Triazolopyridazine LRRK2 kinase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2013.02.043

Leucine-rich repeat kinase 2 (LRRK2) has been implicated in the pathogenesis of Parkinson's disease (PD). Inhibition of LRRK2 kinase activity is a therapeutic approach that may lead to new treatments for PD. Herein we report the discovery of a series of [

Full text of DOI:10.1016/j.bmcl.2013.02.043

Bioorganic & Medicinal Chemistry Letters 23 (2013) 1967–1973
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Triazolopyridazine LRRK2 kinase inhibitors
a
b
a
a
Maurizio Franzini a, , Xiaocong M. Ye , Marc Adler , Danielle L. Aubele , Albert W. Garofalo ,
Shawn Gauby c, Erich Goldbach c, Gary D. Probst a, Kevin P. Quinn c, Pam Santiago c, Hing L. Sham a,
Danny Tam d, Anh Truong a, Zhao Ren d
⇑
a Department of Medicinal Chemistry, Elan Pharmaceuticals, 180 Oyster Point Blvd, South San Francisco, CA 94080, United States
b Department of Molecular Design, Elan Pharmaceuticals, 180 Oyster Point Blvd, South San Francisco, CA 94080, United States
c Department of Drug Metabolism and Pharmacokinetics, Elan Pharmaceuticals, 180 Oyster Point Blvd, South San Francisco, CA 94080, United States
d Department of Assay Development, Elan Pharmaceuticals, 180 Oyster Point Blvd, South San Francisco, CA 94080, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Leucine-rich repeat kinase 2 (LRRK2) has been implicated in the pathogenesis of Parkinson’s disease (PD).
Inhibition of LRRK2 kinase activity is a therapeutic approach that may lead to new treatments for PD.
Herein we report the discovery of a series of [1,2,4]triazolo[4,3-b]pyridazines that are potent against both
wild-type and mutant LRRK2 kinase activity in biochemical assays and show an unprecedented selectiv-
ity towards the G2019S mutant. A structural rational for the observed selectivity is proposed.
Ó 2013 Elsevier Ltd. All rights reserved.
Received 13 December 2012
Revised 1 February 2013
Accepted 7 February 2013
Available online 14 February 2013
Keywords:
LRRK2
Parkinson’s
Inhibition
Kinase
Treatment
Discovery
Wild-type
G2019S mutant
Phosphorylation
SAR
Activity
Selectivity
Bioisostere
Hinge
HTS
Oxidative metabolism
Binding mode
Synthesis
Mutations in most genes associated with Parkinson’s disease
(PD) have been correlated with early onset or pathologically atyp-
ical forms of the disease. A recently identified PD-associated PARK
gene that encodes for leucine-rich repeat kinase 2 (LRRK2) has
been related to late onset PD1 and has been indicated as the single
most prevalent cause of monogenic PD.2 Because the clinical phe-
notype deriving from LRRK2 mutations is very similar to idiopathic
PD, LRRK2 has emerged as one of the most relevant targets for
pharmacological intervention in PD pathogenesis.3 Mutations in
LRRK2 are dominantly inherited and several have been shown to
enhance kinase activity.4 The most prevalent, causative LRRK2
mutation, G2019S, located in the kinase activation loop, leads to
a two- to threefold in vitro increase in kinase activity.5 Acute
over-expression of G2019S LRRK2 can be toxic in SH-SY5Y neuro-
blastoma cells of primary cortical neurons.6 Other experiments
have shown that over-expression of mutant LRRK2 negatively im-
pacts neurite outgrowth.7 Although a few research groups have re-
ported on new LRRK2 small molecule inhibitors,8 to the best of our
knowledge no prior work has been published on the selective inhi-
bition of G2019S LRRK2. These observations have enticed us to
pursue the development of novel mutant G2019S LRRK2 kinase
inhibitors as potential new therapeutic agents for the treatment
of PD.
At the onset of our efforts in this project, we screened an
in-house kinase inhibitor focused library of compounds as an expe-
dient way to find LRRK2 inhibitor classes. The screen used an
⇑
Corresponding author.
E-mail addresses: mauriziof@alumni.stanford.edu, mfranzini@berkeley.edu
(M. Franzini).
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.02.043

1968
M. Franzini et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1967–1973
1
N
8
N
N
N
3
N
OMe
N
N
S
N
S
CF3
6
O
S
2
1
µ
M
µ
0.242 M
LRRK2wt IC50= 0.535
LRRK2GS IC50= 0.200
wt/G2019S = 2.7
LRRK2wt IC50
=
µ
µ
M
LRRK2GS IC50= 0.050
wt/G2019S = 4.8
M
µ
Av. solubility 72 M
µ
Av. solubility 9
M
Figure 1. Initial leads.
homogeneous time-resolved fluorescence (HTRF) assay measuring
the inhibition of phosphorylation of LRRKtide.9 A series of 3-aryl-6-
arylthio-[1,2,4]triazolo[4,3-b]pyridazines emerged from the high
throughput (HTS) screen as leads. Compound 1 was attractive
due to its low molecular weight (<400) and encouraging in vitro
potency with an interesting mutant (G2019S) versus wild type
(wt) selectivity (Fig. 1). A test on a re-synthesized batch of 1 in a
time-resolved fluorescence resonance energy transfer (TR-FRET)
Figure 3. Homology model showing the proposed binding mode of 1 in the ATP
pocket of the LRRK2 kinase domain.
series of 3-thienyl analogs (vide infra). We opted to maintain this
moiety while investigating the effect of N-atom deletions in the
fused heterocyclic core (Fig. 2). Of all variations tested, only imi-
dazo[1,2-b] pyridazine 4 maintained comparable activity to the
triazolo[4,3-b]pyridazine analog 3. The most potent compound
prepared during the early stages of HTS lead expansion was the
corresponding 6-alkylthio derivative 8, with a 7.6:1 G2019S/wt
selectivity, a result that could be attributed to a potentially stron-
ger interaction with the hinge binding region through the N-1
atom within the imidazole moiety. However, compound 8 suffered
from a surprising loss of selectivity as measured in a representative
kinase panel; therefore this core modification was not pursued any
further.
Our homology model of LRRK2 binding was developed on the
basis of MLK1 (pdb ID: 3DTC).12 At the time, the 3DTC structure
had the closest homology to LRRK2 as judged by the BLAST score
(1e-25, with 31% identities). However, the 3DTC contained several
ligand induced conformational changes that were undesirable.
These included a collapsed P-loop and a partially disordered DFG
sequence. The P-loop was remodeled using homologous structures
of kinases with ATP analogues. The DYG sequence was built to
assay confirmed a biochemical potency of 0.242
and 0.050 M in G2019S LRRK2, with an inhibition ratio of 4.8 in
favor of the mutant kinase. Most remarkably, compound
lM in wt LRRK2
l
1
emerged as a selective kinase inhibitor for LRRK2. This compound
was tested against a panel of 40 selected kinases in a single-point
inhibition assay at 1 lM and found to cleanly inhibit LRRK2 with
no other kinase in the test panel showing inhibition over 18%.10
These preliminary promising results spurred our efforts into build-
ing an SAR with the goal of further optimizing the activity profile of
this scaffold while improving the low solubility11 and the poor per-
meability (21 nm/s) for 1, which was on the same order of magni-
tude as poor permeability control atenolol (6 3 nm/s).
HTS lead 2 showed comparable activity also in preference for
G2019S LRRK2 along with higher solubility. A negligible P-gp lia-
bility (efflux ratio of 0.8) was observed for 2, which was also very
permeable (280 nm/s) and comparable to the high permeability
control propranolol (170 50 nm/s). These properties represented
a significant improvement over 1. The replacement of the phenyl
ring at C-3 on compound 1 with a bioisosteric 2-thienyl group en-
hanced potency which is clearly seen when compound 3 is com-
pared with 1 (Fig. 2). This was found to be a general trend in a
resemble PKC-
i (pdb ID: 3A8X), one of the few pdb kinase struc-
tures that contain a DYG sequence.13 We propose that compound
Figure 2. Variations on the scaffold core.

M. Franzini et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1967–1973
1969
Table 1
Variations on the C-3 heterocyclic substituent
N
N
N
N
S
CF3
3
Ar
Compd
Ar
LRRK2 wt IC50
(lM)
LRRK2 G2019S IC50
(lM)
wt/G2019S
9
1H-Pyrazol-5-yl
Thiazol-5-yl
Thiazol-2-yl
Thiazol-4-yl
3-Bromothiophen-2-yl
Pyridin-2-yl
Pyridin-3-yl
Pyridin-4-yl
2-Chlorophenyl
3-Chlorophenyl
3-(Piperazin-1-yl)phenyl
4-Fluorophenyl
0.053
0.074
0.088
0.140
0.613
0.164
0.211
0.859
3.305
0.265
0.314
0.674
0.013
0.018
0.025
0.035
0.134
0.045
0.064
0.253
1.023
0.051
0.055
0.179
4.1
4.1
3.5
4.0
4.6
3.6
3.3
3.4
3.2
5.2
5.7
3.8
10
11
12
13
14
15
16
17
18
19
20
Table 2
Aliphatic variations on C-3
N
N
N
N
S
CF3
R
Compd
R
LRRK2 wt IC50
(lM)
LRRK2 G2019S IC50
(l
M)
wt/G2019S
21
22
23
24
25
Cyclopropyl
Cyclopentyl
Tetrahydro-2H-pyran-4-yl
Piperidin-4-yl
Thiophen-2-ylmethyl
0.994
1.549
5.329
>20
0.309
0.511
2.020
22.128
19.542
3.2
3.0
2.6
—
>20
—
Table 3
Variations on the C-6 arylthio substituent
N
N
N
N
N
Ar
S
N
Ar
N
N
S
S
A
B
N
Compd
Series
Ar
LRRK2 wt IC50
(l
M)
LRRK2 G2019S IC50
(
lM)
wt/G2019S
26
27
28
29
30
31
32
33
34
35
36
A
A
B
B
B
B
B
B
B
B
B
4-(Trifluoromethyl)phenyl
2-(Trifluoromethyl)phenyl
3-Bromophenyl
3-Chlorophenyl
3-Fluorophenyl
5.897
3.168
0.032
0.031
0.169
0.138
0.075
0.281
0.117
0.183
0.159
2.026
2.470
0.006
0.008
0.077
0.026
0.030
0.067
0.101
0.033
0.141
3.0
1.3
5.3
3.9
2.2
5.3
2.5
4.2
1.2
5.5
1.1
3-Methoxyphenyl
3-tert-Butylphenyl
3-(Trifluoromethoxy)phenyl
pyrimidin-2-yl
4-(Trifluoromethyl)pyrimidin-2-yl
Phenyl
1 binds to the kinase domain of LRRK2 as shown in Figure 3, where
the most relevant interaction is a H-bond in which the triazolo ring
accepts a H-bond from hinge residue A1950.
A variety of compounds with different aryl, heteroaryl, and alkyl
groups on C-3 were prepared and tested, while maintaining the
3-(trifluoromethyl)phenylthio substituent on C-6, as shown in
Tables 1 and 2.14 We observed that five-membered heterocycles
with a variety of substitution patterns (e.g., 9 through 12) were gen-
erally more potent than six-membered rings (e.g., 14 through 20). In
all cases selectivity in favor of the inhibition of mutant LRRK2 is pre-
served. Of comparable potency to this subclass of compounds, the 2-
pyridyl derivative 14 suggests a potential role played by the dihedral

1970
M. Franzini et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1967–1973
mutant vs. wild type IC50 ratio suffered significant erosion. This
observation supported the assumption that hydrophobic, aromatic
functionality in a specific orientation was necessary for optimal
inhibition of the G2019S mutant kinase. When tested in a 40-kinase
N
N
N
N
N
N
N
S
CF3
N
S
CF3
S
S
panel at 1 lM, 37 (solubility = 12 lM) showed the same excellent
selectivity for LRRK2 as seen for 1. Methyl ester 38 was found to
be about fivefold less potent than the corresponding amide, 37.
The configuration of the stereocenter in 38 does not seem to be rel-
evant, in that enantiomers 39 and 40 were virtually equipotent. Sim-
ple homologation of the alkyl substituent, as in 41, also did not
change the IC50 significantly. However, further branching into an
isopropyl group (as in 42) depressed potency. Incorporation of a
46
47
µ
µ
M
LRRK2wt IC50
=
>4
M
LRRK2wt IC50
=
>100
µ
µ
M
LRRK2GS IC50= 2.068
Av. solubility 3
M
LRRK2GS IC50= >4
Av. solubility 2
µ
µ
M
M
CO2Et
N
N
quaternary
a-carbon (43) strongly reduced activity, and virtually
N
N
N
erased any wt vs. mutant selectivity. Isosteres of the carboxylate
functionality, such as oxadiazole 44 and pyrazole 45, proved only
marginally beneficial. The average kinetic solubility of C-6 alkylthio
analogs was significantly improved for some ester derivatives
N
N
CO2Et
S
N
S
CF3
S
48
49
LRRK2wt IC50
LRRK2GS IC50= >100
µ
µ
LRRK2wt IC50
=
>100
M
M
µ
>100 M
=
(99
isosteres (88
(12 M for 37).
l
M for 39, as a representative example) and for heterocyclic
µ
M
LRRK2GS IC50= >100
µ
lM for 44), but remained low for primary amides
µ
Av. solubility 13 M
Av. solubility 31
M
l
Figure 4. C-7 and C-8 substituted analogs.
With the intent to probe the available space surrounding C-7 and
C-8, 7- and 8- methyl substituted derivatives of 3 (46 and 47, respec-
tively) were prepared and found to be virtually inactive, so were
analogs 48 and 49 (Fig. 4), an indication of the possible tight packing
in the binding pocket around this region. Further evidence for a con-
strained binding pocket came from the observed inactivity of a ser-
ies of analogs in which the thioether was replaced. Modeling
suggests that the dihedral angle around the thioether should be
close to 90°. Analogs incorporating O, as in 50, N, as in 51, or C, as
in 52 and 53, should have a wider dihedral angle and all were sub-
stantially less active (Table 5).
Oxidative metabolism soon emerged as a major liability with the
large majority of all the compounds in the sulfide series. Incubation
of 1 in NADPH-supplemented rat liver microsomes at 37 °C for
30 min completely consumed the parent molecule resulting in the
putative sulfoxide 54 and sulfone 55 metabolites. However, this oxi-
dation pathway was less extensive in human liver microsomes. The
latter compounds were independently synthesized, tested, and
found to be very weak inhibitors (Fig. 5). They were also significantly
more resistant to degradation in the microsomal stability assays.
The replacement of the arylthio moiety with carbon-based and
small heterocyclic bioisosteric groups is the focus of current syn-
thetic efforts.
angle between the C-3 heteroaryl group and the [1,2,4]triazolo[4,3-
b]pyridazine core. Indeed, although G2019S IC50 values for parent
compound 1 and meta-substituted derivatives 18 and 19 are very
close to the corresponding values for 14 (within standard deviation),
ortho-substitution on the phenyl ring, such as that seen in 17, in-
duced a 10- to 20-fold decrease in potency. Similarly, 3-bromo-2-
thienyl derivative 13 showed a six- to eightfold decrease in potency
as compared to 3. Conversely, minor substitutions at the para posi-
tion on the phenyl ring, such as the fluorine in 20, come with a three-
fold decrease in potency, suggesting a tight space within the enzyme
pocket surrounding the phenyl ring on C-3. The presence of a hydro-
gen bond acceptor at the para position was not favorable either, as
shown by a further decrease in potency with compound 16 (to be
compared with 14 and 15).
Small cycloalkyl groups at C3, such as the cyclopropane ring in
21 or the cyclopentane in 22, are detrimental to potency (Table 2).
Modest expansion to a THP ring, as in 23, further reduces potency
by a factor of 3 to 4. The presence of a H-bond donor, as in piper-
idine analog 24, completely erased kinase inhibition; the same ef-
fect was observed upon introduction of a one-carbon linker
between the core and the thienyl ring (25).
In regard to a cellular assay, inhibition of LRRK2 kinase is known
to decrease autophosphorylation at S935. This measurement can be
utilized to assess LRRK2 kinase activity in cells.15 Our cellular assay
employed HEK293 cells stably transfected with LRRK2 (G2019S).
Concerning the substitution pattern on the thioaryl group linked
to C-6, the meta position of the trifluoromethyl substituent seemed
highly preferred (compare 26 and 27 with 1, Table 3). Most rewar-
dingly, the combination of a C-3 thiazole and a C-2 3-halo-
phenylthioether afforded compounds in this series (28, 29) with
single digit nanomolar potency against mutant LRRK2 while main-
taining selectivity vs. wild type LRRK2. This gain in potency was
not preserved in the meta-fluoro analog 30, an indication that steric
bulk and polarizability of the substituent in this position might be
relevant factors to maintaining potency. Electron-donating and
withdrawing groups, as in 31 and 33, respectively, or bulkier groups,
as in 32, seemed well tolerated. Introduction of heteroatoms in the
arylthio ring, to give for instance pyrimidine derivative 35, only
marginally affected potency. Exclusion of ring substituents, as in
34 or 36, reduced potency especially against the mutant, thus eras-
ing the selectivity gained in most other cases, an intriguing finding
in our SAR.
Several compounds exhibited better than 1 lM potency and we ob-
served some correlation with in vitro biochemical results, although
with a 100-fold discrepancy possibly due to poor permeability (Ta-
ble 6).
Synthesis
Based on several experimental findings that emerged during
our synthetic efforts, intermediate 59 behaves more like an acyl
chloride than as an aryl halide due to the high electron deficiency
of the fused heterocycle. Displacement by good nucleophiles, such
thiophenates and thiolates, typically occur under mild conditions
in the presence of inorganic weak bases (rt–80 °C) to give the tar-
get sulfides in generally good yield.
As a representative example, synthesis of compound 21 from
pyridazine 56 is shown in Scheme 1. The preparation of C-3 (het-
ero)aryl derivatives (such as 3) can take advantage of an expedi-
tious one-pot cyclocondensation of 3,6-dichloropyridazine 61
As with HTS hit 2, and in contrast to the observed trend with sub-
stitution on the left-hand side of the molecule, the aryl group linked
to C-6 via the sulfide can be replaced by a polar aliphatic chain, fea-
turing an ester or a primary amide (Table 4). In this series, primary
amide 37 was of comparable potency to aryl analog 3 although the

M. Franzini et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1967–1973
1971
Table 4
Variations on the C-6 alkylthio substituent
N
N
3
N
R
N
S
6
Ar
Compd
Ar
R
LRRK2 wt IC50
0.098
(lM)
LRRK2 G2019S IC50
0.048
(lM)
wt/G2019S
NH2
37
2.0
S
O
OMe
OMe
38
39
40
0.535
0.458
0.463
0.200
0.215
0.187
2.7
2.1
2.5
S
S
O
O
OMe
OMe
S
S
O
O
41
42
43
44
0.338
1.294
2.420
0.384
0.167
0.581
3.117
0.162
2.0
2.2
0.8
2.4
NH2
NH2
S
O
O
S
S
O
N
N
45
1.662
0.798
2.1
HN
N
Cl
Installment of substituents at C-7 (as in 48) required the regio-
selective functionalization of a 3-phenyl-6-chloro-[1,2,4]triazol-
o[4,3-b]pyridazine 61 via Knochel’s zincation with TMPZnClÁLiCl,
Scheme 2.16 Substituents at C-8 (as in 47 and 49) had to be
introduced at the onset of the synthetic route, on the 3,6-dichloro-
pyridazine starting material.
On a number of cases, significant amounts of bis-thiolated side-
products were isolated whereby position C-8 was found decorated
with the thiol utilized for the substitution on the C-6 position. Such
unusual reactivity on C-8 can be rationalized by partial charges
computation on the [1,2,4]triazolo[4,3-b]pyridazine core, and is
described in the literature to occur via a vicarious nucleophilic
substitution.17
Table 6
Selected cell activity values
Compd
LRRK2 G2019S IC50
(l
M)
G2019S S935 cell IC50 (lM)
3
9
0.023
0.013
0.018
0.025
0.035
0.006
0.008
0.030
0.048
1.656
1.016
0.780
1.727
2.112
0.683
1.397
1.393
3.032
10
11
12
28
29
32
37
In summary, we have designed and synthesized a series of
[1,2,4]triazolo[4,3-b]pyridazines that are potent and selective
LRRK2 inhibitors. A homology model of LRRK2 binding was pro-
posed and an inhibitor binding mode postulated based on this
model. An SAR campaign was developed giving rise to a series of
compounds with unprecedented selectivity for the inhibition of
LRRK2 (G2019S) over wild type LRRK2. In particular, analogs 28
and 29 gave single digit nM inhibition in the biochemical assay,
with a hydrazide (62), to give the corresponding C-6 chloride (63)
in good yield. Disappointingly, transition-metal catalyzed cross
coupling of chlorides such as 63 with boronic acids and their deriv-
atives (Suzuki conditions) generally proved challenging and low-
yielding at best, mostly due to competitive hydrolysis of the C-6
chloride.

1972
M. Franzini et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1967–1973
Table 5
Variations on the C-6 linker
N
N
N
N
Y
Compd
Y
LRRK2 wt IC50
>4
(lM)
LRRK2 G2019S IC50
1.555
(lM)
50
O
CF3
CF3
51
52
>100
25
>100
4.57
N
H
CF3
CF3
O
53
26
18
CN
N
N
N
N
O
S
N
N
N
N
N
N
CF3
N
S
CF3
N
S
CF3
O
O
1
54
LRRK2wt IC50 = 2.58 M
55
µ
µ
LRRK2wt IC50 = 0.242 M
LRRK2GS IC50 = 0.050
µ
µ
LRRK2wt IC50 = 4.73 M
µ
LRRK2GS IC50 = 2.27
M
µ
M
LRRK2GS IC50 = 1.23
M
Figure 5. Metabolic oxidation of 1 to sulfoxide (54) and sulfone (55).
H
b
a
N
H2NHN
N
c
HS
CF3
N
Cl
HN
N
N
+
+
N
N
N
N
N
S
CF3
N
Cl
N
Cl
Cl
O
O
N
Cl
60
56
57
58
59
21
N
N
Cl
c
HS
CF3
d
NHNH2
N
N
+
+
N
N
N
N
S
CF3
N
Cl
S
N
O
S
S
63
60
61
62
3
Scheme 1. Reagents and conditions: (a) Et3N, CH2Cl2, rt, 85%; (b) HOAc, 100 °C, 68%; (c) K2CO3, DMF, 80 °C, 87%; (d) Et3NHCl, C6H5CF3, 140 °C, 65%.
I
a
N
b
N
N
N
c
OEt
N
N
N
N
+
HS
N
N
N
N
S
N
N
Cl
N
Cl
N
Cl
O
OEt
O
61
62
63
48
Scheme 2. Reagents and conditions: (a) TMPZnClÁLiCl, THF, rt; then I2, rt, 57%; (b) PhB(OH)2, cat [Pd(dppf)Cl2], 2 M Na2CO3, DME, 90 °C, 35%; (c) K2CO3, rt, 77%.

M. Franzini et al. / Bioorg. Med. Chem. Lett. 23 (2013) 1967–1973
1973
7. Wang, L.; Xie, C.; Greggio, E.; Parasiadou, L.; Shim, H.; Sun, L.; Chandran, J.; Lin,
X.; Lai, C.; Yang, W. J.; Moore, D. J.; Dawson, T. M.; Dawson, V. L.; Chiosis, G.;
Cookson, M. R.; Cai, H. J. Neurosci. 2008, 28, 3384.
8. See, for instance. Chen, H.; Chan, B.; Drummond, J.; Estrada, A. A.; Gunzner-
Toste, J.; Liu, X.; Liu, Y.; Moffat, J.; Shore, D.; Sweeney, Z.; Tran, T.; Wang, S.;
Zhao, G.; Zhu, H.; Burdick, D. J. Med. Chem. 2012, 55, 5536; Also: Reith, A. D.;
Bamborough, P.; Jandu, K.; Andreotti, D.; Mensah, L.; Dossang, P.; Choi, H. G.;
Deng, X.; Zhang, J.; Alessi, D. R.; Gray, N. S. Bioorg. Med. Chem. Lett. 2012, 22,
5625.
9. Homogeneous time-resolved fluorescence (HTRF) assay using GST-LRRK2.
Detection was done using biotinylated LRRKtide, Eu(K)-Ab (p-Moesin)/SA-
XL665 measured at 620 and 665 nm.
10. Compounds were tested at Invitrogen in a single-point assay. Selected tested
kinases included Abl1, Akt1, Aurora A, CDK1/cyclin B, CDK5/p25, Chek 1, CSF1R
and showed interesting potency in the cell-based assay. However,
the poor oxidative metabolic stability of this series could not be
overcome despite our efforts, thus preventing the in vivo testing
of the compounds. As a result of these limitations, further medici-
nal chemistry efforts on this series have been suspended.
Acknowledgments
The authors thank Wayman Chan for performing difficult puri-
fications. We also want to acknowledge the work of Pui Seto, Mon-
ica You, Jeanne Baker, and Kevin Tanaka for the production and
purification of antibodies used in our assays.
(Fms), ErbB4 (Her4), Flt3, GSK3b, Jak1, various MAPKs, PDGFR
(p110 ), PLK1 through 3, Raf1 (cRAF), Ret, Syk, Ttk.
a, PIK3CA
a
11. Kinetic solubility, measured in isotonic buffer at pH 7.4, average of two
determinations.
12. Hudkins, R. L.; Diebold, J. L.; Tao, M., et al J. Med. Chem. 2008, 51(18), 5680.
13. Takimura, T.; Kamata, K.; Fukasawa, K., et al Acta Crystallogr., Sect. D 2010, D66,
577.
References and notes
1. Paisan-Ruiz, C. et al Neuron 2004, 44, 595; Zimprich, A. et al Neuron 2004, 44,
601.
2. Paisan-Ruiz, C.; Nath, P.; Washecka, N.; Gibbs, J. R.; Singleton, A. B. Hum. Mutat.
2008, 29, 485.
3. Wang, G.; Pan, J.; Chen, S.-D. Prog. Neurobiol. 2012, 98, 207.
4. Vancraenenbroeck, R.; Lobbestael, E.; De Mayer, M.; Baekelandt, V.; Taymans,
J.-M. CNS Neurol. Disord. Drug Targets 2011, 10, 724.
5. Greggio, E.; Cookson, M. R. ASN Neuro. 2009, 1; West, A. B.; Moore, D. J.; Choi, C.;
Andarbi, S. A.; Li, X.; Dikeman, D.; Biskup, S.; Zhang, Z.; Lim, K. L.; Dawson, V. L.;
Dawson, T. M. Hum. Mol. Genet. 2007, 16, 223.
14. Enzyme activity was determined using GST-LRRK2(970-2527), GST-
(G2019S)LRRK2(970-2527), and LRRKtide in the presence of 100
lM ATP.
15. Dzamko, N.; Deak, M.; Hentati, F., et al Biochem. J. 2010, 430, 405.
16. Mosrin, M.; Knochel, P. Org. Lett. 2009, 11, 1837; Bresser, T.; Mosrin, M.;
Monzon, G.; Knochel, P. J. Org. Chem. 2010, 75, 4686.
17. Katriusiak, A. A. et al Pol. J. Chem. 1996, 70, 1279. Tetrahedron Lett. 2006, 47,
4259.
6. Smith, W. W.; Pei, Z.; Jiang, H.; Moore, D. J.; Liang, Y.; West, A. B.; Dawson, V. L.;
Dawson, T. M.; Ross, C. A. Proc. Natl. Acad. Sci. U.S.A. 2005, 105, 18676.