New 6-Aminoquinoxaline Derivatives with Neuroprotective Effect on Dopaminergic Neurons in Cellular and Animal Parkinson Disease Models

Article abstract of DOI:10.1021/acs.jmedchem.6b00297

Parkinson's disease (PD) is a neurodegenerative disorder of aging characterized by motor symptoms that result from the loss of midbrain dopamine neurons and the disruption of dopamine-mediated neurotransmission. There is currently no curative treatment for this disorder. To discover druggable neuroprotective compounds for dopamine neurons, we have designed and synthesized a second-generation of quinoxaline-derived molecules based on structure-activity relationship studies, which led previously to the discovery of our first neuroprotective brain penetrant hit compound MPAQ (5c). Neuroprotection assessment in PD cellular models of our newly synthesized quinoxaline-derived compounds has led to the selection of a better hit compound, PAQ (4c). Extensive in vitro characterization of 4c showed that its neuroprotective action is partially attributable to the activation of reticulum endoplasmic ryanodine receptor channels. Most interestingly, 4c was able to attenuate neurodegeneration in a mouse model of PD, making this compound an interesting drug candidate for the treatment of this disorder.

Full text of DOI:10.1021/acs.jmedchem.6b00297

Subscriber access provided by United Arab Emirates University | Libraries Deanship
Article
New 6-Aminoquinoxaline Derivatives with Neuroprotective Effect on
Dopaminergic Neurons in Cellular and Animal Parkinson Disease Models
Gael Le Douaron, Laurent Ferrié, Julia E. Sepulveda-Diaz, Majid Amar, Abha Harfouche,
Blandine Séon-Méniel, Rita Raisman-Vozari, Patrick P. Michel, and Bruno Figadère
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00297 • Publication Date (Web): 24 Jun 2016
Downloaded from http://pubs.acs.org on June 25, 2016
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 free 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 accessible to all
readers and 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.
Journal of Medicinal Chemistry 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 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
New 6-Aminoquinoxaline Derivatives with
Neuroprotective Effect on Dopaminergic
Neurons in Cellular and Animal Parkinson
Disease Models
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Gael Le Douaron†‡, Laurent Ferrié†, Julia E. SepulvedaꢀDiaz‡, Majid Amar†‡, Abha
Harfouche†, Blandine SéonꢀMéniel†, Rita RaismanꢀVozari‡, Patrick P. Michel*‡, Bruno
Figadère*†
†
BioCIS, Univ. ParisꢀSud, CNRS, Université ParisꢀSaclay, 92290, ChâtenayꢀMalabry,
France.
‡
Sorbonne Universités, Université Pierre et Marie Curie Paris 06, INSERM U1127, CNRS
UMR7225, Institut du Cerveau et de la Moelle Epinière, Paris, France.
KEYWORDS
Neurodegenerative diseases, phenotypic screening, medicinal chemistry, heterocycle
synthesis, primary neuronal cultures.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 91
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
Parkinson disease (PD) is a neurodegenerative disorder of ageing characterized by motor
symptoms that result from the loss of midbrain dopamine neurons and the disruption of
dopamineꢀmediated neurotransmission. There is currently no curative treatment for this
disorder. To discover druggable neuroprotective compounds for dopamine neurons, we have
designed and synthesized a secondꢀgeneration of quinoxalineꢀderived molecules based on
structure–activity relationship studies, which led previously to the discovery of our first
neuroprotective brain penetrant hit compound MPAQ (5c). Neuroprotection assessment in
PD cellular models of our newly synthesized quinoxalineꢀderived compounds has led to the
selection of a better hit compound, PAQ (4c). Extensive in vitro characterization of 4c
showed that its neuroprotective action is partially attributable to the activation of reticulum
endoplasmic ryanodine receptor channels. Most interestingly, 4c was able to attenuate
neurodegeneration in a mouse model of PD, making this compound an interesting drug
candidate for the treatment of this disorder.
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
INTRODUCTION
Parkinson disease (PD) is a neurodegenerative disorder of ageing, characterized by motor
symptoms resulting from the loss of dopamine (DA) neurons in the substantia nigra (SN) and
1
the depletion of the neurotransmitter DA in the striatum. The molecular mechanisms at the
origin of the disease are not yet completely understood, but several hypotheses are currently
considered. In particular, mitochondrial dysfunction, oxidative and excitotoxic stresses,
protein misfolding, calcium dyshomeostasis and neuroinflammation processes have been
ACS Paragon Plus Environment

Page 3 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
,1,3,4
reported to be involved in disease progression.
Most likely, these pathological processes
interact with one another to lead to DA cell demise.
The pharmacological treatment of PD can be divided into neuroprotective and
symptomatic therapies. In practice, nearly all of the available treatments are symptomatic
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
ones, as they act principally, if not uniquely, on the motor symptoms of the disease.
Neuroprotective agents, having the potential to slow down or reverse the natural course of the
disease, are therefore greatly needed. Many agents developed for neuroprotection in PD have
shown great promise in preclinical assays, but none of them resulted in a treatment for
6
patients until now. This is probably partly due to the fact that a large number of these
compounds have poor brain bioavailability. Another reason is our limited understanding of
7
PD pathomechanisms. Although the number of mechanisms targeted by PD drug candidates
is steadily increasing, the search for an effective neuroprotective agent is continuing.
In the search for neuroprotective molecules for PD treatment, our laboratory has
synthesized small nonꢀpeptide chemicals, based on structure–activity relationship (SAR)
8
,9
studies of molecules with neuroprotective/neuritogenic effects. We initially synthesized
hydrids of melatonin and fatty acids, i.e., natural products previously described for their
1
0,11
antioxidative, neuroprotective or neurotrophic effects.
As a result, we obtained an inꢀ
house collection of Nꢀacylated tryptamine derivatives that was screened in a PD cellular
model of DA cell death. This work led to the identification of good antioxidative and
neuroprotective/neuritogenic compounds, but, unfortunately, these products did not possess
all the physicochemical properties expected for a good bloodꢀbrain barrier (BBB)
9
permeation. Therefore, Nꢀalkylꢀ6ꢀaminoquinoxaline derivatives were synthesized in order to
obtain better BBB permeability. Interestingly, a compound with no aliphatic side chain
exerted both neuroprotective and neuritogenic activities on DA neurons in midbrain cultures,
12
through a mechanism of action, that was apparently indirect, as it involved astroglial cells.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 91
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
This product, 5c, had almost all the physicochemical prerequisites for good BBB
12
permeation. The ability of this product to cross the BBB was confirmed by two
complementary analytical methods: matrixꢀassisted laser desorption/ionization/timeꢀofꢀflight
(MALDIꢀTOF) mass spectrometry imaging of brain tissue sections and highꢀperformance
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
liquid chromatographyꢀtandem mass spectrometry (HPLCꢀMS/MS) quantification of brain
12
tissue homogenates. Interestingly, this compound also exerted antioxidant effects.
In order to improve the neuroprotective activity of compound 5c, we designed and
synthesized derivatives and carried out their screening using the same PD cellular model as
12
before. For that, several structural modifications were designed based on our previous
studies, which reported that the phenyl ring in position 3 of the quinoxaline core structure is
1
2
important for neuroprotection. In addition, the presence of a C16 aliphatic chain on an Nꢀ
1
2
alkylꢀ6ꢀaminoquinoxaline derivative was correlated with a neuritogenic activity. On this
basis, we prepared several 5c derivatives with either an aryl or a heteroaryl ring in position 3
of the quinoxaline. The role of the methyl substituent in position 2 was also studied. We also
synthesized 6ꢀaminoꢀquinoxaline derivatives functionalized by a chlorine and/or bromine
atom in position 5 and 7, as halogenation of quinolines can be associated with
13
neuroprotection, as previously reported in the case of clioquinol. Nꢀalkyl derivatives of 5c
were also synthesized since a number of Nꢀalkylꢀ6ꢀaminoquinoxaline derivatives and
propargylamineꢀderived compounds, such as rasagiline exhibit interesting pharmacological
features; rasagiline is a monoamine oxidase inhibitor used as symptomatic treatment for PD
1
4,15
and this compound is also believed to delay disease progression .
ACS Paragon Plus Environment

Page 5 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
After characterization of the neuroprotective activity of our derivatives on DA cells, we
also studied the activity of our new hit compounds on glial cells, because neuroprotection can
12
be indirect in this model system as in the case of 5c. The protective effect of compound 4c,
one of our hit compounds that was closely related structurally to 5c, had no effect on glial
cells, but seemed to operate directly on DA neurons. Its effect was due in part to the
activation of ryanodine receptor (RyR) calcium release channels. Such a mechanism of action
has been previously described for paraxanthine, the main neuroprotective metabolite of
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
16,17
caffeine, which demonstrated in vivo neuroprotection.
Interestingly, 4c also exerted
substantial neuroprotection against DA cell death triggered by trophic support withdrawal or
by induction of lowꢀlevel oxidative stress, suggesting that this compound may be effective
against different PDꢀrelated mechanisms. Most interestingly, these neuroprotective properties
were still observed in vivo as 4c exerted a robust neuroprotective action towards nigrostriatal
DA neurons in the 1ꢀmethylꢀ4ꢀphenylꢀ1,2,3,6ꢀtetrahydropyridine (MPTP) mouse model of
PD.
RESULTS
Chemistry. Diꢀsubstituted 6ꢀaminoquinoxalines were obtained by various methods.
Hinsberg condensation between a dianiline (e.g. 4ꢀnitrophenyleneꢀ1,2ꢀdiamine) and a 1,2ꢀ
dicarbonyl compound in refluxing water gave the corresponding quinoxaline in good yield.
Ketoaldehyde (e.g. pyruvaldehyde, phenylglyoxal) gave preferentially the 2ꢀsubstitutedꢀ6ꢀ
1
8
aminoquinoxaline (Scheme 1). After reduction of the nitro group to the corresponding
aniline (H , Pd/C in EtOH), the major regioisomer was obtained as a pure product after
column chromatography.
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 91
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 1. Synthesis of 6-aminoquinoxaline and 2-substituted-6-aminoquinoxalines by
Hinsberg condensation.
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
Reagents and conditions: (a) H
flash chromatography.
2 2
O, reflux, (b) H , Pd/C in EtOH, 60 °C, then purification by
Starting from 2ꢀsubstitutedꢀ6ꢀaminoquinoxalines, which are prepared by Hinsberg
condensation, various 2,3ꢀdisubstituted quinoxalines were obtained by addition of an
19
organolithium reagent at –78 °C and subsequent oxidation by manganese dioxide (Scheme
). Yields are low to moderate, depending on the nature of the organolithium reagent;
2
aryllithium gave the expected compounds in low yields. Indeed, the reactivity of the
organolithium reagent in this reaction is correlated with its basicity, which explains these
contrasting results.
Scheme 2. Synthesis of 2,3-disubstituted quinoxalines.
ACS Paragon Plus Environment

Page 7 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Reagents and conditions: (a) R
1
Li (2.5 equiv.), THF, –78 °C or tBuLi (8 equiv.), R
(5 equiv.), CHCl , reflux.
1
Br (4
equiv), –78 °C, Et O/THF if R is aryl (b) MnO
2
1
2
3
Because of the weak reactivity of aryl lithium reagents, 3ꢀarylꢀ6ꢀaminoquinoxalines were
produced by an alternative approach. Regioselective nitration at position 7 of quinoxalinꢀ2ꢀol
was performed by treatment with nitric acid in acetic acid. Then, 7ꢀnitroquinoxalinol 1l was
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
0
3 2
treated with POCl to yield 2ꢀchloroꢀ7ꢀnitroquinoxaline 1m. SnCl mediated reduction of
nitro function gave 2ꢀchloroꢀ7ꢀaminoquinoxaline 4m. Finally, palladiumꢀcatalyzed crossꢀ
coupling reaction of the latter with diverse arylboronic acids led to the expected 3ꢀarylꢀ6ꢀ
aminoquinoxalines in excellent yields. Alternatively, an acetylene function was introduced in
position 3 through a Sonogashira cross coupling reaction, albeit in a low yield. These steps
are presented in Scheme 3.
Scheme 3. Synthesis of 3-aryl-6-aminoquinoxalines from quinoxalin-2-ol.
Reagents and conditions: (a) HNO
EtOAc, reflux. (d) ArB(OH) or ArB(MIDA) (1.3 equiv.), (PPh
2.6 equiv.), dioxane/water, reflux. (e) TMSꢀacetylene or 4ꢀmethoxyꢀphenylacetylene,
PdCl (PPh , CuI, Et N, THF, reflux. When R = TMS subsequent basic treatment (K CO
3
(2 equiv.), acetic acid, rt. (b) POCl
3
, reflux. (c) SnCl
2
,
2
3
)
2
PdCl (0.03 equiv.), K
2
2
CO
3
(
2
3
)
2
3
2
3
,
MeOH, reflux 10 min) furnishes compound 4s. MIDA = Nꢀmethyliminodiacetic acid.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 91
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
Starting from the above synthesized 6ꢀaminoquinoxalines, several other derivatives were
obtained. For instance, regioselective bromination in position 5 of compound 5c was
prepared in almost quantitative yield by reaction of 5c with one equivalent of bromine in
AcOH or with a slight excess of NBS in CH Cl at room temperature (rt) to give 7c.
2
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
Chlorination of 5c was very effective as well by using NCS in CH Cl , at rt, to give
2
2
regioselectively, compound 8c. Position 7 of 8c could be halogenated (e.g. bromine in
excess) with more difficulties to give for instance compound 9c (Scheme 4).
Scheme 4. Regioselective halogenation of 6-aminoquinoxaline 5c.
Reagents and conditions: (a) Br
2
(1 equiv.), AcOH, rt., 1 h or NCS / NBS (1.2 equiv.),
DCM, rt., 2 h. (b) Excess Br , AcOH, rt, 2 h.
2
Nꢀpropargylꢀ6ꢀaminoquinoxaline derivatives have been also prepared. Thus, Nꢀpropargylꢀ6ꢀ
aminoquinoxaline derivatives 10c, 10e, 11c, 11d, 11e, 11f, 11g, 11h, 11n, 11r, were
respectively obtained by reaction of compounds 4c, 4e, 5c, 5d, 5e, 5f, 5g, 5h, 5j, 5r, with
2 3
propargylꢀbromide and K CO in hot DMF. In a similar manner, propargyl derivative 12c
was prepared from 5c and 3ꢀchloroꢀ3ꢀmethylꢀ1ꢀbutyne in presence of a catalytic amount of
21
CuCl and trimethylamine.
Nꢀpropargylation of 5ꢀbromoꢀ2ꢀmethylꢀ3ꢀphenylꢀ6ꢀ
aminoquinoxaline derivative 7c was also performed, affording compound 13c. These steps
are presented in Scheme 5.
ACS Paragon Plus Environment

Page 9 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 5. Synthesis of N-propargyl-6-aminoquinoxaline derivatives.
H N
N
N
R2
HN
N
N
2
R2
R1
a
R1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4c, R = H, R =Ph
10c, 49%
10e, 41%
11c, 40%
11d, 71%
11e, 28%
11f, 34%
11g, 40%
11h, 28%
11n, 50%
11r, 42%
1
2
4e, R =H, R =4-F-phenyl
1
2
5
5
5
5
5
5
5
5
c, R =Me, R =Ph
1 2
d, R =Me, R =4-MeO-phenyl
1
2
e, R =Me, R =4-F-phenyl
1
2
f, R =Me, R =4-Cl-phenyl
1
2
g, R =Me, R =4-Me-phenyl
1
2
h, R =Me, R =2-Naphtyl
1
2
n, R =Me, R =3,4-Cl -phenyl
1
2
2
r, R =Me, R =3,4,5-(MeO) -phenyl
1
2
3
HN
HN
N
N
H N
2
N
N
b
65%
1
2c
5c
Br
Br
5
5
6
3
2
6
H N
2
N
N
N
N
3
2
a
14 %
7
7
7c
13c
Reagents and conditions: (a) propargyl bromide (1.5 equiv.), KI (1 equiv.), K CO (1
equiv.), DMF, 70 °C. (b) 1,1ꢀdimethylꢀpropargyl chloride (1.4 equiv.), CuCl (cat.), Et N (1,4
2
3
3
equiv.), THF/water, rt.
Neuroprotective potential of newly synthesized compounds using a culture system that
models DA cell loss in PD. All synthesized compounds (4c, 4e, 4h, 4m, 4n, 4o, 4p, 4q, 4r,
4
s, 4t, 5a-k, 6a, 6d, 7c, 8c, 9c, 10c, 10e, 11c-h, 11n, 11r, 13c, 12c) were evaluated for their
neuroprotective potential using a PD culture model in which DA neurons die spontaneously
and selectively as a consequence of a mechanism involving immature astrocytes and calcium
2
2,23
dyshomeostasis.
Cultures were maintained for 10 days in the presence or absence of the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 91
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
different test compounds, and neuroprotection was then evaluated by counting surviving TH
+
immunopositive (TH ) neurons.
+
Consistent with previous results, we observed that more than 60% of TH neurons were
22,23
generally lost after 10 days of culture in the absence of any treatment.
Nꢀ(6), 2'ꢀOꢀ
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
dibutyryladenosine 3':5' cyclic monophosphate (DbcAMP), a lipophilic analog of cAMP that
24
is commonly used as a reference molecule in this model system exhibited robust protective
effects for DA neurons in the present study (Figure 1). Among a series of 25 6ꢀ
aminoquinoxaline derivatives (4c, 4e, 4h, 4m, 4n, 4o, 4p, 4q, 4r, 4s, 4t, 5b-k, 6a, 6d, 7c, 8c,
9c), 14 of them showed a significant neuroprotective effect at 50 µM (4m, 4c, 4p, 4r, 4e, 4h,
4n, 4o, 4q, 5e, 5j, 6a, 7c, 8c). This effect was either lower or higher than that provided by 5c,
used at 100 µM. Based on these results, we noticed that all synthesized compounds with a 2ꢀ
hydrogenꢀ3ꢀarylꢀ6ꢀaminoquinoxaline core structure (4c, 4e, 4h, 4n,,4o, 4p, 4q, 4r) were
significantly neuroprotective, thus suggesting that the lack of a substituent at position 2 of the
quinoxaline ring was always associated with neuroprotection. When comparing 6a to its
regioisomer 4c, we found that these two compounds had strong and similar neuroprotective
activities (compound 4c at 50 µM: 226.0 ± 9.4 % vs compound 6a at 50 µM: 187.3 ± 6.1 %).
This suggests that a lack of substitution in position 2 is not a necessary requirement for
neuroprotection. Instead, to be optimally effective 6ꢀaminoquinoxalines need to be monoꢀ
substituted in position 2 or 3 by an aryl group.
Halogenation of the quinoxaline ring (7c, 8c) or substitution of the phenyl ring in position 3
(4e, 5e, 5j) with a halogen group (chlorine or bromine) led generally to active compounds.
Replacement of the phenyl ring by a chlorine atom led also to an active compound (4m),
while replacement of the phenyl ring by an alkyl or alkyne group led to inactive (5b) or even
toxic compounds (4s, 4t, data not shown). The effects of 6ꢀaminoquinoxaline derivatives,
which are protective for DA neurons at 50 µM, are described in Figure 1. The effects of the
ACS Paragon Plus Environment

Page 11 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
other test compounds (4s, 4t, 5b, 5d, 5f, 5g, 5h, 5i, 5k, 6d, 9c) with no protective activity are
not shown. Chemical structures of all test compounds (active and inactive) are depicted in
Chart 1.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Screening of 6ꢀaminoquinoxaline derivatives in a cellular system that models
+
selective DA cell death in midbrain cultures. Number of TH neurons surviving after 10 days
in vitro (DIV) in midbrain cultures treated or not with 6ꢀaminoquinoxalines (50 µM, except
1
2
(
5c) 100 µM). Note that 5c is inactive at 50 µM. DbcAMP (1 mM) is used as reference
protective molecule for DA neurons. Inactive 6ꢀaminoquinoxaline derivatives (4s, 4t, 5b, 5d,
f, 5g, 5h, 5i, 5k, 6d, 9c) are not included in the figure. A 185 % increase in survival at 50
5
µM, is our cutꢀoff for the selection of the most active compounds (above red line). *, p< 0.05
vs controls.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 91
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
F
H N
N
N
H2N
N
N
H N
N
N
H2N
N
Cl
2
2
N
N
OH
4
c
4e
4h
4m
N
H N
N
N
H N
2
N
N
H N
N
N
H N
N
N
2
2
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
4
n
4o
4p
4q
OMe
OMe
OMe
OMe
H N
N
N
H N
N
H N
N
N
2
2
2
N
4r
4s
4t
OMe
F
H N
N
N
H N
N
H N
2
N
H N
N
N
2
2
2
N
N
5d
5b
5c
5e
Cl
H N
N
H N
N
N
H2N
N
N
2
2
N
5f
5g
5h
OMe
OMe
Cl
Cl
H N
2
N
N
H N
2
N
N
H N
2
N
N
5
i
5j
5k
OMe
Br
H N
N
N
H N
2
N
N
H N
N
N
2
2
6a
6d
7c
Cl
Cl
H N
2
N
N
H N
N
2
Br
N
8c
9c
Chart 1. Chemical structures of all 6ꢀaminoquinoxaline derivatives tested in the cellular PD
model of spontaneous DA cell death. Compounds having significant protective effects
ACS Paragon Plus Environment

Page 13 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
towards DA neurons at 50 µM are colored in purple. Protective compounds selected for
further investigations (see Figure 1 for selection criteria) are colored in green.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In order to find 6ꢀaminoquinoxaline derivatives more potent and efficient than 5c, we
selected 6 compounds of the present series (4c, 4p, 4r, 4e, 6a, 7c) on the basis of their
efficacy to protect DA neurons, at 50 µM. More specifically, we selected compounds
improving the survival rate of DA neurons at 10 DIV by more than 185 % in comparison to
corresponding control cultures, a protection similar or superior to that provided by 5c, at 100
µM. The selected compounds were then tested at 10 µM, using the same cellular assay (see
Figure 2).
Only compounds 4c, 4e and 6a remained neuroprotective at 10 µM. These compounds
possessed similar efficacy at 10 µM (compound 4c vs 4e vs 6a: 151.5 ± 6.2 % vs 144.2 ± 7.0
%
vs 134.4 ± 4.9 %), with, however, a slightly better protective effect for 4c. Interestingly
compound 4c, which has a core structure very close to that of 5c (4c: 3ꢀphenylꢀ6ꢀ
aminoquinoxaline vs 5c: 2ꢀmethylꢀ3ꢀphenylꢀ6ꢀaminoquinoxaline), was significantly more
effective than 5c at 100 µM (5c 100 µM: 199.2 ± 11.8 % vs 4c 50 µM: 226.0 ± 9.4 %). This
confirms the idea that a lack of substituent in position 2 of the quinoxaline ring is beneficial
for neuroprotection.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 91
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 2. Evaluation of previously selected 6ꢀaminoquinoxaline derivatives for their efficacy
at a concentration of 10 µM. Selected 6ꢀaminoquinoxaline derivatives were applied to
midbrain cultures at 10 µM and the survival rate of DA neurons was assessed at 10 DIV.
Derivatives not providing protection at 10 µM (4p, 4r, 7c) are not shown in this figure.
DbcAMP (1 mM) was used as reference protective molecule for DA neurons. *, p< 0.05 vs
controls.
Among another series of 12 Nꢀpropargylꢀ6ꢀaminoquinoxaline derivatives (10c, 10e, 11c,
11d, 11e, 11f, 11g, 11h, 11n, 11r, 13c, 12c), 5 compounds showed a significant
neuroprotective effect at 50 µM (10c, 10e, 11c, 11e, 13c). The protective effect was either
lower or higher than that provided by 5c, at 100 µM. The activity of Nꢀpropargylꢀ6ꢀ
aminoquinoxaline derivatives at 50 µM and their chemical structures are described in Figure
3. The effects of the other test compounds (11f, 11g, 11h, 11n, 12c) with no protective
activity are not shown. Chemical structures of all test compounds (active and inactive) are
presented in Chart 2.
ACS Paragon Plus Environment

Page 15 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Consistent with previous results obtained with 2,3ꢀdisubstitutedꢀ6ꢀaminoquinoxalines, we
noticed that among five Nꢀpropargylꢀ6ꢀaminoquinoxaline derivatives (10c, 10e, 11c, 11e,
1
3c) that are neuroprotective for DA neurons, two are not substituted in position 2 (10c, 10e),
which confirms our hypothesis that lack of substitution at position 2 of the quinoxaline ring is
beneficial for neuroprotection.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Screening of Nꢀpropargylꢀ6ꢀaminoquinoxaline derivatives in a cellular system that
+
models selective DA cell death in midbrain cultures. Number of TH neurons surviving after
1
0 DIV in cultures treated or not with Nꢀpropargylꢀ6ꢀaminoquinoxaline derivatives (50 µM).
DbcAMP (1 mM) is used as reference protective molecule for DA neurons. Inactive Nꢀ
propargylꢀ6ꢀaminoquinoxaline derivatives (11d, 11f, 11g, 11h, 11n, 11r, 12c) are not
included in this figure. A 185 % increase in survival at 50 µM, is our cutꢀoff for active
compounds (above red line). *, p< 0.05 vs controls.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 91
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
Chart 2. Chemical structures of all Nꢀpropargylꢀ6ꢀaminoquinoxaline derivatives synthesized
and tested in the cellular PD model of spontaneous DA cell death. Compounds having
significant protective effects towards DA neurons at 50 µM are colored in purple. Protective
compounds selected for further investigations (see Figure 3 for selection criteria) are colored
in green.
In order to find Nꢀpropargylꢀ6ꢀaminoquinoxaline derivatives more potent and efficient than
5c, we selected 2 compounds of this series (10c, 10e) on the basis of their efficacy to protect
DA neurons, at 50 µM. More specifically, we selected compounds improving the survival
rate of DA neurons at 10 DIV by more than 185 % in comparison to corresponding control
cultures, a protection similar or superior to that provided by 5c, at 100 µM. The selected
compounds were then tested at 10 µM, using the same cellular assay (see Figure 4).
ACS Paragon Plus Environment

Page 17 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Compounds 10c and 10e conserved a significant neuroprotective activity at 10 µM.
Compound 10c had a better efficacy than 10e and a similar efficacy to dbcAMP, at 1mM
(dbcAMP 1 mM: 225.3 ± 13.1 % vs 10c 10 µM: 211.6 ± 7.6 %). Interestingly, compound
10c, which has a core structure very close to 5c, like 4c, was significantly more potent and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
efficient than 5c (5c 100 µM: 199.2 ± 11.8 % vs 10c 10 µM: 211.6 ± 7.6 %).
Figure 4. Evaluation of previously selected Nꢀpropargylꢀ6ꢀaminoquinoxaline derivatives for
their efficacy at 10 µM. Selected Nꢀpropargylꢀ6ꢀaminoquinoxaline derivatives were applied
to midbrain cultures at 10 µM and the survival rate of DA neurons was assessed at 10 DIV.
DbcAMP (1 mM) is used as reference protective molecule for DA neurons. *, p < 0.05 vs
controls.
Effect of quinoxaline derivatives on glial cell proliferation. In the cellular model used
herein, we reported previously that neuroprotection could result from a repressive effect on
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 91
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
dividing glial cells, in particular on astrocytes and their precursor cells, which represent the
2
5,26,27
bulk of proliferating cells.
In particular, we have shown that the reference compound
12,28
dbcAMP and 5c were protective by decreasing glial cell proliferation.
Therefore, we
wished to have first a rough estimate of the impact that 6ꢀaminoquinoxaline derivatives may
have on glial cells. For that, we determined semiꢀquantitatively whether test compounds had
an effect on the density and morphology of glial cells that were labelled with the glial marker
protein, vimentin, using fluorescence microscopic examination of the cultures. We tested at
two concentrations (10 and 50 µM) the three 6ꢀaminoquinoxaline and the two Nꢀpropargylꢀ6ꢀ
aminoquinoxaline derivatives that were selected on the basis of their neuroprotective effect
for DA neurons at 10 µM. Data are presented in Table 1.
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
Semiꢀquantitative effect of test treatments on the density
+
and morphology of vimentin glial cells
Compound
ꢀ
10 µM
50 µM
ꢀ
a
dbcAMP
+++
b
5
4
c
c
ꢀ
ꢀ
++
+/ꢀ
+
4e
ꢀ
6
a
ꢀ
+/ꢀ
++
++
1
0c
+
+/ꢀ
10e
Table 1: Impact of 6ꢀaminoquinoxaline and Nꢀpropargylꢀ6ꢀaminoquinoxaline derivatives on
the density and the morphology of labeled glial cells with glial cell marker vimentin. (+/ꢀ) no
ACS Paragon Plus Environment

Page 19 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
+
a
effect or only weak effects, (+) weak effects, and (++) marked effects on vimentin cells.
b
DbcAMP (1 mM) is used as reference molecule. tested at 100 µM.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Almost all the neuroprotective compounds had at least a weak effect on glial cells. Similar
to 5c at 100 µM, Nꢀpropargyl derivatives exerted the strongest effects on the astrocytic glial
cell population. Thus, we may assume that neuroprotective effects of compounds 10c and 10e
resulted from a glialꢀdependent mechanism, as observed with the first generation compound,
5c. By contrast, compounds 6a and 4c, which possess a strong neuroprotective effect at 50
µM and 10 µM, did not seem to exert an effect on glial cells. This result is surprising for
compound 4c, since this product is structurally very close to the previous hit compound 5c.
Overall 4c presents a number of interesting features; (i) it is very effective for
neuroprotection, (ii) it is apparently inactive towards glial cells, which suggests that it may
operate via a direct effect on DA neurons and (iii) it is the closest structural homolog of 5c,
which is a druggable candidate in particular due to its capability to penetrate the brain
12
parenchyma. To further demonstrate that 4c had no significant effects on glial cells, we used
3
[
[
methylꢀ H]ꢀthymidine as a marker of proliferation of glial cells. The amount of incorporated
3
28
methylꢀ H]ꢀthymidine was measured as previously described. As shown in Figure 5, 4c
did not significantly affect the incorporation of tritiumꢀlabeled thymidine in midbrain cultures
93.8 ±1.7 % vs control 100.0 ± 1.5 %), which is consistent with our previous semiꢀ
(
quantitative analysis. This result suggests that compound 4c may be protective through a
direct effect on DA neurons. This is a key advantage, as it may reduce the risk of side effects
possibly associated with systemic drug administration in an animal model of PD and
ultimately in PD patients.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 91
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 5. 4c does not reduce glial cell proliferation in midbrain cultures. Quantification of
3
[
H]ꢀthymidine incorporation in 8 DIV cultures, treated or not with 4c (10ꢀ50 µM). DbcAMP
at 1 mM was used as a positive control. NS: no significant effect.
Characterization of the effects of 4c towards DA neurons: morphological and
functional effects. We wished to further characterize the effect of 4c towards DA neurons. In
particular, we were interested to determine (i) whether rescued DA neurons remained
functional and (ii) whether 4c exerted neuritogenic effects on these neurons. To characterize
these effects, we performed experiments in which midbrain cultures treated with different
concentrations of 4c (10, 25, 50 µM) were processed either for TH immunodetection or for
29
tritiumꢀlabeled DA uptake, a marker of both DA cell function and differentiation.
The protective effects of 4c observed between 10ꢀ50 µM were associated with an increase
of DA uptake in the same range of concentrations (see Figure 6A, B, C), suggesting that 4cꢀ
+
treated DA neurons were fully functional. Also, the rate of DA uptake per TH neuron was
significantly increased in 4cꢀtreated cultures (see Figure 6D). This latter observation
ACS Paragon Plus Environment

Page 21 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
suggested that DA neurons rescued by 4c were not only functional but that they were also
29
more differentiated, as DA uptake sites are preferentially localized on neuritic extensions.
+
Consistent with this observation, microscopic examination of the cultures revealed that TH
neurons treated with 4c had generally a more developed neuritic network in comparison to
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
+
control neurons (Figure 6B). Note that TH cell survival and DA uptake were increased in
similar proportions in cultures treated with dbcAMP. This is in agreement with the absence of
+
neuritogenic effects of this compound on TH neurons (Figure 6B).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 91
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

Page 23 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 6. Neuroprotective and neuritogenic effects of 4c onto DA neurons in midbrain
+
cultures. (A) Number of TH neurons, in 10 DIV cultures, treated or not with 4c (1ꢀ50 µM).
+
(
B) Illustration showing TH neurons, in 10 DIV cultures, treated or not with 4c (10ꢀ50 µM).
+
Illustrations of the effects of 4c on TH cells are presented under an inverted format. Scale
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
bars = 60 µm. In (A), (B), (C) and (D) DbcAMP (1mM) was used as reference protective
3
molecule. (C) Quantification of [ H]ꢀDA uptake, in 10 DIV cultures, treated or not with 4c
+
(
10ꢀ50 µM). (D) Rate of DA uptake per TH neuron in 10 DIV cultures treated or not with 4c
10ꢀ50 µM). *P < 0.05 vs control cultures.
(
Mechanism of action of 4c: role of intracellular calcium stores. Previous studies have
shown that DA neurons can be rescued in this culture system by compounds that have a
30
repressive effect on glial cells, i.e., immature astrocytes. Such a mechanism is unlikely for
c, as this compound does not prevent glial cell proliferation, as shown in Figure 5. Another
4
set of neuroprotective molecules is represented by depolarizing compounds that have the
ability of maintaining cytosolic calcium levels within a neuroprotective range of
23
concentrations. To explore the possible impact of 4c on intracellular calcium levels, we
2+
exposed 4cꢀtreated cultures to molecules having the capacity to prevent Ca influx at the
plasma membrane. We used blockers of Lꢀtype (nifedipine) and Nꢀtype (ωꢀconotoxin
MVIIA) voltage gated calcium channels. None of these compounds caused a decrease of 4c
neuroprotective effects. However, the blockade of endoplasmic reticulum calcium release
channels, namely ryanodine receptor channels (RyRs), by dantrolene led to a substantial
reduction of the effects of 4c. This indicates that RyRs may represent a potential target for
neuroprotection by 4c (see Figure 7). This is reminiscent of previous observations showing
that activation of this receptor by paraxanthine, the primary metabolite of caffeine, is
31,32,16
protective for DA neurons in experimental PD models.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 91
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 7. Neuroprotection by 4c requires calcium mobilization through RyRs. Number of
+
TH neurons in 10 DIV cultures chronically exposed or not to 4c (50 µM) in the presence or
not of nifedipine (NIF, 20µM), ωꢀconotoxin MVIIA (ωꢀCON, 0.5µM) or dantrolene (DAN,
30 µM). *P < 0.05 vs. no treated cultures. #P < 0.05 vs 4c treated cultures.
The neuroprotective effect of 4c on DA neurons is independent of Glial cell-Derived
Neurotrophic Factor (GDNF). GDNF is a trophic peptide that exerts potent neuroprotective
33
and neuritogenic effects on DA neurons. Interestingly, we observed that 4c had about the
same efficacy as GDNF in rescuing DA neurons in midbrain cultures. Furthermore, like
GDNF, 4c promoted neuritogenesis. This led us to postulate that 4c could operate by
stimulating the endogenous production of GDNF. To test this possibility, we assessed the
activity of 4c in the presence of an antiꢀGDNF antibody (ABꢀ212ꢀNA; 10 µg/ml) that
12,22
neutralizes the biological activity of the neurotrophic peptide, as previously described.
Whereas the antibody was sufficient to prevent the increase in DA cell survival resulting
ACS Paragon Plus Environment

Page 25 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
from a treatment with GDNF, it failed to reduce neuronal survival in the presence of 50 µM
of 4c (Figure 8). This indicates that the effects of 4c are unrelated to that of GDNF.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
+
Figure 8. The protective effect of 4c is unrelated to that of GDNF. Number of TH neurons,
in 10 DIV cultures, treated or not with 4c (50 µM), in presence or not of an antiꢀGDNF
antibody (ABꢀ212ꢀNA, 10 µg/mL). GDNF at 20 ng/ml was used as a positive control. *P <
#
0
.05 vs control cultures. P < 0.05 vs GDNFꢀtreated cultures.
Mature DA neurons deprived of GDNF can be efficiently rescued by 4c
To verify that the effect of 4c was not restricted to a short developmental period after plating,
we used midbrain cultures in which the spontaneous death of DA neurons was prevented by
1
2,22
chronic application of 20 ng/mL trophic peptide GDNF.
Removal of GDNF from these
+
22
cultures at 10 DIV, leads to a massive loss of TH neurons within the next 5 days. We
confirmed this finding and established that 4c was potently protective for DA neurons in this
particular setting. DA cell rescue was already highly significant at 10 µM and reached an
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 91
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
optimum between 25ꢀ50 µM. The neuroprotective effect of 4c following GDNF withdrawal
is depicted in Figure 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
2
3
4
5
6
7
8
9
0
Figure 9. Mature DA neurons dependent on GDNF for their survival can be rescued by 4c.
Rescuing effect of 4c (10ꢀ50 µM) in midbrain cultures exposed to 20 ng/ml GDNF for 10
days and then deprived of the trophic peptide between 11 DIV and 15 DIV. Comparison to
cultures maintained continuously with GDNF up to 15 DIV. *P < 0.05 vs GDNFꢀtreated
cultures. #P < 0.05 vs GDNFꢀdeprived cultures.
Antioxidant potential of 4c for DA neurons. 5c, a close structural analog of 4c is a potent
12
antioxidant molecule. Thus, we tested whether 4c could also operate via an antioxidant
34
effect. For this purpose, we used a culture setting where the death of DA neurons is caused
by a Fentonꢀtype reaction that is generated by the presence in the culture medium of
35
catalytically active iron (i.e., ferrous iron) reacting with molecular oxygen. Like 5c, we
found that 4c was strongly protective for DA neurons. Optimal effects were obtained at 5 µM
ACS Paragon Plus Environment

Page 27 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(see Figure 10); 4c at 5 µM was as effective as desferrioxamine (10 ꢀM), a molecule with
36,37
iron chelating properties that has a clinical use.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 10. Antioxidant effect of 4c. Neuroprotective effect of 4c (1ꢀ10 µM) for DA neurons
in a model system of chronic oxidative stress. Desferrioxamine (Desf 10 µM) is used as a
positive control. *P < 0.05 vs control cultures.
Neuroprotective effects of 4c on SN DA neurons in the MPTP mouse model of PD. One of
the most commonly used animal models for DA depletion in PD is achieved through
systemic administration of the neurotoxin MPTP in mice. This PD model reproduces a
number of pathological processes, such as oxidative stress, mitochondrial dysfunction, and
38,39
inflammation that are thought to participate in disease progression.
Herein, C57BL/6 mice
received a subꢀchronic MPTP treatment in order to induce the degeneration of nigrostriatal
40
DA neurons and a depletion in striatal DA and related metabolites.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 91
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
SN DA neurons were detected by immunostaining the TH enzyme using an antiꢀTH
antibody, followed a revelation step with the chromogenic substrate 3,3ꢀdiaminobenzidine
40,41
(
DAB). In line with previous studies,
a subꢀchronic MPTP treatment, in C57BL/6 mice,
led to a massive degeneration of nigrostriatal DA neurons (MPTP/vehicle: 61 ± 5% vs
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
+
vehicle: 100 ± 2% TH neurons). Following 11 days of 4c treatment, at the dose of 2x25
+
mg/kg/day (per os), there was a partial but substantial rescue of TH neurons in comparison
to MPTP/vehicleꢀtreated mice (MPTP/4c (2x25 mg/kg/day): 78 ± 7% vs MPTP/vehicle: 61 ±
+
5
% TH neurons). At the highest dosage (2x50 mg/kg/day), 4c prevented almost entirely the
toxic effect of MPTP on DA neurons (MPTP/4c (2x50 mg/kg/day): 91 ± 5% vs vehicle: 100
+
± 2% TH neurons, no significant difference; P
>
0.05). The neuroprotective effect of 4c onto
DA neurons in the SN of MPTPꢀtreated mice is depicted in Figure 11. Importantly, we found
that 4c was only a poor inhibitor of MAOꢀB, the enzyme that converts MPTP into its active
+
38,39
metabolite MPP within the brain.
More specifically, IC50 values for the reference MAOꢀ
B inhibitor selegiline and 4c were estimated to be 1.86 and 513 µM, respectively. This
signifies that the neuroprotective effect of 4c against MPTP is unlikely to result from a
+
reduced bioavailability of MPP for target DA neurons.
ACS Paragon Plus Environment

Page 29 of 91
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment