

conformational changes in the chloride ionophore caused

by allosteric modulation effected by the binding of the

test compound to the benzodiazepine receptor. The

resultant value, expressed as a ratio of that for the test

Partial Agonists at GABA_A/ Benzodiazepine Receptor

J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 161

Sch em e 2^a

^aReagents: (i) phosgene in toluene, CH₂Cl₂or THF, EtNiPr₂; (ii) HNu, EtNiPr₂; (iii) 1 N tBuOK, THF, diethyl chlorophosphate; (iv) 5,

1 N tBuOK, THF.

Ta ble 1. Physical Data for Imidazo[1,5-a]quinoxaline Amides, Carbamates, Ureas, and Thiocarbamates

compd

R⁴

R⁵

R⁶

R⁷

mp (°C)

method

yield (%)

formula

anal.

34

36

37

38

39

40

41

42

43

44

45

46

47

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

67

68

69

70

72

73

74

75

76

77

H

CF₃

2-pyridine

2-furan

2-pyrrole

NH₂

NHMe

NMe₂

NHEt

NEt₂

NHiPr

pyrrolidine

morpholine

aniline

OMe

OiPr

SMe

SEt

SPh

NMe₂

H

F

H

Cl

H

F

H

Cl

H

F

H

Me

H

F

H

Cl

H

F

H

Cl

H

F

Cl

H

Cl

H

175-177

239-240

190-191

252-253

114-115

214-215

196-197

189-190

173-174

210-211

181.5-182.5

198-198.5

211-212.5

187-188

207-208

176-177

169-170

222.5-223.5

194-195

227-228.5

226-228

254.5-255.5

220.5-221.5

203-204

243-244.5

202-204

198.5-201

224-225

209-210

265-265.5

168-169

126-127

160-161

200.5-201.5

163-164

162-165

207-208

190.5-191

195-196

226-227.5

C

A

I, J

A

F

D

B

F

B

D

F

D

E

G

F

H, J

F

H, J

F

H, J

F

G

71

30

70, 47

64

88

83

49

75

37

58

92

85

67

65

58

95

69

86

88

81

63

80, 55

92

82

75

82, 54

87

75

57, 59

91

81

92

75

93

C₁₇H₁₂N₅O₂F₃

C, H, N, F

C, H, N

C

₂₁H₁₆N₆O₂‚¹/₄H₂O

₂₀H₁₅N₅O₃‚¹/₄H₂O

₂₀H₁₆N₆O₂

C₁₆H₁₄N₆O₂

₁₇H₁₆N₆O₂‚¹/₈H₂O

C₁₈H₁₈N₆O₂‚¹/₄H₂O

₁₈H₁₈N₆O₂

C₂₀H₂₂N₆O₂‚¹/₄H₂O

₁₉H₂₀N₆O₂‚H₂O

C

C, H, N

C

C, H, N^a

C, H, N

C₂₀H₂₀N₆O₂

C

₂₀H₂₀N₆O₃‚¹/₃H₂O

₂₂H₁₈N₆O₂

₁₇H₁₅N₅O₃

C, H, N

C, H, N, S

C, H, N, S^b

C, H, N, S^c

C, H, N

₁₉H₁₉N₅O₃

₁₇H₁₅N₅O₂S

₁₈H₁₇N₅O₂S‚¹/₄H₂O

C₂₂H₁₇N₅O₂S‚¹/₂H₂O

C₁₈H₁₇N₆O₂F

NMe₂

C₁₈H₁₇N₆O₂F‚¹/₄H₂O

C₁₈H₁₇N₆O₂Cl

C, H, N

C, H, N, Cl

C, H, N

C₁₈H₁₇N₆O₂Cl‚¹/₂₀CH₂Cl₂

pyrrolidine

morpholine

NH₂

NHMe

NMe₂

pyrrolidine

SMe

NMe₂

C

₂₀H₁₉N₆O₂F

₂₀H₁₉N₆O₂Cl

C, H, N

C, H, N, Cl

C, H, N

C, H, N, Cl

C, H, N

₂₀H₁₉N₆O₃F

₂₀H₁₉N₆O₃F‚¹/₃H₂O

₂₀H₁₉N₆O₃Cl‚¹/₈H₂O

₂₁H₂₂N₆O₃‚¹/₈H₂O

Me

H

C₁₈H₁₈N₆O₂‚³/₈H₂O

C

C, H, N

₁₉H₂₀N₆O₂

C₂₀H₂₂N₆O₂

C

₂₂H₂₄N₆O₂

₁₉H₁₉N₅O₂S‚¹/₈H₂O

C, H, N, S^d

C, H, N, Cl

C, H, N

H, J

71, 70

73, 68

73, 44

73, 62

C₂₀H₂₁N₆O₂Cl

pyrrolidine

pyrrolidine (R^5′) Et)

pyrrolidine (R^5′) iPr)

pyrrolidine (R^5′) tBu)

C₂₂H₂₃N₆O₂Cl

C₁₉H₂₀N₆O₂

C₂₀H₂₂N₆O₂

C₂₁H₂₄N₆O₂

H

C, H, N

C,H,N

a

b

d

N: calcd, 21.98; found, 21.50. S: calcd, 8.62; found, 8.02. ^cS: calcd, 7.55; found, 7.09. S: calcd, 8.36; found, 7.87.

U-78875, bretazenil, and abecarnil,²²although to a

lesser degree.

The effects of substituents at the 5-position in the

simple unsubstituted ring system were examined ini-

162 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1

J acobsen et al.

Sch em e 3^a

^aReagents: (i) glyoxylic acid, MeOH; (ii) NaBH₄, EtOH; (iii) tBuNH₂, EtOH; (iv) hydrazine hydrate, Raney Ni, EtOH; (v) (1) ClCH₂CO₂Cl,

EtNiPr₂, THF, (2) EtNiPr₂, CH₃CN, NaI; (vi) 2 N H₂SO₄; (vii) NaBH₄, EtOH.

Sch em e 4^a

cant loss of affinity, whereas less extended but bulky

groups (tBu, phenyl, morpholine) were well tolerated.

While changes in the 5-substituent had negligible effects

on affinity, dramatic differences in efficacy between

compounds were noted. By TBPS measurement, most

of the amide derivatives were full agonists, particularly

those with an electron-rich heterocycle (37, 38). How-

ever, the simple benzamide 35 was a partial agonist,

while the 2-pyridyl analog 36 was an inverse agonist.

The chloride current value for 38 was consistent with

this compound being a full agonist, as was its activity

in the metrazole antagonism assay. The remaining

amide derivatives were substantially less effective as

metrazole antagonists, in contrast to what their in vitro

activity would predict. As a class the urea derivatives

were more consistent in this assay. The in vitro efficacy

for these compounds varied from partial agonists to full

agonists based on both the TBPS and chloride current

assays. By TBPS measurement, several compounds

were full agonists, with 47 having the greatest efficacy.

All of the urea analogs were partial agonists in the

chloride current screen. No general trends between

efficacy and size or the degree of N-alkyl substitution

were obvious. In the metrazole assay, the cyclic ureas

were the most potent anticonvulsant agents followed by

the monoalkyl urea analogs, with the disubstituted

compounds least effective. Within the urea subseries,

several analogs (39, 41, 42, 45, 46) stood out as having

excellent metrazole activity combined with partial ago-

nist properties in vitro. Analogs in the carbamate and

thiocarbamate classes tended to be full agonists as

indicated by both the TBPS and chloride current assays.

The tert-butyl carbamate analog 48 by the TBPS assay

had 3-fold greater efficacy than diazepam, further

emphasizing the high degree of agonism for this sub-

series. The metrazole antagonism screen was consistent

with the in vitro activity, as 48 was extremely effective

(0.06 mg/kg). The other carbamate and thiocarbamate

analogs were also quite active as metrazole antagonists.

Interestingly, thiophenol analog 53, like the aniline

^aReagents: (i) 2-amino-2-methyl-1-propanol, 110 °C; (ii) H₂CrO₄,

acetone; (iii) MeI, K₂CO₃, DMF; (iv) TiCl₃, H₂O, NaOAc, MeOH;

(v) sulfided Pt on carbon, EtOH.

Sch em e 5^a

^aReagents: (i) BrCH₂CO₂Et, EtNiPr₂, 140 °C; (ii) H₂, Pd/C,

EtOH.

tially. As noted in Table 2, the amides, carbamates,

ureas, and thiocarbamates had relatively high affinity

for the benzodiazepine receptor on the GABA_Acomplex,

ranging from 0.43 to 16 nM and comparing favorably

to standards such as diazepam and zolpidem. The only

exception proved to be the electronegative trifluoro-

methyl acetamide 34. Other substituents at this posi-

tion varied substantially in electron-donating character

and size yet resulted in only minor changes in affinity.

Only the aniline and thiophenol analogs had a signifi-

Partial Agonists at GABA_A/ Benzodiazepine Receptor

J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 163

Sch em e 6^a

a

Reagents: (i) RCOCl, CH₂Cl₂, Et₃N; (ii) H₂O, ∆; (iii) POCl₃, Et₃N, CH₂Cl₂.

Ta ble 2. [³H]Fnz Binding, TBPS Shift, ³⁶Cl^-Uptake, and Metrazole Antagonism Data for Imidazo[1,5-a]quinoxaline Amides,

Carbamates, Ureas, and Thiocarbamates

metrazole^dED₅₀

compd

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

R⁵

K_i(nM)^a

[

³⁵S]TBPS shift^b,c

³⁶Cl^-uptake^b,c

(mg/kg, ip)

Me

CF₃

Ph

2-pyridyl

2-furan

2-pyrrole

NH₂

NHMe

NMe₂

NHEt

1.5

143

4.3

1.1

0.43

1.1

7.2

2.8

1.0

3.5

0.85

3.4

0.45

0.81

16

0.94

0.73

0.54

-0.27

0.98

0.92

0.61

0.95

0.69

0.72

0.82

1.1

0.41

0.84

1.3

2.9

0.7

35

42

>50

21

0.55

1.1

4.4

0.75

0.53

0.68

0.22

0.62

13

11

21

5.3

15

11

0.24

0.84

>50

NEt₂

NHiPr

pyrrolidine

morpholine

aniline

OtBu

OMe

OiPr

SMe

SEt

0.50

0.15

0.29

48

49

50

51

52

53

1.9

1.5

1.8

1.3

2.1

11

0.06

4.4

1.9

6.2

1.3

1.1

0.92

1.1

1.0 ( 0.2

0.06

0.73

1.1

0.50

1.0 ( 0.15

0.04

SPh

>50

diazepam

U-78875

4.9

1.6

0.50 (0.38-1.2)

1.6

a

Mean binding affinity against [³H]flunitrazepam; see ref 17 and the Experimental Section for methods. The standard error was

b

<(10% 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. ^cSee the Experimental Section. Antagonism of metrazole-induced clonic convulsions in the rat

d

after ip injection; see the Experimental Section.

analog in the urea series, was completely inactive in the

metrazole screen, indicating possible rapid metabolism

or poor CNS bioavailability for these aryl analogs.

As selected urea analogs had a desirable partial

agonist profile combined with good activity in the

primary in vivo screen, further modifications at other

positions on this template usually maintained the

5-substituent as either a dimethylamino, pyrrolidino,

or morpholino urea. As shown in Table 3, the incorpo-

ration of either a fluoro or chloro group at the 6- or

7-position had little effect on affinity. Only the 7-chloro

analogs show a slight decrease in binding affinity within

each urea subseries. With the dimethylamino and

pyrrolidino urea analogs, halogenation generally in-

creased efficacy by TBPS measurement, except for the

6-chloro derivatives, which had similar values to the

unsubstituted parents. The 7-fluoro and 7-chloro de-

rivatives generally had higher chloride current ratios

than the 6-halo or unsubstituted analogs. In the

metrazole antagonism screen, most of the halogenated

analogs in this subseries were much less effective than

the parent compounds, even though their in vitro

efficacy was often greater. The only exceptions were

54 and 61, which were roughly equivalent to the

unsubstituted analogs. As a class, the morpholine ureas

were somewhat dissimilar. By TBPS measurement,

only the 7-fluoro analog 63 had greater efficacy than

the parent. In contrast, the halogenated and 6-methyl

analogs all had greater efficacy by chloride current

measurement, with 63 and 64 being full agonists. In

addition, all the morpholine derivatives tested were

active in antagonizing metrazole-induced seizures (al-

though to a lesser degree than unsubstituted 46), in

contrast to the dimethylamino and pyrrolidino analogs.

The 6-methyl (65) and 6-fluoro (62) analogs were

particularly impressive having potent metrazole activ-

ity, greater than expected considering their in vitro

partial agonist profile. Nonetheless, within this series,

substitution at the 6- or 7-position with chloro, fluoro,

or methyl groups, while often providing increased ef-

ficacy, usually resulted in diminished activity in vivo.

Biological data for compounds incorporating gem-

dimethyl groups at the 4-position are provided in Table

4. As compared to the unsubstituted analogs, gem-

164 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1

J acobsen et al.

Ta ble 3. [³H]Fnz Binding, TBPS Shift, ³⁶Cl^-Uptake, and Metrazole Antagonism Data for Imidazo[1,5-a]quinoxaline Ureas with

Fluoro, Chloro, or Methyl A-Ring Substituents

metrazole^dED₅₀

compd

R⁵

NMe₂

pyrrolidine

morpholine

R⁶

R⁷

K_i(nM)^a

[

³⁵S]TBPS shift^b,c

³⁶Cl^-uptake^b,c

(mg/kg, ip)

41

54

55

56

57

45

58

59

60

61

46

62

63

64

65

H

F

H

F

H

Cl

H

F

H

Cl

H

F

1.0

0.69

0.84

0.99

0.62

0.88

0.41

1.0

1.25

0.60

0.93

0.84

0.66

0.96

0.71

0.51

0.68

0.60

0.95

0.83

21

0.89

0.61

0.71

3.7

0.45

0.64

0.56

0.73

1.1

0.81

0.48

0.68

2.3

H

Cl

H

F

H

Cl

H

F

>50

0.24

>50

0.60

0.84

6.2

0.59

0.69

0.56

0.15

0.42

1.1

1.0

0.70

H

Me

6.2

Cl

H

4.6

5.2

a

Mean binding affinity against [³H]flunitrazepam; see ref 17 and the Experimental Section for methods. The standard error was

b

<(10% 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. ^cSee the Experimental Section. Antagonism of metrazole-induced clonic convulsions in the rat

d

after ip injection; see the Experimental Section.

Ta ble 4. [³H]Fnz Binding, TBPS Shift, ³⁶Cl^-Uptake, and Metrazole Antagonism Data for Imidazo[1,5-a]quinoxaline Amides,

Carbamates, Ureas, and Thiocarbamates with gem-4,4-Dimethyl Substitution

metrazole^dED₅₀

compd

R⁵

R⁷

K_i(nM)^a

[

³⁵S]TBPS shift^b,c

³⁶Cl uptake^b,c

(mg/kg, ip)

66

67

68

69

70

71

72

73

74

Me

NH₂

NHMe

NMe₂

pyrrolidine

OMe

SMe

NMe₂

pyrrolidine

H

Cl

4.9

86

21

3.1

2.9

6.8

5.1

3.1

3.3

0.70

0.60

0.71

0.97

0.84

0.8

0.80

0.79

0.67

30

>50

7.4

0.22

0.67

0.34

0.50

0.4

1.0

1.33

1.04

35

a

Mean binding affinity against [³H]flunitrazepam; see ref 17 and the Experimental Section for methods. The standard error was

b

<(10% 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. ^cSee the Experimental Section. Antagonism of metrazole-induced clonic convulsions in the rat

d

after ip injection; see the Experimental Section.

dimethyl substitution lowered affinity substantially,

between 3- and 12-fold. The only exception proved to

be 73, which had a K_iof 3.1 nM, roughly comparable to

that of 57. In contrast, the primary urea analog 67

suffered the greatest loss of affinity with increased

substitution. The effects of the gem-dimethyl groups on

efficacy were mixed. By TBPS shift measurement, both

69 and 70 had significantly increased efficacy, while

most of the other analogs displayed little change over

the unsubstituted parent compounds. Interestingly, the

methylurea analog 68 had significantly lower efficacy

as compared to 40. A lack of a definitive trend was also

noted for these analogs in the chloride current assay.

However, even though gem-dimethyl substitution usu-

ally maintained or increased efficacy by in vitro analy-

sis, almost all of the compounds were inactive in the

metrazole screen. The only effective compound was 73,

with an ED₅₀of 7.4 mg/kg, while 74 and 66 were

marginally active, although 74 was 50-fold less potent

than the unsubstituted analog. For the most part,

substitution with methyl groups at the 4-position had

a minimal effect on in vitro activity while having a

deleterious effect on in vivo activity. This diminished

activity for the 4,4-dimethyl analogs in the metrazole

and other in vivo assays may relate to increased

metabolism for this substitution pattern.

The 3′-substituent on the oxadiazole was also varied,

with data for several representative structures provided

in Table 5. Reasonable binding affinity was maintained

across the series except for the tert-butyl analog, which

had 15-fold decreased affinity as compared to the other

analogs. A general trend in the in vitro properties for

both the TBPS and chloride current assays was noted,

as compounds containing bulkier substituents had

Partial Agonists at GABA_A/ Benzodiazepine Receptor

J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 165

Ta ble 5. [³H]Fnz Binding, TBPS Shift, ³⁶Cl^-Uptake, and

Metrazole Antagonism Data for Imidazo[1,5-a]quinoxaline

Ureas with Selected 5′-Alkyloxadiazole Substituents

Ta ble 6. [³H]Fnz Binding in R₁â₂γ₂, R₃â₂γ₂, and R₆â₂γ₂

Subtypes and ³⁶Cl^-Uptake in R₁â₂γ₂and R₃â₂γ₂Subtypes for

Imidazo[1,5-a]quinoxaline Ureas and Amides

K_i(nM)^a

Cl^-uptake^c,d

R₁â₂γ₂R₃â₂γ₂

b

compd

R₁â₂γ₂R₃â₂γ₂R₆â₂γ₂

33

35

39

41

42

46

55

58

62

63

64

65

67

68

73

2.7

0.3

124

1140

0.75

0.68

0.52

0.74

1.7

4.5

3.5

15

0.15

0.95

0.59

0.55

0.83

0.36

K_i

[

³⁵S]TBPS

shift^b,c

³⁶Cl^-

metrazole^d

compd R^5′(nM)^a

uptake^b,cED₅₀(mg/kg, ip)

75

76

45

77

Et

iPr

cPr

0.65

1.1

0.45

0.39

0.69

0.41

0.76

0.25

0.40

0.50

1.2

>50

11

0.24

18

0.17

0.30

0.22

4.7

0.51 13%

0.70

1.1

0.60

2.0

30

0%

tBu 14.7

0.22

0.67

1.3

0.0

0.78

0.62

Mean binding affinity against [³H]flunitrazepam; see ref 17

and the Experimental Section for methods. The standard error

a

b

was <(10% of the mean. Diazepam is defined as a full agonist

diazepam

Ro 15-4513 15 ( 4

U-78875

13

10

5.7

>7000

1.0 ( 0.15 1.0 ( 0.15

which gives a value of 1. Antagonists are defined as having a shift

5.4 ( 0.4

value of 0; partial agonists are intermediate. ^cSee the Experi-

3.5

603

d

mental Section. Antagonism of metrazole-induced clonic convul-

sions in the rat after ip injection; see the Experimental Section.

a

The ability of the compounds to displace [³H]flunitrazepam

was measured in membranes from Sf-9 insect cells expressing the

R₁â₂γ₂and R₃â₂γ₂subtypes and to displace [³H]Ro 15-4513 in insect

cell membranes expressing the R₆â₂γ₂subtype, as described in the

enhanced efficacy. This correlation of efficacy to steric

bulk was particularly evident in the chloride current

assay where the 5′-ethyl analog was nearly an antago-

nist, while the tert-butyl derivative was a full agonist.

On the other hand, activity in the metrazole assay did

not parallel the in vitro results. Instead the cyclopropyl

analog was superior to the other more efficacious

analogs. While the diminished activity of 77 in this

assay is explainable by affinity differences, the lack of

potent anticonvulsant activity for the other oxadiazole

analogs is less clear.

The frequent lack of correlation between the TBPS

shift ratio, Cl^-current, and in vivo activity for com-

pounds within this series, and benzodiazepine ligands

in general, can be partially attributed to heterogeneity

of the GABA_Areceptor population. Classical benzo-

diazepines interact with the R₁â₂γ₂, R₂â₂γ₂, R₃â₂γ₂, and

R₅â₂γ₂subtypes with nearly equal affinity,²³while

atypical BzR ligands such as Cl 218 872 and Zolpidem

have moderate selectivity for the R₁â₂γ₂subtype.^4a,23

However, classical benzodiazepines do not interact with

the R₆â₂γ₂subtype located in cerebellar granule cells,

whereas ligands such as Ro 15-4513 have high affinity

for this subtype.²⁴Furthermore, these BzR ligands

often display efficacy selectivity, where Cl^-current is

modulated to varying degrees in these receptor sub-

types.²⁵The search for selective ligands for these

subtypes as well as the elucidation of their function

remains an important goal in this area.

A few analogs from this series were examined in

several different benzodiazepine receptor subtypes.^26,27

As reported previously,^12,13these quinoxalines have

excellent affinity for the R₁â₂γ₂subtype with binding

data for a selected number of compounds shown in Table

6. The affinity of all analogs evaluated in the R₃â₂γ₂

subtype was also quite high, although usually 2-9-fold

less than for R₁â₂γ₂. Most analogs within this series

were not evaluated in these binding assays given the

general lack of selectivity observed between these two

subtypes. Affinity of selected analogs for the R₆â₂γ₂

subtype was extremely poor with only the small 5-acet-

amide substituent of 33 tolerated.^12,13Even then, at

least a 45-fold drop in affinity between the R₁â₂γ₂and

R₆â₂γ₂subtypes was noted.

Experimental Section. Percent inhibition at 100 nM. ^cDiazepam

b

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. See the Experimental Section.

d

While all analogs were evaluated for efficacy using

the Cl^-current assay in the R₁â₂γ₂subtype, several

derivatives were screened in the R₃â₂γ₂subtype. Three

analogs (46, 63, 73) had significantly different efficacy

between the two subtypes. Interestingly with morpho-

line 46, efficacy was enhanced in the R₃â₂γ₂subtype,

whereas 63 and 73 displayed full agonist efficacy in the

R₁â₂γ₂subtype and a partial agonist profile in R₃â₂γ₂.

Not surprisingly, 46 and 73 also displayed the greatest

differences between the TBPS (evaluated in the native

GABA_Areceptor population) and Cl^-current (R₁â₂γ₂)

values. Several analogs were also identified which had

a partial agonist profile in both subtypes. Of particular

interest were partial agonists 39, 41, and 46 (R₁â₂γ₂and

R₃â₂γ₂subtypes), which also displayed potent activity

in the metrazole assay. Analogs within this series were

not evaluated for efficacy in the R₆â₂γ₂subtype due to

low affinity.

While only a limited study was carried out for

compounds within this series in these receptor subtypes,

the binding differences between the R₆â₂γ₂and other

subtypes are consistent with a sterically restricted

region in the R₆â₂γ₂subtype that cannot accommodate

the rather large (out-of-plane) 5-substituent of these

analogs.¹³In addition, like the benzodiazepines and

related compounds, little binding or efficacy selectivity

between the R₁â₂γ₂and R₃â₂γ₂subtypes was evident for

most of the analogs in this series. Nonetheless, signifi-

cant differences in efficacy for three compounds between

these subtypes were noted, providing promise for the

future discovery of efficacy selective analogs within the

quinoxaline urea class.

On the basis of the results of the in vitro (binding,

TBPS, chloride current) and metrazole assays, several

of the most promising compounds were evaluated

further for typical benzodiazepine-type side effects, with

data provided in Table 7. Two of the more undesirable

side effects typically promoted by benzodiazepine full

166 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1

J acobsen et al.

Ta ble 7. Benzodiazepine Antagonism, Ethanol Potentiation, and Acute Physical Dependence Data for Imidazo[1,5-a]quinoxaline

Ureas

ethanol potentiation ED₅₀(mg/kg)

BZD antag traction^a

ED₅₀(mg/kg)

acute electroshock physical dependence

compd

40

41

42

43

51

54

61

62

63

73

74

traction^a

2.7

27

LRR

activity^a(drug MA₅₀/V122 MA₅₀

)

22

1

11

>30

A (18.9/21.5)

IA (23.2/24.9)

IA (19.7/20.4)

IA (20.9/21.5)

A (17.7/22.4)

A (17.7/20.3)

IA (21.0/22.4)

A (14.4/21.2)

A (20.7/24.1)

A (17.2/19.3)

IA (19.4/19.3)

IA (22.1/22.1)

A^b(18.2/25.4)

IA (21.2/22.2)

>30

4

>30

4

>30

0.3

3

0.4

1

2

3

>30

17

9

>30

0.2

>30

76

17

>30

diazepam

U-78875

0.54 (0.54-0.54)

6.6 (3.3-13)

0.86 (0.5-1.4)

>30

1.7 (0.8-3.8)

a

b

See the Experimental Section. Tested at 15 mg/kg.

agonists are ethanol potentiation and physical depend-

ence. The potentiation of the effects of alcohol by the

test compounds was determined by measuring the drug

ED₅₀s for traction and the loss of righting after co-

administration. In this assay, diazepam was quite

potent, having ED₅₀values of 0.54 and 0.86 mg/kg for

traction and loss of righting, respectively. The degree

of acute physical dependence was assessed through an

electroshock seizure assay.²⁸Withdrawal 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. In this

assay, the lowering of electroshock thresholds precipi-

tated by a benzodiazepine antagonist after an acute

regimen of test compound was used to quantify the

development of physical dependence. Mice received the

test drug for 3 days (150 mg/kg/day). Twenty-four hours

after the last administration, the mice received an iv

injection of a benzodiazepine antagonist (flumazenil).

Five minutes later, the mice were assessed for electro-

shock seizure thresholds (MA₅₀compared to vehicle

MA₅₀), which typically were lower for mice undergoing

flumazenil-precipitated withdrawal. Standard benzo-

diazepine ligands such as diazepam were active in this

paradigm at doses as low as 15 mg/kg/day. Finally,

most of the compounds were evaluated for their ability

to antagonize the muscle relaxation effects of a benzo-

diazepine full agonist, triazolam. Traction was assessed

30 min after separate administration of triazolam and

the test compound. The antagonist flumazenil, at 2.0

mg/kg, was effective in blocking the muscle-relaxing

effects of triazolam, as was the partial agonist U-78875

(1.7 mg/kg). Traditional full agonist benzodiazepine

ligands are not effective in blocking muscle relaxation

in this assay.

pounds were tested at 10 times the dose of diazepam,

any significant activity in this assay was considered

undesirable. The two analogs that significantly altered

the loss of righting reflex were also found to be active

in the physical dependence screen and, not unexpect-

edly, were full agonists in the in vitro efficacy assays.

However, for other compounds, the correlation of the

TBPS and chloride current assays to activity in the

physical dependence screen was quite poor, perhaps due

to differences in pharmacokinetics or varying receptor

subtype interactions. Many of the quinoxalines tested

were active to some degree in antagonizing the effects

of triazolam. The most potent compounds were 41, 43,

54, and 63. Nonetheless, only two of these analogs were

inactive in the acute electroshock dependence assay,

indicating that even a moderately potent benzodiazepine

antagonist can still produce undesirable side effects.

On the basis of its overall profile in these screens, 41

was selected for detailed evaluation in several other

assays useful for predicting anxiolytic activity. In

Vogel’s punished licking assay^10a(Figure 2), this qui-

noxaline was effective in increasing the number of

shocks taken at doses ranging between 0.3 and 10 mg/

kg. The activity of 41 in this assay was equieffective

to that of 1.0 mg/kg diazepam (DZ). Similarly, in the

Geller-Cook conflict assay,^10a41 was active in punished

animals at doses of 0.3, 1.0, and 3.0 mg/kg (Figure 3).

Significant activity was also nearly achieved at lower

doses. No effects were observed in the nonpunished 41

group. Again the activity of this compound compared

quite favorably to that achieved by diazepam (data not

shown). The morpholine analog 63 was also active in

both of these assays as shown in Figures 2 and 3.

Molecu la r Con for m a tion s. The low-energy con-

formers of a representative set of compounds were

determined by molecular mechanics methods. Calcula-

tions were performed using both the AMBER* and

MM2* force fields, which were developed for the Macro-

Model²⁹system of programs as extensions of the original

AMBER³⁰and MM2³¹parameter sets. Initial structures

to be minimized were generated by means of an internal

coordinate Monte Carlo procedure³²in which torsional

angles around all rotatable bonds were allowed to vary.

The analogs examined by these methods are listed in

Table 8.

Most of the quinoxaline analogs that were evaluated

in the ethanol potentiation assay had little effect on the

loss of righting reflex, consistent with their partial

agonist profile. Only 40 and 73 were active, although

to a much lesser degree than diazepam. However, like

U-78875, many of the quinoxalines did have effects on

traction, although usually to a lesser degree than

diazepam. Compounds 41-43 in particular stand out

as being relatively free of ethanol interaction. Roughly

one-half of the compounds tested were active in the

physical dependence assay. Even though these com-

In Figure 4, structures of the two lowest-energy

conformers of 41 obtained with the MM2* force field are

shown. The structures are identical except for the

Partial Agonists at GABA_A/ Benzodiazepine Receptor

J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 167

Figure 5 shows structure A for three urea analogs (39,

41, 45) in an edge view of the imidazoquinoxaline ring.

For each analog shown, as well as all others unsubsti-

tuted at the 4-position, the ring system is nearly planar,

with the oxadiazole ring rotated out of the imidazole

plane by only 5°. For the 4,4-gem-dimethyl-substituted

analogs 67, 69, and 70, the preferred orientation of the

oxadiazole ring is ca. 30° out of the imidazole plane due

to electrostatic interactions between N_4′and the methyl

groups. The R⁵substituent is seen to occupy an out-

of-plane region, similar to that of the C₅-phenyl group

in classical benzodiazepines. The position of the R⁵

group is not significantly altered by 4,4-gem-dimethyl

substitution.

While the molecular structures obtained from the

respective force fields were found to be very similar, the

same does not apply to relative energies. Specifically,

the data in Table 8 show that structure A is the lowest-

energy conformer according to MM2*, but AMBER*

favors structure B as that of lowest energy except for

the gem-dimethyl-substituted analogs 67, 69, and 70.

Thus, in general, the preferred conformation cannot be

determined unequivocally from the present studies. In

many cases, the computed energy difference between

structures A and B is not large, and a mixture of

conformations may exist. However, as discussed further

below, the N₅-C rotation barrier is such that conversion

between the two conformers may not readily occur. In

this regard, the orientation of the carbonyl group

appears to be unimportant as a factor in receptor

binding, since it can be physically constrained to point

in relatively different orientations as in 1 and isomers

cyclized from N₅-C₆via lactam rings³³as illustrated

for 78 (Figure 6). Analogs from both of these con-

strained ring systems have excellent affinity for the

BzR.

In order to obtain an indication of the flexibility of

the present analogs, estimates of selected rotation

barriers were obtained from a series of constrained

molecular mechanics energy minimizations. The fol-

lowing gives ranges of values obtained from studies on

the analogs in Table 8 using both AMBER* and MM2*

parameters. The cyclopropyl group is free to rotate

about the C_5′-C bond with a barrier of only 1.4-1.7

kcal/mol. Rotation of the oxadiazole group about C₃-

C_3′is subject to a moderate barrier of 5-7 kcal/mol. The

methyl groups in the gem-dimethyl-substituted analogs

do not change the rotation barrier significantly. As may

be expected, rotation about N₅-C bond appeared to be

severely restricted, and barrier heights could not be

obtained in most cases. Calculations performed on the

N₅-formyl analog, in which changes in formyl orienta-

tion were accompanied by significant distortions of the

quinoxaline ring, yielded an estimate of 12-17 kcal/mol.

F igu r e 2. Vogel conflict assay. Bar graphs of the mean

number of shocks received after drinking for each dose group.

Compound 41 or 63 was injected ip 30 min prior to the test

session with 8-10 animals/group. Statistical comparison

between vehicle- and drug-treated animals was carried out by

one-way analysis of variance followed by the nonparametric

Wilcoxon rank sum test for each treatment group (* ) p <

0.05; ** ) p < 0.01).

P h a r m a cop h or e Mod els. A number of models have

been proposed in order to elucidate the steric and

electronic properties necessary for BzR recognition and

activation.³⁴For the present analogs, both structures

A and B of Figure 4 exhibit several features consistent

with general models proposed by Wermuth et al.³⁵and

Gardner.³⁶As depicted in Figure 7, the fused phenyl

component of the quinoxaline ring and the N₂atom act

as the aromatic region and hydrogen acceptor site (δ₁),

respectively, regarded as the principal features neces-

sary for receptor recognition.^34-37The cyclopropyl oxa-

orientations of the R⁵substituent, which differ by a 180°

rotation about the N₅-C bond. Calculations using the

AMBER* force field yielded structures with no signifi-

cant geometrical differences from those obtained with

MM2*, both for 41 and all other analogs for which

calculations were performed. It should be noted that

for each structure shown, there exists a pair of confor-

mational enantiomers of identical energy, which differ

geometrically due to the puckering of the imidazo-

quinoxaline ring.

168 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1

J acobsen et al.

F igu r e 3. Geller-Cook conflict assay. Bar graphs of the mean response rate of lever pressing for food reinforcement. Animals

were trained to press a lever to receive a food pellet for a 5 min period followed by a 2 min punished period where food was

accompanied by an electric shock, signaled by a light over the response lever. Compounds 41 and 63 were injected ip 30 min

before the test session with 6 animals/dose group. Wilcoxon signed-rank tests were used to compare data from each dosing group

with their own control data (* ) p < 0.05).

diazole group in the present analogs correlates with both

the “freely-rotating aromatic” group in the Wermuth

model, and the ester group cited in the Gardner model.

This feature is absent in classical benzodiazepines but

is present in a number of forms including aromatic,

ester, acyl, or oxadiazole groups in other BzR ligands.

According to both models, efficacy may be modulated

by altering either the lipophilic character and/or the

physical size of this substructure, which is supported

in the present study by the limited data given in Table

5. Perhaps most significantly, the R⁵substituent which

distinguishes this series from most other BzR ligands,

corresponds to the “out-of-plane” region described by

Wermuth. This site is thought to correlate with full

agonist activity,^34,35which may explain how the 5-sub-

stituent in the present analogs can successfully modu-

late efficacy. The only other major class of BzR ligands

that have a nonaromatic substituent at this site is the

cyclopyrrolones,^35,36with the piperazine side chain of

Zopiclone located in the same region as the R⁵substitu-

ent (e.g., 45) when an overlay of the δ₁and aromatic

regions is performed. Interestingly, the electronegative

carbonyl group of the R⁵substituent probably does not

correspond to the δ₂hydrogen acceptor site in the

Wermuth model (see Figure 7) as it is located out of the

aryl plane and has a significantly greater distance to

the imidazole hydrogen acceptor site (δ₁) than typically

observed: 6.0-6.5 Å, depending on the conformational

Partial Agonists at GABA_A/ Benzodiazepine Receptor

J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 169

Ta ble 8. Relative Energies of Low-Energy Conformers of

Selected Imidazo[1,5-a]quinoxaline Amides, Ureas, and

Carbamates

F igu r e 6.

relative energies (kcal/mol)

AMBER*

MM2*

compd

R⁵

A

B

A

B

33

35

39

41

45

48

49

50

67^a

69^a

70^a

Me

phenyl

NH₂

NMe₂

pyrrolidine

OtBu

OMe

OiPr

NH₂

NMe₂

1.29

1.00

0.00

1.03

1.12

2.94

3.04

2.96

0.00

0.24

0.00

1.72

3.42

3.54

0.00

2.08

1.79

2.61

2.52

2.13

1.67

1.96

1.63

3.82

6.79

7.00

pyrrolidine

a

4,4-gem-Dimethyl substituted.

F igu r e 7. Structure of 41 showing elements of the Wermuth

and Gardner pharmacophore models: (a) aromatic region, (b)

hydrogen acceptor site δ₁, (c) freely-rotating unit, and (d) out-

of-plane region. The second hydrogen acceptor site (δ₂, e) is

absent in the present analogs. Distances shown between a, b,

and e are given in angstroms.

effects of the R³substituent coplanar with the quinoxa-

line ring and the out-of-plane R⁵substituent.

F igu r e 4. Molecular structures of the two lowest-energy

conformers of 41.

Con clu sion

Compounds within this imidazo[1,5-a]quinoxaline

series are high-affinity ligands for the GABA_A/benzo-

diazepine receptor. By control of the substituents at the

4-, 5-, 6-, and 7-positions, analogs spanning a wide range

of efficacy, from antagonists to full agonists, were

prepared. Most importantly, particularly for the 5-urea

quinoxalines, compounds with a partial agonist profile

were identified. Many of these compounds were active

in selected animal models of anxiety or sedation.

Furthermore, in line with the partial agonist hypothesis,

side effects such as physical dependence, ethanol po-

tentiation, and muscle relaxation were significantly

reduced as compared to benzodiazepine full agonists.

From this series, 41 stands out as having minimal

benzodiazepine side effects yet is equieffective to di-

azepam in classical animal models of anxiety such as

the Vogel and Geller paradigms. The continued evalu-

ation of 41 and the further exploration of this series in

the search of GABA_A-derived anxiolytic partial agonists

will be reported in due course.

F igu r e 5. Edge view of molecular structures of conformer A

for 39, 41, and 45.

isomer vs 3.2 Å for standard BZDs. Given the similar

affinity of compounds from this series to 1 and 78, it is

arguable that if the carbonyl group is acting as the δ₂

hydrogen acceptor, then this site in this series plays a

greater role in modulating efficacy than affinity. In

summary, the present analogs are consistent with a

model in which an aromatic region and single hydrogen

acceptor site act as necessary components for receptor

recognition, while efficacy is governed by the cooperative

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

combustion 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.

170 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1

J acobsen et al.

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 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 times, dry the

organic layers with the indicated drying agent, filter off the

drying agent, and remove solvent using a rotary evaporator

under reduced pressure. Tetrahydrofuran (THF) and ether

were distilled from sodium and benzophenone. Dichloro-

methane 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 chlorophos-

phate 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.

divided, Aldrich 22,167-8, slurry in water at pH 10), and 540

mL of ethanol. The mixture was stirred for 90 min at 0 °C.

Filtration and aqueous workup (Et₂O, Na₂SO₄) gave 91.7 g

(98%) of 18 as a brown oil which was used without further

purification: ¹H NMR (300 MHz, CDCl₃) δ 6.82 (apparent t, J

) 7.8 Hz, ArH), 6.75 (dd, J ) 7.9, 1.6 Hz, ArH), 6.60 (dd, J )

7.7, 1.7 Hz, ArH), 4.11 (br s, NH₂), 3.02 (br s, NH), 1.26 (s,

C(CH₃)₃).

4-ter t-Bu t yl-5-ch lor o-1,2,3,4-t et r a h yd r oq u in oxa lin -2-

on e (19). Chloroacetyl chloride (55.2 mL, 693 mmol) was

added dropwise with stirring over 1 h to a -78 °C solution of

diamine 18 (91.7 g, 462 mmol) and diisopropylethylamine (241

mL, 1.38 mol) in 900 mL of THF. The mixture was allowed

to warm slowly and stirred at room temperature for 18 h. Basic

workup (CH₂Cl₂, NaHCO₃, Na₂SO₄) of the concentrated mix-

ture gave an uncyclized amide intermediate as a black oil: ¹H

NMR (300 MHz, CDCl₃) δ 9.74 (br s, CONH), 8.33 (d, J ) 7.6

Hz, ArH), 7.05-7.2 (m, ArH, 2 H), 4.20 (s, CH₂), 3.07 (s, NHC-

(CH₃)₃), 1.24 (s, C(CH₃)₃).

A solution of the intermediate and diisopropylethylamine

(120 mL, 689 mmol) in 1.50 L of acetonitrile was heated at

reflux in the presence of sodium iodide (9.00 g, 60.0 mmol) for

18 h. Aqueous workup (CH₂Cl₂, Na₂SO₄) of the concentrated

mixture gave the crude product as a black, oily solid. Flash

chromatography (10% acetone/hexane) and trituration (CH₂Cl₂)

gave 12.1 g of 19 as a white solid (mp 197.5-198.5 °C). The

filtrate was concentrated, rechromatographed, and triturated

(5:30:65, v/v, MeOH:Et₂O:hexane) to give 18.8 g of additional

product. Flash chromatography of the filtrate (1% MeOH/

CH₂Cl₂) gave an additional 1.22 g of 19. The total yield of 19

was 32.1 g (29%): IR (mineral oil) 1677, 1579, 1367, 1187, 778

P r ep a r a tion of Am id e Tem p la tes 3. 1,2,3,4-Tetrahydro-

quinoxalin-2-one (3a ), 3,3-dimethyl-1,2,3,4-tetrahydroquinoxa-

lin-2-one (3g), 5-fluoro-1,2,3,4-tetrahydroquinoxalin-2-one (3c),

and 6-fluoro-1,2,3,4-tetrahydroquinoxalin-2-one (3b) were pre-

pared as previously described.¹²

1

cm^-1; H NMR (300 MHz, CDCl₃) δ 8.05 (br s, NH), 7.0-7.15

(m, ArH, 2 H), 6.73 (d, J ) 7.6 Hz, ArH), 3.72 (ABq, J _AB

)

17.3 Hz, ∆ν ) 169 Hz, CH₂), 1.27 (s, C(CH₃)₃); MS (EI) m/ e

238, 223, 182, 153, 90, 57.

6-Ch lor o-1,2-d ih yd r oqu in oxa lin -2-on e (14). Following

the procedure of Lumma,¹⁵glyoxylic acid (29.4 mL, 266 mmol,

50% aqueous) was added to a solution of 4-chloro-1,2-phen-

ylenediamine (13; 38.0 g, 266 mmol) and MeOH (1.87 L) at 15

°C. The dark black solution was stirred for 24 h at room

temperature and concentrated. The residue was washed with

water (4 × 608 mL) and 2-propanol (145 mL) and dried in

vacuo to give 45.1 g of a black solid. Two successive recrys-

tallizations from hot (ca. 90 °C) 2-methoxyethanol (1.49 and

0.927 L) gave 18.0 g (37%) of 14 as a gray powder: ¹H NMR

(300 MHz, DMSO-d₆) δ 12.55 (br s, NH), 8.22 (s, NdCH), 7.86

(narrow m, ArH), 7.62 (dd, J ) 8.4, 1.5 Hz, ArH), 7.32 (d, J )

8.7 Hz, ArH).

6-Ch lor o-1,2,3,4-tetr a h yd r oqu in oxa lin -2-on e (15). So-

dium borohydride (5.10 g, 135 mmol) was added to a mixture

of imine 14 (5.60 g, 31.0 mmol) and ethanol (200 mL). The

resultant solution was stirred for 2.5 h at room temperature

and concentrated. Aqueous workup (ethyl acetate, MgSO₄)

and recrystallization from ethyl acetate-hexane gave (two lots)

4.19 g of amine 15. The filtrate was concentrated and purified

by flash chromatography (1:1 hexane:ethyl acetate) to provide

451 mg (after recrystallization) of additional product (4.64 g

total, 82%) as a light brown powder (mp 171-174 °C): IR

(mineral oil) 1687, 1517, 1408, 1307, 1299 cm^-1; ¹H NMR (300

MHz, CDCl₃-MeOD) δ 6.6-6.75 (m, ArH), 3.95 (s, NCH₂CO);

MS (EI) m/ e 182, 153.

5-Ch lor o-1,2,3,4-tetr a h yd r oqu in oxa lin -2-on e (21).

A

suspension of tert-butyl-protected amide 19 (32.1 g, 134 mmol)

in 500 mL of 2 N H₂SO₄was stirred at room temperature for

18 h. The solids were filtered, washed with aqueous NaHCO₃

and water, and dried at room temperature to give 23.6 g (96%)

of a 23:1 mixture of 21 and the corresponding imine 20 as

determined by ¹H NMR integration of the amide hydrogen

resonances at δ 12.62 for 20 and δ 10.47 for 21.

Sodium borohydride (1.84 g, 48.6 mmol) was added in one

portion to a 0 °C slurry of the crude amine/imine mixture (20.5

g, 113 mmol) in 680 mL of EtOH. The mixture was allowed

to warm to room temperature and stirred for 2 h. The ethanol

was evaporated, and the solids were triturated with water,

filtered, washed with water, and dried to give 19.5 g (95%) of

21 as a tan solid (mp 191-193 °C): IR (mineral oil) 3407, 1690,

1

1503, 1426, 771 cm^-1; H NMR (300 MHz, DMSO-d₆) δ 10.47

(s, CONH), 6.89 (d, J ) 8.0 Hz, ArH), 6.70 (d, J ) 7.0 Hz,

ArH), 6.59 (apparent t, J ) 7.9 Hz, ArH), 5.84 (s, CH₂NH),

3.80 (d, J ) 1.8 Hz, NHCH₂); MS (EI) m/ e 182, 153.

2-[N-(5-Ch lor o-2-n it r op h en yl)a m in o]-2-m et h yl-1-p r o-

p a n ol (23). A solution of 2,4-dichloronitrobenzene (22; 20.0

g, 0.104 mol) and 2-amino-2-methyl-1-propanol (80 mL, 0.84

mol) was stirred at 110 °C for 68 h. After cooling to room

temperature, aqueous workup (CH₂Cl₂, MgSO₄) gave 26.9 g

of 23, sufficiently pure (>90%) to be carried on crude. An

analytical sample was prepared by recrystallization from ethyl

acetate-hexane to provide orange crystals (mp 102-104 °C):

IR (mineral oil) 3335, 1610, 1577, 1492, 1331, 1251, 1154, 748

cm^-1; ¹H NMR (300 MHz, CDCl₃) δ 8.55 (br s, NH), 8.13 (d, J

) 9.2 Hz, ArH), 7.09 (d, J ) 2.1 Hz, ArH), 6.59 (dd, J ) 9.2,

2.0 Hz, ArH), 3.71 (s, CH₂OH), 2.11 (br s, OH), 1.48 (s,

C(CH₃)₂); MS (EI) m/ e 244, 213, 166.

N-(5-Ch lor o-2-n itr oph en yl)-2-m eth ylalan in e (24). Jones

reagent¹⁶(2.67 M) was added in aliquots (40, 20, 20, 20, and

10 mL) every 15 min to a solution of crude alcohol 23 (26.9 g)

and acetone (2.05 L) at 0 °C until the reaction was done by

TLC analysis. 2-Propanol (150 mL) was added, and the

mixture was allowed to warm to room temperature. The

mixture was filtered, and the solids washed with acetone

several times. The combined filtrates were concentrated and

partitioned between ether (800 mL) and 10% potassium

N-ter t-Bu tyl-6-ch lor o-2-n itr oa n ilin e (17). A mixture of

2,3-dichloronitrobenzene (16; 92.5 g, 482 mmol), tert-butyl-

amine (138 mL, 1.31 mol), and 50 mL of ethanol was heated

in a bomb at 150 °C for 3 days. After cooling to room

temperature, concentration and aqueous workup (CH₂Cl₂,

Na₂SO₄) gave 107 g (97%) of 17 as a brown oil: IR (neat) 2970,

1

1532, 1484, 1447, 1366, 1348, 1192, 755 cm^-1; H NMR (300

MHz, CDCl₃) δ 7.73 (dd, J ) 8.2, 1.6 Hz, ArH), 7.61 (dd, J )

8.0, 1.6 Hz, ArH), 7.07 (apparent t, J ) 8.1 Hz, ArH), 4.77 (s,

NH), 1.22 (s, C(CH₃)₃); MS (EI) m/ e 172, 154, 142, 126, 114,

99, 90.

2-(ter t-Bu tyla m in o)-3-ch lor oa n ilin e (18). A solution of

hydrazine monohydrate (115 mL, 2.37 mol) in 270 mL of

ethanol was added dropwise with stirring over 30 min to a 0

°C mixture of 17 (108 g, 472 mmol), Raney nickel (27 g, finely

Partial Agonists at GABA_A/ Benzodiazepine Receptor

J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 171

hydroxide (3 × 100 mL). The basic layers were acidified (3 N

HCl) and extracted with CH₂Cl₂(3 × 180 mL). The CH₂Cl₂

layers were dried (MgSO₄), filtered, and concentrated to

provide 16.2 g (60% from 22) of carboxylic acid 24 as a yellow

powder (mp 146-147 °C): IR (mineral oil) 3363, 1707, 1613,

(14.1 mL, 101 mmol) was added to a mixture of formamide

31d (6.18 g, 33.7 mmol) and methylene chloride (77 mL) at 0

°C. Phosphorus oxychloride (3.14 mL, 33.7 mmol) was added

slowly and the solution stirred for 1 h at 0 °C. The reaction

was quenched with aqueous sodium carbonate (3.57 g Na₂CO₃

in 39 mL of H₂O), and the mixture was stirred for an additional

15 min at 0 °C before being allowed to warm to room

temperature. Aqueous workup (CH₂Cl₂, MgSO₄) and purifica-

tion by flash chromatography (2:1 hexane:ethyl acetate) gave

4.71 g (85%) of 32d as an amber liquid: ¹H NMR (300 MHz,

CDCl₃) δ 4.73 (s, CNCH₂), 1.46 (s, tBu).

1

1573, 1495, 1336, 1272, 1241, 753 cm^-1; H NMR (300 MHz,

CDCl₃) δ 8.48 (br s, NH), 8.16 (d, J ) 9.1 Hz, ArH), 6.68 (dd,

J ) 9.1, 2.0 Hz, ArH), 6.64 (d, J ) 2.0 Hz, ArH), 1.74 (s,

C(CH₃)₂); MS (EI) m/ e 258, 213.

N-(5-Ch lor o-2-n itr oph en yl)-2-m eth ylalan in e Meth yl Es-

ter (25). A mixture of acid 24 (22.3 g, 86.2 mmol), DMF (260

mL), potassium carbonate (35.6 g, 0.258 mol), and iodomethane

(26.9 mL, 0.432 mol) was stirred at room temperature for 16

h. Aqueous workup (ether, water and brine washes, MgSO₄)

gave 22.0 g (94%) of ester 25 as an orange solid (mp 87-89

°C): IR (mineral oil) 3349, 1737, 1606, 1489, 1330, 1263, 1230,

1149, 752 cm^-1; ¹H NMR (300 MHz, CDCl₃) δ 8.48 (br s, NH),

8.15 (d, J ) 9.1 Hz, ArH), 6.65 (dd, J ) 9.1, 2.1 Hz, ArH), 6.54

(d, J ) 2.1 Hz, ArH), 3.77 (s, OCH₃), 1.69 (s, C(CH₃)₂); MS

(EI) m/ e 272, 213.

Exa m p le P r oced u r e for th e Syn th esis of 3-(5-Cyclo-

p r op yl-1,2,4-oxa d ia zol-3-yl)-4,5-d ih yd r oim id a zo[1,5-a ]-

qu in oxalin es 6: 6-Ch lor o-3-(5-cyclopr opyl-1,2,4-oxadiazol-

3-yl)-4,5-dih ydr oim idazo[1,5-a ]qu in oxalin e (6e). A solution

of amide 3e (2.63 g, 14.4 mmol) in 30 mL of THF was cooled

to -40 °C, and potassium tert-butoxide (1.0 M in THF, 15.8

mL, 15.8 mmol) was added dropwise over 5 min. The mixture

was allowed to warm to room temperature over 30 min and

then cooled to -50 °C. Diethyl chlorophosphate (2.70 mL, 18.7

mmol) was added dropwise over 4 min, and the mixture was

allowed to warm from -50 to -30 °C over 1 h and then allowed

to warm to room temperature over 30 min. The solution was

cooled to -78 °C, and isocyanide 5 (2.58 g, 17.3 mmol) was

added. Potassium tert-butoxide (17.3 mL, 17.3 mmol) was

added dropwise over 10 min. The mixture was allowed to

warm slowly to room temperature and stirred at room tem-

perature overnight. The mixture was diluted with water; the

solids were filtered, washed, and dried to give 3.26 g (72%) of

6e as a white solid (mp 205-207 °C): IR (mineral oil) 3327,

6-Ch lor o-3,3-d im eth yl-1,2,3,4-tetr a h yd r oqu in oxa lin -2-

on e (26). Aqueous titanium trichloride (20%, 260 mL) was

added dropwise over 10 min to a mixture of nitro compound

25 (14.1 g, 51.7 mmol), sodium acetate (240 g, 2.93 mol), MeOH

(504 mL), and water (156 mL). The light blue mixture was

stirred for 2.5 h at room temperature. Basic workup (ethyl

acetate, NaHCO₃, brine wash, MgSO₄) gave 10.8 g (99%) of

amide 26 as a tan powder (mp 166-170 °C) homogeneous by

TLC analysis: IR (mineral oil) 3311, 1658, 1614, 1505, 1404,

1

1355 cm^-1; H NMR (300 MHz, CDCl₃) δ 8.67 (br s, CONH),

6.73 (dd, J ) 8.3, 2.0 Hz, ArH), 6.6-6.7 (m, ArH, 2 H), 3.78 (s,

1571, 1493, 1305, 773 cm^-1

; ¹H NMR (300 MHz, CDCl₃) δ 8.06

(s, ArH), 7.34 (d, J ) 8.0 Hz, ArH), 7.19 (d, J ) 7.0 Hz, ArH),

6.78 (apparent t, J ) 8.1 Hz, ArH), 4.89 (d, J ) 1.6 Hz,

NHCH₂), 4.67 (s, NH), 2.2-2.3 (m, CHCH₂), 1.2-1.4 (m,

CH₂CH₂); MS (EI) m/ e 313, 244, 229.

NH), 1.42 (s, C(CH₃)₂); MS (FAB) m/ e 210, 195, 167.

N-(2-Meth yl-6-n itr op h en yl)glycin e Eth yl Ester (28). A

mixture of 2-methyl-6-nitroaniline (27; 2.00 g, 13.1 mmol),

ethyl bromoacetate (3.5 mL, 32 mmol), and diisopropylethyl-

amine (3.5 mL, 20 mmol) was heated at reflux (140 °C) for 8

h. The resultant solution was allowed to cool to room tem-

perature. After a basic workup (CHCl₃, NaHCO₃, MgSO₄), the

residue was resubmitted to the above reaction conditions for

an additional 16 h. After workup, purification by flash

chromatography (10:1 f 5:1 hexane:ethyl acetate) gave 681

mg (22%) of 28 as a glassy yellow solid (mp 49-52 °C): IR

(mineral oil) 3365, 1731, 1536, 1475, 1344, 1236, 1207, 1104,

3-(5-Cyclopropyl-1,2,4-oxadiazol-3-yl)-4,4-dimethyl-4,5-dihy-

droimidazo[1,5-a]quinoxaline (6g), 3-(5-cyclopropyl-1,2,4-oxa-

diazol-3-yl)-4,5-dihydroimidazo[1,5-a]quinoxaline (6a ), 3-(5-

cyclopropyl-1,2,4-oxadiazol-3-yl)-7-fluoro-4,5-dihydroimidazo[1,5-

a]quinoxaline (6b), and 3-(5-cyclopropyl-1,2,4-oxadiazol-3-yl)-

6-fluoro-4,5-dihydroimidazo[1,5-a]quinoxaline (6c) were

synthesized as previously reported.¹²

Exa m p le P r oced u r es for th e Acyla tion of Qu in oxa lin e

6. Meth od A: 3-(5-Cyclop r op yl-1,2,4-oxa d ia zol-3-yl)-4,5-

d ih yd r o-5-(2-p yr r olylca r b on yl)im id a zo[1,5-a ]q u in oxa -

lin e (38). The pyrrole acid chloride was prepared by the

procedure of Maxim.³⁸A mixture of 6a (600 mg, 2.15 mmol),

the acid halide (389 mg, 3.00 mmol), diisopropylethylamine

(1.05 mL, 6.03 mmol), and THF (15.0 mL) was stirred at room

temperature for 16 h. Additional acid halide (200 mg, 1.54

mmol) and diisopropylethylamine (0.50 mL, 2.9 mmol) were

added, and the mixture was stirred for an additional 5 h, at

which time no starting material was evident by TLC analysis.

1

1022 cm^-1; H NMR (300 MHz, CDCl₃) δ 7.88 (d, J ) 8.5 Hz,

ArH), 7.33 (d, J ) 8.8 Hz, ArH), 6.8-7.0 (m, NH), 6.86

(apparent t, J ) 7.8 Hz, ArH), 4.17 (q, J ) 7.1 Hz, CH₂CH₃),

3.94 (d, J ) 6.0 Hz, NHCH₂), 2.39 (s, CH₃Ar), 1.23 (t, J ) 7.1

Hz, CH₂CH₃); MS (EI) m/ e 238, 165, 118, 107, 91.

6-Meth yl-1,2,3,4-tetr a h yd r oqu in oxa lin -2-on e (29).

A

mixture of nitro compound 28 (675 mg, 2.83 mmol), ethanol

(60.0 mL), and 10% Pd/C (150 mg) was hydrogenated (48 psi)

at room temperature for 3.5 h. The mixture was filtered, the

residue was washed with ethanol, and the combined filtrates

were concentrated to provide 425 mg (93%) of amide 29 as a

tan powder (mp 168-171 °C): IR (mineral oil) 3382, 1671,

1490, 1444, 1393, 1295, 770 cm^-1; ¹H NMR (300 MHz, CDCl₃)

δ 8.68 (br s, NHCO), 6.79 (d, J ) 7.7 Hz, ArH), 6.6-6.75 (m,

ArH, 2 H), 4.04 (s, NHCH₂), 3.76 (br s, NHCH₂), 2.17 (s,

CH₃Ar); MS (EI) m/ e 162, 133.

Exa m p le P r oced u r e for Oxa d ia zole 31: 3-[(N-F or m yl-

a m in o)m eth yl]-5-ter t-bu tyl-1,2,4-oxa d ia zole (31d ). A so-

lution of trimethylacetyl chloride (6.82 mL, 55.4 mmol) in

methylene chloride (10 mL) was added dropwise over 10 min

to a cold suspension of (N-formylamino)acetamide oxime

triethylamine¹⁴(30; 10.9 g, 49.9 mmol) and triethylamine (11.9

mL, 85.4 mmol) in methylene chloride (80 mL) at 0 °C. The

resultant slurry was stirred at 0 °C for 1 h and at room

temperature for 22 h. The mixture was concentrated, diluted

with 75 mL of water, and heated at reflux for 48 h. After

cooling to room temperature, aqueous workup (CH₂Cl₂, MgSO₄)

and purification by flash chromatography (1:1 ethyl acetate:

methylene chloride) gave 6.18 g (68%) of 31d as a yellow

liquid: ¹H NMR (300 MHz, CDCl₃) δ 8.32 (s, CHO), 6.0-6.3

(m, NH), 4.63 (d, J ) 5.6 Hz, NHCH₂), 1.44 (s, tBu).

Basic workup (ethyl acetate and then CHCl₃

, NaHCO , MgSO )

3

4

and two successive recrystallizations from hot ethyl acetate/

MeOH gave 365 mg of 38 as white plates (mp 252-253 °C).

The filtrate was concentrated and recrystallized as above to

provide an additional 144 mg (509 mg total, 64%) of the

amide: IR (mineral oil) 3120, 1643, 1579, 1509, 1401, 1359,

1294, 1037, 767 cm^-1; ¹H NMR (300 MHz, CDCl₃) δ 9.49 (br s,

NH), 8.15 (s, ArH), 7.59 (apparent t, J ) 9.2 Hz, ArH, 2 H),

7.2-7.4 (m, ArH, 2 H), 6.95 (s, ArH), 6.30 (s, ArH), 6.17

(narrow m, ArH), 5.49 (s, CH₂N), 2.2-2.35 (m, CHCH₂CH₂),

1.15-1.4 (m, CHCH₂CH₂); MS (EI) m/ e 372, 304, 279, 210,

195, 94. Anal. (C₂₀H₁₆N₆O₂) C, H, N.

Met h od B: 3-(5-Cyclop r op yl-1,2,4-oxa d ia zol-3-yl)-5-

[(d im et h yla m in o)ca r b on yl]-4,5-d ih yd r oim id a zo[1,5-a ]-

qu in oxa lin e (41). A mixture of dimethylcarbamyl chloride

(539 mg, 5.01 mmol), pyridine (10 mL), and imidazoquinoxa-

line 6a (700 mg, 2.51 mmol) was stirred for 10 min at room

temperature. Basic workup (CH₂Cl₂, NaHCO₃, Na₂SO₄) pro-

vided a mixture of product and starting material. The residue

was combined with dimethylcarbamyl chloride (10.0 mL, 109

mmol) and 30 mL of pyridine, and the mixture was stirred at

room temperature for 19 h. Identical workup gave a solid that

was absorbed onto 10 g of Celite. Flash chromatography with

Exa m p le P r oced u r e for Isocya n id e 32: 3-(Isocya n o-

m eth yl)-5-ter t-bu tyl-1,2,4-oxa d ia zole (32d ). Triethylamine

172 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1

J acobsen et al.

15% EtOAc/CH₂Cl₂followed by 50% EtOAc/CH₂Cl₂afforded

431 mg (49%) of 41 as an off-white solid (mp 196-197 °C): IR

(mineral oil) 1664, 1573, 1500, 1275, 1202, 756 cm^-1; ¹H NMR

(300 MHz, CDCl₃) δ 8.11 (s, ArH), 7.54 (d, J ) 6.7 Hz, ArH),

7.2-7.3 (m, ArH), 7.05-7.2 (m, ArH, 2 H), 4.99 (s, NCH₂), 2.85

(s, NMe₂), 2.2-2.3 (m, CH), 1.2-1.4 (m, CH₂CH₂); MS (EI)

m/ e, 350, 306, 278, 263, 238, 210, 195, 72. Anal. (C₁₈H₁₈N₆O₂‚

¹/₄H₂O) C, H, N.

Meth od C: 3-(5-Cyclop r op yl-1,2,4-oxa d ia zol-3-yl)-4,5-

dih ydr o-5-(tr iflu or oacetyl)im idazo[1,5-a ]qu in oxalin e (34).

Trifluoroacetic anhydride (0.28 mL, 2.0 mmol) was added to a

mixture of 6a (500 mg, 1.79 mmol) and pyridine (4.0 mL). The

resultant solution was stirred for 16 h at room temperature.

Basic workup (CHCl₃, NaHCO₃, MgSO₄) and recrystallization

from ethyl acetate-hexane (two crops) gave 477 mg (71%) of

the trifluoroacetamide as a light brown powder (mp 175-177

°C): IR (mineral oil) 1707, 1604, 1512, 1498, 1432, 1424, 1419,

1283, 1218, 1211, 1205, 1194, 1163, 1154, 1087, 768 cm^-1; ¹H

NMR (300 MHz, CDCl₃) δ 8.16 (s, ArH), 7.88 (br s, ArH), 7.63

(d, J ) 8.4 Hz, ArH), 7.35-7.55 (m, ArH, 2 H), 5.38 (s, NCH₂),

2.2-2.35 (m, CHCH₂), 1.2-1.4 (m, CH₂CH₂); MS (EI) m/ e 375,

306, 278, 210. Anal. (C₁₇H₁₂N₅O₂F₃) C, H, N, F.

181.5-182.5 °C): IR (mineral oil) 3007, 1648, 1513, 1501, 1409,

1204, 760, 756 cm^-^{1 1}H NMR (300 MHz, CDCl₃) δ 8.10 (s,

;

ArH), 7.54 (d, J ) 7.7 Hz, ArH), 7.1-7.3 (m, ArH, 3 H), 5.04

(s, ArNCH₂), 3.28 (br s, NCH₂CH₂, 4 H), 2.2-2.3 (m, CH),

1.75-1.9 (m, NCH₂CH₂, 4 H), 1.15-1.4 (m, CHCH₂CH₂); MS

(EI) m/ e 376, 306, 278, 238. Anal. (C₂₀H₂₀N₆O₂) C, H, N.

With Th iols, Meth od G: 3-(5-Cyclop r op yl-1,2,4-oxa -

diazol-3-yl)-4,5-dih ydr o-5-[(m eth ylth io)car bon yl]im idazo-

[1,5-a ]qu in oxa lin e (51). Sodium thiomethoxide (158 mg,

2.25 mmol) was added to a slurry of carbamyl chloride 9a (700

mg, 2.05 mmol) in 10 mL of THF at 0 °C. The mixture was

allowed to warm to room temperature and stirred for 19 h.

The concentrated material was stirred in a mixture of water

and 20% ether/hexane. The resulting solids were filtered,

washed with water and hexane, and dried to give 690 mg (95%)

of 51 as a white solid (mp 176-177 °C): IR (mineral oil) 1660,

1497, 1422, 1357, 1206, 761 cm^-1; ¹H NMR (300 MHz, CDCl₃)

δ 8.12 (s, ArH), 7.85 (d, J ) 7.4 Hz, ArH), 7.58 (d, J ) 7.4 Hz,

ArH), 7.3-7.45 (m, ArH, 2 H), 5.34 (s, NCH₂), 2.36 (s, SCH₃),

2.25-2.4 (m, CH), 1.2-1.4 (m, CH₂CH₂); MS (EI) m/ e 353,

306, 278, 210, 195. Anal. (C₁₇H₁₅N₅O₂S‚¹/₄H₂O) C, H, N, S.

Exa m p le P r oced u r e for 12. Meth od H: 6-Ch lor o-3,3-

d im e t h yl-1,2,3,4-t e t r a h yd r o-4-(p yr r olid in oca r b on yl)-

qu in oxa lin -2-on e (12h , Nu ) p yr r olid in e). Phosgene (60.0

mL, 72.0 mmol, 12.5% in toluene) was added to a solution of

amine 3h (4.50 g, 21.4 mmol), THF (25.0 mL), imidazole (50

mg), and diisopropylethylamine (4.50 mL, 25.8 mmol). The

solution was stirred at room temperature for 96 h. Additional

phosgene (20.0 mL, 24.0 mmol), diisopropylethylamine (2.00

mL, 11.5 mmol), and CH₂Cl₂(10 mL) were added after 24 h.

Aqueous workup (ethyl acetate, MgSO₄) and trituration of the

residue with 20:1 hexane:ether gave 5.64 g (96%) of carbamoyl

chloride 11h as a tan powder: ¹H NMR (300 MHz, CDCl₃) δ

9.05-9.25 (br s, NH), 7.42 (d, J ) 2.1 Hz, ArH), 7.26 (dd, J )

8.5, 2.2 Hz, ArH), 6.91 (d, J ) 8.5 Hz, ArH), 1.64 (s, C(CH₃)₂).

Pyrrolidine (2.25 mL, 27.0 mmol) was added to a solution

of the crude carbamoyl chloride 11h (5.64 g, 20.7 mmol), THF

(100 mL), and diisopropylethylamine (4.70 mL, 27.0 mmol) at

0 °C. The solution was stirred for 1 h at 0 °C and for 15 h at

room temperature. Aqueous workup (ethyl acetate, MgSO₄)

and crystallization from ethyl acetate-hexane gave (two crops)

5.14 g of the desired product as white crystals (mp 192-193

°C). The filtrate was concentrated and purified by flash

chromatography (1:1 ethyl acetate:hexane) to provide an

additional 391 mg (5.53 g total, 84% from amine) of product:

IR (mineral oil) 3203, 3077, 1691, 1661, 1503, 1416, 1390, 1380,

Exa m p le P r oced u r e for th e Rea ction of 6 w ith Isocy-

a n a tes. Meth od D: 3-(5-Cyclop r op yl-1,2,4-oxa d ia zol-3-

yl)-4,5-d ih yd r o-5-[(m et h yla m in o)ca r b on yl]im id a zo[1,5-

a ]qu in oxa lin e (40). Imidazoquinoxaline 6a (600 mg, 2.15

mmol) was stirred with 10 mL of methyl isocyanate at 40 °C

for 3 days. The mixture was diluted with hexane and filtered.

The solids were triturated with copious amounts of hexane and

dried to give 600 mg (83%) of the product as a white solid (mp

214-215 °C): IR (mineral oil) 3374, 3100, 1655, 1575, 1503,

767 cm^-^{1 1}H NMR (300 MHz, CDCl₃) δ 8.10 (s, ArH), 7.5-

;

7.65 (m, ArH, 2 H), 7.2-7.4 (m, ArH, 2 H), 5.26 (s, CH₂), 5.06

(s, NH), 2.75-2.90 (m, NCH₃), 2.2-2.3 (m, CH), 1.15-1.4 (m,

CH₂CH₂); MS (EI) m/ e 336, 279, 238, 210, 195, 69. Anal.

(C₁₇H₁₆N₆O₂‚¹/₈H₂O) C, H, N.

E xa m p le P r oced u r e for Ca r b a m oyl Ch lor id e 9:

5-(Ch lor oca r b on yl)-3-(5-cyclop r op yl-1,2,4-oxa d ia zol-3-

yl)-4,5-d ih yd r oim id a zo[1,5-a ]qu in oxa lin e (9a ). Phosgene

(39.4 mL, 43.0 mmol, 1.09 M in toluene) was added dropwise

to a slurry of amine 6a (4.00 g, 14.3 mmol), THF (80 mL), and

diisopropylethylamine (7.5 mL, 43 mmol). The mixture was

stirred at room temperature for 19 h. The excess phosgene

was decomposed with ice and NaHCO₃. Aqueous workup

(CH₂Cl₂, Na₂SO₄) and crystallization from CH₂Cl₂/hexane gave

4.15 g (85%) of 9a as a white solid (mp 164-165 °C): ¹H NMR

(300 MHz, CDCl₃) δ 8.16 (s, ArH), 7.8-7.95 (m, ArH), 7.6-

7.65 (m, ArH), 7.35-7.55 (m, ArH, 2 H), 5.50 (br s, NCH₂),

2.2-2.35 (m, CH), 1.2-1.4 (m, CH₂CH₂).

1

1354, 1301 cm^-1; H NMR (300 MHz, CDCl₃) δ 8.54 (s, NH),

6.90 (dd, J ) 8.3, 2.1 Hz, ArH), 6.76 (d, J ) 8.3 Hz, ArH), 6.68

(d, J ) 2.1 Hz, ArH), 3.58 (br s, 2 H), 3.17 (br s, 2 H), 1.2-2.0

(m, 10 H); MS (EI) m/ e 307, 98.

Exa m p le P r oced u r es for th e Rea ction of Ca r ba m oyl

Ch lor id e 9 w ith Selected Nu cleop h iles. With Alcoh ols,

Meth od E: 3-(5-Cyclop r op yl-1,2,4-oxa d ia zol-3-yl)-4,5-d i-

h yd r o-5-(m et h oxyca r b on yl)im id a zo[1,5-a ]q u in oxa lin e

(49). A solution of carbamoyl chloride 9a (500 mg, 1.46 mmol)

and THF (3.0 mL) was added to a solution of sodium methoxide

(95 mg, 1.8 mmol) and MeOH (2.0 mL) at 0 °C. The solution

was allowed to warm to room temperature and stirred for 2

h. The reaction was quenched with 1 mL of water, and the

solution was concentrated. Aqueous workup (EtOAc, MgSO₄)

and recrystallization of the crude solid from EtOAc/hexane

provided 318 mg (65%) of 49 as a yellow powder (mp 187-

188 °C): IR (mineral oil) 1714, 1514, 1504, 1444, 1301, 1291,

1221, 1207, 764, 757 cm^-1; ¹H NMR (300 MHz, CDCl₃) δ 8.11

(s, ArH), 7.78 (d, J ) 7.0 Hz, ArH), 7.54 (dd, J ) 7.6, 1.4 Hz,

ArH), 7.2-7.4 (m, ArH, 2 H), 5.27 (s, NCH₂), 3.83 (s, OCH₃),

2.2-2.35 (m, CHCH₂CH₂), 1.2-1.4 (m, CHCH₂CH₂); MS (EI)

m/ e 337, 278, 268, 209, 195. Anal. (C₁₇H₁₅N₅O₃) C, H, N.

With Am in es, Meth od F : 3-(5-Cyclop r op yl-1,2,4-oxa -

d ia zol-3-yl)-4,5-d ih yd r o-5-(p yr r olid in oca r bon yl)im id a zo-

[1,5-a ]qu in oxa lin e (45). A solution of carbamyl chloride 9a

(1.00 g, 2.93 mmol) and diisopropylethylamine (1.02 mL, 5.86

mmol) in 20 mL of THF was cooled in an ice bath. Pyrrolidine

(0.37 mL, 4.4 mmol) was added, and the mixture was allowed

to warm to room temperature and stir overnight. Basic

workup (CH₂Cl₂, NaHCO₃, Na₂SO₄) and recrystallization from

CH₂Cl₂/hexane gave 1.02 g (92%) of 45 as a white powder (mp

Exa m p le P r oced u r e for Acyla tion of 3. Meth od I: 4-(2-

F u r oyl)-1,2,3,4-t et r a h yd r oq u in oxa lin -2-on e (12a , Nu )

2-fu r a n ). 2-Furoyl chloride (0.49 mL, 5.0 mmol) was added

to a solution of amine 3a (710 mg, 4.79 mmol), triethylamine

(0.80 mL, 5.7 mmol), and THF (20 mL) at 0 °C. The mixture

was stirred for 1.5 h at 0 °C and for 2 h at room temperature.

Basic workup (ethyl acetate and then CHCl₃, NaHCO₃, MgSO₄)

and recrystallization from hot ethyl acetate gave 815 mg (70%)

of the desired product as off-white needles (mp 229-230 °C):

IR (mineral oil) 1684, 1648, 1504, 1474, 1391, 1373, 756 cm^-1

;

¹H NMR (300 MHz, CDCl₃) 8.53 (br s, NH), 7.39 (s, ArH), 7.1-

7.25 (m, ArH), 6.85-7.05 (m, ArH, 4 H), 6.45 (dd, J ) 3.5, 1.7

Hz, ArH), 4.59 (s, NCH₂); MS (EI) m/ e 242, 95.

Exa m p le P r oced u r e for th e Rea ction of 12 w ith Iso-

cya n id e 5. Met h od J : 7-Ch lor o-3-(5-cyclop r op yl-1,2,4-

o x a d ia zo l-3-y l)-4,4-d im e t h y l-4,5-d ih y d r o -5-(p y r r o li-

d in oca r bon yl)im id a zo[1,5-a ]qu in oxa lin e (74). Potassium

tert-butoxide (2.05 mL, 2.05 mmol, 1.0 M in THF) was added

to a solution of amide 12h (Nu ) pyrrolidine; 626 mg, 2.03

mmol) and THF (3.2 mL) at -25 °C. The solution was allowed

to warm to 0 °C over 45 min and recooled to -40 °C. Diethyl

chlorophosphate (0.30 mL, 2.1 mmol) was added, and the

solution was allowed to warm to 0 °C over 1 h. After cooling

to -78 °C, a solution of isocyanide 5 (333 mg, 2.23 mmol) and

THF (0.5 mL) was added. Potassium tert-butoxide (2.05 mL,

2.05 mmol) was then added dropwise over several minutes.

Partial Agonists at GABA_A/ Benzodiazepine Receptor

J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 173

The reddish solution was stirred at -78 °C for 30 min and

allowed to warm slowly to room temperature. After 16 h,

aqueous workup (ethyl acetate, MgSO₄), trituration (10:1

hexane:ether), and recrystallization of the resultant solid from

hot ethyl acetate-hexane gave 620 mg (70%, three crops) of

74 as off-white needles (mp 207-208 °C): IR (mineral oil)

³⁶Cl^-Up ta k e Stu d ies. ³⁶Cl^-uptake in rat cerebrocortical

synaptoneurosomes was measured by a rapid filtration tech-

nique using Whatman GF/B filters as described elsewhere.^18,19

A typical incubation medium contained 0.2 µCi Na³⁶Cl/mL, 118

mM NaCl, 5 mM KCl, 1.8 mM MgSO₄, and 20 mM Hepes/

Tris, pH 7.0, with or without test drugs. Drugs were added

in concentrated methanol solutions, and the level of methanol

did not exceed 0.4% and was maintained constant in all tubes.

The membrane suspensions were preincubated for 5 min at

30 °C. The reaction was initiated by mixing equal volumes

(125 µL) of the membrane suspension (1 mg of protein) and

the reaction mixture containing ³⁶Cl^-at 30 °C. After 5 s, the

reaction was terminated by adding ice cold NaCl incubation

buffer. The mixture was filtered over a Whatman GF/B filter

under vacuum, and the filters were washed four times with 5

mL of ice cold NaCl incubation buffer. The radioactivity on

the filters was counted in the presence of Instagel (15 mL).

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 R₁â₂γ₂subtype,

as described earlier.²⁰Briefly, patch pipettes made of boro-

silicate glass tubes were fire-polished and showed a tip

resistance of 0.5-2 MΩ when filled with a solution containing

(mM) 140 CsCl, 11 EGTA, 4 MgCl₂, 2 ATP, and 10 Hepes, pH

7.3. The cell-bathing external solution contained (mM) 135

NaCl, 5 KCl, 1 MgCl₂, 1.8 CaCl₂, and 5 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 Instru-

ment 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 temperature (21-

24 °C).

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.²¹Briefly, 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. ED₅₀s for

protection against tonic seizure were calculated by the method

of Spearman-Karber (Finney, D. J . Statistical Methods in

Biological Assay).

Eth a n ol P oten tia tion . Male CF-1 Charles River mice

were injected orally with 7.5 mL/kg 50% aqueous ethanol and

simultaneously received the test compound by the subcutane-

ous route (10 mL/kg in 0.25% aqueous methyl cellulose).

Thirty minutes later, they were tested for the loss of traction

response (muscle relaxation) and loss of righting reflex (an-

esthesia). The former test consists of determining if the mouse

after suspension by the forepaws on an 18 g wire can place

one of its hindlimbs on the wire within 10 s. The latter test

determines whether the same animal will roll over onto its

feet within 10 s of release on its back. Six mice were tested

at each dose, and active compounds were retested at several

1

3118, 1653, 1588, 1518, 1416, 1187, 945 cm^-1; H NMR (300

MHz, CDCl₃) δ 8.07 (s, ArH), 7.41 (d, J ) 8.5 Hz, ArH), 7.00

(dd, J ) 8.5, 2.1 Hz, ArH), 6.80 (d, J ) 2.1 Hz, ArH), 3.59 (br

s, 2 H), 2.8-3.4 (m, 2 H), 2.2-2.35 (m, CHCH₂), 1.5-2.2 (m,

C(CH₃)₂, NCH₂CH₂CH₂), 1.15-1.45 (m, CHCH₂CH₂); MS (EI)

m/ e 438, 423, 98. Anal. (C₂₂H₂₃N₆O₂Cl) C, H, N, Cl.

GABA_ARecep 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.²⁶

The Sf-9 cells were infected at a multiplicity of infection of 3

plaque-forming units of 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 R₁or R₃, R₆, â₂, and γ₂subunits of GABA_Awere

derived by transfection of plasmids containing cDNA and a

plasmid encoding G418 resistance into human kidney cells

(A293 cells) as described elsewhere.²⁷After 2 weeks of

selection in 1 mg/mL G418, cells positive for all three GABA_A

receptor mRNAs by Northern blotting were used for electro-

physiology to measure GABA-induced Cl^-currents. For

equilibrium binding measurements, Sf-9 cells infected with

baculovirus-carrying cDNAs for R₁or R₃, R₆, â₂, and γ₂subunits

were harvested in 2 L batches 60 h postinfection. The

membranes were prepared following the procedure described

previously.²⁶Briefly, the membranes were prepared in normal

saline after homogenization with Polytron PT 3000 homog-

enizer (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 GABA_Areceptors 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).

GABA_ARecep tor Bin d in g. The in vitro binding affinity

of the imidazo[1,5-a]quinoxalines for GABA_Awas determined

as previously described with minor modification.¹⁷Freshly

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.

doses to establish an ED₅₀

.

Acu te P h ysica l Dep en d en ce.²⁸Male CF-1 mice (10/

group) were dosed sc twice daily with the test compound, once

at 0800 and again at 1600 (with twice the 0800 dose) for 3

days. Twenty-four hours after the last dose, they received an

intravenous injection of the benzodiazepine antagonist fluma-

zenil (2.5 mg/kg in 5% aqueous N,N-dimethylacetamide). Five

minutes later, the electroshock seizure threshold was esti-

mated by the up-down method in which the stimulus current

was lowered or elevated by 0.05 log interval if the preceding

animal did or did not convulse, respectively. From the data

thus generated, a threshold (effective current 50) was calcu-

lated. 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 significantly lowered (below 95% confidence in-

[

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 incu-

bated 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 radioisotope and counted for radioactivity. Nonspecific

binding was estimated in the presence of 2 µM diazepam or

unlabeled TBPS and subtracted to compute specific binding.¹⁸

174 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1

J acobsen et al.

(6) (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. Psychopharma-

col. Bull. 1991, 27, 171-179.

(7) (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. TIPS 1990,

11, 452-456.

(8) 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.

(9) 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.

(10) (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-Amino-

butyric Acid_AReceptors. Mol. Pharmacol. 1992, 42, 294-301.

(11) DN-2327. Drugs Future 1993, 18 (5), 414-417.

(12) 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

GABA_A/Benzodiazepine Receptor. J . Med. Chem. 1994, 37, 758-

768.

(13) Im, W. B.; Im, H. K.; Pregenzer, J . F.; Hamilton, B. J .; Carter,

D. B.; J acobsen, E. J .; TenBrink, R. E.; VonVoigtlander, P. F.

Differential Affinity of Dihydroimidazoquinoxalines and Diimid-

azoquinazolines to the R₁â₂γ₂and R₆â₂γ₂Subtypes of Cloned

GABA_AReceptors. Br. J . Pharmacol. 1993, 110, 677-680.

(14) Watjen, F.; Baker, R.; Engelstoff, M.; Herbert, R.; MacLeod, A.;

Knight, A.; Merchant, K.; Moseley, J .; Saunders, J .; Swain, C.

J .; Wong, E.; Springer, J . P. Novel Benzodiazepine Receptor

Partial Agonists: Oxadiazolylimidazobenzodiazepines. J . Med.

Chem. 1989, 32, 2282-2291.

terval of parallel control group treated for 3 days with saline

and injected intravenously with flumazenil) the seizure thresh-

old, were considered to have caused physical dependence.

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).

Vogel’s P u n ish ed Lick in g Assa y. Following a previously

reported procedure,^10aexperimentally naive, male F344 rats

were deprived of drinking water for 48 h before being placed

in a chamber with a drinking spout. Brief electric shock (0.5

mA, 0.1 s) was delivered through the spout during every 10th

lick. The session terminated 3 min after the first shock or 15

min after the session started if no drinking occurred. A test

drug was injected ip 30 min before the test session, with at

least 8 animals/dose group. The median number of shocks for

each dose group was compared to that of a parallel vehicle-

treated group, using a one-way analysis of variance followed

by the nonparametric Wilcoxon rank sum test.

Geller -Cook Con flict Test. Following a previously re-

ported procedure,^10amale Sprague-Dawley rats were trained

to press a lever for food reinforcement (45 mg pellet). A 5 min

period of VI-60" schedule alternated with a 2 min period of

FR-10 punishment schedule, where each food delivery was

accompanied by electrical shock (0.4 mA, 0.5 s) from the grid

floor. The punishment component was always signaled by a

light over the response lever. After performance stabilized,

test drugs 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 the increase of shocks compared to the

immediately preceding vehicle session. Statistical significance

for each dose was estimated by the Wilcoxon signed-rank test.

Refer en ces

(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) (a) Pritchett, D. B.; Sontheimer, H.; Shivers, B. D.; Ymer, S.;

Kettenmann, H.; Schonfield, P. R.; Seeburg, P. H. Importance

of a Novel GABA_AReceptor 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.

(5) Skolnick, P.; Wong, G. In Imidazopyridines in Anxiety Disor-

ders: A Novel Experimental and Therapeutic Approach; Zivkovic,

B., Langer, S., Bartholini, G., Eds.; Raven Press: New York,

1992.

(15) Lumma, W. C.; Hartman, R. D.; Saari, W. S.; Englehardt, E. L.

Piperazinylquinoxalines With Central Serotoninmimetic Activ-

ity. J . Med. Chem. 1981, 24, 93-101.

(16) Djerassi, C.; Engle, R. R.; Bowers, A. The Direct Conversion of

Steroidal ∆⁵-3â-Alcohols to ∆⁵- and ∆⁴-3-Ketones. J . Org. Chem.

1956, 21, 1547-1549.

(17) Sethy, V. H.; Harris, D. W. Determination of Biological Activity

of Alprazolam, Triazolam and Their Metabolites. J . Pharm.

Pharmacol. 1982, 34, 115-116.

(18) Im, W. B.; Blakeman, D. P. Correlation Between γ-Aminobutyric

Acid_AReceptor Ligand-Induced Changes in t-Butylbicyclo-

phosphoro[³⁵S]thionate Binding and ³⁶Cl^-Uptake in Rat Cere-

brocortical Membranes. Mol. Pharmacol. 1991, 39, 394-398.

(19) 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.

(20) 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, â₂γ₂and

R₁â₂γ₂Subtypes of Cloned GABA_AReceptors. Mol. Pharmacol.

1993, 44, 866-870.

(21) Rudzik, A. D.; Hester, J . B.; Tang, A. H.; Straw, R. N.; Friis, W.

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.

(22) These standards had ED₅₀s in the range of 0.17-4.5 mg/kg.

(23) (a) Pritchett, D. B.; Seeburg, P. H. γ-Aminobutyric Acid_A

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 GABA_A-Benzodiazepine Receptors Produced in Transfected

Cells. Science 1989, 245, 1389-1392.

Partial Agonists at GABA_A/ Benzodiazepine Receptor

J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 175

(24) Luddens, H.; Pritchett, D. B.; Kohler, M.; Killisch, I.; Keinanen,

K.; Monyer, H.; Sprengel, R.; Seeburg, P. H. Cerebellar GABA_A

Receptor Selective for a Behavioral Alcohol Antagonist. Nature

1990, 346, 648-651.

(25) Puia, G.; Vicini, S.; Seeburg, P. H.; Costa, E. Influence of

Recombinant γ-Aminobutyric Acid_AReceptor Subunit Composi-

tion on the Action of Allosteric Modulators of γ-Aminobutyric

Acid_A-Gated Cl^-Currents. Mol. Pharmacol. 1991, 39, 691-696.

(26) 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 GABA_AChloride Channels and Benzodiazepine

Binding Sites in Baculovirus Infected Insect Cells. Bio/ Technol-

ogy 1992, 10, 679-681.

(27) Hamilton, B. J .; Lennon, D. J .; Im, H. K.; Im, W. B.; Seeburg,

P. H.; Carter, D. B. Stable Expression of Cloned Rat GABA_A

Receptor Subunits in a Human Kidney Cell Line. Neurosci. Lett.

1993, 153, 206-209.

(28) 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.

(29) 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.

(31) Allinger, N. L. Conformational Analysis. 130. MM2. A Hydro-

carbon Force Field Utilizing V₁and V₂Torsional Terms. J . Am.

Chem. Soc. 1977, 99, 8127-8134.

(32) 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.

(33) Mickelson, J . W.; J acobsen, E. J .; Im, H. K.; Im, W. B.; Schreur,

P. K. D.; Sethy, V. H.; Tang, A. H.; VonVoigtlander, P. F.; McGee,

J . E.; Petke, J . D. Unpublished observations.

(34) Villar, H. O.; Davies, M. F.; Loew, G. H.; Maguire, P. A.

Molecular Models for Recognition and Activation at the Benzo-

diazepine Receptor: A Review. Life Sci. 1991, 48, 593-602.

(35) Tebib, S.; Bourguignon, J .-J .; Wermuth, C.-G. The Active Analog

Approach Applied to the Pharmacophore Identification of Ben-

zodiazepine Receptor Ligands. J . Comput.-Aided Mol. Des. 1987,

1, 153-170.

(36) 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.

(37) Fryer, R. I. Ligand Interactions at the Benzodiazepine Receptor.

In Comprehensive Medicinal Chemistry Volume 3; Hansch, C.,

Sammes, P. G., Taylor, J . B., Eds.; Pergamon Press: New York,

1990.

(38) Maxim, N.; Zugravescu, I.; Fulga, I. Sur L’acide R-Pyrrole-

Carbonique et Quelques Amides Derivees. (R-Pyrrole Carboxylic

Acid and Some Amide Derivatives.) Bull. Soc. Chim. 1938, 5,

44-48.

(30) 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.

J M940765F

Article Doi

High-affinity partial agonist imidazo[1,5-a]quinoxaline amides, carbamates, and ureas at the γ-aminobutyric acid A/benzodiazepine receptor complex

DOI: 10.1021/jm940765f

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm940765f

Full text of DOI:10.1021/jm940765f

Products guided by the article

R&D Labs maybe for 148010-70-2

Relevant to this article

Hot Product