New insight into the central benzodiazepine receptor-ligand interactions: Design, synthesis, biological evaluation, and molecular modeling of 3-substituted 6-phenyl-4 H -imidazo[1,5- a][1,4]benzodiazepines and related compounds

ARTICLE

pubs.acs.org/jmc

New Insight into the Central Benzodiazepine ReceptorꢀLigand

Interactions: Design, Synthesis, Biological Evaluation, and Molecular

Modeling of 3-Substituted 6-Phenyl-4H-imidazo[1,5-a]-

[1,4]benzodiazepines and Related Compounds^†

Maurizio Anzini,*^,‡Salvatore Valenti,^‡Carlo Braile,^‡Andrea Cappelli,*^,‡Salvatore Vomero,^‡Stefano Alcaro,^§

Francesco Ortuso,^§Luciana Marinelli,^||Vittorio Limongelli,^||Ettore Novellino,^||Laura Betti,^{^}

Gino Giannaccini,^{^}Antonio Lucacchini,^{^}Simona Daniele,^{^}Claudia Martini,^{^}Carla Ghelardini,^#

Lorenzo Di Cesare Mannelli,^#Gianluca Giorgi,^∞Maria Paola Mascia,^b,1and Giovanni Biggio^b,1

^‡Dipartimento Farmaco Chimico Tecnologico and European Research Centre for Drug Discovery and Development,

Universitꢀa degli Studi di Siena, Via A. Moro, 53100 Siena, Italy

§

ꢀ

Dipartimento di Scienze Farmacobiologiche, Universitꢀa “Magna Græcia” di Catanzaro, Complesso Nini Barbieri, 88021 Roccelletta di

Borgia (CZ), Italy

Dipartimento di Chimica Farmaceutica e Tossicologica, Universitꢀa di Napoli “Federico II”, Via D. Montesano 49, 80131 Napoli, Italy

^{^}Dipartimento di Psichiatria, Neurobiologia Farmacologia e Biotecnologie, Universitꢀa di Pisa, Via Bonanno 6, 56126 Pisa, Italy

^#Dipartimento di Farmacologia Preclinica e Clinica “M. Aiazzi Mancini”, Universitꢀa degli Studi di Firenze, Viale G. Pieraccini 6,

50139 Firenze, Italy

^∞Dipartimento di Chimica, Universitꢀa degli Studi di Siena, Via A. Moro, 53100 Siena, Italy

^bDipartimento di Biologia Sperimentale “B. Loddo”, Sezione Neuroscienze, Universitꢀa degli Studi di Cagliari, Cittadella Universitaria,

S.S. 554, Km. 4.500, 09042 Monserrato (CA), Italy

¹Istituto di Neuroscienze, Consiglio Nazionale delle Ricerche, Cittadella Universitaria, S.S. 554, Km 4.500, 09042 Monserrato (CA), Italy

S Supporting Information

b

ABSTRACT: 3-Substituted 6-phenyl-4H-imidazo[1,5-a][1,4]benzodiazepines and related com-

pounds were synthesized as central benzodiazepine receptor (CBR) ligands. Most of the

compounds showed high aﬃnity for bovine and human CBR, their K_ivalues spanning from the

low nanomolar to the submicromolar range. In particular, imidazoester 5f was able to promote a

massive ﬂow of ³⁶Cl^ꢀin rat cerebrocortical synaptoneurosomes overlapping its eﬃcacy proﬁle

with that of a typical full agonist. Compound 5f was then examined in mice for its

pharmacological eﬀects where it proved to be a safe anxiolytic agent devoid of the unpleasant

myorelaxant and amnesic eﬀects of the classical 1,4-benzodiazepines. Moreover, the selectivity

of some selected compounds has been assessed in recombinant R₁β₂γ₂L, R₂β₁γ₂L, and

R₅β₂γ₂L human GABA_Areceptors. Finally, some compounds were submitted to molecular

docking calculations along with molecular dynamics simulations in the Cromer’s GABA_A

homology model.

’ INTRODUCTION

(3β-hydroxypregn-5-ene-20-one), allopregnenolone (3β-hy-

droxy-5R-pregn-16-en-20-one)], and the barbiturates [e.g.,

pentobarbital (5-ethyl-5-(1-methylbutyl)barbituric acid)], are

prescribed as sedatives, muscle relaxants, anxiolytics, and

anticonvulsants. Conversely, negative GABA_Amodulators,

such as BDZ inverse agonists [e.g., β-carboline DMCM (1)

and FG-7142 (2), Chart 1] and the neuroactive steroids [e.g.,

pregnenolone sulfate, dehydroepiandrosterone sulfate (3β-

hydroxyandrost-5-en-17-one)], have anxiogenic and convul-

sant eﬀects.^3ꢀ6Neutral modulators such as the central benzo-

diazepine receptor (CBR) antagonist ﬂumazenil (3) (ethyl

The action of γ-aminobutyric acid (GABA) on the GABA_A

chloride channel complex is capable of controlling the excitability

of many central nervous system (CNS) pathways, and it is

estimated that over 20ꢀ50% of all central synapses utilize GABA

as their transmitter, depending on the brain region.¹GABA_A

receptors are ligand-gated chloride ion channels that contain

allosteric sites in addition to the agonist binding site, through

which other agents can modulate receptor function.²Positive

modulators of the GABA_Areceptors, such as the classical

1,4-benzodiazepines [BDZ, e.g., diazepam (7-chloro-1,3-dihydro-1-

methyl-5-phenyl-2H-1,4-benzodiazepin-2-one), lorazepam (7-

chloro-5-(o-chlorophenyl)-1,3-dihydro-3-hydroxy-2H-1,4-benzo-

diazepin-2-one)], the neuroactive steroids [e.g., pregnenolone

Received: February 11, 2011

Published: July 14, 2011

r

2011 American Chemical Society

5694

dx.doi.org/10.1021/jm2001597 J. Med. Chem. 2011, 54, 5694–5711

|

Journal of Medicinal Chemistry

ARTICLE

Chart 1. Structures of Reference Compounds

Chart 2. Structures of 3-Substituted 6-Phenyl-4H-imidazo-

[1,5-a][1,4]benzodiazepines 5 and Related Compounds

6 and 7

account for the sedative and muscle relaxant eﬀects of non-

selective BDZ agonists, whereas those with an R₂or R₃subunit

mediate the anxiolytic and anticonvulsant eﬀects.^21ꢀ23

In situ hybridization and immunocytochemistry studies have

shown that the R₅GABA_Areceptor subtype is preferentially

expressed at high levels in the hippocampus, although it is also

found to a lesser degree within the cortex and the olfactory bulb.^24,25

The receptors are thought to be located extrasynaptically²⁶and

play a role in tonic inhibition of hippocampal CA1 pyramidal

neurons.²⁷Furthermore, the hippocampal R₅-mediated tonic

conductance is highly sensitive to anesthetics, which are known

to produce a pronounced amnesic eﬀect in vivo.²⁸Although

many regions of the CNS are involved in cognitive functions,

the hippocampus clearly plays a key role.²⁹It has been sugges-

ted that BDZ agonists may exert their amnesic eﬀects by

modulating hippocampal function, since the anterograde

rather than retrograde amnesia is similar to deﬁcits induced

by hippocampal lesions in animals and man.³⁰It has therefore

been hypothesized^11,31that R₅subtype-selective inverse ago-

nists could be procognitive but devoid of anxiogenic and

proconvulsant eﬀects associated with activity at other GABA_A

receptor subtypes. Up to now, only modest progress has been

made in the design of receptor-subtype selective ligands,

although numerous papers detailing success in achieving

selectivity in the synthesis of BDZ ligands that exhibit 100-

fold selectivity toward particular GABA receptor isoforms have

been published.^32ꢀ35Unfortunately, as Harris et al. stated in a

very recent paper: “BDZ ligands that exhibit true speciﬁcity to a

single GABA_Areceptor isoform (aﬃnities 1000-fold higher for

one GABA_Areceptor isoform over all other isoforms) are to

date rare.”³⁶Thus, within a large research project aimed at

deepening the study of ligandꢀCBR interactions and in order

to extend knowledge about our previously reported binding

mode³⁷of two imidazo[1,5-a][1,4]benzodiazepine esters (4a,b)

(Chart 1), we focused our attention on the synthesis of a

new series of 3-substituted 6-phenyl-4H-imidazo[1,5-a][1,4]-

benzodiazepines (5aꢀu) and related compounds (6 and 7)

(Chart 2), the biological activity of which was assayed in vitro

and in vivo and the results of these studies were rationalized

through extensive computational studies.

8-ﬂuoro-5,6-dihydro-5-methyl-6-oxo-4H-imidazo[1,5-a][1,4]-

benzodiazepine-3-carboxylate) (Chart 1) bind to GABA_Areceptor

but have no intrinsic activity, although ﬂumazenil antagonizes the

eﬀects of both positive andnegative GABA_Amodulators that act via

the CBR.

Positive GABA_Amodulators are useful in inducing not only

sedation and muscle relaxation prior to surgical procedures but

also amnesia.⁷The amnesic eﬀects of the positive modulators in

animal and man are well established,^8ꢀ10and thus, it has been

hypothesized that the negative modulators being procogni-

tive^11,12may have the opposite eﬀect. Indeed improvements in

learning and memory dysfunction have been reported after

dehydroepiandrosterone administration to individuals with low

dehydroepiandrosterone sulfate levels,¹³but further studies have

failed to demonstrate a signiﬁcant eﬀect of these steroids on

cognitive abilities.^14,15Thus, the therapeutic use of neurosteroids

for cognitive dysfunction is of uncertain value.¹⁶Similarly, non-

selective BDZ inverse agonists cannot be used to treat neurolo-

gical disorders associated with cognitive impairment because of

their profound anxiogenic and proconvulsant liabilities.¹⁷The

discrete localization of the diﬀerent BDZ-sensitive GABA_A

receptors subtypes provides restricted targets within the brain,

potentially oﬀering improvement in the therapeutic window for

negative GABA_Areceptors modulators.

At present, a total of 21 subunits (R_1ꢀ6, β_1ꢀ4, γ_1ꢀ4, δ, ε, π, θ,

and F_1ꢀ3) have been cloned and sequenced with the majority of

GABA_Areceptors comprising of R, β, and γ subunits arranged in

a 2:2:1 stoichiometry. The subunits may theoretically coassemble

to form a plethora of structurally diﬀerent ion pores, but among

the multitude of possible receptor subtype combinations, only

those that contain a γ₂or γ₃¹⁸subunit in conjunction with R_1,

R_2,R₃, or R₅seem to bind BDZ ligands with signiﬁcant aﬃnity.

In particular, BDZ pharmacological eﬀects are strictly related to

the R isoform.¹⁹

The benzodiazepine binding pocket appears to be located at

the interface of γ and R subunits, with the residues on both

proteins making signiﬁcant contributions to the active site. Since

the GABA binding site requires contributions from both an R

and β subunit, we therefore need only consider channels com-

posed of R, β, and γ subtypes for our purposes.²⁰

In particular, diﬀerent halogen atoms or small substituents

were introduced into the benzo ring and the pendent phenyl with

the aim of modulating aﬃnity and intrinsic activity of imidazo-

[1,5-a][1,4]benzodiazepine esters 5. Furthermore, compounds

5pꢀu were synthesized with the aim of probing a more hindered

or a more polar side chain than carboxyethyl moiety. Naphthyl

derivative 5o and naphtho-fused compound 6 were synthesized

to investigate the steric tolerance in correspondence to the benzo

Studies using molecular genetic or pharmacological approaches

have indicated that GABA_Areceptors containing an R₁subunit

5695

dx.doi.org/10.1021/jm2001597 |J. Med. Chem. 2011, 54, 5694–5711

Journal of Medicinal Chemistry

ARTICLE

Scheme 1. Imidazo Annulation of Diazepinone Derivatives 8aꢀf,iꢀo and 10^a

^aReagents: (i) t-BuOK, ClPO(OEt)₂, THF; (ii) CNCH₂COOEt or CNCH₂COO-t-Bu, t-BuOK.

Scheme 2. Synthesis of 3-Substituted Benzodiazepin-2-one

Derivatives 8e,i and 9^a

Scheme 3. Synthesis of 7-Chloro-5-naphthyl-1,4-benzodia-

zepinone 8o^a

^aReagents: (i) 1-bromonaphthalene, Mg, dry EtOEt; (ii) BrCH₂COBr,

CH₂Cl₂; (iii) liquid NH₃, MeOH.

^aReagents: (i) CH₃CH(NH₂)COOEt HCl or MeOOCCH(NH₂)-

3

CH₂COOMe HCl, pyridine.

3

Scheme 4. Synthesis of 5-(4-Chlorophenyl)naphthodiazepinone

Derivative 10^a

and the pendent phenyl ring, respectively. Finally, conforma-

tional constrained derivatives 5e,i and 7 were designed in order

to evaluate our hypothesis on the orientation of the carboxyethyl

group with respect to the imidazo ring.³⁷

’ CHEMISTRY

The synthesis of imidazoesters 5aꢀf,iꢀp and 6 was carried

out as depicted in Scheme 1. Brieﬂy, the imidazo annulation of

diazepinone derivatives 8aꢀf,i-o and 10 was accomplished via

the corresponding phosphonate intermediate by reaction with

ethyl isocyanoacetate or tert-butyl isocyanoacetate in the pre-

sence of potassium tert-butoxide, providing the imidazoesters

5aꢀf,i-p and 6 in satisfactory yield.³⁸

^aReagents: (i) 4-bromochlorobenzene, Mg, dry EtOEt; (ii) BrCH₂COBr,

CH₂Cl₂; (iii) liquid NH₃, MeOH.

The synthetic route used to prepare benzodiazepinones

8aꢀd,f,jꢀn starting from the suitable 2-aminobenzophenones

involved the sequential treatment with bromoacetyl bromide in

dichloromethane to give the corresponding bromoacetamides

and then with ammonia in ethanol at reﬂux or the reaction with

glycine ethyl ester in pyridine at reﬂux.^39,40On the other hand,

racemic 3-substituted 1,4-benzodiazepinones 8e,i and 9 were

prepared by condensation of the suitable benzophenone deriva-

tive (11e,i) with alanine ethyl ester or aspartic acid dimethyl ester

hydrochloride in pyridine (Scheme 2).

The naphtho-fused diazepinone 10 was synthesized as sket-

ched in Scheme 4. In this sequence, the conversion of 1-amino-

2-naphthonitrile 14⁴¹into R-aminonaphthophenone 15 was

accomplished by treatment with Grignard’s reagent, 4-chloro-

phenylmagnesium bromide, and the condensation of 15 with

bromoacetyl bromide and successive cyclization of the bromoa-

cetamide in the presence of liquid NH₃to accomplish the ring

closure as for compound 8o aﬀorded naphthodiazepinone 10.

The alkaline hydrolysis of imidazo ester 5k aﬀorded corre-

sponding acid 16, which by sequential treatment with isobutyl

chloroformate and the proper amine (or ammonia) was converted

into expected carboxamides 5qꢀs as depicted in Scheme 5.

Dehydration of amide 5q, by means of phosphorus oxychlor-

ide in dry benzene, aﬀorded imidazocarbonitrile 5t in good yield,

as the precursor of tetrazole derivative 5u. Compound 5t under-

went a 1,3-dipolar cycloaddition with trimethyltin azide to give,

after protection with triphenylchloromethane and deprotection

with HCl, imidazotetrazole derivative 5u (Scheme 6).

The synthesis of benzodiazepinone 8o was carried out starting

from 2-amino-5-chlorobenzonitrile 12, which was treated with

Grignard’s reagent, 1-naphthylmagnesium bromide, in diethyl

ether to aﬀord the expected naphthyl derivative 13 (Scheme 3).

Condensation of 13 with bromoacetyl bromide in dichloro-

methane gave the respective bromoacetamide, which was in turn

cyclized into the expected lactam 8o in the presence of liquid

ammonia in methanol at reﬂux.

5696

dx.doi.org/10.1021/jm2001597 |J. Med. Chem. 2011, 54, 5694–5711

Journal of Medicinal Chemistry

ARTICLE

Scheme 5. Synthesis of Carboxamides 5qꢀs^a

^aReagents: (i) 2 N NaOH, EtOH; (ii) ClCOO-i-Bu, TEA, CHCl₃; (iii) NH₃or NH₂CH₃or NH(CH₃)₂.

Scheme 6. Synthesis of 1H-Tetrazole Derivative 5u^a

^aReagents: (i) POCl₃, benzene; (ii) (CH₃)₃SnN₃, xylene; (iii) 10 N NaOH, CH₂Cl₂, THF, and then ClC(C₆H₅)₃; (iv) 10% HCl, THF.

The synthesis of conformationally constrained derivative 7

was achieved starting from 9, which was reduced in the presence

of lithium aluminum hydride to aﬀord the corresponding hydro-

xyethyl derivative 17. After the protection of the alcoholic

function by means of tert-butylchlorodimethylsilane,⁴²com-

pound 18 was cyclized to the imidazo ester 19 and deprotected

in the presence of (n-Bu)₄NF to give alcohol 20 which, by

treatment with sodium hydride in DMF, gave the expected

imidazolactone 7 in satisfactory yield (Scheme 7).

interaction at the CBR binding site. In particular, the presence

of small atoms (e.g., H or F) in position 8 is required for

nanomolar CBR aﬃnity, whereas the presence of the bulkier

chlorine atom produces a drop in CBR aﬃnity that achieves

about 1 order of magnitude when F, Cl, or CH₃is present in

the 4⁰position.

A particular case is represented by the 8-Cl-3⁰-NO₂derivative

5n, which shows aﬃnity in the low nanomolar range slightly

higher than its 8-ﬂuoro analogue 5g. This result conﬁrms that

the electronic properties of NO₂group have a dominant role in

the interaction with the polar residues of the CBR binding

site.³⁷

’ RESULTS AND DISCUSSION

In Vitro Binding. The binding affinity of the imidazo[1,5-a]-

[1,4]benzodiazepine derivatives 5aꢀu, 6, and 7 to the benzo-

diazepine receptor in bovine and human cortical membranes was

determined by means of competition experiments against the

radiolabeled antagonist [³H]flumazenil and expressed as K_i

values when the compounds inhibited radioligand binding by

more than 80% at the fixed concentration of 10 μM. The in vitro

efficacy of the tested compounds was measured by the GABA

ratio (GR), which predicts the pharmacological profile of a CBR

ligand.^43ꢀ45The resulting value, expressed as a ratio of K_iwithout

GABA to K_iwith GABA, is nearly 2 for a full agonist and around 1

for an antagonist; partial agonists show intermediate values

between 1 and 2, while a GABA ratio value below 1 is typical

for inverse agonists. The data summarized in Table 1 show that

most of the newly synthesized compounds interact with CBR, show-

ing affinities from the low nanomolar to the submicromolar ranges.

The structureꢀaﬃnity relationship analysis shows that sub-

stituents in positions 8 and 4⁰of the 6-phenylimidazo[1,5-a]-

[1,4]benzodiazepine scaﬀold play a key role in modulating the

In the series of 8-unsubstituted derivatives 5aꢀd, 5v (Ro-15-

1624), and 5x (Ro-23-2896) (Chart 3), the introduction of

substituents in the 4⁰-position produces a slight aﬃnity decrease

when F, Cl, and CH₃substituents are taken into consideration,

while the eﬀect is more marked in the case of 4⁰-methoxy

derivative 5d. Diﬀerently, in the series of 8-chloro derivatives

5jꢀm, 5y (Ro-15-8670, Chart 3) the aﬃnity decrease of about 1

order of magnitude is also observed in 4⁰-ﬂuoro derivative 5j, 4⁰-

chloro derivative 5k, and 4⁰-methyl derivative 5l. This discre-

pancy in the structureꢀaﬃnity relationship trends suggests that

the introduction of a chlorine atom in position 8 of the tricyclic

nucleus leads to a charge redistribution in the benzofused ring,

modulating the polar interaction within the binding site. A

comparison of GABA ratio values suggests that the presence of

a hydrogen atom in position 8 produces antagonist-like proﬁles.

On the other hand, the imidazobenzodiazepine derivatives bear-

ing a chlorine atom in the same position show diﬀerent eﬃcacy

proﬁles spanning from the described⁶⁰agonist properties of 5y

to the antagonist ones shown by 4⁰-methyl derivative 5l and by

5697

dx.doi.org/10.1021/jm2001597 |J. Med. Chem. 2011, 54, 5694–5711

Journal of Medicinal Chemistry

ARTICLE

Scheme 7. Synthesis of Conformationally Constrained

Derivative 7^a

synaptoneurosomes.^47,48The synaptic chloride conductance

effected by GABA activating the GABA_Areceptor complex is

modulated by ligands acting at the CBR. Full agonists increase

current, and antagonists have no effect, while inverse agonists

decrease ion flow. Among the title compounds, 5a,b,c,f,n were

selected and compared to flunitrazepam, flumazenil, and ethyl

β-carboline, a full agonist, an antagonist, and an inverse agonist,

respectively (Figure 2).

On the basis of the results obtained with these standards, it was

possible to show the course of the uptake and to make a direct

comparison with the compounds under examination. In parti-

cular, compounds 5a,b,n did not show a signiﬁcant modiﬁcation

of ³⁶Cl^ꢀinﬂux, behaving in a manner similar to that of the

antagonist ﬂumazenil. Although a very slight decrease for com-

pound 5b and a modest increase for compound 5n could be

recognized, no eﬀect on the ion ﬂow seems to be elicited by

compound 5a. Thus, with some discrepancy with respect to their

GABA ratio (GR) values [e.g., 5a (GR = 0.77) and 5b (GR =

0.84)], these compounds appear to behave as CBR antagonists as

well as compound 5n, the GABA ratio of which is very close

to 1.0.

A quite diﬀerent behavior is shown by compounds 5c,f, the

eﬀect of which on ³⁶Cl^ꢀinﬂux is surprising if their GABA ratio

values, 0.97 and 1.0, respectively, are taken into account. In fact,

the massive ion ﬂow promoted by these compounds is comple-

tely unexpected, making the eﬃcacy proﬁle of 5c similar to the

one of a partial agonist, while that of 5f, overlapping the ﬂunitra-

zepam proﬁle, is therefore typical of a full agonist. The rationa-

lization of these results is far from obvious in light of the discre-

pancies observed when using the two diﬀerent intrinsic eﬃcacy

determination methods.

In Vivo Efficacy. In order to assess the unexpected agonist

properties, compound 5f was examined in mice for its pharma-

cological effects. Four potential benzodiazepine actions were

considered: potential anxiolytic effect was screened by means of

light/dark box test, while the myorelaxant effect was measured by

means of the rotarod test; the hole-board test was performed to

verify the effect of 5f on mouse spontaneous motility and

explorative activity, and finally the mouse learning and memory

impairment was evaluated by passive avoidance test.

Eﬀects on mouse anxiety of the newly synthesized molecule

and diazepam were studied by means of a light/dark box

apparatus. In our experiments compound 5f showed a statisti-

cally signiﬁcant anxiolytic-like eﬀect starting from a dose of

1.0 mg kg^ꢀ1po, demonstrated by the increasing of time spent

in the light compartment of the light/dark apparatus (Figure 3).

At 10 mg/kg^ꢀ1po, compound 5f reached an eﬀect comparable to

that exhibited by the reference drug diazepam at 1.0 mg kg^ꢀ1ip,

showing the same eﬃcacy but lower potency.

The eﬀects of compound 5f on animal motor coordination

were investigated by means of the mouse rotarod test as a

screening method to discover any myorelaxant eﬀect (Figure 4).

Compound 5f at a dose of 10 mg kg^ꢀ1po was unable to modify

motor coordination, since 5f did not increase the number of

falls from the rotating rod. Diazepam at 1.0 mg kg^ꢀ1ip showed

a signiﬁcant muscle relaxant eﬀect (data not shown).

The hole board test, which is known to highlight neurological

or muscular alterations, demonstrates that 10 mg kg^ꢀ1po

compound 5f did not show signiﬁcant diﬀerences either in

spontaneous motility (number of movements) or in explorative

activity (number of inspections) with respect to the carboxy-

methylcellulose (CMC) treated control mice (Figure 5).

^aReagents: (i) LiAlH₄, THF; (ii) TBDMSCl, imidazole, CH₂Cl₂; (iii)

t-BuOK, ClPO(OEt)₂, THF; (iv) CNCH₂COOEt, t-BuOK; (v)

n-Bu₄NF, THF; (vi) NaH, DMF.

3⁰-nitro derivative 5n through the partial agonist proﬁles of 4⁰-ﬂuoro

derivative 5j and 4⁰-chloro derivative 5k.

In order to obtain information on the bioactive conformation

of these 6-phenylimidazo[1,5-a][1,4]benzodiazepine deriva-

tives, the ester carbonyl of compound 5v was constrained in cis

position with respect to imidazole nitrogen as in lactone 7

(Figure 1). Unfortunately, compound 7 shows a CBR aﬃnity

in the submicromolar range very similar to that shown by

4-methyl derivative 5e and about 1 order of magnitude lower

than the unconstrained parent compound 5v. Therefore, this

conformational constraining approach is unsuitable for study of

the interaction of these imidazodiazepine derivatives with CBR.

The idea is that the introduction of a carbon atom in the

4-position of the tricyclic nucleus is not tolerated by the CBR

binding site (compare 5v with 5e and 5f with 5i) and the

alteration in the binding phenomenon (leading to submicro-

molar aﬃnity) could mask the true eﬀect of the carbonyl group

constraining.

The modiﬁcation of the ester side chain in 8,4⁰-dichloro

derivative 5k produced inactive compounds 5pꢀu. Moreover,

the benzo fusion of the pendent phenyl ring of 5y leads to an

aﬃnity decrease of about 1 order of magnitude (compare 5o with

5y) and the same structural modiﬁcation leading to compound 6

is not tolerated (compare 6 with 5b).

In Vitro Efficacy. A second and more direct measure of in vitro

efficacy was determined by a ³⁶Cl^ꢀuptake assay in rat cerebrocortical

5698

dx.doi.org/10.1021/jm2001597 |J. Med. Chem. 2011, 54, 5694–5711

Journal of Medicinal Chemistry

ARTICLE

Table 1. Inhibition of [³H]Flumazenil Speciﬁc Binding to Bovine and Human Cortical Membranes and GABA Ratios of the Title

and Reference Compounds

K_i( SEM^a(nM)

GABA ratio^b

compd

R

R₁

R₂

Ar

bovine

human

bovine

human

5a

COOEt

H

F

H

Me

H

Me

H

p-ﬂuorophenyl

p-chlorophenyl

p-methylphenyl

p-methoxyphenyl

phenyl

11.5 ( 3.8

12.7 ( 1.1

11.4 ( 3.8

110 ( 25

453 ( 68.2

10.9 ( 2.5

4.4 ( 0.3

14.8 ( 1.4

510 ( 90.3

188 ( 35.2

236 ( 84.6

199 ( 44.8

210 ( 73.2

1.7 ( 0.2

186 ( 78.2

1299 ( 280

>10000

20.8 ( 0.2

22.5 ( 2.9

18.5 ( 2.5

120 ( 10.8

750 ( 65.8

22.3 ( 0.3

9.7 ( 0.7

0.77

0.84

0.97

0.78

1.30

1.0

0.96

0.87

1.10

5b (Ro-22-9735)^c

COOEt

COO-t-Bu

CONH₂

CONHMe

CONMe₂

CN

5c

5d

5e

5f

5g^d

5h^d

p-ﬂuorophenyl

m-nitrophenyl

p-nitrophenyl

p-ﬂuorophenyl

p-chlorophenyl

p-methylphenyl

p-methoxyphenyl

m-nitrophenyl

1-naphthyl

1.05

1.20

0.86

F

1.19

0.89

1.19

1.25

1.30

1.05

F

20.0 ( 1.8

550 ( 50.7

212 ( 38.6

210 ( 23

5i

F

5j

Cl

5k

5l

178 ( 38.3

227 ( 45.4

2.3 ( 0.3

5m

5n

0.99

1.30

0.98

5o

216 ( 61.5

5p

p-chlorophenyl

5q

5r

>10000

5s

>10000

5t

1654 ( 220

>10000

5u

5-(1H-tetrazole)

6

>10000

7

283 ( 118

5.17 ( 0.2

1.9 ( 0.09

0.95

1.68

1.03

ﬂunitrazepam

6.5 ( 0.5

1.60

1.01

ﬂumazenil

2.1 ( 0.08

^aK_ivalues are the mean ( SEM of three independent determinations. ^bGABA ratio = (K_iwithout GABA)/(K_iwith 50 μM GABA). See ref 40.

c

^dCompounds 5g,h are referred to in the introduction as 4a,b, respectively (see ref 37).

Chart 3. Structures, Aﬃnity Values for CBR, and Intrinsic Activity of Compounds 5vꢀy⁴⁶

In order to investigate the eﬀect of compound 5f on learning

and memory, the mice performance on “passive avoidance test”,

an experiment in which the animal learns to avoid a noxious event

by suppressing a particular behavior, was investigated. In this

assay, the diﬀerence between the retention latencies of 5f-treated

mice and CMC-treated controls was not statistically signiﬁcant at

any tested dose (0.1, 1.0, 3.0, and 10 mg kg^ꢀ1po, data not

shown), showing no amnesic eﬀect. Moreover, the eﬀect of 5f as

antiamnesic compound was evaluated as the capability to prevent

the amnesic eﬀect of 1.5 mg kg^ꢀ1scopolamine ip. At 3 mg kg^ꢀ1po,

compound 5f could signiﬁcantly increase the retention session

latency and at 10 mg kg^ꢀ1po completely prevented the amnesic

eﬀect of scopolamine (Figure 6).

Interaction with GABA Receptor Subtypes. On the basis of

the above results dealing with binding affinity, intrinsic activity,

and behavioral tests, we decided to verify and evaluate the binding

5699

dx.doi.org/10.1021/jm2001597 |J. Med. Chem. 2011, 54, 5694–5711

Journal of Medicinal Chemistry

ARTICLE

Figure 4. Eﬀect on motor coordination 45 min after the treatment.

Each value represents the mean of at least 15ꢀ20 mice: (∧) P < 0.05, (/)

P < 0.01 vs saline/CMC-treated mice.

Figure 1. Crystal structure of compound 7. Ellipsoids enclose 50%

probability.

Figure 5. Eﬀects on neurological or muscular alterations. Each value

represents the mean of at least 15ꢀ20 mice: (∧) P < 0.05, (/) P < 0.01 vs

saline/CMC-treated mice.

Figure 2. ³⁶Cl^ꢀuptake measured in rat cerebrocortical synaptoneuro-

somes for compounds 5a (magenta), 5b (bleu), 5c (cyan), 5f (green),

5n (red), ﬂunitrazepam (empty circles), ﬂumazenil (empty diamonds),

and ethyl β-carboline (empty squares).

Figure 6. Learning and memory test. Each value represents the mean

of at least 15ꢀ20 mice: (∧) P < 0.05, (/) P < 0.01 vs saline/CMC-

treated mice.

of CBR ligands, were evaluated by means of two different expe-

rimental approaches based on the two-electrode voltage clamp

at GABA_Areceptor subtypes and radioligand assays, performed

on HEK293 cells stably transfected with R₂β₂γ₂or R₅β₃γ₂

isoform.

The two-electrode voltage-clamp technique was used as a ﬁrst

screening of compounds 5b,c,f,g in order to assess their ability

to modulate GABA-evoked Cl^ꢀcurrents in Xenopus laevis oocy-

tes expressing recombinant R₁β₂γ₂L, R₂β₁γ₂L, and R₅β₂γ₂L

human GABA_Areceptors (Table 2). Compounds 5b,c,f,g (1ꢀ

10 μM) did not modify the responses of a GABA concentration

equal to EC_5ꢀ10(approximately 3ꢀ6 μM GABA) and did not

show selectivity for a speciﬁc GABA subunit combination.

Figure 3. Anxiolytic eﬀect. Each value represents the mean of at least

15ꢀ20 mice: (∧) P < 0.05, (/) P < 0.01 vs saline/carboxymethylcellu-

lose (CMC) treated mice.

properties and the pharmacologic profiles of the most repre-

sentative imidazoesters 5b,c,f,g toward the GABA receptor

subtypes. In particular, isoforms of the R subunit (R₁, R₂, and

R₅), which are the most implicated in the pharmacological effects

5700

dx.doi.org/10.1021/jm2001597 |J. Med. Chem. 2011, 54, 5694–5711

Journal of Medicinal Chemistry

ARTICLE

Table 2. Modulation of GABA-Evoked Cl^ꢀCurrents in Xenopus laevis Oocytes Expressing Recombinant R₁β₂γ₂L, R₂β₁γ₂L,

R₅β₂γ₂L Human GABA_AReceptors Elicited by Compounds 5b,c,f,g at 1ꢀ10 μM^a

R₁β₂γ₂L

R₁β₁γ₂L

R₅β₂γ₂L

μM

5b

5c

5f

5g

5b

5c

5f

5g

5b

5c

5f

5g

0.01 +13 ( 7.3 ND

0.1 +1.0 ( 7.0 ND

ND

1

+3.6 ( 4.1 +8.0 ( 39 ꢀ13 ( 9.2 +14 ( 15 ꢀ7.0 ( 7.4 0.0 ( 6.9 ꢀ7.5 ( 10 +10 ( 11 ꢀ0.5 ( 0.5 ꢀ4.0 ( 13 ꢀ1.0 ( 3.0 +27 ( 19

10 +6.3 ( 11 ꢀ4.0 ( 9.4 ꢀ12 ( 5.0 ꢀ24 ( 7.2 ꢀ18 ( 8.0 +8.3 ( 7.6 +6.5 ( 3.0 ꢀ3.3 ( 2.7 0.0 ( 5.0

ꢀ22 ( 6.5 ꢀ16 ( 5.5 ꢀ2.5 ( 9.5

^aEach compound was preapplied for 60s before being coapplied for 30s with an EC_5ꢀ10of GABA (3ꢀ6 μM). Data, expressed as percent of control, are

the mean ( SEM of two to three oocytes.

GABA in a reversible manner. However, the potentiation of the

GABA response induced by diazepam was blocked when the

oocytes were exposed to 5b,c,f,g (10 μM). The ability of these

compounds to block the eﬀect of the benzodiazepine diazepam

clearly demonstrates that these molecules act at the GABA_A

receptor via the benzodiazepine recognition site and that they

are antagonists.

Subsequently, compounds 5c and 5f were studied in radio-

ligand assays in HEK293 cells stably transfected with R₂β₂γ₂or

R₅β₃γ₂GABA_Areceptor subtypes. The results showed that both

compounds were able to bind to R₂β₂γ₂receptor subtype with

an aﬃnity value in the nanomolar range (5c, K_i= 105 ( 10 nM;

5f, K_i= 60 ( 5 nM). In particular, the ability of compound 5f to

bind to the R₂subunit highlights the anxiolytic eﬀects shown in

the light/dark box test (Figure 3). In contrast, both compounds

showed a signiﬁcantly lower aﬃnity toward R₅β₃γ₂receptor

subtype (5c, K_i= 908 ( 90 nM; 5f, K_i= 772 ( 70 nM). Anyway,

5f binding to the R₅subunit could result in positive eﬀects in

learning and memory, as shown in the “passive avoidance test”.

Molecular Modeling Studies. In order to investigate at the

molecular level the interactions that govern the recognition and

binding of our ligands to the CBR, docking calculations by means

of AutoDock 4.0 have been carried out using as target protein

Cromer’s and the Ernst’s GABA_Ahomology model.⁵⁰In these

calculations, only the interface between the R and γ subunits,

where the benzodiazepine binding site is known to be localized,

has been considered as the docking area. Subsequently, the

energy profile and the stability of the most interesting complexes

as obtained from docking studies were assessed through exten-

sive molecular dynamics (MD) studies and ab initio calculations.

Docking experiments, in the Cromer model, proposed similar

binding conformations for compounds 5aꢀc,f,j, where the fused

imidazole ring engages a πꢀπ interaction with Tyr210 and the

fused benzene ring interacts with Phe100 and His102 residues

(Figure 8A). Finally, the pendent phenyl moiety is involved in

electron transfer interactions with Phe77 and Tyr58 rings. Even if

the distance between the nitrogen at position 4 of the benzodia-

Figure 7. Eﬀects of compounds 5b,c,f,g (10 μM) on the diazepam

(DZ) (1.0 μM) enhancement of the GABA_Areceptor function in

Xenopus laevis oocytes expressing (a) R₁β₂γ₂L, (b) R₂β₁γ₂L, and (c)

R₅β₂γ₂L human GABA_Areceptors. Data are expressed as percentage

potentiation of the response induced by GABA at EC₁₀and represent

the mean ( SEM of two to three diﬀerent oocytes.

zepine ring and Thr142 is high for an H-bond formation (∼4 Å),

their proximity would suggest a possible interaction. All these

favorable interactions contribute to achieve a good docking score

(e.g., 5c AutoDock binding energy of ꢀ10.43 kcal/mol) and an

inhibitory activity in the nanomolar range as it results from the

experimental data.

Furthermore, in order to evaluate their pharmacological pro-

ﬁle, the eﬀects of the above-cited compounds (10 μM) were

tested on the diazepam (1.0 μM) enhancement of the GABA_A

receptor function (Figure7). Inagreement with previous reports,⁴⁹

diazepam potentiated the action of an EC_5ꢀ10concentration of

Interestingly, the introduction of a bulkier side chain at

position 3 such as the tert-butyl chain (5p) hampers the proper

location of the ligand in the binding pocket, resulting in a decreased

docking score (AutoDock binding energy of ꢀ6.53 kcal/mol).

This is probably due to the steric clashes of the tert-butyl group

5701

dx.doi.org/10.1021/jm2001597 |J. Med. Chem. 2011, 54, 5694–5711

Journal of Medicinal Chemistry

ARTICLE

Figure 8. Binding conformation of 5f in the CBR receptor as resulted from docking (A) and from subsequent MD (B). The ligand and the interacting

residues are represented as green and cyan sticks, respectively, while the protein is represented as bronze cartoon. (B) The two interacting waters are

shown as red spheres, while hydrogen atoms are not displayed for the sake of clarity.

with the backbone atoms of Tyr141 and Thr142, which are

strongly held in place as they are part of a β-sheet. The

decreased docking score of this compound is in line with its

CBR reduced aﬃnity, which is in the micromolar range, as

measured in the binding assays. Since docking results obtained

using Cromer’s homology model might be aﬀected by the

approximation in the deﬁnition of the protein structure, we

repeated the docking calculation of 5f using Ernst’s model. In

the latter model, it was found that the benzodiazepine ring of 5f

is approximately located in the same region previously occu-

pied in Cromer’s model, but the pendent phenyl ring and the

ester chain inverted their relative position. Thus, in Ernst’s

model, the ester chain ﬁlls the aromatic pocket formed by

Phe77 and Tyr58 and the pendent phenyl ring ﬁlls the region

formed by polar residues such as Thr141 and Thr207, where

Figure 9. (Top) Plot of the rmsd of the heavy atoms of 5j during MD

simulation. The ligand changes its binding mode only after 1 ns of

simulation. (Bottom) Plot of the rmsd of the heavy atoms of 5f during

usually polar groups are bound. Moreover, while docking

results in Cromer’s model showed that only short ester chains

are tolerated because of the presence of Tyr141 and Thr142

MD simulation. This pose was very stable during the whole simulation.

backbone, in Ernst’s model this chain points outside the

receptor and even bulkier ester groups might be allowed,

in stark contrast with structureꢀactivity relationship (SAR)

data herein presented. The above-mentioned considerations

prompted us to discard the docking results coming from

Ernst’s model. The major cause of the docking failure in this

model was likely the orientation of the side chains of some

residues such as (R)-Arg144, (R)-Tyr160, and (γ)-Asn128

that occupy the binding pocket, reducing the space available

for the ligand binding. Thus, we decided to redo the docking

calculations in the Ernst’s model, optimizing these side chain

positions to enlarge the region available for the ligand binding.

Interestingly, when docking of 5f is performed in such a model

the best scored pose is highly similar to that found previously in

Cromer’s model. This supports the reliability of the binding

mode proposed for 5f and suggests that either model can be

used for the study of the benzodiazepine binding once the

binding pocket has been properly optimized.

geometric proﬁles, reinforcing the signiﬁcance of our docking

results and also addressing the relevant biological activity (see

Table 1). In fact, the binding mode of 5f does not substantially

diverge from that calculated by the AutoDock program,⁵¹con-

serving all the main interactions with the protein (Figure 8). This

is also clearly shown by the low root-mean-square deviation

(rmsd) calculated on the heavy atoms of the ligand with respect

to their coordinates at the beginning of the simulation (Figure 9).

In the found binding mode, the fused benzene ring of 5f

engages stable stacking interactions with Phe100 and His102

with the ﬂuorine atom placed in a fruitful position to engage an

halogenꢀπ interaction⁵²with the imidazole ring of His102. The

fused imidazole ring during the MD improves the geometry of its

πꢀπ interaction with Tyr210 side chain, while the pendent

phenyl moiety is involved in hydrophobic contacts with Phe77

and Tyr58 via a stacking and a T-shaped interaction, respectively.

Moreover, during the whole simulation the carbonyl oxygen of

the carbethoxy group is constantly involved in a H-bond inter-

action either directly with Thr207 or through a water bridge with

Thr207 and Tyr210. Similarly, the nitrogen at position 4 of the

benzodiazepine ring has been found to establish a H-bond

with Thr142 either directly or via a neighboring water molecule.

Subsequently, with the aim to assess the energy proﬁle and the

stability of CBR (Cromer’s model) in complex with compounds

representative of the series such as 5a,f,j, molecular dynamics

(MD) studies were carried out. The three complexes showed

diﬀerent behaviors during MD simulations as detailed below.

The complex between 5f and CBR showed stable energetic and

5702

dx.doi.org/10.1021/jm2001597 |J. Med. Chem. 2011, 54, 5694–5711

Journal of Medicinal Chemistry

ARTICLE

imidazobenzodiazepine (IBDZ) nucleus of 5f. The chlorine

substituent is certainly bulkier than the ﬂuorine one of 5f, and

the presence of a His102 side chain in the region of the binding

site that hosts the halogen atom does not allow a comfortable

allocation of the bulkier chlorine atom. Thus, the molecule

slightly moves up in the binding site with the fused benzene

ring located under the upper ﬂexible loop and the ﬂuorophenyl

moiety pointing toward the inner part of the protein (Figure 10).

Although most of the previously described ligandꢀprotein inter-

actions are conserved in this new binding conformation, some of

them are either badly oriented or lost, like for instance the

interaction between the ﬂuorophenyl ring and the Tyr58 side chain.

In order to evaluate the diﬀerences in the electrostatic

potential proﬁle between 5f and 5j, ab initio calculations were

performed to map the electrostatic potential of the two com-

pounds onto the respective molecular surfaces representing the

electron densities (Figure 11). While the ﬂuorine atom of 5f is

entirely electronegative because of its small atomic radius, the

bulkier chlorine atom of 5j shows a considerably less negative

charge. As a consequence, the halogenꢀπ interaction formed by

5f with His102 side chain is stronger than that established by 5j.

On the basis of our MD and ab initio calculations, we can

conclude that the diﬀerent electrostatic proﬁles of 5f and 5j

together with the weaker interactions established by 5j with CBR

are at the base of the lower aﬃnity of 5j for CBR.

Figure 10. Representation of the binding conformation of 5j in the

CBR receptor at the beginning (red) and at the end (green) of the MD

simulation. During the simulation, the ligand changes its conformation,

either losing or weakening key interactions with the protein. The ligand

is displayed as licorice under a transparent surface, while His102 is

shown as space ﬁll. The protein is represented as bronze cartoon, and

hydrogen atoms are not displayed for the sake of clarity.

Our results on the binding modes of 5f and 5j are also useful to

understand the behavior of compound 5a within CBR. Here,

although the halogenꢀπ interaction with His102 is no longer

possible because of the presence of a hydrogen atom in place of

the halogen at position 8 of the IBDZ nucleus, the small size of

the hydrogen does not alter the placement of the ligand in the

binding site. Consequently, similar to 5f, the binding mode of 5a

during the MD simulation is stable conserving most of the above-

described interactions with the surrounding residues and thus

explaining the good aﬃnity of this compound for CBR.

The coherence between docking and MD results supports the

reliability of the proposed binding mode of 5f to CBR.

Figure 11. Electrostatic potential mapped onto the molecular surface of

5f (left) and 5j (right). The calculations were computed at the

HartreeꢀFock level of theory and using 6-31G* as basis set. The scale

of the electrostatic potential is ꢀ0.031 (red) to 0.031 hartree (blue).

’ CONCLUSION

A series of 3-substituted 6-phenyl-4H-imidazo[1,5-a][1,4]-

benzodiazepines and related compounds were designed and

synthesized as central benzodiazepine receptor ligands. Among

the newly synthesized compounds, 3-carboxyethyl esters showed

high aﬃnity for CBR with K_ivalues ranging from the low

nanomolar to the submicromolar range and GABA ratio values

spanning from 0.77 to 1.30. In particular, ethyl 8-ﬂuoro-6-(4-

ﬂuorophenyl)-4H-imidazo[1,5-a][1,4]benzodiazepine-3-car-

boxylate (5f), in spite of its antagonist proﬁle indicated by a

GABA ratio value of 1.0, behaved as a full agonist in the classical

animal model of anxiety light/dark box test and was devoid of

the unpleasant side eﬀects such as myorelaxation and amnesia.

On the other hand, the CBR agonist proﬁle of compound 5f

Notably, the described binding mode shares all the main inter-

actions present in the reﬁned benzodiazepine pharmacophore

model.^52bParticularly, the basic polar points, namely, H1 and H2

in the pharmacophore model, are here represented by the

interactions with Thr207/Tyr210 and Thr142, respectively.

The important lipophilic interaction L1 is that engaged with

Ph100, while the additional lipophilic regions, namely, L2 and

L3, are here formed by His102 and Tyr58/Phe77, respectively.

As for 5j, MD simulation showed a weak stability of the

binding conformation found by docking simulation (Figure 9).

In fact, while AutoDock placed the ligand in a conformation very

similar to that of 5f in the CBR binding site, only after 1 ns of MD

simulation, 5j left the starting position adopting a diﬀerent

binding conformation with the ﬂuorophenyl moiety pointing

toward the inner part of the protein. The diﬀerent behavior of 5f

and 5j during the MD simulation, combined with the signiﬁcantly

lower CBR aﬃnity shown by 5j with respect to 5f, prompted

us to a more detailed investigation of the diﬀerences between

the two ligands. Compound 5j diﬀers from 5f for the presence

of a chlorine atom in place of the ﬂuorine in position 8 of the

36

was conﬁrmed when its eﬃcacy was tested by means of

Cl^ꢀuptake and measured in rat cerebrocortical synaptoneuro-

somes. In this test 5f was capable of promoting a massive ion

ﬂux, mimicking the eﬃcacy proﬁle of a full agonist such as

ﬂunitrazepam.

Moreover, the selectivity of compounds 5b,c,f,g has been

assessed in recombinant R₁β₂γ₂L, R₂β₁γ₂L, and R₅β₂γ₂L hu-

man GABA_Areceptors, measuring their eﬀects on the diazepam

enhancement of the GABA_Areceptor function. These experiments

5703

dx.doi.org/10.1021/jm2001597 |J. Med. Chem. 2011, 54, 5694–5711

Journal of Medicinal Chemistry

ARTICLE

suggested that the selected compounds interacted with the

benzodiazepine recognition site of GABA_Areceptor showing

antagonist proﬁles. These results were conﬁrmed when com-

pounds 5c and 5f were studied in radioligand assays in

HEK293 cells stably expressing R₂β₂γ₂or R₅β₃γ₂GABA_A

receptor subtype. The results showed that both compounds

could bind to R₂β₂γ₂receptor subtype with an aﬃnity in the

nanomolar range. In particular, the ability of compound 5f to

bind to the R₂subunit supports the anxiolytic eﬀects shown in

the light/dark box test. In contrast, both compounds showed a

signiﬁcantly lower aﬃnity toward R₅β₃γ₂receptor subtype.

Anyway, 5f binding to the R₅subunit could result in the

positive eﬀects in learning and memory, as shown in the

“passive avoidance test”.

Finally, the most interesting compounds were studied in

molecular docking simulations along with molecular dynamics

experiments by using Cromer’s GABA_Ahomology model in

order to better understand the ligand interactions at CBR. The

docking experiments showed that only small substituents are well

tolerated in the 4⁰position. Thus, compounds 5aꢀc,f were able

to interact proﬁtably with key amino acids involved in the

binding site, in agreement with the nanomolar range resulting

from the experimental data, whereas the introduction of a bulkier

lateral side chain (as in compound 5p), with respect to the

carboxyethyl one, prevents the correct binding mode from being

established. The molecular dynamics calculation performed with

the binary complexes formed by 5a, 5f, or 5j with CBR demon-

strated the stability of the binding mode and the main inter-

molecular interactions based on a large-time-scale simulation.

The complex between 5f and CBR showed stable energetic and

geometric proﬁles, reinforcing the signiﬁcance of our docking

results and also addressing the relevant biological activity.

dried over Na₂SO₄, filtered, and concentrated in vacuo. The residue was

purified by flash chromatography with a suitable eluent to afford the

expected imidazo ester, which after recrystallization from the suitable

solvent gave an analytical sample.

Ethyl 6-(4-Fluorophenyl)-4H-imidazo[1,5-a][1,4]benzodi-

azepine-3-carboxylate (5a). The title compound was prepared

starting from 8a and following the above-cited general procedure (yield

29%). Colorless crystals were obtained from benzeneꢀcyclohexane, mp

163ꢀ164 ꢀC. HPLC (method A): retention time, 4.7 min; purity,

97.2%. ¹H NMR (CDCl₃): 1.39 (t, 3H, J = 7.4), 4.02 (d, 1H, J = 12.5),

4.40 (m, 2H), 6.00 (d, 1H, J = 12.4), 7.02 (m, 2H), 7.43ꢀ7.55 (m,

4H), 7.59ꢀ7.71 (m, 2H), 7.91 (s, 1H). MS (ESI): m/z 350 (M + H⁺).

Ethyl 6-(4-Chlorophenyl)-4H-imidazo[1,5-a][1,4]benzodi-

azepine-3-carboxylate (5b). The title compound was resynthesized⁴⁰

starting from 8b and following the above-cited procedure (yield 50%).

Colorless crystals were obtained from cyclohexaneꢀchloroform, mp

176ꢀ177 ꢀC. HPLC (method A): retention time, 4.3 min; purity,

96.3%. ¹H NMR (CDCl₃): 1.41 (t, 3H, J = 7.0), 4.07 (d, 1H, J = 12.1 Hz),

4.41 (m, 2H), 6.06 (d, 1H, J = 12.3 Hz), 7.30ꢀ7.73 (m, 8H,) 7.92 (s, 1H).

MS (ESI): m/z 366 (M + H⁺).

Ethyl 6-(4-Methylphenyl)-4H-imidazo[1,5-a][1,4]benzodi-

azepine-3-carboxylate (5c). The title compound was prepared

starting from 8c and following the above-cited procedure (yield 49%).

Colorless crystals were obtained from diethyl ether, mp 141ꢀ142 ꢀC.

1

HPLC (method A): retention time, 4.1 min; purity, 98.4%. H NMR

(CDCl₃): 1.41 (t, 3H, J = 7.0), 2.35 (s, 3H), 4.03 (d, 1H, J = 12.3 Hz),

4.41 (m, 2H), 6.03 (d, 1H, J = 12.6 Hz), 7.15ꢀ7.67 (m, 8H), 7.91 (s,

1H). MS (ESI): m/z 346 (M + H⁺).

Ethyl 6-(4-Methoxyphenyl)-4H-imidazo[1,5-a][1,4]benzodi-

azepine-3-carboxylate (5d). The title compound was prepared

starting from 8d and following the above-cited procedure to obtain a

pale yellow oil (yield 45%). HPLC (method A): retention time, 3.7 min;

purity, 98.2%. ¹H NMR (CDCl₃): 1.44 (t, 3H, J = 7.0 Hz), 3.81 (s, 3H),

4.01 (d, 1H, J = 12.5 Hz), 4.37 (m, 2H), 5.98 (d, 1H, J = 12.4 Hz), 6.85

(d, 2H, J = 8.7 Hz), 7.42ꢀ7.52 (m, 5H), 7.58ꢀ7.65 (m, 1H), 7.87 (s,

1H). MS (ESI): m/z 362 (M + H⁺).

’ EXPERIMENTAL SECTION

Ethyl 4-Methyl-6-phenyl-4H-imidazo[1,5-a][1,4]benzodi-

azepine-3-carboxylate (5e). The title compound was prepared

starting from 8d and following the above-cited procedure (yield 58%).

Colorless crystals were obtained from cyclohexane, mp 168ꢀ169 ꢀC.

Chemistry. All chemicals used were of reagent grade. Yields refer to

purified products and are not optimized. Melting points were deter-

mined in open capillaries on a Gallenkamp apparatus and are un-

corrected. Merck silica gel 60 (230ꢀ400 mesh) was used for column

chromatography. Merck TLC plates and silica gel 60 F₂₅₄were used

for TLC. ¹H NMR spectra were recorded with a Bruker AC 200 spectro-

meter in the indicated solvent (TMS as internal standard). The values of

the chemical shifts are expressed in ppm, and the coupling constants (J) are

expressed in Hz. Mass spectra were recorded on a VG 70-250S (EI, 70 eV),

a Varian Saturn 3, or a ThermoFinningan LCQdeca spectrometer.

The purity of compounds 5aꢀf,iꢀu, 6, and 7 was assessed by RP-

HPLC and was found to be higher than 95%. A VWR_Hitachi L-2130

pump system equipped with a VWR_Hitachi L-2400 and with a Merck

LiChroCART 125-4 C18 column was used in the HPLC analysis with

(method A) acetonitrileꢀwaterꢀmethanol (50:20:30) or (method B)

methanolꢀacetonitrile (20:80) as the mobile phase at a ﬂow rate of

0.7 mL/min. UV detection was achieved at 210 nm.

General Procedure for the Synthesis of Compounds 5aꢀf,

iꢀp. A solution of the suitable 1,4-benzodiazepin-2-one 8aꢀf,iꢀo

(1.2 mmol) and potassium tert-butoxide (2.7 mmol) in dry THF

(30 mL) was stirred at 0 ꢀC for 10 min under argon and then treated

with diethyl chlorophosphate (4.6 mmol). After the mixture was stirred

for 30 min, the suitable isocyanoacetate (6.1 mmol) was added. The

resulting solution was stirred at 0 ꢀC for 1 h and then allowed to stir at

room temperature overnight. Acetic acid (1.8 mL) was added, and the

mixture was stirred for an additional 20 min and then poured onto

crushed ice to give a brownish solid. After filtration, the solid was

dissolved in dichloromethane and the organic layer washed with brine,

1

HPLC (method A): retention time, 5.3 min; purity, 97.2%. The H

NMR spectrum showed the existence of two diasteromeric conforma-

tions; for the sake of simplicity the integral intensities have not been

given. ¹H NMR (CDCl₃): 1.21 (d, J = 7.4), 1.39 (t, J = 7.6), 2.13 (d, J =

6.9), 4.26ꢀ4.44 (m), 6.63 (q, J = 7.2), 7.30ꢀ7.63 (m), 7.89 (s). MS

(EI): m/z 345 (M⁺, 45).

Ethyl 8-Fluoro-6-(4-fluorophenyl)-4H-imidazo[1,5-a]-

[1,4]benzodiazepine-3-carboxylate (5f). The title compound

was prepared starting from 8f and following the above-cited procedure

(yield 28%). Yellow crystals were obtained from benzeneꢀcyclohexane,

mp 197ꢀ198 ꢀC. HPLC (method A): retention time, 5.4 min; purity,

98.2%¹H NMR (CDCl₃): 1.40 (t, 3H, J = 6.9), 4.03 (br d, 1H, J = 12.3),

4.35 (m, 2H), 6.03 (br d, 1H, J = 12.2), 7.00ꢀ7.15 (m, 3H), 7.32ꢀ7.42

(m, 1H), 7.47ꢀ7.60 (m, 3H), 7.87 (s, 1H). MS (EI): m/z 367 (M⁺, 49).

Ethyl 8-Fluoro-6-(4-fluorophenyl)-4-methyl-4H-imidazo-

[1,5-a][1,4]benzodiazepine-3-carboxylate (5i). The title com-

pound was prepared starting from 8i and following the above-cited

procedure (yield 30%). Yellow crystals were obtained from benzeneꢀ

cyclohexane, mp 245ꢀ246 ꢀC. HPLC (method A): retention time,

6.5 min; purity, 97.5%. The ¹H NMR spectrum showed the existence of

two diasteromeric conformations; for the sake of simplicity the integral

1

intensities have not been given. H NMR (CDCl₃): 1.21 (d, J = 7.5),

1.39 (t, J = 7.4), 2.12 (d, J = 6.9), 4.23ꢀ4.44 (m), 6.64 (q, J = 7.2), 7.05

(m), 7.35 (m), 7.54 (m), 7.84 (s). MS (EI): m/z 381 (M⁺, 35).

5704

dx.doi.org/10.1021/jm2001597 |J. Med. Chem. 2011, 54, 5694–5711

Journal of Medicinal Chemistry

ARTICLE

Ethyl 8-Chloro-6-(4-fluorophenyl)-4H-imidazo[1,5-a]-

[1,4]benzodiazepine-3-carboxylate (5j). The title compound

was prepared starting from 8j and following the above-cited procedure

(yield 66%). Colorless crystals were obtained from benzeneꢀcyclohexane,

mp 203ꢀ204 ꢀC. HPLC(methodA): retention time, 5.3 min; purity, 96.2.

¹H NMR (CDCl₃): 1.41 (t, 3H, J = 6.9), 4.03 (d, 1H, J = 12.4), 4.41 (m,

2H), 6.03 (d, 1H, J = 12.2), 7.01ꢀ7.65 (m, 7H), 7.89 (s, 1H). MS (EI):

m/z 383 (M⁺, 64).

Ethyl 8-Chloro-6-(4-chlorophenyl)-4H-imidazo[1,5-a]-

[1,4]benzodiazepine-3-carboxylate (5k). The title compound

was prepared starting from 8k and following the above-cited procedure

(yield 58%). Colorless crystals were obtained from benzeneꢀcyclohexane,

mp 229ꢀ230 ꢀC. HPLC (method A): retention time, 54 min; purity,

97.2%. ¹H NMR (CDCl₃): 1.41 (t, 3H, J = 7.0), 4.03 (d, 1H, J = 12.0), 4.42

(m, 2H), 6.04 (d, 1H, J = 12.3), 7.34ꢀ7.56 (m, 6H), 7.65 (dd, 1H, J = 8.3,

2.1), 7.90 (s, 1H,). MS (ESI): m/z 400 (M + H⁺).

4.44 (m, 2H), 6.05 (d, 1H, J = 12.3), 7.37 (m, 3H), 7.52 (m, 2H), 7.71

(m, 2H), 7.90 (d, 1H, J = 8.6), 8.01 (m, 1H), 8.10 (s, 1H), 8.14 (m,

1H). MS (EI): m/z 415 (M⁺, 10).

General Procedure for the Synthesis of Diazepinone

Derivatives 8aꢀd,f,jꢀo and 10. The synthesis of diazepinone

derivatives 8aꢀd,f,j-o and 10 was carried out starting from the suitable

2-aminobenzophenones following well-known procedures.^37,39,40

General Procedure for the Synthesis of Carboxamides

5qꢀs. A solution of acid 16 (0.49 mmol) in CHCl₃(3.0 mL) and

TEA (0.070 mL, 0.95 mmol) was cooled at 15 ꢀC and treated with of

isobutyl chloroformate (0.51 mmol) while sirring at 15 ꢀC for 1 h. The

reaction mixture was treated with a suitable amine (0.50 mmol) (for 5q,

gaseous NH₃was bubbled into the reaction mixture until an insoluble

residue appeared), stirred at the same temperature for 30 min, poured

into iceꢀwater, and then extracted with CH₂Cl₂. The organic layer was

washed with brine, dried over Na₂SO₄, and evaporated under reduced

pressure. The resulting solid was recrystallized from a suitable solvent to

afford an analytical sample of the expected compound.

8-Chloro-6-(4-chlorophenyl)-4H-imidazo[1,5-a][1,4]benzo-

diazepine-3-carboxamide (5q). The title compound was pre-

pared following the above-cited procedure (yield 83%). White crystals

were obtained from benzeneꢀcyclohexane, mp >300 ꢀC. HPLC (method

A): retention time, 2.6 min; purity, 97.6%. ¹H NMR (CDCl₃): 4.03 (d, 1H,

J = 13.3), 5.32 (br s, 1H), 6.21 (d, 1H, J = 13.8 Hz), 6.93 (br s, 1H),

7.31ꢀ7.39 (m, 3H), 7.43ꢀ7.52 (m, 3H), 7.62 (dd, 1H, J = 8.9, 2.6), 7.80

(s, 1H).

Ethyl 8-Chloro-6-(4-methylphenyl)-4H-imidazo[1,5-a]-

[1,4]benzodiazepine-3-carboxylate (5l). The title compound

was prepared starting from 8l and following the above-cited procedure

(yield 30%). Colorless crystals were obtained from benzene, mp

232ꢀ233 ꢀC. HPLC (method A): retention time, 4.4 min; purity,

97.5%. ¹H NMR (CDCl₃): 1.39 (t, 3H, J = 7.2), 2.36 (s, 3H), 4.02 (d,

1H, J = 12.5), 4.39 (m, 2H), 6.02 (d, 1H, J = 12.4), 7.16 (d, 2H, J = 7.8),

7.36 (d, 2H, J = 7.9), 7.41 (d, 1H, J = 1.7), 7.49 (d, 1H, J = 8.5), 7.61

(dd, 1H, J = 8.8, 1.8), 7.87 (s, 1H). MS (EI): m/z 379 (M⁺, 49).

Ethyl 8-Chloro-6-(4-methoxyphenyl)-4H-imidazo[1,5-a]-

[1,4]benzodiazepine-3-carboxylate (5m). The title compound

was prepared starting from 8m and following the above-cited proce-

dure (yield 44%). Colorless crystals were obtained from diethyl ether,

mp 180ꢀ181 ꢀC. HPLC (method A): retention time, 3.6 min; purity,

99.2%. ¹H NMR (CDCl₃): 1.40 (t, 3H, J = 7.0), 3.81 (s, 3H) 3.99 (d,

1H, J = 12.5), 4.40 (m, 2H), 5.98 (d, 1H, J = 12.5), 6.86 (d, 2H, J = 8.7),

7.41ꢀ7.52 (m, 4H), 7.61 (dd, 1H, J = 8.5, 2.2), 7.87 (s, 1H). MS (EI):

m/z 395 (M⁺, 54)

Ethyl 8-Chloro-6-(3-nitrophenyl)-4H-imidazo[1,5-a][1,4]ben-

zodiazepine-3-carboxylate (5n). The title compound was prepared

starting from 8n and following the above-cited procedure (yield 66%).

Colorless crystals were obtained from benzeneꢀcyclohexane, mp

258ꢀ260 ꢀC. HPLC (method A): retention time, 3.7 min; purity, 96.1%.

¹H NMR (CDCl₃): 1.41 (t, 3H, J = 7.0), 4.10 (d, 1H, J = 12.7), 4.42 (m,

2H), 6.12 (d, 1H, J = 12.8), 7.35 (d, 1H, J = 2.0), 7.55ꢀ7.73 (m, 3H),

7.87ꢀ7.93 (m, 2H), 8.27ꢀ8.37 (m, 2H). MS (EI): m/z 410 (M⁺, 70).

Ethyl 8-Chloro-6-(1-naphthyl)-4H-imidazo[1,5-a][1,4]ben-

zodiazepine-3-carboxylate (5o). The title compound was prepared

starting from 8o and following the above-cited procedure (yield 38%).

Colorless crystals were obtained from benzeneꢀcyclohexane, mp

193ꢀ194 ꢀC. HPLC (method A): retention time, 3.9 min; purity,

97.2%. ¹H NMR (CDCl₃): 1.39 (t, 3H, J = 7.1), 4.20 (br s, 1H), 4.39

(q, 2H, J = 7.2), 6.10 (br s, 1H), 7.18 (s, 1H), 7.32ꢀ7.56 (m, 7H),

7.83ꢀ7.92 (t, 2H, J = 8.8), 8.00 (s, 1H). MS (EI): m/z 415 (M⁺, 80).

tert-Butyl 8-Chloro-6-(4-chlorophenyl)-4H-imidazo[1,5-a]-

[1,4]benzodiazepine-3-carboxylate (5p). The title compound

was prepared starting from 8k and following the above-cited proce-

dure (yield 30%). Colorless crystals were obtained from benzeneꢀ

cyclohexane, mp 230ꢀ231 ꢀC. HPLC (method A): retention time, 6.6

min; purity, 99.6%. ¹H NMR (CDCl₃): 1.61 (s, 9H), 4.00 (d, 1H, J =

12.5), 6.02 (d, 1H, J = 12.8 Hz), 7.32ꢀ7.52 (m, 6H), 7.62 (dd, 1H, J =

8.5, 2.1), 7.86 (s, 1H).

8-Chloro-6-(4-chlorophenyl)-N-methyl-4H-imidazo[1,5-a]-

[1,4]benzodiazepine-3-carboxamide (5r). The title compound

was prepared following the above-cited procedure (yield 60%). Colorless

crystals were obtained from benzeneꢀcyclohexane, mp 267ꢀ268 ꢀC.

1

HPLC (method A): retention time, 3.4 min; purity, 98.8%. H NMR

(CDCl₃,): 2.95 (d, 3H, J= 4.9), 4.03 (d, 1H, J= 11.9), 6.24 (d, 1H, J= 12.0),

7.07 (br s, 1H), 7.30ꢀ7.55 (m, 6H), 7.62 (dd, 1H, J=8.6, 2.6), 7.77(s, 1H).

8-Chloro-6-(4-chloropenyl)-N,N-dimethyl-4H-imidazo-

[1,5-a][1,4]benzodiazepine-3-carboxamide (5s). The title com-

pound was prepared following the above-cited procedure (yield 79%).

Colorless crystals were obtained from benzeneꢀcyclohexane, mp

252ꢀ253 ꢀC. HPLC (method A): retention time, 3.8 min; purity,

99.8%. ¹H NMR (CDCl₃): 3.07 (s, 3H), 3.31 (s, 3H), 4.05 (d, 1H, J =

12.8), 5.85 (d, 1H, J = 12.8), 7.32ꢀ7.37 (m, 3H), 7.42ꢀ7.52 (m, 3H),

7.62 (dd, 1H, J = 8.7, 2.8), 7.82 (s, 1H).

8-Chloro-6-(4-chlorophenyl)-3-(1H-tetrazol-5-yl)-4H-

imidazo[1,5-a][1,4]benzodiazepine (5u). A mixture of 5t

(0.27 mmol) in dry xylene (4.0 mL) containing 0.53 mmol of trimethyltin

azide was heated at 115 ꢀC for 41 h under argon. The precipitate was

collected by filtration, washed with hot toluene, and dried under reduced

pressure. A mixture of the solid in CH₂Cl₂(5.0 mL) and THF (1.0 mL)

was treated with 10 N NaOH (0.02 mL, 0.2 mmol) and stirred at room

temperature for 1 h. Triphenylmethyl chloride (0.114 mmol) was then

added, andthe reaction mixture was stirredatthe sametemperaturefor3h,

then poured into iceꢀwater and extracted with CH₂Cl₂. The organic layer

was washed with brine, dried over Na₂SO₄, and concentrated under

reduced pressure. The resulting residue was dissolved in THF (4.0 mL)

and 10% HCl (2.0 mL), and the resulting mixture was stirred at room

temperature for 4 h, then neutralized with 10% NaOH. The solvent was

evaporated under reduced pressure to afford a solid, which was recrys-

tallized from benzeneꢀcyclohexane to give the analytical sample of 5u as

colorless crystals (mp >300 ꢀC, yield 37%). HPLC (method B): retention

time, 6.6 min; purity, 96.2%. ¹H NMR (DMSO-d₆): 4.30 (d, 1H, J = 12.0),

5.87 (d, 1H, J = 12.0), 7.43 (m, 1H), 7.47 (m, 4H), 7.85ꢀ7.99 (m, 2H),

8.55 (s, 1H). MS (ESI): m/z 395 (M ꢀ H⁺).

Ethyl 6-(4-Chlorophenyl)-4H-imidazo[1,5-a]naphtho[2,1-f]-

[1,4]diazepine-3-carboxylate (6). The title compound was pre-

pared starting from 10 and following the above-cited procedure (yield

48%). Colorless crystals were obtained from ethanolꢀethyl acetate,

mp 277ꢀ278 ꢀC. HPLC (method A): retention time, 5.2 min; purity,

99.0%. ¹H NMR (CDCl₃): 1.44 (t, 3H, J = 7.1), 4.08 (d, 1H, J = 12.5),

8-Chloro-6-(4-chlorophenyl)-4H-imidazo[1,5-a][1,4]ben-

zodiazepine-3-carbonitrile (5t). A mixture of compound 5q

(0.4 mmol) in dry benzene (7.0 mL) was treated while stirring with

5705

dx.doi.org/10.1021/jm2001597 |J. Med. Chem. 2011, 54, 5694–5711

Journal of Medicinal Chemistry

ARTICLE

POCl₃(4 mmol) and refluxed for 7 h. The mixture was then treated with

of CH₂Cl₂(2.0 mL), a saturated solution of NaHCO₃(2.0 mL) and

stirred at room temperature for 12 h. The organic layer was separated

and washed until neutral, dried over Na₂SO₄, and evaporated under

reduced pressure to give a solid, which was recrystallized from benzeneꢀ

cyclohexane to afford 5t colorless crystals (mp 249ꢀ250 ꢀC, yield 47%).

[³H]Flumazenil binding studies were performed as previously

reported.⁵⁸The [³H]ﬂumazenil binding was performed in triplicate by

incubating aliquots of the membrane fractions (0.2ꢀ0.3 mg of protein)

at 0 ꢀC for 90 min in 0.5 mL of 50 mM Tris-citrate buﬀer, pH 7.4, with

approximately 0.2 nM [³H]ﬂumazenil. Nonspeciﬁc binding was deﬁned

in the presence of 10 μM diazepam. After incubation, the samples were

diluted at 0 ꢀC with 5 mL of the assay buﬀer and immediately harvested

onto GF/B ﬁlters (Brandel) by means of a harvester and washed with

ice-cold assay buﬀer. The ﬁlters were washed twice with 5 mL of the

buﬀer, dried, and an amount of 4 mL of Ready Protein Beckman

scintillation cocktail was added; radioactivity was counted in a Packard

LS 1600 liquid-phase scintillation β counter.

1

HPLC (method A): retention time, 3.8 min; purity, 98.4%. H NMR

(CDCl₃): 4.12 (d, 1H, J = 11.8), 5.39 (d, 1H, J = 11.9), 7.33ꢀ7.46

(m, 5H), 7.53 (d, 1H, J = 8.7) 7.66 (dd, 1H, J = 8.2, 2.0), 7.90 (s, 1H). MS

(EI): m/z 352 (M⁺, 100).

Lactone 7. A solution of compound 20 (0.14 mmol) in dry DMF

(7.0 mL) was cooled at 0 ꢀC, treated with NaH (0.20 mmol), and kept at

this temperature for 30 min. The ice bath was removed, and the reaction

mixture was stirred at 120 ꢀC for 5 h, poured into iceꢀwater, and then

extracted with CH₂Cl₂. The organic layer was washed with water and

brine, then dried over Na₂SO₄, and concentrated in vacuo. The resulting

brown oil, when treated with a mixture of diethyl etherꢀn-hexane, gave a

yellowish solid which after recrystallization from ethyl acetate afforded

compound 7 as yellow crystals (yield 43%, mp 258ꢀ259 ꢀC). HPLC

(method A): retention time, 3.4 min; purity, 98.4%. ¹H NMR (CDCl₃):

2.58 (m, 1H), 2.78 (m, 1H), 4.40 (m, 1H), 4.59 (dd, 1H, J = 13.0, 6.4),

4.93 (dd, 1H, J = 12.9, 8.6), 7.31ꢀ7.51 (m, 7H), 7.59ꢀ7.74 (m, 2H),

8.04 (s, 1H). MS (EI): m/z 329 (M⁺, 100).

X-ray Crystallography. A single crystal of 7 was submitted to

X-ray data collection by means of a Siemens P4 four-circle diffractometer

at 293 K equipped with a graphite monochromated Mo KR radiation

(λ = 0.710 69 Å). The structure was solved by direct methods imple-

mented in the SHELXS-97 program.⁵³The refinements were carried out

by full-matrix anisotropic least-squares on F²for all reflections for non-H

atoms by means of the SHELXL-97 program.⁵⁴

Crystallographic data (excluding structure factors) of this crystal

structure have been deposited with the Cambridge Crystallographic

Data Centre assupplementary publication numberCCDC 805673. Copies

of the data can be obtained, free of charge, on application to CCDC, 12

Union Road, Cambridge CB2 1EZ, U.K. [fax, 144-(0)1223-336033;

e-mail, deposit@ccdc.cam.ac.uk].

Radioligand Binding Studies in Native Bovine and Human

Cerebral Receptors. [³H]Flumazenil (specific activity of 70.8

Ci/mmol) was obtained from Perkin-Elmer Life Science (Milano, Italy).

All other chemicals were at reagent grade and were obtained from com-

mercial suppliers.

Compounds were routinely dissolved in DMSO and added to the

assay mixture to a ﬁnal volume of 0.5 mL. Blank experiments were

carried out to determine the eﬀect of the solvent (2%) on binding. At

least six diﬀerent concentrations spanning 3 orders of magnitude,

adjusted approximately for the IC₅₀of each compound, were used.

IC₅₀values, computer-generated by a nonlinear formula on a computer

program (GraphPad, San Diego, CA), were converted to K_ivalues, the

K_dvalues of radioligand in these diﬀerent tissues calculated by the Cheng

and Prusoﬀ equation being known.⁵⁹The K_dof [³H]ﬂumazenil binding

to cortex membrane from bovine and human was 0.85 and 0.91 nM,

respectively. The GABA ratio was determined by calculating (K_iwithout

GABA)/(K_iwith 50 μM GABA) for each compound.

Functional Efficacy Studies (³⁶Cl^ꢀUptake Studies). ³⁶Cl^ꢀ

(specific activity 9.69 μCi/g) was obtained from Perkin-Elmer Life

Science (Milan, Italy). All other chemicals were reagent grade and were

obtained from commercial suppliers.

³⁶Cl^ꢀuptake was measured in rat cerebrocortical synaptoneuro-

somes as described by Schwartz et al.,⁴⁷with minor modiﬁcations.

Brieﬂy, cerebral cortex was dissected from SpragueꢀDawley male rats

suspended 1:10 with ice-cold solution containing 145 mM NaCl, 5 mM

KCl, 5 mM MgC₂, 1 mM CaCl₂, 10 mM HEPES, pH 7 (T1 buﬀer), and

10 mM ^D-glucose; they were homogenized with a glassꢀglass homo-

genizer (ﬁve strokes) and ﬁltered through three layers of nylon mesh

(160 μm) and a 10 μm Millipore ﬁlter. The ﬁltrates were centrifuged at

1000g for 15 min. After the supernatant was discarded, the pellet was

gently resuspended in T1 buﬀer and washed once more by centrifuga-

tion (1000g for 15 min). The ﬁnal pellet containing the synaptoneuro-

somes was suspended 1:2 in T1 buﬀer and was kept on ice until ready for

assay (no longer than 30 min).

Aliquots of synaptoneurosome suspensions (1.5ꢀ2 mg of protein)

were preincubated at 30 ꢀC for 10 min prior to the addition of 0.2 μCi

³⁶Cl^ꢀ. Drugs were added simultaneously with the ³⁶Cl^ꢀ(0.35 mL total

assay volume). ³⁶Cl^ꢀuptake was stopped 10 s later by the addition of

5 mL of ice-cold HEPES, followed by vacuum ﬁltration through glass

ﬁber ﬁlters (Whatman GF/B) that had been soaked with 0.05%

polyethylenimine to reduce nonspeciﬁc binding of ³⁶Cl^ꢀ. The ﬁlters

were washed three more times with 5 mL of ice-cold buﬀer and placed

into scintillation vials containing 4 mL of Ready Protein Beckman

scintillation cocktail, and radioactivity was counted in a Packard LS 1600

liquid-phase scintillation β counter. Data are expressed as percent

stimulation of ³⁶Cl^ꢀuptake above basal level.

Pharmacological Methods. The experiments were carried out in

accordance with the Animal Protection Law of the Republic of Italy, DL

No. 116/1992, based on the European Communities Council Directive

of November 24, 1986 (86/609/EEC). All efforts were made to

minimize animal suffering and to reduce the number of animals involved.

Male CD-1 albino mice (22ꢀ24 g) and male Swiss Webster (20ꢀ26 g)

(Morini, Italy) were used. Twelve mice were housed per cage and fed a

standard laboratory diet, with tap water ad libitum for 12 h/12 h light/

dark cycles (lights on at 7:00). The cages were brought into the

experimental room the day before the experiment for acclimatization

purposes. All experiments were performed between 10:00 and 15:00.

Bovine cortex was obtained from the local slaughterhouse. Human

cortex samples were taken post-mortem at the Department of Patholo-

gical Anatomy, University of Pisa, Italy, during autopsy sessions. The

subjects had died from causes not primarily involving the brain and had

not suﬀered from any psychiatric or neurological disorders. The time

between death and tissue dissection/freezing ranged from 18 to 36 h.

The samples were immediately packed in dry ice and stored in a ꢀ80 ꢀC

freezer. The study was approved by the Ethics Committee of the

University of Pisa, Italy.

Bovine and human cerebral cortex membranes were prepared in

accordance with Martini et al.⁵⁵Brieﬂy, cerebral cortex was homoge-

nized in 10 volumes of ice cold 0.32 M sucrose containing protease

inhibitors. The homogenate was centrifuged at 1000g for 10 min at 4 ꢀC.

The resulting pellet was discarded, and the supernatant was recentri-

fuged at 48000g for 15 min at 4 ꢀC. Then the pellet was osmotically

shocked by suspension in 10 volumes of 50 mM Tris-citrate buﬀer at pH

7.4 containing protease inhibitors and recentrifuged at 48000g for 15

min at 4 ꢀC, The resulting membranes were frozen and washed by means

of a procedure previously described for removing endogenous GABA

from cerebral cortex.⁵⁶Finally, the pellet was suspended in 10 volumes

of 50 mM Tris-citrate buﬀer, pH 7.4, and used in the binding assay.

Protein concentration was assayed by the method of Lowry et al.⁵⁷by

means of bovine serum albumin as the standard.

5706

dx.doi.org/10.1021/jm2001597 |J. Med. Chem. 2011, 54, 5694–5711

Journal of Medicinal Chemistry

ARTICLE

Rotarod Test. The integrity of the animals’ motor coordination was

assessed using a rotarod apparatus (Ugo Basile, Varese, Italy) at a

rotating speed of 16 rpm. The treatment was performed before the test.

The numbers of falls from the rod were counted for 30 s, 30 min after

drug administration, and the test was performed according to the

method described by Vaught et al.⁶⁰

Light/Dark Box Test. The apparatus (50 cm long, 20 cm wide, and

20 cm high) consisted of two equal acrylic compartments, one dark and

one light, illuminated by a 60 W bulb lamp and separated by a divider

with a 10 cm ꢁ 3 cm opening at floor level. Each mouse was tested by

placing it in the center of the lighted area, away from the dark one, and

allowing it to explore the novel environment for 5 min. The number of

transfers from one compartment to the other and the time spent in the

illuminated side were measured. This test exploited the conflict between

the animal’s tendency to explore a new environment and its fear of

bright light.⁶¹

Passive-Avoidance Test. The test was performed according to the

step-through method described by Jarvik et al.⁶²The apparatus consisted

of a two-compartment acrylic box with a lighted compartment connected

to a darkened one by a guillotine door. As soon as the mouse entered the

dark compartment, it received a thermal shock punishment. The latency

times for enteringthe dark compartmentwere measuredin thetraining test

and after 24 h in the retention test. The maximum entry latency allowed in

the training and retention sessions was, respectively, 60 and 180 s.

Hole Board Test. The hole board test consisted of a 40 cm²plane

with 16 flush-mounted cylindrical holes (3 cm diameter) distributed 4 ꢁ

4 in an equidistant, grid-like manner. Mice were placed on the center of

the board one by one and allowed to move about freely for a period of 5

min each. Two electric eyes, crossing the plane from midpoint to

midpoint of opposite sides, thus dividing the plane into four equal

quadrants, automatically signaled the movement of the animal (counts

in 5 min) on the surface of the plane (spontaneous motility). Miniature

photoelectric cells in each of the 16 holes recorded (counts in 5 min) the

exploration of the holes (exploratory activity) by the mice. A total of

12ꢀ15 mice per group were tested.⁶³

with modified Barth solution (MBS) containing (in mM) 88 NaCl,

1 KCl, 2.4 NaHCO₃, 10 HEPES, 0.82 MgSO₄,0.33 Ca(NO₃)₂, 0.91

CaCl₂. GABA was dissolved in MBS and applied for 30s. Diazepam, 5b,

5c, 5f, or 5g was first dissolved in dimethylsulfoxide (DMSO) and then

diluted in MBS. The final DMSO concentration in MBS was 0.1%. This

concentration did not affect the GABA responses. Experiments were

performed by using a concentration of GABA corresponding to the

EC_5ꢀ10(i.e., a concentration that produced peak currents equal to

5ꢀ10% of a maximal GABA concentration). Drugs were preapplied for

60s before being coapplied with GABA for 30 s.

Radioligand binding studies at GABA_Areceptor subtypes expressed

in HEK293 cells were assessed as previously described.^63,66Clonal

mammalian cell lines, expressing relatively high levels of GABA_A

receptor subtypes (R₂β₂γ₂, R₅β₃γ₂), were maintained in Eagle’s mini-

mum essential medium with EBSS, supplemented with 10% fetal calf

serum, ^L-glutamine (2 mM), penicillin (100 U/mL), and streptomycin

(100 μg/mL) in a humidiﬁed atmosphere of 5% CO₂, 95% air at 37 ꢀC,

After removal, the cells were harvested by centrifugation at 500g. The

crude membranes were prepared after homogenization in 10 mM

potassium phosphate, pH 7.4, and diﬀerential centrifugation at 48000g

for 30 min at 4 ꢀC, The pellets were washed twice in this manner before

ﬁnal resuspension in 10 mM potassium phosphate, pH 7.4, that

contained 100 mM potassium chloride and [³H]ﬂumazenil. Binding

assays to transfected cell membranes were carried out as previously

described. Brieﬂy, the cell line membranes were incubated in a volume of

500 μL, which contained [³H]ﬂumazenil at 1ꢀ2 nM and the test

compound in the 10^ꢀ9ꢀ10^ꢀ5M range. Nonspeciﬁc binding was deﬁned

by 10^ꢀ5M diazepam. Assays were incubated to equilibrium for 1 h at

4 ꢀC. The compounds were dissolved in DMSO, the level of which did

not exceed 1% and which was kept constant in all tubes. At least six

diﬀerent concentrations of each compound were used. The data of n = 3

experiments carried out in triplicate were analyzed by means of an

iterative curve-ﬁtting procedure (program Prism, GraphPad, San Diego,

CA), which provided IC₅₀, K_i, and SEM values for tested compounds,

the K_ivalues being calculated from the Cheng and Prusoﬀ equation.⁵⁹

Docking Calculations. All the ligands were docked into the CBR

model by means of Autodock 4.0 program. Docking simulations of the

compounds were carried out by means of the Lamarckian genetic

algorithm⁵¹and applying a protocol with an initial population of 50

randomly placed individuals, a maximum number of 1.0 ꢁ 10⁶energy

evaluations, a mutation rate of 0.02, a crossover rate of 0.80, and an

elitism value of 1. The pseudo Solis and Wets algorithm with a maximum

of 300 interactions was applied for the local search. Then 100 independent

docking runs were carried out for each ligand, and the resulting con-

formations that differ by less than 1.0 Å in positional root-mean-square

deviation (rmsd) were clustered together. The conformer with the lowest

free binding energy was taken as the representative of each cluster.

Ligand Setup. The structures of the ligands were generated by

means of Discovery Studio software, version 1.5. Minimizations energies

were achieved by means of the CHARMm force field⁶⁷as implemented

in Discovery Studio. Minimizations were carried out by means of 50 steps

of steepest descent and 10 000 steps of conjugate gradient as mini-

Drugs. Diazepam (Valium 10, Roche) and pentylenetetrazole (PTZ,

Sigma) were used. All drugs were dissolved in isotonic (NaCl 0.9%)

saline solution and injected sc/ip. The new compound was administered

by the po route and was suspended in 1% carboxymethylcellulose

sodium salt and sonicated immediately before use. Drug concentrations

were prepared in such a way that the necessary dose could be ad-

ministered in a 10 mL/kg volume of carboxymethylcellulose (CMC) 1%

by the po, ip, or sc route.

Statistical Analysis. All experimental results are given as the mean

( SEM. Each value represents the mean of 25 mice. An analysis of

variance, ANOVA, followed by Fisher’s protected least significant

difference procedure for post hoc comparison, was used to verify

significance between two mean values of behavioral results. The data

were analyzed with the StatView software for Macintosh (1992). P

values of less than 0.05 were considered significant.

Radioligand Binding Studies in Recombinant r₁β₂γ₂L,

r₂β₁γ₂L, and r₅β₂γ₂L Human GABA_AReceptors Expressed

in Xenopus laevis Oocytes and in HEK293 Cells. For Xenopus

laevis oocyte experiments, the cDNAs encoding human R₁, R₂, R₅, β₁,

β₂, or γ₂L GABA receptor subunits subcloned in pCDM8 vector were

used for nuclear injections. Isolation of Xenopus laevis oocytes, cDNA

injections, and two-electrode voltage-clamp recordings at a holding

potential of ꢀ70 mV were performed as previously described.⁶⁴The

R₁β₂γ₂L, R₂β₁γ₂L, R₅β₂γ₂L receptor subunit combinations were

injected into the animal poles of the oocytes according to the blind

method described by Colman⁶⁵in a 1:1 ratio (1.5 ng/30 nL). Measure-

ments were performed in oocytes 1ꢀ4 days after injection. Oocytes

expressing R₁β₂γ₂L, R₂β₁γ₂L, R₅β₂γ₂L receptors were placed in a

chamber (approximately 100 μL volume) and perfused (2 mL/min)

ꢁ

mization algorithms, with a rms convergence criterion of 0.01 Å.

Partial atomic charges were assigned by means of the Gasteigerꢀ

Marsili formalism.⁶⁸All the relevant torsion angles were treated as

rotatable during the docking process, thus allowing a search of the

conformational space.

Protein Setup. The GABA_Areceptor model was set up for docking

as follows: polar hydrogens were added by means of Discovery Studio

software, version 1.5, and Kollman united-atom partial charges were

assigned. ADDSOL utility of the AutoDock program was used to add

salvation parameters to the protein structures, and the grid maps

representing the proteins in the docking process were calculated by

means of AutoGrid. The grids, one for each atom type in the ligand plus

5707

dx.doi.org/10.1021/jm2001597 |J. Med. Chem. 2011, 54, 5694–5711

Journal of Medicinal Chemistry

ARTICLE

one for electrostatic interactions, were chosen to be large enough to

include not only the hypothetical benzodiazepine binding site but also a

significant part of the protein around it. As a consequence, for all docking

’ AUTHOR INFORMATION

Corresponding Author

*For M.A.: phone, +39 577 234173; faxs +39 577 234333; e-mail,

anzini@unisi.it. For A.C.: phone, +39 577 234320; e-mail,

cappelli@unisi.it.

ꢁ

calculations, the dimensions of grid map was 46 Å ꢁ 50 Å ꢁ 56 Å with a

ꢁ

grid-point spacing of 0.375 Å.

Energy Refinement of the CBR/Ligand Complexes. The

complexes obtained were subjected to energy minimization by means of

CHARMm force field as implemented in Discovery Studio. Energy

optimizations were carried out by means of 200 steps of steepest descent

followed by 10 000 steps of conjugated gradient with an rms of 0.01 as

gradient value.

Molecular Electrostatic Potential Calculations. For all com-

pounds the electrostatic potential was calculated by means of Gaussian

03⁶⁹and mapped onto the electron density surface for each compound.

The isovalue of 0.0004 electron/bohr³was chosen for the definition of

the density surface, while electrostatic potential was computed using

HartreeꢀFock functional and 6-31G* basis set with a scale of ꢀ0.031

(red) to 0.031 hartree (blue), ꢀ19.45 and 19.45 kcal mol^ꢀ1, respectively.

Molecular Dynamics Methods. The pose representative of the

most populated cluster for each compound (5a, 5f, and 5j) was used as

starting point for further molecular dynamics simulations.

’ ACKNOWLEDGMENT

We are grateful to Dr. Laura Salvini (C.I.A.D.S., University of

Siena, Italy) for the recording of the mass spectra, to Prof.

Stefania D’Agata D’Ottavi for the careful reading of the manu-

script, and to INSTM (Consorzio Interuniversitario Nazionale

per la Scienza e la Tecnologia dei Materiali) for the access to

Accelrys software. We are also grateful to Prof. Cromer for kindly

providing us with the homology model of the GABA_Areceptor.

This work was ﬁnancially supported by CNR (Consiglio Nazionale

delle Ricerche) and MIUR (Ministero dell’Universitꢀa e della

Ricerca), PRIN (Programmi di Ricerca di Rilevante Interesse

Nazionale).

The binary complex between the benzodiazepine receptor and the

ligand has been subjected to a molecular dynamics study, in explicit

solvent and periodic boundary conditions, by means of NAMD2

molecular dynamics simulation code.⁷⁰Protein and ligand were para-

metrized by using Xleap module of AMBER 9⁷¹in the parm99 version⁷²

of the all-atom AMBER force ﬁeld.^73,74The complex was subjected to

the following procedure prior to molecular dynamics data collection: it

was surrounded by a box of 6 Å layer of TIP3P31 pre-equilibrated water

molecules⁷⁵and neutralized by redistributing the net charge of the

system equally on the backbone atoms of the proteins. A two-step

preliminary minimization was carried out by using a sophisticated

conjugate gradient and line search algorithm, ﬁrst 2000 steps minimizing

water, counterions, and hydrogens, then in the following 2000 steps the

loops and all the side chains of the protein setting a convergence

criterion on the gradient of 0.0001 kcal mol^ꢀ1Å^ꢀ1. The microcanonical

NVE ensemble (constant volume and energy) was used to raise the

system temperature to 300 K during 43 ps of Langevin dynamics

simulations with starting 5 kcal/(mol Å²) restraints on protein backbone

atoms and on the ligand heavy atoms, gradually decreased during the

heating. The system was then equilibrated with respect to volume

running for 70 ps until all the restraints were left oﬀ in NPT conditions

(constant pressure and temperature) at 1 atm and 300 K using the

Langevin dynamics to keep the temperature constant, while the pressure

was controlled by using the Langevin piston NoseꢀHoover method.^76,77

Eventually, the production run in the NPT ensemble was carried out.

van der Waals and short-range electrostatic interactions were esti-

mated within a 10 Å cut-oﬀ with the switch value set to 8 Å and the pair

list distance extended to 13.5 Å, whereas the long-range electrostatic

interactions were assessed by using the particle mesh Ewald (PME)

method,⁷⁸with a grid size set to 120, 120, and 120 for x, y and z axes,

respectively, and interpolated by a fourth-order function and by setting

’ DEDICATION

^†In honor of the 80th birthday of R. Ian Fryer, Professor Emeritus

of Medicinal Chemistry at the Department of Chemistry, Rutgers

The State University of New Jersey, Newark, New Jersey 07102,

United States.

’ ABBREVIATIONS USED

CBR, central benzodiazepine receptor; BDZ, benzodiazepine;

GABA, γ-aminobutyric acid; CNS, central nervous system; GR,

γ-aminobutyric acid ratio; CMC, carboxymethylcellulose; DZ,

diazepam; MD, molecular dynamics; SAR, structureꢀactivity

relationship; rmsd, root-mean-square deviation; IBDZ, imidazo-

benzodiazepine

’ REFERENCES

(1) Sieghart, W. Structure and Pharmacology of Gamma-Aminobu-

tyric Acid A Receptor Subtypes. Pharmacol. Rev. 1994, 47, 181–234.

(2) Smith, G. B.; Olsen, R. W. Functional Domains of GABA_A

Receptors. Trends Pharmacol. Sci. 1995, 16, 162–168.

(3) Argyropulos, S. V.; Nutt, D. J. The Use of Benzodiazepines in

Anxiety and Other Disorders. Eur. Neuropsychopharmacol. 1999, 9,

391–392.

(4) Park-Chung, M.; Malayev, A.; Purdy, R. H.; Gibbs, T. T.; Farb,

D. H. Sulfated and Unsulfated Steroids Modulate γ-Aminobutyric Acid

Receptor A Function through Distinct Sites. Brain Res. 1999, 830,

72–87.

(5) Rupprecht, R.; Holsboer, F. Neuroactive Steroids: Mechanisms

of Action and Neuropsychoharmacological Perspectives. Trends Neu-

rosci. 1999, 22, 410–416.

(6) Dorow, R. FG7142 and Its Anxiety Inducing Eﬀects in Humans.

Br. J. Clin. Pharmacol. 1987, 23, 781–782.

(7) Williams, T. J.; Bowie, P. E. Midazolam Sedation To Produce

Complete Amnesia for Bronchoscopy: 2 Years’ Experience at a District

General Hospital. Respir. Med. 1999, 93, 361–365.

(8) Stewart, S. A. The Eﬀects of Benzodiazepines on Cognition.

J. Clin. Psychiatry 2005, 66, 9–13.

(9) Lister, R. G. The Amnestic Action of Benzodiazepines in Man.

Neurosci. Biobehav. Rev. 1985, 9, 87–94.

(10) Ghoneim, M. M.; Mewaldt, S. P. Benzodiazepines and Human

Memory: A Review. Anestesiology 1990, 72, 926–938.

(11) Chambers, M. S.; Atack, J. R.; Broughton, H. B.; Collinson, N.;

Cook, S.; Dawson, G. R.; Hobbs, S. C.; Marshall, G.; Maubach, K. A.;

the direct sum tolerance to 10^ꢀ5

.

All the simulations of the solvated complex were performed with a

time step of 2 fs in combination with the SHAKE⁷⁹algorithm to

constrain bond lengths involving hydrogen atoms.

The VMD program was used for visualization and data analysis.⁸⁰

’ ASSOCIATED CONTENT

S

Supporting Information. Experimental details for the syn-

b

thesis and characterization of intermediates 8e,i,o, 9, 10, 13,

and 15ꢀ20. This material is available free of charge via the

Internet at http://pubs.acs.org.

5708

dx.doi.org/10.1021/jm2001597 |J. Med. Chem. 2011, 54, 5694–5711

Journal of Medicinal Chemistry

ARTICLE

Pillai, G. V.; Reeve, A. J.; MacLeod, A. M. Identiﬁcation of a Novel,

Selective GABA_AR₅Receptor Inverse Agonist Which Enhances Cogni-

tion. J. Med. Chem. 2003, 46, 2227–2240.

(12) Sarter, M.; Bruno, J. P.; Berntson, G. G. Psychotogenic Proper-

ties of Benzodiazepine Receptor Inverse Agonists. Psychopharmacology

2001, 156, 1–13.

(13) Wolkowitz, O. M.; Reus, V. I.; Roberts, E.; Manfredi, F.; Chan,

T.; Ormiston, R.; Johnson, J.; Canick, J.; Brizendine, H.; Weingartner, H.

Antidepressant and Cognition-Enhancing Eﬀects of DHEA in Major

Depression. Ann. N.Y. Acad. Sci. 1995, 774, 337–339.

(14) Bloch, M.; Schmidt, P. J.; Danaceau, M. A.; Adams, L. F.;

Rubinow, D. R. Dehydroepiandrosterone Treatment of Midlife Dysthymia

Biol. Psychiatry 1999, 45, 1533–1541.

(15) Wolf, O. T.; K€oster, B.; Kirschbaum, C.; Pietrowsky, R.; Kern,

W.; Hellhammer, D. H.; Born, J.; Fehm, H. L. A Single Administration of

Dehydroepiandrosterone Does Not Enhance Memory Performance in

Young Healthy Adults, but Immediately Reduces Cortisol Levels. Biol.

Psychiatry 1997, 42, 845–848.

(16) Vallꢁee, M.; Mayo, W.; Le Moal, M. Role of Pregnenolone,

Dehydroepiandrosterone and Their Sulfate Esters on Learning and

Memory in Cognitive Aging. Brain Res. Rev. 2001, 37, 301–312.

(17) Dorow, R.; Horowski, R.; Paschelke, G.; Amin, M.; Braestrup,

C. Severe Anxiety Induced by FG 7142, a β-Carboline Ligand for

Benzodiazepine Receptors. Lancet 1983, 322, 98–99.

(28) Caraiscos, V. B.; Newell, J. G.; You-Ten, K. E.; Elliott, E. M.;

Rosahl, T. W.; Waﬀord, K. A.; MacDonald, J. F.; Orser, B. A. Selective

Enhancement of Tonic GABAergic Inhibition in Murine Hippocampal

Neurons by Low Concentrations of the Volatile Anesthetic Isoﬂurane. J.

Neurosci. 2004, 24, 8454–8458.

(29) Gilman, S.; Newman, S. The Hypothalamus and Limbic

System. In Essentials of Clinical Neuroanatomy and Neurophysiology;

Manter, J., Gatz, A., Eds.; Davis: Philadelphia, PA, 1992; pp 219ꢀ225.

(30) McNaughton, N.; Morris, R. G. M. Chlordiazepoxide, an

Anxiolytic Benzodiazepine, Impairs Place Navigation in Rats. Behav.

Brain Res. 1991, 24, 39–46.

(31) Maubach, K. GABA_AReceptor Subtype Selective Cognition

Enhancers. Curr. Drug Targets: CNS Neurol. Disord. 2003, 2, 233–239.

(32) Carling, R. W.; Moore, K. W.; Street, L. J.; Wild, D.; Isted, C.;

Leeson, P. D.; Thomas, S.; O’Connor, D.; McKernan, R. M.; Quirk, K.;

Cook, S. M.; Atack, J. R.; Waﬀord, K. A.; Thompson, S. A.; Dawson,

G. R.; Ferris, P.; Castro, J. L. 3-Phenyl-6-(2-pyridyl)methyloxy-1,2,4-

triazolo[3,4-a]phthalazines and Analogues: High-Aﬃnity γ-Aminobutyric

Acid-A Benzodiazepine Receptor Ligands with R₂, R₃, and R₅-Subtype

Binding Selectivity Over R₁. J. Med. Chem. 2004, 47, 1807–1822.

(33) Huang, Q.; He, X.; Ma, C.; Liu, R.; Yu, S.; Dayer, C. A.; Wenger,

G. R.; McKernan, R.; Cook, J. M. Pharmacophore/Receptor Models for

GABA_A/BzR Subtypes (R₁β₃γ₂, R₅β₃γ₂, and R₆β₃γ₂) via a Compre-

hensive Ligand-Mapping Approach. J. Med. Chem. 2000, 43, 71–95.

(34) Liu, R.; Hu, R. J.; Zhang, P.; Skolnick, P.; Cook, J. M. Synthesis

and Pharmacological Properties of Novel 8-Substituted Imidazobenzo-

diazepines: High-Aﬃnity, Selective Probes for R5-Containing GABA_A

receptors. J. Med. Chem. 1996, 39, 1928–1934.

(18) Knoﬂach, F.; Rhyner, Th.; Villa, M.; Kellenberger, S.; Drescher,

U.; Malherbe, P.; Sigel, E.; Mohler, H. The γ3-Subunit of the GABA_A-

Receptor Confers Sensitivity to Benzodiazepine Receptor Ligands.

FEBS Lett. 1991, 293, 191–194.

(19) Barnard, E. A.; Skolnick, P.; Olsen, R. W.; Molhler, H.; Sieghart,

W.; Biggio, G.; Braestrup, C.; Bateson, A. N.; Langer, S. Z. International

Union of Pharmacology. XV. Subtypes of γ-Aminobutyric Acid A

Receptors: Classiﬁcation on the Basis of Subunit Structure and Receptor

Function. Pharmacol. Rev. 1998, 50, 291–313.

(20) Ernst, M.; Brauchart, D.; Boresch, S.; Sieghart, W. Comparative

Modeling of GABA-A Receptor: Limits, Insights, Future Developments.

Neuroscience 2003, 119, 933–943.

(21) McKernan, R. M.; Rosahl, T. W.; Reynolds, D. S.; Sur, C.;

Waﬀord, K. A.; Atack, J. R.; Farrar, S.; Myers, J.; Cook, G.; Ferris, P.;

Garrett, L; Bristow, L.; Marshall, G.; Macaulay, A.; Brown, N.; Howell,

O.; Moore, K. W.; Carling, R. W.; Street, L. J.; Castro, J. L.; Ragan, C. I.;

Dawson, G. R.; Whiting, P. J. Sedative but Not Anxiolytic Properties of

Benzodiazepines Are Mediated by the GABA_AReceptor R1 Subtype.

Nat. Neurosci. 2000, 3, 587–592.

(35) Liu, R.; Zhang, P.; McKernan, R. M.; Waﬀord, K.; Cook, J. M.

Synthesis of Novel Imidazobenzodiazepines Selective for the R₅β₂γ₂

(Bz5)/Benzodiazepine Receptor Subtype. Med. Chem. Res. 1995, 5,

700–709.

(36) Harris, D; Clayton, T.; Cook, J.; Sahbaie, P.; Halliwell, R. F.;

Furtm€uller, R.; Huck, S.; Sieghart, W.; DeLorey, T. M. Selective

Inﬂuence on Contextual Memory: Physiochemical Properties Asso-

ciated with Selectivity of Benzodiazepine Ligands at GABA_AReceptors

Containing the R5 Subunit. J. Med. Chem. 2008, 51, 3788–3803.

(37) Anzini, M.; Braile, C.; Valenti, S.; Cappelli, A.; Vomero, S.;

Marinelli, L.; Limongelli, V.; Novellino, E.; Betti, L.; Giannaccini, G.;

Lucacchini, A.; Ghelardini, C.; Galeotti, N.; Makovec, F.; Giorgi, G.;

Fryer, R. I. Ethyl 8-Fluoro-6-(3-nitrophenyl)-4H-imidazo[1,5-a][1,4]benzo-

diazepine-3-carboxylate as Novel, Highly Potent and Safe Antianxiety

Agent. J. Med. Chem. 2008, 51, 4730–4743.

(22) Rudolph, U.; M€ohler, H. Analysis of GABA_AReceptor Func-

tion and Dissection of the Pharmacology of Benzodiazepines and

General Anaesthetics through Mouse Genetics. Annu. Rev. Pharmacol.

Toxicol. 2004, 44, 475–498.

(23) Atack, J. R.; Hutson, P. H.; Collinson, N.; Marshall, G.; Bentley,

G.; Moyes, C.; Cook, S. M.; Collins, I.; Waﬀord, K.; McKernan, R. M.;

Dawson, G. R. Anxiogenic Properties of an Inverse Agonist Selective for

R3 Subunit-containing GABA_AReceptors. Br. J. Pharmacol. 2005,

144, 357–366.

(24) Wisden, W.; Laurie, D. J.; Monyer, H.; Seeburg, P. H. The

Distribution of 13 GABA_AReceptor Subunit mRNAs in the Rat Brain. I.

Telencephalon, Diencephalon, Mesencephalon. J. Neurosci. 1992, 12,

1040–1062.

(38) Fryer, R. I.; Gu, Z. Q.; Wang, C. G. Synthesis of Novel,

Substituted 4H-Imidazo[1,5-a][1,4]benzodiazepine. J. Heterocycl. Chem.

1991, 28, 1661–1669.

(39) Walsh, D. A. The Synthesis of 2-Aminobenzophenones. Synth-

esis 1980, 32, 677–688.

(40) Walser, A.; Fryer, R. I. Bicyclic Diazepines, Chapter VII:

Dihydro-1,4-benzodiazepinones and Thiones. In The Chemistry of

Heterocyclic Compounds; Fryer, R. I., Ed.; John Wiley & Sons Inc.:

New York, 1991; Vol. 50, pp 713ꢀ836 and references therein.

(41) Zhang, W.; Koehler, K. F.; Harris, B.; Skolnick, P.; Cook, J. C.

Synthesis of Benzo-Fused Benzodiazepines Employed as Probes of the

Agonist Pharmacophore of Benzodiazepine Receptors. J. Med. Chem.

1994, 37, 745–757.

(42) Okabe, M.; Chu, S. R.; Tam, S. Y. K.; Todaro, L. J.; Coﬀen, D. L.

Synthesis of the Dideoxynucleosides “ddC” and “CNT” from Glutamic

Acid, Ribonolactone, and Pyrimidine Bases. J. Org. Chem. 1988, 53,

4780–4786.

(43) Braestrup, C.; Nielsen, M.; Honorꢁe, T.; Jensen, L. H.; Petersen,

E. N. Benzodiazepine Receptor Ligands with Positive and Negative

Eﬃcacy. Neuropharmacology 1983, 22, 1451–1457.

(44) Braestrup, C.; Nielsen, M. Benzodiazepine Receptors. In Hand-

book of Psychopharmacology; Iversen, L. L., Iversen, S. D., Snyder, S. H.,

Eds.; Plenum Press: New York, 1983; Vol. 17, pp 285ꢀ384.

(45) Fryer, R. I.; Rios, R.; Zhang, P.; Gu, Z. Q.; Basile, A. S.; Skolnick,

P. StructureꢀActivity Relationships of 2-Pyrazolo[4,3-c]quinoline-3-ones

(25) Thompson, C. L.; Bodewitz, G.; Stephenson, F. A.; Turner,

J. D. Mapping of GABA_AReceptor R5 and R6 Subunit-like Immunor-

eactivity in Rat Brain. Neurosci. Lett. 1992, 144, 53–56.

(26) Br€unig, I.; Scotti, E.; Sidler, C. Intact Sorting, Targeting, and

Clustering of Gamma-Butyric Acid Receptor Subtypes in Hippocampal

Neurons in Vitro. J. Comp. Neurol. 2002, 443, 43–55.

(27) Caraiscos, V. B.; Elliott, E. M.; You-Ten, K. E.; Cheng, V. Y.;

Belelli, D.; Newell, J. G.; Jackson, M. F.; Lambert, J. J.; Rosahl, T. W.;

Waﬀord, K. A.; MacDonald, J. F.; Orser, B. A. Tonic Inhibition in Mouse

Hippocampal CA1 Pyramidal Neurons Is Mediated by R5 Subunit-

Containing γ-Aminobutyric Acid Type A Receptors. Proc. Natl. Acad.

Sci. U.S.A. 2004, 101, 3662–3667.

5709

dx.doi.org/10.1021/jm2001597 |J. Med. Chem. 2011, 54, 5694–5711

Journal of Medicinal Chemistry

ARTICLE

and Their N- and O-Methyl Analogues at Benzodiazepine Receptors.

Med. Chem. Res. 1993, 3, 122–130.

Malmberg-Aiello, P. Benzodiazepine Receptor Ligands. 8: Synthesis and

Pharmacological Evaluation of New Pyrazolo[5,1-c][1,2,4]benzo-

triazine-5-oxide 3- and 8-Disubstituted: High Aﬃnity Ligands Endowed

with Inverse-Agonist Pharmacological Eﬃcacy. Bioorg. Med. Chem.

2006, 14, 758–775.

(64) Mascia, M. P.; Machu, T. K.; Harris, R. A. Enhancement of

Homomeric Glycine Receptor Function by Long-Chain Alcohols and

Anaesthetics. Br. J. Pharmacol. 1996, 119, 1331–1336.

(46) Fryer, R. I. Ligand Interactions at Central Benzodiazepine

Recetor. In Comprehensive Medicinal Chemistry; Hansch, C., Sammes,

P. G., Taylor, J. B., Eds; Pergamon: Oxford, U.K., 1990; Vol. 3, pp

539ꢀ566.

(47) Schwartz, R. D.; Suzdak, P. D.; Paul, S. M. R-Aminobutyric Acid

(GABA)- and Barbiturate-Mediated ³⁶Cl^ꢀUptake in Rat Brain Synap-

toneurosomes: Evidence for Rapid Desensitization of the GABA Re-

ceptor-Coupled Chloride Ion Channel. Mol. Pharmacol. 1986, 30,

419–426.

(65) Colman, A. Expression of Exogenous DNA in Xenopus

Oocytes. In Transcription and Translation: A Practical Approach; Hames,

B. D., Higgins, S. J., Eds.; Oxford Press: Washington, DC, 1984; pp

49ꢀ59.

(48) Im, H. K.; Im, W. B.; Hamilton, B. J.; Carter, D. B.; Von-

Voigtlander, P. F. Potentiation of GABA-Induced Chloride Currents by

Various Benzodiazepine Site Agonists with R₁γ₂, β₂γ₂and R₁β₂γ₂

Subtypes of Cloned GABA_AReceptors. Mol. Pharmacol. 1993, 44,

866–870.

(66) Besnard, F.; Even, Y.; Itier, V.; Granger, P.; Partiseti, M.;

Avenet, P.; Depoortere, H.; Graham, D. Development of Stable Cell

Lines Expressing Diﬀerent Subtypes of GABA_AReceptors. J. Recept.

Signal Transduction Res. 1997, 17, 99–113.

(49) M€ohler, H.; Fritschy, J. M.; Rudolph, U. A New Benzodiazepine

Pharmacology. J. Pharmacol. Exp. Ther. 2002, 300, 2–8.

(50) (a) Cromer, B. A.; Morton, J. C.; Parker, M. W. Anxiety over

GABA_AReceptor Structure Relieved by AChBP. Trends Biochem. Sci.

2002, 27, 280–287. (b) Ernst, M.; Brauchart, D.; Boresch, S.; Sieghart,

W. Comparative Modeling of GABA(A) Receptors: Limits, Insights,

Future Developments. Neuroscience 2003, 119, 933–943.

(51) Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart,

W. E.; Belew, R. K.; Olson, A. J. Automated Docking Using a Lamarckian

Genetic Algorithm and Empirical Binding Free Energy Function. J.

Comput. Chem. 1998, 19, 1639–1662.

(67) Brooks, B. R.; Bruccoleri, R. E.; Olafson, D. B.; States, D. J.;

Swaminathan, S.; Karplus, M. CHARMM: A Program for Macromol-

ecular Energy, Minimization and Dynamics Calculations. J. Comput.

Chem. 1983, 4, 187–217.

(68) Gasteiger, J.; Marsili, M. Iterative Partial Equalization of Orbital

Electronegativity: A Rapid Access to Atomic Charges. Tetrahedron 1980,

36, 3219–3228.

(69) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;

Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.;

Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;

Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson,

G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;

Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.;

Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.;

Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;

Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.;

Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski,

V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick,

D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui,

Q.; Baboul, A. G.; Cliﬀord, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.;

Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith,

T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.;

Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.;

Pople, J. A. Gaussian 03, revision C.02; Gaussian, Inc.: Wallingford, CT,

2004.

(52) (a) Saraogi, I.; Vijay, V. G.; Das, S.; Sekar, K.; Guru Row, T. N.

CꢀHalogen π Interactions in Proteins: a Database Study. Cryst. Eng.

3 3 3

2003, 6, 69–77. (b) Kahnberg, P.; Lager, E.; Rosenberg, C.; Schougaard,

J.; Camet, L.; Sterner, O.; Nielsen, E.; Nielsen, M.; Liljefors, T.

Reﬁnement and Evaluation of a Pharmacophore Model for Flavone

Derivatives Binding to the Benzodiazepine Site of the GABAA Receptor.

J. Med. Chem. 2002, 45, 4188–4201.

(53) Sheldrick, G. M. SHELXS-97: A Program for Automatic Solution

of Crystal Structures, release 97-2; G€ottingen University: G€ottingen,

Germany, 1997.

(54) Sheldrick, G. M. SHELXL-97: A Program for Crystal Structure

Reﬁnement, release 97-2; G€ottingen University: G€ottingen, Germany,

1997.

(55) Martini, C.; Lucacchini, A.; Ronca, G.; Hrelia, S.; Rossi, C. A.

Isolation of Putative Benzodiazepine Receptors from Rat Brain Mem-

branes by Aﬃnity Chromatography. J. Neurochem. 1982, 38, 15–19.

(56) Martini, C.; Rigacci, T.; Lucacchini, A. [³H]Muscimol Binding

Site on Puriﬁed Benzodiazepine Receptor. J. Neurochem. 1983,

41, 1183–1185.

(57) Lowry, O. H.; Rosenbrouh, N. J.; Farr, A. L.; Randall, R. J.

Protein Measurement with the Folin Phenol Reagent. J. Biol. Chem.

1951, 193, 265–275.

(58) Bertelli, L.; Biagi, G.; Giorgi, I.; Manera, C.; Livi, O.; Scartoni,

V.; Betti, L.; Giannaccini, G.; Trincavelli, L.; Barili, P. L. 1,2,3-Triazolo-

[1,5-a]quinoxalines: Synthesis and Binding to Benzodiazepine and

Adenosine Receptors. Eur. J. Med. Chem. 1998, 33, 113–122.

(59) Cheng, Y. C.; Prusoﬀ, W. H. Relationship Between the Inhibi-

tion Constant (K_i) and the Concentration of Inhibition Which Causes

50 Percent Inhibition (IC₅₀) of an Enzyme Reaction. Biochem. Pharma-

col. 1973, 22, 3099–3108.

(60) Vaught, J.; Pelley, K.; Costa, L. G.; Setler, P.; Enna, S. J. A

Comparison of the Antinociceptive Responses to GABA-Receptor

Agonist THIP and Baclofen. Neuropharmacology 1985, 4, 211–216.

(61) Walsh, D. M.; Stratton, S. C.; Harvey, F. J.; Beresford, I. J.;

Hagan, R. M. The Anxiolytic-like Activity of GR159897, a Non-Peptide

NK2 Receptor Antagonist, in Rodent and Primate Models of Anxiety.

Psychopharmacology 1995, 122, 186–191.

(70) Phillips, J. C.; Braun, R.; Wang, W.; Gumbart, J.; Tajkhorshid,

E.; Villa, E.; Chipot, C.; Skeel, R. D.; Kale, L.; Schulten, K. Scalable

Molecular Dynamics with NAMD. J. Comput. Chem. 2005, 26,

1781–1802.

(71) Case, D. A.; Darden, T. A.; Cheatham, T. E., III; Simmerling,

C. L.; Wang, J.; Duke, R. E.; Luo, R.; Merz, K. M.; Pearlman, D. A.;

Crowley, M.; Walker, R. C.; Zhang, W.; Wang, B.; Hayik, S.; Roitberg,

A.; Seabra, G.; Wong, K. F.; Paesani, F.; Wu, X.; Brozell, S.; Tsui, V.;

Gohlke, H.; Yang, L.; Tan, C.; Mongan, J.; Hornak, V.; Cui, G.; Beroza,

P.; Mathews, D. H.; Schafmeister, C.; Ross, W. S.; Kollman, P. A.

AMBER 9; University of California: San Francisco, CA, 2006.

(72) Wang, J.; Cieplak, P.; Kollman, P. A. How Well Does a

Restrained Electrostatic Potential (RESP) Model Perform in Calculat-

ing Conformational Energies of Organic and Biological Molecules?

J. Comput. Chem. 2000, 21, 1049–1074.

(73) Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.; Merz,

K. M.; Ferguson, D. M.; Spellmeyer, D. C.; Fox, T.; Caldwell, J. W.;

Kollman, P. A. A Second Generation Force Field for the Simulation of

Proteins, Nucleic Acids, and Organic Molecules. J. Am. Chem. Soc. 1995,

117, 5179–5197.

(74) Hornak, V.; Abel, R.; Okur, A.; Strockbine, B.; Roitberg, A.;

Simmerling, C. Comparison of Multiple Amber Force Fields and

Development of Improved Protein Backbone Parameters. Proteins

2006, 65, 712–725.

(75) Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey,

R. W.; Klein, L. M. Comparison of Simple Potential Functions for

Simulating Liquid Water. J. Chem. Phys. 1983, 79, 926–935.

(62) Jarvik, M. E.; Kopp, R. An Improved One-Trial Passive

Avoidance Learning Situation. Psychol. Rep. 1967, 21, 221–224.

(63) Guerrini, G.; Costanzo, A.; Ciciani, G.; Bruni., F.; Selleri,

S.; Costagli., C.; Besnard., F.; Costa, B.; Martini, C.; De Siena, G.;

5710

dx.doi.org/10.1021/jm2001597 |J. Med. Chem. 2011, 54, 5694–5711

Journal of Medicinal Chemistry

ARTICLE

(76) Martyna, G. J.; Tobias, D. J.; Klein, M. L. Constant Pressure

Molecular Dynamics Algorithms. J. Chem. Phys. 1994, 101, 4177–4189.

(77) Feller, S. E.; Zhang, Y.; Pastor, R. W.; Brooks, B. R. Constant

Pressure Molecular Dynamics Simulation: The Langevin Piston Method

J. Chem. Phys. 1995, 103, 4613–4621.

(78) Essmann, U.; Perera, L.; Berkowitz, M. L.; Darden, T.; Lee, H.;

Pedersen, L. G. A Smooth Particle Mesh Ewald Method. J. Chem. Phys.

1995, 103, 8577–8593.

(79) Ryckaert, J. P.; Ciccotti, G.; Berendsen, H. J. C. Numerical

Integration of the Cartesian Equations of Motion of a System with

Constraints: Molecular Dynamics of n-Alkanes. J. Comput. Phys. 1977,

23, 327–341.

(80) Humphrey, W.; Dalke, A.; Schulten, K. Visual Molecular

Dynamics. J. Mol. Graphics 1996, 14, 33–38.

5711

dx.doi.org/10.1021/jm2001597 |J. Med. Chem. 2011, 54, 5694–5711

Article Doi

New insight into the central benzodiazepine receptor-ligand interactions: Design, synthesis, biological evaluation, and molecular modeling of 3-substituted 6-phenyl-4 H -imidazo[1,5- a][1,4]benzodiazepines and related compounds

DOI: 10.1021/jm2001597

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jm2001597

Full text of DOI:10.1021/jm2001597

Products guided by the article

R&D Labs maybe for 1318262-06-4

Relevant to this article

Hot Product