Mixed-lineage kinase 1 and mixed-lineage kinase 3 subtype-selective dihydronaphthyl[3,4-a]pyrrolo[3,4-c]carbazole-5-ones: Optimization, mixed-lineage kinase 1 crystallography, and oral in vivo activity in 1-methyl-4-phenyltetrahydropyridine models

Article abstract of DOI:10.1021/jm8005838

The optimization of the dihydronaphthyl[3,4-a]pyrrolo[3,4-c]carbazole-5-one R² and R¹² positions led to the identification of the first MLK1 and MLK3 subtype-selective inhibitors within the MLK family. Compounds 14 (CEP-5104) and 16 (CEP-6331) displayed good potency for MLK1 and MLK3 inhibition with a greater than 30- to 100-fold selectivity for related family members MLK2 and DLK. Compounds 14 and 16 were orally active in vivo in a mouse MPTP biochemical efficacy model that was comparable to the first-generation pan-MLK inhibitor 1 (CEP-1347). The MLK1 structure-activity relationships were supported by the first-reported X-ray crystal structure of MLK1 bound with 16.

Full text of DOI:10.1021/jm8005838

5680
J. Med. Chem. 2008, 51, 5680–5689
Mixed-Lineage Kinase 1 and Mixed-Lineage Kinase 3 Subtype-Selective
Dihydronaphthyl[3,4-a]pyrrolo[3,4-c]carbazole-5-ones: Optimization, Mixed-Lineage Kinase 1
Crystallography, and Oral in Vivo Activity in 1-Methyl-4-phenyltetrahydropyridine Models
Robert L. Hudkins,*,† James L. Diebold,† Ming Tao,† Kurt A. Josef,† Chung Ho Park,† Thelma S. Angeles,† Lisa D. Aimone,†
Jean Husten,† Mark A. Ator,† Sheryl L. Meyer,† Beverly P. Holskin,† John T. Durkin,† Alexander A. Fedorov,‡
Elena V. Fedorov,‡ Steven C. Almo,‡ Joanne R. Mathiasen,† Donna Bozyczko-Coyne,† Michael S. Saporito,† Richard W. Scott,†
and John P. Mallamo†
DiscoVery Research, Cephalon, Incorporated, 145 Brandywine Parkway, West Chester, PennsylVania 19380, and
Albert Einstein College of Medicine, Bronx, New York 10461
ReceiVed May 17, 2008
The optimization of the dihydronaphthyl[3,4-a]pyrrolo[3,4-c]carbazole-5-one R2 and R12 positions led to
the identification of the first MLK1 and MLK3 subtype-selective inhibitors within the MLK family.
Compounds 14 (CEP-5104) and 16 (CEP-6331) displayed good potency for MLK1 and MLK3 inhibition
with a greater than 30- to 100-fold selectivity for related family members MLK2 and DLK. Compounds 14
and 16 were orally active in vivo in a mouse MPTP biochemical efficacy model that was comparable to the
first-generation pan-MLK inhibitor 1 (CEP-1347). The MLK1 structure-activity relationships were supported
by the first-reported X-ray crystal structure of MLK1 bound with 16.
Introduction
The mixed-lineage kinases (MLKs) are critical upstream
activating components of the stress-activated c-jun-N-terminal
kinase (JNK) pathway, and they function to phosphorylate MAP
kinase kinases MKK4 and MKK7 of the cascade. MKK4 and
MKK7 activate the JNKs (JNK1-3) that phosphorylate the
transcription factor cJun on Ser63 and Ser73, which is an early
initiating event in the cell-death process.2,3 c-Jun phosphory-
lation mediates neuronal apoptosis in response to a variety of
stimuli, and it may contribute to the cell death that is associated
with neurodegenerative pathological conditions.2 In Parkinson’s
disease (PD), compelling postmortem evidence indicates that
neuronal apoptosis may be the underlying mode of cell death
of nigrostriatal dopaminergic neurons, which is a primary cause
of the hallmark locomotor deficits in the disease (akinesia,
tremor, and postural instability).4 The characteristic features of
PD suggest that a drug that is capable of attenuating nigrostriatal
dopaminergic degeneration could theoretically slow the disease
progression by providing therapy beyond that of neurotransmitter
replacement, which is an elusive unmet medical need.
Our efforts in the design and synthesis of MLK inhibitors in
Figure 1. Structures of indolocarbazoles and dihydronaphthylcarba-
zoles.
the indolocarbazole class identified 1 (CEP-1347)5 as a first-
generation molecule that advanced through phase III clinical
trials (Figure 1). 1 promoted neuronal survival in numerous in
vitro and in vivo models5a,6 including attenuating the loss of
tyrosine hydroxylase (TH) immunoreactivity and dopamine
(DA) transporter density in mice and monkeys following the
administrationof1-methyl-4-phenyl-tetrahydropyridine(MPTP).5a,7
The chemistry objective for a subsequent follow-on, second-
generation compound was to identify a fully synthetic entity to
circumvent requirements for the fermented glycosolyated natural
product (+)K-252a (2), which has a lower molecular weight
and enhanced pharmacokinetic properties. Recently, we reported
our work that identified the MLK pharmacophore in (+)K-252a
that led to an MLK1 and MLK3 subtype-selective dihydronaph-
thyl[3,4-a]pyrrolo[3,4-c]carbazole (DHN) scaffold.8 The 7-oxo
derivative 3 demonstrated both enzyme and cellular activity with
good selectivity. However, early preclinical studies showed that
it suffered from metabolic instability and poor pharmacokinetic
properties. We therefore focused our attention on optimizing
the potency and properties of the 5-oxo regiomer 4. In this
article, we report the optimization of positions R2 and R12, the
MLK1 crystallography, the pharmacokinetic properties, and the
oral in vivo activity for the MLK1 and MLK3 family-selective
DHN analogs.
* To whom correspondence should be addressed. E-mail: rhudkins@
cephalon.com. Tel: 610-738-6283. Fax: 610-738-6558.
Chemistry
†
The 5-oxo carbazole analogs were synthesized via a
Diels-Alder reaction as outlined in Scheme 1 using our
Cephalon, Incorporated.
‡
Albert Einstein College of Medicine.
10.1021/jm8005838 CCC: $40.75
2008 American Chemical Society
Published on Web 08/21/2008

MLK1 and MLK3 Subtype-SelectiVe Inhibitors
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18 5681
Scheme 1a
or under phase-transfer conditions (NaOH, N-tetrabutylammo-
nium bromide, DCM/water) to produce 18-22 and 24-26
(Scheme 2).
Results and Discussion
In Vitro MLK1 and MLK3 Structure-Activity Relation-
ships. The MLK1 and MLK3 enzyme inhibitory data are shown
in Table 1. Analogs were screened for inhibition against
baculoviral glutathione-S-transferase (GST) fusion proteins that
expressed the active human MLK1 and MLK3 kinase domains
by using myelin basic protein as substrate.6c,8 The pan-MLK
inhibitor 1 was used as a reference standard, and it displayed
IC50 values of 38 and 64 nM for MLK1 and MLK3,
respectively.5b 5-Oxo 4 was about 30 and 11 times weaker for
MLK 1 and MLK3, respectively, compared to 1 (IC50 ) 1.1
and 0.72 µM, respectively.) A methoxy group was scanned to
probe SAR around the B ring initially. This tactic revealed that
the preferred position for substitution was 2 > 3 . 1. Compound
8 (2-methoxy) showed a 3- to 5-fold improvement in MLK1
and MLK3 potency, whereas substitution at the 1 position was
not tolerated, and the 3-OMe analog was significantly weaker
(data not shown). A 2-alkoxy group proved to be important as
the OiPr in 9 improved potency 16-fold for MLK1 and 5-fold
for MLK3 compared with 8, whereas the 2-OH in 10 showed
less than 10% inhibition at 1 µM concentration (Table 1). A
set of alkanols were next installed at N12 in accord with our
previous work that demonstrated that the chirality of the (+)K-
252a 3′-sugar alcohol was critical for trkA tyrosine kinase
activity.10 Unfortunately, the ethanol analog 13 had only weak
inhibitory activity versus MLK3 (IC50 ) 1 µM, comparable in
potency to 4), whereas analogs with increased alkyl chain length
or branching were devoid of activity (data not shown).
Conversely, our exploring combinations of R2-alkoxy with R12
ethanol produced analogs with greater-than-additive potency.
Ethanol 14 and propanol 15 displayed a 10- to 15-fold
improvement in MLK potency compared with 4. The size of
the O-alkyl ether at position 2 was critical for potency in the
order iPr > Me > nPr > cAlkyl > Et > H. An optimum balance
for MLK1 and MLK3 activity was obtained with R2 OMe (14)
(IC50 ) 103 and 55 nM for MLK1 andMLK3, respectively)
and OiPr (16) (IC50 ) 26 and 67 nM for MLK1 and MLK3,
respectively). Substituting with larger, nonether groups at R2
or probing the region with heterocycles or donor and acceptor
groups decreased but did not abolish the activity (see 18-26).
(For examples, see 18-26.) The R12 donor OH group was
critical for MLK activity because O-alkyl ethers or the reomoval
of the hydroxyl dramatically decreased MLK activity (data not
shown).
a Reagents and conditions: (a) ethyl cis-ꢀ-cyanoacrylate, YtBr3, toluene;
(b) DDQ, toluene, rt; (c) RaNi, H2, DMF/MeOH f 12a-c; (d) 10 N NaOH,
BrCH2CH2OBn, acetone, reflux; and (e) Pd(OH)2, HCl, DMF.
published methods for 3 and 4.8,9 The Diels-Alder reaction
with ethyl cis-ꢀ-cyanoacrylate and 2-(3,4-dihydronaphth-2-yl)-
1H-indole 5 in the presence of YtBr3 in toluene proceeded in a
regioselective manner to produce the 3-ethoxycarbonyl isomer
6 (Scheme 1). The aromatization of tetrahydrocarbazoles using
DDQ in toluene or acetonitrile cleanly produced the cyanoester
carbazoles 7. Reductive cyclization (RaNi, H2, DMF, MeOH)
produced lactams 8 and 9. The regiochemistry was confirmed
by NMR experiments and crystallography, as previously de-
scribed.9 A key synthetic transformation in the elaboration of
the N12 structure-activity relationships (SAR) was the clean
alkylation of the indole NH by the use of NaOH in acetone
without hydrolysis of the ester or nitrile. The alkylation of
cyanoester carbazole 7 with 2-bromoethyl benzyl ether produced
the O-benzyl-protected 11 in >90% yield. On a large scale, it
was more efficient to use a two-step hydrogenation process to
produce the target alcohol lactams 14 and 16 than it was to use
a one-pot procedure. Accordingly, the reductive cyclization of
11a and 11b by the use of RaNi/H2 in DMF/MeOH produced
the O-benzyl lactam 12. O-debenzylation using Pd(OH)2/H2 in
DMF produced the alcohol lactam in high yield and purity. We
needed an efficient scalable method to produce 2-hydroxy-12-
ethanol 17 in high yield and purity so that it could be used as
an intermediate to analog the 2-ether position. The best
conditions we found to demethylate 14 were provided by a
mixture of AlCl3 and ethane thiol in DCE. Similar conditions
were used to convert 8 to 10. The alkylation of 17 with various
alkyl halides was accomplished by the use of Cs2CO3 in CH3CN
MLK1 Cell Activity in Vitro. The potent examples 9, 14,
16, and 19 were evaluated in an MLK1 cell-based assay for the
inhibition of MKK4 phosphorylation. MLK1 was transfected
in Chinese hamster ovary (CHO) cells with dominant-negative
MKK4 cDNA, and phospho-MKK4 was measured in an ELISA-
based format, as previously described.6c,8 Compounds 14 and
16 (MLK1 enzyme IC50 ) 26 and 103 nM, respectively)
displayed MLK1 cell activity that was relatable to the isolated
enzyme activity with cell IC50 values of 127 and 33 nM,
respectively (Table 1). Compounds 9 and 19 displayed ap-
proximately 3- and 7-fold shifts from the cell free-MLK1
enzyme activity. The cell translation of 14 and 16 compared
favorably to the optimized semisynthetic clinical compound 1
(MLK1 enzyme IC50 ) 38 nM; MLK1 cell IC50 ) 72 nM).5b,6c
MLK1/16 X-Ray Crystal Structure. To identify the im-
portant interactions and to assist in the design of new inhibitors,

5682 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18
Hudkins et al.
Scheme 2a
a Reagents and conditions: (a) (i) AlCl3, EtSH, DCE, 0f50 °C, (ii) 1N HCl; (b) NaOH (1.5 equiv), N-tetrabutylammonium bromide, R2-Br, CH2Cl2
/H2O; and (c) R2-Br or R2-Cl,Ce2CO3, CH3CN, 80 °C.
Table 1. MLK1 and MLK3 Activity of Dihydronaphthylcarbazole
Analogs
Table 2. X-ray Crystallography Data Collection and Refinement
Statisticsa
Data Collection
space group
P21212
cell dimensions
data redundancy
unique reflns
A ) 56.93, b ) 126.30, c ) 41.36 Å
5.1
9475
resolution (Å)a
completeness (%)a
Rmerge (%)a
2.0-2.6 (2.69-2.60)
96.8 (92.5)
7.9 (31.9)
19.3 (3.5)
a
average I/σ
entry
R2
R12
MLK1a MLK3a MLK1 cellsb
Refinement
1
4
8
9
38
64
72
resolution (Å)
2.0-2.6
no. reflns
8673
H
H
H
H
H
1106
188
66
724
274
159
0%
1000
55
OMe
OiPr
OH
H
OMe
OMe
OiPr
OH
working set
test set
Rwork/Rfree (%)
196
683
10
13
14
15
16
17
18
19
20
21
22
23
24
25
26
9%
21.8/28.2
no. atoms
1913
43
32
5
33.2
rms deviations
0.007
(CH2)2OH
(CH2)2OH
(CH2)3OH
(CH2)2OH
(CH2)2OH
(CH2)2OH
(CH2)2OH
(CH2)2OH
(CH2)2OH
14%
103
101
26
32%
30%
49
35%
36%
52%
9%
127
33
protein
solvent
inhibitor
87
65
30%
32%
32%
4%
30%
40%
0%
46%
8%
46%
SO4
average B factor (Å2)
OCH2cPr
OcPentyl
OcHexyl
OCH2(2-Pyr)
OCH2CH2OCH3 (CH2)2OH
OCH2CH2OBn
OCH2CH2OH
OCH2CN
345
bond lengths (Å)
bond angles (deg)
a Numbers in parentheses indicate values for the highest-resolution shell.
1.42
(CH2)2OH
(CH2)2OH
(CH2)2OH
(CH2)2OH
hinge in a manner that is analogous to that of 111 and to the
conformation of (+)K-252a (2) bound to the tyrosine kinase
c-Met12 (Figure 2). 16 anchors at the hinge via two key hydrogen
bonds (1.8 Å); the lactam N-H forms a strong hydrogen bond
with the carbonyl backbone of Gly221 and the lactam CdO
accepts a hydrogen bond with the backbone amide of Ala223
(Figure 3). The MLK1 ATP pocket is formed by hydrophobic
residues Ile150, Phe155, Val158, Ala169, Phe195, and Ile204
from the N-terminal lobe; Leu275 and Phe295 from the
C-terminal lobe; and Met220, Phe222, and Ala223 from the
hinge region. Moreover, the apolar atoms of Gly151, Gly153,
Gly226, Thr293, and Asp294 also participate in the formation
of the pocket. This DFG-in structure had no density in the loop
between strand ꢀ and helix RC (chain segment 174-180) in
the C-terminal domain and in part of the activation loop
(294-319) chain segment 296-309. The binding conformation
of 16 positions the 2-O-isopropyl moiety at the solvent channel
where the ether oxygen forms a hydrogen bond with the Arg230
side chain (∼2.5 Å) (Figure 3). An additional hydrophobic
interaction between the Phe221 side chain and the iPr may
account for the increased potency of 16 over 14. The N12 ethanol
OH serves as an H-bond donor with the carbonyl backbone of
Ser272 (2.1 Å) and is also in close proximity for electrostatic
contacts to the polar side-chain residue of Asn273 (∼3.4 Å)
and potentially Thr293 (∼4 Å). The very flexible glycine-rich
NT
30%
58%
O(CH2)3CN
a IC50 values are reported in nM as the average of duplicates agreeing
within 20%; % inhibition at 1 µM. NT ) not tested. b MLK1 cell inhibitions
are IC50 values reported in nM, as described previously (refs 5c and 7).
we solved a cocrystal structure of 16 with the unphosphorylated
MLK1 kinase domain (MLK1[T312A](136-406)) at 2.6 Å
resolution (Table 2). This represents the first reported crystal
structure that was solved in the MLK family. The topology of
the MLK1 kinase domain in complex with 16 is similar to other
protein kinases wherein the structure consists of a smaller
N-terminal lobe, which is largely composed of ꢀ sheets, and a
C-terminal lobe, which is predominantly R-helical. The N-
terminal domain comprises one R helix, RC (181-197), and a
six-stranded ꢀ sheet, ꢀ0 (137-139), ꢀ1 (144-150), ꢀ2
(158-163), ꢀ3 (166-174), ꢀ4 (206-211), and ꢀ5 (215-220)
(Figure 2). The six-stranded ꢀ sheet is also found in GSK-3ꢀ
and p38, in contrast with the five-stranded ꢀ sheet that is found
in protein kinases such as FGFR, cMet, Abl, VEGFR2, CDK2,
and EGFR. The additional strand ꢀ0 makes antiparallel ꢀ-sheet
hydrogen bonding interactions with ꢀ5. The C-terminal lobe
comprises two ꢀ strands and seven R helices. The two lobes
are connected by a single-polypeptide-strand hinge region
(220-224). Compound 16 is bound in the ATP pocket at the

MLK1 and MLK3 Subtype-SelectiVe Inhibitors
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18 5683
Figure 2. Ribbon diagram of the MLK1-16 X-ray crystal structure.
Table 3. Kinase Selectivity Profile for 14 and 16
R3 of 209, 43, and 39 nM, respectively. When tested against a
broader panel of 254 kinases at 1 µM concentration, 14 inhibited
only 4 kinases (1.6%), whereas 16 inhibited 10 kinases (3.9%)
with greater than 90% inhibition.
kinasea
14
16
kinasea
VEGF-R3 >10 000 2459 ( 569
14
16
MLK1
MLK2
MLK3
DLK
TrkA
JNK
MKK4/7 >3000
PKC >10 000
VEGF-R1 >10 000
103 ( 20
26 ( 12
1582 ( 46 1855 ( 87 PDGF-Rꢀ >10 000 >10 000
55 ( 17
>3000
1270 ( 203 1055 ( 224 IRK
>30 000
65 ( 9
>3000
FGF-R1
P38R
>10 000 >10 000
>10 000 >10 000
>10 000 >10 000
>10 000 >10 000
>10 000 >10 000
Rat Pharmacokinetic Properties for 14 and 16. On the
basis of the MLK1 and MLK3 enzyme and cellular potencies,
14 and 16 were evaluated in the rat for pharmacokinetic
properties (Table 4). In a rat liver S9 preparation, both
compounds showed ∼44% of parents remaining after 2 h of
incubation in the presence of NADPH. The oral bioavailability
for 16 was estimated to be 37% by determining the AUCs
following iv (1 mg/kg) and po (5 mg/kg) administration over a
6 h period. The iv terminal half life was 0.8 h with a volume of
distribution of 2.0 L/kg and a clearance rate of 28 mL/min/kg.
The oral Cmax was 339 ng/mL (0.8 µM), and the brain-to-plasma
ratio was approximately 0.5. Compound 14 also showed good
oral bioavailability after iv (2 mg/kg) and po (4 mg/kg)
>30 000
>3000
>6000
>10 000
EGF-R
CDKs
CHK1/2 >10 000 >10 000
Lck
6400
3400
>10 000
VEGF-R2 4715 ( 603 3378 ( 585 Fyn
1190
a IC50 values in nM.
nucleotide-binding loop 151-158 is closer to the active site in
MLK1 compared with other kinases. A unique side-chain shift
with Phe155 from the phosphorylation loop forms a hydrophobic
interaction with the N12 ethyl side chain and also functions to
cap the cleft and block solvent accessibility to the inhibitor.
Kinase Selectivity Profile for 14 and 16. The first-generation
clinical compound 1 was a pan-MLK inhibitor with IC50 values
of 38, 64, 23 and 114 nM for MLK1, MLK2, MLK3, and DLK,
respectively.5b,6c,8 Compound 14 had IC50 values of 103 and
55 nM for MLK1 and MLK3, respectively, 1.6 µM for MLK2,
and >3 µM for DLK. Compound 16 had IC50 values of 26 and
65 nM for MLK1 and MLK3, respectively, 1.9 µM for MLK2,
and >3 µM for DLK. Neither compound inhibited MLK6 or
MLK7 (IC50 > 30 µM), which indicates a >30- to 100-fold
selectivity for MLK1 or MLK3 (Table 3). Moreover, weak
inhibition was demonstrated for other kinases within the stress-
activated pathway, such as MKK4, MKK7, or JNK, and for
serine/threonine and tyrosine kinases that are inhibited by (+)K-
administration measured over a 6 h period (F ) 57%; Cmax
572 ng/mL) (Table 4).
)
In Vitro and in Vivo Activity in MPTP Models. MPTP is
a potent and selective nigrostriatal DA neurotoxin that produces
PD-like symptoms in humans, nonhuman primates, and mice.14
MPP+, the neurotoxic metabolite of MPTP, is selectively taken
up in DA neurons and inhibits complex I of the mitochondrial
electron transport.15 The subsequent activation of the JNK
signaling pathway leads to apoptotic neuronal death in various
cell-culture models of neurodegeneration.1,2 The MLKs function
upstream of JNK to phosphorylate the dual-specificity MAPK
kinases MKK4 and MKK7, which activate the MAP kinase
JNK1 (see Figure 5). Correspondingly, MPP+ addition to SH-
SY5Y cells results in the activation of the JNK pathway16 and
apoptotic cell death including nuclear chromatin condensation,
membrane blebbing, and DNA laddering.17 Furthermore, MPTP
administration to mice activates the JNK pathway in brain7b
and elicits an increase in neurons with double-strand DNA
breaks and chromatin clumping, which are additional markers
252a or 1, such as protein kinase C5c (PKC) (mixture of Ca2+
-
dependent isozymes R, ꢀ, and δ), vascular endothelial growth
factor receptor-2 (VEGF-R2),10b and trkA5c,10b,13 (Table 3). 16
markedly better demonstrated FGF-R1 (>10 µM vs 210 nM)
and VEGF-R1-3 selectivity compared with clinical compound
1, which had IC50 values for VEGF-1, VEGF-R2, and VEGF-

5684 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18
Hudkins et al.
Figure 3. MLK1-16 important interactions.
Table 4. Pharmacokinetic Properties in Rat
compd
14a
16b
0.8
iv
t1/2 (h)
0.6
AUC (ng ·h/mL)
1731
0.9
19
579
2.0
28
Vd (L/kg)
CL (mL/min/kg)
po
AUC (ng ·h/mL)
Cmax (ng/mL)
t1/2 (h)
1988
572
1.4
1073
339
1.1
F (%)
57
37
a Administered at 2 and 4 mg/kg for iv and po, respectively. b Administered
at 1 and 5 mg/kg for iv and po, respectivley.
Figure 5. Activity of 14 and 16 in the MPTP-mediated increases in
pMKK4 levels in mice.
induced loss of nigrostriatal dopaminergic neurons in mice7a
and primates.5a 1 is a pan-MLK inhibitor and therefore cannot
elucidate which MLKs are involved in the MKK4/JNK/c-Jun
cell-death mechanism that is induced by MPTP. Previously, the
pharmacological tools were not available to address these
questions. The MLK1/3-selective inhibitors 14 and 16 were
evaluated in vitro and in vivo in MPTP models of neurotoxicity.
In vitro, 16 rescued human SH-SY5Y cells in a concentration-
dependent manner following the treatment with MPP+ with an
IC50 of 43 nM.16a In this assay, 16 was less potent than but
equally as efficacious as 1 (Figure 4). 14 displayed an IC50 of
32 nM. Whereas 1 showed a pronounced inverted U-shaped
dose response, this effect was not seen with 14 and was only
moderately seen with 16 up to 1 µM concentration, which may
be a reflection of the high kinase-selectivity profile.
Figure 4. Activity of 1 and 16 in MPP+-induced SH-SY5Y cell death.
of apoptosis.18 Preclinically, 1 showed a broad neuroprotective
and survival-promoting profile in vitro and in vivo.5-7 In MPTP-
induced models of PD, 1 attenuated MPP+-mediated SH-SY5Y
cell death in vitro16a and blocked the MPTP-mediated activation
of JNK and MKK4 in a biochemical efficacy model in vivo.7b
In neuroprotective experiments in vivo, 1 blocked the MPTP-
Previously, we reported that MPTP administration to mice
increased levels of phosphoMKK4 (pMKK4) and phosphoJNK
(pJNK) in the striatum and substantia nigra.7b Blocking the

MLK1 and MLK3 Subtype-SelectiVe Inhibitors
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18 5685
and was treated with trimethyl borate (38.0 mL, 0.338 mol). The
resulting mixture was stirred for 1 h at -65 °C. To the reaction
was added Na2CO3 solution (338 mL of 2 M solution, 0.676 mol),
dichlorobis(triphenylphosphine)-palladium(II) (1.58 g, 2.25 mmol),
and 3-bromo-7-methoxy-1,2-dihydronaphthalene (55.55 g, 0.23
mol). The mixture was mechanically stirred at reflux for 18 h and
was then cooled to ambient temperature (rt), and the solvent was
removed at reduced pressure. The residue was dissolved in EtOAc
(1.5 L) and was washed with 2 N HCl (500 mL) and water. After
drying over MgSO4, the EtOAc layer was evaporated to produce a
tan solid. This solid was stirred in DCM/MeOH (1:1, 500 mL) for
0.5 h. The DCM was distilled at rt, and the product was collected
to produce a tan solid (58.6 g, 95%). 1H NMR (DMSO-d6, δ): 11.23
(s, 1H), 7.44-7.41 (d, 1H, J ) 8 Hz), 7.30-7.27 (d, 1H, J ) 8
Hz), 7.05-7.00 (t, 3H, J ) 7.1 Hz), 6.93-6.90 (t, 1H, J ) 7 Hz),
6.73-6.69 (t, 2H, J ) 7 Hz), 6.56 (s, 1H), 3.80 (s, 3H), 2.84-2.79
(t, 2H, J ) 8 Hz), 2.64-2.59 (t, 2H, J ) 7.7 Hz). LCMS m/z: 276
(M + 1).
2-(6-Isopropoxy-3,4-dihydronaphth-2-yl)-1H-indole (5c). This
compound was prepared from indole (22.82 g, 0.16 mol) and
3-bromo-7-isopropoxy-1,2-dihydronaphthalene (34.0 g, 0.13 mol),
as described for 2-(6-methoxy-3,4-dihydronaphth-2-yl)-1H-indole.
The product was isolated as a tan solid (33 g, 81% yield). 1H NMR
(DMSO-d6, δ): 11.22 (s, 1H), 7.44-7.41 (d, 1H, J ) 8 Hz),
7.30-7.27 (d, 1H, J ) 8 Hz), 7.05-7.00 (t, 3H, J ) 7 Hz),
6.93-6.90 (t, 1H, J ) 7 Hz), 6.73-6.69 (t, 1H, J ) 7 Hz), 6.56 (s,
1H), 3.8 (m, 1H), 2.84-2.81 (t, 2H, J ) 8 Hz), 2.64-2.59 (t, 2H,
J ) 7.8 Hz), 1.23-1.21 (d, 6H, J ) 6 Hz). LCMS m/z: 304
(M + 1).
4-Cyano-3-ethoxycarbonyl-1,2,3,4-tetrahydro-(6-methoxy-1,2-
dihydronaphthyl)[3,4-a]-9H-carbazole (6b). 5b (200 g, 0.726
mol), ytterbium bromide (30 g, 0.072 mol), and ethyl cis-ꢀ-ethyl-
cyanoacrylate (272 g, 2.17 mol) in toluene (4 L) were heated at
reflux for 1 h. The product that precipitated when the mixture was
allowed to cool to rt was collected by filtration and was washed
with toluene (400 mL) and ether (2 × 300 mL). The solid was
washed thoroughly with water (3 L) and was dried to produce 165.6
g (57%) of a white solid that was used directly in the next step.
LCMS m/z: 401 (M + 1).
4-Cyano-3-ethoxycarbonyl-(6-methoxy-1,2-dihydronaphth-
yl)[3,4-a]-9H-carbazole (7b). To 6b (60 g, 0.150 mol) suspended
in toluene (1.5 L) was added DDQ (71.4 g, 0.32 mol) at rt with
vigorous stirring. The temperature of the reaction mixture gradually
raised to 33 °C over 1 h before returning to rt. The solid was
collected by filtration, washed thoroughly with toluene, and dried
in air. It was then dispersed in 2 L of water with vigorous stirring,
and 80 g of solid sodium bicarbonate was added portion-wise. After
it was stirred for 3 h, the mixture was filtered, and the solid was
washed thoroughly with water until the washings were a neutral
pH. The solid was dried to produce 59 g (99%). mp >250 °C. 1H
NMR (DMSO-d6, δ): 12.1 (s, 1H), 8.45 (d, 1H), 7.60 (m, 2H),
7.25-7.4 (m, 2H), 7.0 (s, 1H), 6.9 (s, 1H), 4.35 (q, 2H), 3.8 (s,
3H), 3.15 (m, 2H), 2.9 (m, 2H), 1.25 (t, 3H). LCMS m/z: 397
(M + 1).
MPTP-mediated increase in pMKK4, the immediate downstream
target of the MLKs was used as a biochemical efficacy measure
of peripherally administrated MLK inhibitors in the brain.
Compounds 14 and 16 were administered orally at 6.7 mg/kg
2 h pre-MPTP and 2 h post-MPTP (40 mg/kg, sc). The
substantia nigra from treated and control groups were dissected
2 h later, and levels of pMKK4 were measured by immunoblot.7
14 and 16 inhibited pMKK4 activation by 68 and 64%,
respectively (Figure 5). The plasma levels 2 h postdose (4 h
after MPTP) were 0.25 and 0.58 µM for 14 and 16, respectively.
For comparison, the administration of 1 at 1.0 and 10 mg/kg sc
inhibited pMKK4 activation by 45-50%.5a,7b Similar to 1,
compounds 14 and 16 did not inhibit the MPP+ formation via
MAO-B or the extent or rate of uptake by inhibition of the DA
transporter (unpublished results).
Two different models of MPTP-induced neurotoxicity can
be produced by differential dosing of MPTP. The administration
of a low dose (20 mg/kg sc) of MPTP produces an ap-
proximately 50% loss of striatal DA terminal markers, as
measured by the specific DA uptake site ligand GBR-12935
binding, without affecting DA cell bodies in the substantia nigra,
as measured by TH immunoreactivity.5a,7a Alternatively, the
administration of a higher dose (40 mg/kg sc) results in a
complete loss of striatal DA markers and about a 50% loss of
substantia nigra cell bodies. Compounds 14 and 16 were
evaluated for neuroprotection under both toxin-dosing condi-
tions. The protection afforded by compound 16 was limited to
striatal DA terminals after b.i.d. oral dosing in the low-dose
MPTP model, whereas compound 14 protected both striatal DA
terminals and cell bodies at oral doses that were as low as 2
mg/kg q.d. in the low-dose model.
Conclusions
The optimization of the R2 and R12 positions of the DHN
scaffold showed that combining an R2-methoxy or an R2-
isopropoxy with an R12 ethanol produced analogs with greater-
than-additive potency and identified 14 and 16 as potent MLK1
and MLK3 subtype-selective inhibitors. 14 and 16 demonstrated
oral in vivo activity in the mouse MPTP-activated MKK4
biochemical efficacy model, which was comparable to 1. In
MPTP neuroprotection models, 14 was superior to 16 and is
the subject of a pending detailed full pharmacology publication.
The orally bioavailable MLK1/3 subtype inhibitors 14 and 16
are valuable tools that are helpful in delineating the critical role
of the MLK kinases in the JNK pathway and in neuroprotection
in various cell types and neuronal populations.
Experimental Section
Chemistry. All reagents and anhydrous solvents were obtained
from commercial sources and were used as received. 1H NMR
specta were obtained on a Bruker 400 MHz instrument in the solvent
indicated with tetramethylsilane as an internal standard. We ran
preparative chromatography using silica gel GF 20 × 20 cm2 1000
µm plates (Analtech). Coupling constants (J) are in Hertz (Hz).
We ran analytical HPLC by using a Zorbax RX-C8 5 × 150 mm2
column eluting with a mixture of acetonitrile and water containing
0.1% trifluoroacetic acid with a gradient of 10-100%.
2-(6-Methoxy-3,4-dihydronaphth-2-yl)-1H-indole (5b). To a
3 L three-necked round-bottomed flask containing indole (40.0 g,
0.4 mol) in dry THF (1.1 L) under a nitrogen atmosphere at -65
°C was added n-BuLi (1.6 M in hexanes, 225.3 mL). After the
mixture was stirred for 1 h, CO2(g) was passed through via a syringe
for 10 min. The solution was then concentrated to half-volume at
reduced pressure to remove excess CO2. The THF (400 mL) was
replaced, and t-BuLi (235 mL of 1.7 M solution in hexanes) was
then added dropwise at -65 °C. The solution was stirred for 1 h
2-Methoxy-13,14-dihydronaphthol[2,1-a]pyrrolo[3,4-c]carba-
zole-5-one (8). 7b (2.8 g, 7.1 mmol) was added to Raney nickel
catalyst (ca. 2 g, wet form) in DMF/MeOH (100 mL, 3:1) and was
hydrogenated at 50 psi on a Parr apparatus for 18 h. The solution
was diluted with DMF (50 mL), filtered through celite, and
concentrated at reduced pressure to produce 2.1 g (85%). The
product was crystallized from MeOH-ether and was dried to
produce a white solid. mp >250 °C. 1H NMR (DMSO-d6, δ): 11.6
(s, 1H), 8.4 (s, 1H), 7.95 (d, 1H), 7.85 (d, 1H, J ) 7.8 Hz), 7.6 (d,
1H, J ) 8.2 Hz), 7.45 (t, 1H, J ) 7.8 Hz), 7.3 (t, 1H, J ) 7.8 Hz),
6.9 (s, 1H), 6.8 (d, 1H, J ) 6.6 Hz), 4.8 (s, 2H), 3.8 (s, 3H), 3.3
(t, 2H, J ) 7.2 Hz), 2.87 (t, 2H, J ) 5.6 Hz). LCMS m/z: 355 (M
+ 1). Anal. Calcd for C23H18N2O2 (C, H, N).
4-Cyano-3-ethoxycarbonyl-1,2,3,4-tetrahydro-(6-isopropoxy-
1,2-dihydronaphthyl)-[3,4-a]-9H-carbazole (6c). This compound
was synthesized using the method for 6b by using 5c (30 g, 0.1
mol), ytterbium bromide (4 g, 0.01 mol), and ethyl cis-ꢀ-ethyl

5686 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18
Hudkins et al.
cyanoacrylate (37 g, 0.3 mol) in toluene (1 L) to produce 21.5 g
(50%) of a white solid that was directly used in the next step. LCMS
m/z: 429 (M + 1).
2-Isopropoxy-12-(2-benzyloxyethyl)-13,14-dihydronaphthol[2,1-
a]pyrrolo[3,4-c]carbazole-5-one (12c). This compound was syn-
thesized using the method for 12b starting with 11c (47.8 g, 82
mmol) to produce 41 g (97%) of a solid that contained <4% 16.
1H NMR (DMSO-d6, δ): 8.4 (s, 1H), 7.95 (d, 1H), 7.85 (d, 1H),
7.7 (d, 1H), 7.50 (m, 1H), 7.4 (m, 1H), 7.15 (bq, 3H), 7.05 (m,
2H), 6.85 (s, 1H), 6.80 (d, 1H), 4.8 (s, m, 4H), 4.75 (m, 1H), 4.3
(s, 2H), 3.75 (m, 2H), 3.3 (m, 2H), 2.87 (t, 2H), 1.3 (d, 6H). LCMS
m/z: 517 (M + 1). This material was directly used in the next step
without further purification.
2-Isopropoxy-12-(2-hydroxyethyl)-13,14-dihydronaphthol[2,1-
a]pyrrolo[3,4-c]carbazole-5-one (16). This compound was syn-
thesized using the method for 14 starting with 12c (41 g, 79 mmol)
to produce 33 g (98%) as an off-white solid. mp >250 °C. 1H NMR
(DMSO-d6, δ): 8.4 (s, 1H), 7.95 (d, 1H), 7.85 (d, 1H), 7.7 (d, 1H),
7.50 (t, 1H), 7.3 (t, 1H), 6.9 (s, 1H), 6.8 (d, 1H), 5.0 (b, 1H), 4.8
(s, 2H), 4.65 (m, 3H), 3.8 (bs, 2H), 3.3 (bt, 2H), 2.75 (bt, 2H),
1.35 (d, 6H). LCMS m/z: 426 (M + 1). Anal. Calcd for C27H26N2O3
(C, H, N).
2-Methoxy-12-(3-hydroxypropyl)-13,14-dihydronaphthol[2,1-
a]pyrrolo[3,4-c]carbazole-5-one (15). This compound was syn-
thesized using 3-bromopropylbenzyl ether and the general method
for 14 to produce 15 as an off-white solid. mp >250 °C. 1H NMR
(DMSO-d6, δ): 8.4 (s, 1H), 7.95 (d, 1H), 7.85 (d, 1H), 7.7 (d, 1H),
7.50 (t, 1H), 7.3 (t, 1H), 6.9 (s, 1H), 6.8 (d, 1H), 5.0 (bs, 1H), 4.8
(s, 2H), 4.65 (bt, 2H), 3.8 (m, 5H), 3.3 (m, 2H), 2.75 (m, 2H), 1.8
(m, 2H). LCMS m/z: 413 (M + 1).
2-Hydroxy-12-(2-hydroxyethyl)-13,14-dihydronaphthol[2,1-
a]pyrrolo[3,4-c]carbazole-5-one (17). To a stirred mixture of AlCl3
(800 mg, 6 mmol) in DCE (8 mL) at 0 °C was added EtSH (1.40
mL), followed by 14 (398 mg, 1 mmol). After stirring for 48 h at
50 °C, 1 N HCl (5 mL) was added and was stirred at rt for 0.5 h.
The product was collected, washed with water, and dried to produce
240 mg (63%) of a tan solid. 1H NMR (DMSO-d6, δ): 9.4 (s, 1H),
8.4 (s, 1H), 7.92 (d, 1H), 7.80 (d, 1H), 7.7 (d, 1H), 7.5 (t, 1H), 7.3
(t, 1H), 6.7 (s, 1H), 6.6 (m, 1H), 5.0 (t, 1H), 4.80 (s, 2H), 4.65 (m,
2H), 3.80 (m, 2H), 3.30 (m, 2H), 2.50 (m, 2H). LCMS m/z: 385
(M + 1).
4-Cyano-3-ethoxycarbonyl-(6-isopropoxy-1,2-dihydronaph-
thyl)[3,4-a]-9H-carbazole (7c). This compound was synthesized
by the method for 7b starting with 6c (38.4 g, 89.5 mmol) and
DDQ (40.6 g, 180 mmol) in CH3CN (1.15 L). The solid was
collected, washed thoroughly with CH3CN and water, and dried to
produce 37 g (97%) of an off-white solid. mp >250 °C. 1H NMR
(DMSO-d6, δ): 12.1 (s, 1H), 8.5 (d, 1H), 7.60 (m, 2H), 7.25-7.4
(m, 2H), 7.0 (s, 1H), 6.9 (s, 1H), 4.35 (q, 2H), 3.8 (m, 1H), 3.2 (m,
2H), 2.8 (m, 2H), 1.2 (m, 9H). LCMS m/z: 425 (M + 1).
2-Isopropoxy-13,14-dihydronaphthol[2,1-a]pyrrolo[3,4-c]car-
bazole-5-one (9). This compound was synthesized using the method
for 8 to produce an off-white solid. mp >250 °C. 1H NMR (DMSO-
d6, δ): 11.6 (s, 1H), 8.4 (s, 1H), 7.95 (d, 1H), 7.85 (d, 1H, J ) 7.8
Hz), 7.6 (d, 1H, J ) 8.2 Hz), 7.45 (t, 1H, J ) 7.8 Hz), 7.3 (t, 1H,
J ) 7.8 Hz), 6.9 (s, 1H), 6.8 (d, 1H, J ) 6.6 Hz), 4.8 (s, 2H), 3.7
(m, 1H), 3.3 (t, 2H, J ) 7.2 Hz), 2.87 (t, 2H, J ) 5.6 Hz), 1.2 (d,
6H, J ) 7.2 Hz). LCMS m/z: 382 (M + 1). Anal. Calcd for
C25H22N2O2 (C, H, N).
4-Cyano-3-ethoxycarbonyl-(6-methoxy-1,2-dihydronaphth-
yl)[3,4-a]-9-2-benzyloxyethyl-carbazole (11b). A mixture of 7b
(50 g, 0.126 mol), 2-bromoethylbenzyl ether (120 g, 0.56 mol),
and 10 N NaOH (250 mL) in acetone (1450 mL) was heated at
reflux for 14 h. The acetone was removed at reduced pressure, a
water/hexane (500 and 1250 mL, respectively) biphase was added,
and the mixture was vigorously stirred for 0.5 h. The resulting solid
was collected and was thoroughly washed with water until the
washing was neutral. The solid was dried under vacuum and was
then washed with hexane to produce 60 g (90%) of a solid. 1H
NMR (DMSO-d6, δ): 8.55 (d, 1H), 7.85 (d, 1H), 7.15 (t, 1H), 7.4
(t, 1H), 7.25 (d, 1H), 7.15 (m, 3H), 7.0 (s, 1H), 6.9 (m, 3H), 4.9
(bs, 2H), 4.30 (m, 4H), 3.8 (m, 5H), 3.45 (t, 2H), 2.75 (t, 2H), 1.2
(t, 3H). LCMS m/z: 531 (M + 1).
2-Methoxy-12-(2-benzyloxyethyl)-13,14-dihydronaphthol[2,1-
a]pyrrolo[3,4-c]carbazole-5-one (12b). A solution of 11b (59.8
g, 0.113 mol) in DMF/MeOH (1.1 L, 10:1) was hydrogenated over
Raney nickel (100 g) at 55 psi on a Parr apparatus for 2 days. The
catalyst was removed by filtration, the filtrate was concentrated
under reduced pressure, and the resulting semisolid was triturated
with ether (1.8 L) overnight to produce 53.6 g (93%) of a white
2-Hydroxy-13,14-dihydronaphthol[2,1-a]pyrrolo[3,4-c]carba-
zole-5-one (10). This compound was synthesized using the method
1
for 17 to produce a tan solid. mp >250 °C. H NMR (DMSO-d6,
δ): 11.6 (s, 1H), 9.45 (s, 1H), 8.3 (s, 1H), 8.06 (d, 1H, J ) 5.5
Hz), 7.95 (d, 1H, J ) 5 Hz), 7.55 (d, 1H, J ) 5 Hz), 7.43 (t, 1H,
J ) 7 Hz), 7.23 (t, 1H, J ) 7 Hz), 6.70 (s, 1H), 6.64 (d, 1H, J )
6 Hz), 4.8 (s, 2H), 3.3 (t, 2H, J ) 7.2 Hz), 2.97 (t, 2H, J ) 5 Hz).
LCMS m/z: 340 (M - 1).
1
solid that contained <3% of 14. HNMR (DMSO-d6, δ): 8.45 (s,
1H), 7.95 (d, 1H), 7.85 (d, 1H), 7.7 (d, 1H), 7.50 (t, 1H), 7.3 (t,
1H), 7.15 (bq, 3H), 7.05 (bq, 2H), 6.85 (s, 1H), 6.80 (d, 1H), 4.8
(s, 2H), 3.8 (s, 3H), 3.3 (t, 2H), 2.87 (t, 2H). LCMS m/z: 489 (M
+ 1). This material was directly used in the next step without further
purification.
General Method for 18-20. To a mixture of 17 (38.4 mg, 0.1
mmol), NaOH (6.0 mg, 1.5 equiv), and N-tetrabutylammonium
bromide (3.2 mg, 0.1 equiv) in 0.5 mL of CH2Cl2 and 0.5 mL of
water was added the appropriate alkyl bromide under a N2
atmosphere. The mixture was stirred at rt for 14-72 h and was
concentrated, and the residue was washed with water and was dried
over magnesium sulfate. Purification by preparative TLC with
CH2Cl2/MeOH afforded the desired compound as solids.
2-[2-Cyclopropylmethoxy]-12-(2-hydroxyethyl)-13,14-tetrahy-
dro-naphthol[2,1-a]pyrrolo[3,4-c]carbazole-5(6H)-one (18). 1H
NMR (DMSO-d6, δ): 8.45 (s, 1H), 7.96 (d, 1H), 7.85 (d, 1H), 7.67
(d, 1H), 7.50 (t, 1H), 7.29 (t, 1H), 6.90 (s, 1H), 6.76 (d, 1H), 5.05
(m, 1H), 4.85 (s, 1H), 4.65 (m, 3H), 3.89 (m, 4H), 3.3 (m, 2H),
2.77 (m, 2H), 1.30 (m, 1H), 0.65 (m, 2H), 0.48 (m, 2H). LCMS
m/z: 439 (M + 1).
2-[2-Cyclopentyloxy]-12-(2-hydroxyethyl)-13,14-tetrahydro-
naphthol[2,1-a]pyrrolo[3,4-c]carbazole-5(6H)-one (19). 1H NMR
(DMSO-d6, δ): 8.41 (s, 1H), 7.96 (d, 1H), 7.85 (d, 1H), 7.67 (d,
1H), 7.50 (t, 1H), 7.29 (t, 1H), 6.86 (s, 1H), 6.75 (d, 1H), 5.03 (m,
1H), 4.90 (m, 1H), 4.79 (s, 2H), 4.64 (m, 2H), 3.84 (m, 3H), 2.77
(m, 2H), 1.61-2.0 (m, 8H). LCMS m/z: 453 (M + 1).
2-Methoxy-12-(2-hydroxyethyl)-13,14-dihydronaphthol[2,1-
a]pyrrolo[3,4-c]carbazole-5-one (14). A solution of 12b (43.6 g,
89.2 mmol) in DMF (1 L) containing 5 drops of 12 M hydrochloric
acid was hydrogenated over palladium hydroxide (2.2 g) at 50 psi
on a Parr apparatus. The catalyst was removed by filtration through
a bed of celite, and the filtrate was concentrated under reduced
pressure. The resulting semisolid was triturated with ether (2 L)
overnight to produce 14 as an off-white solid (36 g, 100%). mp
>250 °C. 1H NMR (DMSO-d6, δ): 8.4 (s, 1H), 7.95 (d, 1H), 7.85
(d, 1H), 7.7 (d, 1H), 7.50 (t, 1H), 7.3 (t, 1H), 6.9 (s, 1H), 6.8 (d,
1H), 5.0 (bs, 1H), 4.8 (s, 2H), 4.65 (bt, 2H), 3.8 (bs, 5H), 3.3 (bt,
2H), 2.75 (bt, 2H). LCMS m/z: 399 (M + 1). Anal. Calcd for
C25H22N2O3 (C, H, N).
4-Cyano-3-ethoxycarbonyl-(6-isopropoxy-1,2-dihydronaph-
thyl)[3,4-a]-9-2-benzyloxyethyl-carbazole (11c). This compound
was synthesized using the method for 11b starting with 7c (35.9 g,
85 mmol), 2-bromoethylbenzyl ether (82 g, 380 mmol), and 10 N
NaOH (170 mL) in acetone (1.9 L) to produce 47.2 g (97%) of an
off-white solid. 1H NMR (DMSO-d6, δ): 8.5 (d, 1H), 7.85 (d, 1H),
7.55 (t, 1H), 7.4 (t, 1H), 7.3 (d, 1H), 7.2 (m, 3H), 6.95 (s, 1H), 6.9
(m, 3H), 4.9 (b, 2H), 4.75 (m, 1H), 4.30 (m, 4H), 3.8 (m, 2H), 3.3
(m, 2H), 2.75 (t, 2H), 1.2 (t, 9H). LCMS m/z: 559 (M + 1).
2-Cyclohexyloxy-12-(2-hydroxyethyl)-13,14-dihydro-naph-
thol[2,1-a]pyrrolo[3,4-c]carbazole-5-one (20). 1H NMR (DMSO-
d6, δ): 8.41(s, 1H), 7.96 (d, 1H), 7.85 (d, 1H), 7.67 (d, 1H), 7.50
(t, 1H), 7.29 (t, 1H), 6.86 (s, 1H), 6.75 (d, 1H), 5.03 (m, 1H), 4.90

MLK1 and MLK3 Subtype-SelectiVe Inhibitors
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18 5687
(m, 1H), 4.79 (s, 2H), 4.64 (m, 2H), 3.84 (m, 4H), 2.77 (m, 2H),
1.61-2.0 (m, 10H). MS m/z: 467 (M + 1).
500 µg/mL myelin basic protein. The reaction was initiated by
adding purified recombinant kinases (50, 150, and 100 ng of GST-
MLK1KD, GST-MLK2KD/LZ, and GST-MLK3KD, respectively).
Samples were incubated for 15 min at 37 °C. The reaction was
stopped by adding ice-cold 50% TCA, and the proteins were
allowed to precipitate for 30 min at 4 °C. The plates were then
washed with ice-cold 25% TCA. A supermix scintillation cocktail
was added, and the plates were allowed to equilibrate for 1 to 2 h
prior to counting by the Wallac 1450 MicroBeta-Plus scintillation
counter. The inhibition curves were generated for the compounds
by plotting percent control activity versus log10 of the concentration
of compound. The IC50 values were calculated by nonlinear
regression using the sigmoidal dose-response (variable slope)
equation in GraphPad Prism as follows: y ) bottom + (top -
bottom)/(1 + 10(log IC50 - x) × hillslope), where y is the percent
kinase activity at a given concentration of the compound, x is the
logarithm of the concentration of the compound, “bottom” is the
percent of control kinase activity at the highest compound
concentration tested, and “top” is the percent of control kinase
activity at the lowest compound concentration examined. The values
for bottom and top were fixed at 0 and 100, respectively. IC50 values
were reported as the average of duplicates that agree within 20%.
Other Kinase Assays. Inhibition assays for recombinant receptor
tyrosine kinases (EGFR, FGFR1, ꢀIRK, PDGFRꢀ, TIE2, TrkA,
VEGF-R2) and serine/threonine kinases (CDK1, CDK2, DLK,
JNK1, p38R, PKC) were performed as described previously.10b,19,20
Protein-Complex Preparation, Crystallization, and X-ray
Data Collection. MLK111 (1.0 mg/mL in 20 mM HEPES, 25 mM
NaCl, 1 mM DTT, and 1 mM EDTA (pH 7.4)) was incubated with
1 mM 16 for 30 min at 4 °C. Cocrystals of the enzyme11 and 16
were obtained by mixing 2 µL of the protein complex (1.7 mg/ml)
in 20 mM HEPES (pH 7.4), 25 mM NaCl, 1 mM DTT, and 1mM
EDTA with 1 µL of the reservoir solution of 12% PEG20000, 200
mM Am2(SO4), and 100 mM MES (pH 6.5). The prepared complex
was crystallized by hanging-drop vapor diffusion by combining 3
µL of the protein-ligand solution with 1 µL of a reservoir solution
containing 20% PEG 3350 and 0.2 M MgSO4 (pH 5.9). The crystals
appeared in 5 to 6 days as thin plates and exhibited diffraction that
was consistent with that of the orthorhombic space group P21212
(a ) 56.93, b ) 126.30, and c ) 41.36 Å) with 1 molecule of
enzyme-inhibitor complex per asymmetric unit. Prior to data
collection, the crystals were transferred to cryoprotectant solutions
composed of their mother liquors and 20% glycerol. After incuba-
tion for about 15 s, the crystals were flash-cooled in a nitrogen
stream. X-ray diffraction data were collected at 2.6 Å resolution at
the NSLS X4A beamline (Brookhaven National Laboratory) on an
ADSC CCD detector. We integrated and scaled diffraction intensi-
ties by using the Denzo and Scalepack programs.21 The final 2.6
Å data set for the crystal of the complex was 96.8% complete with
an Rmerge of 0.079 (Table 2).
2-(2-Pyridylmethyl)-12-(2-hydroxyethyl)-13,14-dihydro-naph-
thol[2,1-a]pyrrolo[3,4-c]carbazole-5-one (21). A mixture of 17
(30 mg, 0.08 mmol), 2-picolyl chloride hydrochloride (20 mg, 0.12
mmol), and cesium carbonate (150 mg, 0.468 mmol) in acetonitrile
(8 mL) was heated to 80 °C in a sealed tube for 18 h. The solvent
was evaporated, and water was added to produce a solid, which
was collected. We purified the product by silica gel preparative
chromatography using ethyl acetate to produce 7 mg (19%) of a
tan solid; 1H NMR (DMSO-d6, δ): 8.56 (s, 1H), 8.36 (s, 1H), 7.92
(d, 1H), 7.85 (m, 2H), 7.66 (d, 1H), 7.51 (d, 1H), 7.48 (t, 1H),
7.33 (m, 1H), 7.27 (m, 1H), 6.97 (s, 1H), 6.85 (d, 1H), 5.20 (s,
2H), 4.97 (m, 1H), 4.75 (s, 2H), 4.62 (m, 2H), 3.6-3.8 (M, 4H),
2.8 (m, 2H). LCMS m/z: 476 (M + 1).
2-Methoxyethoxy-12-(2-hydroxyethyl)-13,14-dihydronaph-
thol[2,1-a]pyrrolo[3,4-c]carbazole-5(6H)-one (22). This com-
pound was synthesized using the method for 21 starting with 17
(50 mg, 0.013 mmol) and 2-bromoethyl methyl ether (32 mg, 0.221
mmol) to produce a tan solid (25 mg, 43%). 1H NMR (DMSO-d6,
δ): 8.36 (s, 1H), 7.90 (d, 1 h), 7.83 (d, 1H), 7.64 (d, 1H), 7.45 (t,
1H), 7.24 (t, 1H), 6.87 (s, 1H), 6.77 (d, 1H), 4.97 (t, 1H), 4.75 (s,
2H), 4.61 (s, 2H), 4.11 (s, 2H), 3.77 (d, 2H), 3.65 (s, 2H), 2.73 (s,
2H). LCMS m/z: 443 (M + H).
2-(2-Benzyloxyethoxy)-12-(2-hydroxyethyl)-13,14-dihydronaph-
thol[2,1-a]pyrrolo[3,4-c]carbazole-5(6H)-one (23). This com-
pound was synthesized using a method similar to that for 21 starting
with 17 (19.5 mg, 0.05 mmol), K2CO3 (34.6 mg, 0.25 mmol), KI
(8.7 mg, 0.52 mmol), and benzyl 2-bromoethyl ether (8.3 µL, 0.52
mmol) in acetone/DMF (2 mL; 4:1). Purification by preparative
TLC using CH2Cl2/MeOH (95:5) afforded the product as a solid
1
(10 mg, 39%). H NMR (DMSO-d6, δ): 8.5 (s, 1H), 8.0 (m, 1H),
7.90 (m, 1H), 7.7 (m, 1H), 7.5 (m, 1H), 7.3 (m, 6H), 7.0 (s, 1H),
6.9 (m, 1H), 5.0 (m, 1H), 4.80 (s, 2H), 4.65 (m, 6H), 4.05 (s, 2H),
3.80 (m, 2H), 3.30 (m, 2H), 2.50 (m, 2H). LCMS m/z: 519
(M + 1).
2-(2-Hydroxyethoxy)-12-(2-hydroxyethyl)-13,14-dihydronaph-
thol[2,1-a]pyrrolo[3,4-c]carbazole-5(6H)-one (24). A mixture of
compound 23 (5 mg, 0.01 mmol), 10% Pd(OH)2/C, and 0.1 mL of
12 N HCl in ethanol (1.0 mL) was hydrogenated at 42 psi on a
Parr apparatus for 24 h at rt. After filtration and concentration, 2.2
mg (27%) of the product was isolated as a solid. 1H NMR (DMSO-
d6, δ): 8.5 (s, 1H), 8.0 (m, 1H), 7.9 (m, 1H), 7.7 (m, 1H), 7.5 (m,
1H), 7.30 (m, 1H), 6.99 (s, 1H), 6.9 (m, 1H), 5.01 (s, 1H), 4.90
(m, 2H), 4.80 (m, 2H), 4.62 (m, 2H), 4.0 (s, 2H), 3.9 (m, 4H), 3.3
(m, 2H), 2.80 (m, 2H). LCMS m/z: 451(M + 1).
2-Acetonitrile-12-(2-hydroxyethyl)-13,14-dihydro-naphthol[2,1-
a]pyrrolo[3,4-c]carbazole-5-one (25). This compound was syn-
thesized using the method for 21 starting with 17 and 2-bromoac-
1
etonitrile to produce an off-white solid. H NMR (DMSO-d6, δ):
8.42 (s, 1H), 7.94 (m, 2H), 7.65 (m, 1H), 7.64 (m, 1H), 7.45 (t,
1H, J ) 7.4 Hz), 7.26 (t, 1H, J ) 7.4 Hz), 7.0 (s, 1H), 6.90 (d, 1H,
J ) 6.6 Hz), 5.2 (s, 2H), 4.97 (m, 1H), 4.77 (s, 2H), 4.61 (m, 2H),
3.80 (m, 2H), 3.3 (m, 2H), 2.77 (m, 2H). LCMS m/z: 424
(M + H).
Structure Determination and Model Refinement. The structure
of the MLK1-16 complex was solved by molecular replacement
using the EPMR program.22 The APO-MLK1 monomer structure
(unpublished results) was used as the search model. The solution
was found as one clear monomer in a search that used all data
between 15 and 4 Å. The model was subsequently refined with
CNS.23 Bound 16 could be clearly seen in the electron-density maps
immediately after the first cycle of rigid-body refinement of the
protein molecule alone. Iterative cycles of manual rebuilding with
TOM24 and refinement with CNS resulted in a model of the
complex at Rcryst ) 0.218 and Rfree) 0.282. The final structure
2-(3-Butyronitrile)-12-(2-hydroxyethyl)-13,14-dihydro-naph-
thol[2,1-a]pyrrolo[3,4-c]carbazole-5-one (26). This compound was
synthesized using the method for 21 starting with 17 and 4-bromo
1
butyronitrile to produce an off-white solid. H NMR (DMSO-d6,
δ): 8.36 (s, 1H), 7.82-7.92 (m, 2H), 7.64 (d, 1H, J ) 8.4), 7.46 (t,
1H, J ) 7.4 Hz), 7.24 (t, 1H, J ) 7.5 Hz), 6.89 (s, 1H), 6.79 (d,
1H, J ) 6.6 Hz), 4.97 (m, 1H), 4.75 (s, 2H), 4.61 (m, 2H), 4.07
(m, 2H), 3.80 (m, 2H), 3.3 (m, 2H), 2.62-2.73 (m, 4H), 2.0 (m,
2H). LCMS m/z: 452 (M + H).
Mixed-Lineage Kinase Assays. The MLK1, MLK2 and MLK
assays were performed by using the Millipore multiscreen trichlo-
roacetic acid (TCA) in-plate format, as described previously.6c,8
Each 50 µL assay mixture contained 20 mM Hepes (pH 7.2), 5
mM EGTA, 15 mM MgCl2, 1 mM DTT, 25 mM ꢀ-glycerophos-
phate, Km level of ATP (60 µM for MLK1 or 100 µM for MLK2
and MLK3), 0.25 µCi [γ-32P]ATP, 0.1% BSA, 2% DMSO, and
-2
contains 1913 protein atoms, one inhibitor molecule, one SO4
ion and 43 water molecules for a single monomer of the complex
in the asymmetric unit. None of the nonglycine residues lie in the
disallowed regions of the Ramachandran plot. The X-ray data
collection and crystallographic refinement statistics are represented
in Table 2.
Coordinates. The coordinates for MLK1-16 have been deposited
in the Protein Data Bank (PDB code is 3DTC).
Rat Pharmacokinetics. Adult male Sprague-Dawley (Charles
River, Kingston, NY) rats (n ) 3/treatment group) were dosed via

5688 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18
Hudkins et al.
Vaught, J. L.; Neff, N. T. CEP-1347/KT7515 prevents motor neuronal
programmed cell death and injury-induced dedifferentiation in vivo.
J. Neurobiol. 1998, 35, 361–370. (b) Borasio, G. D.; Hostmann, S.;
Anneser, J. M. H.; Neff, N. T.; Glicksman, M. A. CEP-1347/KT7515,
a JNK pathway inhibitor, supports the in vitro survival of chick
embryonic neurons. NeuroReport 1998, 9, 1435–1439. (c) Maroney,
A. C.; Finn, J. P.; Connors, T. J.; Durkin, J. T.; Angeles, T.; Gessner,
G.; Xu, Z.; Meyer, S. L.; Savage, M. J.; Greene, L. A.; Scott, R. W.;
Vaught, J. L. CEP-1347 (KT-7515), a semisynthetic inhibitor of the
mixed lineage kinase family. J. Biol. Chem. 2001, 276, 25302–25308.
(7) (a) Saporito, M. S.; Brown, E. M.; Miller, M. S.; Carswell, S. CEP-
1347/KT7515, an inhibitor of c-Jun N-terminal kinase activation,
attenuates the 1-methyl-4-phenyl tetrahydropyridine-mediated loss of
nigrostriatal dopaminergic neurons in vivo. J. Pharmacol. Exp. Ther.
1999, 288, 421–427. (b) Saporito, M. S.; Thomas, B. A.; Scott, R. W.
MPTP activates c-jun NH2-terminal kinase (JNK) and its upstream
regulator kinse MKK4 in nigrostriatal neurons in vivo. J. Neurochem.
2000, 75, 1200–1208.
(8) Hudkins, R. L.; Johnson, N. W.; Angeles, T. S.; Gessner, G.; Mallamo,
J. P. Synthesis and mixed lineage kinase activity of pyrrolocarbazole
and isoindolone analogs of (+)K-252. J. Med. Chem. 2007, 50, 433–
441.
(9) Hudkins, R. L.; Park, C.-H. Synthesis of indeno[2,1-a]pyrrolo[3,4-
c]carbazole lactam regioisomers using ethylcis-ꢀ-cyanoacrylate as a
dienophile and lactam precursor. J. Heterocycl. Chem. 2003, 40, 135–
142.
the lateral tail vein with the indicated dose for intravenous
administration (3% DMSO, 30% solutol, and 67% phosphate-
buffered saline) or via oral gavage (50% Tween 80, 40% propylene
carbonate, and 10% propylene glycol) at the indicated dose. Rats
were fasted overnight prior to oral dose administration. Serial blood
samples were collected from the lateral tail vein in heparinized
collection tubes (approximately 0.25 mL) at seven sampling times
over a 6 h period. The plasma was separated by centrifugation,
and the sample was prepared for analysis HPLC/MS by protein
precipitation with acetonitrile. The plasma samples were analyzed
for drug and internal standard via LCMS/MS protocol. The
pharmacokinetic parameters were calculated by a noncompartmental
method using WinNonlin software (professional version 4.1,
Pharsight Corporation, Palo Alto, CA, 1997).
Acknowledgment. We acknowledge technical support from
George Gessner, Thomas Connors, Shi Yang, Beth McKenna,
and Beth Ann Thomas and insightful discussions and support
from Drs. Jeffry Vaught and Ed Bacon from Cephalon, Inc.
and Ejner Moltzen and Erik Nielsen from Lundbeck A/S.
Supporting Information Available: Elemental analysis of key
compounds. This material is available free of charge via the Internet
at http://pubs.acs.org.
(10) (a) Gingrich, D. E.; Hudkins, R. L. Synthesis and kinase inhibitory
activity of 3′-(S)-epi-K-252a. Bioorg. Med. Chem. Lett. 2002, 12,
2829–2831. (b) Gingrich, D. L.; Yang, S. X.; Gessner, G. W.; Angeles,
T. S.; Hudkins, R. L. Synthesis, modeling and in Vitro activity of 3′-
(S)-epi-K-252a analogs. Elucidating the stereochemical requirements
of the 3′-sugar alcohol on trkA tyrosine kinase activity. J. Med. Chem.
2005, 48, 3776–3783.
References
(1) (a) Maroney, A. C.; Saporito, M. A.; Hudkins, R. L. Mixed lineage
kinase family, potential targets for preventing neurodegeneration. Curr.
Med. Chem.: Cent. NerV. Syst. Agents 2002, 2, 143–155. (b) Wang,
L. H.; Besirli, C. G.; Johnson, E., Jr. Mixed lineage kinases: a target
for the prevention of neurodegeneration. Annu. ReV. Pharmacol.
Toxicol. 2004, 44, 451–474. (c) Gallo, K. A.; Johnson, G. L. Mixed-
lineage kinase control of JNK and p38 MAPK pathways. Nat. ReV.
Mol. Cell. Biol. 2002, 3, 663–672.
(2) (a) Bozyczko-Coyne, D.; Saporito, M. S.; Hudkins, R. L. Targeting
the JNK pathway for therapeutic benefit in CNS disease. Curr. Drug
Targets: CNS Neurol.Disord. 2002, 1, 31–49. (b) Bogoyevitch, M. A.;
Boehm, I.; Oakley, A.; Ketterman, A. J.; Barr, R. K. Targeting the
JNK MAPK cascade for inhibition: basic science and therapeutic
potential. Biochim. Biophys. Acta 2004, 1697, 89–101. (c) Harper,
S. J.; LoGrasso, P. Signaling for survival and death in neurons. The
role of stress-activated kinases, JNK and p38. Cell. Signalling 2001,
13, 299–310. (d) Mielke, K.; Herdegen, T. JNK and p28 stress kinases:
degenerative effectors of signal-transduction-cascades in the nervous
system. Prog. Neurobiol. 2000, 61, 45–60.
(3) (a) Gupta, S.; Barrett, T.; Whitmarsh, A. J.; Cavanagh, J.; Sluss, H. K.;
Derijard, B.; Davis, R. J. Selective interaction of JNK protein kinase
isoforms with transcription factors. EMBO J. 1996, 15, 2760–2770.
(b) Kallunki, T.; Su, B.; Tsigelny, I.; Sluss, H. K.; Derijard, B.; Moore,
G.; Davis, R.; Karin, M. JNK2 contains a specificity-determining
region responsible for efficient c-Jun binding and phosphorylation.
Genes DeV. 1994, 8, 2996–3007. (c) Kallunki, T.; Deng, T.; Hibi,
M.; Karin, M. c-Jun can recruit JNK to phosphorylate dimerization
partners via specific docking interactions. Cell 1996, 87, 1–20.
(4) (a) Hirsch, E. C.; Hunot, S.; Faucheux, B.; Agid, Y.; Mizuno, Y.;
Mochizuki, H.; Tatton, W. G.; Tatton, N.; Olanow, W. C. Dopamin-
ergic neurons degenerate by apoptosis in Parkinson’s disease. MoV.
Disord. 1999, 14, 383–385. (b) Hartman, A.; Hunot, S.; Michel, P. P.;
Muriel, M. P.; Vyas, S.; Faucheux, B.; Mouatt-Prigent, A.; Turmel,
H.; Srinivasan, A.; Ruberg, M.; Evan, G. I.; Agid, Y.; Hirsch, E. C.
Caspace-3: a vulnerability factor and final effector in apoptotic death
of dopaminergic neurons in Parkinson’s disease. Proc. Natl. Acad.
Sci. U.S.A. 2000, 97, 2875–2880.
(5) (a) Saporito, M. A.; Hudkins, R. L.; Maroney, A. C. Discovery of
CEP-1347/KT-7515: an inhibitor of the cJun N-terminal kinase
pathway for the treatment of neurodegenerative diseases. Prog. Med.
Chem. 2002, 40, 23–62. (b) Murakata, C.; Kaneko, M.; Gessner, G.;
Angeles, T. S.; Ator, M. A.; O’Kane, T. M.; McKenna, B. A. W.;
Thomas, B. A.; Mathiasen, J. R.; Saporito, M. S.; Bozyczko-Coyne,
D.; Hudkins, R. L. Mixed lineage kinase activity of indolocarbazole
analogs. Bioorg. Med. Chem. Lett. 2002, 12, 147–150. (c) Kaneko,
M.; Saito, Y.; Saito, H.; Matsumoto, T.; Matsuda, Y.; Vaught, J. L.;
Dionne, C. A.; Angeles, T. S.; Glicksman, M. A.; Neff, N. T.; Rotella,
D. P.; Kauer, J. C.; Mallamo, J. P.; Hudkins, R. L.; Murakata, C.
Neurotrophic 3,9-bis[(alkylthiomethyl]- and -[bis(alkoxymethyl)]-K-
252a Derivatives. J. Med. Chem. 1997, 40, 1863–1869.
(11) The CEP-1347-MLK1 and apo-MLK1 crystal structures are the subject
of a separate publication.
(12) Schiering, N.; Knapp, S.; Marconi, M.; Flocco, M. M.; Cui, J.; Perego,
R.; Rusconi, L.; Cristiani, C. Crystal structure of the tyrosine kinase
domain of the hepatocyte growth factor receptor c-Met and its complex
with the microbial alkaloid K-252a. Proc. Natl. Acad. Sci. U.S.A. 2003,
22, 12654–12659.
(13) Ruggeri, B. A.; Miknyoczki, S. J.; Singh, J.; Hudkins, R. L. The role
of neurotrophin-trk interactions in oncology: The antitumor efficacy
of potent and selective trk tyrosine kinase inhibitors on preclinical
tumor models. Curr. Med. Chem. 1999, 6, 845–857.
(14) (a) Langston, J. W. The etiology of Parkinson’s disease with emphasis
on the MPTP story. Neurology 1996, 47, S153-S160. (b) Heikkila,
R. E.; Manzino, L.; Cabbat, F. S.; Duvoisin, R. C. Protection against
the dopaminergic neurotoxicity of 1-methyl-4-phenyl-1,2,3,6-tetrahy-
dropyridine by monoamine oxidase inhibitors. Nature 1984, 311, 467–
469. (c) Heikkila, R. E.; Hess, A.; Duvoisin, R. C. Dopaminergic
neurotoxicity of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine in mice.
Science 1984, 224, 1451–1453. (d) Blanchet, P. J.; Konitsiotis, S.;
Hyland, K.; Arnold, L. A.; Pettigrew, K. D.; Chase, T. N. Chronic
exposure to MPTP as a primate model of progressive Parkinsonism:
a pilot study with a free radical scavenger. Exp. Neurol. 1998, 153,
214–222.
(15) Nicklas, W. J.; Vyas, I.; Heikkila, R. E. Inhibition of NADH-linked
oxidation in brain mitochondria by 1-methyl-4-phenylpyridine, a
metabolite of the neurotoxin, 1-methyl-4-phenylpyridine. Life Sci.
1985, 36, 2503–2508.
(16) (a) Mathiasen, J. R.; McKenna, B. A. W.; Saporito, M. S.; Ghadge,
G. D.; Roos, R. P.; Holskin, B. P.; Wu, Z.-L.; Trusko, S. P.; Connors,
T. P.; Maroney, A. C.; Thomas, B. A.; Thomas, J. C.; Bozyczko-
Coyne, D. Inhibition of mixed lineage kinase 3 attenuates MPP+-
induced neurotoxicity in SH-SY5Y cells. Brain Res. 2004, 1003, 86–
97. (b) Cassarino, D. S.; Halvorsen, E. M.; Swerdlow, R. H.;
Abramova, N. N.; Parker, W. D., Jr.; Sturgill, T. W.; Bennett, J. P.,
Jr. Interaction among mitochondria, mitogen-activated protein kinases
and nuclear factor kB in cellular models of Parkinson’s disease.
J. Neurochem. 2000, 74, 1384–1392.
(17) (a) Sheehan, J. P.; Palmer, P. E.; Helm, G. A.; Tuttle, J. B. MPP+
induced apoptotic cell death in SH-SY5Y neuroblastoma cells: an
electron microscope study. J. Neurosci. Res. 1997, 48, 226–37. (b)
Dipasquale, B.; Marini, A. M.; Youle, R. J. Apoptosis and DNA
degradation induced by 1-methyl-4-phenylpyridinium in neurons.
Biochem. Biophys. Res. Commun. 1991, 181, 1442–1448.
(18) Tatton, N. A.; Kish, S. J. In situ detection of apoptotic nuclei in the
substantia nigra compacta of 1-methyl-4-phenyl-1,2,3,6-tetrahydro-
pyridine-treated mice using terminal deoxynucleotidyl transferase
labelling and acridine orange staining. Neuroscience 1997, 77, 1037–
1048.
(6) (a) Glicksman, M. A.; Chiu, A. Y.; Dionne, C. A.; Kaneko, M.;
Murakata, C.; Oppenheim, R. W.; Prevette, D.; Sengelaub, D. R.;
(19) Angeles, T. S.; Steffler, C.; Bartlett, B. A.; Hudkins, R. L.; Stephens,
R. M.; Kaplan, D. R.; Dionne, C. A. Enzyme linked immunosorbant

MLK1 and MLK3 Subtype-SelectiVe Inhibitors
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 18 5689
assay for TrkA tyrosine kinase activity. Anal. Biochem. 1996, 236,
49–55.
(22) Kissinger, C. R.; Gehlhaar, D. K.; Fogel, D. B. Rapid automated
molecular replacement by evolutionary search. Acta Crystallogr., Sect.
D: Biol. Crystallogr. 1999, 55, 484–491.
(23) Jones, A. T. Interactive computer graphics: FRODO. Methods Enzymol.
1985, 115, 157–171.
(24) Brunger, A. T.; Adams, P. D.; Clore, G. M.; DeLano, W. L.; Gros,
P.; Grosse-Kunstleve, R. W.; Jiang, J. S.; Kuszewski, J.; Nilges, M.;
Pannu, N. S.; Read, R. J.; Rice, L. M.; Simonson, T.; Warren, G. L.
Crystallography & NMR system: Aa new software suite for macro-
molecular structure determination. Acta Crystallogr., Sect. D: Biol.
Crystallogr. 1998, 54, 905–921.
(20) Ruggeri, B.; Singh, J.; Gingrich, D.; Angeles, T.; Albom, M.; Chang,
H.; Robinson, C.; Hunter, K.; Dobrzanski, P.; Jones-Bolin, S.; Aimone,
L.; Klein-Szanto, A.; Herbert, J.-M.; Bono, F.; Casellas, P.; Bourie,
B.; Pili, R.; Isaacs, J.; Ator, M.; Hudkins, R.; Dionne, C.; Mallamo,
J.; Vaught, J. CEP-7055: a novel, orally-active pan inhibitor of vascular
endothelial growth factor receptor tyrosine kinases with potent anti-
angiogenic activity and anti-tumor efficacy in pre-clinical models.
Cancer Res. 2003, 63, 5978–5991.
(21) Otwinowski, Z.; Minor, W. Processing of X-ray Diffraction Data
Collected in Oscillation Mode. In Methods in Enzymology; Carter,
C. W. J., Sweet, R. M., Abelson, J. N., Simon, M. I., Eds.; Academic
Press: New York, 1997; 307-326.
JM8005838