Design and synthesis of brain penetrant selective JNK inhibitors with improved pharmacokinetic properties for the prevention of neurodegeneration

Article abstract of DOI:10.1016/j.bmcl.2011.06.100

The SAR of a series of brain penetrant, trisubstituted thiophene based JNK inhibitors with improved pharmacokinetic properties is described. These compounds were designed based on information derived from metabolite identification studies which led to compounds such as 42 with lower clearance, greater brain exposure and longer half life compared to earlier analogs.

Full text of DOI:10.1016/j.bmcl.2011.06.100

Bioorganic & Medicinal Chemistry Letters 21 (2011) 5521–5527
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and synthesis of brain penetrant selective JNK inhibitors
with improved pharmacokinetic properties for the prevention of
neurodegeneration
Simeon Bowers a, , Anh P. Truong a, , R. Jeffrey Neitz a, Roy K. Hom a, Jennifer M. Sealy a, Gary D. Probst a,
David Quincy a, Brian Peterson a, Wayman Chan a, Robert A. Galemmo Jr. a, Andrei W. Konradi a,
Hing L. Sham a, Gergely Tóth b, Hu Pan b, May Lin b, Nanhua Yao b, Dean R. Artis b, Heather Zhang c,
Linda Chen c, Mark Dryer c, Bhushan Samant c, Wes Zmolek c, Karina Wong c, Colin Lorentzen c,
Erich Goldbach c, George Tonn c, Kevin P. Quinn c, John-Michael Sauer c, Sarah Wright d, Kyle Powell d,
Lany Ruslim d, Zhao Ren d, Frédérique Bard d, Ted A. Yednock a,b,c,d, Irene Griswold-Prenner d
⇑
⇑
a Department of Chemical Sciences, Elan Pharmaceuticals, 180 Oyster Point Boulevard, South San Francisco, CA 94080, United States
b Department of Molecular Design, Elan Pharmaceuticals, 180 Oyster Point Boulevard, South San Francisco, CA 94080, United States
c Department of Lead Finding, Drug Disposition, and Safety Evaluation, Elan Pharmaceuticals, 800 Gateway Boulevard, South San Francisco, CA 94080, United States
d Department of Biology, Elan Pharmaceuticals, 1000 Gateway Boulevard, 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:
The SAR of a series of brain penetrant, trisubstituted thiophene based JNK inhibitors with improved
pharmacokinetic properties is described. These compounds were designed based on information derived
from metabolite identification studies which led to compounds such as 42 with lower clearance, greater
brain exposure and longer half life compared to earlier analogs.
Received 19 May 2011
Revised 20 June 2011
Accepted 22 June 2011
Available online 7 July 2011
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
c-Jun N-terminal kinase inhibitor
JNK
Neurodegeneration
Trisubstituted thiophene
The c-Jun N-terminalkinases (JNKs) are members of the mitogen-
activated protein kinase (MAPK) family and activate transcription
factors such as c-jun by phosphorylation of serine/threonine resi-
dues.1 The JNKs regulate signal transduction in response to stress
stimuli such as UV light, reactive oxygen species, cytokines, hypoxia
and protein misfolding.1,2 Three isoforms of JNK have been identi-
fied: JNK1, JNK2 and JNK3.1 JNK1 and JNK2 are ubiquitously ex-
pressed while JNK3 is localized primarily in the brain. The up
regulation of the JNKs is implicated in numerous disease states such
as obesity, cancer, Type 2 diabetes and more specifically, JNK3 has
been implicated in playing important roles in neurodegenerative
diseases such as Alzheimer’s disease, Parkinson’s disease, Hunting-
ton’s disease, and cerebral ischemia.3 JNK3 knockout mice are viable
and are resistant to induced neurodegeneration compared to wild
type animals4 suggesting that inhibition of JNK may be tolerated
and neuroprotective.3
For a JNK inhibitor to be neuroprotective, it must be brain pene-
trant and have a high degree of kinase selectivity to avoid potential
toxicity. In the case of JNK this is particularly challenging due to
the closely related MAP kinases p38a and ERK2. In our two previ-
ous papers,5 we reported the development of a series of di- and tri-
substituted thiophenes which started from micromolar hits and
resulted in low nanomolar JNK inhibitors. Compounds 1 and 2
(Fig. 1) are typical examples from this series. They are essentially
equipotent on the JNK isoforms but possess excellent selectivity
for JNK over p38a and ERK2. Additionally, compound 1 has a clean
profile when screened against a panel of 38 kinases.5b Furthermore,
compound 1 demonstrated robust in vivo oral JNK activity as mea-
sured by phospho-c-jun reduction in our kainic acid mouse model,
a
crucial step towards the development of neuroprotective
agents.5b Reported in this Letter are our efforts to improve the
pharmacokinetic properties of this scaffold while retaining the
excellent kinase selectivity of this class of compounds.
Compound 1 displayed only modest stability when incubated
with liver microsomes and moderate P-gp efflux in MDR–MDCK
cells which resulted in sub-optimal brain levels when dosed orally.
⇑
Corresponding authors.
E-mail addresses: simeon.bowers@elan.com (S. Bowers), anh.truong@elan.com
(A.P. Truong).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.06.100

5522
S. Bowers et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5521–5527
O
S
O
S
Cl
CN
N
H
N
H
O
N
N
N
HN
HN
N
N
N
1
2
JNK3 IC50 = 11 nM
JNK2 IC50 = 43 nM
JNK1 IC50 = 14 nM
JNK3 IC50 = 7 nM
JNK2 IC50 = 8 nM
JNK1 IC50 = 9 nM
α
μ
α
μ
p38 IC50 > 50
M
μ
p38 IC50 > 50
M
μ
M
ERK2 IC50 > 50
HBSS solubility = 25
OxMet% m, r, h = 29, 9, 50
M
ERK2 IC50 > 50
μ
M
μ
M
HBSS solubility = 95
OxMet% m, r, h = 37, 0, 39
MDR MDCK A>B = 270 nm/sec
P-gp efflux ratio = 1.6
MDR MDCK A>B = 179 nm/sec
P-gp efflux ratio = 4.6
Figure 1. In vitro characteristics of trisubstituted thiophene based JNK inhibitors.6
O
S
Cl
NH
N
H
N
N
Figure 3. Crystal structure of compound
2 in green bound to JNK3 (2.4 Å
N
resolution) The PBD deposition code is 3RTP. For experimental conditions see Ref. 6.
1
mouse liver microsomes
CO2H
Br
F
Br
O
S
O
S
O
S
b
d
a
c
Cl
NH
Cl
NH
Cl
NH
N
N
N
N
H
N
N
H
H
N
N
F
8
N
N
N
N
N
N
N
6
7
O
OH
O
OH
CO2H
Br
Br
3
4
5
Figure 2. Metabolites of compound
1 identified by LC–MS/MS analysis of
N
F
H2N
F
N
F
compound (10 mol) incubated with mouse liver microsomes.
l
9
10
11
Scheme 1. Reagents and conditions: (a) (i) KNO3, H2SO4 (aq), 0 °C, 20 min; (ii) Fe,
NH4Cl, MeOH, 100 °C, 3 h; (iii) NaNO2, HBF4, H2O, 0 °C, 1 h then 150 °C, 18 h; (b)
DIPA, nBuLi, Pd2dba3, 20-(dicyclohexylphosphino)-N,N-dimethylbiphenyl-2-amine,
t-butylacetate, toluene, rt; (ii) HCO2H, CH2Cl2, 50 °C, 16 h; (c) glycerol, nitroben-
zene, H2SO4, FeSO4ꢀ7H2O, 80 °C, 18 h; (d) (i) Pd2(dba)3, Q-Phos, (2-tert-butoxy-2-
oxoethyl)zinc(II) bromide, THF, rt–50 °C; (ii) HOAc, HCl, 60 °C, 18 h.
In order to locate the major sites of metabolism, compound 1 was
incubated with mouse liver microsomes for 2 h and the resulting
metabolites were analyzed by LC–MS/MS. Based on this analysis
the putative structures of the major metabolites 3, 4 and 5, were
derived (Fig. 2). Interestingly, no evidence of oxidation on the thi-
ophene portion of the molecule was observed but the isoquinoline
region of the inhibitor was extensively oxidized and revealed itself
to be the metabolic soft spot in this scaffold.
It was envisioned that the incorporation of heteroatoms and/or
electron withdrawing substituents on to the biaryl ring would re-
duce metabolism on the aryl acetamide portion of the molecule.
To determine the sites of the inhibitor where modification of the
structure could take place without affecting the JNK3 potency,
the co-crystal structure of trisubstituted thiophene 2 in complex
with JNK3 in the presence of the JIP1 peptide was examined
(Fig. 3).
In the first method, fluorine substituents were incorporated into
the bicyclic ring via a Schiemann8 reaction with an appropriate
bicyclic amine followed by introduction of the acetic acid via al-
pha-arylation of the bromoheterocycle
7 in the presence
Pd2(dba)3.9 Alternatively, quinoline analogs could be accessed
using the Skraup quinoline synthesis10 starting with an appropri-
ately substituted aniline (9) followed by elaboration to the aryl
acetamide using the conditions mentioned above. This method
was used for analogs 21, 23 and 24, (Table 1).
With the aryl acetic acids in hand it remained to couple them to
an aminothiophene. Our previous synthesis of trisubstituted thio-
phenes5b involved the conversion of an ester into a triazole in a
three-step procedure at the end of the synthesis. While this
method allowed access to many compounds it suffered from low
yielding reactions over multiple steps and therefore we sought
an alternative route whereby the triazole would be formed prior
to coupling with the aryl acetic acids. This was accomplished as
shown in Scheme 2. Aminothiophene 12 was converted into thie-
no[2,3-d]pyrimidin-4(3H)-one 13 followed by dibromination and
selective debromination at the 2-position of the thiophene to give
compound 14. Treatment with POCl3 followed by hydrazine gave
compound 15 which was subsequently converted into tricyclic
intermediate 16 upon reaction with triethyl orthoformate. Reac-
tion of compound 16 with N-methylethane-1,2-diamine gave the
key synthetic intermediate, aminothiophene 17. Compound 17
Compound 2 forms a hydrogen bond between the NH of Met149
and the carbonyl of the inhibitor (Fig. 3). The nitrile at the 4-
position of the thiophene lies within proximity to the gatekeeper
residue (Met146) and previous SAR studies showed that chloro-,
bromo- and methyl- substituents were also tolerated in this posi-
tion.5b The fused bicyclic ring points toward solvent with the
carbonyl of the quinolinone engaging in a hydrogen bond with
Asn152. It is clear from this crystal structure that substituents
may be tolerated on the solvent exposed edge of the bicyclic ring
system with minimal risk of reduction in potency.
The first series of analogs that were targeted intended to im-
prove the metabolic stability of the quinoline and isoquinoline ser-
ies of aryl acetamides by the incorporation of fluoro and
trifluoromethyl substituents. The aryl acetamides were prepared
using either one of two general methods outlined in Scheme 1.7

S. Bowers et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5521–5527
5523
Table 1
JNK inhibition and in vitro metabolic stability for compounds 1 and 19–24
O
S
R'
NH
N
H
R
N
N
a
a
Compds
R
R0
JNK3 IC50 (nM)
JNK1 IC50a (nM)
14
JNK2 IC50 (nM)
OxMet% (m, r, h)b
1
Cl
11
43
68
29, 9, 50
34, 5, 44
N
19
Br
24
16
N
F
F
20
21
22
23
Br
Br
Br
Br
30
17
92
2, 17, 39
5, 18, 42
3, 20, 37
3, 16, 31
N
41
14
140
164
337
N
F
F3C
52
21
N
280
178
N
N
CF3
24
Cl
183
81
24
3, 1, 32
CF3
a
See Ref. 6 Values are means of at least three experiments at 10
See Ref. 6 Percent remaining after 30 min incubation with liver microsomes (m = mouse, r = rat and h = human).
l
M ATP concentration.
b
csubstituent at the 3-position of the quinoline (compound 22)
and substitution at the 7 and 8-positions was avoided in future
analogs.
N
O
N
O
N
S
S
S
S
H2N
NC
12
d, e
a
b, c
We next turned our attention to the quinolinone series of inhib-
itors. As reported in our earlier Letter,5b the quinolinone analogs
were often more potent against JNK3 compared to the quinolines
and isoquinolines and we endeavored to take advantage of this po-
tency increase in our search for more metabolically stable analogs.
A series of naphthyridinones was prepared and it was expected
that the additional heteroatom in the bicyclic ring would lower
the log P of the compounds and therefore improve their metabolic
stability. These compounds were prepared as outlined in Scheme 3.
HN
HN
N
Br
Br
NH
H2N
13
14
15
N
S
O
S
S
N
f
g
h
Br
NH
Br
N
N
H2N
H
R
Br
N
NH
N
N
N
N
16
17
18
Scheme 2. Reagents and conditions: (a) HCO2H, concd H2SO4, microwave, 100 °C,
15 min.; (b) Br2, AcONa, AcOH, reflux; (c) Zn, AcOH, H2O; (d) POCl3, 100 °C,
microwave, 0.5 h; (e) hydrazine, EtOH, 75 °C; (f) HC(OEt)3, EtOH, 100 °C; (g) N1-
methylethane-1,2-diamine, MeOH, 60 °C; (h) RCH2CO2H, POCl3, pyridine, 0 °C, 0.5 h.
O
O
OH
O
OH
O
H
N
N
NH2
Br
N
O
e
a, b
c, d
N
N
N
N
25
26
27
28
was then coupled with the aryl acetic acids generated as shown in
Scheme 1 to give the analogs shown in Table 1.
O
The introduction of a single fluorine substituent (19, 20 and 21)
resulted in a slight decrease in the JNK3 activity of the inhibitors
compared to compound 1 and did not improve the compounds’
metabolic stabilities. Similarly, the introduction of trifluoromethyl
substituents into the 3, 7 and 8-positions on the quinoline (22, 23,
and 24, respectively), resulted in compounds with reduced po-
tency and less metabolic stability compared to the parent com-
pound 1. Interestingly, analogs with trifluoromethyl substituents
at the 7- or 8-position of the quinoline (23 and 24) were four to
fivefold less potent than the analog with the trifluoromethyl
OH
O
N
H
NH2
N
O
N
O
f
g, h
N
NH
N
N
N
N
O
N
29
30
31
32
Scheme 3. Reagents and Conditions: (a) ethyl acrylate, Pd(OAc)2, P(o-tol)3, Et3N,
DMF, 120 °C, 24 h; (b) NaOEt, EtOH, reflux, 2 h; (c) methyl 2-bromoacetate, LiHMDS,
DMF, rt, 0.5 h; (d) NaOH, H2O, THF, 70 °C, 0.5 h; (e) Pd-C, H2 (40 psi), MeOH, 18 h; (f)
NaOEt, EtOH, 60 °C, 2 h; (g) methyl 2-bromoacetate, NaH, DMF, rt, 1 h; (h)
Me3SnOH, 1,2-dichloroethane, reflux, 2 h.

5524
S. Bowers et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5521–5527
Heck coupling of 3-amino-2-bromopyridine (25) with ethyl acry-
late followed by cyclization gave 1,5-naphthyridin-2(1H)-one
(26).11 Alkylation of 26 with methyl 2-bromoacetate followed by
ester hydrolysis gave the desired aryl acetic acid 27 which could
be hydrogenated to give 28. This route was used to prepare all
the naphthyridinones shown in Table 2. Pyrazolopyrimidines were
prepared by reaction of 3-aminopyrazole (29) with 1,3-dimethyl-
uracil to give intermediate 3112 which was alkylated as described
above and converted to the desired carboxylic acid 32 upon reac-
tion with trimethyltin hydroxide.13 These aryl acetic acids were
coupled with aminothiophene 17 as shown in Scheme 2 to give
the analogs displayed in Table 2.
In this series of analogs, the naphthyridinones were slightly
more potent than quinolinone 33 but more importantly, they were
significantly more stable when incubated with liver microsomes.
For example, compound 34 had 83% compound remaining after a
30 min incubation with human liver microsomes whereas com-
pound 33 only had 24% compound remaining. Cyanothiophene
35 was approximately twofold more potent than the correspond-
ing bromothiophene 34 and also possessed greater metabolic sta-
bility in lower species. As observed with the quinoline and
isoquinoline analogs (Table 1), the addition of trifluoromethyl sub-
stituents (37 and 38) did not provide a further increase in meta-
bolic stability. Alternative bicyclic ring systems were also
explored with pyrazolopyrimidines 39 and 40 which had similar
potency to the naphthyridinones but did not offer any advantages
in terms of metabolic stability.
and cyano (Table 3). These analogs were accessed using the same
route outlined in Scheme 3 using anilines in place of aminopyri-
dines. Difluoroquinolinone 41 provided a modest increase in stabil-
ity compared to 33 (Table 2) but the addition of trifluoromethyl
substituents was more effective with substitution at the 6-position
(42) providing the greatest increase in stability. The addition of cy-
ano substituents into the quinoline ring afforded extremely potent
compounds with compound 47 (JNK3 IC50 = 2 nM) being 14-fold
more potent than its parent 33 (JNK3 IC50 = 28 nM).
Finally, we turned our attention to the triazole region of the
inhibitor in an effort to optimize the potency and metabolic stabil-
ity of the compounds. Our earlier report8a identified both oxazoles
and thiazoles as viable triazole replacements and these thiophene
substituents were combined with the favored quinolinone and
naphthyridinone aryl acetamides. These analogs were prepared
by Stille coupling of bromothiophene 4914 with the appropriate
heterocyclic stannane followed by reduction of the nitro group
and POCl3 mediated amide formation (Scheme 4).
With the dihydronaphthyridione aryl acetamide (A) (Table 4)
both oxazol-2-yl (52) and thiazol-2-yl (53) were less stable than
their triazole counterpart (34), although, they maintained the
excellent potency against JNK3. N-Methyltriazole (54) was fourfold
more potent than NH-triazole (34) but suffered a loss in metabolic
stability presumably due to N-dealkylation. Our earlier Letter5b
demonstrated that 4-cyanothiophenes are generally more stable
that the 4-bromothiophenes and therefore this substituent was
combined with the triazole replacements. With the naphthyridi-
none aryl acetamide, cyanothiophene (55) was not significantly
more stable than (53) suggesting that the thiazole ring was causing
The SAR of the quinolinones was expanded by the addition of
electron withdrawing substituents such as trifluoromethyl, fluoro
Table 2
JNK inhibition and in vitro metabolic stability for compounds 33–40
O
S
R'
N
H
R
NH
N
N
a
a
a
Compds
R
R0
JNK3 IC50 (nM)
JNK1 IC50 (nM)
JNK2 IC50 (nM)
OxMet% (m, r, h)b
0, 0, 24
N
N
O
O
33
Br
28
25
9
42
32
34
35
Br
14
6
55, 50, 83
84, 81, 80
N
N
N
O
CN
5
18
N
O
36
37
Br
Br
7
6
4
4
8
51, 60, 85
43, 43, 77
N
F3C
N
N
O
O
21
N
38
Br
14
8
38
35, 2, 64
F3C
N
N
O
39
40
Br
Br
8
8
3
74
36
50, 71, 72
3, 42, 48
N
N
N
O
14
F3C
N
N
a
See Ref. 6 Values are means of at least three experiments at 10
See Ref. 6 Percent remaining after 30 min incubation with liver microsomes (m = mouse, r = rat and h = human).
l
M ATP concentration.
b

S. Bowers et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5521–5527
5525
Table 3
JNK inhibition and in vitro metabolic stability data for compounds 41–48
O
S
Br
NH
N
H
R
N
N
OxMet%c (m, r, h)b
a
a
a
Compds
R
JNK3 IC50 (nM)
JNK1 IC50 (nM)
JNK2 IC50 (nM)
F
F
N
O
41
5
3
15
14, 9, 38
N
O
O
42
43
9
9
2
6
5
58, 35, 67
29, 25, 44
F3C
F3C
N
O
21
N
44
71
31
193
27, 1, 49
CF3
N
O
O
45
46
47
48
6
2
12
68
10
10
16, 39, 64
10, 26, 29
60, 46, 69
44, 62, 59
F3CO
F3CO
N
19
2
11
1
N
N
O
NC
NC
O
2
1
a
See Ref. 6 Values are means of at least three experiments at 10
See Ref. 6 Percent remaining after 30 min incubation with liver microsomes (m = mouse, r = rat and h = human).
l
M ATP concentration.
b
in this series, none of these new compounds had any measurable
IC50 against ERK2 or p38 at the concentrations tested. In order
to determine the broader kinase selectivity of our latest generation
of thiophene based inhibitors, compound 42 was subjected to a
screen of 36 kinases. This screen revealed an exceptionally clean
O
S
S
a
S
X
c
X
a, b
X
N
H2N
H
O2N
R
N
Br
N
O
O
X = Br or CN
profile with only JNK3 being inhibited at 10
lM compound
49
50
51
concentration.6
Scheme 4. Reagents and Conditions: (a) 2-(tributylstannyl)oxazole, Pd(PPh3)4,
DMF, 90 °C, 18 h; (b) Fe, AcOH, 50 °C, 1 h; (c) RCH2CO2H, POCl3, pyridine, 0 °C, 0.5 h.
On the basis of their improved in vitro stability, select com-
pounds were evaluated in in vivo pharmacokinetic studies and
the results are shown in Table 6. The brain exposures of these ana-
logs track well with the in vitro derived P-gp efflux ratios. For
example, compounds with P-gp ratios greater than 1.6 showed
brain/plasma ratios less than 0.5 and low exposure in the brain
(1, and 34) when dosed at 1 mg/kg in mice. Both compounds 41
and 42 had low P-gp efflux ratios and displayed brain/plasma ra-
tios close to 1. Compounds with good in vitro stability (42 and
34) generally had low clearance and longer half lives compared
to less stable analogs such as 41.
the degradation in stability. With the quinolinone aryl acetamide
(B) the stability of the triazole replacements improved. Within this
series, oxazole (56) and thiazol-4-yl (58) were considerably less
stable than thiazol-2-yl (57), which had stability similar to the par-
ent triazole 42.
In order for these inhibitors to be effective neuroprotective
agents, they must be able to pass through the blood–brain barrier
to reach the target and therefore, must have a high rate of perme-
ability and a low P-gp efflux. Illustrated in Table 5 are the in vitro
derived permeability and P-gp efflux ratios of the more metaboli-
cally stable analogs. Naphthyridinones generally had higher P-gp
efflux in MDR–MDCK cells compared to isoquinoline 1. Both diflu-
oroquinolinone 41 and 6-trifluoromethyl quinolinone 42 had low
P-gp efflux ratios but with lower permeability compared to the
naphthyridinone 34. Replacement of the trifluoromethyl substitu-
ent of 42 with a nitrile (47) resulted in a dramatic increase in the
P-gp efflux of the inhibitor. Consistent with previous compounds
Compound 42 displayed the most favorable combination of
high brain exposure, good brain/plasma ratio and low clearance
and the neuroprotective properties of this inhibitor were evaluated
in a human cortical neuronal amyloid-b neurotoxicity assay.6,15 In
this assay, compound 42 has an EC50 of 2.9 lM suggesting that
this series JNK inhibitors maybe viable neuroprotective agents.
In conclusion, we have prepared trisubstituted thiophene
based JNK inhibitors with improved pharmacokinetic properties
compared to earlier analogs. This was accomplished by identify-
ing the metabolites generated upon treatment of our earlier

5526
S. Bowers et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5521–5527
Table 4
JNK inhibition and in vitro metabolic stability data for compounds 52–58. SAR of the heterocycle
O
S
R3
N
H
R2
R1
O
N
O
N
F3C
N
A
B
a
a
a
Compds
R1
A
R2
R3
Br
JNK3 IC50 (nM)
JNK1 IC50 (nM)
JNK2 IC50 (nM)
OxMet% (m, r, h)b
0, 0, 11
52
6
7
66
O
S
N
53
54
A
A
Br
Br
15
3
13
1
145
0, 0, 3
N
N
N
13
33, 37, 46
N
Me
S
55
56
57
A
B
B
CN
CN
CN
11
8
10
4
62
29
53
7, 1, 16
N
N
N
4, 2, 24
O
S
9
7
52, 76, 75
58
B
CN
8
2
17
17, 8, 32
N
S
a
See Ref. 6 Values are means of at least three experiments at 10
See Ref. 6 Percent remaining after 30 min incubation with liver microsomes (m = mouse, r = rat and h = human).
l
M ATP concentration.
b
Table 5
P-gp, permeability and kinase selectivity data for select analogs
O
S
Br
NH
N
H
R
N
N
OxMet% (m, r, h)b
29, 9, 50
Papp (nm/s)
P-gp Effluxb
1.6
a
a
a
b
Compds
R
JNK3 IC50 (nM)
11
P38
>50
a
IC50
(lM)
ERK2 IC50 (lM)
1c
>50
270
N
N
O
34
41
42
47
14
5
>50
>50
>50
>50
>0.5
55, 50, 84
14, 19, 38
42, 39, 62
60, 46, 69
368
1.8
N
F
F
N
O
>50
>50
>50
83.1
36.2
34.3
0.68
0.61
4.18
N
O
5
F3C
NC
N
O
2
a
b
c
See Ref. 6 Values are means of at least three experiments at 10 lM ATP concentration.
Percent remaining after 30 minute incubation with liver microsomes (m = mouse, r = rat and h = human).
4-Chlorothiophene instead of 4-bromothiophene.
generation of inhibitors with liver microsomes and using this
information to design more metabolically stable analogs. This
led to compound 42, a brain penetrant selective JNK inhibitor
which had vastly improved brain exposure and low clearance

S. Bowers et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5521–5527
5527
Table 6
In vivo pharmacokinetic data for select analogs
Cmpds
Mouse (1 mg/kg IV)
Rat
AUC brain (ng h/mL)
Brain/plasma
IV (1 mg/kg)
Clearance (mL/h/kg)
PO(2 mg/kg)
AUC (nmol h/mL)
t1/2 (min)
F%
1
88
35
160
235
0.4
na
na
na
na
91
40
34
34
41
42
0.24
1.4
0.7
2.5
2.3
4.6
660
1351
554
2720
571
1314
Bao, J.; Banasiak, K. J.; Haddad, G. G.; Flavell, R. A.; Davis, R. J.; Rakic, P. Proc.
Natl. Acad. Sci. U.S.A. 2003, 100, 15184.
compared to earlier analogs and had demonstrated in vitro neuro-
protection in human cortical neurons. Future efforts to develop
these compounds as neuroprotective agents will be reported in
due course.
5. (a) Hom, R. K.; Bowers, S.; Sealy, J. M.; Truong, A. P.; Probst, G. D.; Neitzel, M. L.;
Neitz, R. J.; Fang, L.; Brogley, L.; Wu, J.; Konradi, A. W.; Sham, H. L.; Tóth, G.; Pan,
H.; Yao, N.; Artis, D. R.; Quinn, K.; Sauer, J.-M.; Powell, K.; Ren, Z.; Bard, F.;
Yednock, T.; Griswold-Prenner, I. Bioorg. Med. Chem. Lett. 2010, 20, 7303; (b)
Bowers, S.; Truong, A. P.; Neitz, R. J.; Neitzel, M.; Probst, G. D.; Hom, R. K.;
Peterson, B.; Galemmo, R. A.; Konradi, A. W.; Sham, H. L.; Tóth, G.; Pan, H.; Yao,
N.; Artis, D. R.; Brigham, E. F.; Quinn, K. P.; Sauer, J.-M.; Powell, K.; Ruslim, L.;
Ren, Z.; Bard, F.; Yednock, T. A.; Griswold-Prenner, I. Bioorg. Med. Chem. Lett.
2011, 21, 1838.
Acknowledgments
We would like to thank Jacek Jagodzinski, Lee Latimer, Ann Qin,
Shawn Gauby, Danny Tam, and Michael P. Bova for their contribu-
tions to this work.
6. See the Supplementary data.
7. For details on the synthesis of these analogs see: Sham, H. L.; Konradi, A. W.;
Hom, R. K.; Probst, G. D.; Bowers, S.; Truong, A.; Neitz, R. J.; Sealy, J.; Toth, G.
WO2010091310, 2010.
8. (a) Balz, G.; Schiemann, G. Ber. 1927, 60B, 1186; (b) Roe, A. Org. React. 1949, 5,
193; (c) Furuya, T.; Klein, J.; Ritter, T. Synthesis 2010, 11, 1804.
9. (a) Jørgensen, M.; Lee, S.; Liu, X.; Wolkowski, J. P.; Hartwig, J. F. J. Am. Chem. Soc.
2002, 124, 12557; (b) Hama, T.; Liu, X.; Culkin, D. A.; Hartwig, J. F. J. Am. Chem.
Soc. 2003, 125, 11176.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.06.100.
10. (a) Skraup, Z. H. Berichte 1880, 13, 2086; (b) Yamashkin, S. A.; Oreshkina, E. A.
Chem. Heterocycl. Comp. 2006, 42, 701.
References and notes
11. Sakamoto, T.; Kondo, Y.; Yamanaka, H. Chem. Pharm. Bull. 1985, 33, 4764.
12. Gavrin, L. K.; Lee, A.; Provencher, B. A.; Massefski, W. W.; Huhn, S. D.;
Ciszewski, G. M.; Cole, D. C.; McKew, J. C. J. Org. Chem. 2007, 72, 1043.
13. Nicolaou, K. C.; Estrada, A. A.; Zak, M.; Lee, S. H.; Safina, B. S. Angew. Chem., Int.
Ed. 2005, 44, 1378.
1. For recent reviews on the JNK signaling pathway, see (a) Siddiqui, M. S.; Reddy,
P. A. J. Med. Chem. 2010, 53, 3005; (b) Bogoyevitch, M. A.; Ngoei, K. R. W.; Zhao,
T. T.; Yeap, Y. Y. C.; Ng, D. C. H. Biochim. Biophys. Acta 2010, 1804, 463; (c)
Haeusgen, W.; Boehm, R.; Zhao, Y.; Herdegen, T.; Waetzig, V. Neuroscience
2009, 161, 951; (d) LoGrasso, P.; Kamenecka, T. Mini-Rev. Med. Chem. 2008, 8,
755; (e) Bogoyevitch, M. A.; Arthur, P. G. Biochim. Biophys. Acta 2008, 1784, 76;
(f) Bogoyevitch, M. A.; Kobe, B. Microbiol. Mol. Biol. Rev. 2006, 70, 1061.
2. (a) Manning, A. M.; Davis, R. J. Nat. Rev. Drug Disc. 2003, 2, 554; (b) Weston, C.
R.; Davis, R. J. Curr. Opin. Cell Biol. 2007, 19, 142.
3. For reviews on the use of JNK inhibitors as neuroprotective agents, see: (a)
Borsello, T.; Forloni, G. Curr. Pharm. Design 2007, 13, 1875; (b) Wang, W.; Ma,
C.; Mao, Z.; Li, M. Drug News Perspect. 2004, 17, 646; (c) Resnick, L.; Fennell, M.
Drug Discovery Today 2004, 9, 932; (d) Harper, S. J.; Wilkie, N. Expert Opin. Ther.
Targets 2003, 7, 187.
14. Wagner, R.; Donner, P. L.; Kempf, D. J.; Maring, C. J.; Stoll, V. S.; Ku, Y.-Y.; Pu, Y.-
M. WO2008011337, 2008.
15. (a) Wright, S.; Malinin, N. L.; Powell, K. A.; Yednock, T.; Rydel, R. E.; Griswold-
Prenner, I. Neurobiol. Aging 2007, 28, 226; (b) Wogulis, M.; Wright, S.;
Cunningham, D.; Chilcote, T.; Powell, K.; Rydel, R. E. J. Neurosci. 2005, 25,
1071; (c) Probst, G. D.; Bowers, S.; Sealy, J. M.; Truong, A. P.; Hom, R. K.;
Galemmo, R. A., Jr.; Konradi, A. W.; Sham, H. L.; Quincy, D. A.; Pan, H.; Yao, N.;
Lin, M.; Tóth, G.; Artis, D. R.; Zmolek, W.; Wong, K.; Qin, A.; Lorentzen, C.;
Nakamura, D. F.; Quinn, K. P.; Sauer, J.-M.; Powell, K.; Ruslim, L.; Wright, S.;
Chereau, D.; Ren, Z.; Anderson, J. P.; Bard, F.; Yednock, T. A.; Griswold-Prenner,
I. Bioorg. Med. Chem. Lett. 2011, 21, 315.
4. (a) Yang, D. D.; Kuan, C. Y.; Whitmarsh, A. J., et al Nature 1997, 389, 865; (b)
Kuan, C. Y.; Whitmarsh, A. J.; Yang, D. D.; Liao, G.; Schloemer, A. J.; Dong, C.;