Structural Determinants for the Mode of Action of Imidazopyridine DS2 at δ-Containing γ-Aminobutyric Acid Type A Receptors

DS2 at δ ‑Containing γ ‑Aminobutyric Acid Type A Receptors

Article

Structural Determinants for the Mode of Action of Imidazopyridine

Frederik Rostrup, Christina B. Falk-Petersen, Kasper Harpso ̷e, Stine Buchleithner, Irene Conforti,

Sascha Jung, David E. Gloriam, Tanja Schirmeister, Petrine Wellendorph, and Bente Fro ̷lund *

Cite This: J. Med. Chem. 2021, 64, 4730 −4743

Read Online

ACCESS

Metrics & More

Article Recommendations

sı Supporting Information

ABSTRACT: Despite the therapeutic relevance of δ-containing γ-

aminobutyric acid type A receptors (GABA Rs) and the need for

δ-selective compounds, the structural determinants for the mode

and molecular site of action of δ-selective positive allosteric

modulator imidazo[1,2-a]pyridine DS2 remain elusive. To guide

the quest for insight, we synthesized a series of DS2 analogues

guided by a structural receptor model. Using a ﬂuorescence-based

ﬂuorometric imaging plate reader membrane potential assay, we

found that the δ-selectivity and the pharmacological proﬁle are

severely aﬀected by substituents in the 5-position of the

imidazopyridine core scaﬀold. Interestingly, the 5-methyl, 5-

bromo, and 5-chloro DS2 analogues, 30, 35, and 36, were

shown to be superior to DS2 at α4β1δ as mid-high nanomolar

potency δ-selective allosteric modulators, displaying 6−16 times higher potency than DS2. Of these, 30 also displayed at least 60-fold

selectivity for α4β1δ over α4β1γ2 receptor subtypes representing a potential tool for the selective characterization of δ-containing

GABA Rs in general.

INTRODUCTION

example, alcoholism, epilepsy, and major depression

■

disorder, these receptors have been in focus in the last few

decades as potential drug targets (for reviews, see, e.g., refs 11.

and 12). To date, few compounds exhibit δ-selectivity

γ-Aminobutyric acid (GABA) is a major inhibitory neuro-

transmitter in the central nervous system, and it mediates its

eﬀect via the ionotropic γ-aminobutyric acid type A receptors

13,14

including imidazo[1,2-a]pyridine DS2 (Figure 1), shown by

Jensen et al. to be a functionally selective α4/6β3δ positive

allosteric modulator (PAM), relative to its action at α4β3γ2

(

GABA Rs) and the metabotropic γ-aminobutyric acid type B

receptors. GABA Rs are clinically employed targets for

numerous drugs, including anesthetics, barbiturates, and

and α1β3γ2 receptors. In addition, DS2 has been shown to

selectively modulate α4β1/3δ over α4β1/3 receptors in 1−10

μM concentrations. Although DS2 does not appear to pass the

benzodiazepines because of their sedative, anxiolytic, and

anticonvulsant eﬀects. GABA Rs form transmembrane hetero-

pentameric complexes from at least 19 diﬀerent subunits: α1−

blood−brain barrier in rats and mice, we recently reported

, β1−3, γ1−3, ρ1−3, δ, θ, π, and ε. The combination of

11 11

uptake of a C-labeled DS2 analogue, [ C]DS2OMe (Figure

subunits aﬀects the subcellular localization and type of

) into the pig brain.

inhibition being mediated. The majority of GABA Rs have

the general stoichiometry of 2α, 2β, and 1γ subunits and are

A previous structure−activity relationship (SAR) study on

the direct agonist eﬀect of DS2 and the 6,8-dibromo analogue,

DS1 (Figure 1), reported that DS1, but not DS2, is able to

increase [ H]-ethynylbicyclooethobenzoate ([ H]-EBOB)

binding to δ-containing GABA Rs in the absence of

GABA. EBOB is a potent noncompetitive GABA R

typically located synaptically mediating fast and transient

inhibition. The synaptic GABA Rs are sensitive to high

concentrations of GABA and are prone to desensitization.

Extrasynaptic GABA Rs typically incorporate a δ-subunit.

Although it is generally accepted that the δ-subunit in its

cognate receptors simply replaces the γ-subunit with respect to

26,29

Received: December 15, 2020

Published: April 13, 2021

arrangement, this is still not unequivocally established.

Pharmacologically, extrasynaptic receptors are sensitive to low

concentrations of GABA and mediate tonic inhibition, distinct

from the fast and transient synaptic inhibition, and are less

prone to desensitization. As the δ-subunit-containing

GABA Rs have been proposed to be implicated in, for

2021 American Chemical Society

Journal of Medicinal Chemistry

730

J. Med. Chem. 2021, 64, 4730−4743

of GABA. However, both compounds lost their δ -selectivity

Article

in the presence of GABA, thus limiting their use for in vivo

studies.

Despite DS2 being an important tool compound

regardless of extensive investigations, its molecular site of

14,22,23

and

,15

action remains elusive.

Herein, we describe our eﬀorts to

investigate the molecular determinants for the pharmacology

of DS2 at α4β1δ GABA Rs by probing the δ-selective

modulatory eﬀect on the activity of GABA using a ﬂuorometric

imaging plate reader (FLIPR) membrane potential (FMP)

functional assay. Ultimately, we sought to assist future

modeling and site-directed mutagenesis studies that could

lead to the revelation of the undiscovered DS2 binding

15,17

site.

Consequently, we expand the pharmacophore for the

δ-selective allosteric eﬀect on GABA modulation by introduc-

ing various synthesis strategies to structurally diversify the

imidazopyridine scaﬀold of DS2 including a structural receptor

model.

Figure 1. Chemical structures of the δ-selective GABA R allosteric

RESULTS AND DISCUSSION

■

modulator, DS2, the positron emission tomography (PET) ligand

Ligand Design. Based on the structural similarity between

DS2 and the standard benzodiazepine binding site ligand

zolpidem (Figure 1), we hypothesized that DS2 may bind in a

benzodiazepine-like binding pocket at the intersubunit α4/δ-

candidate, [ C]DS2OMe, the GABA R allosteric agonist, DS1, and

the standard benzodiazepine binding site ligand zolpidem.

interface at α4β1δ GABA Rs. This extracellular domain

antagonist that binds to the picrotoxinin binding site within the

(ECD) pocket has been referred to as the C-loop pocket,

and the existence of such a pocket at δ-containing GABA Rs

GABA R ion channel. [ H]-EBOB binding is sensitive to

15,25

conformational changes in the chloride-conducting channel,

which can be mediated by increasing concentrations of GABA

and/or aﬀected by modulators in the presence or absence of

has previously been debated.

The binding hypothesis is

further challenged by the subunit stoichiometry and assembly

26−29

of δ-GABA Rs, which continues to be poorly understood.

9−21

GABA.

In the referred SAR study, two DS1 analogues

To rationally guide ligand design, we constructed a

homology model of the ECD C-loop pocket at the α4/δ-

interface using the cryogenic electron microscopy (cryoEM)

were identiﬁed as δ-selective allosteric agonists in the absence

Figure 2. (A and B) Putative binding model of DS2 (white carbon atoms) in the extracellular part of the GABA R interface between the α4 (purple

carbon atoms) and δ (orange carbon atoms) subunits showing charge-assisted hydrogen bonds and π-cation and π-π interactions between DS2 and

the receptor (yellow, green, and cyan dotted lines, respectively), highlighted in another view in (B). The amino acid backbone of GABA Rs is

depicted as a cartoon and selected residues as sticks, while the van der Waals surface is shown in transparent gray. In addition to the carbon atoms,

which are colored according to molecules, oxygen, nitrogen, sulfur, and chlorine are red, blue, yellow, and green, respectively. (C) Schematic

overview of the explored sites in the DS2 scaﬀold and numbering of the positions.

731

J. Med. Chem. 2021, 64, 4730−4743

Journal of Medicinal Chemistry

Article

structure of the human α1β2γ2 GABA R in complex with

Scheme 2. Synthesis of 2-Substituted Imidazo[1,2-

a]Pyridine-3-amine

ﬂumazenil (PDB-code: 6D6T) as a template for our model.

This was the ﬁrst structure showing a binding in the

benzodiazepine site, and though later published structures

seem more relevant as templates, that is, containing positive

modulators, diazepam, and alprazolam, structural comparison

shows very little variation in the binding site (Figure S1). The

model was subsequently used for induced ﬁt and extra

precision (XP)-Glide docking of previously reported com-

pounds. The putative binding mode obtained for DS2

Figure 2A,B) predicts the following speciﬁc interactions: (i)

(

Reagents and conditions: (a) 2-chloroacetic acid, Et N/H O, 90 °C

charge-assisted hydrogen bonds from the central amide to

α4R135 and δE71, (ii) π-cation interaction from the

imidazo[1,2-a]pyridine to δR157 and δH204, and (iii) π-

stacking between the chlorophenyl and δF90. Additionally,

other surrounding amino acids delineate the binding site,

which are considered as a van der Waals surface, some in direct

contact and some further away allowing space for substituents.

As expected, there is a positional overlap of DS2 and diazepam

to give 2-(2-iminopyridin-1(2H)-yl)acetic acid (19); (b) POCl₃,

toluene, and reﬂux; (c) Conc. H SO , HNO , and 5 °C to rt.; (d)

Pd(PPh ) , DME/H O, Na CO , and reﬂux; (e) SnCl , MeOH, and

reﬂux_.

from the binding site leaving space for substituents (Figure

B), which was challenged by compound 25 (Scheme 3). In

contrast, the binding mode model shows limited to a fair

amount of room for substitution possibilities in the 6- and 5-

positions of the imidazo[1,2-a]pyridine, which was explored by

compounds 27, 28, and 30−36 (Scheme 3).

(

plus the very similar alprazolam) in the α4/δ- and α1/γ2-

interfaces with similar dimer interfaces and C-loop closure, but

distinct interactions reﬂecting diﬀerent modulators and

binding sites (Figure S2).

Synthesis of Target Compounds. The new DS2

analogues 24−34 and 40 were synthesized as depicted in

Schemes 1−4. Compound 35 and 36 were synthesized

To test the predicted binding mode, we systematically varied

the structural scaﬀold of DS2, as depicted in Figure 2C. If our

predicted binding mode is true, the amide functionality in

position 3 (Figure 2C) with both a hydrogen bond donor and

acceptor function should be crucial to modulation, which can

be probed by bioisosteric replacement, exempliﬁed by

compounds 40 and 42 (Scheme 4). To this end, we examined

the lack of aromatic interactions to the thiophene in position 2

according to procedures reported in the literature.

The 2-substituted imidazo-[1,2-a]pyridin-3-amides (24−34)

were all obtained in a two-step procedure via the

corresponding amines (10−18). Initially, one-step GBB-

31−33

multicomponent reactions

with potassium cyanide, the

respective 2-aminopyridines, aldehydes, and diﬀerent Lewis

acids were attempted to obtain 10. However, byproduct

formation, including formation of bicondensated adducts,

and poor starting material conversion resulted in low yields

and unsuccessful workups. Alternatively, the tert-butyl-

protected intermediates, imidazo-[1,2-a]pyridine-3-amines

(

Figure 2C) and the available space from almost protruding

out of the binding site by increasing the aromatic ring size and

removing the aromaticity by compounds 18 and 22 (Schemes

and 2). Likewise, the chlorophenyl of DS2 is predicted to

point into an unexplored area under the C-loop and even out

(

1−9), were obtained in medium to high yield through similar

Scheme 1. Imidazo[1,2-a]Pyridine-3-Amine Synthesis Via

31−33

GBB-multicomponent reactions

using the appropriate

the Groebke−Blackburn−Bienayme (GBB)-

Multicomponent Reaction

Scheme 3. Synthesis of N-(2-Substituted Imidazo[1,2-

a]Pyridin-3-yl)amides

Reagents and conditions: (a) tert-butyl isocyanide, NH Cl, toluene,

and reﬂux; (b) 5 M HBr and 110 °C. (c) Pd(PPh ) , Na CO ,

Reagents and conditions: (a) Toluene/pyridine and rt. The table

PhB(OH) , DME/H O, and reﬂux. (d) Pd(PPh ) , Na CO , 2-

shows the structural details of the synthesized compounds 24−34.

Compounds 35 and 36 have been reported previously.

thienylB(OH) , DME/H O, and reﬂux.

732

J. Med. Chem. 2021, 64, 4730−4743

Journal of Medicinal Chemistry

enabling introduction of an array of substituents in the 2-

Article

Scheme 4. Synthesis of Amide Bioisosteres of DS2^a

position of the imidazo[1,2-a]pyridine core scaﬀold (Scheme

). The intermediate 21 was synthesized according to

procedures reported followed by a Suzuki−Miyaura cross-

coupling aﬀording the corresponding 2-phenyl analogue, 22, in

high yield (88%). To explore a broader use of 21 for further

structural elaboration, we showed that 21 readily undergoes

nucleophilic aromatic substitution with morpholine at room

temperature in high yield (not shown).

6,37

However, as others before us,

we encountered

diﬃculties in controlling the apparent substituent-dependent

outcome of the reduction of the nitro group. Moreover, the

isolation of the resulting amine, under the given reaction

conditions, proved complicated. In our hands, using tin(II)-

chloride in reﬂuxing methanol proved to be the most feasible

method and gave the desired amine 23.

Reagents and conditions: (a) 2-bromo-1-phenylethan-1-one,

NaHCO , MeOH, and reﬂux; (b) MeCN, NBS for X = Br, NIS for

X = I, and rt.; (c) TMSA, Pd(PPh ) Cl , CuI, Et N, and 80 °C; (d)

Finally, the 2-substituted imidazo-[1,2-a]pyridin-3-amides

24−34) were obtained from the corresponding amines, 10−

8 and 23, and the appropriate acyl chlorides in a mixture of

TBAF, THF, and rt.; (e) phenylazide, CuSO ·5H O, sodium

(

ascorbate, THF/H O, and rt.; (f) ethyl 3-oxo-3-phenylpropanoate,

CBr , MeCN, and 80 °C; (g) 4-chloro-N′-hydroxybenzimidamide,

DMSO, NaOH, and rt.

toluene and pyridine (Scheme 3). Multiple of the amides

proved troublesome to purify using standard column

chromatography puriﬁcation methods. However, classical

methods such as trituration or recrystallization aﬀorded the

pure target compounds.

substituted 2-aminopyridines, tert-butyl isocyanide, and

commercially available aliphatic or aromatic aldehydes

(Scheme 1). To introduce aromatic moieties into the pyridine

ring, the bromide in 1 was used as a handle in a Suzuki−

Miyaura cross-coupling reaction with commercially available

boronic acids, aﬀording 7 and 8 in decent yield (52−60%).

Subsequent acidic deprotection using hydrobromic acid

aﬀorded the intermediate amines 10−18.

To expand the ﬂexibility of the design and diversity of the

substituents, a complementary strategy was explored based on

a key intermediate, 21, containing a chloride as a handle

The amide bioisosteres, 1,2,3-triazole and 1,2,4-oxadiazole,

were obtained as illustrated in Scheme 4. The imidazo[1,2-

a]pyridine core scaﬀold of the intermediates 37 and 38 was

synthesized through an initial alkylation−condensation reac-

tion using 2-bromo-1-phenylethan-1-one.

To introduce a handle for the subsequent cross-coupling

reaction aﬀording 39, a regioselective halogenation using N-

bromo- or N-iodosuccinimide was performed to give 37 and

38 , respectively, in high yield. Subsequent Sonogashira cross-

Figure 3. Initial testing of agonists and PAM activity of compounds 24−34 and 40 at α4β1δ and compounds 30−34 at α4β1γ2 receptors. Agonist

activity of 10 μM compounds at (A) α4β1δ and (C) α4β1γ2. PAM activity induced by 10 μM compounds at (B) α4β1δ and (D) α4β1γ2 in the

presence of GABA (EC , see Table 1). (E) Concentration−response curve showing the PAM activity of 30−32 at α4β1δ. The data are shown as

means ± SD. Data in (A), (B), and (E) are representative of two−three independent experiments, and data shown in (C) and (D) are from a single

experiment. The dotted line represents the response from GABA EC coapplied with the analogues, when testing the PAM activity in (B), (D),

and (E). Selected compounds investigated further are highlighted in blue.

733

J. Med. Chem. 2021, 64, 4730−4743

Journal of Medicinal Chemistry

Article

coupling followed by a TBAF-mediated deprotection gave

access to the terminal alkyne. Interestingly, we found that the

bromo analogue, 37, was superior to the iodo analogue, 38, as

a cross-coupling partner, according to the yield of the

compounds obtained, as dehalogenation was observed when

using the latter. Finally, the 1,2,3-triazole was installed through

a copper-catalyzed alkyne-azide cycloaddition, which furnished

Table 1. PAM Potencies of Selected DS2 Analogues

compound

EC₅₀(μM), [pEC₅₀± SEM], n

α4β1δ

α4β1γ2

α1β2γ2

DS2

0.98

0.16

1.10

1.92

0.14

0.060

[6.01 ± 0.078], 4

[6.79 ± 0.082], 4

[5.96 ± 0.063], 3

[5.72 ± 0.046], 3

[6.84 ± 0.140], 5

[7.22 ± 0.079], 4

>10; n = 3

n = 2

0 in 78% yield.

The ester 41 was prepared as previously described using

ethyl 3-oxo-3-phenylpropanoate and subsequent cycloconden-

n = 2

sation with 4-chloro-N′-hydroxybenzimidamide aﬀording the

Concentration−response curves were obtained in the FMP assay.

,2,4-oxadiazole 42.

Compounds were coapplied with a concentration of GABA

corresponding to approximately GABA EC20 determined from full

GABA concentration−response curves at the respective receptors.

SAR of the Target Compounds at α4β1δ and α4β1γ2

GABA Rs. The compounds were subjected to functional

Concentrations used to induce EC response were α4β1δ: 0.06 μM,

characterization in the FMP assay. Initially, the compounds

were tested in a single concentration of 10 μM as agonists at

α4β1γ2: 0.6 μM, and α1β2γ2: 0.5 μM. No potentiation in the tested

concentration range of 0.01−10 μM. Inactive (potentiation was less

than 10% of the GABA max in the tested concentration range of

α4β1δ and α4β1γ2 GABA Rs and as PAMs (applied together

with a concentration of GABA corresponding to EC ), to

.01−10 μM). Not tested.

evaluate intrinsic activity as well as selectivity (Figure 3).

The compounds 27, 29, and 34 were found to have no

activity at α4β1δ and were therefore not pursued further. In

addition, the poor solubility of the 1,2,4-oxadiazole (42)

precluded the compound from pharmacological character-

ization. Compounds 24−26 and 33 demonstrated agonist

activity as these compounds induced responses at α4β1δ in the

absence of GABA, making these compounds ago-PAMs.

However, because we were primarily interested in PAMs,

these were not characterized further. Instead, compounds 28

and 30−32 were singled out as displaying no or limited agonist

activity (Figure 3B). The most eﬃcacious of these, 30−32,

were further indicated to display some selectivity for α4β1δ

over α4β1γ2 (Figure 3D). The bioisosteric replacement of the

amide with a triazole was tolerated as 40 revealed PAM activity

at α4β1δ, although seemingly with a relatively low eﬃcacy.

The two most eﬃcacious agonists, 24 and 33, in this study

display potent modulatory eﬀects as well (Figure 3A,B).

However, the SAR obtained for the modulatory eﬀect on

GABA activity, in this speciﬁc study and described

responses at γ-containing receptors for some of our

compounds (Table 1) may seem contradictory to a binding

site in the α δ subunit interface. However, considering the

possibility of weak binding to a corresponding binding site in

an α1/4γ2S subunit interface, the resulting SAR was still

interpreted in reference to our proposed binding model for

DS2 in the α4δ subunit interface (Figure 2). The 5-position

(

R ) in the pyridine part of the scaﬀold (Figure 2C) was

investigated in most detail and proved to be important for the

δ-selectivity of the compounds. This is evident as 30−32,

containing a methyl, triﬂuoromethyl, or a ﬂuorine at this

position, potently potentiated α4β1δ responses, while exhibit-

ing no or negligible activity at α4β1γ2 (Table 1). Surprisingly,

the iodinated compound 33, as the only one of this series, was

shown to be a dual allosteric agonist and modulator at both

α4β1δ and α4β1γ2 (Figure 3 and Table 1), which is not the

typical signature for a ligand binding in the ECD

benzodiazepine binding site. Additionally, as demonstrated

by the apparently diﬀerential eﬀects on potentiation versus

activation (Figure 3; 33 vs 30 and 24 vs 25), the 5-position

previously, seems to diﬀer from the corresponding SAR for

the allosteric agonist eﬀect. This is especially pronounced for

the 5-substituted analogues 28 and 30−32, showing a

signiﬁcant modulatory eﬀect but no direct agonist eﬀect.

Next, we generated full concentration−response curves of

the 5-methyl, 5-triﬂuoromethyl, and 5-ﬂuoro DS2 analogues

together with R substituents may be used to optimize the

PAM activity over allosteric agonism or vice versa. In contrast

to the markedly diminished allosteric agonist eﬀect, the

retained PAM activity of 25 with the 4-OPh instead of the

4-Cl in the parent compound, DS2, is consistent with our

predicted binding model showing additional available space for

0−32 in the presence of GABA (EC ) (exempliﬁed in Figure

E, EC values for 30−32 collected in Table 1, and

representative raw data in Figure S3 ). In addition, because of

signiﬁcant structural overlap, we included the previously

identiﬁed 5-chloro and 5-bromo DS2 analogues, 35 and 36,

substitution (Figure 2B), and this clearly indicates further R -

substitution possibilities and warrants additional future

investigation of other and even bigger and more bulky

in this series as well. The determined EC₅₀values of the

PAM activities are shown in Table 1 and representative raw

data in Figure S3 . The methyl analogue 30 was found to stand

out both in terms of potency and selectivity. It displayed a

PAM potency of 160 nM (EC ) (Figure 3E), which is

substituents in this position. For the 5-position (R ), a ﬁve-

membered ring seem to be the largest allowed, as the 2-thienyl

(28) retains PAM activity while a phenyl substituent (27) no

longer potentiates the GABA EC response (Figure 3). Again,

these results are consistent with our model showing available

space at the 5-position (Figure 2A).

approximately 6 times higher than that of DS2 itself at α4β1δ.

Αdditionally, 30 showed no activity at 10 μM at α4β1γ2 (>60-

fold selectivity for α4β1δ). Compounds 35 and 36 were

equally potent as 30 but lacked the same degree of selectivity.

Compounds 31 and 32 were neither more potent than DS2

itself at α4β1δ, nor did they match the selectivity observed for

DS2.

Though only a single compound explores R substitution

(34), the lack of eﬀect with a bromo-substituent in this speciﬁc

position corroborates with the fairly tight ﬁt between DS2 and

the receptor in our model, predicting very limited space for

substitution. However, when it comes to the lack of activity for

29 (Figure 3), which has an R phenyl substituent, our model

Structural Rationalization of Major SAR Observa-

tions. The fact that we observe slight potentiation of GABA

fails to provide an explanation. No apparent speciﬁc

interactions were obtained in the docking, and there seems

734

J. Med. Chem. 2021, 64, 4730−4743

Journal of Medicinal Chemistry

containing receptor subtypes.

Article

to be ample space for the slightly larger phenyl compared to

the 2-thienyl. However, it is tempting to speculate from our

pharmacological results that this part of the compounds should

bind in a tight pocket encompassing a ﬁve-membered ring, also

nonaromatic (26), but no larger than that. On the other hand,

compound 40, which shows moderate PAM activity, has a

phenyl in the corresponding position on a scaﬀold with a

triazole bioisostere replacing the amide. Thus, it could also be

EXPERIMENTAL SECTION

■

Chemistry. General Procedures. Melting points (mp) were

determined on an SRS OptiMelt apparatus with open capillary

tubes. The melting points are uncorrected. Triethylamine was stored

over KOH pellets. Thin layer chromatography (TLC) was performed

^oⁿ^p^r^e^c^o^a^t^e^d^sⁱ^lⁱ^c^a⁶⁰^g^e^l^F254 from Merck and visualized using a UV-

lamp at 254/366 nm. Flash column chromatography was conducted

manually in glass columns loaded with silica gel 60 (40−63 μm)

supplied by Merck. An internal solvent puriﬁcation system provided

anhydrous solvents such as THF, DCM, and DMF. PET refers to the

fraction boiling at 40−60 °C. Nuclear magnetic resonance (NMR)

spectra were obtained at 300 K on either of two systems. A Bruker

Avance II 400 MHz spectrometer equipped with a 5 mm broad band

BBFO probe was used. Also, a Bruker Avance III HD 600 MHz

spectrometer equipped with a 5 mm cryogenically cooled CPDCH

the common absence of an R substituent that causes the lack

of potentiation by 29 and 34. Further studies are needed to

investigate this. Additionally, the triazole bioisoteric replace-

ments of the amide (40) conﬂict with the binding mode of

DS2 as it is unable to form both of the charge-assisted

hydrogen bonds observed for the amide to α4R135 and δE71,

bridging the two subunits across the interface.

Based on the above SAR analysis (summarized in Figure 4),

where the pharmacological activity of the majority of the

compounds agrees, but a single compound contradicts our

binding mode, it remains nonconclusive whether this is the site

of action for DS2. To further investigate this hypothesis,

structural-guided site-directed mutagenesis studies are cur-

rently being analyzed in our lab.

C( H)Z-GRD probe was used. 16−64 scans were collected for H-

NMR spectra, depending on the concentration, with a relaxation delay

of 1.0 s. 256−3072 scans were collected for C-NMR spectra using a

relaxation delay of 2.0 or 4.0 s. The solvent peak was used as the

reference in which CDCl (7.26 and 77.16 ppm), MeOD-d (3.31 and

49.00 ppm), and DMSO-d (2.50 and 39.52 ppm) were applied,

respectively, for H- and C-NMR. The obtained spectra were

processed such as phase- and baseline-corrected in MNova software,

version 11.0. The spectral data are reported in the given order:

Chemical shift (δ), multiplicity (singlet (s), doublet (d) double

doublet (dd), triplet (t), quartet (q), pentet (p), triplet of doublets

Irrespective of our binding site hypothesis, our novel

analogues of DS2 show that the 5-position of the scaﬀold is

clearly vital for the pharmacological proﬁle of the analogues

(

td), doublet of doublet of doublets (ddd), broad (b) and multiplet

m)), coupling constant(s) J(Hz), and number of protons. The purity

of the ﬁnal compounds (≥95% unless otherwise stated) was

determined on either of two high-performance liquid chromatography

HPLC) systems. The samples were dissolved in 1 mL of (MeCN/

H O)(1:1) or DMSO in cases in which the solubility of the

compounds was found to be insuﬃcient in (MeCN/H O)(1:1) and

(

PAM/ago-PAM) as well as for the preference for α4β1δ

versus α4β1γ2. Such dual activity is a known phenomenon,

(

where subtle structural diﬀerences result in diverse pharmaco-

logical activity proﬁles and selectivity as previously reported

42,43

within the cys-loop receptor research ﬁeld.

to avoid precipitation of the compounds on the column. HPLC

system 1 was a LaChrome reversed phase system from Merck,

Germany, Darmstadt that had a Chromolith SpeedROD RP-18

column (4.6 × 50 mm). Detection was performed at 254 nm via an L-

CONCLUSIONS

■

In summary, we have synthesized a series of novel DS2

analogues through complementary synthesis strategies allowing

structural diversiﬁcation of the imidazopyridine core scaﬀold

and identiﬁed novel compounds with better potency at, and

400 UV detector. The samples were injected using an L-7200 auto

sampler, injecting 1−10μL. An L7100 pump operating with either a

ﬂow rate of 3 or 4 mL/min was used. A linear gradient of A (99.9%

H O/0.1% TFA) and B (90% MeCN/10% H O/0.1% TFA) from

higher selectivity for, δ-containing GABA Rs. We ﬁnd that

100% A to 90% B over 5 min was applied. The system used

EZChromeElite software for data processing and collection. HPLC

system 2 was a Dioxnex Ultimate HPLC system, which had an LPG-

substituents in the 5-position of the imidazopyridine core

scaﬀold severely aﬀect both the δ-selectivity and modulatory

activity. Interestingly, the 5-methyl, 5-bromo, and 5-chloro

DS2 analogues, 30, 35, and 36, are superior to DS2 at α4β1δ

as PAMs with 6- to 16 times increased potency, whereas the

structurally closely related 5-iodo analogue, 33, is an ago-PAM.

This illustrates how even small diﬀerences in the imidazopyr-

idine core scaﬀold of DS2 can give rise to diverse

pharmacological proﬁles. Compound 30 provides a useful

tool with improved sensitivity and selectivity for studying the

speciﬁc subtype α4β1δ, recently proposed to have a functional

3400A pump with a ﬂow rate of 1 mL/min, using the previously

described mobile phases A and B. A linear gradient system from 100%

A to 100% B over 15 min was used. 6 min column equilibration was

performed after each run. The column was a Germini-NX C18

column (4.6 × 250 mm, 3μm, 110 Å). A WPS-3000SL auto sampler

was used, injecting 1−10 μL. This system had a DAD3000D diode

array detector set to 225, 240, 254, and 290 nm. Data processing and

collection were conducted using chromeleon software version 6.80.

Ultraperformance liquid chromatography-electrospray ionization-

mass spectrometry (UPLC-ESI-MS) was performed on an Agilent

100 HPLC system equipped with a C-18 column (2.1 mm × 50 mm

role in the hippocampus and possibly toward other δ-

1.7 μm), which was coupled to a Hewlett Packard 1100 series mass

spectrometer with an electrospray ionization source. MassLynx mass

spectrometry software version 4.1 was used for data processing. The

samples were dissolved in 1 mL of (MeCN/H O) (1:1) and ﬁltered

through a 0.22 μm ﬁlter. The samples were injected via an Acquity

FTN Autosampler. A ﬂow rate of 0.8 mL/min was applied. A linear

gradient system of A (MeCN/H O/HCOOH) (0.05:0.95:0.01) and

B (MeCN/HCOOH) (0.99:0.01), which rose from 100% A to 100%

B over 3.5 min and then 1 min at 100%, maintaining a ﬂow rate of 0.8

mL/min.

General GBB Procedure for N-(tert-butyl)-imidazo[1,2-a]-

pyridin-3-amine Derivatives. The aldehyde (1 equiv) was

Figure 4. Summarized SAR for the DS2 scaﬀold in terms of δ-

selective allosteric modulation.

dissolved in toluene (40−100 mL). NH Cl (1 equiv), tert-butyl

isocyanide (1.2 equiv), and 2-aminopyridine (1 equiv) were then

735

J. Med. Chem. 2021, 64, 4730−4743

Journal of Medicinal Chemistry

Article

added in that order. The reaction mixture was heated to reﬂux under

argon and monitored by TLC. The reaction mixture was cooled to

room temperature, reduced in vacuo, and reevaporated with 10 mL

toluene to yield the crude product, which was recrystallized to aﬀord

the pure N-(tert-butyl)-imidazo[1,2-a]pyridin-3-amines.

4.8 Hz). ESI-MS (m/z) calculated [M + H]⁺for C H FN S =

15 17 3

290.11, found 290.1.

N-(tert-butyl)-6-iodo-2-(thiophen-2-yl)imidazo[1,2-a]-

pyridin-3-amine (5). Obtained from 2-thiophenecarboxaldehyde

(

0.4 mL, 4.3 mmol), NH Cl, (0.24 g, 4.5 mmol), tert-butyl isocyanide

0.8 mL, 7.2 mmol), and 2-amino-5-iodopyridine (1.00 g, 4.5

N-(tert-butyl)-6-bromo-2-(thiophen-2-yl)imidazo[1,2-a]-

pyridin-3-amine (1). Obtained from 2-thiophenecarboxaldehyde

mmol), using the general GBB procedure. The reaction mixture was

reﬂuxed for 6 h. The crude product was recrystallized from MeCN/

(

0.7 mL, 7.1 mmol), NH Cl (0.38 g, 7.1 mmol), tert-butyl isocyanide

(1.0 mL, 8.4 mmol), and 2-amino-5-bromopyridine (1.24 g, 7.1

H O (6: 1), which furnished the product as a white solid (1.18 g, 69%

mmol), using the general GBB procedure. Additional tert-Butyl

isocyanide (0.2 mL, 1.8 mmol) was added after 18 h. The reaction

mixture was reﬂuxed for 48 h in total. The crude product was

yield): mp 180−181 °C. R = 0.7 (PET/EtOAc) (1: 1). H NMR

(400 MHz, Chloroform-d) δ 8.44 (t, J = 1.3 Hz, 1H), 7.59 (d, J = 2.6

Hz, 1H), 7.37−7.28 (m, 3H), 7.11 (dd, J = 5.1, 3.6 Hz, 1H), 3.09 (s,

H), 1.19 (s, 9H). C NMR (151 MHz, Chloroform-d) δ 140.73,

recrystallized from MeCN/H O (16:1). The resulting solid was

37.03, 135.30, 132.30, 128.91, 127.52, 125.43, 125.31, 122.80,

18.37, 74.49, 57.01, 30.70. ESI-MS (m/z) calculated [M + H] for

lyophilized to yield the product as white needles (2.31 g, 93%): mp

84 °C, R = 0.5 (DCM : EtOAc) (20 : 1). H NMR (400 MHz,

C H IN S = 398.02, found 398.0.

Chloroform-d) δ 8.34−8.29 (m, 1H), 7.56 (dd, J = 1.0 Hz, 1H),

15 17

-bromo-N-(tert-butyl)-2-(thiophen-2-yl)imidazo[1,2-a]-

.44−7.38 (m, 1H), 7.33 (dd, J = 1.0 Hz, 1H), 7.17 (dd, J = 2.0 Hz,

pyridin-3-amine (6). Obtained from 2-thiophenecarboxaldehyde

H), 7.10 (dd, J = 3.7 Hz, 1H), 3.09 (s, 1H), 1.19 (s, 9H). C NMR

(

0.4 mL, 4.3 mmol), NH Cl, (0.23 g, 4.3 mmol), tert-butyl isocyanide

0.60 mL, 5.2 mmol), and 2-amino-4-bromopyridine (0.74 g, 4.3

(151 MHz, Chloroform-d) δ 139.68, 136.21, 134.87, 126.80, 126.52,

24.40, 124.25, 122.89, 122.27, 117.04, 105.51, 56.02, 29.70. ESI-MS

mmol), using the general GBB procedure. The reaction mixture was

(m/z) calculated [M + H] for C H BrN S = 350.03, found 350.0

15 16 3

reﬂuxed for 38 h. The crude product was recrystallized from MeOH/

and [M + H + 2] = 352.0.

N-(tert-butyl)-6-methyl-2-(thiophen-2-yl)imidazo[1,2-a]-

pyridin-3-amine (2). Obtained from 2-thiophenecarboxaldehyde

H O (10: 1) to furnish the product as oﬀ-white needles (0.87 g, 52%

yield). This product was slightly contaminated with impurities and

was used in the subsequent reaction without further puriﬁcation. mp

(

0.6 mL, 6.0 mmol), NH Cl (0.32 g, 6.0 mmol), tert-butyl isocyanide

72−173 °C. R = 0.5 (PET/EtOAc) (2: 1). H NMR (400 MHz,

(0.8 mL, 7.1 mmol), and 2-amino-5-methylpyridine (0.65 g, 6.0

Chloroform-d) δ 8.06 (dd, J = 7.3, 0.8 Hz, 1H), 7.70 (s, 1H), 7.57 (d,

J = 3.6 Hz, 1H), 7.33 (dd, J = 5.1, 1.1 Hz, 1H), 7.11 (dd, J = 5.1, 3.6

Hz, 1H), 6.85 (dd, J = 7.2, 1.9 Hz, 1H), 3.09 (s, 1H), 1.18 (s, 9H).

mmol) using the general GBB procedure. The reaction mixture was

reﬂuxed for 23 h. The crude product was recrystallized from MeCN/

H O (3: 10), which furnished the product as a gray crystalline solid,

ESI-MS (m/z) calculated [M + H + ] for C H BrN S = 350.03,

(

1.03 g, 60% yield): mp 193−194 °C. R = 0.6 (PET/EtOAc) (1: 1).

found 350.0 and [M + H + 2] = 352.1.

H NMR (400 MHz, Chloroform-d) δ 7.98−7.95 (m, 1H), 7.56 (dd,

J = 3.6, 1.2 Hz, 1H), 7.42 (dd, J = 9.1, 1.0 Hz, 1H), 7.29 (dd, J = 5.1,

General Suzuki−Miyaura Cross-Coupling Procedure. An aryl

halide, such as 1 (1.2 g, 3.4 mmol, 1 equiv), was dissolved in DME

.2 Hz, 1H), 7.09 (dd, J = 3.6 Hz, 1H), 6.97 (dd, J = 9.1, 1.7 Hz, 1H),

.06 (s, 1H), 2.33 (s, 3H), 1.18 (s, 9H). C NMR (151 MHz,

(

75 mL) and water (38 mL). A boronic acid, such as phenylboronic

acid (0.498 g, 4.1 mmol, 1.2 equiv), Na CO (0.723 g, 6.8 mmol, 2.0

Chloroform-d) δ 141.46, 138.02, 134.83, 127.57, 127.37, 124.76,

equiv), and Pd(PPh ) (0.393 g, 0.34 mmol, 10 mol %) were added,

24.70, 122.77, 121.35, 121.07, 116.73, 56.84, 30.70, 18.57.

N-(tert-butyl)-2-(thiophen-2-yl)-6-(triﬂuoromethyl)imidazo-

respectively. The ﬂask was ﬁtted with a rubber septum and purged

with argon. The reaction mixture was then heated to reﬂux until

completion of the reaction monitored by HPLC, cooled to room

temperature, and diluted with water (70 mL) and DCM (100 mL).

The aqueous layer was extracted with DCM (100 + 200 mL). The

[1,2-a]pyridin-3-amine (3). Obtained from 2-thiophenecarboxalde-

hyde (0.9 mL, 9.3 mmol), NH Cl, (0.50 g, 9.3 mmol), tert-butyl

isocyanide (1.3 mL, 11 mmol), and 5-(triﬂuoromethyl)pyridin-2-

amine (1.51 g, 9.3 mmol), using the general GBB procedure. The

reaction mixture was reﬂuxed for 41 h. The crude product was

combined organic layers were washed with H O (110 mL), dried over

Na SO , and reduced in vacuo. The crude product was dissolved in

recrystallized from MeCN/H O (2: 1), which furnished the product

DCM, concentrated onto celite, and puriﬁed by column chromatog-

raphy to yield the product and/or puriﬁed by recrystallization.

N-(tert-butyl)-6-phenyl-2-(thiophen-2-yl)imidazo[1,2-a]-

pyridin-3-amine (7). Obtained from 1 (1.2 g, 3.4 mmol) and

phenylboronic acid (0.498 g, 4.1 mmol), using the general Suzuki−

Miyaura cross-coupling procedure. Puriﬁcation of the crude product

was performed by column chromatography (silica gel 60, 40−63 μM)

as beige to orange needles (2.29 g, 72% yield): mp 153−154 °C. R =

.5 (PET/EtOAc) (2: 1). H NMR (400 MHz, Chloroform-d) δ

.61−8.55 (m, 1H), 7.64−7.56 (m, 2H), 7.36 (dd, J = 5.1, 1.1 Hz,

H), 7.29−7.24 (m, 1H), 7.13 (dd, J = 5.1, 3.6 Hz, 1H), 3.16 (s, 1H),

.20 (s, 9H). ¹³C NMR (151 MHz, Chloroform-d) δ 141.85, 136.71,

36.65, 127.57, 125.72, 125.63, 123.99, 123.90 (q, J = 270.9 Hz),

22.80 (q, J = 5.8 Hz), 120.29 (q, J = 2.7 Hz), 117.85, 115.97 (q, J =

(

DCM/EtOAc) (20: 1), which furnished the product as a white solid,

4.0 Hz), 57.05, 30.63. F NMR (376 MHz, Chloroform-d) δ

0.62 g, 52% yield): mp 147−148°C. R = 0.3 (DCM/EtOAc) (20:

−

62.08. ESI-MS (m/z) calculated [M + H] for C H F N S =

1). R = 2.08 min. (4 mL/min ﬂow rate). H NMR (400 MHz,

40.11, found 340.1.

N-(tert-butyl)-6-ﬂuoro-2-(thiophen-2-yl)imidazo[1,2-a]-

pyridin-3-amine (4). Obtained from 2-thiophenecarboxaldehyde

Chloroform-d) δ 8.38 (dd, J = 1.2 Hz, 1H), 7.63−7.54 (m, 4H),

.53−7.44 (m, 2H), 7.43−7.37 (m, 2H), 7.32 (dd, J = 5.1, 1.2 Hz,

H), 7.11 (dd, J = 5.1, 3.6 Hz, 1H), 3.14 (s, 1H), 1.23 (s, 9H). C

(

0.9 mL, 9.3 mmol), NH Cl, (0.50 g, 9.4 mmol), tert-butyl isocyanide

1.3 mL, 11 mmol, 1.2 equiv), and 2-amino-5-ﬂuoropyridine (1.04 g,

NMR (151 MHz, Chloroform-d) δ 141.72, 137.96, 137.79, 135.59,

29.25, 127.83, 127.45, 126.97, 125.88, 125.09, 125.02, 124.94,

23.34, 121.02, 117.21, 56.93, 30.79.

N-(tert-butyl)-2,6-di(thiophen-2-yl)imidazo[1,2-a]pyridin-3-

9.3 mmol), using the general GBB procedure. The reaction mixture

was heated for 37 h. The crude product was recrystallized from

MeCN/H O (1.2: 1), to furnish the product as beige needles, (1.77 g,

amine (8). Obtained from 1 (0.83 g, 2.4 mmol) and 2-thienylboronic

acid (0.37 g, 2.9 mmol), using the general Suzuki−Miyaura cross-

coupling procedure. Puriﬁcation of the crude product was performed

by column chromatography (silica gel 60, 40−63 μM) (DCM/

MeOH)(40: 1 MeOH). The product was obtained as an oﬀ-white

solid. The impure fractions containing the product were combined,

reduced in vacuo and recrystallized from MeCN, and washed with

cold MeCN and cold water to yield the product as oﬀ-white needles

6% yield): mp 188−189 °C. R = 0.4 (PET: EtOAc) (2: 1). H

NMR (400 MHz, Chloroform-d) δ 8.12 (dd, J = 4.4, 2.5 Hz, 1H),

.56 (dd, J = 3.6, 1.1 Hz, 1H), 7.49 (dd, J = 9.7, 5.0 Hz, 1H), 7.33

dd, J = 5.1, 1.1 Hz, 1H), 7.10 (dd, J = 5.1, 3.6 Hz, 1H), 7.04 (ddd, J

(

10.0, 7.8, 2.5 Hz, 1H), 3.10 (s, 1H), 1.19 (s, 9H). C NMR (151

MHz, Chloroform-d) δ 153.21 (d, J = 236.0 Hz), 139.81, 137.33,

136.48 (d, J = 2.3 Hz), 127.47, 125.24, 125.10, 124.30, 117.78 (d, J =

8.8 Hz), 116.44 (d, J = 25.9 Hz), 110.30 (d, J = 41.5 Hz), 56.97,

0.67. ¹⁹F NMR (376 MHz, Chloroform-d) δ −140.64 (dt, J = 8.5,

(0.474 g, 60% yield): mp 204−205 °C. R = 0.4 (DCM/MeOH)(40:

1). R = 2.39 min (3 mL/min ﬂow rate) (>90% pure by HPLC system

736

J. Med. Chem. 2021, 64, 4730−4743

Journal of Medicinal Chemistry

Article

). H NMR (400 MHz, Chloroform-d) δ 8.43 (dd, J = 1.9, 1.0 Hz,

H), 7.60 (dd, J = 1.0 Hz, 1H), 7.53 (dd, J = 9.3, 0.9 Hz, 1H), 7.40

124.53 (q, J = 270.9 Hz), 124.27, 123.90, 122.51, 122.13 (q, J = 5.9

Hz), 117.16, 116.60 (q, J = 2.6 Hz), 113.76 (q, J = 33.1 Hz).

(

dd, J = 9.3, 1.9 Hz, 1H), 7.32 (ddd, J = 7.2, 5.0, 1.1 Hz, 2H), 7.28

dd, J = 3.6, 1.2 Hz, 1H), 7.11 (td, J = 3.8 Hz, 2H), 3.13 (s, 1H), 1.23

s, 9H). ¹³C NMR (151 MHz, Chloroform-d) δ 141.57, 140.85,

NMR (376 MHz, DMSO-d ) δ −60.43. ESI-MS (m/z) calculated [M

+ H] for C H F N S = 284.05, found 284.0.

6-ﬂuoro-2-(thiophen-2-yl)imidazo[1,2-a]pyridin-3-amine

(13). Obtained from 4 (0.40 g, 1.4 mmol) using the general tert-butyl

deprotection procedure. The reaction mixture was cooled in an ice

bath upon adjustment of pH with 5 M NaOH. The product was

isolated as a greenish crystalline solid (0.278 g, 85% yield): mp 201−

37.65, 135.72, 128.33, 127.48, 125.10, 124.99, 124.97, 124.27,

23.73, 123.40, 119.91, 119.89, 117.38, 56.94, 30.81. ESI-MS (m/z)

calculated [M + H] for C H N S = 354.11, found 354.1.

3 2

-bromo-N-(tert-butyl)-2-cyclopentylimidazo[1,2-a]pyridin-

-amine (9). Obtained from cyclopentanecarbaldehyde (1.0 mL, 9.3

mmol), NH Cl, (0.50 g, 9.3 mmol), tert-butyl isocyanide (1.3 mL,

202 °C (decomp.). R = 0.6 (PET/EtOAc) (1: 3) + 1% Et N. H

NMR (400 MHz, Chloroform-d) δ 7.96−7.93 (m, 1H), 7.56 (dd, J =

3.7, 1.1 Hz, 1H), 7.51−7.45 (m, 1H), 7.34 (dd, J = 5.1, 1.1 Hz, 1H),

7.14 (dd, J = 5.1, 3.6 Hz, 1H), 7.02 (ddd, J = 10.1, 8.0, 2.4 Hz, 1H),

1.5 mmol), and 2-amino-5-bromopyridine (1.61 g, 9.3 mmol) using

the general GBB procedure. The reaction mixture was reﬂuxed for 46

h. The crude product was recrystallized from MeCN/H O (3: 1),

which furnished the product as large, slightly gray crystals (2.10 g,

3.33 (s, 2H). C NMR (151 MHz, Chloroform-d) δ 153.52 (d, J =

236.4 Hz), 138.95, 137.14, 132.00 (d, J = 2.0 Hz), 127.88, 124.96,

123.89, 122.71, 117.83 (d, J = 8.9 Hz), 115.76 (d, J = 25.9 Hz),

7% yield): mp 143 °C. R = 0.6 (PET/EtOAc) (2: 1). H NMR (400

MHz, Chloroform-d) δ 8.26 (dd, J = 1.9, 0.8 Hz, 1H), 7.36 (d, J = 9.4

Hz, 1H), 7.11 (dd, J = 9.4, 2.0 Hz, 1H), 3.23−3.11 (m, 1H), 2.75 (s,

108.74 (d, J = 41.4 Hz). F NMR (376 MHz, Chloroform-d) δ

−140.25 (q, J = 5.4 Hz). ESI-MS (m/z) calculated [M + H] for

H), 2.00−1.90 (m, 6H), 1.76−1.63 (m, 2H), 1.21 (s, 9H). C NMR

C H FN S = 234.05, found 234.0.

(101 MHz, Chloroform-d) δ 146.08, 140.70, 126.54, 123.59, 123.35,

6-iodo-2-(thiophen-2-yl)imidazo[1,2-a]pyridin-3-amine

17.66, 105.68, 55.39, 37.72, 33.92, 30.48, 26.24. ESI-MS (m/z)

(14). Obtained from 5 (0.450 g, 1.1 mmol) using the general tert-

butyl deprotection procedure. The reaction mixture was cooled in an

ice bath upon adjustment of pH with 5 M NaOH. The product was

calculated [M + H] for C H BrN = 336.1 found 336.1 and [M + H

2] = 338.2.

General tert-Butyl Deprotection Procedure. The tert-butyl-

isolated as a yellow crystalline solid (0.322 g, 83% yield). R

= 0.7

(Heptane/EtOAc) (1: 3) + 1% Et N. H NMR (400 MHz,

protected imidazo[1,2-a]pyridin-3-amine was suspended in 10−15

mL of 5 M aqueous HBr. The reaction mixture was heated to 110 °C

until completion monitored by HPLC and/or TLC and cooled to

room temperature, and the pH of the reaction mixture was adjusted to

pH ∼12−14 via aqueous 5 M NaOH. For some compounds, the

reaction mixture was cooled in an ice bath during this adjustment of

the pH as speciﬁed for relevant compounds. The aqueous layer was

extracted with EtOAc (3 × 50 mL). The combined organic layers

were dried over Na SO and subsequently reduced in vacuo to yield

Chloroform-d) δ 8.34−8.25 (m, 1H), 7.57 (dd, J = 3.6, 1.1 Hz,

1H), 7.35 (dd, J = 5.1, 1.1 Hz, 1H), 7.33−7.27 (m, 2H), 7.14 (dd, J =

5.1, 3.6 Hz, 1H), 3.33 (s, 2H). C NMR (151 MHz, DMSO-d ) δ

138.20, 136.83, 128.88, 127.90, 126.93, 125.53, 123.92, 123.13,

122.12, 117.54, 74.89.ESI-MS (m/z) calculated [M + H]⁺for

7-bromo-2-(thiophen-2-yl)imidazo[1,2-a]pyridin-3-amine

15). Obtained from 6 (0.40 g, 1.2 mmol,1 equiv) using the general

S = 341.96, found 341.9.

(

tert-butyl deprotection procedure. The reaction mixture was cooled in

an ice bath upon adjustment of pH with 5 M NaOH. The product was

isolated as an orange solid (0.313 g, 87% yield). This product was

products as solids. The product was used in the subsequent reaction

without further puriﬁcation unless otherwise speciﬁed.

-bromo-2-(thiophen-2-yl)imidazo[1,2-a]pyridin-3-amine

10). Obtained from 1 (0.71 g, 2.0 mmol), using the general tert-butyl

deprotection procedure and was isolated as a yellow crystalline solid

0.58 g, 98% yield): mp 194−195 °C. R = 0.5 (Heptane/EtOAc) (3:

(

used in the subsequent reactions without further puriﬁcation. R = 0.6

(PET/EtOAc) (1: 1.5). H NMR (400 MHz, Chloroform-d) δ 7.91

(d, J = 7.2 Hz, 1H), 7.72 (s, 1H), 7.59 (d, J = 3.6 Hz, 1H), 7.36 (dd, J

= 5.1, 1.1 Hz, 1H), 7.15 (dd, J = 5.1, 3.6 Hz, 1H), 6.92 (dd, J = 7.2,

(

). R = 1.27 min. (4 mL/min ﬂow rate). H NMR (400 MHz,

Chloroform-d) δ 8.16 (dd, J = 2.0, 0.9 Hz, 1H), 7.57 (dd, J = 3.6, 1.2

Hz, 1H), 7.41 (dd, J = 9.4, 0.9 Hz, 1H), 7.35 (dd, J = 5.1, 1.1 Hz,

1.8 Hz, 1H), 3.35 (s, 2H). ESI-MS (m/z) calculated [M + H] for

BrN

S = 293.97, found 293.9 and [M + H + 2] = 295.9.

H), 7.18−7.12 (m, 2H), 3.34 (s, 2H). C NMR (151 MHz,

6-phenyl-2-(thiophen-2-yl)imidazo[1,2-a]pyridin-3-amine

16). Obtained from 7 (0.56 g, 1.6 mmol, 1 equiv) and 5 M aq. HBr

70 mL), using the general tert-butyl deprotection procedure. The

product was isolated as a yellow solid (0.435 g, 92% yield): mp 189−

(

Chloroform-d) δ 139.76, 136.91, 131.33, 127.93, 127.21, 125.17,

24.16, 122.38, 121.63, 117.98, 106.87. ESI-MS (m/z) calculated [M

H] for C H BrN S = 293.97, found 293.9 and [M + H + 2] =

11 9 3

91 °C. R = 0.7 (PET/EtOAc) (1: 3). H NMR (600 MHz,

95.9.

Chloroform-d) δ 8.21 (dd, J = 1.9, 1.0 Hz, 1H), 7.64−7.57 (m, 4H),

-methyl-2-(thiophen-2-yl)imidazo[1,2-a]pyridin-3-amine

(11). Obtained from 2 (0.400 g, 1.4 mmol), using the general tert-

7.48 (dd, J = 8.5, 6.9 Hz, 2H), 7.42−7.39 (m, 2H), 7.35 (dd, J = 5.1,

butyl deprotection procedure. Short reaction time (1.5 h) adjustment

of pH at 0 °C proved vital. The product was isolated as a yellow

1.1 Hz, 1H), 7.16 (dd, J = 5.1, 3.6 Hz, 1H), 3.39 (s, 2H). C NMR

(151 MHz, Chloroform-d) δ 140.71, 137.76, 137.43, 130.71, 129.22,

crystalline solid (0.319 g, 99% yield): mp = 174−175 °C. R = 0.3

127.91, 127.85, 127.00, 126.40, 124.74, 124.55, 123.78, 121.91,

(Heptane/EtOAc) (1: 2). R = 1.3 min. (4 mL/min ﬂow rate). H

119.36, 117.10. ESI-MS (m/z) calculated [M + H] for C17

S =

NMR (400 MHz, Chloroform-d) δ 7.79 (t, J = 1.6 Hz, 1H), 7.56 (t, J

0.9 Hz, 1H), 7.43 (d, J = 9.1 Hz, 1H), 7.31 (dd, J = 0.9 Hz, 1H),

.14 (d, J = 3.7 Hz, 1H), 6.96 (d, J = 1.6 Hz, 1H), 3.32 (s, 2H), 2.35

s, 3H). ¹³C NMR (151 MHz, Chloroform-d) δ 140.49, 137.67,

292.09, found 292.1.

(

2,6-di(thiophen-2-yl)imidazo[1,2-a]pyridin-3-amine (17).

Obtained from 8 (0.110 g, 0.30 mmol 1 equiv) using the general

tert-butyl deprotection procedure, 5 M aq. HBr (26 mL) and was

29.92, 127.80, 127.02, 124.48, 123.53, 121.66, 121.32, 119.73,

isolated as a yellow crystalline solid (0.087 g, 94% yield): mp 197−

16.71, 18.52. ESI-MS (m/z) calculated [M + H] for C H N S =

198 °C. R

= 0.6 (PET/EtOAc) (1: 3). R

= 1.95 min (3 mL/min). H

30.7, found 230.5.

-(thiophen-2-yl)-6-(triﬂuoromethyl)imidazo[1,2-a]pyridin-

-amine (12). Obtained from 3 (0.47 g, 1.4 mmol) using the general

NMR (400 MHz, Chloroform-d) δ 8.23 (dd, J = 1.8, 1.0 Hz, 1H),

7.59 (dd, J = 3.6, 1.0 Hz, 1H), 7.54 (dd, J = 9.3, 1.0 Hz, 1H), 7.38

(dd, J = 9.3, 1.9 Hz, 1H), 7.35−7.29 (m, 3H), 7.15 (dd, J = 5.1, 3.6

Hz, 1H), 7.12 (dd, J = 5.1, 3.6 Hz, 1H), 3.38 (s, 2H). ¹C NMR (151

MHz, Chloroform-d) δ 140.62, 140.58, 137.33, 130.93, 128.30,

127.87, 125.10, 124.83, 123.88, 123.84, 123.75, 121.93, 120.40,

tert-butyl deprotection procedure. The reaction mixture was cooled in

an ice bath upon adjustment of pH with 5 M NaOH. The product was

isolated as a yellow crystalline solid, (0.329 g, 84% yield): mp 208−

09 °C. R = 0.7 (Heptane/EtOAc) (1: 3). R = 8.80 min (>95% pure

118.16, 117.28. ESI-MS (m/z) calculated [M + H] for C H N S =

15 11

3 2

by HPLC system 2). H NMR (400 MHz, Chloroform-d) δ 8.46−

298.05, found 298.0.

.38 (m, 1H), 7.66−7.57 (m, 2H), 7.38 (dd, J = 5.1, 1.1 Hz, 1H),

6-bromo-2-cyclopentylimidazo[1,2-a]pyridin-3-amine (18).

Obtained from 9 (0.242 g, 0.7 mmol, 1 equiv) using the general tert-

butyl deprotection procedure. The extraction was performed with

.29−7.23 (m, 1H), 7.16 (dd, J = 5.1, 3.6 Hz, 1H), 3.41 (s, 2H).

NMR (151 MHz, DMSO-d ) δ 137.90, 137.72, 128.00, 127.21,

737

J. Med. Chem. 2021, 64, 4730−4743

Journal of Medicinal Chemistry

Article

EtOAc (2 × 200 mL) and once with a mixture of MeOH (30 mL) in

EtOAc (200 mL) to obtain complete extraction of the product. The

MS (m/z) calculated [M + H] for C H BrN OS = 398.00, found

18 12 3

398.5 and [M + H + 2] = 400.5.

compound was isolated as an orange solid (0.196 g, 100% yield). R =

N-(6-bromo-2-(thiophen-2-yl)imidazo[1,2-a]pyridin-3-yl)-4-

phenoxybenzamide (25). Obtained from 10 (0.140 g, 0.50 mmol)

and 4-phenoxybenzoyl chloride (0.140 g, 0.6 mmol) using the general

acylation procedure for imidazo[1,2-a]pyridine benzamides. The

reaction mixture was reduced in vacuo, and the resulting solid was

.5 (Heptane/EtOAc) (1: 3) + 1% Et N. R = 1.57 (3 mL/min ﬂow

3 t

rate). H NMR (400 MHz, Chloroform-d) δ 8.14 (dd, J = 2.0, 0.9 Hz,

H), 7.37 (dd, J = 9.4, 0.9 Hz, 1H), 7.09 (dd, J = 9.5, 1.9 Hz, 1H),

.28−3.16 (m, 1H), 3.00 (s, 2H), 2.07−2.00 (m, 2H), 1.96−1.81 (m,

washed with cold H O and cold acetone and dried in vacuo. The

H), 1.75−1.66 (m, 2H). C NMR (101 MHz, Chloroform-d) δ

product was puriﬁed by trituration from MeOH and subsequently

41.41, 139.35, 125.75, 122.13, 121.80, 117.65, 106.19, 37.72, 33.04,

washed with H O and MeOH and dried under high vacuum. The

5.99. ESI-MS (m/z) calculated [M + H] for C H BrN = 280.04,

product was isolated as a white solid (0.047 g, 20% yield): mp 266 °C.

found 280.0 and [M + H + 2] = 282.0.

R = 0.6 (PET/EtOAc) (1: 2) + 1% Et N. R = 2.35 min (3 mL/min

-nitro-2-phenylimidazo[1,2-a]pyridine (22).

Obtained

ﬂow rate) (>99% pure by HPLC system 1). H NMR (400 MHz,

DMSO-d ) δ 10.59 (s, 1H), 8.53 (d, J = 2.3 Hz, 1H), 8.21−8.13 (m,

from 21 (0.500 g, 2.5 mmol), phenyl boronic acid (0.370 g, 3.0

mmol), Na CO (0.268 g, 2.5 mmol, 1 equiv), and Pd(PPh ) (0.288

3 4

H), 7.64−7.54 (m, 2H), 7.51−7.42 (m, 4H), 7.28−7.23 (m, 1H),

g, 0.25 mmol, 10 mol %) using the general Suzuki−Miyaura cross-

coupling procedure. The crude product was dissolved in DCM,

concentrated onto silica gel 60 (40−63 μm), and subsequently

puriﬁed by column chromatography (EtOAc/Heptane) (1: 1) to

.22−7.10 (m, 5H). C NMR (151 MHz, DMSO-d ) δ 165.89,

60.50, 155.39, 140.52, 135.67, 134.80, 130.55, 130.33, 128.45,

27.92, 127.52, 126.55, 124.78, 124.55, 123.94, 119.72, 117.68,

117.44, 115.03, 106.48. ESI-MS (m/z) calculated [M + H] for

C H BrN O S = 490.0, found 490.0 and [M + H + 2] = 492.0.

24 16

furnish the product as a yellow solid (0.525 g, 88% yield). H NMR

(

400 MHz, Chloroform-d) δ 9.52 (dt, J = 7.1, 1.2 Hz, 1H), 7.96−7.87

N-(6-bromo-2-cyclopentylimidazo[1,2-a]pyridin-3-yl)-

benzamide (26). Obtained from 18 (0.223 g, 0.8 mmol) and

benzoyl chloride (0.1 mL, 1.1 mmol) using the general acylation

procedure for imidazo[1,2-a]pyridine benzamides. No precipitate

formed upon addition of 1.5 mL of water to the reaction mixture. The

reaction mixture was reduced in vacuo and then concentrated on celite

and puriﬁed by column chromatography (silica gel 60, 40−63 μM)

m, 2H), 7.85 (dt, J = 8.9, 1.1 Hz, 1H), 7.66 (ddd, J = 8.9, 7.0, 1.3 Hz,

H), 7.57−7.47 (m, 3H), 7.29 (td, J = 7.0, 1.3 Hz, 1H). C NMR

151 MHz, Chloroform-d) δ 150.42, 145.29, 132.02, 130.98, 130.33,

(

30.18, 129.10, 128.32, 128.28, 118.46, 116.61. The spectral data

were consistent with previously reported data.

-phenylimidazo[1,2-a]pyridin-3-amine (23). 22 (1.00 g,

.2 mmol) was dissolved in MeOH (39 mL) and then SnCl (3.98 g,

(

DCM/MeOH/Et N) (20: 0.5: 0.01). Selected pure fractions were

20.0 mmol) was added. The reaction mixture was stirred overnight at

combined and reduced, and the resulting product was then triturated

from MeCN, reduced in vacuo, then under high vacuum, and ﬁnally

lyophilized to yield the product as a white solid (0.012 g, 4% yield). R_f

reﬂux (65 °C). The reaction mixture was then cooled to room

temperature, and the solvent was removed in vacuo. A small amount

of EtOAc was then added, and the pH of the solution was adjusted

with 5 M aqueous NaOH to pH∼7. The reaction mixture was left

stirring for 2 h. Additional EtOAc was added, and the solution was

ﬁltered through celite. The compound was then extracted through

continuous extraction overnight using EtOAc and reduced in vacuo to

0.5 (DCM/MeOH) (40: 1) + 1% Et N. R = 10.40 min (>99% pure

3 t

by HPLC system 2). H NMR (400 MHz, DMSO-d ) δ 10.30 (s,

H), 8.35 (d, J = 1.9 Hz, 1H), 8.11−8.04 (m, 2H), 7.70−7.61 (m,

(

dd, J = 9.5, 1.9 Hz, 1H), 3.14 (p, J = 8.1 Hz, 1H), 1.94 (dtd, J = 10.7,

obtain the product as a white solid (0.740 g, 83% yield). H NMR

.6, 5.2, 2.7 Hz, 2H), 1.83−1.71 (m, 4H), 1.60 (dt, J = 8.6, 4.7 Hz,

(

400 MHz, Chloroform-d) δ 8.04−7.94 (m, 3H), 7.55 (dt, J = 9.1, 1.1

H). C NMR (151 MHz, DMSO-d ) δ 166.51, 145.55, 139.96,

Hz, 1H), 7.47 (dd, J = 8.4, 7.0 Hz, 2H), 7.37−7.28 (m, 1H), 7.11

33.12, 132.12, 128.45, 128.08, 126.61, 123.41, 117.58, 115.66,

(ddd, J = 9.1, 6.7, 1.3 Hz, 1H), 6.81 (td, J = 6.7, 1.1 Hz, 1H), 3.40 (s,

05.51, 37.06, 32.28, 25.34. ESI-MS (m/z) calculated [M + H] for

H). C NMR (151 MHz, Methanol-d ) δ 141.62, 135.56, 130.66,

C H BrN O = 384.07, found 384.1 and [M + H + 2] = 386.0.

129.60, 128.16, 127.91, 126.69, 124.67, 123.53, 116.95, 112.92. The

N-(6-phenyl-2-(thiophen-2-yl)imidazo[1,2-a]pyridin-3-yl)-

spectral data in Chloroform-d were consistent with previously

benzamide (27). Obtained from 16 (0.150 g, 0.5 mmol) and

benzoyl chloride (0.07 mL, 0.6 mmol) using the general acylation

procedure for imidazo[1,2-a]pyridine benzamides. The reaction

mixture was reduced in vacuo, and the resulting solid was

recrystallized from THF to yield the product as a white solid

reported data.

General Acylation Procedure for the Synthesis of Imidazo-

1,2-a]Pyridine Benzamides. Flame- or oven-dried glassware was

[

used. The imidazo[1,2-a]pyridine-3-amines (0.5 mmol, 1 equiv) were

suspended in dry toluene (3 mL) and dry pyridine (1 mL). The ﬂask

was ﬁtted with a rubber septum and purged with argon. The acyl

chloride (1.2 equiv) was added at once through the septum, and the

reaction mixture was stirred at room temperature and monitored by

TLC, UPLC-ESI-MS, and/or HPLC. The crude product was

obtained and puriﬁed as speciﬁed.

(

0.034 g, 17% yield): mp 270−271°C. R = 0.4 (PET/EtOAc) (1: 3).

R = 11.37 min. (>95% pure by HPLC system 2). H NMR (600

MHz, Methanol-d ) δ 8.76 (t, J = 1.4 Hz, 1H), 8.36 (dd, J = 9.3, 1.7

Hz, 1H), 8.19 (d, J = 7.4 Hz, 2H), 8.04 (d, J = 1.0 Hz, 1H), 7.80

(

ddd, J = 12.5, 4.4, 1.1 Hz, 2H), 7.76−7.70 (m, 3H), 7.64 (t, J = 7.8

Hz, 2H), 7.54 (dd, J = 8.3, 6.6 Hz, 2H), 7.52−7.45 (m, 1H), 7.29 (t, J

N-(6-bromo-2-(thiophen-2-yl)imidazo[1,2-a]pyridin-3-yl)-

benzamide (24). Obtained from 10 (0.145 g, 0.5 mmol) and

benzoyl chloride (0.07 mL, 0.6 mmol) using the general acylation

procedure for imidazo[1,2-a]pyridine benzamides. The crude product

was obtained by adding 1.5 mL of water to the reaction mixture,

which was subsequently stirred for 30 min. The reaction mixture was

cooled in an ice bath, and the precipitate was collected by ﬁltration

and washed with cold water (5 °C, 3 × 3 mL) and cold acetone (−18

3.9 Hz, 1H). C NMR (151 MHz, Methanol-d ) δ 170.42, 139.02,

36.17, 135.73, 134.46, 133.58, 133.39, 131.42, 130.54, 130.40,

30.05, 129.74, 129.53, 129.30, 128.65, 128.55, 127.71, 123.58,

17.88, 113.17. ESI-MS (m/z) calculated [M + H] for C H N OS

396.12, found 396.2 and 397.2.

N-(2,6-di(thiophen-2-yl)imidazo[1,2-a]pyridin-3-yl)-

benzamide (28). Obtained from 17 (0.23 g, 0.9 mmol) and benzoyl

chloride (0.1 mL 1.0 mmol) using the general acylation procedure for

imidazo[1,2-a]pyridine benzamides. The crude product was obtained

by adding 0.5 mL of water to the reaction mixture and stirred for 30

min. The reaction mixture was cooled in an ice bath, and the

precipitate was collected by ﬁltration and washed with cold water to

yield the crude product (0.247 g). 0.125 g of the crude product was

recrystallized from EtOH, then dried in vacuo, and ﬁnally dried under

high vacuum. The product was isolated as a white solid (0.038 g, 32%

yield): mp 253 °C. R = 0.3 (DCM/ MeOH) (40: 1) + 1% Et N. R =

C, 3 mL) to yield the crude product. The crude product was

recrystallized from toluene, which furnished the product as a white

solid: mp 269 °C. R = 0.4 (DCM: EtOAc) (5: 1). R = 1.58 min (4

mL/min ﬂow rate) (>95% pure by HPLC system 1). H NMR (400

MHz, Methanol-d ) δ 8.27 (t, J = 1.3 Hz, 1H), 8.19−8.12 (m, 2H),

.74−7.65 (m, 1H), 7.65−7.57 (m, 3H), 7.55−7.45 (m, 3H), 7.13

(dd, J = 5.1, 3.7 Hz, 1H). C NMR (151 MHz, Methanol-d ) δ

70.17, 142.95, 136.78, 135.80, 134.03, 134.00, 130.84, 129.97,

29.27, 128.69, 127.56, 126.73, 124.85, 118.23, 115.98, 108.66. ESI-

11.27 min (>99% pure by HPLC system 2). H NMR (400 MHz,

738

J. Med. Chem. 2021, 64, 4730−4743

Journal of Medicinal Chemistry

Article

DMSO-d ) δ 10.67 (s, 1H), 8.35 (d, J = 1.6 Hz, 1H), 8.20−8.13 (m,

mmol) and 4-chlorobenzoyl chloride (0.1 mL 0.8 mmol) using the

general acylation procedure for imidazo[1,2-a]pyridine benzamides.

The crude product was obtained by adding 1.5 mL of water to the

reaction mixture and stirred at 0 °C for 30 min. The precipitating

solid was collected by ﬁltration and washed with water and dried in

vacuo to yield the crude product. The crude product was triturated

from EtOAc(15 mL), puriﬁed by recrystallizations, ﬁrst from EtOAc

(15 mL) and then from MeOH (14 mL), and dried in vacuo to furnish

H), 7.75−7.55 (m, 8H), 7.51 (d, J = 3.6 Hz, 1H), 7.15 (ddd, J = 8.6,

.1, 3.6 Hz, 2H). ¹³C NMR (151 MHz, DMSO-d ) δ 166.70, 141.27,

39.14, 136.08, 134.80, 133.04, 132.38, 128.66, 128.50, 128.13,

27.88, 126.29, 126.02, 124.90, 124.76, 124.52, 119.97, 118.98,

17.06, 115.01. ESI-MS (m/z) calculated [M + H] for C H N OS

402.07, found 402.1.

N-(2-phenylimidazo[1,2-a]pyridin-3-yl)benzamide (29).

Benzoyl chloride (0.141 g, 1.0 mmol) in DCM (0.7 mL) was slowly

added to the stirring solution of DIPEA (0.174 g, 1.35 mmol) and 23

the product as a white solid (0.052 g, 16% yield): mp 289 °C. R = 0.4

(PET/EtOAc) (2: 1). R = 10.73 min (>99% pure by HPLC system

(0.15 g, 0.7 mmol) in DCM at 0 °C. The reaction mixture was left

2). H NMR (400 MHz, DMSO-d ) δ 10.73 (s, 1H), 8.51 (dd, J =

stirring for 3 h at room temperature and monitored by TLC. The

reaction mixture was reduced in vacuo, and the resulting product was

puriﬁed by column chromatography (EtOAc/Heptane) (6: 4) + 1%

4.5, 2.4 Hz, 1H), 8.18−8.11 (m, 2H), 7.73−7.65 (m, 3H), 7.56 (dd, J

= 5.1, 1.2 Hz, 1H), 7.48 (dd, J = 3.7, 1.2 Hz, 1H), 7.42 (ddd, J = 10.1,

8.3, 2.5 Hz, 1H), 7.13 (dd, J = 5.1, 3.6 Hz, 1H). C NMR (151 MHz,

DMSO-d ) δ 165.66, 152.88 (d, J = 233.8 Hz), 139.86, 137.22,

135.90, 135.46 (d, J = 2.1 Hz), 131.82, 130.10, 128.66, 127.89,

126.32, 124.53, 117.47 (d, J = 9.1 Hz), 117.14 (d, J = 25.8 Hz),

115.71 (d, J = 1.9 Hz), 111.09 (d, J = 41.8 Hz). F NMR (376 MHz,

DMSO-d ) δ −140.31 (dt, J = 9.2, 4.8 Hz). ESI m/z calculated [M +

H] for C H ClFN OS = 372.04, found 372.0 and [M + H + 2] =

18 12 3

28.68, 128.55, 128.00, 127.68, 126.63, 125.20, 123.77, 116.92,

374.1.

15.44, 112.24. ESI-MS (m/z) calculated [M + H] for C H N O =

4-chloro-N-(6-iodo-2-(thiophen-2-yl)imidazo[1,2-a]pyridin-

3-yl)benzamide (33). Obtained from 14 (0.272 g, 0.8 mmol) and 4-

chloro benzoyl chloride (0.159 g, 0.9 mmol) using the general

acylation procedure for imidazo[1,2-a]pyridine benzamides. The

crude product was obtained by adding water (1.5 mL) to the

reaction mixture, which was then cooled in an ice bath for 30 min.

The precipitate was collected by ﬁltration, washed with water, and

dried in vacuo to yield the crude product. Recrystallization from

MeOH (26 mL) and drying in vacuo furnish the product as white

14.13, found 314.1.

-chloro-N-(6-methyl-2-(thiophen-2-yl)imidazo[1,2-a]-

pyridin-3-yl)benzamide (30). Obtained from 11 (0.130 g, 0.57

mmol) and 4-chlorobenzoyl chloride (0.090 mL 0.70 mmol, 1.2

equiv) using the general acylation procedure for imidazo[1,2-

a]pyridine benzamides. The crude product was obtained by adding

0.9 mL of distilled water to the reaction mixture and stirred at 0 °C

for 30 min. The precipitating white solid was collected by ﬁltration

and washed with water and recrystallized from EtOAc. Filtration and

drying in vacuo and freeze drying furnished the product as a white

needles (0.087 g, 23% yield): mp 279−280 °C. R

EtOAc) (1: 2). R = 11.55 min (>99% pure by HPLC system 2). H

NMR (400 MHz, DMSO-d ) δ 10.68 (s, 1H), 8.59 (d, J = 1.5 Hz,

= 0.5 (Heptane/

crystalline solid (0.095 g, 46% yield): mp = 276 °C. R = 0.2

(

Heptane/EtOAc) (1: 2) + 1% Et N. R = 10.61 min (>98% pure by

1H), 8.18−8.10 (m, 2H), 7.70 (d, J = 8.5 Hz, 2H), 7.59−7.50 (m,

HPLC system 2). H NMR (400 MHz, DMSO-d ) δ 10.67 (s, 1H),

2H), 7.50−7.42 (m, 2H), 7.12 (dd, J = 5.1, 3.6 Hz, 1H). C NMR

.19−8.12 (m, 2H), 8.03−7.98 (m, 1H), 7.74−7.66 (m, 2H), 7.53

(151 MHz, DMSO-d ) δ 165.75, 140.74, 137.22, 135.66, 134.36,

(

dt, J = 6.8, 1.6 Hz, 2H), 7.45 (dd, J = 3.7, 1.2 Hz, 1H), 7.21 (dd, J =

.2, 1.7 Hz, 1H), 7.11 (dd, J = 5.1, 3.6 Hz, 1H), 2.30 (s, 3H).

133.12, 131.81, 130.13, 128.64, 128.37, 127.90, 126.49, 124.73,

117.88, 114.06, 76.49. ESI m/z calculated [M + H] for

NMR (151 MHz, DMSO-d ) δ 165.67, 141.12, 137.25, 136.26,

11ClIN

OS = 479.94, found 480.0 and [M + H + 2] = 482.0.

33.84, 131.80, 130.01, 128.74, 128.56, 127.79, 125.94, 124.15,

N-(7-bromo-2-(thiophen-2-yl)imidazo[1,2-a]pyridin-3-yl)-4-

chlorobenzamide (34). Obtained from 15 (0.216 g, 0.7 mmol) and

-chlorobenzoyl chloride (0.129 g, 0.7 mmol) using the general

21.91, 121.19, 115.97, 113.82, 17.52. ESI m/z calculated [M + H]

for C H ClN OS = 368.06, found 368.1 and [M + H + 2] 370.1.

acylation procedure for imidazo[1,2-a]pyridine benzamides. The

crude product was obtained by adding water (2 mL) to the reaction

mixture, which was then cooled in an ice bath for 30 min. The

precipitate was collected by ﬁltration, washed with water, and dried in

vacuo to yield the crude product. Recrystallization from MeOH (14

-chloro-N-(2-(thiophen-2-yl)-6-(triﬂuoromethyl)imidazo-

1,2-a]pyridin-3-yl)benzamide (31). Obtained from 12 (0.166 g,

.59 mmol) and 4-chlorobenzoyl chloride (0.090 mL 0.70 mmol)

[

using the general acylation procedure for imidazo[1,2-a]pyridine

benzamides. The crude product was obtained by adding 1 mL of

water to the reaction mixture and stirred at 0 °C for 30 min. The

precipitated solid was collected by ﬁltration and washed with water

and then dried in vacuo. To the ﬁltrate was added water (15 mL), and

this aqueous layer was extracted with DCM (2 × 15 mL). The

combined organic layers were dried over Na SO , combined with the

mL) and drying in vacuo with Et

performed, which furnished the product as white needles (0.071 g,

23% yield): mp 266 °C. R = 0.6 (Heptane/EtOAc) (1: 1). R = 11.41

min (>95% pure by HPLC system 2). H NMR (400 MHz, DMSO-

) δ 10.77 (s, 1H), 8.18 (d, J = 7.2 Hz, 1H), 8.14 (d, J = 8.6 Hz, 2H),

7.96 (d, J = 1.8 Hz, 1H), 7.75−7.65 (m, 2H), 7.58 (dd, J = 5.1, 1.2

O (5 mL) added twice were

initially collected precipitate, and reduced in vacuo to furnish the

crude product. The crude product was concentrated on celite and

puriﬁed by column chromatography (silica gel 60, 40−63 μM)

Hz, 1H), 7.50 (dd, J = 3.6, 1.2 Hz, 1H), 7.19−7.08 (m, 2H).

NMR (151 MHz, DMSO-d ) δ 165.68, 142.21, 137.34, 135.60,

(

Heptane/EtOAc) (1: 0.9), reduced under high vacuum, and freeze-

134.69, 131.63, 130.02, 128.76, 127.94, 126.58, 125.00, 124.83,

dried. The product was isolated as a white solid (0.034 g, 14% yield):

mp 268 °C. R = 0.4 (Heptane/EtOAc) (1: 0.9). R = 12.89 min

>99% pure by HPLC system 2). H NMR (400 MHz, DMSO-d ) δ

118.57, 118.51, 115.76, 114.62. ESI m/z calculated [M + H] for

12BrClN

OS = 431.96, found 432.0 and [M + H + 2] = 434.0.

(

3-bromo-2-phenylimidazo[1,2-a]pyridine (37).

49,50

0.79 (s, 1H), 8.93−8.82 (m, 1H), 8.23−8.09 (m, 2H), 7.83 (d, J =

phenylimidazo[1,2-a]pyridine

(0.822 g, 4.2 mmol) was dissolved

.4 Hz, 1H), 7.75−7.68 (m, 2H), 7.63−7.56 (m, 2H), 7.54 (dd, J =

in MeCN (25 mL), and NBS (0.839 g, 4.7 mmol) was added at once.

The reaction mixture was stirred at room temperature in the dark

until completion by TLC (1.5 h). The reaction mixture was reduced

in vacuo, and the residue was taken up in DCM (150 mL). The

organic layer was washed with 2 M NaOH (80 mL), with a sat.

.6, 1.2 Hz, 1H), 7.15 (dd, J = 5.1, 3.6 Hz, 1H). C NMR (151 MHz,

DMSO-d ) δ 165.81, 141.92, 137.29, 135.84, 135.26, 131.77, 130.18,

28.66, 128.01, 127.03, 125.31, 123.87 (q, J = 5.5 Hz), 121.14 (q, J =

71.3 Hz), 120.90 (q, J = 3.3, 2.0 Hz), 117.66, 115.83, 115.07 (q, J =

3.6 Hz). ¹⁹F NMR (376 MHz, DMSO-d ) δ −59.90. ESI m/z

over MgSO

O (80 mL). The organic layer was dried

calculated [M + H] for C H ClF N OS = 422.03, found 422.1 and

[

M + H + 2]⁺424.1.

which solidiﬁed over time to glossy dark crystals (1.13 g, 98% yield).

-chloro-N-(6-ﬂuoro-2-(thiophen-2-yl)imidazo[1,2-a]-

This product was used in the subsequent reactions without further

pyridin-3-yl)benzamide (32). Obtained from 13 (0.210 g, 0.9

puriﬁcation. R = 0.5 (Heptane/EtOAc) (1: 1) + 1% Et N. H NMR

739

J. Med. Chem. 2021, 64, 4730−4743

Journal of Medicinal Chemistry

Article

(

400 MHz, Chloroform-d) δ 8.19 (dt, J = 7.0, 1.1 Hz, 1H), 8.17−8.11

temperature, in the dark, and monitored by TLC (17.5 h). The

m, 2H), 7.65 (dt, J = 9.0, 1.1 Hz, 1H), 7.52−7.46 (m, 2H), 7.42−

reaction mixture was poured into H O (100 mL) and EtOAc (100

.36 (m, 1H), 7.29−7.22 (m, 1H), 6.94 (td, J = 6.8, 1.1 Hz, 1H).

mL). The aqueous layer was extracted with EtOAc (2 × 100 mL).

NMR (151 MHz, Chloroform-d) δ 145.61, 142.85, 133.02, 128.61,

The combined organic layers were dried over Na SO and then

28.44, 128.04, 125.22, 124.11, 117.79, 113.19, 91.85. The spectral

reduced in vacuo. The crude product was concentrated on celite and

puriﬁed by column chromatography (40−63 μm silica gel 60) (PET/

EtOAc) (1: 1) to yield the product as a white solid (0.137 g, 78%

yield): mp 188−189 °C. R = 0.6 (PET/EtOAc) (1: 2). R = 10.62

52,53

data were consistent with previously reported data.

z) calculated [M + H] for C H BrN = 273.00, found 273.0 and [M

ESI-MS (m/

H + 2] = 275.0.

-iodo-2-phenylimidazo[1,2-a]pyridine (38). Obtained via a

min (>99% pure by HPLC system 2). H NMR (400 MHz, DMSO-

procedure similar to the synthesis of 37, from 2-phenylimidazo[1,2-

a]pyridine (0.860 g, 4.4 mmol) and N-iodosuccinimide (1.15 g, 5.4

mmol.). Also, this reaction was conducted under argon. The product

was isolated as a pale brown solid, 1.22 g, 3.8 mmol, 85% yield. This

d ) δ 9.19 (s, 1H), 8.46 (dd, J = 6.9, 1.2 Hz, 1H), 8.05−7.97 (m, 2H),

7.85−7.78 (m, 2H), 7.75−7.71 (m, 1H), 7.65 (dd, J = 8.6, 7.1 Hz,

2H), 7.59−7.51 (m, 1H), 7.45−7.37 (m, 3H), 7.36−7.29 (m, 1H),

7.00 (td, J = 6.8, 1.2 Hz, 1H). C NMR (151 MHz, DMSO-d ) δ

144.79, 143.86, 136.85, 136.48, 133.88, 129.93, 128.96, 128.43,

127.95, 127.76, 125.97, 125.21, 123.81, 120.23, 116.87, 112.97,

110.71. ESI-MS (m/z) calculated [M + H] for C H N = 338.1,

.09−8.03 (m, 2H), 7.63 (dt, J = 9.0, 1.1 Hz, 1H), 7.53−7.44 (m,

found 338.1.

3-(4-chlorophenyl)-5-(2-phenylimidazo[1,2-a]pyridin-3-yl)-

H), 7.43−7.37 (m, 1H), 7.27 (ddd, J = 9.0, 6.8, 1.3 Hz, 1H), 6.94

(td, J = 6.8, 1.2 Hz, 1H). C NMR (151 MHz, Chloroform-d) δ

1,2,4-oxadiazole (42) 41. (0.171 g, 0.64 mmol) and 4-chloro-N′-

48.29, 148.25, 133.73, 128.66, 128.49, 128.47, 126.66, 125.66,

hydroxybenzimidamide (0.073 g, 0.43 mmol) were dissolved in

DMSO (0.5 mL). Powdered NaOH (0.026 g, 0.63 mmol) was added

at once, and the reaction mixture was stirred at room temperature for

7 h and monitored by TLC. Water (8 mL) was added, and the

reaction mixture was stirred for 15 min after which the precipitate was

collected by ﬁltration and washed with water (15 mL). The solid was

dried in vacuo and puriﬁed by recrystallization from EtOAc (7 mL)

and dried in vacuo and freeze-dried. The product was isolated as a

117.76, 113.29, 59.57. The spectral data were consistent with

55,56

previously reported data.

for C H IN = 320.99, found 320.9.

ESI-MS (m/z) calculated [M + H]

-phenyl-3-((trimethylsilyl)ethynyl)imidazo[1,2-a]pyridine

(39). This compound was synthesized from 37 (0.400 g, 1.5 mmol),

which was placed in a microwave tube. CuI, (0.043 g, 0.23 mmol, 15

mol %) and Pd(PPh ) Cl (0.154 g, 0.22 mmol, 15 mol %) were

added. The tube was sealed and purged with argon. Ethynyltrime-

thylsilane (0.25 mL, 1.8 mmol) and triethylamine (4.5 mL) were

added. The reaction mixture was heated to 80 °C until completion by

TLC (23 h) and cooled to room temperature. The reaction mixture

white solid (0.103 g, 64% yield): mp 242 °C. R

= 0.8 (DCM/

MeOH) (10: 1). R = 14.55 min (>98% pure by HPLC 2). This

sample was dissolved in 1.0 mL of DMSO. H NMR (400 MHz,

Chloroform-d) δ 9.70 (d, J = 6.9 Hz, 1H), 8.13 (d, J = 8.4 Hz, 2H),

7.93−7.89 (m, 2H), 7.86 (d, J = 9.0 Hz, 1H), 7.59−7.44 (m, 6H),

was diluted with water (30 mL) and Et O (100 mL). The aqueous

layer was extracted with Et O (2 × 80 mL). The combined organic

7.20 (t, J = 6.9 Hz, 1H). C NMR (151 MHz, Chloroform-d) δ

168.53, 167.21, 152.70, 147.60, 137.69, 132.80, 129.93, 129.78,

129.41, 129.01, 128.76, 128.49, 128.31, 125.32, 117.85, 114.91,

layers were dried over Na SO and reduced in vacuo. The crude

product was concentrated on celite and puriﬁed by column

chromatography (silica gel 60, 40−63 μM) (PET/EtOAc) (3: 1).

107.54. ESI-MS (m/z) calculated [M + H] for C21

H14ClN O =

The product was isolated as a pale yellow solid (0.211 g, 50% yield).

373.09, found 373.1 and [M + 2 + H] = 375.1.

R = 0.3 (PET: EtOAc) (3: 1). H NMR (400 MHz, Chloroform-d) δ

Computational Chemistry. Homology Model. The model was

built using the Maestro Schrodinger, package version 10.7.015, release

2016−3, using Prime, Structure Prediction Wizard. The 3.86 Å cryoEM

structure of the human GABA R α1β2γ2 heteropentamer in complex

.39−8.32 (m, 2H), 8.29 (dt, J = 6.8, 1.2 Hz, 1H), 7.65 (dt, J = 8.9,

.1 Hz, 1H), 7.50−7.42 (m, 2H), 7.41−7.34 (m, 1H), 7.32−7.27 (m,

H), 6.93 (td, J = 6.8, 1.1 Hz, 1H), 0.35 (s, 9H). C NMR (151

MHz, Chloroform-d) δ 148.65, 145.16, 133.53, 128.71, 128.56,

27.37, 126.46, 125.47, 117.60, 113.07, 108.17, 105.04, 93.83, 0.07.

-phenyl-3-(1-phenyl-1H-1,2,3-triazol-4-yl)imidazo[1,2-a]-

pyridine (40). A ﬂask was charged with previously lyophilized 39

0.200 g, 0.69 mmol) and TBAF (0.345 g, 1.3 mmol). The ﬂask was

ﬁtted with a rubber septum and purged with argon. Then, dry THF

10 mL) was added through the septum. The reaction mixture was

with GABA and ﬂumazenil (PDB-code: 6D6T) was used as a

template for this ECD homology model. In particular, we used the

α1γ2 subunit interface, which contains the bound ﬂumazenil, to avoid

modeling a binding site that could not encompass a ligand in the

following docking. The human sequences of the ECDs of the α4- and

δ-subunits (Entries and P48169 and O14764) were extracted from

the UniProt database and aligned with the template (Chain D α1-

subunit and Chain E γ2-subunit, respectively), using ClustalW within

Maestro (Figure S4 ).

(

stirred at room temperature until completion by TLC (1.5 h). The

reaction mixture was reduced in vacuo, and the residue was taken up

in a 10% aqueous solution of NH Cl (20 mL) and DCM (20 mL).

The residues 36−43 of the α4-subunit and the residues 17−39 of

the δ-subunit were deleted as they were poorly modeled because of

the lack of a template and expected to be far from the C-loop binding

pocket, based on the built monomers. The sequence identities of the

ECD of the α4- and δ-subunits to the templates were found to be 69

and 36%, respectively.

The aqueous layer was extracted with DCM (2 × 20 mL), and the

combined organic layers were dried over Na SO and reduced in

vacuo to yield the crude product. The crude product was concentrated

on celite and puriﬁed by column chromatography (silica gel 60, 40−

63 μM) (PET/EtOAc) (2: 1) to furnish 3-ethynyl-2-phenylimidazo-

[

1,2-a]pyridine as a white solid (0.123 g, 82% yield): mp 141−142 °C

The model was initially built as monomers using the energy-based

method. The monomers were then combined into a heterodimer,

using the Prime Multimer function. After ensuring correct dimer

construction by comparing to the template (Figure S5 ), the

heterodimer was prepared using the Protein Preparation Wizard, to

ﬁx any problems such as steric clashes.

(

decomp). R = 0.4 (PET: EtOAc) (2: 1). H NMR (400 MHz,

Chloroform-d) δ 8.36−8.28 (m, 3H), 7.67 (dt, J = 9.1, 1.1 Hz, 1H),

.51−7.45 (m, 2H), 7.42−7.36 (m, 1H), 7.30 (ddd, J = 9.1, 6.8, 1.3

Hz, 1H), 6.94 (td, J = 6.8, 1.2 Hz, 1H), 4.07 (s, 1H). C NMR (151

MHz, Chloroform-d) δ 149.11, 145.41, 133.41, 128.92, 128.76,

27.46, 126.64, 125.39, 117.74, 113.27, 103.96, 90.04, 73.45. A ﬂask

was charged with 3-ethynyl-2-phenylimidazo[1,2-a]pyridine (0.114 g,

.52 mmol), CuSO ·5H O (0.0261 g, 0.10 mmol, 20 mol %) and

The expected protonation state of the amino acid residues at pH =

7.4 was then calculated using PropKa. Afterward, the model was

energy-minimized using OPLS3 force ﬁelds.

The model was assessed by ProCheck in which it was found that

88.9, 10.6, and 0.5% of the modeled amino acid residues were in most

favored regions, additional allowed regions, and generously allowed

regions, respectively (Supporting information). The outliers in the

Ramachandran plot and the residues GLU72(A) and GLN162(B)

(+)-Sodium L-ascorbate (0.0534 g, 0.27 mmol, 51 mol %). The ﬂask

was ﬁtted with a rubber septum and purged with argon. THF (15

mL), H O (10 mL), and 1.0 mL of a ∼0.5 M solution of

azidobenzene in tert-butyl methyl ether (∼0.52 mmol) were then

added through the septum. The reaction mixture was stirred at room

740

J. Med. Chem. 2021, 64, 4730−4743

Journal of Medicinal Chemistry

The Supporting Information is available free of charge at

Article

were visually inspected and found to be nowhere near the binding site

of investigation.

analogues were coapplied with a concentration of GABA correspond-

ing to approximately GABA EC₂₀values determined from full GABA

concentration−response curves at the respective receptor subtypes

(see Table 1). The obtained ΔRFU responses were normalized to the

ΔRFU response induced by 100 μM GABA (maximum response)

after subtraction of the buﬀer response and given as %Response of

Docking. The ligands were manually drawn within Maestro

(

Release 2016−3: Maestro, Schro

016) and preprocessed using LigPrep (Release 2016−3: LigPrep,

Schrodinger, LLC, New York, NY, 2016), and the structures were

dinger, LLC, New York, NY,

energy-minimized using OPLS3 forced ﬁelds. The protonation state of

the ligands was determined by Epik at pH =7.4 ± 2.

GABA_m_a_x.

Concentration−response curves used to determine the PAM

potencies (EC50-values) of the analogues and the agonist potencies

of GABA used to determine GABA EC20 values were ﬁtted to the

four-parameter concentration−response model:

To ensure a binding model that can encompass analogues of

diﬀerent sizes from the previous publication, the relatively large 4-

butoxy-N-(6-chloro-2-(thiophen-2-yl)imidazo[1,2-a]pyridin-3-yl)-

benzamide was initially docked using induced ﬁt docking (Release

top − bottom

1 + 10

2016−3: Induced Fit Docking protocol; Glide & Prime, Schrodinger,

Response = bottom +

[

(logEC −A)·n_H]

LLC, New York, NY, 2016). The calculations were performed with

the OPLS3 force ﬁelds and XP Glide for redocking, and the centroid

of the box was speciﬁed from the residues, α4R135 and δF90;

otherwise, default settings were applied. Among the very similar three

top-ranked poses according to the IFD score, we selected the pose

with the best Emodel score. The ligand was deleted, and the resulting

protein was used as a docking template for XP docking (Release

where EC₅₀is the concentration of the compound A resulting in the

half-maximum response (response halfway between top and bottom)

and n_His the Hill coeﬃcient of the curve. The data analysis was

performed using GraphPad Prism v. 8.4.3 (GraphPad software Inc.,

San Diego, CA, USA).

016−3: Glide, Schrodinger, LLC, New York, NY, 2016) of a

ASSOCIATED CONTENT

Supporting Information

https://pubs.acs.org/doi/10.1021/acs.jmedchem.0c02163.

selection of analogues, that is, (N-(6-bromo-2-(thiophen-2-yl)-

imidazo[1,2-a]pyridin-3-yl)-4-methoxybenzamide, N-(6,8-dibromo-2-

■

sı

(thiophen-2-yl)imidazo[1,2-a]pyridin-3-yl)-4-methoxybenzamide, 4-

butoxy-N-(6-chloro-2-(thiophen-2-yl)imidazo[1,2-a]pyridin-3-yl)-

benzamide, and DS2 (Figures S5, S6 ). The DS2 pose with the best

Emodel displayed a binding mode consistent with high scoring poses

of the other compounds and was selected for the design of new

compounds.

Pharmacology. Compounds. GABA and DS2 (4-chloro-N-[2-(2-

thienyl)imidazo[1,2-a]pyridin-3-yl]benzamide) were obtained from

Tocris Bioscience (Bristol, UK).

Structural comparison of the employed homology model

template and published templates (S1), structural

comparison of the binding modes of DS2 (S2),

representative concentration−response curves of the

PAM activity of 30−32, 35, and 36 (S3), the sequence

alignment (S4), structural comparison of the homology

model (S5), and structural illustration of the binding of

Cell Culturing and Transient Transfection. All used HEK293 cell

lines were maintained in DMEM containing Gluta-MAX-I supple-

mented with 10% fetal bovine serum and 1% penicillin−streptomycin

0−32 (S6) (PDF)

(

all from Life Technologies, Paisley, UK) in an incubator at 37 °C

Molecular formula string (CSV)

with a humidity of 5% CO . The HEK293 Flp-In cell line stably

expressing the human δ-subunit (δ-HEK) and the background HEK

cell line stably expressing the G-protein coupled receptor NPBWR2

were positively selected using 200 μg/mL hygromycin B as reported

AUTHOR INFORMATION

■

Corresponding Author

Bente Frolund − Department of Drug Design and

previously.

The α4β1δ receptor was selected as the model receptor based on

Pharmacology, Faculty of Health and Medical Sciences,

University of Copenhagen, Copenhagen DK-2100, Denmark;

orcid.org/0000-0001-5476-6288 ;

previous success with establishing a reproducible, uniform receptor

population for this subtype. Recombinant human α4β1δ and

α4β1γ2 GABA Rs were expressed by transfection of δ-HEK cells and

HEK background cells, respectively. δ-HEK cells were transfected in a

Phone: +45.35336495; Email: bfr@sund.ku.dk

:1 ratio of human α - and β -subunits (pUNIV) to express α4β1δ as

4 1

Authors

described previously. HEK background cells were transfected in a

:1:2 ratio of human α4-, β1-, and γ2-subunit (pcDNA3.1/Zeo) to

Frederik Rostrup − Department of Drug Design and

Pharmacology, Faculty of Health and Medical Sciences,

University of Copenhagen, Copenhagen DK-2100, Denmark

Christina B. Falk-Petersen − Department of Drug Design and

Pharmacology, Faculty of Health and Medical Sciences,

University of Copenhagen, Copenhagen DK-2100, Denmark

express α4β1γ2 using Polyfect transfection reagent (Qiagen, West

Sussex, UK) as described by the manufacturer, except for using half

volumes. A cell line stably expressing α1β2γ2 receptors in HEK293

cells as previously used was a gift from Marianne L. Jensen,

Neurosearch.

FMP Assay. Testing of compounds in the ﬂuorescence-based FMP

Kasper Harpso̷e − Department of Drug Design and

assay was performed exactly as described previously. Data were

obtained as relative changes in ﬂuorescence units (ΔRFU) given as

the diﬀerence between the baseline ﬂuorescence signal before

compound addition and the peak/top plateau in the ﬂuorescence

signal obtained after buﬀer/compound addition. All raw data traces

were inspected manually, and signals resulting from buﬀer/compound

additions and signal artifacts were omitted from the analysis.

Generally, we could conﬁrm the presence of the γ2 subunit in the

functional receptors from the higher response level and numerically

lower GABA potencies at α4β1γ2 compared to α4β1 binary receptors

in the FMP assay.

Pharmacology, Faculty of Health and Medical Sciences,

University of Copenhagen, Copenhagen DK-2100, Denmark

Stine Buchleithner − Department of Drug Design and

Pharmacology, Faculty of Health and Medical Sciences,

University of Copenhagen, Copenhagen DK-2100, Denmark

Irene Conforti − Department of Drug Design and

Pharmacology, Faculty of Health and Medical Sciences,

University of Copenhagen, Copenhagen DK-2100, Denmark

Sascha Jung − Institute of Pharmaceutical and Biomedical

Sciences, Johannes Gutenberg University Mainz, Mainz D-

For agonist testing, the obtained ΔRFU responses were normalized

to the ΔRFU response induced by 100 μM GABA (maximum

response) after subtraction of the buﬀer response and given as %

Response of GABA_m_a_x. When testing for the PAM eﬀect, the

55128, Germany

David E. Gloriam − Department of Drug Design and

Pharmacology, Faculty of Health and Medical Sciences,

741

J. Med. Chem. 2021, 64, 4730−4743

Journal of Medicinal Chemistry

orcid.org/0000-0002-4299-7561

Article

University of Copenhagen, Copenhagen DK-2100, Denmark;

Tanja Schirmeister − Institute of Pharmaceutical and

Biomedical Sciences, Johannes Gutenberg University Mainz,

Mainz D-55128, Germany

Petrine Wellendorph − Department of Drug Design and

Pharmacology, Faculty of Health and Medical Sciences,

University of Copenhagen, Copenhagen DK-2100, Denmark;

orcid.org/0000-0002-5455-8013

(10) Holm, M. M.; Nieto-Gonzalez, J. L.; Vardya, I.; Henningsen,

K.; Jayatissa, M. N.; Wiborg, O.; Jensen, K. Hippocampal GABAergic

dysfunction in a rat chronic mild stress model of depression.

Hippocampus 2011, 21, 422−433.

11) Brickley, S. G.; Mody, I. Extrasynaptic GABA(A) receptors:

their function in the CNS and implications for disease. Neuron 2012,

3, 23−34.

12) Lee, V.; Maguire, J. The impact of tonic GABAA receptor-

mediated inhibition on neuronal excitability varies across brain region

and cell type. Front. Neural Circuits 2014, 8, 1−27.

(13) Solomon, V. R.; Tallapragada, V. J.; Chebib, M.; Johnston, G.

Complete contact information is available at:

https://pubs.acs.org/10.1021/acs.jmedchem.0c02163

A. R.; Hanrahan, J. R. GABA allosteric modulators: An overview of

recent developments in non-benzodiazepine modulators. Eur. J. Med.

Chem. 2019, 171, 434−461.

Author Contributions

(

14) Lee, H. J.; Absalom, N. L.; Hanrahan, J. R.; van

The manuscript was written through contributions of all

authors. F.R. and K.H. performed the modeling studies. F.R.,

I.C., and S.J. synthesized the compounds. C.F.P., S.B., and

P.W. performed FMP assays. B.N. performed radioligand

binding assay. B.F. devised the study and supervised the work

together with P.W. All authors have given approval to the ﬁnal

version of the manuscript.

Nieuwenhuijzen, P.; Ahring, P. K.; Chebib, M. A pharmacological

characterization of GABA, THIP and DS2 at binary alpha4beta3 and

beta3delta receptors: GABA activates beta3delta receptors via the

beta3(+)delta(−) interface. Brain Res. 2016, 1644, 222−230.

(15) Jensen, M. L.; Wafford, K. A.; Brown, A. R.; Belelli, D.;

Lambert, J. J.; Mirza, N. R. A study of subunit selectivity, mechanism

and site of action of the delta selective compound 2 (DS2) at human

recombinant and rodent native GABAA receptors. Br. J. Pharmacol.

Notes

(

013, 168, 1118−1132.

16) L’Estrade, E. T.; Hansen, H. D.; Falk-Petersen, C.; Haugaard,

A.; Griem-Krey, N.; Jung, S.; Luddens, H.; Schirmeister, T.;

Erlandsson, M.; Ohlsson, T.; Knudsen, G. M.; Herth, M. M.;

Wellendorph, P.; Frolund, B. Synthesis and pharmacological

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

■

This work was supported by a Lundbeck Foundation (grant

R230-2016-2562 to C.B.F.-P.) and the Drug Research

Academy.

evaluation of [11C]4-Methoxy-N-[2-(thiophen-2-yl)imidazo[1,2-a]-

pyridin-3-yl]benzamide as a brain penetrant PET ligand selective for

the δ -subunit-containing γ -aminobutyric acid type A receptors. ACS

Omega 2019, 4, 8846−8851.

(17) Yakoub, K.; Jung, S.; Sattler, C.; Damerow, H.; Weber, J.;

Kretzschmann, A.; Cankaya, A. S.; Piel, M.; Rösch, F.; Haugaard, A.

ABBREVIATIONS

■

CryoEM, Cryogenic Electron Microscopy; ECD, extracellular

domain; FMP, ﬂuorometric imaging plate reader (FLIPR)

S.; Fro ̷lund, B.; Schirmeister, T.; Luddens, H. Structure −function

evaluation of imidazopyridine derivatives selective for δ -subunit-

containing γ -aminobutyric acid type A (GABAA) receptors. J. Med.

Chem. 2018, 61, 1951−1968.

membrane potential; GABA, γ-aminobutyric acid; GABA Rs,

γ-aminobutyric acid type A receptors; [ H]-EBOB, [ H]-

ethynylbicyclooethobenzoate; PAM, positive allosteric modu-

lator; SEM, Standard Error of the Mean; SAR, structure−

activity relationship; XP, Xtra precision.

(18) Cole, L. M.; Casida, J. E. GABA-gated chloride channel:

binding-site for 4 ′-ethynyl-4-n-[2,3-3H2]propylbicycloorthobenzoate.

Pestic. Biochem. Physiol. 1992, 44, 1−8.

19) Uusi-Oukari, M.; Maksay, G. Allosteric modulation of

[

3H]EBOB binding to GABAA receptors by diflunisal analogues.

Neurochem. Int. 2006, 49, 676−682.

20) Yagle, M. A.; Martin, M. W.; de Fiebre, C. M.; de Fiebre, N. C.;

Drewe, J. A.; Dillon, G. H. [3H]Ethynylbicycloorthobenzoate

[3H]EBOB) binding in recombinant GABAA receptors. Neuro-

toxicology 2003, 24, 817−824.

21) Maksay, G.; Biro, T. High affinity, heterogeneous displacement

REFERENCES

■

2) Tija, C. J.; Guido, M. GABAA receptor trafficking and its role in

1) Owens, D. F.; Kriegstein, A. R. Is there more to gaba than

synaptic inhibition? Nat. Rev. Neurosci. 2002 , 3 , 715 −727.

the dynamic modulation of neuronal inhibition. Nat. Rev. Neurosci.

008, 9, 331−343.

3) Chiara, D. C.; Jounaidi, Y.; Zhou, X.; Savechenkov, P. Y.; Bruzik,

(

of [3H]EBOB binding to cerebellar GABA A receptors by

neurosteroids and GABA agonists. Neuropharmacology 2005, 49,

K. S.; Miller, K. W.; Cohen, J. B. General anesthetic binding sites in

human α 4 β 3 δ γ -aminobutyric acid type A receptors (GABAARs). J.

Biol. Chem. 2016, 291, 26529−26539.

31−438.

(22) Stephen, G. B.; Zhiwen, E.; Thomas, P. M.; Catriona, M. H.

4) Hanson, S. M.; Czajkowski, C. Structural mechanisms underlying

The contribution of delta subunit-containing GABAA receptors to

phasic and tonic conductance changes in cerebellum, thalamus and

neocortex. Front. Neural Circuits 2013, 7, 1−8.

benzodiazepine modulation of the GABA(A) receptor. J. Neurosci.

008, 28, 3490−3499.

5) Möhler, H.; Fritschy, J. M.; Rudolph, U. A new benzodiazepine

pharmacology. J. Pharmacol. Exp. Ther. 2002 , 300 , 2 −8.

6) Gravielle, M. C. Regulation of GABAA receptors by prolonged

exposure to endogenous and exogenous ligands. Neurochem. Int. 2018,

18, 96−104.

7) Farrant, M.; Nusser, Z. Variations on an inhibitory theme: phasic

and tonic activation of GABAA receptors. Nat. Rev. Neurosci. 2005, 6,

15−229.

8) Chuang, S.-H.; Reddy, D. S. Genetic and molecular regulation of

(23) Hartiadi, L. Y.; Ahring, P. K.; Chebib, M.; Absalom, N. L. High

and low GABA sensitivity α 4 β 2 δ GABAA receptors are expressed in

Xenopus laevis oocytes with divergent stoichiometries. Biochem.

Pharmacol. 2016, 103, 98−108.

Steyaert, J.; Miller, K. W.; Aricescu, A. R. GABAA receptor signalling

(24) Masiulis, S.; Desai, R.; Uchanski, T.; Serna Martin, I.; Laverty,

D.; Karia, D.; Malinauskas, T.; Zivanov, J.; Pardon, E.; Kotecha, A.;

mechanisms revealed by structural pharmacology. Nature 2019, 565,

454−459.

extrasynaptic GABA-A receptors in the brain: Therapeutic insights for

epilepsy. J. Pharmacol. Exp. Ther. 2018 , 364 , 180 −197.

9) Roberto, M.; Varodayan, F. P. Synaptic targets: Chronic alcohol

(25) Ahring, P. K.; Bang, L. H.; Jensen, M. L.; Stro ̷bæk, D.; Hartiadi,

L. Y.; Chebib, M.; Absalom, N. A pharmacological assessment of

actions. Neuropharmacology 2017, 122, 85−99.

agonists and modulators at α 4 β 2 γ 2 and α 4 β 2 δ GABAA receptors:

742

J. Med. Chem. 2021, 64, 4730−4743

Journal of Medicinal Chemistry

modulators with close chemical similarity but diverse pharmacological

Article

The challenge in comparing apples with oranges. Pharmacol. Res.

properties. Mol. Pharmacol. 2012, 81, 710−718.

26) Kaur, K. H.; Baur, R.; Sigel, E. Unanticipated structural and

016, 111, 563−576.

(44) Dalby, N. O.; Falk-Petersen, C. B.; Leurs, U.; Scholze, P.; Krall,

J.; Frolund, B.; Wellendorph, P. Silencing of spontaneous activity at

alpha4beta1/3delta GABAA receptors in hippocampal granule cells

reveals different ligand pharmacology. Br. J. Pharmacol. 2020, 177,

3975−3990.

functional properties of delta-subunit-containing GABAA receptors. J.

Biol. Chem. 2009, 284, 7889−7896.

(27) Baur, R.; Kaur, K. H.; Sigel, E. Structure of alpha6 beta3 delta

GABAA receptors and their lack of ethanol sensitivity. J. Neurochem.

009, 111, 1172−1181.

28) Barrera, N. P.; Betts, J.; You, H.; Henderson, R. M.; Martin, I.

(

(45) Martínez-Urbina, M. A.; Zentella, A.; Vilchis-Reyes, M. A.;

Guzmán, A.; Vargas, O.; Ramírez Apan, M. T.; Ventura Gallegos, J. L.;

L.; Dunn, S. M. J.; Edwardson, J. M. Atomic force microscopy reveals

the stoichiometry and subunit arrangement of the alpha4beta3delta

GABA(A) receptor. Mol. Pharmacol. 2008, 73, 960 −967.

Díaz, E. 6-Substituted 2-(N-trifluoroacetylamino)imidazopyridines

induce cell cycle arrest and apoptosis in SK-LU-1 human cancer

cell line. Eur. J. Med. Chem. 2010, 45, 1211−1219.

29) Patel, B.; Mortensen, M.; Smart, T. G. Stoichiometry of δ

(46) Schwerkoske, J.; Masquelin, T.; Perun, T.; Hulme, C. New

multi-component reaction accessing 3-aminoimidazo[1,2-a]pyridines.

Tetrahedron Lett. 2005, 46, 8355−8357.

subunit containing GABAA receptors. Br. J. Pharmacol. 2014, 171,

85−994.

30) Zhu, S.; Noviello, C. M.; Teng, J.; Walsh, R. M., Jr.; Kim, J. J.;

Hibbs, R. E. Structure of a human synaptic GABAA receptor. Nature

018, 559, 67−72.

31) Bienaymé, H.; Bouzid, K. A new heterocyclic multicomponent

(

(47) Yan, R.-L.; Yan, H.; Ma, C.; Ren, Z.-Y.; Gao, X.-A.; Huang, G.-

S.; Liang, Y.-M. Cu(I)-Catalyzed synthesis of imidazo[1,2-a]pyridines

from aminopyridines and nitroolefins using air as the oxidant. J. Org.

Chem. 2012, 77, 2024−2028.

reaction for the combinatorial synthesis of fused 3-aminoimidazoles.

(48) Reynoso Lara, J. E.; Salgado-Zamora, H.; Bazin, M.-A.;

Campos-Aldrete, M. E.; Marchand, P. Design and synthesis of

imidazo[1,2-a]pyridines with carboxamide group substitution and in

silico evaluation of their interaction with a LuxR-type quorum sensing

receptor. J. Heterocyclic Chem. 2018, 55, 1101−1111.

Angew. Chem., Int. Ed. 1998, 37, 2234−2237.

(32) Groebke, K.; Weber, L.; Mehlin, F. Synthesis of imidazo[1,2-a]

annulated pyridines, pyrazines and pyrimidines by a novel three-

component condensation. Synlett 1998, 1998, 661−663.

(49) Kour, D.; Khajuria, R.; Kapoor, K. K. Iodine-ammonium

(33) Blackburn, C.; Guan, B.; Fleming, P.; Shiosaki, K.; Tsai, S.

acetate promoted reaction between 2-aminopyridine and aryl methyl

ketones: a novel approach towards the synthesis of 2-arylimidazo[1,2-

a]pyridines. Tetrahedron Lett. 2016, 57, 4464−4467.

Parallel synthesis of 3-aminoimidazo[1,2-a]pyridines and pyrazines by

a new three-component condensation. Tetrahedron Lett. 1998, 39,

34) Guchhait, S. K.; Chaudhary, V.; Madaan, C. A chemoselective

635−3638.

(50) Sayeed, I. B.; Lakshma Nayak, V.; Shareef, M. A.; Chouhan, N.

K.; Kamal, A. Design, synthesis and biological evaluation of

imidazopyridine −propenone conjugates as potent tubulin inhibitors.

Med. Chem. Commun. 2017, 8, 1000−1006.

Ugi-type reaction in water using TMSCN as a functional isonitrile

equivalent: generation of heteroaromatic molecular diversity. Org.

Biomol. Chem. 2012, 10, 9271−9277.

(51) Salgado-Zamora, H.; Velazquez, M.; Mejía, D.; Campos-

(35) Bazin, M.-A.; Marhadour, S.; Tonnerre, A.; Marchand, P.

Aldrete, M. E.; Jimenez, R.; Cervantes, H. Influence of the 2-aryl

group on the ipso electrophilic substitution process of 2-arylimidazo-

Exploration of versatile reactions on 2-chloro-3-nitroimidazo[1,2-

a]pyridine: expanding structural diversity of C2- and C3-function-

alized imidazo[1,2-a]pyridines. Tetrahedron Lett. 2013, 54, 5378−

[

1,2-a]pyridines. Heterocycl. Commun. 2008, 14, 27−32.

(

52) Zhou, X.; Yan, H.; Ma, C.; He, Y.; Li, Y.; Cao, J.; Yan, R.;

Huang, G. Copper-mediated aerobic oxidative synthesis of 3-bromo-

382.

imidazo[1,2-a]pyridines with pyridines and enamides. J. Org. Chem.

36) Misra, A. P.; Raj, K.; Bhaduri, A. P. In search of imidazo [1,2-a]

pyridine derivatives exhibiting resistance for catalytic hydrogenation.

(

016, 81, 25−31.

Synth. Commun. 1999, 29, 3227−3236.

53) Delaye, P. O.; Pénichon, M.; Allouchi, H.; Enguehard-Gueiffier,

(37) Tokuhara, H.; Imaeda, Y.; Fukase, Y.; Iwanaga, K.; Taya, N.;

C.; Gueiffier, A. Regiocontrolled functionalization of 2,3-

dihalogenoimidazo[1,2- a ]pyridines by SuzukiMiyaura and Sonoga-

shira cross-coupling reactions. Org. Biomol. Chem. 2017, 15, 4199−

Watanabe, K.; Kanagawa, R.; Matsuda, K.; Kajimoto, Y.; Kusumoto,

K.; Kondo, M.; Snell, G.; Behnke, C. A.; Kuroita, T. Discovery of

benzimidazole derivatives as orally active renin inhibitors: Opti-

mization of 3,5-disubstituted piperidine to improve pharmacokinetic

profile. Bioorg. Med. Chem. 2018, 26, 3261−3286.

54) Ghosh, P.; Ganguly, B.; Kar, B.; Dwivedi, S.; Das, S. Green

204.

procedure for highly efficient, rapid synthesis of imidazo[1,2-

a]pyridine and its late stage functionalization. Synth. Commun.

(38) El Akkaoui, A.; Bassoude, I.; Koubachi, J.; Berteina-Raboin, S.;

Mouaddib, A.; Guillaumet, G. Pd-catalyzed regiocontrolled Sonoga-

shira and Suzuki cross-coupling reaction of 3,6-dihalogenoimidazo-

55) Zhang, Y.; Chen, Z.; Wu, W.; Zhang, Y.; Su, W. CuI-catalyzed

018, 48, 1076−1084.

[

1,2-a]pyridines: one-pot double-coupling approach. Tetrahedron

011, 67, 7128−7138.

39) Huo, C.; Tang, J.; Xie, H.; Wang, Y.; Dong, J. CBr4 Mediated

aerobic oxidative α -aminaton cyclization of ketones to access aryl or

alkenyl-substituted imidazoheterocycles. J. Org. Chem. 2013, 78,

Oxidative C −N Bond Formation: Applied in the synthesis of

2494−12504.

imidazo[1,2-α ]pyridines and imidazo[1,2-α ]pyrimidines. Org. Lett.

56) Chen, Z.; Cao, G.; Zhang, F.; Li, H.; Xu, J.; Miao, M.; Ren, H.

Metal-free mediated C-3 methylsulfanylation of imidazo[1,2-a]-

pyridines with dimethyl sulfoxide as a methylsulfanylating agent.

Synlett 2017, 28, 1795−1800.

(

016, 18, 1016−1019.

40) Mayer, J.; Sauer, A.; Iglesias, B. A.; Acunha, T. V.; Back, D.;

Rodrigues, O.; Dornelles, L. Ferrocenylethenyl-substituted 1,3,4-

oxadiazolyl-1,2,4-oxadiazoles: Synthesis, characterization and DNA-

binding assays. J. Organometal. Chem. 2017 , 841 , 1 −11.

(57) The UniProt Consortium. UniProt: a worldwide hub of protein

knowledge. Nucleic Acids Res. 2019, 47, D506−D515.

(58) Laskowski, R. A.; MacArthur, M. W.; Moss, D. S.; Thornton, J.

(41) Schwartz, T. W.; Holst, B. Allosteric enhancers, allosteric

M. PROCHECK: a program to check the stereochemical quality of

protein structures. J. Appl. Crystallogr. 1993, 26, 283−291.

agonists and ago-allosteric modulators: where do they bind and how

do they act? Trends Pharmacol. Sci. 2007, 28, 366−373.

(

59) Falk-Petersen, C. B.; So ̷gaard, R.; Madsen, K. L.; Klein, A. B.;

(42) Madjroh, N.; Olander, E. R.; Bundgaard, C.; Soderhielm, P. C.;

Fro ̷lund, B.; Wellendorph, P. Development of a robust mammalian

cell-based assay for studying recombinant α β δ GABA receptor

subtypes. Basic Clin. Pharmacol. Toxicol. 2017, 121, 119−129.

Jensen, A. A. Functional properties and mechanism of action of

PPTQ, an allosteric agonist and low nanomolar positive allosteric

modulator at GABAA receptors. Biochem. Pharmacol. 2018, 147,

43) Gill, J. K.; Dhankher, P.; Sheppard, T. D.; Sher, E.; Millar, N. S.

53−169.

A series of alpha7 nicotinic acetylcholine receptor allosteric

743