Benzothiazole-Based LRRK2 Inhibitors as Wnt Enhancers and Promoters of Oligodendrocytic Fate

Article abstract of DOI:10.1021/acs.jmedchem.9b01752

Leucine rich repeat kinase 2 (LRRK2) is an enigmatic enzyme and a relevant target for Parkinson's disease (PD). However, despite the significant amount of research done in the past decade, the precise function of LRRK2 remains largely unknown. Moreover, the therapeutic potential of its inhibitors is in its infancy with the first clinical trial having just started. In the present work, the molecular mechanism of LRRK2 in the control of neurogenesis or gliogenesis was investigated. We designed and synthesized novel benzothiazole-based LRRK2 inhibitors and showed that they can modulate the Wnt/β-catenin signaling pathway. Furthermore, compounds 5 and 14 were able to promote neural progenitors proliferation and drive their differentiation toward neuronal and oligodendrocytic cell fates. These results suggest potential new avenues for the application of LRRK2 inhibitors in demyelinating diseases in which oligodendrocyte cell-death is one of the pathological features.

Full text of DOI:10.1021/acs.jmedchem.9b01752

Subscriber access provided by BIU Pharmacie | Faculté de Pharmacie, Université Paris V
Article
BENZOTHIAZOLE-BASED LRRK2 INHIBITORS AS WNT
ENHANCERS AND PROMOTERS OF OLIGODENDROCYTIC FATE
Josefa Zaldivar-Diez, Lingling Li, Ana M. Garcia, Wenning Zhao, Cristina Medina-
Menendez, Stephen J. Haggarty, Carmen Gil, Aixa V. Morales, and Ana Martínez
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b01752 • Publication Date (Web): 11 Dec 2019
Downloaded from pubs.acs.org on December 11, 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 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
REVISED 5 12.11.2019
BENZOTHIAZOLE-BASED LRRK2 INHIBITORS AS WNT ENHANCERS
AND PROMOTERS OF OLIGODENDROCYTIC FATE
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
1
3
Josefa Zaldivar-Diez, Lingling Li, Ana M. Garcia, Wen-Ning Zhao, Cristina
2
3
1,4
2*
Medina-Menendez, Stephen. J. Haggarty, Carmen Gil, Aixa V. Morales, Ana
Martinez1,4*
1
Centro de Investigaciones Biológicas-CSIC, Ramiro de Maeztu 9, 28040 Madrid
(Spain)
2
Instituto Cajal-CSIC, Av. Doctor Arce, 37, 28002 Madrid (Spain)
3
Chemical Neurobiology Lab, Center for Genomic Medicine, Departments of
Neurology and Psychiatry, Massachusetts General Hospital, Harvard Medical School,
1
85 Cambridge Street, Boston, MA 02114 (USA)
4Centro de Investigación Biomédica en Red de Enfermedades Neurodegenerativas
(CIBERNED), Instituto Carlos III, 28031 Madrid (Spain).
*Correspondence to:
Prof A Martinez (ana.martinez@csic.es)
Dr A. V. Morales (aixamorales@cajal.csic.es)
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 62
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
Leucine Rich Repeat Kinase 2 (LRRK2) is an enigmatic enzyme and a relevant target
for Parkinson’s Disease (PD). However, despite the significant amount of research done
in the last decade, the precise function of LRRK2 remains largely unknown. Moreover,
the therapeutic potential of its inhibitors is in its infancy with the first clinical trial
having just started. In the present work, the molecular mechanism of LRRK2 in the
control of neurogenesis or gliogenesis was investigated. We designed and synthesized
novel benzothiazole-based LRRK2 inhibitors and showed that they can modulate the
Wnt/β-catenin signaling pathway. Furthermore, compounds 5 and 14 were able to
promote neural progenitors proliferation and drive their differentiation towards neuronal
and oligodendrocytic cell fates. These results suggest potential new avenues for the
application of LRRK2 inhibitors in demyelinating diseases in which oligodendrocyte
cell-death is one of the pathological features.
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
Keywords: LRRK2 inhibitors, adult neurogenesis, Wnt enhancers, oligodendrocyte
differentiation
2
ACS Paragon Plus Environment

Page 3 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Introduction
Leucine Rich Repeat Kinase 2 (LRRK2) is an enigmatic enzyme that is considered one
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
of the most attractive targets for Parkinson’s Disease (PD). Since its discovery in
2
2004, LRRK2 research has progressed significantly in recent years with a growing
number of reports of LRRK2 interactors and more selective and potent LRRK2 kinase
3
inhibitors. However, the potential pathophysiological roles of LRRK2 is much more
complicated and it can be related to other neurodegenerative disorders, opening the
translational potential of its inhibitors beyond PD. It is crucial to take into consideration
that LRRK2 functions have been mainly studied in neurons, where the expression of
endogenous LRRK2 is low, and therefore, more studies are needed to have a better
understanding of the role played by LRRK2 inside and outside the central nervous
sytem (CNS).
LRRK2 is considered to be a large protein constituted of 2527 amino acids and
classified as a member of the ROCO superfamily, which is characterized by the
presence of a Ras/GTPase C-domain (ROC), a kinase domain and a carboxy-terminal of
4
Roc sequence (COR), which linking them together. In addition, domains related to
protein-protein interaction are found: the N-terminal region of LRRK2 is formed by the
5
armadillo (ARM) domain, the ankyrin (ANK) domain comprising seven ankyrin-type
6
repeats, thirteen leucine-rich repeats in the LRR domain, followed by the catalytic
domain with both GTPase and activity (ROC-COR-Kinase) and finally the WD40
domain, which consists of C-terminal repeats.7
The physiological functions of LRRK2 have been largely related to its two enzymatic
activities: the kinase domain that catalyzes phosphorylation and the ROC-GTPase
domain that is involved in GTP-GDP hydrolysis. In addition to those catalytic
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 62
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
functions, multiple protein-protein interaction regions indicate that LRRK2 may act as a
scaffolding protein contributing to the formation of a multiprotein signaling complex.8
A step forward in understanding the physiological role of LRRK2 was the discovery of
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
Rab GTPases as substrates of LRRK2 catalytic activity. Additionally, growing
evidence implicates LRRK2 function in remarkably diverse pathways, which involve
1
0
11
12
the regulation of transcription, regulation of translation, mitochondrial function,
neurite outgrowth,13 cytoskeletal dynamic,14 vesicle trafficking,15 autophagic protein
16
17
18
degradation, inflammation and regulation of adult neurogenesis.
19
In a previous work from our group, we presented a new family of indolinone-like
LRRK2 inhibitors that showed a proneurogenic effect as they increased the proliferation
rate of neural precursor cells isolated from adult mice ventricular-subventricular zone.
However, it is unknown what the underlying molecular mechanism of LRRK2 in the
control of neurogenesis or gliogenesis entails and if LRRK2 regulates the generation of
new neurons or glial cells in the context of adult brain. One of the possible mechanisms
could be through the control of Wnt/β-catenin signaling pathway. Wnt/β-catenin
2
0
signaling is known as one of the major players in all stages of brain development and
in the generation of new neurons and oligodendrocytes in the adult brain.21-22 Using a
2
3
Wnt signaling reporter mouse, Wnt/β-catenin signaling activity has been observed in
neural stem cells and neural progenitor cells of the ventricular-subventricular zone (V-
SVZ). Also, Wnt/β-catenin activity is upregulated as an endogenous repairing
2
4
25
mechanism in ischemic stroke and traumatic nerve injury models.
Growing evidence suggests that deficiencies in Wnt signaling in the adult brain may be
linked to the etiology of several neurodegenerative diseases.26 Furthermore, Wnt/β-
catenin activation with specific small molecules modulators, such as some glycogen
synthase kinase-3β (GSK-3β) inhibitors, has been shown to promote neuroregeneration
4
ACS Paragon Plus Environment

Page 5 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
27
in ageing mouse models of PD by rescuing midbrain dopaminergic stem cells. Taken
together, these data suggest that pharmacological modulation of the Wnt signaling in the
adult brain might lead to new therapeutic strategies for neurological diseases.28
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
2
9
LRRK2 plays a key role in Wnt signaling, with reported direct physical interactions
between low-density lipoprotein receptor-related protein 6 (LRP6), the cytoplasmic
phosphoproteins known as DVL1 (Dishevelled 1) and components of the β-catenin
destruction complex.30 Pathogenic mutations in LRRK2 (R1441C, Y1699C and
3
1
G2019S) weakened those interactions with Wnt pathway components, while recently,
the gain-of-function LRRK2 mutations have been demonstrated to enhance the
3
2
repression of β-catenin signaling showing the involvement of this kinase in canonical
Wnt pathway.33
Based on this collective evidence, we aimed to find new chemically diverse LRRK2
inhibitors that could be used in a chemical genetic approach for a deep interrogation of
the role of LRRK2 in adult neurogenesis. The goal was to have pharmacological tools
able to probe the induction of Wnt signaling in human cells and to determine their
potential as regulators of adult neurogenesis using neural stem cells isolated from the
neurogenic niche of adult mouse V-SVZ both in proliferation and in differentiation
experiments. In this work, we present a class of promising novel LRRK2 inhibitors that
fulfill these characteristics and therefore may open new indications for LRRK2
inhibitors outside the PD spectrum.
Results and discussion
Design and synthesis of new compounds
3
4
Previous reported computational studies of LRRK2, led to the identification of two
hits (1-2) together with their main interactions with the protein due to a pair of strong
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 62
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
hydrogen bonds and a number of hydrophobic interaction. With these results in mind,
we designed a small series of amino[6+5]benzo-annelated compounds with a phenyl-
morpholino tail able to maintain these main interactions (Figure 1). Following a
convergent synthesis pathway, derivatives 3-5 were easily obtained as described in the
experimental section, and evaluated as LRRK2 inhibitors using a commercial
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
5
methodology. The three new compounds differ in the heteroatoms in the heterocyclic
framework. Remarkably, when the heteroatom was a sulfur atom, promising activity
was found. In fact, benzothiazole 5 showed an IC50 value of 0.70 µM for LRRK2
inhibition and was considered as a new lead for further optimization (Figure 1).
H
H
H
N
N
N
N
N
N
H C
3
O
N
N
N
1
2
Cl
O
H C
3
CH3
HO
N
X
NH
X=O
X=NH 4
X=S
3
N
O
O
5
Figure 1. Design of the first LRRK2 inhibitors.
Several structural modifications of benzothiazole 5 were proposed including
substituents with different steric and electrostatic chemical properties in different
positions of the heterocyclic ring together with the replacement of the morpholino
moiety by tails of different nature (Figure 2).
6
ACS Paragon Plus Environment

Page 7 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
OMe, OEt, OPr,
CF Me, iPr, Br,
N
3,
R
OMe
NH
Cl, F, CO Et,
O
2
X
N
MeSO2
O
5
O
Cl
O
Me
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. Proposed modifications for benzothiazole 5.
The synthesis of the N-(benzothiazol-2-yl)-4-morpholinobenzamide derivatives was
performed by a coupling reaction between the corresponding 2-amino-benzothiazole
and the 4-morpholinobenzoic acid (Scheme 1). Each 2-amino-benzothiazole was mixed
with the 4-morpholinobenzoic acid in the presence of 1-ethyl-3-(3-
dimethylaminopropyl)carbodiimide (EDCI) as a coupling agent and 4-dimethyl-
aminopyridine (DMAP) as a basic catalyzer. Previously, the 4-morpholinobenzoic acid
was activated with the coupling agent for a certain time and then the corresponding 2-
amino-benzothiazole was added. Following this procedure, it was possible to obtain the
corresponding N-(benzothiazol-2-yl)-4-morpholinobenzamide derivatives with low to
moderate yields (<50%). A total of thirteen new benzothiazoles (6-17) were synthesized
with different substituents in the benzo-annelated ring in order to see the influence of
their nature and position in the biological activity. In an attempt of increasing the
reaction yield, a different methodology was assayed for the synthesis of compound 18.
For that purpose, we used (benzotriazol-1-yloxy)tripyrrolidinophosphonium
hexafluorophosphate (PyBOP) and 1-hydroxybenzotriazole (HOBt) with the 4-
morpholinobenzoic acid, dissolved in THF to activate the acid under microwave
irradiation (50 ºC) during 1 hour and then adding the corresponding amino-
benzothiazole and heating the resulting mixture during 5 hours at 110ºC under
microwave irradiation. However, the yield was similar to the previous conditions used.
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 62
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
R4
R4
N
S
O
N
S
i) or ii)
NH
NH2
N
O
R6
R6
HO
N
O
O
6
-18
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
Scheme 1. i) EDCI, DMAP, Et N, CH Cl , r.t., ii) PyBOP, HOBt, MW (110 C, 5h), THF
3 2 2
To study the influence of the morpholino moiety in the kinase inhibition and to confirm
the relevance of this chemical fragment in the biological activity, the synthesis of N-(6-
flurobenzothiazol-2-yl)-benzamide derivatives (19-23) was proposed and executed
following the methodologies described above (Scheme 2).
O
i) or ii)
N
S
N
S
NH
R
NH3
HO
F
R
F
O
1
9-23
o
Scheme 2. i) EDCI, DMAP, Et
3
N, CH Cl , r.t., ii) PyBOP, HOBt, MW (110 C, 5h), THF
2 2
The chemical structures of all the newly synthetized compounds were assigned
unequivocally based in their analytical and spectroscopic data (see experimental
section).
LRRK2 inhibition, SAR and docking studies
The potential LRRK2 inhibition was evaluated for all the new benzothiazole derivatives
5
-23. When some activity was found, the compounds were also tested as potential
inhibitors of LRRK2 G2019S, a superactive kinase form carrying the most common
3
6
LRRK2 mutation found in familial PD patients. The assays were performed using a
fluorescent-based immunoassay for the detection of ADP formed by the kinase
35
reaction. Initially all the compounds were evaluated at a fixed concentration of 10 µM,
and the dose response curves were obtained and IC values calculated only for those
5
0
8
ACS Paragon Plus Environment

Page 9 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compounds showing higher than 50% inhibition. Only when the compounds showed
inhibition on the wild type LRRK2, the potential inhibitory activities on the mutant
LRRK2 G2019S were assayed. Results are depicted in Table 1 using the known LRRK2
inhibitor Lrrk2-in-1 as reference standard.37
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
Table 1. LRRK2 enzymatic activity, permeability (Pe) in PAMPA-BBB assay and Wnt signal
enhancement expressed as normalized fold change to DMSO of benzothiazole compounds 5-23.
R4
N
NH
R6
S
R
O
Co
mp
R4
R6
R
LRRK2
%inh@ IC (µM)
LRRK2
LRRK2
G2019S
%inh@ IC (µM)
LRRK2
G2019S
Pe
Fold increase
Wnt 10µM
1µM
a,b
-6
-1
5
0
10 cm.s
a,b
1
0µM
50
1
0µM
5
H
H
N
O
75%
0.70 ±0.13
85%
0.37±0.01
13.4±2.4
CNS+
n.d.
3.25
2.20
6
7
8
H
H
H
OMe
N
N
N
O
O
O
73%
51%
61%
0.37±0.01
2.27±1.32
0.48±0.03
93%
69%
71%
0.44±0.29
0.28±0.10
0.33±0.07
-
-
-
-
CF
3
n.d.
Me
19.0±2.6
CNS+
-
1.76
1.79
9
H
H
Cl
F
N
N
O
O
60%
66%
0.37±0.06
0.17±0.03
50%
55%
-
-
-
1
0
0.18 ±0.05
5.1±0.1
CNS+
-
1.58
2.00
1
1
1
1
1
2
3
4
Me
Cl
H
H
H
N
N
N
N
O
O
O
O
18%
32%
80%
75%
-
-
-
-
-
-
-
-
-
-
-
-
-
OEt
Br
0.30±0.06
0.51±0.14
85%
70%
0.16±0.02
0.26±0.15
n.d.
H
4.2±0.1
CNS+
-
1.25
2.00
1
5
H
H
OPr
iPr
N
N
O
O
13%
48%
-
-
-
-
-
16
17
18
3.49±0.52
70%
1.12±0.43
4.3±3.4
CNS+
4.3±2.7
CNS+
-
1.19
1.22
H
SO
2
Me
N
N
O
O
77%
2.50±0.76
80%
0.98±0.17
1.85
1.86
H
H
CO
2
Et
22%
81%
-
-
-
-
-
1
9
F
0.52±0.08
79%
0.52±0.08
15.0±1.3
CNS+
n.d.
1.43
2.29
2
2
0
1
H
H
H
H
F
F
F
F
68%
13%
16%
11%
4.46±0.97
67%
5.64±1.88
-
-
-
-
-
-
-
-
Cl
O
-
-
-
-
-
-
-
-
-
-
-
-
Me
22
OMe
O
2
3
a
standard compound Lrrk2-in-1: IC50 values: 0.02 µM and 0.02 µM in LRRK2 and LRRK2 G2019S respectively, b
IC50Values ± Standard deviation (ten different concentrations per compound, in two independent experiments)
Origin 9.0 software used; n.d.: not possible to determine in the experimental conditions
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 62
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 great majority of the newly prepared compounds were able to inhibit LRRK2 in the
submicromolar range being equipotent or at times having enhanced inhibition with the
mutant superactive LRRK2 G2019S form. Structure-activity relationships studies
pointed to a complete lack of activity when a substituent was attached to the position 4
of the benzothiazole moiety (see compounds 11 and 12 versus 8 and 9) or when a
phenyl ring was not attached to the carbonyl linker (see compounds 21-23 versus 19-
20).
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
In order to gain insight into the binding mode of these benzothiazole-based LRRK2
inhibitors, a docking study using Glide XP (Schrödinger suite program) that allows
examination of a potential binding mode to the enzyme was carried out. Although the
kinase domain of LRRK2 has not been crystallized, a LRRK2 homology model has
34
19
been published and we have reproduced it and used with success in previous work.
All the compounds showed a similar binding mode and the structure-activity
relationships could be easily explained. Derivative 10 is the one that shows the highest
activity in the wild type form of LRRK2 (IC =0.19 µM). The nitrogen in the
5
0
benzothiazole core is able to form a hydrogen bond with Asp2017 in the catalytic site.
Furthermore, the fluorine atom at position 6 of the benzothiazole may form a halogen
bond with Glu1920. In addition, the side chain of Arg1957 is flexible and it may,
therefore, form a hydrogen bond with the oxygen atom of the morpholino moiety
(Figure 3A).
1
0
ACS Paragon Plus Environment

Page 11 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
A
B
10
Asp1887
11
Asp1887
Lys1906
Glu1920
Lys1906
Glu1920
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
Arg1957
Arg1957
Asp2017
Asp2017
Figure 3. (A) Docking of compound 10 in the LRRK2-homology model. Dashed lines show
hydrogen-bonds and halogen-bonds. (B) Docking of benzothiazole 11.
However, when substitution takes place at position 4 of the benzothiazole moiety
instead of position 6, activity dramatically decreases (see 11-12 vs 8-9). This lack of
activity could be explained based on steric hindrance: the benzothiazole core is not able
to fit inside the cavity, therefore, the phenyl-morpholino moiety is the part of the
molecule allocated towards the cavity whereas the benzothiazole core is oriented to the
external part exposed to the solvent. In this last binding mode, interactions with
Glu1920, Arg1957 and Asp2017 are lost (Figure 3B). The docking of compound 11 is
shown as an example.
Regarding the nature of the substituent in the exocyclic nitrogen atom, both morpholino
or phenyl moieties were allowed by the enzyme (see 10, 19, 20 vs 21-23). The docking
of 19 is shown as an example (Figure 4A). However, when the moiety attached to
exocyclic nitrogen is bigger and/or it contains more than one aromatic ring it results in a
lack of activity. Docking of compound 21 is shown as an example, and in this case, the
loss of activity could be explain that in order to allocate the aliphatic chain attached to
the exocyclic nitrogen the benzothiazole core changes its orientation: the nitrogen atom
is now pointed to the exterior of the cavity and the sulphur atom towards the interior.
With this binding mode, both the hydrogen-bond between the nitrogen and Asp2017
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 62
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
and the halogen-bond between the fluorine and Glu1920 are impossible to be formed
(Figure 4B).
Asp1887
A
B
Asp1887
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
Lys1906
Glu1920
21
1
9
Lys1906
Glu1920
Arg1957
Arg1957
Asp2017
Asp2017
Figure 4. (A) Docking of compound 19 in the LRRK2 homology model. Dashed lines show
hydrogen-bonds and halogen-bonds. (B) Docking of compound 21 in the homology model.
These docking studies performed can be used to further optimize this family of
compounds in a lead-to-candidate process. The introduction of halogens in position 5 of
the benzothiazole core can be studied to determine if two halogen-bonds with the
enzyme could be formed and thus increase the enzymatic activity.
Prediction of blood-brain barrier (BBB) penetration
Blood-brain barrier (BBB) penetration is an essential drug-like property for any
compound developed to target neurodegenerative and neurological diseases. Here, we
used a Parallel Artificial Membrane Permeability Assay (PAMPA) as a high throughput
technique to predict the passive permeability into the brain of the biological active
3
8
benzothiazole derivatives that were synthesized. We employed a brain lipid porcine
membrane in the assay. The in vitro permeability (Pe) of ten commercial drugs through
lipid membrane extract together with the tested compounds was determined. An assay
validation was made comparing the reported permeability values (Pe) of human
commercial drugs with the experimental data obtained (Table S1). A good correlation
between experimental-described values was obtained Pe (exp) = 1.3711 (bibl) – 1.4509
1
2
ACS Paragon Plus Environment

Page 13 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
(
R = 0.972) (Figure S1). From this equation and following the pattern established in the
3
9
literature for BBB permeation prediction, compounds could be classified as CNS+
when they present a permeability > 4.03 x 10-6 cm.s-1. Based on these results
compounds 5, 8, 14, 10, 16, 17 and 19, with Pe values in that range, are predicted to be
able to cross the BBB by passive permeation (Table 1). In contrast, compounds 6, 7, 13
and 20 precipitated in the experimental conditions employed or lacked a clear UV
spectrum after filtration. Thus, it was not possible to determine their BBB permeability
prediction using this methodology.
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
Wnt signaling modulation
The pharmacological modulation of Wnt signaling is very relevant for inducing the
generation of new neurons and glial cells in the aging or neurodegenerative adult
4
0
brain. LRRK2 has been shown to repress canonical Wnt signaling acting in interaction
with key Wnt signaling components (DVL1, LRP6, amongst others), whereas many of
the pathogenic mutated forms of LRRK2 (R1441C, G2019S) decrease, even more, Wnt
3
1
signaling activity. Therefore, we tested the LRRK2 inhibitors synthesized here and
found to be predicted as BBB penetrant (Table 1) in a high-throughput-screen to
determine their ability to promote Wnt modulation. For that purpose, we used human
neural progenitor cells (NPC), derived from induced pluripotent stem cell (iPSC), that
stably express a TCF/LEF-luciferase reporter gene to monitor Wnt signaling activity.41
The expression of LRRK2 was confirmed by quantitative RT-PCR both in the iPSCs
and in the NPC line used (Figure 5).
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 62
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. Real time RT-PCR of LRRK2 expression in induced Pluripotent Stem Cells (iPSC)
and Neural Progenitor Cells (NPC). Quantitative values were normalized to iPSC (set to 1) and
GAPDH. Values represent averages of triplicates. Error bars indicate standard deviations (SD)
of the mean. *p≤ 0.05
Compound activity was assessed in the presence of Wnt3a protein delivered as Wnt3a-
expressing cells conditioned media. Compounds were tested at two fixed concentrations
of 1 and 10 µM and the results were expressed in fold increase of TCF/LEF reporter
activity with respect to vehicle DMSO as control (Table 1).
In all the assayed compounds a modest, but consistent increase of Wnt signaling was
observed. Compounds 5, 10, 14 and 19 showed more than two fold change in Wnt
activity and were selected to perform dosage dependent curves from 0.01 µM to 20.00
µM. The assay was performed in the same way as before but this time serial dilutions of
the compounds were used and each concentration of individual compound was repeated
eight times. Results were expressed in average fold increase of TCF/LEF reporter
activity with respect to that of DMSO ± SD and plotted in dose-dependent curves.
CHIR-99021, a GSK3β inhibitor and consequently a strong positive modulator of Wnt
signaling, was used as reference control (Figure 6).42
1
4
ACS Paragon Plus Environment

Page 15 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CHIR-99021
5
10
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
Log concentration (µM)
Log concentration (µM)
Log concentration (µM)
14
19
Log concentration (µM)
Log concentration (µM)
Figure 6. Dose-dependent curves of Wnt modulation (y axis represents fold change and x axis
compound concentration) of compounds 5, 10, 14 and 19 and CHIR-99021 as positive control.
We noted that a number of the observed dose responses showed steep hill coefficients,
i.e. compounds went from showing little activity to a rapid increase in activity upon
increasing the dose. This was particularly the case of the reference control GSK3
inhibitor CHIR99021 and compound 5, while compounds 10, 14 and 19 show a more
traditional concentration-dependent effect on Wnt pathway activity. Together, we
observed a clear correlation between the concentration of the compound and the fold-
change in Wnt activity. Together, these experiments demonstrate that benzothiazole-
based LRRK2 inhibitors produce a modest, but dose-dependent increase of Wnt/β-
catenin signaling in human NPCs. These data suggest that LRRK2’s kinase activity may
function to repress Wnt signaling. Mechanistically, since we measure changes in
transcription through a TCF/LEF-dependent reporter gene we surmise the inhibition of
LRRK2 leads to accumulation of nuclear β-catenin leading to transcriptional induction.
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 62
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 and to assess the modulation of Wnt signaling pathway, Axin-2 mRNA
expression, one of the known Wnt/β-catenin direct target genes, was quantified for
compounds 5, 10, 14 and 19. In all the cases we have observed an increase in mRNA
levels which confirms the activation of Wnt signaling activity previously determined
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). The four inhibitors tested have an IC value in LRRK2 inhibition in the
50
same range of activity. Thus, differences on Axin-2 expression may be associated to the
different physico-chemical properties of each compound that may influence in
solubility, cell penetration, protein binding, etc. This moderate Wnt signaling
promotion, exhibited by our benzothiazole-based LRRK2 inhibitors, could be very
relevant to rescue the pathogenic LRRK2 activity in neurodegenerative diseases where
Wnt signaling is impaired.
Figure 7. Real time RT-PCR of Axin2 expression in Neural Progenitor Cells (NPC) of
compounds 5, 10, 14 and 19. Quantitative values were normalized to DMSO (set to 1) and
GAPDH. Values represent averages of triplicates. Error bars indicate standard error of the mean
(SEM).. *p≤ 0.05
Additionally, the standard Wnt reporter assay was also run at basal condition (without
adding Wnt3a) (data not shown). In this condition, our compounds did not have activity
on their own consistent with the fact that like many compounds in human iPSC-derived
1
6
ACS Paragon Plus Environment

Page 17 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
NPCs there is often a requirement for co-stimulation with a Wnt ligand to observe
strong activation. The fact that the compounds do not activate the reporter on their own
also provides a functional signature that distinguishes them from other classes of known
Wnt pathway activators tested in the same assay, including GSK-3 and HDAC
inhibitors, as both activate basal Wnt signaling and enhance the effects of adding
Wnt3a. Thus, the lack of activity of the compounds on their own provides a measure of
specificity of the compounds.
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
Neurogenic properties of LRRK2 inhibitors
Adult NSCs continuously generate neurons throughout life in two main brain regions:
the subgranular zone of the hippocampus and the V-SVZ, adjacent to the lateral
4
3
ventricles. The V-SVZ is the largest germinal region in the adult mammalian brain
and generates olfactory bulb interneurons and oligodendrocytes. Moreover, in human
brain, it can be the source of the neuroblasts and interneurons found in the adjacent
4
4
striatum. Thus, the finding of new molecules able to increase the neurogenic potential
of the V-SVZ niche, is of great relevance. Although LRRK2 expression has been
described in different brain regions such as cortex, substantia nigra, caudate putamen,
olfactory bulb and hippocampus, amongst others,45 it has been poorly described in
4
6
neurogenic niches. To clarify which cells of the V-SVZ niche of the adult brain could
express LRRK2, we performed immunohistochemical analysis in 2.5 months old mice
brain coronal sections. Within the V-SVZ, glial fibrillary acidic protein (GFAP)-
positive type B cells with hallmark features of astrocytes are neural stem cells and have
multipotent self-renewing capacity in vitro.43
LRRK2 is clearly expressed in most of the NSCs or type B cells with radial-like
morphology expressing GFAP and Sox2 (Figure 8A and 8B) while only a
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 62
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
subpopulation of neuroblasts or type A cells, doublecortin (Dcx) positive cells expresses
LRRK2. Furthermore, we confirmed that LRRK2 was also expressed in the subgranular
zone of the dentate gyrus (Figure 9). Altogether, these results suggest a possible
involvement of this kinase in the control of the neurogenic process as it is present in
both adult brain niches.
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 8. (A, B) Coronal sections of a 2.5 months old brain mouse showing expression of
GFAP (green) and Sox2 (white) in radial glia-like NSCs in the SVZ of the lateral ventricles.
The majority of those cells coexpress LRRK2 (red, yellow arrows). (C, D) Similar sections
showing Dcx expression (green cells, arrowheads) in neuroblasts cells. LRRK2 is expressed in
the same neurogenic area (red cells, white arrows) but only in few neuroblasts was coexpressed
with Dcx (yellow arrows). The nuclei were stained with bisbenzimide (blue) in all the panels. A
total of three mice (2 males and 1 female) were used.
1
8
ACS Paragon Plus Environment

Page 19 of 62
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 9. Expression of LRRK2 in the DG of the hippocampus in the adult mouse brain.
(A, B) Coronal sections of a 2,5 months old brain mouse showing Dcx expression (green cells,
arrowheads) in neuroblasts cells and immature granule cells in the subgranular zone of the
dentate gyrus (DG). LRRK2 is also expressed in cells of the subgranular zone (red cells, white
arrows), including Dcx expressing cells (yellow arrows). LRRK2 is also expressed by cells in
the Hilus (Hi) and in pyramidal neurons of the CA3 region of the hippocampus. . A total of
three mice (2 males and 1 female) were used.
As modulation of Wnt signaling could have an impact on neurogenesis, we evaluated in
a neurosphere proliferation assay the new LRRK2 inhibitors that positively impact on
Wnt pathway. We used an in vitro neurospheres culture model derived from NSCs and
neural progenitors.47 The V-SVZ of adult mice was employed to isolate the above
mentioned stem cells. These primary cells were cultured with and without the selected
LRRK2 inhibitors at two concentrations (1 and 5 µM) using the selective and brain
48
permeable LRRK2 inhibitor PF06447475 (PF-475) as control. In the first step, we
determined the potential effect of the selected LRRK2 inhibitors on the formation of
neurospheres grown for 4 days in the presence of mitogens (FGF and EGF). The size of
neurospheres (estimated as total area of the neurospheres) was significantly higher in
those cultures treated with the selected small molecules compared to controls treated
with DMSO (Figure 10).
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 62
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 10. LRKK2 inhibitors promoted proliferation of NSCs and progenitors from adult
V-SVZ. Representative bright field micrographs showing the size of primary neurospheres after
4 days in culture in the presence of DMSO as a control (A) or LRRK2 inhibitor 14 (B). (C)
Quantitative analysis of the neurospheres in culture. The total neurosphere area was measured in
each field using an especially designed macro for IMAGE J. Results are mean values ±
standard error of the mean (SEM), represented as relatives to control value (=100), of 6 different
fields each one from triplicates and from three independent experiments. *p≤0.05; **p≤0.005.
Neurospheres were prepared separately in three experiments from a total number of four mice (2
females and 2 males).
The increase, ranging from 1.8- to 2.4-fold in the total area of the formed neurospheres,
was a reflection of the proliferating capacity of the NSCs and neural progenitors that
form the neurospheres. Thus, treatment with benzothiazole-based LRRK2 inhibitors
promotes the proliferation of NSCs and neural progenitors from adult SVZ.
2
0
ACS Paragon Plus Environment

Page 21 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In order to determine whether the most promising LRRK2 inhibitors (5 and 14) could
be acting through the modulation of Wnt signaling pathway, we performed
neurospheres proliferation experiments in the presence of a widely used Wnt signaling
inhibitor, XAV-939. This compound selectively inhibits Wnt/β-catenin-mediated
transcription through tankyrase1/2 inhibition, and thus, XAV-939 increases the protein
levels of the Axin-GSK3β complex and promotes the degradation of β-catenin.
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
Adult SVZ-derived neurospheres grown in the presence of 1 µM XAV-939 for 4 days,
showed a 30% decrease in their size in comparison with control DMSO treated cells, as
expected for a Wnt signaling inhibitor (Figure 11). Moreover, the proliferating activity
promoted by LRRK2 inhibitors 5 and 14 was completely reverted in the presence of the
Wnt signaling inhibitor XAV939 (Figure 11). In conclusion, our benzothiazole-based
LRRK2 inhibitors may promote the proliferation of adult NSCs and neural progenitors,
at least in part, through the activation of Wnt/β-catenin signaling pathway.
2
50
00
50
00
**
**
2
1
*
1
50
0
O
V
5
4
1
5
4
1
S
A
+
X
+
M
V
D
A
V
X
A
X
Figure 11. LRKK2 inhibitors promote proliferation of NSCs and progenitors
through the activation of Wnt canonical pathway. Quantitative analysis of total
neurosphere area after 4 days in culture in the presence of DMSO as a control, 1 µM
XAV393 Wnt signaling inhibitor (XAV), 1 µM LRRK2 inhibitors 5 or 14, or the
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 62
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
combination of both, as indicated. The total neurosphere area was measured in each
field using an especially designed macro for IMAGE J. Results are mean values ±
standard error of the mean (SEM), represented as relatives to control value (=100), of 6
different fields each one from triplicates and from three independent experiments.
Student’s test comparing each treatment with that of DMSO; *p≤0.05; **p≤0.005.
Neurospheres were prepared from 2 female mice
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
Next, we analyzed whether the most promising LRRK2 inhibitors (5 and 14) together
with the control reference PF-475 could regulate neural cell differentiation after
adhesion of dissociated cells from SVZ neurospheres to a poly-ornithine//fibronectin
coverslide.49 Neurospheres were cultured for 6 days with the compounds at a
th
concentration of 1µM without mitogens. On the 7 day cells were fixed and immuno-
stained using β3-tubulin as a neuronal marker, CNPase as an oligodendrocytic marker,
GFAP for astrocytes and bisbenzimide for nuclei. Images were taken by confocal
microscopy and the number of positive cells for each marker was counted. Treatment
with LRRK2 inhibitors increased significantly the number of differentiated CNPase
positive oligodendrocytes, whereas in the case of neurons, although there was a constant
tendency to increase the number of positive β3-tubulin cells compared to control, the
compound treatment did not reached statistical significance (Figure 12). Regarding the
number of GFAP positive cells, it remained constant despite treatment with the LRRK2
inhibitors (data not shown).
2
2
ACS Paragon Plus Environment

Page 23 of 62
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 12. Analysis of neuronal differentiation in NSCs and progenitors from V-SVZ adult
brain grown in the presence of LRRK2 inhibitors. Dissociated cells from adult SVZ
neurospheres grown for 7 days without mitogens in the presence of control (DMSO) (A, A’),
benzothiazole derivative 5 (B, B’), 14 (C, C’) and the control PF-475 (D,D’). Representative
immunofluorescent field micrographs showing the different neural cell types generated:
oligodendrocytes expressing CNPase (green) and neurons expressing 3-tubulin (red). The
nuclei are stained with bisbenzimide (blue) in all the panels. (E, F) Quantification analysis of
the differentiated cells in culture. Results are mean value ± SEM, represented as relatives to
total number of cells (=100), of 4 different fields each one from triplicates and from two
independent experiments. Bar in A: 50 µm. Neurospheres were prepared separately in three
experiments from a total number of four mice (2 females and 2 males).
These results suggest that benzothiazole-based LRRK2 inhibitors have the potential to
drive differentiation of neural stem cells mostly toward an oligodendrocytic fate,
opening new fields for future therapeutic application.
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 62
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
Target engagement and kinase profiling
Finally, in order to confirm that this oligodendrocytic and neurogenic potential was
indeed driven by LRRK2 inhibition, we performed target engagement experiments with
the adult V-SVZ neurospheres. Two LRRK2 auto-phosphorylation sites have been used
as a measurement of LRRK2 activity: Ser935 and Ser1292. However, some authors
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
0
report that Ser935 is not a direct LRRK2 auto-phosphorylation site. Furthermore, one
report that compared LRRK2 Ser935 and Ser1292 phosphorylation showed that the
4
8
same LRRK2 inhibitor affected differently at these epitopes. Recent experiments show
that a possible better way of measuring LRRK2 activity and thus, LRRK2 inhibitor
engagement is by measuring its downstream substrate, the phosphorylation of Rab10.51-
5
2
For this reason, we quantified the amount of phosphorylated Rab10 in neurospheres
cultures when the cells are treated with the selected benzothiazole-based LRRK2
inhibitors by Western blotting. These experiments showed that in three different
replicates the phosphorylation of Rab10 was decreased as result of LRRK2 catalytic
activity being inhibited when the cells were treated with LRRK2 inhibitors 5 and 14 at a
dose of 10µM compared to control (Figure 13).
2
4
ACS Paragon Plus Environment

Page 25 of 62
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 13. (A) LRRK2 inhibitor treatment reduces Rab10 phosphorylation. Neurosphere
cultures grown for 7 days in the presence of mitogens were treated either with DMSO (control)
or with the LRRK2 inhibitors 5 and 14. Cell pellets were obtained as described in the
experimental section. The levels of p-Rab10 were analysed by immunoblotting using a
monoclonal antibody to p-Rab10 (T73) in three different experiments. (B) Quantification
analysis. Results are mean value ± SEM of the three experiments. *p<0.05. Neurospheres were
prepared from 2 male mice.
Finally, since the catalytic domain of protein kinases is highly conserved along the
kinome, selectivity of one of the active benzothiazole-based LRRK2 inhibitors here
studied (compound 5) was determined. The percentage of protein kinase activity at a
fixed dose of 10 µM for compound 5 was calculated on a panel of 50 different protein
kinases including tyrosine and serine/threonine kinases. Moreover, CK-1δ and CK-1ε
were specifically included based in the benzothiazole central scaffold which is also
present in others potent CK-1 inhibitors.53-54 Compound 5 shows a good selectivity with
a final selectivity score or “S” score, a quantitative measure of compound selectivity
calculated by dividing the number of kinases that compounds bind to by the total
number of distinct kinases tested, excluding mutant variants,55 of 0.02 as it only
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 62
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
interacts with 1 out of 50 kinases evaluated with a percentage of activity below 35%
(Figure 14).
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 14. Protein kinase profile for benzothiazole-based LRRK2 inhibitor 5.
Conclusions
In this work a new family of LRRK2 inhibitors, based on the benzothiazole scaffold,
have been synthesized. In addition, their inhibitory activity was determined and a
potential binding mode of these new molecules and the SAR were studied using a
homology LRRK2 model. The ability of all the active compounds to penetrate the BBB
was predicted using the in vitro PAMPA assay. Those compounds that were predicted
to be BBB penetrant were used as pharmacological tools to determine if the
pharmacological inhibition of LRRK2 could promote adult neurogenesis. This was first
performed using an HTS assay to determine if the LRRK2-inhibitors modulated Wnt
signaling in the presence of Wnt3a protein with the selected compounds then further
evaluated to determine their dose dependency. The most promising inhibitors,
compounds 5, 10, 14 and 19, were then evaluated in a proliferation assay of
neurospheres prepared from the SVZ of adult mice. Using the size of the neurosphere
measured as total area of the sphere as a read out of proliferation, it was found that
2
6
ACS Paragon Plus Environment

Page 27 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
LRRK2 inhibition by chemically diverse inhibitors promoted the proliferation of NSCs
and neural progenitors. In relation to this aspect, the proliferating activity of two
selected LRRK2 inhibitors, benzothiazoles 5 and 14, was exerted, at least in part, by
enhancing Wnt signaling pathway. Moreover, LRRK2 inhibitors 5 and 14, promoted
oligodendrogliogenesis and a tendency toward neurogenesis. Lastly, in order to confirm
that this effect was indeed dependent on LRRK2 inhibition, target engagement
experiments were performed by quantification of the decrease in a Rab10
phosphorylation, a known substrate of LRRK2.
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
Taken together, our findings shown that the pharmacological inhibition of LRRK2 with
the benzothiazole-based inhibitors has the potential to activate Wnt-signaling, promote
the proliferation of NSCs and neural progenitors and drive the differentiation towards
neuronal and oligodendrocytic cell fates. This is of crucial importance in the
neurodegeneration field as these inhibitors may act as multifunctional drugs. The fact
that LRRK2 inhibitors are able to promote the differentiation towards oligodendrocytes
opens a new avenue for the application of these inhibitors in multiple sclerosis or other
demyelinating diseases in which oligodendrocyte cell-death is one of the features of the
pathology.
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 62
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
Experimental section
Chemistry. Substrates were purchased from commercial sources (Sigma Aldrich,
Acros, Fluorochem and TCI-Europe) and used without further purification. The
microwave equipment used was an Initiator TM from Biotage®. Crude residues were
purified with the indicated solvent as eluent by column chromatography carried out at
medium pressure using silica gel (E. Merck, Grade 60, particle size 0.040 – 0.063 mm,
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
30 – 240 mesh ASTM) or by IsoleraOne flash purification system from Biotage®
using columns SF10-4g Si 50 AX1368-8 from Agilent Technologies and KP-Sil 10
from Biotage®. Thin Layer Chromatography was done in chromatographic sheets
(Merck type 60 F254 with a width of 0.2 mm). Compounds were detected with UV light
1
(254 nm). H-NMR spectra were obtained on the Bruker AVANCE-300 spectrometer
from the CAI-UCM working at 300 MHz. Typical spectral parameters: spectral width
1
3
1
6 ppm, pulse width 9 μs (57°), data size 32 K. C-NMR experiments were carried out
on the Bruker AVANCE-300 spectrometer operating at 75 MHz or on a Bruker AV 500
MHz spectrometer working at 125 MHz from the CAI-UCM. The acquisition
parameters: spectral width 16 kHz, acquisition time 0.99 s, pulse width 9 μs (57°), data
size 32 K. Chemical shifts are reported in values (ppm) relative to internal Me Si and J
4
values are reported in Hz. The multiplicity of the signals (s: singlet, d: doublet, dd:
doublet of doublets, t: triplet, td: triplet of doblets, q: quadruplet, qt: quintuplet, m:
multiplet), the coupling constants (J) and the specific characterization of each
compound are shown. HPLC analyses were performed on Alliance Waters 2690
equipment, with a UV detector photodiode array Waters 2996 with MS detector
MicromassZQ (Waters), using an Sunfire column C18, 3.5 μm (50 mm × 4.6 mm) and
acetonitrile and MilliQ water (with 0.1% formic acid) as mobile phase or in an HPLC-
2
8
ACS Paragon Plus Environment

Page 29 of 62
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
MS equipment form Thermo Fisher coupled to a Finnigan TM LXQ TM detector using
an Sunfire column C18, 3.5 μm (50 mm × 4.6 mm) and acetonitrile and MilliQ water
(with 0.1% formic acid) as mobile phase. The standard gradient consisted of a 10 min
-1
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
run from 15% to 95% of acetonitrile at a flow rate of 1 mL·min . Melting points were
determined with a Buchi M560 apparatus. High resolution mass spectrometry (HRMS)
was done in a spectrometer LC/MSD 100 Hewlett Packard using positive Electro Spray
techniques (ESI). Values are expressed in mass units (m/z). Elemental analyses were
performed by the analytical department at CENQUIOR (CSIC) and CAI-UCM, and the
results obtained were within ± 0.4% of the theoretical values using the analyzer Heraeus
CHN-O-rapid. All the final compounds have a purity ≥95% tested by HPLC.
(Benzoxazol-2-yl)-4-morpholinobenzamide (3). 2-Aminobenzoxazole (64.0 mg, 0.3
mmol) were dissolved in anhydrous tetrahydrofuran under N atmosphere in a bath with
2
acetone and dry ice (-78 ºC). After 5 minutes, 1H-benzo[d][1,2,3]triazol-1-yl 4-
morpholinobenzoate (100.0 mg, 0.3 mmol) dissolved in anhydrous tetrahydrofuran (330
µL) of nBuLi solution (1.4 M in hexane) were added onto the first solution. The mixture
was stirred for 1 hour, and then the crude was washed with a saturated NaCl solution.
Afterwards, the solvent was removed under reduced pressure and the remaining solid
was purified by flash column chromatography using CH Cl /MeOH (20:1) as eluents, to
2
2
1
afford the desired product as a white solid (63.0 mg, 63%) M.p.: 220.0 - 222.0 ºC. H-
NMR (300 MHz, CDCl ): δ 8.14 - 7.93 (m, 2 H), 7.46 - 7.34 (d, J = 5.6 Hz, 1 H), 7.29 -
3
1
3
7.08 (m, 3 H), 6.81 (d, J = 8.8 Hz, 2 H), 3.83 - 3.71 (m, 4 H), 3.29 - 3.12 (m, 4 H). C-
NMR (75 MHz, CDCl ) δ 162.5, 154.2, 130.6, 124.9, 123.9, 113.6, 113.5, 110.6, 66.6,
3
+
4
7.7. Purity HPLC: 99%. MS (ESI+): m/z 324 [M+H] . Anal. (C H N O ) Calculated:
1
8
17
3
3
C 66.86%, H 5.30%, N 13.00%. Found: C 66.60%, H 5.55%, N 13.22%.
2
9
ACS Paragon Plus Environment