Antioxidant-Conjugated 1,2,4-Triazolo[4,3- a]pyrazin-3-one Derivatives: Highly Potent and Selective Human A2A Adenosine Receptor Antagonists Possessing Protective Efficacy in Neuropathic Pain

Article abstract of DOI:10.1021/acs.jmedchem.9b00778

New 8-amino-6-aryl-1,2,4-triazolo[4,3-a]pyrazin-3-ones were designed to obtain dual antioxidant-human A_2A adenosine receptor (hA_2A AR) antagonists. Two sets of compounds were synthesized, the first featuring phenol rings at the 6-pos

Full text of DOI:10.1021/acs.jmedchem.9b00778

Subscriber access provided by Nottingham Trent University
Article
Antioxidant-conjugated 1,2,4-Triazolo[4,3-a]pyrazin-3-one Derivatives:
Highly Potent and Selective Human A2A Adenosine Receptor
Antagonists Possessing Protective Efficacy in Neuropathic Pain
Matteo Falsini, Daniela Catarzi, Flavia Varano, Costanza Ceni, Diego Dal Ben, Gabriella
Marucci, Michela Buccioni, Rosaria Volpini, Lorenzo Di Cesare Mannelli, Elena
Lucarini, Carla Ghelardini, Gianluca Bartolucci, Marta Menicatti, and Vittoria Colotta
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b00778 • Publication Date (Web): 27 Aug 2019
Downloaded from pubs.acs.org on August 27, 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 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Antioxidant-conjugated 1,2,4-Triazolo[4,3-a]pyrazin-3-one
Derivatives: Highly Potent and Selective Human A2A
Adenosine Receptor Antagonists Possessing Protective
Efficacy in Neuropathic Pain
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
a
a
a
b
Matteo Falsini, Daniela Catarzi, Flavia Varano, Costanza Ceni, Diego Dal Ben, Gabriella
b
b
b
c
c
Marucci, Michela Buccioni, Rosaria Volpini, Lorenzo Di Cesare Mannelli, Elena Lucarini,
c
a
a
a*
Carla Ghelardini, Gianluca Bartolucci, Marta Menicatti, Vittoria Colotta
aDipartimento di Neuroscienze, Psicologia, Area del Farmaco e Salute del Bambino, Sezione di Farmaceutica e
Nutraceutica, Università degli Studi di Firenze, Via Ugo Schiff, 6, 50019 Sesto Fiorentino, Italy.
bScuola di Scienze del Farmaco e dei Prodotti della Salute, Università degli Studi di Camerino, via S. Agostino 1,
6
2032 Camerino (MC), Italy.
cDipartimento di Neuroscienze, Psicologia, Area del Farmaco e Salute del Bambino, Sezione di Farmacologia e
Tossicologia, Università degli Studi di Firenze, Viale Pieraccini 6, 50139 Firenze, Italy.
Key words: G protein-coupled receptors, A2A adenosine receptor antagonists, 1,2,4-triazolo[4,3-
a]pyrazin-3-one, neuropathic pain, ligand-adenosine receptor modeling studies.
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 69
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
New 8-amino-6-aryl-2-phenyl-1,2,4-triazolo[4,3-a]pyrazine-3-ones were designed to obtain dual
antioxidant-human A2A adenosine receptor (hA2A AR) antagonists. Two sets of compounds were
synthesized, the first featuring phenol rings at the 6-position, the second bearing the lipoyl and 4-
hydroxy-3,5-di-terbut-benzoyl residues appended by different linkers on the 6-phenyl ring. Several
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
new triazolopyrazines (1-21) were potent and selective hA2A AR antagonists (K = 0.17-54.5 nM).
i
Compounds 11, 15 and 21, featuring antioxidant moieties, and compound 12, lacking the
antioxidant functionality, reduced oxaliplatin–induced toxicity in microglia cells, the most active
being the lipoyl-derivative 15 and the (4-hydroxy-3,5-di-tert-butyl)phenyl- analogue 21 which were
effective in reducing the oxygen free radical level. The lipoyl-derivative 15 was also able to revert
oxaliplatin-induced neuropathy in mouse. In vivo efficacy of 15 makes it a promising
neuroprotective agent in oxidative stress-related diseases.
2
ACS Paragon Plus Environment

Page 3 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
The endogenous nucleoside adenosine affects many pathophysiological conditions through
activation of G protein-coupled receptors classified as A , A , A and A receptors. A AR
1
2A
2B
3
2A
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
stimulation increases adenylate cyclase activity and cAMP production, thus activating protein
kinase A and the mitogen-activated protein kinases p38, ERK1/ 2 and JNK1/2.1,2
The A AR subtype is expressed in the central nervous system (CNS) showing the greatest density
2
A
in the striatum, olfactory tubercle and nucleus accumbens while lower levels are present in the
cortex and hippocampus. In periphery, the A2A AR is abundant in heart, lung, blood vessels and in
the immune system.
The A2A AR plays a key role in the regulation of inflammatory processes both in the CNS and in
periphery.1-4 At peripheral level, it activates an anti-inflammatory cascade through a reduced
functionality of the immune system cells and inflammatory cells. In particular, the A2A AR
decreases the functions of neutrophils, T cells activation, migration of mast cells and macrophages
and the release of cytokines. The A2A AR-mediated immune suppressive effect accounts for the
profitable role of the A2A stimulation in inflammatory processes. However, in some types of solid
cancer, in which hypoxia enhances adenosine concentration, this effect can exacerbate. Hence,
suppression of the immune responses in the tumor microenvironment, in particular those T cell-
mediated, produces deleterious effects since protects cancer cells from death, thus promoting tumor
growth and metastasis. As a consequence, A2A AR antagonists, being effective in removing the
adenosine-mediated immune escape, are considered as novel therapeutic agents in the
immunotherapy of cancer.5
In the CNS, A2A AR activation can exert opposite effects to the peripheral ones. The A2A AR is
present on both pre- and post-synaptic neurons and also in glial cells where it stimulates pro-
inflammatory functions, particularly by inducing activation of both microglia and astrocytes in pro-
1
,4
inflammatory phenotype. Under physiological conditions, the A AR expression in microglia and
2
A
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 69
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
astrocyte is usually low while it increases after brain insults, nerve injury and inflammatory
6,7
signals. Induction of glial A AR expression takes part in an important feed-forward mechanism
2
A
8
,9
to locally control neuroinflammatory responses in the brain. Activation of A ARs in microglia
2
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
has mixed effects on proliferation of these cells, and clearly shows a facilitating action on the
release of pro-inflammatory cytokines, such as IL-1, TNF, IL-2 and IL-6, and of ROS, all
associated with neuronal damage occurring in Parkinson’s (PD) and Alzheimer’s (AD) diseases.9
Consequently, the blockade of the A2A AR by antagonists induces neuroprotection in these CNS
disorders in which neuroinflammatory and oxidative processes play a significant role.1-3
Neuroprotection attributed to A2A AR antagonists have been associated also with their ability to
reduce glutamate levels by decreasing its release10-12 and enhancing its glial uptake.6,13,14
A2A AR is also involved in the development of neuropathic pain and its blockade confers
1
5
protection. Neuropathic pain is a common type of chronic pain, which occurs in several disorders
and results in several factors leading to impairment in nerve function. Its pathophysiology is quite
complex and involves both central and peripheral mechanisms with alterations in the ion channel
expression, neurotransmitter release, and pain pathways.16 Although the molecular basis of
neuropathic pain is not completely understood, oxidative stress might contribute to its
development.17-19 In pain following spinal cord injury, beside dysfunction of neurons, other
pathogenic events occur, including microglia activation and enhanced extracellular glutamate
2
0
which, in turn, activates intracellular pathways such as ROS formation. Platinum-based anticancer
drugs can cause peripheral neuropathy involving sensory nerves and it has been demonstrated that
the treatment with this kind of drugs induces, among others, ROS generation, damage at nuclear and
2
0
mitochondrial DNA, loss in antioxidant enzymes, and nerve tissue impairment. In accordance,
systemic administration of antioxidant17,18 or ROS scavenger17 produces pain relief in different
animal model of neuropathic pain.
4
ACS Paragon Plus Environment

Page 5 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The role of the A AR in pain is still controversial because several studies support both its pro- and
2
A
21
anti-nociceptive role, depending on the receptor localization and the type of pain. Coherent with a
pro-nociceptive role it was observed that after peripheral nerve injury, A AR stimulation induces
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
both activation and proliferation of microglia and astrocytes responsible of inflammation occurring
in neuropathic pain, while genetic deletion of the A2A AR decreases all the behavioral and
histological signs of pain.15 Several studies also showed that systemic22,23 and spinal24
administration of the selective A2A AR antagonist 2-(furan-2-yl)-7-phenethyl-7H-pyrazolo[4,3-e]-
1
,2,3-triazolo[1,5-c]pyrimidin-5-amine (SCH58261) produced antinociception in several preclinical
models. Moreover, potent hA inverse agonists belonging to our thiazolo[5,4-d]pyrimidine series
2
A
showed an anti-nociceptive effect equal to or greater than morphine in acute pain models.25
Taking into account these premises, over the last few years we have directed a part of our research
to obtaining new A2A AR antagonists belonging to bicyclic25-33 and monocyclic34 heterocyclic
classes. Among the former, the 8-amino-2-phenyl-1,2,4-triazolo[4,3-a]pyrazine-3-one series27,33
was investigated and several potent hA AR antagonists were identified, some of which proved to
2
A
be neuroprotective in PD27 and AD in vitro models. In this paper we describe new 1,2,4-
triazolo[4,3-a]pyrazines designed as hA2A AR antagonists bearing an unsubstituted phenyl ring at
position 2 and different moieties at position 6 (1-21). The former group was chosen since it proved
33
2
7
to be an important feature to obtain an efficient hA2A receptor-ligand interaction, while the 6-
substituents were mostly selected to obtain dual acting antioxidant-A AR antagonists. Compounds
2
A
endowed with this mixed activity have attracted our attention since they would possess a potentially
increased protective effect both in neurodegenerative diseases and in neuropathic pain. The new
triazolopyrazines can be subdivided into two sets, depending on the type of the 6-substituent.
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 69
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
NH2
N
NH2
N
N
R2
N
N
R2
R5
N
N
N
R3
R3
O
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
O
HO
R5
R O
4
7-11
1-6
R = H, OMe
2
OMe
HO
HO
R = R = H, OMe, Me, t-Bu
3
5
HO
HO
R = Me
4
HO
OH
R -R = O-CH
2
3
4
Me
HO
HO
Me
t-Bu
Figure 1. New 8-amino-2-phenyl-1,2,4-triazolo[4,3-a]pyrazin-3-ones 1-11.
In the first set (1-11, Figure 1), derivatives 7-11 bear phenolic and polyphenolic rings at the 6-
position. These kinds of substituents were chosen since they are a common feature of both natural
and synthetic antioxidant agents. Among them, naturally occurring hydroxycinnamic acids, such as
caffeic and ferulic acids, and resveratrol (Figure 2) were proven to exert diverse bioactivities
affording neuroprotective effects.35-37 3,5-Di-tert-butyl-4-hydroxytoluene (BHT) is a synthetic
3
7
antioxidant used for food and pharmaceuticals. Like other hindered phenols, BHT can exert
biological functions for its ability to intercept and react with free radicals through atom transfer.38
OH
Me
HO
RO
COOH
HO
HO
HO
R= H Caffeic acid
R= Me Ferulic acid
OH Resveratrol
BHT
Figure 2. Some natural and synthetic antioxidant agents.
Considering our triazolopyrazines 7-11, electron-donating groups were also introduced on the 6-(4-
hydroxyphenyl) ring, in particular at the ortho position of the hydroxy group (Me, tert-But). These
6
ACS Paragon Plus Environment

Page 7 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
substituents might have a role in improving radical scavenging activity, which may be mainly
related to their ability to delocalize/stabilize the resulting phenoxyl radical.37
The second set of triazolopyrazines (12-21, Figure 3) was synthesized to obtain derivatives 15-17,
20, 21, in which antioxidant moieties were spaced by different linkers from the para position of the
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
-phenyl ring.
NH2
N
N
N
NH2
N Ph
N
N
Antioxidant
moiety
N Ph
O
N
linker
15-17, 20, 21
O
R
12-14, 18, 19
R= see Table 2
O
O
N
2 O
H
S
O
S
H
N
HO
N
H
N
2
H
Figure 3. New 8-amino-2-phenyl-1,2,4-triazolo[4,3-a]pyrazin-3-ones 12-21.
As antioxidant pendants, we selected -lipoic and 3,5-di-tert-butyl-4-hydroxybenzoic acid residues.
The latter was chosen for its structural similarity to BHT, the former because, besides being a
naturally occurring compound present in food and used as dietary integrator, it emerged in
preclinical studies as a promising agent for the treatment and/or prevention of neurodegenerative
disorders.39-41 At a molecular level, -lipoic acid is effective in scavenging free radicals and
reducing oxidative stress. It also increases or maintains cellular glutathione levels by acting as a
transcriptional inducer of genes governing glutathione synthesis. Clinical studies investigating the
effect of -lipoic acid on diabetic neuropathy have revealed its efficacy in relieving neuropathic
pain symptoms.42,43
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 69
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
All the newly synthesized triazolopyrazines 1-21 were evaluated for their affinity at ARs. These
derivatives include not only the target compounds, bearing antioxidant moieties, but also their
synthetic precursors and some derivatives prepared to broaden SAR studies.
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
RESULTS AND DISCUSSION
Chemistry. The 1,2,4-triazolopyrazin-3-one derivatives 1-21 were prepared as depicted in Schemes
1
1
-3. Compounds 1-11 (Scheme 1) were obtained starting from ethyl 5-oxo-1-phenyl-4,5-dihydro-
4
4
4
H-1,2,4-triazole-3-carboxylate 22 which was regioselectively N -alkyated with the suitable -
bromoketones 23-28. Of the latter, 23-26 were previously reported,45-48 while 27 and 28 were newly
synthesized in the same conditions employed to obtain 23-26, i.e. by brominating the corresponding
acetophenone derivatives 29 and 30. Compound 29 was commercially available while 30 was
4
9
synthesized by methylation of (4-hydroxy-3,5-di-tert-butyl)phenylethanone. The N -alkyltriazole
4
derivatives 31-36 were cyclized with ammonium acetate, by heating in a sealed tube, to give the
1
,2,4-triazolo[4,3-a]pyrazine-3,8-dione derivatives 37-42 which were chlorinated with phosphorus
oxychloride, under microwave irradiation, to give the related 8-chloro derivatives 43-48. Their
treatment with a saturated ethanolic solution of ammonia gave the desired 8-amino-1,2,4-
triazolo[4,3-a]pyrazine-3-one derivatives 1-6. Reaction of the 6-(2,4-dimethoxy)phenyl derivative 1
with BBr 1M dichloromethane solution produced demethylation of the methoxy group at position
3
4, yielding to the 4-hydroxy-2-methoxyphenyl derivative 7. Its structure was established by
NOESY experiment, indicating the spatial proximity between the methoxy group and the sole
aromatic proton at position 3.
8
ACS Paragon Plus Environment

Page 9 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 1a
O
O
R
t-But
CH3
c
CH3
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
H CO
HO
3
R
t-But
O
2
3
9 R= CH3
0 R= t-But
b
O
O
N
EtO
N
EtO
N Ph
Br
a
N Ph
N
+
R6
HN
O
O
O
23-28
2
2
R6
31-36
d
NH2
Cl
O
N
N
N
N
e
N
HN
R6
f
N Ph
N
N Ph
Ph
N
N
N
R6
R6
O
O
O
1-6
43-48
37-42
g
7
-11
R6
1, 23, 31, 37, 43 C H -2,4-diOCH
R6
C H -2-OCH -4-OH
7
6
4
3
3
6
4
3
2
, 24, 32, 38, 44 C H -3,4-diOCH
8
C H -3,4-diOH
6 4
6
4
3
, 25, 33, 39, 45 C H -3,4-OCH O
9
C H -3,4,5-triOH
6 4
6
4
2
4, 26, 34, 40, 46 C H -3,4,5-triOCH
10
11
C H -4-OH-3,5-diCH
6 4 3
6
4
3
5
6
, 27, 35, 41, 47 C H -4-OCH -3,5-diCH
C H -4-OH- tBu
6 4
6
4
3
3
, 28, 36, 42, 48 C H -4-OCH -3,5-di-tBu
6
4
3
a
Reagents and conditions: (a) K CO , DMF/CH CN, rt; (b) Br , CHCl /Et O, 0 °C-
2
3
3
2
3
2
rt; (c) CH I, K CO 2-butanone, reflux; (d) NH OAc, 140 °C sealed tube; (e) POCl , mw 170 °C;
3
2
3,
4
3
(
f) NH , absolute EtOH; (g) BBr , anhydrous CH Cl , 0 °C- rt.
3 3 2 2
Demethylation of the (methoxyphenyl) derivatives 2 and 4, 5 with BBr (1M dichloromethane
3
solution) gave the corresponding hydroxyphenyl-substituted compounds 8-10. These conditions did
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 69
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
not work to demethylate the 6-(3,5-di-tert-butyl-4-methoxy)phenyl derivative 6, probably due to the
steric bulk of the two tert-butyl groups. Reaction was successful in more drastic conditions, i.e. with
4
8% aqueous HBr at reflux which, however, caused the removal of a tert-butyl substituent. Thus,
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 3-tert-butyl-4-hydroxy derivative 11 was obtained, instead of the desired 4-hydroxy-3,5-di-tert-
butyl derivative. Scheme 2 depicts the synthesis of the triazolopyrazines 12-16, of which 15 and 16
were the target compounds linking antioxidant moieties.
Scheme 2a
NH2
NH2
N
N
N
N
N Ph
a
N Ph
N
N
O
O
R
O
HO
1
2 R= CN
49
13 R= CONH2
b
NH2
N
N
N Ph
N
O
O
O
R-HN
S
O
15 R=
S
c
14 R= H
16 R=
OH
a
Reagents and conditions: a) NC-CH -Cl or NH -CO-CH -Cl, K CO , anhydrous acetone, reflux; b)
2
2
2
2
3
LiAlH , anhydrous THF, 0 °C; c) (R, S) lipoic acid or 3,5-di-tert-butyl-4-hydroxybenzoic acid, 1-
4
(
3-(dimethylamino)-propyl))-3-ethylcarbodiimide hydrochloride, NEt , 1-hydroxybenzotriazole,
3
anhydrous DMF, rt.
2
7
The starting material was the previously reported 6-(4-hydroxy)phenyl-triazolopyrazine 49 which
was O-alkylated with the suitable alkyl halides to give the corresponding 6-(4-O-alkylated)
1
0
ACS Paragon Plus Environment

Page 11 of 69
Journal of Medicinal Chemistry
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
compounds 12 and 13. The cyano derivative 12 was reduced at rt with LiAlH to afford the 6-(4-(2-
4
aminoethoxy)phenyl compound 14 which was reacted with (R,S) lipoic acid and 3,5-ditertbutyl-4-
hydroxybenzoic acid, in anhydrous DMF and in presence of 1-hydroxybenzotriazole, 1-
(dimethylamino)-propyl))-3-ethylcarbodiimide hydrochloride and triethylamine, to yield the desired
derivatives 15 and 16.
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 synthesis of the triazolopyrazines 17-21 is shown in Scheme 3. The 6-(4-lipoylaminophenyl)
derivative 17 was obtained by reacting the previously reported 6-(4-aminophenyl) derivative 5033
with (R,S) lipoic acid, in the same conditions described above to prepare 15 from 14.
Scheme 3a
NH2
NH2
N
N
N Ph
N
N
N
N Ph
a
O
N
O
S
N
O
S
H
17
H N
2
50
b
NH2
N
NH2
N
N
N
N
N
N Ph
N Ph
c
O
O
O
O
R-HN
N
H
N
H
18
19
R= H
20 R=
O
O
a
S
S
21 R=
OH
aReagents and conditions: a) (R, S) lipoic acid or 3,5-di-tert-butyl-4-hydroxybenzoic acid, 1-(3-
(dimethylamino)-propyl))-3-ethylcarbodiimide hydrochloride, NEt3, 1-hydroxybenzotriazole,
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 69
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
anhydrous DMF, rt; b) Cl-(CH ) -COOH, 1-(3-(dimethylamino)-propyl))-3-ethylcarbodiimide
2
2
hydrochloride, NEt , anhydrous DMF, rt; c) NH gas/ absolute EtOH, sealed tube, 130 °C.
3
3
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
When the same experimental conditions were employed to react compound 50 with 3-
chloropropionic acid, the 6-(4-acrylamidophenyl) derivative 18 was obtained which was allowed to
react with a saturated solution of ammonia in absolute ethanol to afford the 6-(4-(3-
aminopropanamido)phenyl-derivative 19. This intermediate was transformed into derivatives 20
and 21 by acylation with (R,S) lipoic acid and 3,5-ditertbutyl-4-hydroxybenzoic acid, respectively,
in the same conditions described above to obtain 15 from 14.
Binding and cAMP assays
All the newly synthesized triazolopyrazines 1-21 were evaluated for their affinity at hA , hA and
1
2A
hA ARs, stably transfected in Chinese hamster ovary (CHO) cells, and were tested at the hA AR
3
2B
subtype by determining their inhibitory effects on NECA-stimulated cAMP levels in hA2B CHO
cells (Tables 1 and 2). Derivatives 11, 12, 15, 20 and 21, showing high hA2A AR affinity and
selectivity, were selected to assess their antagonistic profile. Hence, their ability to inhibit or
stimulate the hA2A AR was determined by evaluating their effect on cAMP production in CHO
cells, stably expressing hA ARs (Table 3).
2
A
Structure-Affinity Relationship Studies
The results reported in Tables 1 and 2 displayed that several of the targeted triazolopyrazines
featuring potential antioxidant moieties (7-8, 10, 11 and 15, 17, 20, 21) showed nanomolar hA2A
AR affinity and different degrees of selectivity. Within the first set of compounds (Table 1), the 6-
(
4-hydroxy-3-terbutyl)-phenyl derivative 11 was the most selective for the hA2A AR (K = 8.5 nM).
i
1
2
ACS Paragon Plus Environment

Page 13 of 69
Journal of Medicinal Chemistry
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 other phenolic derivatives (7, 8 and 10) showed nanomolar affinity at the hA AR, compound
2A
1
0 being the most active (K = 2.5 nM), but scarce selectivity since they were able to bind the hA
i
1
subtype with significant affinity (K = 21.3-42.6 nM).
i
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
Table 1. Biological activity of compounds 1-11 at hARs.a
NH2
N
N
N
N
R6
O
Binding experiments
cAMP assays
K (nM)
i
IC (nM)
50
R6
hA1b
hA2Ac
hA3d
hA2Be
>30000
>30000
1
2
2
8 ± 0.3
2.4 ± 0.5
5.7 ± 0.8
118 ± 6.6
80.1 ± 15.8
H CO
OCH3
3
H CO
3
59 ± 12.7
H CO
3
O
O
3
4
13 ± 2.5
55 ± 16
7.4 ± 0.9
3.5 ± 0.8
38 ± 6.7
>30000
>30000
H CO
3
214 ± 4.4
H CO
3
OCH3
H C
5
6
3
4.5 ± 1.4
0.17 ± 0.004
141.6 ± 34
8.6 ± 1.7
>30000
>30000
>30000
H CO
3
CH3
108.5 ± 17
H CO
3
7
2
9.7 ± 1.6
16.8 ± 0.9
11130 ± 975
> 30000
HO
OCH3
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 69
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
HO
8
9
42.6 ± 9.6
175.5 ± 3
5.2 ± 0.5
94.5 ± 21
950 ± 200
>30000
HO
HO
5575 ± 989
17330 ± 3365
>30000
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
HO
OH
H C
3
1
0
1
21.3 ± 7
>30000
2.5 ± 0.8
8.5 ± 1.4
100 ± 0.7
>30000
HO
HO
CH3
1
>30000
a
b
3
Data (n= 3-5) are expressed as means ± standard errors. Displacement of specific [ H]-CCPA binding
c
3
at hA AR expressed in CHO cells. Displacement of specific [ H]-NECA binding at hA AR expressed
1
2A
d
3
in CHO cells. Displacement of specific [ H]-HEMADO binding at hA AR expressed in CHO cells.
3
e
IC values of the inhibition of NECA-stimulated adenylyl cyclase activity in CHO cells expressing
5
0
hA AR.
2B
Derivatives 1-6, including the methoxy synthetic intermediates and the 6-(3,4-
methylenedioxyphenyl) derivative 3, on the whole, showed high affinities for the hA2A AR,
spanning one-digit nanomolar values, and also for the hA subtype. The most active compound at
1
the hA2A AR was derivative 5 (K = 0.17 nM), featuring the 6-(3,5-dimethyl-4-methoxyphenyl)
i
substitution, while its 6-(3,5-di-tert-butyl-4-methoxyphenyl) analogue 6 showed significantly lower
hA2A AR binding activity (K = 141.6 nM), probably due to the steric bulk of the two tert-butyl
i
groups. Compounds 5 and 3 also possess significant affinity for the hA subtype.
3
In the second set of triazolopyrazines (12-21), -lipoic acid and 4-hydroxy-3,5-ditertbutylbenzoic
acid were selected as antioxidant portions and linked by different chains at the para position of the
6-phenyl ring. The choice of this position was based on the results of previous molecular docking
1
4
ACS Paragon Plus Environment

Page 15 of 69
Journal of Medicinal Chemistry
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
studies of this class of compounds at the hA2A AR, highlighting that the presence of hindering
substituents on the 6-phenyl ring favored a binding pose with this moiety pointing towards the
extracellular side of the receptor. Hence, we envisaged that long substituents at the para position
could be well tolerated because they could point away from the binding pocket. The selected chains
were linked through an ethereal (compounds 15, 16) or an amide function (compounds 17, 20 and
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
2
1).
Table 2. Biological activity of derivatives 12-21 and the reference compounds 49 and 50, at hARs.a
NH2
N
N
N
N
O
R
Binding experiments
cAMP assays
K (nM)
i
IC (nM)
50
R
hA1b
hA2Ac
hA3d
hA2Be
N
1
1
1
2
3
4
O
> 30000
8.2 ± 2.3
> 30000
> 30000
O
391.7 ± 104
288.7 ± 54
26 ± 1.7
604 ± 94
> 30000
> 30000
O
H N
2
O
14.9 ± 0.1
2131 ± 173.5
H N
2
O
O
1
1
5
6
378.6 ± 91
2.4 ± 0.3
4097 ± 812
>30000
>30000
>30000
N
S
H
S
O
O
N
H
13670 ± 275
14750 ± 270
HO
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 69
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
O
1
1
1
7
8
9
8.4 ± 0.4
262.7 ± 1.9
479.2 ± 89
5 ± 0.6
>30000
>30000
509 ± 90
>30000
>30000
NH
S
S
O
1.8 ± 0.09
0.59 ± 0.2
H C
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
2
NH
O
9658 ± 1431
>30000
H N
2
NH
O
O
2
2
0
1
1359 ± 284
>30000
36.4 ± 8.2
54.5 ± 7.1
>30000
>30000
N
NH
S
H
S
O
O
N
H
NH
>30000
HO
4
9f
OH
45 ± 10
45 ± 12
53± 13
>30000
>30000
5
0g
NH2
33.5 ± 6.7
22.9 ± 0.2
253.7 ± 67.6
a
b
3
Data (n= 3-5) are expressed as means ± standard errors. Displacement of specific [ H]-CCPA binding
c
3
at hA AR expressed in CHO cells. Displacement of specific [ H]-NECA binding at hA AR expressed
1
2A
d
3
in CHO cells. Displacement of specific [ H]-HEMADO binding at hA AR expressed in CHO cells.
3
e
IC values of the inhibition of NECA-stimulated adenylyl cyclase activity in CHO cells expressing
5
0
f
g
hA AR. Ref. 27. Ref. 33.
2
B
The binding data (Table 2) proved us right. On the whole, all the substituents appended on the 4-
hydroxy- and 4-amino group of derivatives 49 and 50, respectively, increased affinity and /or
selectivity for the hA AR (compare derivatives 12-16 to 49 and compounds 17-21 to 50), with the
2
A
only exception being compound 16, which showed a dropped hA AR binding activity.
2
A
The lipoyl derivatives 15, 17 and 20 resulted in high affinity hA2A AR ligands (K = 2.4-36.4 nM)
i
with different degrees of selectivity versus the hA AR, depending on the nature of the linker.
1
1
6
ACS Paragon Plus Environment

Page 17 of 69
Journal of Medicinal Chemistry
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
Compound 17, bearing the lipoyl residue directly attached on the para-amino group, showed a high
affinity not only for the hA2A receptor but also for the hA AR subtype. Very interesting results
1
were obtained when the lipoyl moiety was spaced from the para-position through the flexible
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
oxyethylamino chain (-O-(CH ) -NH-). In fact, the resulting compound 15 possessed a very high
2
2
affinity for the targeted hA2A receptor (K = 2.4 nM) and also high selectivity, being significantly
i
less active at the hA AR (K = 378.6 nM) When the oxyethylamino spacer was replaced by the
1
i
.
longer and more rigid carboxyamidoethylamino spacer (-NH-CO-(CH ) -NH-), a selective ligand
2 2
for the hA2A AR was still obtained (compound 20) but its affinity and selectivity were lower, with
respect to those of 15. The same two spacers were employed to link the (4-hydroxy-3,5-di-tert-
butyl)benzoyl pendant to the 6-phenyl ring (derivative 16 and 21) but in this case an opposite effect
was obtained since the best results, both in terms of A AR affinity and selectivity, were obtained
2
A
with the carboxyamidoethylamino spacer. In fact, compound 21 showed good affinity for the hA2A
AR (K = 54.5 nM) and high selectivity, while it analogue 16 was almost inactive at all ARs. The
i
significant difference between hA2A affinities of the two compounds obviously depends on the
different spacer. The bit longer carboxyamidoethylamino spacer seems to direct the terminal bulky
aryl ring in a more favorable pose in the receptor binding site (see modeling analysis). Also
derivatives 12-14 and 18, 19, which were not our primary target compounds, resulted in interesting
ligands, showing nanomolar affinity and good to high selectivity for the hA2A AR. In particular,
derivative 19, bearing the carboxyamido-ethylamine substituent at the para position of the 6-phenyl
ring, was the most active (K = 0.59 nM). Derivatives 12 and 18, featuring at the para-position a
i
cyanomethoxy (K = 8.2 nM) and an acrilamido group (K = 1.8 nM), respectively, displayed high
i
i
hA2A AR affinities and selectivity. Reduction of the cyano residue of compound 12 afforded the
amino derivative 14 which maintained the ability to target the hA2A AR with nanomolar affinity
(K = 14.9 nM) but lower selectivity. Transformation of the cyano in amide group (compound 13)
i
also retained affinity but reduced selectivity for the target hA receptor.
2
A
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 69
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
Finally, compounds 11, 12, 15, 20 and 21, showing high hA2A AR affinity and selectivity, were
selected to be further pharmacologically profiled in in vitro studies. Previously, we ascertained their
antagonistic profile by evaluating their effect on cAMP production in CHO cells, stably expressing
hA2A ARs (Table 3). The compounds proved to be able to counteract NECA-stimulated cAMP
production, thus behaving as hA AR antagonists.
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
2
A
Table 3. Potencies of the selected triazolopyrazines 11, 12, 15, 20 and 21 at hA AR.
2A
a
hA AR (IC nM)
2
A
50
1
1
1
2
1
2
5
0
179 ± 53
157 ± 43
116 ± 31
296 ± 66
263 ± 58
21
a
IC values of the inhibition of NECA-stimulated adenylyl cyclase activity in CHO cells expressing
5
0
hA AR. Data are expressed as means ± standard errors.
2
A
Molecular modeling studies. The binding mode of the synthesized compounds at the hA2A AR
cavity was simulated with docking analysis by using the MOE (Molecular Operating Environment
2014.09) software and the CCDC Gold docking tool.50,51 The MOE software analysis was carried
out by selecting the induced fit docking and optimization protocol (schematically, a preliminary
docking analysis providing a set of ligand conformations then energy minimized with the side
chains of the receptor residues in proximity). For the docking tasks, a crystal structure of the hA2A
AR in complex with the antagonist/inverse agonist ZM241385 was employed (http://www.rcsb.org;
52
pdb code: 5NM4; 1.7-Å resolution ). For a subset of compounds, the binding modes at the hA AR
1
5
3
crystal structure (pdb code: 5N2S; 3.3-Å resolution ) were also simulated with the same tools and
protocols.
1
8
ACS Paragon Plus Environment

Page 19 of 69
Journal of Medicinal Chemistry
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 docking conformations generally obtained for the new derivatives at the hA AR are similar to
2A
27
those observed for our previously reported triazolopyrazines and is shown in Figure 4A. In this
binding mode, the bicyclic core is positioned between the side chains of Phe168 (EL2) and
Leu2496.51 and engages non-polar interactions with these residues. The exocyclic amine group
makes H-bond contacts with Asn2536.55 and Glu169 (EL2), while the 2-phenyl substituent is
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
located in the depth of the cavity. The R group is positioned at the entrance of the binding site and
6
oriented toward the extracellular environment. Such compound orientation and interaction resemble
the binding mode of the co-crystallized 4-(2-[7-amino-2-(2-furyl[1,2,4]-triazolo[2,3-
5
2
a][1,3,5]triazin-5ylamino]ethyl)phenol (ZM241385) in the employed crystal structure but also in
5
4,55
other previously reported hA AR X-ray structures.
2A
The presence of substituents on the 6-phenyl ring modulates the interaction with the receptor
residues at the entrance of the cavity and leads to various degrees of affinity for the hA AR (see
2
A
Tables 1 and 2). For previously reported analogues,27 it was observed that a non-polar para-
substituent on this ring was more efficient to improve the hA2A AR affinity than a polar one.
Considering the meta-substituents of the 6-phenyl ring, the affinity data showed that the hA AR
2
A
affinity was not significantly influenced by the chemical-physical profile of the substituent.
Compounds of this new set of triazolopyrazines, differing in polarity of para- or meta-substituent,
present various hA AR affinity. Considering derivatives bearing a small 4-substituent on the 6-
2
A
phenyl moiety, again a non-polar para-substituent on this ring appears more efficient to improve the
hA2A AR affinity than a polar one. As an example, compounds 5, the most active of the herein
reported derivatives, featuring a 4-methoxy and 3,5-di-methyl groups on the 6-phenyl ring, is
endowed with 15-fold higher affinity at the hA AR than 10, which bears a 4-hydroxy and 3,5-di-
2
A
methyl groups.
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 69
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. (A) General binding mode of the synthesized compounds at the hA AR (pdb: 5NM4)
2
A
binding cavity, with indication of some key receptor residues; compound 5 is showed. (B) Top-view
of the hA AR residues at the entrance of the binding cavity and potentially giving interaction with
2A
substituents on the 6-phenyl ring. (C) Molecular surface representation of the entrance of the hA2A
AR binding cavity; dark-to-light color indicates hydrophilic-to-hydrophobic scale.
The substituents inserted on the 6-phenyl ring are located in proximity of Ile662.64, Ser672.65
,
Leu167 (EL2), Leu267 (EL3), Met2707.35 and Tyr2717.36 (Figure 4B-C). Considering the volume,
Figure 4C shows a molecular surface representation of the entrance of the hA AR binding cavity.
2A
From this figure it can be seen that small substituents are allowed in the ortho- and meta-position of
the R group, but the space is too limited to allow the insertion of two or more bulky moieties at the
6
same positions. In fact, compound 6, bearing two tert-butyl groups at the meta-position of the 6-
phenyl ring, has a significantly reduced hA AR affinity compared to the other analogues (Table
2
A
1
). The non-polar profile of the above cited amino acids surrounding the R group allows a
6
2
0
ACS Paragon Plus Environment

Page 21 of 69
Journal of Medicinal Chemistry
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
favorable interaction with non-polar substituents rather than polar ones. Combinations of para- and
meta-substituents or para- and ortho-substituents lead to slight modulation of hA AR affinity.
2
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
Figure 5. (A) General binding mode of the synthesized compounds at the hA AR (pdb: 5N2S)
1
binding cavity, with indication of some key receptor residues; compound 19 is showed. (B) Top-
view of the hA AR residues at the entrance of the binding cavity and potentially giving interaction
1
with substituents on the R aryl ring.
6
Docking results obtained at the hA AR crystal structure are highly similar to the ones obtained at
1
the hA AR (Figure 5). Considering compounds bearing small substituents at the meta- and para-
2A
position of the 6-phenyl group, docking conformations suggest analogue considerations as above
for the impact on the hA AR affinity given by these substituents. This is in agreement with
1
biological evaluation results of compounds 1-5,7-10, which show similar trends of binding affinity
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 69
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
values at hA AR and hA AR. As for the hA AR, compound 5 is the most active of the whole set
1
2A
2A
of derivatives at the hA AR, with 4-fold higher affinity than 10, its para-hydroxy substituted
1
analogue. Even in this case, compounds bearing tert-butyl groups are endowed with lower affinity.
Still considering compounds bearing small substituents at the meta- and para-position of the R6
group, affinities at the hA AR are generally higher than those observed at the hA AR. The set 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
2
A
1
hA AR residues in proximity with the para-substituent is globally more hydrophobic than the hA
1
2
A
AR one, due to the presence of Leu167 (EL2), Leu267 (EL3), Met2707.35 in the hA AR instead of
2A
the Glu170 (EL2), Ser267 (EL3) and Thr2717.35 residues in the respective positions of the hA AR.
1
This factor could play a key role in providing a slight hA AR selectivity (versus the hA AR) for
2
A
1
the compounds described above.
Considering the compounds presenting only small para-substituents on the 6-phenyl ring (12-14,
8, 19), the hA2A AR affinities appear significantly higher than the hA AR ones. The different
1
1
affinity for the two AR subtypes could be due to both the chemical-physical profile of the AR
residues in proximity of the substituent and how much the substituent itself is exposed to the
external environment. Figure 6A-B shows a surface representation of the entrance of the binding
cavities, in proximity to the para-position of the 6-phenyl ring. The light-to-dark regions indicate a
hydrophobic-to-hydrophilic profile of the residues. The more polar profile of the hA AR residues
1
with respect to the hA2A AR ones is due in particular to the presence of a negatively charged
glutamate residue (Glu170, Figure 5B) in hA AR, while the hA AR bears a non-polar leucine
1
2A
(
Leu167, Figure 4B) in the same position. The presence of Glu170 in hA AR leads to a repulsive
1
effect between this residue and the carbonyl group of the compounds bearing a carboxyamidoethyl
spacer (19, 20, 21). This effect is evident from the comparison of the activities of the latter
compounds with the higher affinity data at the hA AR of the corresponding analogues bearing an
1
oxyethyl spacer (14, 15, 16, respectively). At the hA2A AR, both the carboxyamidoethyl and
oxyethyl spacers generally lead to nanomolar affinities.
2
2
ACS Paragon Plus Environment

Page 23 of 69
Journal of Medicinal Chemistry
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 6. A-B Top-view of the binding mode of compound 19 at the hA AR (A, pdb: 5NM4) and
2
A
hA AR (B pdb: 5N2S) binding sites. Molecular surface representations of both binding cavities are
1
represented. Light-to-dark colors of surface correspond to hydrophobic-to-hydrophilic regions. C-
D. Binding modes suggested for compounds bearing large para-substituents in the 6-phenyl ring;
compound 15 at the hA2A AR (pdb: 5NM4) is shown. These molecules may adopt the general
binding mode above described (C) or an alternative, energetically more stable, docking
conformation that makes the para-substituent externally oriented without clashes with receptor
residues (D).
Docking studies performed for the compounds presenting small para-substituents (12-14, 18, 19)
show that these molecules may adopt a binding mode similar to the one described above. The
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 69
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
interaction of the para-substituent with hA2A AR residues is modulated by the chemical-physical
profile of the substituent itself and the receptor residues in proximity. The positively charged amine
function of compound 19 gets located in proximity of both the hydroxyl group of Tyr2717.36 and the
carbonyl group of Ser672.65 (Figures 4B and 6A), thus providing subnanomolar affinity for the hA2A
AR. Compounds bearing a large para-substituent (15-17, 20-21) may adopt as well as the above-
described binding mode (Figure 6C, compound 15), even if the large para-substituent gets located
too close to the receptor residues, giving a steric clash with the protein atoms. Docking results for
these compounds suggest also an alternative binding mode, with the bicyclic scaffold upside-down
oriented to point the 3-carbonyl group toward the amine function of Asn2536.55 (Figure 6D,
compound 15). This binding mode lacks some hydrophilic interactions with the receptor that are
observed in the general compound orientation described above (i.e. with Glu169 at hA AR); on
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
2
A
the other hand, the alternative binding mode favors the pointing of the para-substituent toward the
external environment with a more energetically stable compound conformation. This may explain
the favorable docking scores obtained by the alternative binding mode arrangement for compounds
bearing large para-substituents in the 6-phenyl ring. Considering the largest compounds (15, 16, 20,
2
1), the lowest activity belongs to derivative 16, which is the one presenting a para-substituent of
large volume and the shortest oxyethylamino spacer linking the 6-phenyl ring. In contrast, its
corresponding analogue with the longer carboxyamidoethyl spacer (21) presents a 300-fold
improvement of the hA AR affinity. On the other hand, both derivatives presenting a less bulky
2
A
lipoyl group (15 and 20) are endowed with significantly higher affinity. This suggests that the
length and volume of the large para-substituents appear critical for the receptor affinity, for an
energetically stable accommodation of the substituent itself within the receptor residues.
Neuroprotection Studies on oxaliplatin-induced neurotoxicity in microglia cells. Based on the
affinity data, compounds 11, 12, 15, 20 and 21, potent and selective hA2A AR antagonists, were
2
4
ACS Paragon Plus Environment

Page 25 of 69
Journal of Medicinal Chemistry
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
selected for further pharmacological evaluation. In particular, their protective effect against the
neurotoxicity of the anticancer drug oxaliplatin on rat microglia cells was determined. Neuropathy
induced by oxaliplatin is a common side effect in patients treated with this drug and consists in
paresthesia, dysesthesia, and pain. Such a condition adversely affects quality of life and can lead to
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
56
discontinuation of therapy. It is well-known that glia cells play a key role in the CNS homeostasis
and are strongly involved in the responses to nerve injury. Microglia and astrocytes activate several
mechanisms, such as production of trophic factors, regulation of transmitter and ion concentrations,
which tend to decrease neuronal injury. Nevertheless, in pathological conditions, the maladaptive
plasticity of glial cells strongly sustains negative symptoms like chronic pain. In particular,
microglia functional modifications have the potential to induce neuronal dysfunction, playing a
pivotal role in oxaliplatin neuropathy development.57,58
The selected compounds were chosen taking into account their high affinity and selectivity toward
the hA AR but also the presence in all of them, except compound 12, of antioxidant moieties
2
A
which were thought to counteract oxaliplatin neurotoxicity. In fact, although the molecular basis
underlying the oxaliplatin neuropathy is unclear, some experimental evidence pointed out a
2
0
correlation between oxidative stress damage and neuropathic pain onset, also highlighting efficacy
of the antioxidant silibinin and α-tocopherol in reducing oxaliplatin-dependent pain induced by
5
9
mechanical and thermal stimuli. Compound 12, lacking the antioxidant portion, was tested to
evaluate how the lone blockade of the A AR could affect oxaliplatin toxicity.
2
A
Primary rat microglia cells were treated with oxaliplatin in the absence or in the presence of the
tested compounds. Oxaliplatin damage was evaluated as cell viability and oxidative stress, the latter
60
previously described as the main damage evoked by oxaliplatin. The new synthesized compounds
were tested at 10 μM, the maximum soluble concentration. Oxaliplatin, concentration-dependently,
strongly reduced microglia viability (MTT test) after 24 h incubation (33% and 19% viability with
1
0 and 30 μM, respectively, in comparison to 100% of control condition).
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 69
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 obtained results showed that the lipoic acid-conjugated derivative 15 was the most active in
preventing the oxaliplatin damage, also when incubated at the high oxaliplatin concentration (30
M). Compound 12 was instead effective against 10 μM oxaliplatin. Regarding the other tested
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
compounds, the 6-phenol derivatives 11 and 21 showed partial activity at both oxaliplatin
concentrations, whereas the lipoic derivative 20, contrary to our expectations, turned out to be
ineffective (Table 4). This latter result might be due to a possible instability of 20 in the microglia
assay medium, as reported in the “Chemical stability study” section.
Table 4. Compound effects on microglial cell viabilitya
Cell viability (%)
Oxa 0 μM
Oxa 10 μM
Oxa 30 μM
Control
100.0 ± 7
33.2 ± 1.3**
19.1 ± 0.8**
DMSO 0.75% 90.9 ± 8
1
1
1 10 μM
2 10 μM
46.4 ± 1.4^^
48.7 ± 0.8^^
54.5 ± 1.9^^
38.9 ± 1.6
28.6 ± 1.21^
23.0 ± 0.9
15 10 μM
0 10 μM
21 10 μM
34.2 ± 0.60^
18.3 ± 0.3
2
43.5 ± 1.8^
30.6 ± 2.2^
a
Primary rat microglia cells were plated 4000 cells/well and 24 hours later cells were treated with
oxaliplatin (Oxa) 10 and 30 μM in presence of 11, 12, 15, 20 and 21 at 10 μM for 24 hours. Cell
vitality was assessed via MTT assay. Viability is expressed as % in comparison to the control cells
(arbitrarily set 100% of viable cells). Data are presented as mean ± SEM of three experiments.
*
P<0.05 and **P<0.01 versus control; ^P<0.05 and ^^P<0.01 versus oxaliplatin.
Further investigations were carried out on these compounds by evaluating their ability to prevent
the oxaliplatin-dependent increase of the SOD-inhibitable superoxide anion (cytochrome C assay).
According to the obtained data, compounds 15 and 21 proved to be effective in significantly
2
6
ACS Paragon Plus Environment

Page 27 of 69
Journal of Medicinal Chemistry
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
decreasing the oxygen free radical level thus suggesting a direct antioxidant activity or,
hypothetically, a protective property against mitochondrion (Figure 7).
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 7. Compound effects on SOD-inhibitable O2.− concentrations in rat microglia cells.
5
Microglia cells (5×10 cells/well) were exposed to 100 μM oxaliplatin for 4h in the absence or
presence of tested compounds (10 μM). O2.− concentration was evaluated by cytochrome c assay.
The nonspecific absorbance was measured in the presence of SOD (300 mU/ml) and subtracted
from the total value. Values are expressed as the mean ± SEM of three experiments. *P<0.05 and
**P<0.01 versus control; ^P<0.05 and ^^P<0.01 versus oxaliplatin.
The activity of the detoxifying enzyme catalase was also measured to study the potential effect of
new compounds on peroxisomes, the other intracellular organelle involved in the redox balance. As
shown in Figure 8, oxaliplatin impaired peroxisome functionality, reducing catalase activity while
1
5 and 21 significantly prevented the damage.
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 69
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
5
Figure 8. Compound effects on catalase activity. Microglia cells (5·10 cells/well) were treated
with oxaliplatin (10 µM) in the absence or in the presence of the new compounds (10 µM). Activity
was measured after 24h incubations. Values are expressed as the mean ± S.E.M. of three
experiments. Control condition was arbitrarily set as 100%. *P<0.05 and **P<0.01 versus control;
^P<0.05 and ^^P<0.01 versus oxaliplatin.
Behavioral studies in the oxaliplatin-induced neuropathy model. On the basis of the results
obtained from in vitro tests, we selected compounds 12, 15 and 21 for an in vivo study in a mouse
model of oxaliplatin-induced neuropathy. On day 14, after a repeated treatment with the anticancer
61
drug injected at a clinically-relevant dose, the hypersensivity to a cold non noxious stimulus (cold
plate test) was significantly established (Figure 9). The pain-relieving effects of new synthesized
compounds were tested after a single per os administration. Compounds 12 and 21 (10 and 30 mg
-1
kg ) were able to increase the pain threshold between 15 and 45 min after treatment. Interestingly,
-1
compound 15 induced significant relieving effects starting from the 0.03 mg kg dose. The
efficacy, dose-dependently, increased till it completely reverted oxaliplatin-induced neuropathic
2
8
ACS Paragon Plus Environment

Page 29 of 69
Journal of Medicinal Chemistry
1
2
3
4
-1
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
pain when administered at 1 and 10 mg kg . The effect of 15 began 30 min after treatment, and the
compound was still fully active at 60 min. In the same model, lipoic acid, administered per os at an
-1
-1
equimolar dose (4 mg kg ) to 10 mg kg of compound 15, was completely inactive. Also when
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
tested at 10-fold higher dose (40 mg kg ), lipoic acid was ineffective (Figure 9).
Figure 9. Compound effects against neuropathic pain. Mice were repeatedly treated with
-1
oxaliplatin (2.4 mg kg ; dissolved in 5% glucose solution and i.p. administered). On day 14,
compounds were suspended in carboxymethylcellulose and administered p.o. Pain-related behavior
(
i.e. lifting and licking of the hind paw) was observed and the time (in seconds) of the first sign
was recorded. **P<0.01 vs vehicle + vehicle treated animals; ^P<0.05 and ^^P<0.01 vs oxaliplatin
vehicle treated animals. Each value represents the mean of 10 mice performing in two different
experimental sets.
+
2
9
ACS Paragon Plus Environment