Developing New 4-PIOL and 4-PHP Analogues for Photoinactivation of γ-Aminobutyric Acid Type A Receptors

Article abstract of DOI:10.1021/acschemneuro.9b00478

The critical roles played by GABA_A receptors as inhibitory regulators of excitation in the central nervous system has been known for many years. Aberrant GABA_A receptor function and trafficking deficits have also been associated with several diseases including anxiety, depression, epilepsy, and insomnia. As a consequence, important drug groups such as the benzodiazepines, barbiturates, and many general anesthetics have become established as modulators of GABA_A receptor activity. Nevertheless, there is much we do not understand about the roles and mechanisms of GABA_A receptors at neural network and systems levels. It is therefore crucial to develop novel technologies and especially chemical entities that can interrogate GABA_A receptor function in the nervous system. Here, we describe the chemistry and characterization of a novel set of 4-PIOL and 4-PHP analogues synthesized with the aim of developing a toolkit of drugs that can photoinactivate GABA_A receptors. Most of these new analogues show higher affinities/potencies compared with the respective lead compounds. This is indicative of cavernous areas being present near their binding sites that can be potentially associated with novel receptor interactions. The 4-PHP azide-analogue, 2d, possesses particularly impressive nanomolar affinity/potency and is an effective UV-inducible photoinhibitor of GABA_A receptors with considerable potential for photocontrol of GABA_A receptor function in situ.

Full text of DOI:10.1021/acschemneuro.9b00478

Subscriber access provided by Uppsala universitetsbibliotek
Article
Developing new 4-PIOL and 4-PHP analogues for photo-
inactivation of #-aminobutyric acid type A receptors
Martin Mortensen, Jakob Krall, Kenneth T. Kongstad, Benjamin M. Brygger, Ombretta Lenzi, Pierre
Francotte, Troels E. Sørensen, Birgitte Nielsen, Anders A. Jensen, Trevor G. Smart, and Bente Frølund
ACS Chem. Neurosci., Just Accepted Manuscript • DOI: 10.1021/acschemneuro.9b00478 • Publication Date (Web): 07 Oct 2019
Downloaded from pubs.acs.org on October 8, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 59
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
Developing new 4-PIOL and 4-PHP analogues for
photo-inactivation of -aminobutyric acid type A
receptors
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
§,
†,
†
†
Martin Mortensen ᶲ, Jacob Krall ᶲ, Kenneth T. Kongstad , Benjamin M. Brygger , Ombretta
†
†,‡
†
†
†
Lenzi , Pierre Francotte , Troels E. Sørensen , Birgitte Nielsen , Anders A. Jensen , Trevor G.
§
,*
†.*
Smart , Bente Frølund
§
Department of Neuroscience, Physiology & Pharmacology, University College London, Gower
†
Street, London WC1E 6BT, United Kingdom. Department of Drug Design and Pharmacology,
Faculty of Health and Medical Sciences, University of Copenhagen, Universitetsparken 2, DK-
‡
2
100 Copenhagen. Department of Medicinal Chemistry, Center for Interdisciplinary Research
on Medicines (CIRM), University of Liege, Avenue de l’Hôpital, 1, B36, B-4000 Liège,
Belgium
1
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 2 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Abstract
The critical roles played by GABA receptors as inhibitory regulators of excitation in the central
A
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
nervous system has been known for many years. Aberrant GABA receptor function and trafficking
A
deficits have also been associated with several diseases including anxiety, depression, epilepsy, and
insomnia. As a consequence, important drug groups such the benzodiazepines, barbiturates, and many
general anesthetics, have become established as modulators of GABAA receptor activity.
Nevertheless, there is much we do not understand about the roles and mechanisms of GABAA
receptors at neural network and systems levels. It is therefore crucial to develop novel technologies
and especially chemical entities that can interrogate GABA receptor function in the nervous system.
A
Here, we describe the chemistry and characterization of a novel set of 4-PIOL and 4-PHP analogues
synthesized with the aim of developing a toolkit of drugs that can photo-inactivate GABA receptors.
A
Most of these new analogues show higher affinities/potencies compared to the respective lead
compounds. This is indicative of cavernous areas being present near their binding sites that can be
potentially associated with novel receptor interactions. The 4-PHP azide-analogue, 2d, possesses
particularly impressive nanomolar affinity/potency, and is an effective UV-inducible photo-inhibitor
of GABA receptors with considerable potential for photo-control of GABA receptor function in
A
A
situ.
Keywords
GABA receptors, 4-PIOL, 4-PHP, small ligand organic chemistry, photo-inhibition, single
A
particle tracking, quantum dots.
Introduction
2
ACS Paragon Plus Environment

Page 3 of 59
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
Synaptic inhibition in the brain is largely controlled by aminobutyric acid (GABA) where rapid
inhibition is mediated by the GABA receptors (GABA Rs). This receptor family is abundantly
A
A
expressed throughout the central nervous system, where they control and modulate a plethora of
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
important physiological processes . Evidence indicates that regulating the numbers of cell surface
postsynaptic GABA Rs represents one of the most powerful mechanisms underlying functional
A
2
plasticity of GABAergic synapses . Our understanding of GABA R cellular distribution and cell
A
surface trafficking profiles is still emerging, but it is clear that dysfunction to underlying mechanisms
affecting receptor expression feature in the pathogenesis of a wide range of neurological diseases,
3
-4
especially epilepsy .
Many approaches for studying receptor trafficking have been used such as expressing tagged
recombinant receptors5-8 or employing antibody-based labelling procedures of native receptors9-10
.
Although valuable, these techniques can be limited by their relative inability to discriminate between
functional and non-functional receptors, and between synaptic and extrasynaptic locations, in
addition to recombinant receptor expression potentially affecting native receptor subunit composition
and function. By contrast, a more direct approach for studying native GABA R trafficking is to
A
irreversibly block the orthosteric binding site using a selective high-affinity GABA R photoaffinity
A
ligand. Combining this photoaffinity approach with an associated fluorophore would markedly
enhance our ability to examine the localization and trafficking of GABA Rs with higher time and
A
spatial resolution.
Previously, we reported a series of photochemical probes based on the competitive GABA R
A
antagonist gabazine. In a range of compounds, we integrated the photo-reactive groups aryl azide,
diazirine or benzophenone into the gabazine structure, and demonstrated that the most efficient UV-
3
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 4 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
induced irreversible block of the GABA binding site (and receptor) was observed with an extended
1
1
benzophenone analogue called GZ-B1 .
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Here, we report on the design and synthesis of a novel series of photochemical probes engineered for
irreversible inactivation and imaging of native GABA Rs. The ‘modular-design’ of these molecules
A
is based on the very weak (low efficacy) partial agonists 5-(4-piperidyl)-3-hydroxyisoxazole (4-
PIOL)12-13 and 4-(4-piperidyl)-1-hydroxypyrazole (4-PHP) . Together these novel compounds
14
constitute a class of highly potent and selective orthosteric GABA R ligands. Their potential for
A
cross-linking and visualization of cell surface GABA Rs was also evaluated.
A
Results and Discussion
Design and synthesis of photoactivated GABA R ligands
A
Photoaffinity labelling is a powerful tool to covalently capture the protein target of small highly
1
5
specific molecules . However, the design of photoprobes is often challenging due to the difficulty of
maintaining the potency, efficacy and specificity of the parent lead compounds while imparting the
probes with photo-cross-linking and target visualization capabilities.
4
ACS Paragon Plus Environment

Page 5 of 59
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
A
B
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. (A) Structures of GABA, 4-PIOL, 4-PHP, and derivative target structures 1a–d, 2a–d, and 3a,b.
B) 4-PHP (green carbon atoms) docked into the α13γ2 GABA R homology model (gray carbon atoms)
(
A
1
8
based on the α1β3γ2L cryo-EM structure . The surface of the orthosteric binding site and adjacent cavities
are displayed in gray. H-bonds are depicted by dashed lines.
5
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 6 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Based on comprehensive structure-activity relationship studies using the core scaffolds of 4-PIOL
and 4-PHP (Fig. 1A), we have previously reported on detailed and reliable structural models of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
6-17
the orthosteric GABA R binding site
, now facilitated by the latest GABA R cryo-EM
A
A
structures18-21. Combining all these structures and models, we have identified several cavities in
the GABA (orthosteric) binding site, in close vicinity to the core scaffolds of 4-PIOL and 4-PHP,
which are able to accommodate considerable steric bulk (Fig. 1B). Furthermore, several ‘ligand-
access points’ connecting the GABA binding site (under loop C of the β subunit) to the
1
1
extracellular environment have been identified in GABA Rs. These cavities and ‘access points’
A
have also been verified by docking studies using the GABA binding site of homology models
based on the recent cryo-EM structures of the human α13γ2 heteropentameric GABA R in
A
nanodisc bilayers18-19. By exploring these binding site cavities with different ligands, we have
previously discovered highly potent competitive GABA R antagonists, exemplified by 3-phenyl
A
(K 0.27 µM, IC 1.9 µM) and 5-phenyl (K 0.022 µM, IC 0.15 µM) substituted 4-PHP analogues
i 50 i 50
(
Table 1)14, 17, 22. Interestingly, 3,3-diphenylpropyl-4-PIOL, and the corresponding thio-4-PIOL
analogue, displayed a preference for inhibiting tonic compared to phasic GABA R-mediated
A
currents23-24. Therefore, 4-PIOL and 4-PHP constitute useful scaffolds for introducing
photoactivatable groups/moieties into the GABA binding site.
Our initial aim was to identify the optimal photoactivatable moiety, and its position within the
1
1
target scaffold, for photo-cross-linking to the GABA R protein. As in our previous study , we
A
chose to explore the effects of sequentially incorporating benzophenone or aryl azide as
photoactivatable groups into our lead structures of 4-PIOL and 4-PHP. Despite the known
photochemistry and chemical reactivities of these groups, which ones are eventually adopted for
ACS Paragon Plus Environment
6

Page 7 of 59
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
inclusion is not straightforward. In some cases, a phenylazide-based photoaffinity probe is
preferred, while in other studies benzophenone-based probes are advantageous25-28. Based on
criteria of stability and synthetic feasibility, the benzophenone and phenylazide moieties were
prioritised for incorporation into the 4-position of 4-PIOL (Fig. 1, Scheme 1; 1a–d) and 3- or 5-
position of 4-PHP (Fig.1, Scheme 2; 2c,d), respectively.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The benzophenone derivatives of 4-PIOL and the phenyl-azide derivatives of 4-PHP were
synthesized as described in Schemes 1 and 2, respectively.
The benzophenone derivatives of 4-PIOL (1a–d) were synthesized from compound 4, which was
synthesized as described in the literature29-30, following two distinct strategies (Scheme 1). For the
benzophenone derivatives, with the benzophenone directly attached to the 4-isoxazole ring
(compounds 1a,b), a Suzuki-Miyaura cross coupling was applied between compound 4 and
commercially available (4-benzoylphenyl)boronic acid and phenyl(3-(4,4,5,5-tetramethyl-1,3,2-
3
1
dioxaborolan-2-yl)phenyl)methanone, following a modified procedure . Acidic deprotection of
the intermediates (5a,b) afforded compounds 1a,b.
The 3- and 4-methylene-benzophenone derivatives of 4-PIOL (1c,d) were synthesized by a
Grignard reaction between compound 4 and 3- or 4-benzoylbenzaldehyde, following a modified
procedure29-30. 3- and 4-benzoylbenzaldehyde were synthesized from commercially available 3-
and 4-benzoylbenzyl bromide, respectively, as previously described in the literature by Miziak et
3
2
33
al. (2007) and Suhana and Srinivasan (2003) , respectively. Reduction of the formed hydroxyl
group (intermediate not shown) using triethylsilane and trifluoroacetic acid followed by acid
deprotection of 6c,d afforded compounds 1c,d.
ACS Paragon Plus Environment
7

ACS Chemical Neuroscience
Page 8 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment
8

Page 9 of 59
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
Scheme 1 a
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Reagents and conditions: (i) (4-benzoylphenyl)boronic acid or phenyl(3-(4,4,5,5-tetramethyl-1,3,2-
dioxaborolan-2-yl)phenyl)methanone, Pd(PPh ) Cl , K CO , H O/DMF, 70 C, (ii) 33% HBr in AcOH, 65
3
2
2
2
3
2
i

C, (iii) a) 4, PrMgCl, THF, –30 C to 0 C, b) 3- or 4-benzoylbenzaldehyde, THF, 0 C to rt., (iv) Et SiH,
3
TFA, CH Cl , 50 C.
2
2
ACS Paragon Plus Environment
9

ACS Chemical Neuroscience
Page 10 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The 3- and 5-substituted 4-PHP analogous 2a–d were synthesized as described in Scheme 2 by
1
4
starting from 10a,b, which were obtained following procedures described previously .
The 3-iodinated intermediate 10f, was synthesized from compound 10a by first regio-selective
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
14
iodination in the 3-position using iodine monochloride (10c) followed by selective deprotection
of the N-ethyl-carbamate under basic conditions. The resulting free amine was then re-protected
using Boc anhydride and triethylamine to afford compound 10f. In contrast, the 5-iodinated
intermediates 10e and 10g were synthesized from compounds 10a and 10b using first an ortho-
directed metalation with LDA at –78 C followed by quenching with iodine to afford compound
1
4
10d and 10e . Compound 10d was then converted into compound 10g following an identical
approach as described for the conversion of compound 10c to 10f.
Suzuki-Miyaura cross-couplings of compounds 10c–f with 3-aminophenylboronic acid afforded
compounds 11a–d, where compounds 11a,b were transformed into target compounds 2a,b under
acidic conditions. To incorporate the aromatic azide into the target compounds 2c,d, two
approaches were used. Compound 11c was subjected to catalytic hydrogenation using palladium
on carbon, thereby resulting in the deprotected hydroxypyrazole 12, prior to formation of the
aromatic azide in compound 13a using trimethylsilyl azide and tert-butyl nitrite. For 11d, the
aniline was converted directly into the aromatic azide, compound 13b, in analogy to compound
1
3a. Compounds 13a,b were then deprotected under acidic conditions to afford compounds 2c,d.
Neither of the methods used for synthesizing the comparable aromatic azides 13a and 13b, was
superior.
ACS Paragon Plus Environment
1
0

Page 11 of 59
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
Scheme 2 a
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
i
Reagents and conditions: (i) a) PrMgCl, THF, 0 C, b) 7, THF, 0 C to rt., c) Et SiH, TFA, CH Cl , 65
3
2
2
i

C (ii) a) PrMgCl, THF, 0 C, b) 8, THF, 0 C to rt., c) H , 10% Pd/C, EtOH, rt., d) PMB-Cl, K CO ,
2
2
3
acetone, rt., (iii) ICl, AcOH/H O, 85 C, (iv) a) 10a or 10b, LDA, THF, –78 C, b) I , THF, –78 C to rt.,
2
2
(v) a) KOH, MeOH/H O, reflux, b) Boc O, Et N, CH Cl , (vi) 3-aminophenylboronic acid, K CO ,
2 2 3 2 2 2 3
Pd(PPh ) Cl or Pd(PPh ) , DMF/H O, 90 C, (vii) aq. HCl, reflux, (viii) H , 10% Pd/C, MeOH, rt., (ix)
3
2
2
3 4
2
2
t
TMS-N , BuNO , MeCN, (x) TFA, CH Cl , rt., (xi) HCl, Et O/EtOH, –78 C.
3
2
2
2
2
ACS Paragon Plus Environment
1
1

ACS Chemical Neuroscience
Page 12 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Pharmacological characterization of the photoactive 4-PIOL GABA R ligands.
A
The synthesized 4-PIOL analogues (1a–d) were evaluated for their apparent binding affinities
3
using a [ H]muscimol binding assay with a rat brain cell membrane preparation (Table 1). Using
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
an electrophysiological assay, all the analogues acted as antagonists at recombinant murine

12/32 GABA Rs expressed in HEK293 cells (Figure 2). Therefore, their relative potencies
A
were evaluated by construction of ligand-inhibition curves (Figure 2A) to obtain IC values for
5
0
each analogue (Figure 2B; Table 1). In general, the functional data are in good agreement with the
binding affinity data (Supplementary Fig. S1).
Table 1. Pharmacological data for 4-PIOL, the 4-PIOL analogues 1a-d, 4-PHP and the 4-PHP
analogues 3- and 5-Ph-4-PHP, 2a-d and 3a,b. GABA R binding affinities at rat cortical cell membranes
A
and functional properties at 12/32 GABA R transiently expressed in HEK293 (whole cell patch clamp)
A
or tsA201 cells (FMP assay).
[3H]muscimol binding
IC50 (µM) [pIC50 ± SEM]
K
i
(µM) [pK
i
± SEM]
4
-PIOL
9.1e
> 500 [< 3.30]e
0.008 [8.09 ± 0.12]a
0.10 [6.99 ± 0.11]a
0.18 [6.75 ± 0.11]a
0.24 [6.63 ± 0.14]a
> 500 [< 3.30]f
1a
0.018 [7.75 ± 0.01]
0.16 [6.80 ± 0.05]
0.47 [6.33 ± 0.01]
0.60 [6.22 ± 0.06]
10 [4.99 ± 0.02]f
0.27 [6.57 ± 0.01]f
0.022 [7.65 ± 0.02]f
0.27 [6.57 ± 0.06]
0.017 [7.79 ± 0.07]
0.22 [6.66 ± 0.04]
0.023 [7.66 ± 0.08]
32 [4.54 ±0.10]
1
b
1c
1
d
4-PHP
3
5
2
2
2
2
3
3
-Ph-4-PHP
1.9 [5.73 ± 0.06]c, f
-Ph-4-PHP
0.15 [6.81 ± 0.04]c, f
1.8 [5.75 ± 0.11]d
a
b
c
0.065 [7.19 ± 0.05]d
0.21 [6.68 ± 0.07]b
d
a
b
0.028 [7.55 ± 0.09]b
5.3 [5.28 ± 0.09]b
1.1 [6.00 ± 0.11]
0.73 [6.14 ± 0.12]b
ACS Paragon Plus Environment
1
2

Page 13 of 59
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
The apparent binding affinity constant, Ki, was obtained from binding experiments on rat brain cell
membranes, whereas the functional potencies of the antagonist analogues were obtained from
concentration-inhibition experiments as IC50 values. In functional studies, GABA Rs were transiently
A
a
b
expressed as: α1β2γ2 or α1β3γ2 in HEK293 cells and tested in whole-cell voltage clamp experiments, or
c
d
e
29
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
as human α1β2γ2 in tsA201 cells and tested in FMP Red or Blue assay. Data from Frølund et al., 2005 ,
f
14
and Data from Møller et al., 2010 . Mean values (Ki and IC50) are shown, along with pKi ± SEM and
pIC ± SEM values, from 4 – 8 experiments.
5
0
The benzophenone-4-PIOL photoaffinity probes, 1a–d, all showed binding affinities in the high
to low nanomolar range, most likely because the lipophilic benzophenone affects the desolvation
energy, which is not inherent in the lead molecule, 4-PIOL (K 9.1 µM, Table 1). The observation
i
that the bulky benzophenone analogues 1a–d bind to the orthosteric GABA R binding site with
A
high apparent affinity further supports the presence of cavities capable of accommodating
relatively large substituents near the 4-PIOL scaffold.
The highest binding affinity and functional potency was observed for analogue 1a (pK : 7.75 ±
i
0
.01; and pIC : 8.09 ± 0.12, respectively; Table 1), where the benzophenone substituent is directly
50
connected to the 4-PIOL scaffold via the 3-phenyl position. This affinity/potency was >10-fold
higher than those observed for the longer analogues, 1b and 1c, as well as for the same length
analogue with benzophenone connected to 4-PIOL via the adjacent 4-phenyl position (Fig. 2A-B;
Table 1).
ACS Paragon Plus Environment
1
3

ACS Chemical Neuroscience
Page 14 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
A
B
1
00
GBZ
1a
8
7
6
5
0.01
0.1
1
7
5
2
5
0
5
1b
1c
1
d
2c
2d
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0
10
1
a 1b 1c 1d 2a 2b 2c 2d
[
analogue], M
1
0 µM 1a-d
C
1
0 µM GABA
UV
1
1
1
1
a
b
c
5
4
00 pA
1
0 s
UV
00 pA
10 s
UV
800 pA
UV
1
1
0 s
0 s
d
6
00 pA
Figure 2. (A) Ligand concentration inhibition curves for 4-PIOL [1a-d] and 4-PHP analogues [2a-d]
on α1β2γ2 GABA Rs expressed in HEK293 or tsA201 cells using whole-cell patch clamping or FMP assay
A
(see Table 1 for details); mean ± sem, n = 4-12; (B) Potencies of analogues are shown as pIC ± SEM
5
0
[
linear y-axis, left] and as mean IC [Log y-axis, right]; (C) GABA current examples showing that the 4-
50
PIOL analogues all failed to induce UV (λ ) photo-inhibition (i.e. no depression of the GABA induced
3
75
currents post-UV treatment).
The four 4-PIOL analogues, 1a-d, were examined in whole-cell patch clamp experiments for their
ability to photo-inactivate α1β2γ2 GABA Rs expressed in HEK293 cells (Fig. 2C). A
A
concentration of 10 µM GABA, which induces a sub-maximal current with this receptor isoform,
was efficiently blocked by co-application with 10 µM of either of the competitive antagonist
analogues, 1a-d. The orthosteric binding sites were then fully occupied by 10 µM of 1a-d while
ACS Paragon Plus Environment
1
4

Page 15 of 59
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
being exposed to UV light for 10 s to induce photo-activation of the analogues, with the intention
of covalent bonding to residues within the GABA binding site. After thorough wash-out of non-
covalently bound 4-PIOL analogues, 10 µM of GABA was re-applied to the cells. However, none
of the four analogues (1a-d) irreversibly inhibited the GABA-induced currents which were of
identical amplitudes to those observed prior to UV treatment (Fig. 2C). This suggested a failure of
covalent binding to the GABA site. We next chose to assess if the 4-PHP scaffold would be a
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
better lead molecule in the development of photo-reactive GABA R blockers.
A
Pharmacological characterization of the photoactive 4-PHP GABA R ligands.
A
We have previously observed that introducing a phenyl group into either position 3- or 5- of 4-
PHP, the 5-phenyl-4-PHP analogue displayed a >10-fold higher affinity, and functional potency
as an antagonist at GABA Rs, in comparison with 3-phenyl-4-PHP (pK : 7.65 ± 0.02 and pIC :
A
i
50
9
6.81 ± 0.04 versus pK : 6.57 ± 0.01 and pIC : 5.73 ± 0.06, respectively; Table 1) . This preference
i
50
for the 5-position corresponded well with the data we obtained when introducing an amino group
into the phenyl group of either 3-phenyl-4-PHP or 5-phenyl-4-PHP, 2a (pK : 6.57 ± 0.06 and pIC :
i
50
5
.75 ± 0.11; Table 1) and 2b (pK : 7.79 ± 0.07 and pIC : 7.19 ± 0.05; Fig. 2B, Table 1),
i 50
respectively, where no impact was observed on either ligand affinity for, or potency at, GABA Rs.
A
After establishing that an amino substituent was tolerated in the 3-position on the phenyl ring, we
next replaced the amino group in 2a and 2b with a photo-reactive azide group to generate 2c
(analogous to 2a) and 2d (analogous to 2b). With these analogues we observed no change in
affinity, but a modest reduction in the inhibitory potency when compared with 2a and 2b (2c, pKi:
6.66 ± 0.04 and pIC : 6.68 ± 0.07; and 2d, pK : 7.66 ± 0.08 and pIC : 7.55 ± 0.09; Fig. 2B, Table
5
0
i
50
ACS Paragon Plus Environment
1
5

ACS Chemical Neuroscience
Page 16 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
). As the photo-reactive 5-(3-azidophenyl)-4-PHP, 2d, displayed significantly higher affinity and
potency for α1β2γ2 GABA Rs than 3-(3-azidophenyl)-4-PHP (2c; both p < 0.0001), we decided
A
to proceed by testing the photo-inhibition properties of 2d.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
We performed photo-inhibition experiments similar to those with the benzophenone-4-PIOL
analogues (1a-d), but now using 5-(3-azidophenyl)-4-PHP (2d). Interestingly, a substantive
irreversible inhibition (~70%) of 10 µM GABA currents was observed (Figure 3A, B), indicating
efficient covalent integration of the UV-activated 2d analogue into the GABA binding site. The
GABA EC values before and after UV treatment in the absence of a PHP ligand were identical
50
(pEC : 5.39 ± 0.18 [EC : 4.1 µM, n = 5], and pEC : 5.16 ± 0.24 [EC : 6.8 µM, n = 5],
5
0
50
50
50
respectively; p = 0.48; Figure 3C), which suggested that the function of GABA Rs in the cell
A
preparation were unaffected by UV exposure. However, in the presence of PHP ligand, the
pronounced depression of the maximum GABA current responses was indicative of irreversible
antagonist behavior by 2d at α1β2γ2 GABA Rs after UV; i.e. a covalent block of ~75% of all
A
GABA binding sites (Figure 3C). Interestingly, this new photo-reactive azide-5-phenyl-4-PHP
ligand, 2d, was more effective than our previous photo-antagonist which was based on gabazine,
1
1
GZ-B1, which maximally achieved ~50% irreversible inhibition . We attribute the improved
photo-inhibition to the higher affinity of 2d for the GABA binding site compared with that of GZ-
1
1
B1 (2d: 23 nM; versus GZ-B1: 153 nM ).
It is possible that further photo-inhibition could have been achieved with multiple rounds of UV
exposure with 2d, but since damage to cell health using such protocols is significant and thus
detrimental to the longevity of electrophysiology recordings, we chose not to pursue this.
ACS Paragon Plus Environment
1
6

Page 17 of 59
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
A
B
C
Before UV
1
00
100
80
60
40
20
0
100
After UV in 10 uM 2d
80
8
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
60
6
4
4
0
0
0
2
20
0
-6 -4 -2 0
2
4
6
8 10 12 14 16 18 20
1n 10n 100n 1µ 10µ 100µ 1m 10m
Recording time (min)
Control
2d
[GABA], M
Figure 3. (A) Efficient photo-inactivation of 122 GABA Rs expressed in HEK293 cells by 2d (10
A
µM); 10s of UV [hatched bar], n = 10; (B) level of photo-inhibition 5 min after UV exposure of Krebs
solution [control] or 10 µM 2d and washout of non-covalently bound drug; (C) GABA concentration-
response curves before and after UV exposure with 2d showing irreversible inhibition (‘before-UV’ Hill
curve parameters: Vmax: 98 ± 1.2 %, EC : 4.1 μM, slope: 1.4 ± 0.1, n = 5; ‘after-UV’ Hill curve
5
0
parameters: Vmax: 34 ± 2.8 %, EC : 6.8 μM, slope: 1.1 ± 0.1, n = 5). All data are shown as mean ± sem.
5
0
Given the increased inhibitory efficacy we developed a novel biotin-linked 2d-analogue for
receptor trafficking studies. The advantage of adapting a ligand such as 2d into a trafficking probe,
is that we can study native GABA Rs without the need for prior introduction of epitope tags or
A
mutations to the receptors.
Design, synthesis and pharmacological characterization of photo-conjugates engineered for
photo-cross-linked imaging
In previous trafficking studies of GABA receptors (both GABA and GABA receptors) we have
A
B
relied on the introduction of an α-bungarotoxin binding site mimotope into GABA R subunits and
A
ACS Paragon Plus Environment
1
7

ACS Chemical Neuroscience
Page 18 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
coupling α -bungarotoxin to quantum dots (QDs) via biotin-streptavidin ultra-high affinity
8
binding . The bright fluorescence emitted by the QDs could then be tracked in live imaging studies
3
4
of receptor movements in cell membranes of hippocampal neurons .
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To assess native GABA Rs we successfully developed a gabazine analogue GZ-B1-biotin which
A
4
could be used in trafficking studies . However, despite some success, this ligand did not display
sufficiently robust specific binding. We therefore developed another GABA R photo-probe based
A
on the 4-PHP analogue, 2d, reasoning that the increased photo-inhibition, and higher binding
affinity and potency over GZ-B1 might improve the level of receptor labelling.
The hydrophilic polyethylene glycol (PEG) linker utilized in GZ-B1-biotin was very short (only 2
units long), so we speculated that this could be one reason why specific binding to streptavidin-
3
5
QD655 was compromised. As the QD655 molecule (6-12 nm ) is of a similar size to the extracellular
part of a GABA R (6-9 nm; measurements made on the cryo-EM structure of human α1β3γ2
A
1
8
GABA Rs with truncated N-terminus, pdb: 6I53 ) it is reasonable to assume that more distance
A
between the two globular entities (GABA R ECD and QD -streptavidin/biotin-2d), would be
A
655
required to allow unhindered specific binding by the tethered ligand.
ACS Paragon Plus Environment
1
8

Page 19 of 59
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. 3b bound to the human α1β3γ2 GABA R orthosteric binding site. Receptor carbon atoms are
A
shown in gray, 3b carbon atoms in green, hydrogen bonds shown as yellow dashes.
We estimated the optimal PEG linker length from ligand-receptor docking studies of various target
2
d-photoprobe analogues into the GABA binding site (Figure 4). Allowing for PEG flexibility, for
plenty of space for the PEG linked to the PHP-core to exit the binding site, and for the conjugated
biotin to gain some distance from the GABA R and be able to interact with streptavidin
A
unhindered, we deduced a longer 23-unit PEG (PEG ) linker was required.
2
3
To study the effects of the biotin attachment via a PEG on the binding affinity to and functionality
23
of the GABA R, we synthesized analogues based on both 2c and 2d, leading to the target
A
analogues, 3a and 3b, respectively.
ACS Paragon Plus Environment
1
9

ACS Chemical Neuroscience
Page 20 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Scheme 3 a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Reagents and conditions: (i) CDI, Benzyl glycinate, DMF, 0 C to rt., (ii), (Bpin) , KOAc,
2
Pd(dppf)Cl ·CH Cl , 1,4-Dioxane, 100 C, (iii) Compound 16, 10f or 10g, K CO , Pd(PPh ) , DMF/H O
2
2
2
2
3
3 4
2
t
(2:1), 90 C, (iv) H , 10% Pd/C, DMF, rt., (v) TMS-N , BuNO , MeCN/THF, 0 C to rt., (vi) PyBOP,
2
3
2
DiPEA, Biotin-PEG -NH , DMF, rt., (vii) TFA, CH Cl , 0 C to rt.
2
3
2
2
2
The synthesis was performed as described in Scheme 3 starting from commercially available 4-
bromo-2-nitrobenzoic acid (14). Extending the carboxylic acid with a glycine benzyl ester using
carbonyldiimidazole formed compound 15, which was converted into a suitable Suzuki-Myiaura
cross coupling reagent using bis(pinacolato)diboron (Bpin ) and Pd(dppf)Cl ·CH Cl in anhydrous
2
2
2
2
ACS Paragon Plus Environment
2
0

Page 21 of 59
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
1
,4-dioxane at elevated temperatures (MW irradiation). The resulting compound 16 (black tar) was
observed unstable towards standard purifications (flash column chromatography and standard
normal and reverse phase HPLC, data not shown) and was prepared freshly before each cross-
coupling reaction and without purification.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Suzuki-Myiaura cross-coupling between compound 16 and either 10f or 10g afforded compounds
1
7a and 17b respectively, where the loss of the benzyl ester of the glycine residue was observed.
Compounds 17a,b were subjected to catalytic hydrogenation using palladium on carbon and the
resulting anilines (18a,b) were converted into aromatic azides (19a,b) using tetramethylsilyl azide
and tert-butyl nitrite. Incorporation of the Biotin-PEG -NH moiety in compound 20a,b was
2
3
2
performed by applying peptide chemistry methodologies using PyBOP and DiPEA at room
temperature. The resulting 20a,b were then deprotected under acidic conditions (trifluoroacetic
acid) to afford compounds 3a,b.
Binding and functional experiments with 3a and 3b, re-affirmed that substitution in the 5-position
in 4-PHP, 3b (pK : 6.03 ± 0.11 and pIC : 6.14 ± 0.12), resulted in more potent ligands compared
i
50
to substitution in the 3- position, 3a (pK : 4.54 ± 0.10 and pIC : 5.28 ± 0.09; Figure 5B and Table
i
50
1
). However, the PEG -biotin conjugation resulted in a 25-fold loss of affinity/potency when
23
compared with the parent analogues (2c and 2d; Table 1).
ACS Paragon Plus Environment
2
1

ACS Chemical Neuroscience
Page 22 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
A
3a,b
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
B
C
UV + 3b
1
00
1
00
7
5
0
5
0
80
6
0
0
5
4
GBZ
3
3
2
a
b
20
0
-5
0
5 10 15 20 25 30
[analogue], M
Recording time (min)
Figure 5. (A) Generalised chemical structures of 3a and 3b, showing biotin linked to the parent analogues
[2d and 2c, respectively] via a PEG linker; (B) Ligand concentration inhibition response curves for 3a and
3b in relation to gabazine (GBZ, n = 6 -12); (C) irreversible UV photo-inhibition by 3b (10 µM; n = 7; 10
sec of UV at t = 0 [hatched bar]). Whole-cell patch clamp data in (B and C) was from α1β2γ2 GABA Rs
A
expressed in HEK293 cells; all data are mean ± sem.
Photo-inhibition experiments successfully showed that the level of UV-stimulated irreversible
block of recombinant α1β3γ2 GABA Rs was largely unchanged for 3b (~70%) when compared
A
with 2d (~70–75%), indicating that the photo-reactive azide group remains functionally available
after ligand binding to the orthosteric site.
The final step in our assessment of 3b was to test its utility at labelling and tracking of GABA Rs
A
using QD -streptavidin. Unfortunately, after having exhausted several different methods, and a
655
range of concentrations, incubation times, UV exposures, we failed to obtain any useful labelling
of receptors (Supplementary Fig. S2; video 1). This was very disappointing as controls of our
ACS Paragon Plus Environment
2
2

Page 23 of 59
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
methodology with an engineered GABA 1 subunit with an N-terminus bungarotoxin site
A
(1BBS) expressed with 2 and 2 subunits were perfectly fine with clear demonstration of specific
binding (Supplementary Fig. S3; video 2).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
We deduce that the reason for the lack of specific binding of 3b-QD655 to GABA Rs, probably
A
resides with the long PEG which may have coiled or become adsorbed to the protein surface
2
3
potentially obscuring the biotin molecule, thereby hindering binding to streptavidin. However, the
observed affinity and potency allude to dynamic regions near the orthosteric binding site and
connectors to the extracellular environment, adaptable to substituents of various nature and size.
Conclusion
We have synthesized and characterized a series of novel 4-PIOL and 4-PHP photo-probes. All
these bulky photo-reactive analogues (1a-d, 2c-d) show higher affinity and potency for the GABA
binding site than their lead compounds, indicating considerable spaciousness in the GABA binding
site and additional anchoring points between ligand and the GABA R. The 4-PHP azide-analogue,
A
2
d, showed remarkable nanomolar affinity, as well as efficient photo-inactivation for the GABA R
A
and can serve as a useful research tool of native receptors. Combining this probe with precise UV-
delivery could prove to be a powerful technology in the advancement of our understanding of
GABA R function and effect.
A
ACS Paragon Plus Environment
2
3

ACS Chemical Neuroscience
Page 24 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Methods
Chemistry
Compounds 429-30, 9 , 10a,c,d , 3-benzoylbenzaldehyde , and 4-benzoylbenzaldehyde were all
synthesized as described in the literature. All reagents and solvents (reagent or chromatography
grade) were obtained from commercial suppliers and used without further purification. Air- and/or
moisture-sensitive reactions were performed under a nitrogen atmosphere using syringe-septum
cap techniques and with the use of flame-dried glassware. Anhydrous solvents were obtained by
using a solvent purification system (THF) or by storage over 3 Å molecular sieves. Thin layer
chromatography (TLC) was carried out using Merck silica gel 60 F254 plates, and compounds were
36
7
32
33
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
visualized using UV (254 and 366 nm), KMnO or Ninhydrin spray reagent. Flash column
4
chromatography (FC) and dry column vacuum chromatography (DCVC) were performed using
Merck silica gel (0.040‒0.063 mm or 0.015–0.040 mm, respectively). Preparative reversed phase
HPLC was performed on an Agilent 1100 system, equipped with two preparative solvent delivery
units, a multiple wavelength detector (210 or 254 nm), an autosampler injector (1 mL Loop) and
an optional fraction collector, using a preparative Phenomenex Luna C18(3) column (21.2 mm ×
250 mm, 5 m, 100 Å) and Eluent A (acetonitrile/water/HCOOH, 5:95:0.1) and Eluent B
(acetonitrile/water/HCOOH, 95: 5:0.1) at a flowrate of 20 mL/min. The column was operated at
rt. For HPLC control, data collection, and data handling, ChemStation ver. 3.02 was used. Melting
1
points were recorded on a SRS OptiMelt apparatus in open capillary tubes and are uncorrected. H
13
and C NMR data were recorded on a 300 MHz Varian Mercury 300BB spectrometer equipped
1
with a 5 mm H(BB) probe, a 300 MHz Varian Gemini 2000BB spectrometer equipped with a 5
3
1
13
1
19
mm P, C( H, F) probe, a Bruker Avance 400 MHz spectrometer equipped with a 5 mm
1
19
PABBO BB( H, F) Z-GRD probe, a Bruker Avance 600 MHz spectrometer equipped with a
ACS Paragon Plus Environment
2
4

Page 25 of 59
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
cryogenically cooled 5 mm CPDCH 13C( H) Z-GRD probe, or a Bruker Avance 600 MHz
1
1
13
15
spectrometer equipped with a cryogenically cooled 1.7 mm CPTCI H( C, N) Z-GRD probe at
300 K. Data are tabulated in the following order: chemical shift (δ) [multiplicity (b, broad; s,
singlet; d, doublet; t, triplet; q, quartet; p, pentet; m, multiplet), coupling constant(s) J (Hz), number
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
7
of protons. The solvent residual peak was used as internal reference. Quantification of salt-free
3
a and 3b were performed by the standard Bruker method, ERETIC2 in Topspin 3.5, using pulse
1
width optimized 1D H experiments with long relaxations delay between transients, with cinnamic
acid as external standard. Analytical High performance liquid chromatography (anal. HPLC) was
performed on a Merck-Hitachi HPLC system consisting of an L-7100 pump, an L-7200 auto
sampler, and an L-7400 UV detector (210 or 254 nm), using a Chromolith SpeedROD RP-18
column (4.6 × 50 mm). A linear gradient elution was applied with eluent A (H O/TFA 100:0.01)
2
containing 0% of solvent B (MeCN/H O/TFA, 90:10:0.01) rising to 100% of B over 3.5 min with
2
a flow rate of 4.0 mL/min. The purity of the analyzed compounds is ≥95% unless otherwise stated.
Data were acquired and processed using the EZChrom Elite Software version 3.1.7 by Hitachi.
HPLC-HRMS analyses were performed on a system comprised of an Agilent 1200 HPLC system
comprising of a quaternary pump with a built-in degasser, a thermostated column compartment,
an autosampler, and a photodiode array detector, coupled with a Bruker microOTOF-QII mass
spectrometer equipped with an electrospray ionization (ESI) source and operated via a 1:99 flow
splitter. Mass spectra were acquired in positive ionization mode, using drying temperature of 200
°C, a capillary voltage of −4100V, nebulizer pressure of 2.0 bar, and drying gas flow of 7 L/min.
A solution of sodium formate clusters was injected in the beginning of each run to enable internal
mass calibration. Chromatographic separation was acquired on a Phenomenex Luna C18(2)
column (150 mm × 4.6 mm, 3 μm, 100 Å) maintained at 40 °C, using a flow rate of 0.8 mL/min
ACS Paragon Plus Environment
2
5

ACS Chemical Neuroscience
Page 26 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
and a linear gradient of the binary solvent system water-acetonitrile-formic acid (eluent A:
9
5/5/0.1, and eluent B: 5/95/0.1) rising from 0% to 100% of eluent B over 20 minutes. Data was
acquired using Compass HyStar Ver. 3.2 (Bruker Daltonic GmbH, Germany) and processed using
Compass DataAnalysis Ver. 4.0 (Bruker Daltonic GmbH, Germany).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(3-(3-hydroxy-5-(piperidin-4-yl)isoxazol-4-yl)phenyl)(phenyl)methanone (1a). A solution
of 5a (0.17 g, 0.4 mmol) in 33% HBr in AcOH (4 mL) was stirred at 65 °C for 18 h. Upon cooling
to rt., the reaction mixture was evaporated in vacuo. Recrystallization from EtOH/Et O afforded
2
1
compound 1a as white crystals (0.48 g, 37%): mp 209–211 °C. H NMR (300 MHz, D O): δ 7.76–
2
1
3
7
.47 (m, 9H), 3.46–3.42 (m, 2H), 3.25–3.15 (m, 1H), 3.04–2.94 (m, 2H), 2.08–1.96 (m, 4H). C
NMR (300 MHz, D O): δ 200.1, 170.5, 168.7, 137.4, 136.7, 133.9, 130.8, 130.5, 130.0, 129.4,
2
1
28.8, 128.2, 112.7, 107.2, 43.6, 32.0, 26.1.
4-(3-hydroxy-5-(piperidin-4-yl)isoxazol-4-yl)phenyl)(phenyl)methanone (1b). A solution
(
of 5b (0.56 g, 1.2 mmol) in 33% HBr in AcOH (4 mL) was stirred at 65 °C for 18 h. Upon cooling
to rt., the reaction mixture was evaporated in vacuo. Recrystallization from MeOH/H O afforded
2
1
compound 1b as white crystals (0.25 g, 60%): mp. 227–221 °C (decomp.). H NMR (300 MHz,
DMSO-d ): δ 11.97 (b s, 1H), 8.63 (b s, 1H,), 7.82–7.55 (m, 9H), 3.39–3.34 (m, 3H), 3.12–2.97
6
1
3
(
m, 2H), 2.05–1.91 (m, 4H). C NMR (300 MHz, DMSO-d ): δ 195.9, 171.1, 168.9, 137.7, 136.4,
6
1
33.8, 133.5, 130.9, 130.4, 129.4, 129.4, 106.4, 46.8, 32.5, 27.2.
3-((3-hydroxy-5-(piperidin-4-yl)isoxazol-4-yl)methyl)phenyl)(phenyl)methanone (1c). A
(
solution of 6c (0.25 g, 0.5 mmol) in 33% HBr in AcOH (5 mL) was stirred at 65 °C for 18 h. Upon
cooling to rt., the reaction mixture was evaporated in vacuo. Recrystallization from EtOH/water
1
afforded compound 1c as white crystals (0.17 g, 89%): mp 225–227 °C (decomp.). H NMR (300
MHz, D O): δ 7.69–7.38 (m, 9H), 3.70 (s, 2H), 3.47–3.38 (m, 2H), 3.19–2.91 (m, 3H), 1.99–1.82
2
ACS Paragon Plus Environment
2
6

Page 27 of 59
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
13
(
m, 2H). C NMR (300 MHz, D O): δ 200.9, 171.1, 170.2, 139.4, 137.3, 137.0, 133.7, 133.4,
2
1
30.4, 130.1, 129.2, 128.7, 128.5, 105.0, 43.7, 31.7, 26.0.
(4-((3-hydroxy-5-(piperidin-4-yl)isoxazol-4-yl)methyl)phenyl)(phenyl)methanone (1d). A
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
solution of 6d (0.19 g, 0.4 mmol) in 33% HBr in AcOH (4 mL) was stirred at 65 °C for 18 h. Upon
cooling to rt. the reaction mixture was evaporated in vacuo. Recrystallization from water afforded
1
compound 1d as white crystals (0.14 g, 94%): mp 118–120 °C (decomp.). H NMR (300 MHz,
D O): δ 7.45–7.32 (m, 5H), 7.24 (t, J = 7.0 Hz, 2H), 7.10 (d, J = 6.75 Hz, 2H), 3.60 (s, 2H), 3.42–
2
1
3
3
1
.38 (m, 2H), 3.13–2.92 (m, 3H), 2.01–1.78 (m, 4H). C NMR (300 MHz, D O): δ 197.8, 170.7,
2
69.8, 144.9, 136.7, 134.8, 133.0, 130.6, 129.9, 128.4, 128.3, 104.2, 43.6, 31.6, 30.8, 26.4, 26.1.
3-(3-aminophenyl)-4-(piperidin-4-yl)-1H-pyrazolol (2a). Compound 11a (0.24 g, 0.6 mmol)
was dissolved in 35% aq. HCl (10 mL) and refluxed at 130 C for 2h. The reaction mixture was
cooled to rt. and evaporated in vacuo. The resulting residue was re-evaporated twice with water
(10 mL) and twice with toluene (10 mL). Purification by recrystallization from MeOH/Et O
2
1
brought 2a (115 mg, 58%) as light brown crystals: mp 212–214 C. H NMR (600 MHz, MeOD-
d4): δ 7.71–7.69 (m, 1H), 7.64 (t, J = 1.8 Hz, 1H), 7.62 (t, J = 7.9 Hz, 1H), 7.55 (s, 1H), 7.38 (ddd,
J = 7.7, 2.0, 0.7 Hz, 1H), 3.46–3.41 (m, 2H), 3.25–3.17 (m, 3H), 2.15–2.10 (m, 2H), 1.85–1.77
1
3
(
m, 2H). C NMR (150 MHz; MeOD-d ): δ 140.1, 135.5, 131.2, 130.3, 127.9, 122.0, 454.7, 121.6,
4
1
20.4, 44.0, 30.1, 29.7. Purity by anal. HPLC (254 nm), > 99%.
5-(3-Aminophenyl)-4-(piperidin-4-yl)-1H-pyrazolol (2b). 11b (0.24 mg, 0.6 mmol) was
dissolved in 35% aq. HCl (100 mL) and refluxed at 130 C for 1 h. The reaction mixture was
cooled to rt. and evaporated in vacuo. The resulting residue was re-evaporated with 48% aq. HBr
in vacuo. Purification by recrystallization from MeOH/Et O brought 2b (0.19 g, 23%) as off white
2
1
crystals: mp > 200 C. H NMR (300 MHz, MeOD-d ): δ 7.81–7.70 (m, 3H), 7.70–7.60 (m, 2H),
4
ACS Paragon Plus Environment
2
7

ACS Chemical Neuroscience
Page 28 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
13
3
.43 (d, J = 12.3 Hz, 2H), 3.25–3.00 (m, 3H), 2.12–1.90 (m, 4H). C NMR (75 MHz, MeOD-d ):
4
δ 134.3, 132.5, 131.8, 131.4, 129.8, 129.4, 125.6, 125.3, 122.9, 45.1, 30.8, 30.7. Anal. calcd.
(C H N O·2HBr·1H O): C, 38.38; H, 5.06; N, 12.79. Found: C, 38.50; H, 4.97; N, 12.30.
1
4
18
4
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
-(3-Azidophenyl)-4-(piperidin-4-yl)-1H-pyrazolol (2c). In an amberized vial, 13a (75 mg,
0
3
.2 mmol) was dissolved in CH Cl (6 mL) and carefully added TFA (1 mL) dropwise at rt. After
2 2
.5 h, the reaction mixture was evaporated in vacuo and the resulting residue was lyophilized to
1
afford compound 2c (77 mg, 97%) as amorphous solid. H NMR (600 MHz, DMSO-d ):  12.59
6
(
b s, 1H), 8.76–8.62 (m, 1H), 8.50–8.35 (m, 1H), 7.60 (s, 1H), 7.48 (t, J = 7.7 Hz, 1H), 7.41–7.39
(m, 1H), 7.24–7.23 (m, 1H), 7.09 (ddd, J = 1.0, 2.4, 7.9 Hz, 1H), 3.34–3.29 (m, 2H), 3.06 (tt, J =
13
3
.5, 11.9 Hz, 1H), 3.03–2.96 (m, 2H), 1.99–1.93 (m, 2H), 1.67 (qd, J = 3.9, 13.2 Hz, 2H). C
NMR (150 MHz, DMSO-d ):  158.2 (q, J = 34.2 Hz, TFA), 139.7, 139.1, 135.5, 130.3, 123.9,
6
121.5, 120.2, 117.8, 117.4, 116.3 (q, J = 295.4 Hz, TFA), 29.9, 29.4. HRMS (ESI-TOF): m/z
+
calculated for C H N O [M + H] , 285.1458. Found, 285.1465 (M = 2.3 ppm). Purity by anal.
1
4
17
6
HPLC (254 nm), 96%.
5-(3-azidophenyl)-4-(piperidin-4-yl)-1H-pyrazol-1-ol (2d). A suspension of 13b (110 mg,
0
.22 mmol) in EtOH/Et O (1:1, 10 mL) was cooled to –78 C where upon gaseous HCl was
2
bubbled into the solution for one minute. After 5 minutes, the solution turned red and additional
Et O (10 mL) was added. The stirring was continued for 10 minutes (reaction was followed by
2
TLC) before the solvents were removed in vacuo. Recrystallization from EtOH/Et O afforded
2
1
compound 2d (34 mg, 55%) as amorphous solid. H NMR (300 MHz, D O): δ 7.34 (t, J = 8.26Hz,
2
1H), 7.14 (s, 1H), 7.09–6.99 (m, 2H), 6.95 (m, 1H), 3.23 (d, J = 12.7 Hz, 2H), 2.79 (td, J = 2.3,
1
2.7 Hz, 2H), 2.67 (tt, J = 3.7, 11.8 Hz, 1H), 1.81 (d, J = 14.3 Hz, 2H), 1.61 (qd, J = 4.1, 14.2 Hz,
ACS Paragon Plus Environment
2
8

Page 29 of 59
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
13
2
4
H). C NMR (75 MHz, D O): δ 140.8, 133.1, 130.6, 129.7, 129.5, 126.3, 121.2, 120.2, 112.0,
2
4.4, 29.8, 29.5. Purity by anal. HPLC, >99% (UV).
2-azido-N-(2,76-dioxo-80-(2-oxohexahydro-1H-thieno[3,4-d]imidazol-4-yl)-
,9,12,15,18,21,24,27,30,33,36,39,42,45,48,51,54,57,60,63,66,69,72-tricosaoxa-3,75-
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
diazaoctacontyl)-4-(1-hydroxy-4-(piperidin-4-yl)-1H-pyrazol-5-yl)benzamide (3b). In an
amberized vial and under an argon atmosphere, a solution of 20b (258 mg, 0.15 mmol) in CH Cl
2
2
(
3 mL) was added neat TFA (3 mL) dropwise. The resulting mixture was stirred for 1 h before it
was evaporated in vacuo. Purification of the resulting residue by preparative HPLC (1029% B,
over 23 minutes) afforded compound 3b (73 mg, 26%) as purple oil. 1.83 mg 3b were dissolved
in 492 L methanol-d (corresponding to a theoretical concentration of 2.0 mM) in a 1.7 mm NMR
4
tube and used to quantification by ERETIC2. Using a 4.00 mM external reference sample of
cinnamic acid in methanol-d , the concentration of 3b was found to be 1.92 mM, translating to a
4
1
molecular weight of 3b of 1940.8 g/mol. H NMR (MeOD-d , 600 MHz):  7.97 (d, J = 8.0 Hz,
4
1
H), 7.46 (d, J = 1.3 Hz, 1H), 7.40 (dd, J = 8.0, 1.3 Hz, 1H), 7.27 (s, 1H), 4.50 (dd, J = 7.8, 4.9
Hz, 1H), 4.31 (dd, J = 7.8, 4.5 Hz, 1H), 4.11 (s, 2H), 3.64-3.61 (m, 88H), 3.60 (t, J = 5.4 Hz, 2H),
.54 (t, J = 5.4 Hz, 2H), 3.45 (t, J = 5.4 Hz, 2H), 3.44 (m, 2H), 3.36 (t, J = 5.4 Hz, 2H), 3.20 (dt,
3
J = 9.6, 4.8 Hz, 1H), 3.07 (td, J = 13.2, 2.3 Hz, 2H), 2.94 (m, 1H), 2.92 (dd, J = 12.7, 5.0 Hz, 1H),
2.71 (d, J = 12.7 Hz, 1H), 2.22 (t, J = 7.3 Hz, 2H), 2.05 (br d, J = 14.3 Hz, 2H), 1.84 (qd, J = 13.4,
3
.7 Hz, 2H), 1.74 (m, 1H), 1.67 (m, 2H), 1.61 (m, 1H), 1.44 (p, J = 7.7 Hz, 2H). HRMS (ESI-
3
+
TOF): m/z calculated for C H N O S [M + 3H] , 556.3935. Found, 556.3048 (M = 2.5 ppm).
7
5
134 11 28
2
-azido-N-(2,76-dioxo-80-(2-oxohexahydro-1H-thieno[3,4-d]imidazol-4-yl)-
6
,9,12,15,18,21,24,27,30,33,36,39,42,45,48,51,54,57,60,63,66,69,72-tricosaoxa-3,75-
ACS Paragon Plus Environment
2
9