From Methaqualone and Beyond: Structure - Activity Relationship of 6-, 7-, and 8-Substituted 2,3-Diphenyl-quinazolin-4(3H)-ones and in Silico Prediction of Putative Binding Modes of Quinazolin-4(3H)-ones as Positive Allosteric Modulators of GABAA</su

Article abstract of DOI:10.1021/acschemneuro.0c00600

Methaqualone (2-methyl-3-(o-tolyl)-quinazolin-4(3H)-one, MTQ) is a moderately potent positive allosteric modulator (PAM) of GABAA receptors (GABAARs). In a previous structure-activity relationship (SAR) study probing the importance of 2- and 3-substituent

Full text of DOI:10.1021/acschemneuro.0c00600

pubs.acs.org/chemneuro
Research Article
From Methaqualone and Beyond: Structure−Activity Relationship of
6‑, 7‑, and 8‑Substituted 2,3-Diphenyl-quinazolin-4(3H)‑ones and in
Silico Prediction of Putative Binding Modes of Quinazolin-
4(3H)‑ones as Positive Allosteric Modulators of GABAA Receptors
Peng-Fei Wang, Anders A. Jensen,* and Lennart Bunch*
Cite This: https://dx.doi.org/10.1021/acschemneuro.0c00600
Read Online
Metrics & More
Article Recommendations
ACCESS
ABSTRACT: Methaqualone (2-methyl-3-(o-tolyl)-quinazolin-
4(3H)-one, MTQ) is a moderately potent positive allosteric
modulator (PAM) of GABAA receptors (GABAARs). In a previous
structure−activity relationship (SAR) study probing the impor-
tance of 2- and 3-substituents in the quinazolin-4(3H)-one
scaffold, several potent GABAAR PAMs were identified, including
2,3-diphenylquinazolin-4(3H)-one (PPQ) and 3-(2-chlorophen-
yl)-2-phenylquinazolin-4(3H)-one (Cl-PPQ). Here, PPQ was
applied as lead in a SAR study of 6-, 7-, and 8-substituents in the quinazolin-4(3H)-one by synthesis and functional characterization
of 36 PPQ analogs at various GABAAR subtypes. While none of the new analogs were significantly more potent than PPQ or
displayed pronounced subtype selectivity across the GABAARs tested, several interesting SAR observations were extracted from the
study. In an in silico study, the putative binding modes of MTQ, PPQ, and Cl-PPQ in the transmembrane β2(+)/α1(−) interface of the
α1β2γ2S GABAAR were predicted. Several plausible binding modes were identified for the three PAMs, and rationalization of the
molecular basis for their different modulatory potencies was attempted.
KEYWORDS: GABAA receptors, GABAA positive allosteric modulators (PAMs), structure−activity relationship (SAR) study,
methaqualone, transmembrane domain, docking
potential has partly been realized through the development of
drugs for sleep disorders, anxiety/depression, and epilepsy.7−13
On the other hand, the adverse effects exhibited by the current
GABAAR therapeutics have also been ascribed to these
multiple native GABAAR subtypes and the fairly broad
modulation of these exerted by the drugs. Thus, medicinal
chemistry efforts in the GABAAR field are often focused on the
identification and development of novel ligands characterized
by distinct modulatory properties or more discrete subtype-
selectivity profiles compared to the existing drugs.7,9,14−16
GABA mediates its signaling through the GABAAR via
binding to two orthosteric sites in the extracellular β(+)/α(−)
interfaces of the pentameric complex, which also contains
multiple allosteric sites targeted by numerous modula-
tors.2,17,18 These sites are located in the ECD (e.g., targeted
by benzodiazepines) as well as in the TMD (e.g., targeted by
INTRODUCTION
■
γ-Aminobutyric acid (GABA) is the major inhibitory neuro-
transmitter in the CNS, where it mediates its fast signaling
through the GABAA receptors (GABAARs).1,2 The GABAAR is
a membrane-embedded complex assembled from five subunits,
and it is composed of an extracellular domain (ECD) made up
by the N-termini of the five subunits, a transmembrane domain
(TMD) consisting of the four transmembrane α-helices
(TM1−4) from each of the subunits, and an intracellular
domain made up by the intracellular loops in the subunits.3,4
A
total of 19 subunits (α1−α6, β1−β3, γ1−γ3, δ, ε, π, θ, ρ1−ρ3)
form the basis for the highly heterogeneous family of different
pentameric GABAAR complexes, which are typically assembled
from two α subunits, two β subunits and one additional
subunit, with αβγ2 assemblies being by far the most
predominant GABAARs, organized in a counterclockwise
(−)β(+) (−)α(+) (−)β(+) (−)α(+) (−)γ(+) arrangement (from the
/ / / /
extracellular perspective).1,2,5,6 At least 30 different GABAAR
subtypes are expressed throughout the brain, and the sheer
abundance and the essential roles governed by them as
regulators of neuronal excitability make them highly attractive
drug targets in a wide range of psychiatric, cognitive, and
neurodegenerative disorders. To date, this huge therapeutic
Received: September 15, 2020
Accepted: October 21, 2020
https://dx.doi.org/10.1021/acschemneuro.0c00600
© XXXX American Chemical Society
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX
A

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
general anesthetics and neurosteroids), with one of the most
explored allosteric sites residing in the β(+)/α(−) interface of
the GABAAR TMD.2,17,18 This interface is targeted by a
plethora of different modulators, including the general
anesthetic etomidate and the anticonvulsant loreclezole,19−31
of which many exhibit pronounced selectivity for β2/β3-
containing over β1-containing GABAAR subtypes. The binding
cavity is formed by TM2 and TM3 from the β subunit and
TM1 and TM2 from the α subunit, and a couple of residues
RESULTS AND DISCUSSION
■
Even though the ortho-chloro PPQ analog Cl-PPQ in our
previous study displayed somewhat higher potency as a
GABAAR PAM than PPQ,34 for synthetic simplicity, we
decided to conduct the SAR study on positions 5−8 using
PPQ as the parental scaffold.
Chemistry. Construction of the quinazolin-4(3H)-one core
followed a previously reported procedure35 whereby the
appropriately substituted 2-amino benzoic acid (1) was first
N-acylated to give the amide (2). Thereafter, this intermediate
was cyclized to form the heterocycle 3, which undergoes a
condensation reaction with the appropriate aniline to form 2,3-
substituted quinazolin-4(3H)-ones (4) (Scheme 1). Generally,
the procedure worked smoothly for substituents in positions 6,
7, and 8, with overall yields ranging from moderate to high.
However, we failed to synthesize 5-chloro-2,3-diphenylquina-
zolin-4(3H)-one, as cyclization of intermediate 2-benzamido-
6-chloro-N-phenylbenzamide did not proceed accordingly.
Other commonly used approaches to prepare 2,3-diphenylqui-
nazolin-4(3H)-ones, for example, solvent-free one-pot reaction
assisted by microwave or changing acetic acid to pyridine,
made no difference to the result.36,37 We believe that the
stereoelectronic impact of the chlorine atom with the carbonyl
group prevents the formation of the planar ring structure 3
under the conditions tried.
have been shown to line the modulator binding site (β2-Met286
,
α1-Met236) or to function as a transducer element between
modulator binding and effect on gating (β2 -
Asn265).20,22−24,27,28,30−32
In two previous studies, we have pursued the potential of
quinazolin-4(3H)-one-based allosteric modulators of GA-
BAARs. Our interest in this scaffold was kindled with the
discovery that methaqualone (2-methyl-3-(o-tolyl)-quinazolin-
4(3H)-one, MTQ), a hypnotic-sedative and infamous recrea-
tional drug from the 1960 and 1970s, is a positive allosteric
modulator (PAM) of GABAARs.33 Based on the structural
similarity of MTQ to etomidate and loreclezole and
delineation of the impact of mutations of the β2-Asn265, β2-
Met286, and α1-Met236 residues on its GABAAR modulation, we
proposed that MTQ acts through the β(+)/α(−) TMD interface
in the receptor.33 In a subsequent elaborate structure−activity
relationship (SAR) study of 67 commercially available MTQ
analogs wherein the 2- and 3-substituents on the quinazolin-
4(3H)-one ring were varied systematically, the introduction of
a 2-aryl ring substituent was found to lead to a significant
increase in modulatory potency. Thus, the analog 2,3-diphenyl-
quinazolin-4(3H)-one (PPQ) displayed 40−120-fold increases
in modulatory potency compared to MTQ across α1,2,3,4,5β2γ2S
GABAARs in the fluorescence-based FLIPR Membrane
Potential Blue (FMP) assay (Table 1), and the ortho-chloro
The two 6- and 8-hydroxy analogs 5a and 5b, respectively,
were obtained by saponification of their corresponding ester
precursors (4l and 4m,Scheme 2).
From the pool of analogs in hand, we identified 8-bromo
analog 4p as an attractive starting point to explore the SAR of
the 8 position in even more detail. Thus, by means of three
palladium coupling procedures, 13 additional 8-substituted
analogs (5c−o) were prepared (Scheme 3). On the synthesis
of 5l, an unexpected condensation reaction of the product with
DMF took place in situ to give product 5m (Scheme 4).38
However, by shortening the reaction time from 18 to 8 h, we
were able to isolate target compound 5l.
To increase the diversity of compounds for pharmacological
characterization even further, 5l was hydrolyzed to give acid
6a. With the unexpected analog 5m in hand, we saw the
opportunity to convert it to heterocycle 6b, and finally 5k was
oxidized to ketone 6c (Scheme 4).
Table 1. Functional Properties Exhibited by MTQ, PPQ,
and Cl-PPQ at Human α1β2γ2S, α3β2γ2S, α4β2γ2S, and α5β2γ2S
GABAARs Transiently Expressed in tsA201 Cells in the
a
FMP Assay
EC50 (μM)
compd
α1β2γ2S
α3β2γ2S
α4β2γ2S
α5β2γ2S
MTQ
PPQ
Cl-PPQ
∼30
0.75
0.079
∼30
0.84
0.069
∼30
0.23
0.045
∼30
0.57
0.063
A total of 36 substituted 2,3-diphenylquinazolin-4(3H)-one
(PPQ) analogs were synthesized.
a
SAR. The functional properties of the 36 PPQ analogs
synthesized were determined at human α1β2γ2S, α3β2γ2S,
α4β2γ2S, and either α5β2γ2S (15 of the 36 analogs) or α6β2γ2S
(21 of the 36 analogs) GABAARs transiently expressed in
tsA201 cells in the FMP assay (see Experimental Section for
details). In the following, the functional properties of the
analogs will be presented according to their respective
substitution patterns on the quinazolin-4(3H)-one scaffold,
with data divided into Tables 2, 3, 4, and 5 (6-substituted, 7-
substituted, 8-substituted and disubstituted analogs, respec-
tively).
For the 6-substituted analogs, a counterintuitive SAR profile
was observed for 4a (6-Cl), 4g (6-Br), 4h (6-OMe), and 4l (6-
NO2), as addition of a halide or nitro group in this position
was found to decrease the modulatory potency of PPQ by >30-
fold at the receptors, whereas analog 4h comprising the larger
methoxy group only exhibited slightly reduced potency
compared to PPQ across the subtypes (Table 2). Interestingly
The modulators were co-applied with GABA EC10 (determined on
the day of the experiment) for the respective receptor subtypes (i.e.,
no preincubation with the modulators). Data are from ref 34.
PPQ analog 2-phenyl-3-(o-chlorophenyl)quinazolin-4(3H)-
one (Cl-PPQ) exhibited 5−12-fold further increased modu-
latory potencies at these subtypes compared to PPQ in this
assay (Table 1, Figure 1).34
In the present study, we extend our SAR explorations of the
quinazolin-4(3H)-one class of GABAAR PAMs using the 2,3-
diaryl analog PPQ as a valuable new parental scaffold in a
semirational and systematic SAR study of substituents in
quinazolin-4(3H)-one positions 6, 7, and 8. Furthermore,
based on an in silico study, we propose putative binding modes
of MTQ, PPQ, and Cl-PPQ to the β(+)/α(−) interface of the
GABAAR TMD and address the molecular basis for their
different modulatory potencies.
B
https://dx.doi.org/10.1021/acschemneuro.0c00600
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
Figure 1. Chemical structures of MTQ, PPQ, Cl-PPQ, and 6,7,8-substituted PPQ analogs 4a−r, 5a−o, and 6a−c presented in this study. The
modulatory potencies exhibited by MTQ, PPQ, and Cl-PPQ at the α1β2γ2S GABAAR transiently expressed in tsA-201 cells in the FMP assay are
given.34
a
Scheme 1. Synthetic Route for Compounds 4a−r
a
Reagents and conditions: (a) BzCl, then TEA, DCM, 0 °C to rt, 5 h (90−96%); (b) Ac2O, reflux, (84−92%); (c) aniline, AcOH, reflux, overnight
(35−86%).
a
modulatory potency was observed for the 4q (8-F), 5b (8-
OH), and 5c (8-Me) analogs at the latter receptor subtypes,
and thus these analogs exhibited a preference (∼10-fold) for
the α4,6β2γ2S receptors. For the 8-alkyl analogs 5g−k, a clear
trend of smaller alkyl substituents being tolerated was
observed, whereas increased bulkiness of the alkyl substituent
decreased modulatory potencies across all receptor subtypes
(5g,i,j and 6a,c vs 5h,k). A larger substituent was not well-
tolerated (4o). The analogs 5d, 5e, 5f, and 6b, which all cotain
heterocycles in the 8-position, presented interestingly diverse
modulatory properties. While the presence of a lipophilic 3-
thienyl group (5b) or a 2-methoxy pyrimidine group (5f) in
this position lead to reduced PAM potency, the hydrogen bond
accepting pyrimidine group in 5e was better accommodated.
Finally, introduction of a 5-amino-oxazole group in this
position (6b) yielded another analog displaying ∼10-fold
preference for α4,6β2γ2S over α1,3β2γ2S subtypes (Table 4). As
for the three disubstituted analogs, 4d and 4e, which contain
chloro in the 6- and 8-positions and in the 7- and 8-positions,
respectively, exhibited significantly reduced PAM potency at all
subtypes, whereas 6,7-dimethoxy analog 4i was a weak PAM in
the lower micromolar range with a preference for the α3,4,5β2γ2S
subtypes over α1,3β2γ2S (Table 5).
Scheme 2. Synthetic Routes for Compounds 5a−o
a
Reagents and conditions: (a) KOH, H2O, EtOH, 60 °C, 5 h (27−
34%).
and in support of the activity of 4h, 5a (6-OH) also retained
low micromolar potencies, which could point to the formation
of a favorable hydrogen bond interaction between the OH/
OMe group and the receptor, whereas larger substituents in
the 6-position were not well-tolerated (4j, 4k, and 4n). The
introduction of a chloro (4b) or a trifluoromethyl group (4f)
in the 7-position of PPQ reduced GABAAR PAM activity
substantially (Table 3).
The 8-substituted analogs offered the most interesting
observations in this SAR study (Table 4). Analogously to the
trend observed for the 6-substituted analogs, the 8-Cl analog
(4c) and in particular the 8-Br analog (4p) were substantially
weaker PAMs than PPQ, whereas the 8-OMe analog (4r) was
equipotent with PPQ. While a general trend of 8-substituted
analogs being slightly more potent PAMs at α4,5,6β2γ2S over
α1,3β2γ2S receptors was observed, a 5−10-fold loss in
In summary, none the 36 PPQ analogs exhibited
significantly increased modulatory potencies compared PPQ,
but analogs 4r, 5a, 5l, 5o, and 6c retained the GABAAR PAM
activity and were equipotent with the parent compound. While
C
https://dx.doi.org/10.1021/acschemneuro.0c00600
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
a
of the cleft formed by the β2(+)/α1(−) interface in the GABAAR
TMD was identified as the putative binding pocket. In detail,
the binding pocket is located between β2(+)-TM2/TM3 and
α1(−)-TM1/TM2 α-helices (Figure 2A), and the residues lining
the pocket are either neutral with hydrogen bonding
capabilities or lipophilic (Figure 2B,C). This prediction fits
well with the findings of our previous mutagenesis studies of
the binding site of MTQ and its analogs,33,34 as the binding
pocket was lined by the β2-TM3-Met286 and α1-TM1-Met236
(Figure 3).
Scheme 3
(−)
Binding Mode of MTQ in the β2(+)/α1 TMD Interface.
(−)
Docking of MTQ into the β2(+)/α1 TMD interface resulted
in five binding modes with comparable calculated scores S
(predicted binding affinities) of −5.9 to −5.8 kcal/mol (Figure
3A,B), which differ in terms of how deep MTQ is inserted into
the crevice between the β2-TM3 and the α1-TM2 helices but
also with respect to the overall orientation of the molecule in
the binding pocket (Figure 3A). The five binding modes can
be divided into two general orientations. In two binding
modes, the fused heterocycle of MTQ is oriented toward the
ion channel (Figure 3A, green and purple). The binding mode
depicted in green is clearly positioned deeper and closer to ion
channel, whereas the binding mode shown in purple is facing
the surrounding lipid bilayer. In the three other binding modes,
the 3-(o-tolyl) and 2-methyl groups of MTQ are oriented
toward the ion channel (Figure 3B, green, purple, and yellow).
These three binding modes also differ with regard to how close
MTQ is positioned relative to the ion channel. It is noted that
the in silico predicted binding modes do not involve direct
hydrogen bonding interactions between the carbonyl group
and the aromatic nitrogen of MTQ and receptor residues. As
specific water molecules are not identified in the cryo-EM
structure, it remains unclear how these will affect binding
mode or favor one binding mode over the others. Also an
induced-fit mechanism may result in changes to the shape of
the binding pocket.
a
Reagents and conditions: (a) RB(OH)2, Pd(PPh3)4, Cs2CO3,
toluene, Ar, 100 °C, overnight (43−78%); (b) RSn, Pd(PPh3)4,
toluene, Ar, reflux, overnight, then KF, rt (31−55%); (c) morpholine,
Pd2(dba)3, Xantphos, Cs2CO3, Ar, 100 °C, 4 h (43−77%); (d) 1,3,5-
trimethoxybenzene, CsF, Pd(PPh3)4, 1,2-dichloroethane, acetyltrime-
thylsilane, Ar, 75 °C, 6 h (27%); (e) Pd2(dba)3, P(t-Bu)3, ZnF2,
DMF, Ar, 90 °C, 8 h (47%); (f) Pd2(dba)3, P(t-Bu)3, ZnF2, DMF, Ar,
90 °C, 18 h (79%).
none of the analogs can be claimed to be subtype-selective
PAMs, as mentioned above a couple of analogs that exhibited
∼10-fold preferences for some of the tested subtypes over
others were identified.
Computational Study. An in silico study was performed
to determine plausible binding modes of the quinazolin-
4(3H)-one-based GABAAR PAMs and thereby strengthen the
understanding of quinazolin-4(3H)-one SAR.34 For this, the
recently published cryo-EM structure of the engineered full-
length α1β2γ2 GABAAR was employed (PDB code 6D6U).39
Although this cryo-EM structure has a slightly lower resolution
(3.9 Å) compared to available X-ray structures (5OJM and
5O8F at 3.3 and 3.2 Å, respectively), it comprises the α1 and β2
subunits, which are essential to the modeling study, and thus it
is the best accessible template for this model. The built-in
binding site identifier in MOE was applied, and the upper part
Binding Mode of PPQ in the β2(+)/α1(−) TMD Interface. We
next docked the more potent analog PPQ into the putative
binding pocket. Analogously to the docking of MTQ, five
binding modes for PPQ were identified, with scoring function
S in the interval of −6.5 to −6.4 kcal/mol. The predicted
binding modes of PPQ can be divided into three major
orientations (Figure 4), none of which comprise specific
hydrogen bonding interactions between PPQ and receptor
a
Scheme 4. Synthetic Routes for Compounds 6a−c
a
Reagents and conditions: (a) KOH, H2O, EtOH, 60 °C, 5 h (17%); (b) NH2OH, EtOH, reflux, 1 h (82%); (c) OsO4, THF/H2O, 20 min, then
NaIO4, 16 h (63%).
D
https://dx.doi.org/10.1021/acschemneuro.0c00600
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
Table 2. Functional Properties Displayed by 6-Substituted PPQ Analogs at Human α1β2γ2S, α3β2γ2S, α4β2γ2S, α5β2γ2S, and
b
α6β2γ2S GABAARs Transiently Expressed in tsA201 Cells in the FMP Assay
a
w.a., weak activity. The concentration−response curve for the modulator was not completed within the concentration range tested. The modulator
b
displayed significant effects at the concentration range indicated in the parentheses. The modulators were co-applied with GABA EC10
(determined on the day of the experiment) for the respective receptor subtypes (i.e., no preincubation with the modulators). The data are based on
3−4 independent experiments (n = 3−4). nt, not tested.
Table 3. Functional Properties Displayed by 7-Substituted PPQ Analogs at Human α1β2γ2S, α3β2γ2S, α4β2γ2S, and α5β2γ2S
b
GABAARs Transiently Expressed in tsA201 Cells in the FMP Assay
EC50 (μM) [pEC50 SEM]
compd
R2
α1β2γ2S
α3β2γ2S
α4β2γ2S
α5β2γ2S
α6β2γ2S
a
a
a
4b
4f
-Cl
-CF3
w.a. (6.25−100)
w.a. (6.25−100)
w.a. (6.25−100)
∼10 [∼5.0]
w.a. (25−100)
nt
nt
a
a
a
a
w.a. (25−100)
w.a. (25−100)
w.a. (25−100)
a
w.a., weak activity. The concentration−response curve for the modulator was not completed within the concentration range tested. The modulator
b
displayed significant effects at the concentration range indicated in the parentheses. The modulators were co-applied with GABA EC10
(determined on the day of the experiment) for the respective receptor subtypes (i.e., no preincubation with the modulators). The data are based on
3 independent experiments (n = 3). nt, not tested.
E
https://dx.doi.org/10.1021/acschemneuro.0c00600
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
Table 4. Functional Properties Displayed by 8-Substituted PPQ Analogs at Human α1β2γ2S, α3β2γ2S, α4β2γ2S, α5β2γ2S, and
b
α6β2γ2S GABAARs Transiently Expressed in tsA201 Cells in the FMP Assay
F
https://dx.doi.org/10.1021/acschemneuro.0c00600
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
Table 4. continued
a
w.a., weak activity. The concentration−response curve for the modulator was not completed within the concentration range tested. The modulator
b
displayed significant effects at the concentration range indicated in the parentheses. The modulators were co-applied with GABA EC10
(determined on the day of experiment) for the respective receptor subtypes (i.e., no preincubation with the modulators). The data are based on 3−
4 independent experiments (n = 3−4). nt, not tested.
Table 5. Functional Properties Displayed by Disubstituted PPQ Analogs at Human α1β2γ2S, α3β2γ2S, α4β2γ2S, and α5β2γ2S
b
GABAARs Transiently Expressed in tsA201 Cells in the FMP Assay
EC50 (μM) [pEC50 SEM]
compd
R1
-Cl
-H
-OMe
R2
-H
-Cl
-OMe
R3
α1β2γ2S
α3β2γ2S
α4β2γ2S
α5β2γ2S
α6β2γ2S
a
a
a
a
4d
4e
4i
-Cl
-Cl
-H
w.a. (100)
w.a. (25−100)
w.a. (6.25−100)
w.a. (25−100)
w.a. (6.25−100)
2.9 [5.53 0.04]
w.a. (100)
w.a. (25−100)
6.1 [5.22 0.07]
w.a. (100)
w.a. (25−100)
4.3 [5.37 0.11]
nt
nt
nt
a
a
a
a
a
a
w.a., weak activity. The concentration−response curve for the modulator was not completed within the concentration range tested. The modulator
b
displayed significant effects at the concentration range indicated in the parentheses. The modulators were co-applied with GABA EC10
(determined on the day of the experiment) for the respective receptor subtypes (i.e., no preincubation with the modulators). The data are based on
3−4 independent experiments (n = 3−4). nt, not tested.
Figure 2. Putative binding pocket for the quinazolin-4(3H)-ones in the β2(+)/α1(−) interface of the α1β2γ2 GABAAR TMD (PDB code 6D6U). (A)
(−)
Top view schematic drawing of the α1β2γ2 GABAAR TMD, with the four α-helices in a β2 and an α1 subunit indicated. The β2(+)/α1(−) TMD
interface with the three identified binding modes of MTQ in the computational study is indicated. (B) 3-D structure of the β2(+)/α1 TMD
interface. (C) Schematic drawing of the MTQ binding pocket in the β2(+)/α1(−) TMD interface with the residues lining the pocket indicated. PDB
file, Chain A β2(+); PDB file, Chain B α1(−). Green, apolar side chains; purple, polar side chains (red circle, acidic side chain; blue circle, basic side
chain).
G
https://dx.doi.org/10.1021/acschemneuro.0c00600
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
Figure 3. MTQ binding modes in the β2(+)/α1(−) TMD interface (blue and red helices, respectively). (A) Binding mode no. 1 of MTQ, projecting
the fused heterocycle toward the ion channel. (B) Binding orientation no. 2 of MTQ, projecting the 3-(o-tolyl) and 2-methyl substituents toward
the ion channel.
thus cannot be determined in detail from the present in silico
study.
To validate the cogency of the in silico predicted binding
(−)
modes for MTQ and PPQ in the wild-type β2(+)/α1 TMD
interface, we next performed docking experiments with PPQ in
the binding pocket comprising a β2-Met286Trp mutation. Based
on photolabeling studies, the β2-Met286 residue has previously
been proposed to contribute to or line the etomidate binding
site in the β(+)/α(−) TMD interface of GABAARs,19,31 and
substitution of this Met286 for a Trp residue has been found to
be detrimental to etomidate-mediated PAM activity and to also
reduce the modulatory efficacies of MTQ and 2-phenyl-3-(p-
tolyl)quinazolin-4(3H)-one at GABAARs.23,33,34 The docking
(−)
of PPQ into β2(+)Met286Trp/α1 yielded seemingly reasonable
modulator binding modes in the interface, and the binding
(−)
(+)Met286Trp
modes of PPQ in wild-type β2(+)/α1 and in β2
/
(−)
α1 are superimposed in Figure 5. It can be clearly seen that
in the β2(+)Met286Trp/α1 TMD interface, PPQ is prevented
(−)
from binding deeper into the binding pocket, and thus it does
not reach the β2-TM2/α1-TM2 helices. While this deeper
intrusion into the binding pocket potentially could be
important for the PAM activity exhibited by this modulator
class, it cannot be ruled out that the β2-Met286Trp mutation
simply blocks the entrance of the modulator into the binding
pocket.
(+)
Figure 4. Superimposition of binding modes of PPQ in the β2
/
(−)
α1 TMD interface. In PPQ, binding modes given in green (two
modes, flipped) and yellow face the heterocyclic core faces toward the
ion channel. The PPQ binding mode given in purple projects the two
phenyl rings toward the ion channel. The PPQ binding mode given in
orange is perpendicular in orientation. β2 and α1 helices are given in
blue and in red, respectively.
(−)
Binding Mode of Cl-PPQ in the β2(+)/α1 TMD Interface.
Docking of Cl-PPQ, a 5−12-fold more potent PAM than the
(−)
PPQ analog at GABAARs, into the β2(+)/α1 TMD interface
also resulted in five binding modes, which are superimposed on
the five binding modes of PPQ (Figure 6). This analysis
revealed that the two modulators occupy the pocket in roughly
the same manner, and thus the higher modulatory potency of
Cl-PPQ is again likely due to increased steric bulk and
lipophilicity of the ligand, rather than specific interactions with
receptor residues or a more optimal binding mode.
residues. The 40−120-fold higher modulatory potencies of
PPQ compared to MTQ, are not fully reflected in the
calculated affinities of MTQ and PPQ, although the trend is
confirmed (lower S value for PPQ). The substantially higher
cLogP value of PPQ (4.75) compared to MTQ (3.65) can at
least in part explain the higher potency of PPQ, but the process
of mutually induced fit between ligand and host is complex and
H
https://dx.doi.org/10.1021/acschemneuro.0c00600
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
introduction of even small 7-substituents was detrimental to
PAM activity and only relatively small substituents could be
accommodated in the 6-position of the quinazolin-4(3H)-one,
the SAR for the 8-position was considerably more diverse.
Notably, PPQ potency was retained in analogs comprising
unbranched aliphatic substituents or selected heterocycles in
the 8-position (5l,o and 6b,c), and thus it may be possible to
build further out from these substituents into regions in the
β
(+)/α(−) TMD interface not probed by the analogs in this
study. Finally, after the elaborate delineation of the SAR of
these GABAAR PAMs performed in this and our previous
study,34 the 5-position remains the only unexplored position in
the quinazolin-4(3H)-one scaffold, and it would be interesting
to probe the impact of substituents in this position in future
quinazolin-4(3H)-one analogs provided that a feasible
synthetic route could be established.
(+)
The in silico study of MTQ and PPQ binding to the β2
/
(−)
α1 TMD interface revealed a handful of putative binding
modes for both PAMs. None of these could be excluded based
on available mutagenesis data. The multiple putative binding
modes are perhaps not surprising considering the lipophilic
nature of the modulators and the residues lining this interface,
but it obviously constitutes a challenge in terms of future
structure-based modulator design for this binding site.
Figure 5. Superimposition of computationally determined plausible
(−)
binding modes of PPQ in the wild-type β2(+)/α1 (black) and the
(−)
β2(+)Met286Trp/α1
(purple) TMD interfaces. The β2-Met286Trp
mutation prevents PPQ from binding deeper in the pocket, thus
not reaching the β2TM2/α1TM2 helices.
EXPERIMENTAL SECTION
■
Chemistry. All reactions involving dry solvents or sensitive agents
were performed under a nitrogen atmosphere, and glassware was
dried prior to use. Solvents were dried according to standard
procedures, and reactions were monitored by analytical thin-layer
chromatography (TLC, Merck silica gel 60 F254 aluminum sheets).
Flash chromatography was carried out using Merck silica gel 60A
1
(35−70 μm). H and 13C NMR spectra were recorded on a Bruker
400 MHz spectrometer (1H NMR, 400 MHz; 13C NMR, 100 MHz)
or a Bruker 600 MHz spectrometer (1H NMR, 600 MHz; 13C NMR,
150 MHz). NMR data were obtained in CDCl3 or DMSO-d6
(purchased from Cambridge Isotope Laboratories, Inc.). MS spectra
were recorded using a Waters Acquity UPLC-MS instrument with
dual-wavelength detection with electrospray ionization. Gradients of
5% aqueous MeCN + 0.1% HCO2H (solvent A) and 95% aqueous
MeCN + 0.05% HCO2H (solvent B) were employed. Analytical
HPLC was performed using a Dionex UltiMate 3000 pump and
photodiode array detector (200 and 210 nm, respectively) installed
with an XTerra MS C18 3.5 μm, 4.6 mm × 150 mm column, using a 5
→ 95% MeCN gradient in H2O containing 0.1% TFA. Optical
rotation was measured using a PerkinElmer 241 polarimeter with Na
lamp (589 nm). All commercial chemicals were used without further
purification. The purity of all tested compounds was determined by
HPLC to be >95%. The chemical structures of all final compounds
Figure 6. Superimposition of the in silico predicted five binding
modes of Cl-PPQ (yellow) and the five binding modes of PPQ
(purple).
1
synthesized were confirmed by H NMR, 13C NMR, and UPLC-MS.
Compound Synthesis. General Procedure for the Synthesis of
2,3-Diphenylquinazolin-4-ones 4a−r. To a solution of anthranilic
acid 1a−r (0.4 mmol, 1 equiv) dissolved in dry DCM (15 mL) was
added benzoyl chloride (0.42 mmol, 1.05 equiv), followed by the
dropwise addition of TEA (0.6 mmol, 1.5 equiv) at 0 °C. The
reaction mixture was warmed up to rt and stirred for 5 h before
quenching with 1 × HCl solution (1 mL). The resulting precipitate
was filtered, washed with heptane (3 × 10 mL), and dried under
vacuum to afford the corresponding compounds 2a−r (90−96%).
2-Benzoylamino-benzoic acid 2a−r (0.36 mmol, 1 equiv) was
dissolved in acetic anhydride (5.4 mmol, 15 equiv), and the solution
was refluxed until starting material was consumed (TLC monitoring).
Excess solvent was removed under vacuum, and heptane (20 mL) was
added. The mixture was stirred vigorously for 30 min, filtered, washed
with heptane (3 × 10 mL), and dried under vacuum to afford the
compounds 3a−r (84−92%).
CONCLUSION
■
In our previous search for novel GABAAR PAMs based on the
quinazolin-4(3H)-one scaffold, we exclusively focused on 2-
and 3-substituents and identified a 2-aryl group and ortho-
substituents on this ring as key contributors to high
modulatory potency.34 Although the investigations into the
functional implications of 6-, 7-, and 8-substitutions to the
quinazolin-4(3H)-one scaffold in the present work did not
yield analogs with substantially improved modulatory
potencies than that of the lead PPQ, some interesting SAR
observations could be extracted from the study. While the
I
https://dx.doi.org/10.1021/acschemneuro.0c00600
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
2-Phenyl-benzoxazin-4-one 3a−r (0.3 mmol, 1 equiv) was
dissolved in 10 mL of acetic acid, and aniline (0.315 mmol, 1.05
equiv) was added. The mixture was refluxed overnight before the
solvent was removed under vacuum. The resulting solid was
partitioned between CH2Cl2 (20 mL) and water (20 mL), and then
the aqueous layer was extracted with CH2Cl2 (3 × 20 mL). The
combined organic layers were washed with brine (20 mL), dried over
Na2SO4, filtered, and concentrated under vacuum. The crude residue
was purified by flash chromatography on silica gel (heptane−EtOAc
6:1, isocratic) to afford the title compounds 4a−r (35−86%).
6-Chloro-2,3-diphenylquinazolin-4(3H)-one (4a). White solid,
127.97, 114.27, 108.24, 106.10, 56.38 ppm. MS (ESI+) m/z: 359.2
[M + H]+.
2,3-Diphenyl-6-(p-tolyl)quinazolin-4(3H)-one (4j). White solid,
yield 52%. 1H NMR (400 MHz, chloroform-d) δ 8.49 (d, J = 2.2 Hz,
1H), 7.99 (dd, J = 8.4, 2.1 Hz, 1H), 7.81 (d, J = 8.5 Hz, 1H), 7.55 (d,
J = 7.8 Hz, 2H), 7.31−7.09 (m, 12H), 2.35 (s, 3H) ppm. 13C NMR
(101 MHz, chloroform-d) δ 146.51, 140.18, 137.86, 137.76, 136.77,
135.51, 133.50, 129.76, 129.29, 129.15, 129.03, 128.99, 128.42,
128.23, 128.00, 127.06, 124.72, 121.21, 21.15 ppm. MS (ESI+) m/z:
389.2 [M + H]+.
2,3-Diphenyl-6-(4-(trifluoromethyl)phenyl)quinazolin-4(3H)-one
1
1
(4k). White solid, yield 53%. H NMR (400 MHz, chloroform-d) δ
yield 24%. H NMR (400 MHz, DMSO-d6) δ 8.15 (dd, J = 5.9, 2.7
8.53 (dd, J = 7.9, 4.7 Hz, 1H), 8.11−7.65 (m, 6H), 7.47−7.04 (m,
10H) ppm. 13C NMR (101 MHz, chloroform-d) δ 133.53, 129.46,
129.09, 129.06, 129.01, 128.62, 128.56, 128.05, 127.52, 125.55,
121.36 ppm. MS (ESI+) m/z: 443.1 [M + H]+.
6-Nitro-2,3-diphenylquinazolin-4(3H)-one (4l). Yellow solid, yield
30%. 1H NMR (400 MHz, chloroform-d) δ 9.13 (s, 1H), 8.52 (dd, J =
8.9, 2.7 Hz, 1H), 7.86 (d, J = 8.9 Hz, 1H), 7.30−7.08 (m, 10H) ppm.
13C NMR (101 MHz, chloroform-d) δ 161.20, 158.41, 151.53,
145.98, 136.95, 134.67, 130.09, 129.36, 129.25, 129.02, 128.96,
128.87, 128.78, 128.17, 123.92, 121.15 ppm. MS (ESI+) m/z: 344.1
[M + H]+.
Hz, 1H), 7.95 (dt, J = 7.8, 3.9 Hz, 1H), 7.81 (dd, J = 9.1, 5.7 Hz, 1H),
7.70−7.07 (m, 10H) ppm. 13C NMR (101 MHz, DMSO-d6) δ
155.61, 146.38, 137.92, 137.41, 129.72, 129.60, 129.51, 129.08,
129.01, 128.95, 128.61, 128.04, 120.81 ppm. MS (ESI+) m/z: 333.1
[M + H]+.
7-Chloro-2,3-diphenylquinazolin-4(3H)-one (4b). White solid,
1
yield 69%. H NMR (400 MHz, chloroform-d) δ 8.21 (dd, J = 8.5,
2.4 Hz, 1H), 7.75 (d, J = 2.1 Hz, 1H), 7.41 (dd, J = 8.6, 2.1 Hz, 1H),
7.35−7.03 (m, 10H) ppm. 13C NMR (101 MHz, chloroform-d) δ
161.70, 156.46, 148.48, 140.95, 137.42, 135.14, 129.56, 129.06,
129.04, 128.98, 128.71, 128.59, 128.04, 127.87, 127.34, 119.45 ppm.
MS (ESI+) m/z: 333.1 [M + H]+.
8-Nitro-2,3-diphenylquinazolin-4(3H)-one (4m). Yellow solid,
1
yield 29%. H NMR (400 MHz, chloroform-d) δ 8.59 (d, J = 2.2
8-Chloro-2,3-diphenylquinazolin-4(3H)-one (4c). White solid,
Hz, 1H), 8.44 (d, J = 8.7 Hz, 1H), 8.21 (dd, J = 8.7, 2.2 Hz, 1H),
7.29−7.16 (m, 8H), 7.13−7.07 (m, 2H) ppm. 13C NMR (101 MHz,
chloroform-d) δ 161.17, 157.37, 151.91, 148.09, 137.06, 134.66,
129.94, 129.22, 129.19, 129.01, 128.90, 128.86, 128.14, 125.10,
123.39, 120.83 ppm. MS (ESI+) m/z: 344.1 [M + H]+.
1
yield 43%. H NMR (400 MHz, chloroform-d) δ 8.19 (dd, J = 8.0,
1.5 Hz, 1H), 7.81 (dd, J = 7.7, 1.5 Hz, 1H), 7.42−7.04 (m, 11H)
ppm. 13C NMR (101 MHz, chloroform-d) δ 161.90, 155.62, 144.33,
137.57, 135.14, 134.96, 132.30, 129.64, 129.50, 129.05, 129.02,
128.58, 127.95, 127.20, 126.03, 122.58 ppm. MS (ESI+) m/z: 333.1
[M + H]+.
6,8-Dichloro-2,3-diphenylquinazolin-4(3H)-one (4d). White solid,
yield 55%. 1H NMR (400 MHz, chloroform-d) δ 8.16 (d, J = 2.4 Hz,
1H), 7.80 (d, J = 2.4 Hz, 1H), 7.33−7.20 (m, 6H), 7.14 (dd, J = 8.2,
6.6 Hz, 2H), 7.09−7.04 (m, 2H) ppm. 13C NMR (101 MHz,
chloroform-d) δ 134.96, 132.53, 129.82, 129.46, 129.14, 128.90,
128.76, 128.00, 125.44, 123.10 ppm. MS (ESI+) m/z: 367.0 [M +
H]+.
4-Oxo-2,3-diphenyl-3,4-dihydroquinazolin-6-yl Benzoate (4n).
1
White solid, yield 68%. H NMR (400 MHz, chloroform-d) δ 8.21
(d, J = 7.7 Hz, 2H), 7.62−7.53 (m, 2H), 7.46 (dt, J = 19.3, 7.8 Hz,
3H), 7.28−7.15 (m, 6H), 7.13−7.04 (m, 3H), 6.99 (t, J = 7.6 Hz,
2H) ppm. 13C NMR (101 MHz, chloroform-d) δ 165.30, 161.96,
154.87, 146.95, 137.72, 135.16, 133.54, 130.40, 129.57, 129.47,
129.35, 129.08, 128.98, 128.52, 128.47, 127.57, 127.37, 127.05,
125.02 ppm. MS (ESI+) m/z: 419.2 [M + H]+.
4-Oxo-2,3-diphenyl-3,4-dihydroquinazolin-8-yl Benzoate (4o).
1
7,8-Dichloro-2,3-diphenylquinazolin-4(3H)-one (4e). White solid,
yield 61%. 1H NMR (400 MHz, chloroform-d) δ 8.12 (dt, J = 8.4, 2.2
Hz, 1H), 7.69−7.04 (m, 11H) ppm. 13C NMR (101 MHz,
chloroform-d) δ 129.86, 129.50, 129.11, 128.93, 128.72, 128.38,
128.00, 125.94 ppm. MS (ESI+) m/z: 367.0 [M + H]+.
2,3-Diphenyl-7-(trifluoromethyl)quinazolin-4(3H)-one (4f).
White solid, yield 59%. H NMR (400 MHz, DMSO-d6) δ 8.17 (d,
J = 7.8 Hz, 3H), 7.88 (d, J = 7.8 Hz, 1H), 7.75 (t, J = 7.4 Hz, 1H),
7.68 (t, J = 7.9 Hz, 1H), 7.61 (t, J = 7.7 Hz, 2H), 7.38−7.08 (m, 10H)
ppm. 13C NMR (101 MHz, DMSO-d6) δ 130.31, 129.92, 129.49,
129.44, 129.09, 128.77, 127.93, 127.71, 124.79 ppm. MS (ESI+) m/z:
419.2 [M + H]+.
1
White solid, yield 42%. H NMR (400 MHz, chloroform-d) δ 8.40
8-Bromo-2,3-diphenylquinazolin-4(3H)-one (4p). White solid,
1
(d, J = 8.3 Hz, 1H), 8.05 (d, J = 1.7 Hz, 1H), 7.66 (dd, J = 8.3, 1.7 Hz,
1H), 7.35−7.04 (m, 10H) ppm. 13C NMR (101 MHz, chloroform-d)
δ 161.54, 156.58, 147.52, 137.28, 134.96, 129.70, 129.14, 128.98,
128.96, 128.72, 128.41, 128.08, 125.38, 125.34, 123.31, 123.22,
123.18 ppm. MS (ESI+) m/z: 367.1 [M + H]+.
yield 76%. H NMR (400 MHz, DMSO-d6) δ 8.22 (m, 2H), 7.51
(t, J = 7.8 Hz, 1H), 7.44−7.39 (m, 2H), 7.39−7.20 (m, 8H)) ppm.
13C NMR (101 MHz, DMSO-d6) δ 161.45, 138.56, 138.12, 135.92,
129.87, 129.57, 129.48, 129.06, 128.79, 128.42, 127.99, 126.81,
122.99, 122.55 ppm. MS (ESI+) m/z: 368.1 [M + H]+.
6-Bromo-2,3-diphenylquinazolin-4(3H)-one (4g). White solid,
8-Fluoro-2,3-diphenylquinazolin-4(3H)-one (4q). White solid,
1
1
yield 45%. H NMR (400 MHz, chloroform-d) δ 8.41 (d, J = 2.3
yield 61%. H NMR (400 MHz, DMSO-d6) δ 8.03 (dd, J = 8.0, 1.3
Hz, 1H), 7.82 (dd, J = 8.7, 2.3 Hz, 1H), 7.62 (d, J = 8.6 Hz, 1H),
7.28−7.12 (m, 8H), 7.11−7.04 (m, 2H) ppm. 13C NMR (101 MHz,
chloroform-d) δ 137.92, 129.72, 129.60, 129.51, 129.08, 129.01,
128.95, 128.61, 128.04, 120.81 ppm. MS (ESI+) m/z: 368.1 [M +
H]+.
Hz, 1H), 7.80 (ddd, J = 10.7, 8.1, 1.4 Hz, 1H), 7.60 (td, J = 8.0, 4.8
Hz, 1H), 7.42−7.20 (m, 10H) ppm. 13C NMR (101 MHz, DMSO-
d6) δ 156.34, 138.10, 135.92, 129.89, 129.52, 129.39, 129.06, 128.77,
127.98, 127.89, 123.35, 122.71, 120.70, 120.51 ppm. MS (ESI+) m/z:
317.1 [M + H]+.
6-Methoxy-2,3-diphenylquinazolin-4(3H)-one (4h). White solid,
yield 55%. 1H NMR (400 MHz, chloroform-d) δ 7.72−7.63 (m, 2H),
7.34 (dd, J = 8.9, 2.9 Hz, 1H), 7.27−7.06 (m, 10H), 3.87 (s, 3H)
ppm. 13C NMR (101 MHz, chloroform-d) δ 158.83, 153.04, 142.14,
137.85, 135.53, 129.38, 129.13, 129.10, 129.03, 128.94, 128.36,
127.96, 124.87, 121.78, 106.64, 55.89 ppm. MS (ESI+) m/z: 329.1
[M + H]+.
8-Methoxy-2,3-diphenylquinazolin-4(3H)-one (4r). White solid,
yield 37%. H NMR (400 MHz, DMSO-d6) δ 7.79−7.72 (m, 1H),
1
7.55 (td, J = 8.0, 2.5 Hz, 1H), 7.46 (d, J = 8.0 Hz, 1H), 7.37 (dt, J =
7.9, 2.2 Hz, 2H), 7.35−7.17 (m, 8H), 3.95 (d, J = 2.4 Hz, 3H) ppm.
13C NMR (101 MHz, DMSO) δ 155.18, 154.29, 138.32, 136.26,
129.92, 129.43, 129.23, 129.02, 128.61, 128.01, 127.91, 122.28,
117.84, 115.76, 56.47 ppm. MS (ESI+) m/z: 329.1 [M + H]+.
General Procedure for the Synthesis of 5a and 5b. A solution of
KOH (0.38 mmol, 2 equiv) in 4 mL of H2O was added dropwise to 4l
or 4m (0.19 mmol, 1 equiv) in 1 mL of ethanol. The mixture was
heated at 60 °C for 5 h, cooled to rt, and quenched with dilute 1 ×
HCl solution (1 mL). Afterward, the solvent was removed under
6,7-Dimethoxy-2,3-diphenylquinazolin-4(3H)-one (4i). White
1
solid, yield 39%. H NMR (400 MHz, chloroform-d) δ 7.61 (d, J =
6.2 Hz, 1H), 7.34−7.05 (m, 11H), 4.14−3.85 (m, 6H) ppm. 13C
NMR (101 MHz, chloroform-d) δ 161.61, 155.30, 154.12, 149.47,
143.76, 137.86, 135.61, 129.18, 129.11, 128.95, 128.90, 128.32,
J
https://dx.doi.org/10.1021/acschemneuro.0c00600
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
vacuum. The crude residue was purified by flash chromatography on
silica gel (heptane−EtOAc 1:1, isocratic) to afford the title compound
5a or 5b.
6-Hydroxy-2,3-diphenylquinazolin-4(3H)-one (5a). White solid,
yield 19%. 1H NMR (400 MHz, DMSO-d6) δ 10.17 (s, 1H), 7.65 (d,
J = 8.8 Hz, 1H), 7.50 (d, J = 2.8 Hz, 1H), 7.39−7.15 (m, 11H) ppm.
13C NMR (101 MHz, DMSO-d6) δ 148.88, 138.05, 135.79, 130.64,
136.46, 135.69, 129.96, 129.48, 129.32, 129.00, 128.56, 128.01,
127.02, 124.76, 121.23, 29.45, 22.92 ppm. MS (ESI+) m/z: 355.2 [M
+ H]+.
General Procedure for the Synthesis of Stille Coupling Products
5i−k. In an inert atmosphere of Ar, to a microwave vial was added 4p
(0.3 mmol, 1 equiv), tributyltin (0.45 mmol, 1.5 equiv), Pd(PPh3)4
(0.015 mmol, 5 mol %), and toluene (2 mL). The vial was capped and
refluxed overnight. The reaction mixture was cooled to rt and stirred
with an aqueous solution of saturated KF (10 mL) for an hour. The
mixture was extracted with DCM (3 × 10 mL), and the combined
organic layers were washed with brine (30 mL), dried over Na2SO4,
filtered, and concentrated under vacuum. The crude residue was
purified by flash chromatography on silica gel (heptane−EtOAc 6:1,
isocratic) to afford the title compounds 5i−k (31−55%).
130.12, 129.89, 129.53, 129.40, 129.34, 129.06, 128.76, 127.98,
120.40 ppm. MS (ESI+) m/z: 315.1 [M + H]+.
8-Hydroxy-2,3-diphenylquinazolin-4(3H)-one (5b). White solid,
yield 23%. 1H NMR (400 MHz, DMSO-d6) δ 9.76 (s, 1H), 7.63 (dd,
J = 7.9, 1.4 Hz, 1H), 7.42 (ddd, J = 7.8, 4.9, 3.1 Hz, 3H), 7.36−7.16
(m, 9H) ppm. 13C NMR (101 MHz, DMSO-d6) δ 153.83, 153.61,
138.43, 136.80, 136.13, 130.00, 129.65, 129.24, 128.99, 128.56,
128.18, 127.87, 122.13, 119.36, 116.66 ppm. MS (ESI+) m/z: 315.1
[M + H]+.
2,3-Diphenyl-8-vinylquinazolin-4(3H)-one (5i). White solid, yield
1
53%. H NMR (400 MHz, DMSO-d6) δ 8.17 (ddd, J = 14.4, 7.8, 1.4
Hz, 2H), 7.74−7.58 (m, 2H), 7.46−7.19 (m, 10H), 6.07 (dd, J = 17.9,
1.4 Hz, 1H), 5.46 (dd, J = 11.1, 1.3 Hz, 1H) ppm. 13C NMR (101
MHz, DMSO-d6) δ 138.29, 131.93, 129.94, 129.47, 129.39, 129.03,
128.65, 128.00, 127.38, 126.56, 121.62, 116.99 ppm. MS (ESI+) m/z:
325.1 [M + H]+.
General Procedure for the Synthesis of Suzuki Coupling
Products 5c−h. In an inert atmosphere of Ar, to a microwave vial
was added 4p (0.3 mmol, 1 equiv), borate (0.33 mmol, 1.1 equiv),
Pd(PPh3)4 (0.015 mmol, 5 mol %), Cs2CO3 (0.6 mmol, 3 equiv), and
2 mL of toluene. The vial was capped and heated at 100 °C overnight.
The resulting solid was partitioned between EtOAc (20 mL) and
water (20 mL) and then the aqueous layer was extracted with EtOAc
(3 × 20 mL). The combined organic layers were washed with brine
(30 mL), dried over Na2SO4, filtered, and concentrated under
vacuum. The crude residue was purified by flash chromatography on
silica gel (heptane−EtOAc 6:1, isocratic) to afford the title
compounds 5c−h (43−78%).
8-Allyl-2,3-diphenylquinazolin-4(3H)-one (5j). White solid, yield
1
32%. H NMR (400 MHz, DMSO-d6) δ 8.09 (dd, J = 7.9, 1.5 Hz,
1H), 7.75 (dd, J = 7.4, 1.5 Hz, 1H), 7.54 (t, J = 7.6 Hz, 1H), 7.42−
7.38 (m, 2H), 7.35−7.21 (m, 8H), 6.11 (m, 1H), 5.19−5.01 (m, 2H),
3.82 (d, J = 6.7 Hz, 2H) ppm. 13C NMR (101 MHz, chloroform-d) δ
145.14, 137.99, 137.95, 134.94, 129.44, 129.17, 129.14, 128.91,
128.28, 127.73, 126.84, 125.36, 112.24, 38.32, 22.54 ppm. MS (ESI+)
m/z: 339.2 [M + H]+.
8-Methyl-2,3-diphenylquinazolin-4(3H)-one (5c). White solid,
1
yield 43%. H NMR (400 MHz, DMSO-d6): δ 8.05 (dd, J = 8.0,
8-(2-Methylallyl)-2,3-diphenylquinazolin-4(3H)-one (5k). White
solid, yield 56%. 1H NMR (400 MHz, DMSO-d6) δ 8.10 (dd, J = 7.9,
1.6 Hz, 1H), 7.73 (dd, J = 7.4, 1.6 Hz, 1H), 7.55 (t, J = 7.6 Hz, 1H),
7.44−7.19 (m, 10H), 4.84 (dd, J = 2.4, 1.4 Hz, 1H), 4.73 (dd, J = 2.5,
1.2 Hz, 1H), 3.79 (s, 2H), 1.72 (t, J = 1.1 Hz, 3H) ppm. 13C NMR
(101 MHz, DMSO-d6) δ 162.10, 144.99, 138.38, 137.67, 135.23,
129.96, 129.52, 129.37, 129.02, 128.60, 127.99, 127.24, 125.14,
121.26, 112.75, 38.07, 22.88 ppm. MS (ESI+) m/z: 353.2 [M + H]+.
General Procedure for the Synthesis of 5l and 5m. In an inert
atmosphere of Ar, to a microwave vial containing P(t-Bu)3 (0.02
mmol, 2% mol), Pd2(dba)3 (0.01 mmol, 1% mmol), and 4p (1 mmol,
1 equiv) in DMF (1 mL) was added trimethylsilylacetonitrile (1.2
mmol, 1.2 equiv), followed by ZnF2 (0.6 mmol, 0.6 equiv). The
heterogeneous reaction mixture was stirred at 90 °C for 8 h. The
crude reaction was then allowed to cool to room temperature and
diluted with Et2O (50 mL). The resulting solution was washed with
H2O. The organic phase was dried over Na2SO4, filtered, and
concentrated at reduced pressure. The residue was then purified by
chromatography on silica gel (heptane−EtOAc 10:1, isocratic) to
afford the title compound 5l (47%).
1.5 Hz, 1H), 7.78 (d, J = 7.3 Hz, 1H), 7.50 (t, J = 7.6 Hz, 1H), 7.43−
7.37 (m, 2H), 7.36−7.18 (m, 8H), 2.59 (s, 3H)) ppm. 13C NMR
(101 MHz, DMSO-d6) δ 129.98, 129.36, 129.02, 127.95, 126.94 ppm.
MS (ESI+) m/z: 313.1 [M + H]+.
2,3-Diphenyl-8-(thiophen-3-yl)quinazolin-4(3H)-one (5d). White
1
solid, yield 78%. H NMR (400 MHz, DMSO-d6): δ 8.24−8.10 (m,
3H), 7.78 (d, J = 5.0 Hz, 1H), 7.68−7.58 (m, 2H), 7.46−7.40 (m,
2H), 7.40−7.18 (m, 8H) ppm. 13C NMR (101 MHz, DMSO-d6): δ
162.01, 138.61, 138.28, 136.30, 134.79, 133.35, 129.93, 129.91,
129.49, 129.41, 129.05, 128.64, 128.07, 127.58, 126.04, 125.79,
125.57, 122.06 ppm. MS (ESI+) m/z: 381.1 [M + H]+.
2,3-Diphenyl-8-(pyrimidin-5-yl)quinazolin-4(3H)-one (5e). White
1
solid, yield 67%. H NMR (400 MHz, DMSO-d6): δ 9.18 (d, J = 7.1
Hz, 3H), 8.34 (dd, J = 8.0, 1.5 Hz, 1H), 8.15 (dd, J = 7.5, 1.5 Hz,
1H), 7.76 (t, J = 7.7 Hz, 1H), 7.41−7.27 (m, 7H), 7.25−7.18 (m,
3H) ppm. 13C NMR (101 MHz, DMSO-d6): δ 157.88, 157.45,
138.16, 136.05, 135.78, 132.24, 129.85, 129.54, 129.39, 129.10,
128.75, 128.03, 127.80, 121.93 ppm. MS (ESI+) m/z: 377.2 [M +
H]+.
8-(2-Methoxypyrimidin-5-yl)-2,3-diphenylquinazolin-4(3H)-one
(5f). White solid, yield 73%. 1H NMR (400 MHz, DMSO-d6): δ 9.00
(d, J = 1.2 Hz, 2H), 8.29 (d, J = 7.9 Hz, 1H), 8.09 (d, J = 7.4 Hz, 1H),
7.73 (t, J = 7.7 Hz, 1H), 7.43−7.22 (m, 9H), 3.97 (s, 3H) ppm. 13C
NMR (101 MHz, DMSO-d6): δ 160.38, 144.71, 136.13, 135.24,
129.86, 129.49, 129.39, 129.09, 128.72, 128.05, 127.76, 127.28,
125.76, 55.18 ppm. MS (ESI+) m/z: 407.2 [M + H]+.
When the reaction was allowed to proceed for a longer time (>18
h), an unexpected Knoevenagel reaction occurred and gave the main
product 5m (79%).
2-(4-Oxo-2,3-diphenyl-3,4-dihydroquinazolin-8-yl)acetonitrile
1
(5l). White solid, yield 48%. H NMR (400 MHz, DMSO-d6) δ 8.20
(d, J = 8.0 Hz, 1H), 7.98 (d, J = 7.4 Hz, 1H), 7.63 (td, J = 7.8, 2.0 Hz,
1H), 7.42 (d, J = 7.3 Hz, 2H), 7.39−7.20 (m, 8H), 4.33 (d, J = 2.9
Hz, 2H) ppm. 13C NMR (101 MHz, DMSO-d6) δ 161.70, 145.42,
138.14, 135.96, 134.91, 129.91, 129.63, 129.60, 129.09, 129.05,
128.77, 128.02, 127.58, 126.96, 121.45, 119.40, 19.26 ppm. MS (ESI
+) m/z: 338.1 [M + H]+.
2,3-Diphenyl-8-propylquinazolin-4(3H)-one (5g). White solid,
1
yield 51%. H NMR (400 MHz, DMSO-d6) δ 8.06 (dd, J = 8.0, 1.6
Hz, 1H), 7.75 (dd, J = 7.4, 1.5 Hz, 1H), 7.52 (t, J = 7.6 Hz, 1H), 7.39
(dd, J = 7.7, 2.0 Hz, 2H), 7.36−7.19 (m, 8H), 3.01 (t, J = 7.5 Hz,
2H), 1.73 (q, J = 7.4 Hz, 2H), 0.94 (t, J = 7.3 Hz, 3H) ppm. 13C
NMR (101 MHz, DMSO-d6) δ 140.23, 138.41, 136.44, 134.96,
129.97, 129.49, 129.30, 129.00, 128.57, 128.00, 127.19, 124.68,
121.24, 32.78, 23.83, 14.38 ppm. MS (ESI+) m/z: 341.2 [M + H]+.
8-Isobutyl-2,3-diphenylquinazolin-4(3H)-one (5h). White solid,
3-(Dimethylamino)-2-(4-oxo-2,3-diphenyl-3,4-dihydroquinazo-
1
lin-8-yl)acrylonitrile (5m). White solid, yield 80%. H NMR (400
MHz, DMSO-d6) δ 7.99 (dd, J = 7.9, 1.5 Hz, 1H), 7.84 (s, 1H), 7.77
(dd, J = 7.6, 1.5 Hz, 1H), 7.54 (t, J = 7.8 Hz, 1H), 7.49−7.44 (m,
2H), 7.35−7.22 (m, 8H), 3.20 (s, 6H) ppm. 13C NMR (101 MHz,
DMSO-d6) δ 162.06, 155.97, 153.69, 138.25, 136.19, 135.06, 132.69,
129.85, 129.70, 129.54, 129.08, 128.64, 128.00, 127.49, 123.61,
121.91, 70.88 ppm. MS (ESI+) m/z: 394.2 [M + H]+.
1
yield 53%. H NMR (400 MHz, DMSO-d6) δ 8.07 (dd, J = 7.9, 1.6
Hz, 1H), 7.72 (dd, J = 7.4, 1.6 Hz, 1H), 7.52 (t, J = 7.6 Hz, 1H),
7.42−7.36 (m, 2H), 7.34−7.19 (m, 8H), 2.91 (d, J = 7.1 Hz, 2H),
2.11 (dq, J = 13.5, 6.8 Hz, 1H), 0.91 (d, J = 6.6 Hz, 6H) ppm. 13C
NMR (101 MHz, DMSO-d6) δ 162.16, 146.01, 139.40, 138.42,
8-Acetyl-2,3-diphenylquinazolin-4(3H)-one (5n). A microwave
vial was charged with 1,3,5-trimethoxybenzene (0.03 mmol, 0.1
K
https://dx.doi.org/10.1021/acschemneuro.0c00600
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
equiv), and 4p (0.3 mmol, 1 equiv), CsF (1.2 mmol, 1 equiv), and
Pd(PPh3)4 (0.03 mmol, 10% mol) were added to the vial. The vial
was placed under an atmosphere of Ar and 1,2-dichloroethane (0.5
mL) and acetyltrimethylsilane (0.6 mmol, 2 equiv) were added. The
vial was capped and heated to 75 °C for 6 h. After cooling to rt, the
mixture was diluted with heptane (0.5 mL) and filtered through a plug
of silica gel (10 mL EtOAc eluent), and the volatiles removed under
reduced pressure. The crude residue was purified by flash
chromatography on silica gel (heptane−EtOAc 3:1, isocratic) to
MHz, DMSO-d6) δ 8.12 (dd, J = 8.0, 1.6 Hz, 1H), 7.76 (dd, J = 7.4,
1.6 Hz, 1H), 7.55 (t, J = 7.6 Hz, 1H), 7.39−7.20 (m, 10H), 4.15 (s,
2H), 2.19 (s, 3H) ppm. 13C NMR (101 MHz, DMSO-d6) δ 206.19,
136.61, 134.12, 129.93, 129.48, 129.05, 128.01, 127.21, 45.40, 30.59
ppm. MS (ESI+) m/z: 355.1 [M + H]+.
Computational Study. For all computational work, the software
package MOE 2019.01 was used. The crystal structure of the α1β2γ2
GABAAR (PDB code 6D6U)39 was prepared by applying the
algorithm Protonate3D (standard setup). Thereafter, chain A (β2
subunit) and chain B (α1 subunit) were extracted from the
pentameric receptor complex. The SiteFinder algorithm was applied
to identify the pocket formed at the intersection of β2 TM2+TM3 and
α1 TM1+TM2. Docking was thereafter performed using the function
Dock with standard parameter setup and REFINEMENT set to
Induced Fit.
1
afford the title compound 5n (27%). White solid, yield 27.8%. H
NMR (400 MHz, DMSO-d6) δ 8.37 (dt, J = 7.9, 1.4 Hz, 1H), 8.03
(dt, J = 7.5, 1.5 Hz, 1H), 7.72−7.63 (m, 1H), 7.42 (dt, J = 7.7, 1.4 Hz,
2H), 7.41−7.21 (m, 8H), 2.79 (s, 3H) ppm. 13C NMR (101 MHz,
DMSO-d6) δ 134.18, 130.20, 129.88, 129.67, 129.46, 129.09, 128.79,
128.10, 99.98, 32.92 ppm. MS (ESI+) m/z: 341.1 [M + H]+.
8-Morpholino-2,3-diphenylquinazolin-4(3H)-one (5o). A solution
of 4p (0.3 mmol, 1 equiv) in toluene (1 mL) was added to a mixture
of morpholine (0.3 mmol, 1 equiv), Pd2(dba)3 (7.5 μmol, 2.5 mol %),
Xantphos (15 μmol, 5 mol %), and Cs2CO3 (0.36 mmol, 1.2 equiv)
under an atmosphere of Ar. The reaction mixture was heated at 100
°C for 4 h, and then the cooled reaction mixture was diluted with
ethyl acetate and filtered through Celite, and the filtrate was
concentrated under vacuum. The resultant residue was subjected to
flash chromatography (heptane−EtOAc 3:1, isocratic) to give the title
compound 5o as a white solid, yield 63%. 1H NMR (400 MHz,
DMSO-d6): δ 7.79 (dd, J = 7.9, 1.3 Hz, 1H), 7.49 (t, J = 7.9 Hz, 1H),
7.39 (dd, J = 7.8, 1.8 Hz, 2H), 7.36−7.22 (m, 9H), 3.77 (t, J = 4.6 Hz,
4H), 3.35 (dd, J = 5.9, 3.3 Hz, 4H) ppm. 13C NMR (101 MHz,
DMSO-d6) δ 162.18, 152.65, 148.44, 140.48, 138.38, 136.38, 129.88,
129.49, 129.39, 129.02, 128.57, 128.02, 127.90, 122.38, 121.95,
119.19, 66.67, 52.00 ppm. MS (ESI+) m/z: 384.2 [M + H]+.
Pharmacology. The functional properties of the 36 PPQ analogs
were characterized at human α1β2γ2S, α3β2γ2S, α4β2γ2S, α5β2γ2S, and
α6β2γ2S GABAARs transiently expressed in tsA201 cells in the FLIPR
Membrane Potential Blue (FMP) assay (Molecular Devices, Crawley,
UK) essentially as previously described.34 Briefly, tsA201 cells were
cultured in Dulbecco’s modified Eagle medium Glutamax-I
supplemented with 10% fetal bovine serum, 100 U/mL penicillin,
and 100 μg/mL streptomycin at 37 °C in a humidified 5% CO2
atmosphere. Cells (2 ×106) were split into a 10 cm tissue culture plate
and transfected the following day with a total of 8 μg of cDNA (2 μg
of α-pCDNA3.1 (α1, α3, α4, α5, α6), 2 μg β2-pCDNA3.1, and 4 μg γ2S-
pCDNA3.1) using the Polyfect transfection reagent (Qiagen, Hilden,
Germany). Sixteen to twenty hours after the transfection, the cells
were split into poly(D-lysine)-coated black 96-well plates with clear
bottoms (BD Biosciences, Palo Alto, CA). The following day, the
culture medium was aspirated, and the cells were washed once with
100 μL of assay buffer (140 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl2,
1.2 mM MgCl2, 11 mM HEPES, 10 mM D-glucose, pH 7.4), after
which the cells were incubated in 100 μL of assay buffer
supplemented with FMP dye (0.5 mg/mL) at 37 °C for 30 min.
The 96-well plate was assayed in a FLEXStation3 Benchtop Multi-
Mode Microplate Reader (Molecular Devices, Crawley, U.K.)
measuring emission [in fluorescence units (FU)] at 565 nm caused
by excitation at 525 nm before and up to 90 s after addition of 33.3
μL of assay buffer supplemented with GABA (an EC10 assay
concentration for the specific receptor subtype, determined on the
day of the experiment) and various concentrations of the test
compounds.
2-(4-Oxo-2,3-diphenyl-3,4-dihydroquinazolin-8-yl)acetic Acid
(6a). Compound 5l (0.3 mmol, 1 equiv) was treated with KOH
(0.39 mmol, 1.3 equiv) in 4 mL of water. The mixture was heated at
60 °C for 5 h, cooled to rt, and quenched with dilute 1 × HCl
solution (1 mL). The solvent was removed under vacuum, and the
crude residue was purified by flash chromatography on silica gel
(heptane−EtOAc 1:1, isocratic) to afford the title compound 6a as a
1
white solid, yield 8.1%. H NMR (400 MHz, DMSO-d6) δ 12.27 (s,
1H), 8.12 (dd, J = 8.0, 1.5 Hz, 1H), 7.83 (dd, J = 7.4, 1.5 Hz, 1H),
7.56 (t, J = 7.7 Hz, 1H), 7.44−7.18 (m, 10H), 4.01 (s, 2H) ppm. 13C
NMR (101 MHz, DMSO-d6) δ 172.94, 161.95, 146.07, 136.45,
136.15, 133.62, 129.94, 129.58, 129.47, 129.06, 128.67, 128.0, 127.17,
125.72, 121.07, 36.42 ppm. MS (ESI+) m/z: 357.1 [M + H]+.
8-(5-Aminoisoxazol-4-yl)-2,3-diphenylquinazolin-4(3H)-one
(6b). To a microwave vial containing 5m (0.2 mmol, 1 equiv) and
hydroxylamine hydrochloride (0.3 mmol, 1.5 equiv) was added EtOH
(3 mL). The vial was capped and refluxed for 1 h. The solvent was
removed under vacuum, and the crude residue was purified by flash
chromatography on silica gel (heptane−EtOAc 3:1, isocratic) to
AUTHOR INFORMATION
■
Corresponding Authors
Anders A. Jensen − Department of Drug Design and
Pharmacology, Faculty of Health and Medical Sciences,
University of Copenhagen, Copenhagen 2100, Denmark;
Phone: +45 23 96 24 34; Email: aaj@sund.ku.dk
Lennart Bunch − Department of Drug Design and
Pharmacology, Faculty of Health and Medical Sciences,
University of Copenhagen, Copenhagen 2100, Denmark;
orcid.org/0000-0002-0180-4639; Phone: +45 35 33 62
44; Email: lebu@sund.ku.dk
1
afford the title compound 6b as a white solid, yield 41%. H NMR
(400 MHz, DMSO-d6) δ 8.64 (s, 1H), 8.08 (d, J = 7.8 Hz, 1H), 7.94
(d, J = 7.5 Hz, 1H), 7.60 (t, J = 7.7 Hz, 1H), 7.51−7.42 (m, 2H),
7.41−7.12 (m, 10H) ppm. 13C NMR (101 MHz, DMSO-d6) δ
166.91, 162.04, 154.65, 153.08, 143.72, 136.05, 133.47, 129.87,
129.47, 129.35, 129.04, 128.64, 128.09, 127.61, 124.28, 122.00, 90.41
ppm. MS (ESI+) m/z: 381.1 [M + H]+.
8-(2-Oxopropyl)-2,3-diphenylquinazolin-4(3H)-one (6c). To a
solution of 5k (0.2 mmol, 1 equiv) in a mixture of dry THF (3
mL)/deionized H2O (2 mL) was added 4% OsO4 in deionized H2O
(100 μL). The resulting mixture was stirred at rt for 20 min, and then
NaIO4 (0.9 mmol, 4.5 equiv) was added in several portions. The
reaction mixture was stirred at rt for 16 h and then quenched by the
addition of saturated aqueous Na2SO3 solution (5 mL). The resulting
mixture was extracted with ethyl acetate (2 × 10 mL), and the
combined organic layers were washed with saturated aqueous
NaHCO3 solution (5 mL) and dried over Na2SO4, and the solvent
was removed under reduced pressure. The resultant residue was
subjected to flash chromatography (heptane−EtOAc 3:1, isocratic) to
Author
Peng-Fei Wang − Department of Drug Design and
Pharmacology, Faculty of Health and Medical Sciences,
University of Copenhagen, Copenhagen 2100, Denmark;
School of Pharmaceutical Sciences and Innovative Drug
Research Centre, Chongqing University, Chongqing 400044,
P.R. China
Complete contact information is available at:
https://pubs.acs.org/10.1021/acschemneuro.0c00600
1
give the title compound 6c, white solid, yield 63%. H NMR (400
L
https://dx.doi.org/10.1021/acschemneuro.0c00600
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
(15) Ramerstorfer, J., Furtmuller, R., Sarto-Jackson, I., Varagic, Z.,
Sieghart, W., and Ernst, M. (2011) The GABAA receptor α+β-
interface: a novel target for subtype selective drugs. J. Neurosci. 31,
870−7.
(16) Martinez Botella, G., Salituro, F. G., Harrison, B. L., Beresis, R.
T., Bai, Z., Blanco, M. J., Belfort, G. M., Dai, J., Loya, C. M., Ackley,
M. A., Althaus, A. L., Grossman, S. J., Hoffmann, E., Doherty, J. J., and
Robichaud, A. J. (2017) Neuroactive Steroids. 2. 3alpha-Hydroxy-
3beta-methyl-21-(4-cyano-1H-pyrazol-1′-yl)-19-nor-5beta-pregnan-
20 -one (SAGE-217): A Clinical Next Generation Neuroactive
Steroid Positive Allosteric Modulator of the (gamma-Aminobutyric
Acid)A Receptor. J. Med. Chem. 60 (18), 7810−7819.
(17) Sieghart, W. (2015) Allosteric modulation of GABAA receptors
via multiple drug-binding sites. Adv. Pharmacol. 72, 53−96.
(18) Olsen, R. W. (2015) Allosteric ligands and their binding sites
define γ-aminobutyric acid (GABA) type A receptor subtypes. Adv.
Pharmacol. 73, 167−202.
(19) Li, G. D., Chiara, D. C., Sawyer, G. W., Husain, S. S., Olsen, R.
W., and Cohen, J. B. (2006) Identification of a GABAA receptor
anesthetic binding site at subunit interfaces by photolabeling with an
etomidate analog. J. Neurosci. 26, 11599−605.
(20) Belelli, D., Lambert, J. J., Peters, J. A., Wafford, K., and Whiting,
P. J. (1997) The interaction of the general anesthetic etomidate with
the γ-aminobutyric acid type A receptor is influenced by a single
amino acid. Proc. Natl. Acad. Sci. U. S. A. 94, 11031−6.
(21) Wingrove, P. B., Wafford, K. A., Bain, C., and Whiting, P. J.
(1994) The modulatory action of loreclezole at the γ-aminobutyric
acid type A receptor is determined by a single amino acid in the β2
and β3 subunit. Proc. Natl. Acad. Sci. U. S. A. 91, 4569−73.
(22) Wafford, K. A., Bain, C. J., Quirk, K., McKernan, R. M.,
Wingrove, P. B., Whiting, P. J., and Kemp, J. A. (1994) A novel
allosteric modulatory site on the GABAA receptor β subunit. Neuron
12, 775−82.
(23) Stewart, D., Desai, R., Cheng, Q., Liu, A., and Forman, S. A.
(2008) Tryptophan mutations at azi-etomidate photo-incorporation
sites on α1 or β2 subunits enhance GABAA receptor gating and
reduce etomidate modulation. Mol. Pharmacol. 74, 1687−95.
(24) Stewart, D. S., Hotta, M., Li, G. D., Desai, R., Chiara, D. C.,
Olsen, R. W., and Forman, S. A. (2013) Cysteine substitutions define
etomidate binding and gating linkages in the α-M1 domain of γ-
aminobutyric acid type A (GABAA) receptors. J. Biol. Chem. 288,
30373−86.
(25) Olander, E. R., Madjroh, N., Bunch, L., Soderhielm, P. C., and
Jensen, A. A. (2018) Delineation of the functional properties and the
mechanism of action of AA29504, an allosteric agonist and positive
allosteric modulator of GABAA receptors. Biochem. Pharmacol. 150,
305−319.
(26) Krasowski, M. D., Nishikawa, K., Nikolaeva, N., Lin, A., and
Harrison, N. L. (2001) Methionine 286 in transmembrane domain 3
of the GABAA receptor β subunit controls a binding cavity for
propofol and other alkylphenol general anesthetics. Neuropharmacol-
ogy 41, 952−64.
(27) Halliwell, R. F., Thomas, P., Patten, D., James, C. H., Martinez-
Torres, A., Miledi, R., and Smart, T. G. (1999) Subunit-selective
modulation of GABAA receptors by the non-steroidal anti-
inflammatory agent, mefenamic acid. Eur. J. Neurosci. 11, 2897−905.
(28) Khom, S., Baburin, I., Timin, E., Hohaus, A., Trauner, G.,
Kopp, B., and Hering, S. (2007) Valerenic acid potentiates and
inhibits GABAA receptors: molecular mechanism and subunit
specificity. Neuropharmacology 53, 178−87.
(29) Luger, D., Poli, G., Wieder, M., Stadler, M., Ke, S., Ernst, M.,
Hohaus, A., Linder, T., Seidel, T., Langer, T., Khom, S., and Hering, S.
(2015) Identification of the putative binding pocket of valerenic acid
on GABAA receptors using docking studies and site-directed
mutagenesis. Br. J. Pharmacol. 172 (22), 5403−13.
(30) Chua, H. C., Christensen, E. T., Hoestgaard-Jensen, K.,
Hartiadi, L. Y., Ramzan, I., Jensen, A. A., Absalom, N. L., and Chebib,
M. (2016) Kavain, the major constituent of the anxiolytic kava extract,
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank the International Postdoctoral Exchange
Fellowship Program 2017 of China (No. 20170090) and the
Novo Nordisk Foundation for financial support.
LIST OF ABBREVIATIONS
■
Cl-PPQ, 3-(2-chlorophenyl)-2-phenylquinazolin-4(3H)-one;
FMP, FLIPR Membrane Potential; MTQ, methaqualone (2-
methyl-3-(o-tolyl)-quinazolin-4(3H)-one); PAM, positive al-
losteric modulator; PPQ, 2,3-diphenylquinazolin-4(3H)-one
REFERENCES
■
(1) Olsen, R. W., and Sieghart, W. (2008) International Union of
Pharmacology. LXX. Subtypes of γ-aminobutyric acidA receptors:
classification on the basis of subunit composition, pharmacology, and
function. Update. Pharmacol. Rev. 60, 243−60.
(2) Chua, H. C., and Chebib, M. (2017) GABAA Receptors and the
Diversity in their Structure and Pharmacology. Adv. Pharmacol. 79, 1−
34.
(3) Miller, P. S., and Aricescu, A. R. (2014) Crystal structure of a
human GABAA receptor. Nature 512, 270−275.
(4) Laverty, D., Desai, R., Uchanski, T., Masiulis, S., Stec, W. J.,
Malinauskas, T., Zivanov, J., Pardon, E., Steyaert, J., Miller, K. W., and
Aricescu, A. R. (2019) Cryo-EM structure of the human α1β3γ2
GABAA receptor in a lipid bilayer. Nature 565 (7740), 516−520.
(5) Whiting, P. J. (2003) GABA-A receptor subtypes in the brain: a
paradigm for CNS drug discovery? Drug Discovery Today 8, 445−50.
̈
(6) Hortnagl, H., Tasan, R. O., Wieselthaler, A., Kirchmair, E.,
Sieghart, W., and Sperk, G. (2013) Patterns of mRNA and protein
expression for 12 GABAA receptor subunits in the mouse brain.
Neuroscience 236, 345−72.
(7) Atack, J. R. (2011) GABAA receptor subtype-selective
modulators. I. α2/α3-selective agonists as non-sedating anxiolytics.
Curr. Top. Med. Chem. 11, 1176−1202.
(8) Stephens, D. N., King, S. L., Lambert, J. J., Belelli, D., and Duka,
T. (2017) GABAA receptor subtype involvement in addictive
behaviour. Genes Brain Behav. 16, 149−184.
(9) Atack, J. R. (2011) GABAA receptor subtype-selective
modulators. II. α5-selective inverse agonists for cognition enhance-
ment. Curr. Top. Med. Chem. 11, 1203−1214.
̈
(10) Rudolph, U., and Mohler, H. (2014) GABAA receptor
subtypes: Therapeutic potential in Down syndrome, affective
disorders, schizophrenia, and autism. Annu. Rev. Pharmacol. Toxicol.
54, 483−507.
(11) Engin, E., Benham, R. S., and Rudolph, U. (2018) An Emerging
Circuit Pharmacology of GABAA Receptors. Trends Pharmacol. Sci. 39
(8), 710−732.
(12) Braat, S., and Kooy, R. F. (2015) The GABAA Receptor as a
Therapeutic Target for Neurodevelopmental Disorders. Neuron 86
(5), 1119−30.
(13) Maramai, S., Benchekroun, M., Ward, S. E., and Atack, J. R.
(2020) Subtype Selective γ-Aminobutyric Acid Type A Receptor
(GABAAR) Modulators Acting at the Benzodiazepine Binding Site:
An Update. J. Med. Chem. 63 (7), 3425−3446.
(14) Hoestgaard-Jensen, K., Dalby, N. O., Krall, J., Hammer, H.,
Krogsgaard-Larsen, P., Frølund, B., and Jensen, A. A. (2014) Probing
α4βδ GABAA receptor heterogeneity: differential regional effects of a
functionally selective α4β1δ/α4β3δ receptor agonist on tonic and
phasic inhibition in rat brain. J. Neurosci. 34, 16256−16272.
M
https://dx.doi.org/10.1021/acschemneuro.0c00600
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

ACS Chemical Neuroscience
pubs.acs.org/chemneuro
Research Article
potentiates GABAA Receptors: Functional characteristics and
molecular mechanism. PLoS One 11 (6), No. e0157700.
(31) Chiara, D. C., Jayakar, S. S., Zhou, X., Zhang, X., Savechenkov,
P. Y., Bruzik, K. S., Miller, K. W., and Cohen, J. B. (2013) Specificity
of intersubunit general anesthetic-binding sites in the transmembrane
domain of the human α1β3γ2 γ-aminobutyric acid type A (GABAA)
receptor. J. Biol. Chem. 288, 19343−57.
(32) Hilpert, K., Elliott, M. R., Volkmer-Engert, R., Henklein, P.,
Donini, O., Zhou, Q., Winkler, D. F., and Hancock, R. E. (2006)
Sequence requirements and an optimization strategy for short
antimicrobial peptides. Chem. Biol. 13 (10), 1101−7.
(33) Hammer, H., Bader, B. M., Ehnert, C., Bundgaard, C., Bunch,
L., Hoestgaard-Jensen, K., Schroeder, O. H., Bastlund, J. F.,
Gramowski-Voß, A., and Jensen, A. A. (2015) A multifaceted
GABAA receptor modulator: Functional properties and molecular
mechanism of action of the sedative-hypnotic and recreational drug
methaqualone (Quaalude). Mol. Pharmacol. 88, 401−420.
(34) Madjroh, N., Olander, E. R., Bundgaard, C., Soderhielm, P. C.,
and Jensen, A. A. (2018) Functional properties and mechanism of
action of PPTQ, an allosteric agonist and low nanomolar positive
allosteric modulator at GABAA receptors. Biochem. Pharmacol. 147,
153−169.
(35) Bozdag, M., Alafeefy, A. M., Vullo, D., Carta, F., Dedeoglu, N.,
Al-Tamimi, A.-M. S., Al-Jaber, N. A., Scozzafava, A., and Supuran, C.
T. (2015) Benzenesulfonamides incorporating bulky aromatic/
heterocyclic tails with potent carbonic anhydrase inhibitory activity.
Bioorg. Med. Chem. 23 (24), 7751−7764.
(36) Dandia, A., Singh, R., and Sarawgi, P. (2005) Reinvestigation of
the synthesis of 2-benzyl-3-aryl-quinazolin-4-[3 H]-ones. An
improved multi-component procedure using microwaves. Org. Prep.
Proced. Int. 37 (4), 397−402.
(37) Saravanan, G., Alagarsamy, V., and Kumar, P. D. (2015)
Synthesis and pharmacological investigations of novel 2-phenyl-
quinazolin-4 (3H)-one derivatives. Med. Chem. Res. 24 (1), 408−422.
(38) Wu, L., and Hartwig, J. F. (2005) Mild palladium-catalyzed
selective monoarylation of nitriles. J. Am. Chem. Soc. 127 (45),
15824−15832.
(39) Zhu, S., Noviello, C. M., Teng, J., Walsh, R. M., Kim, J. J., and
Hibbs, R. E. (2018) Structure of a human synaptic GABA A receptor.
Nature 559 (7712), 67−72.
N
https://dx.doi.org/10.1021/acschemneuro.0c00600
ACS Chem. Neurosci. XXXX, XXX, XXX−XXX