Optimization of diaryl amine derivatives as kinesin spindle protein inhibitors

Article abstract of DOI:10.1016/j.bmc.2014.04.008

Structure-activity relationship studies of diaryl amine-type KSP inhibitors were carried out. Diaryl amine derivatives with a pyridine ring or urea group were less active when compared with the parent carboline and carbazole derivatives. Optimization studies of a lactam-fused diphenylamine-type KSP inhibitor revealed that the aniline NH group and 3-CF₃ phenyl group were indispensable for potent KSP inhibition. Modification with a seven-membered lactam-fused phenyl group and a 4-(trifluoromethyl)pyridin-2-yl group improved aqueous solubility while maintaining potent KSP inhibitory activity. From these studies, we identified novel diaryl amine-type KSP inhibitors with a favorable balance of potency and solubility.

Full text of DOI:10.1016/j.bmc.2014.04.008

Bioorganic & Medicinal Chemistry 22 (2014) 3171–3179
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Optimization of diaryl amine derivatives as kinesin spindle protein
inhibitors
a
a
a
b
Tomoki Takeuchi a, Shinya Oishi a, , Masato Kaneda , Ryosuke Misu , Hiroaki Ohno , Jun-ichi Sawada ,
⇑
Akira Asai b, Shinya Nakamura c, Isao Nakanishi c, Nobutaka Fujii a,
⇑
a Graduate School of Pharmaceutical Sciences, Kyoto University, Sakyo-ku, Kyoto 606-8501, Japan
b Graduate School of Pharmaceutical Sciences, University of Shizuoka, Suruga-ku, Shizuoka 422-8526, Japan
c Faculty of Pharmacy, Kinki University, 3-4-1 Kowakae, Higashi-osaka 577-8502, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Structure–activity relationship studies of diaryl amine-type KSP inhibitors were carried out. Diaryl amine
derivatives with a pyridine ring or urea group were less active when compared with the parent carboline
and carbazole derivatives. Optimization studies of a lactam-fused diphenylamine-type KSP inhibitor
revealed that the aniline NH group and 3-CF3 phenyl group were indispensable for potent KSP inhibition.
Modification with a seven-membered lactam-fused phenyl group and a 4-(trifluoromethyl)pyridin-2-yl
group improved aqueous solubility while maintaining potent KSP inhibitory activity. From these studies,
we identified novel diaryl amine-type KSP inhibitors with a favorable balance of potency and solubility.
Ó 2014 Elsevier Ltd. All rights reserved.
Received 12 March 2014
Revised 5 April 2014
Accepted 5 April 2014
Available online 13 April 2014
Keywords:
Diaryl amine
Kinesin spindle protein
Aqueous solubility
1. Introduction
N
H
CF3
CF3
O
N
H
N
H
CF3
Kinesins constitute a superfamily of molecular motor proteins
to move along microtubules.1 Mitotic kinesins are involved in cell
division.2 Non-mitotic kinesins are principally involved in intracel-
lular transport of organelles and vesicles.3 The kinesin spindle pro-
tein (KSP; also known as Eg5) is the mitotic kinesin that belongs to
the kinesin-5 family. The structure of KSP is comprised of three
1
2
O
N
N
H2N
O
N
N
CF3
H
H
H
3a
4c
parts: an N-terminal motor domain, a central a-helical coiled coil
H
H
stalk domain, and a C-terminal tail domain.4 The N-terminal motor
domain contains a catalytic site for ATP hydrolysis and microtu-
bule binding region. KSP moves toward the plus end of the micro-
tubule, just like other kinesins with an N-terminal motor domain,
using the energy generated from the hydrolysis of ATP.5 The KSP
movement is required for centrosome separation and bipolar spin-
dle formation during cell division. Inhibition of KSP leads to mitotic
arrest in the prometaphase with the formation of the monopolar
spindle and subsequent apoptotic cell death.6–9 Therefore, KSP
inhibitors are expected to be favorable agents for cancer chemo-
therapy without neurotoxic side effects.10–13
O
N
N
N
N
CF3
N
CF3
H
H
5a
6
Figure 1. Structures of carbazole- and carboline-type (1–4) and diaryl amine-type
(5, 6) KSP inhibitors.
b-carboline derivative 3a were also identified as highly potent
KSP inhibitors by structure–activity relationship studies of 1.15
However, these inhibitors exhibited limited solubility in the aque-
ous solvents employed for in vivo studies. To satisfy the potent
inhibitory activity requirements as well as better solubility in
aqueous solution, we have designed diphenylamine derivatives
such as 5a by modification of the planar carbazole-type inhibitor
2.16 Diphenylamine 5a exhibited better solubility than carbazole
2 while maintaining potent KSP inhibitory activity. Structural
analysis by single crystal X-ray diffraction studies and free energy
Recently, we reported that carbazole derivative 1 with the 2-CF3
group showed potent KSP inhibitory activity (Fig. 1).14 Carbazole
derivatives, with a lactam ring (2) or urea group (4c), and the
⇑
Corresponding authors. Tel.: +81 75 753 4551; fax: +81 75 753 4570.
E-mail addresses: soishi@pharm.kyoto-u.ac.jp (S. Oishi), nfujii@pharm.kyoto-u.
ac.jp (N. Fujii).
http://dx.doi.org/10.1016/j.bmc.2014.04.008
0968-0896/Ó 2014 Elsevier Ltd. All rights reserved.

3172
T. Takeuchi et al. / Bioorg. Med. Chem. 22 (2014) 3171–3179
calculations demonstrated that the improved solubility of 5a is
attributed to fewer van der Waals interactions in the crystal
packing, as well as a hydrogen-bond acceptor nitrogen in the
aniline moiety for favorable solvation. Interestingly, compound
R1
R2
a
R2
R1
+
X
N
H
H2N
X = Br or OTf
9
10
16
5, 7, 8a,d, 11–14
5a possibly binds to the interface of the
ATP-competitive manner, whereas most KSP inhibitors (e.g., mon-
astrol, S-trityl- -cysteine) bind to the allosteric pocket formed by
helices 2, 3 and loop L5 to show ATP-uncompetitive behav-
a4 and a6 of KSP in an
H
H
O
N
O
O
N
b
c
L
BF4
I
+
a
a
F3C
2
O
CF3
OH
ior.17–19 Replacing the right-hand 3-CF3-phenyl group in 5a with
a pyridine ring provided a more soluble KSP inhibitor 6; however,
this compound showed slightly lower potency than 5a.16 In this
article, we describe the structure–activity relationship study for
novel diaryl amine-type KSP inhibitors with high potency and
aqueous solubility. For this purpose, we performed: (i) modifica-
tion of ring-fused indoles such as 3a and 4c using the same
approach employed for the development of 5a and (ii) intensive
optimization studies of diphenylamine 5a.
15
18
17a
H
N
H
R2
R1
O
N
R2
R1
+
I
SH
S
20a: R1 = CF3, R2 = H
20b R1 = H, R2 = CF3
19a: R1 = CF3, R2 = H
19b: R1 = H, R2 = CF3
:
H
R2
R1
O
N
N
CF3
N
OR
2. Results and discussion
H
H
8a: R1 = NO2, R2 = H
11h: R = Me
f
8d: R1 = H, R2 = NO2
11j: R = H
d
8b: R1 = NH2, R2 = H
2.1. Investigation of diaryl amine-type KSP inhibitors by
modification of ring-fused scaffolds
H
N
: R1 = H, R2 = NH2
8e
8c
O
e
: R1 = NHCONH2, R2 = H
8f: R1 = H, R2 = NHCONH2
N
H
R
We speculated that the poor solubility of carboline and
carbazole derivatives would be attributable to the significant inter-
11k: R = NO2
g
11l: R = NH2
molecular interactions in the crystals (e.g.,
p–p stacking interac-
tions) as seen for compound 2.16 To disrupt the possible crystal
packing of compounds 3a and 4c, the design of less planar analogs
H
H
N
O
N
X
was expected to be
a
promising approach.20 Therefore, we
N
H
CO2R
N
H
CF3
designed diaryl amine derivatives 7 and 8, in which the pyrrole
CAC bond in the central part of carbolines 3 and carbazoles 4
was cleaved (Fig. 2). Diaryl amines 7a,b with a pyridine ring were
designed based on carbolines 3a,b with potent KSP inhibitory
activity (Fig. 2A). Diphenylamines 8a–f with a nitro, amino or urea
group at the 3- or 4-position on the left-hand phenyl ring were
similarly investigated, which represent the cleaved analogs of
carbazoles 4b–f (Fig. 2B). Diaryl amine derivatives 7a,b and 8a,d
were prepared by palladium-catalyzed N-arylation using aryl
bromides 9 and substituted anilines 10 (Scheme 1).21 For the prep-
aration of compounds 8c,f with a urea group, nitro derivatives 8a,d
were reduced to the corresponding amines 8b,e using Pd/C and
11m: R = Me
5a: X = O
h
i
11n
12k
: X = S
: R = H
j
O
N
H
N
H
12a
CF3
O
N
H
N
CF3
H
12c
Scheme 1. Synthesis of diaryl amine derivatives. Reagents and conditions: (a)
Pd2(dba)3, biaryl phosphine ligand, NaOt-Bu, toluene, 100 °C; (b) KOt-Bu, DMF,
40 °C; (c) CuI, ethylene glycol, K2CO3, 2-propanol, 80 °C; (d) Pd/C, HCO2NH4, EtOH,
reflux; (e) KOCN, AcOH, H2O, rt; (f) BBr3, CH2Cl2, rt; (g) Zn, AcOH, rt; (h) LiOHꢀH2O,
MeOH, H2O, 50 °C; (i) Lawesson’s reagent, toluene, reflux; (j) Pd(OAc)2, O2, AcOH,
115 °C.
(A)
ammonium formate, which were converted to the expected com-
pounds 8c,f by KOCN.
First, KSP inhibitory activity of compounds 7a,b, with a pyridine
ring in the left-hand part, was comparatively assessed with the
parent carboline-type inhibitors 3a,b (Table 1). Unfortunately,
the cleavage of the pyrrole CAC bond in carboline led to loss of
N
N
N
3
N
H
CF3
CF3
N
H
CF3
CF3
3a
3b
7a
7b
4
N
KSP ATPase inhibitory activity at 6.3 lM. The solubility of these
N
H
N
H
compounds was evaluated by a thermodynamic method.22 A mix-
ture of EtOH–phosphate buffer (pH 7.4) (1:1) [50% EtOH] and phos-
phate buffer (pH 7.4) were employed as aqueous media.22 In these
solutions, the parent carbazole-type inhibitor 1 was moderately
(B)
R
N
CF3
R
3
N
CF3
soluble (0.424 mg/mL) and insoluble (<1 lg/mL), respectively. N-
H
H
(Pyridin-3-yl)amine 7a showed the anticipated improvement in
thermodynamic solubility, being 30 times more soluble in 50%
EtOH (14.3 mg/mL) compared with the corresponding carboline
3a. N-(Pyridin-4-yl)amine 7b also exhibited approximately 14
times greater solubility in 50% EtOH (24.0 mg/mL) than the parent
carboline 3b. Of note, compound 7b had moderate solubility
4b:
8a: R = NO2
R = NH2
4c: R = NHCONH2
8b: R = NH2
8c: R = NHCONH2
4
R
R
N
H
CF3
N
H
CF3
4d:
8d: R = NO2
8e: R = NH2
R = NO2
(264 lg/mL) even in phosphate buffer, which was 80 times or more
4e: R = NH2
soluble compared with 3b and 3a, respectively. Although these
pyridinylamine derivatives 7a,b were inert in KSP inhibition, it
was suggested that cleavage of the pyrrole CAC bond in carboline
4f:
R = NHCONH2
8f: R = NHCONH2
Figure 2. Design of novel KSP inhibitors 7, 8 with diaryl amine scaffolds.

T. Takeuchi et al. / Bioorg. Med. Chem. 22 (2014) 3171–3179
3173
Table 1
approximately three times lower than that of the parent carbazole
4f.
KSP inhibitory activities and thermodynamic aqueous solubility of diaryl amines with
a pyridine ring and the related compounds
Compound
KSP ATPase
Solubility
2.2. Optimization studies of lactam-fused diaryl amine-type KSP
inhibitors
a,b
IC50
(lM)
50% EtOHc phosphate buffer
(mg/mL)
(pH 7.4) (lg/mL)
Next, we investigated the structure–activity relationship and
the further refinement of 5a for novel potent KSP inhibitors in
terms of: (i) the linkage between the two aryl groups, (ii) the sub-
stituent on the right-hand phenyl group, (iii) the left-hand hetero-
cycle, and (iv) the right-hand aromatic heterocycle (Fig. 3).
1
0.21
0.424
<1
N
H
CF3
CF3
CF3
CF3
CF3
N
N
H
3a 0.052
3b 0.095
7a >6.3d
7b >6.3
0.472
1.76
14.3
24.0
<1
A series of diaryl amine derivatives 5 and 11–14 were prepared
by Buchwald–Hartwig N-arylation using aryl bromides or triflates
9 and substituted anilines 10, except for the compounds (11j,l,n
and 12c,k) indicated below (Scheme 1).21 Diphenylether derivative
17a was obtained by treatment of phenol derivative 15 with
diaryliodonium tetrafluoroborate 16 in the presence of KOt-Bu.23
Diphenylsulfide derivatives 20a,b were prepared by the copper-
catalyzed CAS bond-forming reaction using aryl thiol 18 and
CF3-substituted iodobenzenes 19a,b.24 Compounds 11j, 11l, 11n
and 12k were obtained by simple transformations including
BBr3-mediated demethylation of 11h, Zn-mediated reduction of
11k, saponification of 11m, and thiocarbonylation of 5a using
Lawesson’s reagent,25 respectively. Compound 12c was prepared
by treatment of 12a with O2 in the presence of Pd(OAc)2.
N
3.16
10.8
264
N
H
N
N
H
N
N
H
a
Inhibition of microtubule-activated KSP ATPase activity.
IC50 values were derived from the dose–response curves generated from trip-
b
licate data points.
c
Solubility in an equal volume of EtOH and 1/15 M phosphate buffer (pH 7.4).
d
IC50 was ꢁ7.0
lM.
We initially investigated the type of heteroatom in the central
part of diphenylamine 5a and the position of a substituent on
the right-hand phenyl group (Table 3). Replacing the bridging NH
group in compound 5a with oxygen (17a,b: ether) or sulfur
(20a,b: thioether) resulted in a significant reduction or loss of
KSP inhibitory activity. This indicates that the aniline NH group
of 5a is an indispensable functional group as a hydrogen-bond
donor, which is supported by our previous modeling study.16
Regarding the position of CF3 group on the right-hand phenyl ring
of 5a, 4-CF3 compound 5b was approximately 7 times less potent
than 5a and no KSP inhibition of 2-CF3 compound 5c was observed.
Next, the structure–activity relationship was examined by
replacing 3-substituents on the right-hand phenyl group in 5a
(Table 4). Among compounds 11a–d with or without a 3-alkyl sub-
stituent, the tert-butyl derivative 11d exhibited the most potent
Table 2
KSP inhibitory activities of diphenylamines with a nitro, amino or urea group and the
related carbazoles
a,b
Compound
R
KSP ATPase IC50
(lM)
4b
4c
NH2
NHCONH2
0.27
0.085
R
R
N
CF3
H
4d
4e
4f
NO2
NH2
NHCONH2
0.043
0.55
0.12
N
H
CF3
CF3
8a
8b
8c
NO2
NH2
NHCONH2
>6.3
>6.3
>6.3
R
R
N
H
inhibitory activity (IC50 = 0.16 lM). The structure–activity relation-
8d
8e
8f
NO2
NH2
NHCONH2
3.2
>6.3
0.39
ship of the simple alkyl group correlated with that of the
carbazole-type KSP inhibitors, suggesting that carbazoles (1 and
2) and diphenylamines (5a and 11d) may occupy the same binding
site of KSP.15,16 Substitution with 3,5-di-CF3 (11e), 3-phenyl (11f),
3-phenoxy (11g) and 3-methoxy (11h) groups were not effective.
Introduction of a polar substituent such as 3-hydroxy (11j),
N
H
CF3
a
Inhibition of microtubule-activated KSP ATPase activity.
IC50 values were derived from the dose–response curves generated from trip-
b
licate data points.
and carbazole derivatives could be a promising approach to
improve the solubility in aqueous solution.
Diphenylamine derivatives 8a–f with a nitro, amino or urea
group at the 3- or 4-position on the left-hand phenyl group were
next examined (Table 2). The 3-substituted anilines 8a–c showed
Table 3
KSP inhibitory activities of dihydroquinolinone derivatives
a,b
Compound
X
KSP ATPase IC50
(lM)
H
5a
17a
20a
NH
O
S
0.045
2.0
>6.3
O
O
O
N
no KSP inhibitory activity at 6.3
analogs, compound 8f with a urea group exhibited moderate inhib-
itory activity (IC50 = 0.39 M), while compounds 8d,e with a nitro
or amino group had weak or no potency. The potency of 8f was
lM. Among the 4-substituted
X
X
CF3
CF3
l
H
N
5b
17b
20b
NH
O
S
0.33
>6.3
>6.3
H
N
A
B
(iv) = Aromatic Heterocycles
(iii) = Heterocycles
5c
>6.3
A
B
R
N
H
(ii) R = Various Substituents
X
CF3
n
(i) X = NH, O or S
a
Inhibition of microtubule-activated KSP ATPase activity.
IC50 values were derived from the dose–response curves generated from trip-
b
Figure 3. Strategy for the structure–activity relationship study of diaryl amine-type
licate data points.
KSP inhibitors.

3174
T. Takeuchi et al. / Bioorg. Med. Chem. 22 (2014) 3171–3179
Table 4
KSP inhibitory activities of diphenylamines with a 3-substituent or 3,5-substituents
on the right-hand phenyl group
H
N
5
O
R
3
N
H
R
KSP ATPase
R
KSP ATPase
a,b
a,b
IC50
0.045
>6.3
0.81
(l
M)
IC50
(lM)
3-CF3
H
3-Et
3-i-Pr
3-t-Bu
3,5-di-CF3
3-Ph
5a
3-OMe
3-OCF3
3-OH
3-NO2
3-NH2
11h
11i
11j
11k
11l
11m
11n
>6.3
1.2
>6.3
11a
11b
11c
11d
11e
11f
11g
0.43
0.16
>6.3
>6.3
>6.3
0.44
>6.3
>6.3
>6.3
3-CO2Me
3-CO2H
3-OPh
a
Inhibition of microtubule-activated KSP ATPase activity.
IC50 values were derived from the dose–response curves generated from trip-
b
Figure 5. Plausible binding mode of diphenylamine 11d at the interface of helices
4 and 6.
licate data points.
a
a
group, had slightly reduced potency (IC50 = 0.19
with 5a. Compound 12l, with a seven-membered lactam, showed
approximately equipotent KSP inhibitory activity (IC50 = 0.050 M)
to 5a, suggesting that some flexibility of the lactam carbonyl place-
ment is tolerated. Substitution with a fluorine group on the 5- or
6-position (12m,n) resulted in a reduction in the inhibitory
activity, suggesting that modification at these positions was
inappropriate for favorable interactions with KSP.
lM) compared
3-amino (11l), 3-methoxycarbonyl (11m), or 3-carboxylate (11n)
also gave rise to inactive compounds. 3-Trifluoromethyloxy (11i)
and 3-nitro (11k) derivatives showed moderate inhibitory activity.
It is inferred from these data that the substituent at the 3-position
on the right-hand phenyl part is buried in a relatively large hydro-
phobic pocket of KSP. The possible binding mode of 11d to the
l
interface of the
a4 and a6 helices of KSP is shown in Figure 5.
The tert-butyl group of 11d was buried in the deep hydrophobic
pocket formed by Tyr104, Gly296, Ile299, Thr300, Ile332, Tyr352,
Ala353, and Ala356. Low desolvation energy of tert-butyl group
compared with smaller alkyl groups may also contribute to the
favorable binding of 11d to KSP.
2.3. Analysis of aqueous solubility of potent KSP inhibitors and
further optimization
The thermodynamic solubility of potent diphenylamine deriva-
tives 12i,l in aqueous media was next evaluated (Table 6).22 Diphe-
nylamine 12i with a thiomorpholin-3-one structure was slightly
less soluble (1.70 mg/mL in 50% EtOH) than the parent compound
5a. The longer retention time on a reversed-phase HPLC column
(28.2 min) of compound 12i indicated that the introduction of a
sulfur atom into the lactam ring of 5a resulted in the increased
hydrophobicity, thereby lowering the solubility in aqueous media.
Diphenylamine 12l with a seven-membered lactam ring was
approximately four times more soluble (7.39 mg/mL in 50% EtOH)
than 5a, although a high CLogP value and HPLC retention time
indicated higher hydrophobicity. The lower melting point of com-
pound 12l (140 °C) suggested that the weak crystal packing mainly
contributed to the observed improvement in solubility. These
results indicated that the seven-membered lactam ring on the
In the left part of the molecule, the position of accessory amide
group was crucial for the potency (Table 5). Compounds 12a–d
with an amide group at the 3-position were less active or inactive,
while the parent compound 5a with the 4-position amide group
showed potent KSP inhibitory activity. A five-membered ring thio-
carbamate 12g exhibited moderate inhibitory activity (IC50 = 0.81 -
lM) in contrast to the ineffectiveness in lactam derivative 12e and
carbamate derivative 12f, suggesting that the introduction of a sul-
fur atom into the lactam ring had favorable effects on the bioactiv-
ity. The addition of a sulfur atom (12i) into the six-membered
lactam of 5a also maintained potent KSP inhibitory activity
(IC50 = 0.051 lM), whereas an oxygen atom (12h) decreased the
inhibitory activity. The loss of activity in the N-methylamide deriv-
ative 12j indicates that the lactam NH group at this position is
essential for KSP inhibition. Compound 12k, with a thioamide
O
H
N
H
N
H
O
O
N
N
H
CF3
S
N
H
CF3
N
H
CF3
5a
12i
12l
IC50 = 0.045 µM
solubility in 50% EtOH = 1.80 mg/mL
IC50 = 0.051 µM
IC50 = 0.050 µM
solubility in 50% EtOH = 1.70 mg/mL solubility in 50% EtOH = 7.39 mg/mL
H
N
O
H
H
N
O
N
O
N
N
N
N
H
CF3
S
N
H
13
CF3
N
H
14
CF3
6
IC50 = 0.068 µM
solubility in 50% EtOH = 3.51 mg/mL
Figure 4. Design of novel diaryl amine-type KSP inhibitors 13 and 14.

T. Takeuchi et al. / Bioorg. Med. Chem. 22 (2014) 3171–3179
3175
Table 5
and 12l. Compound 14 exhibited greater solubility in 50% EtOH
(4.82 mg/mL) and in phosphate buffer (8.07 g/mL) than the
corresponding compound 6. Of note, solubility of compound 14
in phosphate buffer was remarkably improved, as we expected,
in comparison with compound 12l (less than 1 lg/mL), although
14 was less soluble than 12l in 50% EtOH. Taken together,
compound 14 was identified to be a novel KSP inhibitor with a
favorable balance of potency and aqueous solubility.
Compounds 5a, 6, 12i,l, 13 and 14 were tested for their inhibi-
tory effect on the proliferation of cancer cell lines: lung cancer cells
A549, colorectal cancer cells HCT-116, and breast cancer cells MCF-
7 (Table 7). Cells were treated with increasing concentrations of
the compounds, and viabilities were measured by the 3-(4,5-
dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulf-
ophenyl)-2H-tetrazolium (MTS) assay. All the tested diaryl amine
derivatives were shown to be effective against these cell lines. In
particular, thiomorpholin-3-one derivative 13 with the highest
KSP inhibitory activity was found to be the most potent against
all the cell lines tested.26
KSP inhibitory activities of diphenylamines with
phenyl group
a
heterocycle on the left-hand
l
5
4
3
6
Ar
N
H
CF3
2
Ar
O
KSP
Ar
KSP
ATPase
ATPase
a,b
a,b
IC50
IC50
(
l
M)
(lM)
H
N
H
O
O
N
5a
0.045
12h
12i
0.29
O
H
N
12a >6.3
12b >6.3
0.051
O
N
H
S
Me
N
O
O
12j
12k
12l
>6.3
O
O
N
H
H
N
3. Conclusions
S
12c
4.6
0.19
N
H
We have performed structure–activity relationship studies for
the development of novel KSP inhibitors using carbolines 3, carbaz-
oles 4 and diphenylamine 5a as the lead compounds. Unfortu-
nately, the nonplanar analogs 7 and 8 of planar carbolines 3 and
carbazoles 4 decreased the potency for KSP inhibition. Optimiza-
tion studies of diphenylamine 5a revealed that bridging NH group,
3-CF3 group on the right-hand phenyl group, and the lactam amide
group at the 4-position on the left-hand phenyl group contributed
to the potent KSP inhibitory activity. Further investigations pro-
vided novel KSP inhibitors 13 with the most potent inhibitory
activity and 14 with the optimal balance of potency and aqueous
solubility in this study.
O
H
N
HN
12d >6.3
0.050
O
H
F
H
N
N
O
O
O
12e
3.6
12m
12n
0.43
0.92
H
N
H
N
F
O
O
12f
>6.3
O
H
N
12g
0.81
4. Experimental
4.1. Synthesis
S
a
Inhibition of microtubule-activated KSP ATPase activity.
IC50 values were derived from the dose–response curves generated from trip-
b
licate data points.
4.1.1. General methods
1H NMR spectra were recorded using a JEOL AL-400 or ECA-500
spectrometer. Chemical shifts are reported in d (ppm) relative to
Me4Si as an internal standard. 13C NMR spectra were referenced
to the residual DMSO signal. Exact mass (HRMS) spectra were
recorded on a JMS-HX/HX 110A mass spectrometer. Melting points
were measured by a hot stage melting point apparatus (uncor-
rected). For flash chromatography, Wakogel C-300E (Wako) or
ChromatorexÒ NH was employed. For analytical HPLC, a Cosmosil
5C18-ARII column (4.6 ꢂ 250 mm, Nacalai Tesque, Inc., Kyoto,
Japan) was employed with a linear gradient of CH3CN containing
0.1% (v/v) TFA at a flow rate of 1 mL/min on a Shimadzu LC-10ADvp
(Shimadzu Corp., Ltd., Kyoto, Japan), and eluting products were
detected by UV at 254 nm. The purity of the compounds was deter-
mined by combustion analysis or HPLC analysis (>95%).
left-hand phenyl group was a promising structural unit for the
development of KSP inhibitors that have a favorable balance of
bioactivity and aqueous solubility.
As such, we identified benzothiomorpholin-3-one 12i and ben-
zoazepin-2-one 12l as potent KSP inhibitors using structure–activ-
ity relationship studies of fused heterocycles on the left-hand
phenyl group. Compound 12l represents the lead compound for
further structural refinements because of the improved solubility.
Separately, examination of the right-hand phenyl group identified
6-pyridine derivative 6 with potent KSP inhibitory activity and
improved aqueous solubility.16 On the basis of these promising
components, two diaryl amine derivatives were then designed
(Fig. 4). Diaryl amine 13 was designed based on compound 12i
with a thiomorpholin-3-one structure and compound 6 with a
6-pyridine ring. Diaryl amine 14 was similarly designed based on
compounds 12l and 6.
4.1.2. General procedure of N-arylation for preparation of diaryl
amine compounds: synthesis of 6-{[3-
Thiomorpholin-3-one 13 exhibited the most potent KSP inhibi-
(trifluoromethoxy)phenyl]amino}-3,4-dihydroquinolin-2(1H)-
one (11i)
tory activity (IC50 = 0.035 lM) among the diaryl amine-type inhib-
itors that were examined in this study (Table 6). However,
Toluene (4.5 mL) was added to a flask containing 6-bromo-3,4-
dihydroquinolin-2(1H)-one (300 mg, 1.33 mmol), 3-(trifluorometh-
compound 13 was less soluble in 50% EtOH (0.669 mg/mL) and in
phosphate buffer (1.11 lg/mL) than the parent compound 6. The
high melting point of compound 13 (217 °C) indicated that the
tight crystal packing might cause this decrease in aqueous solubil-
ity. Benzoazepin-2-one 14 maintained highly potent KSP inhibitory
oxy)aniline (231
l
L, 1.73 mmol), Pd2(dba)3 (15.2 mg, 0.02 mmol),
2-dicyclohexylphosphino-20,40,60-triisopropylbiphenyl
(31.6 mg,
0.07 mmol) and NaOt-Bu (192 mg, 2.00 mmol) under an argon
atmosphere. The mixture was stirred at 100 °C for 9 h. After cooling,
activity (IC50 = 0.050 lM) as seen for the parent compounds 5a, 6

3176
T. Takeuchi et al. / Bioorg. Med. Chem. 22 (2014) 3171–3179
Table 6
KSP inhibitory activities and physicochemical properties of diaryl amine derivatives 5a, 6, 12i,l, 13 and 14
H
H
N
O
H
N
O
N
O
N
H
CF3
S
N
H
CF3
N
H
CF3
5a
12i
12l
KSP ATPase
IC50
M)a,b
(l
0.045
1.80
<1
190
4.2
0.051
1.70
<1
166
4.0
0.050
7.39
<1
140
4.4
Solubility in 50% EtOH (mg/mL)c
Solubility in phosphate buffer (pH 7.4) (
melting point (°C)
lg/mL)
ClogPd
HPLC retention time (min)e
24.4
28.2
27.0
H
N
H
N
H
N
O
O
O
N
N
N
N
H
CF3
S
N
H
CF3
N
H
CF3
6
13
14
KSP ATPase
IC50
(
l
M)a,b
0.068
3.51
6.12
177
3.4
0.035
0.669
1.11
217
3.2
13.8
0.050
4.82
8.07
185
3.6
10.8
Solubility in 50% EtOH (mg/mL)c
Solubility in phosphate buffer (pH 7.4) (
melting point (°C)
lg/mL)
ClogPd
HPLC retention time (min)e
7.0
a
b
c
Inhibition of microtubule-activated KSP ATPase activity.
IC50 values were derived from the dose–response curves generated from triplicate data points.
Solubility in 50% EtOH [an equal volume of EtOH and 1/15 M phosphate buffer (pH 7.4)].
CLogP values were calculated with ChemBioDraw Ultra 12.0.
HPLC analysis was carried out on a Cosmosil 5C18-ARII column (4.6 ꢂ 250 mm) and the material eluted by a linear MeCN gradient (30–70% over 40 min) in 0.1% TFA; flow
d
e
rate of 1 mL/min.
for 2 h. After cooling, the reaction mixture was diluted with EtOAc,
and filtered through a pad of Celite. The filtrate was concentrated
in vacuo. Crude material was purified by flash chromatography with
n-hexane/EtOAc (3:1) to afford compound 8b (165 mg, 97%): brown
oil; IR (neat) cmꢃ1: 3035 (NH), 3377 (NH); 1H NMR (400 MHz,
DMSO-d6) d 4.96 (s, 2H; NH2), 6.19 (d, J = 8.0 Hz, 1H; Ar), 6.28 (d,
J = 8.0 Hz, 1H; Ar), 6.37 (s, 1H; Ar), 6.92 (t, J = 8.0 Hz, 1H; Ar), 7.01
(d, J = 8.0 Hz, 1H; Ar), 7.22 (s, 1H; Ar), 7.25 (d, J = 8.2 Hz, 1H; Ar),
7.37 (t, J = 8.0 Hz, 1H; Ar), 8.13 (s, 1H; NH); 13C NMR (125 MHz,
DMSO-d6) d 104.0, 106.5, 107.8, 111.3, 114.3, 118.8, 124.3 (q),
129.6, 129.8 (q), 130.0, 142.5, 145.2, 149.7; HRMS (FAB): m/z calcd
for C13H11F3N2 (M+) 252.0874; found: 252.0874.
Table 7
Inhibitory effects on cell proliferation of diaryl amine-type KSP inhibitors toward
A549, HCT-116 and MCF-7
a
Compound
IC50
(
l
M)
A549
HCT-116
MCF-7
5a
6
12i
12l
13
14
4.2
2.5
4.1
4.5
1.5
5.0
8.9
4.4
6.8
5.7
2.6
6.5
11
8.0
6.5
3.9
2.8
9.1
a
IC50 values were derived from the dose–response curves generated from trip-
licate data points.
4.1.4. 1-(3-{[3-(Trifluoromethyl)phenyl]amino}phenyl)urea (8c)
To a stirred solution of N1-[3-(trifluoromethyl)phenyl]benzene-
1,3-diamine 8b (80.0 mg, 0.32 mmol) in AcOH (4.0 mL) was added
the reaction mixture was diluted with EtOAc, and filtered through a
pad of Celite. The filtrate was concentrated in vacuo. Crude material
was purified by flash chromatography with n-hexane/EtOAc (2:3) to
afford the desired diaryl amine 11i (292 mg, 68% yield): pale yellow
solid; mp 160–162 °C; IR (neat) cmꢃ1: 1667 (C@O), 3219 (NH), 3315
(NH); 1H NMR (500 MHz, DMSO-d6) d 2.43 (t, J = 6.9 Hz, 2H; CH2),
2.85 (t, J = 6.9 Hz, 2H; CH2), 6.63 (d, J = 8.0 Hz, 1H; Ar), 6.79 (s, 1H;
Ar), 6.81 (d, J = 8.0 Hz, 1H; Ar), 6.91–6.96 (m, 3H; Ar), 7.25 (t,
J = 8.0 Hz, 1H; Ar), 8.25 (s, 1H; NH), 9.99 (s, 1H; NH); 13C NMR
(125 MHz, DMSO-d6) d 25.0, 30.4, 106.3, 109.6, 113.1, 115.8, 118.7,
119.5, 120.1 (q), 124.7, 130.7, 133.1, 136.1, 146.9, 149.4, 169.8; Anal.
Calcd for C16H13F3N2O2: C, 59.63; H, 4.07; N, 8.69. Found: C, 59.51; H,
4.12; N, 8.59.
KOCN (77.2 mg, 0.95 mmol) and water (80.0 lL). The reaction mix-
ture was stirred at room temperature for 18 h, then evaporated to
dryness under vacuum. Crude material was purified by flash chro-
matography with amino silica gel with CHCl3/MeOH (20:1 to 10:1)
to afford compound 8c (31.9 mg, 34% yield): yellow oil; IR (neat)
cmꢃ1: 1661 (C@O), 3225 (NH), 3351 (NH); 1H NMR (500 MHz,
DMSO-d6) d 5.82 (br, 2H; NH2), 6.66 (d, J = 8.0 Hz, 1H; Ar), 6.84
(d, J = 8.0 Hz, 1H; Ar), 7.06 (d, J = 8.0 Hz, 1H; Ar), 7.12 (t,
J = 8.0 Hz, 1H; Ar), 7.28 (s, 1H; Ar), 7.30 (d, J = 8.0 Hz, 1H; Ar),
7.41 (s, 1H; Ar), 7.42 (t, J = 8.0 Hz, 1H; Ar), 8.50 (br, 2H; NH); 13C
NMR (125 MHz, DMSO-d6) d 107.4, 110.7, 110.9, 111.5, 114.8,
118.9, 124.2 (q), 129.3, 129.8 (q), 130.1, 141.5, 142.2, 144.6,
155.8; HRMS (FAB): m/z calcd for C14H12F3N3O (M+) 295.0932;
found: 295.0926.
4.1.3. N1-[3-(Trifluoromethyl)phenyl]benzene-1,3-diamine (8b)
To a stirred solution of 3-nitro-N-[3-(trifluoromethyl)phenyl]
aniline 8a (190 mg, 0.67 mmol) and ammonium formate (509 mg,
8.07 mmol) in EtOH (2.2 mL) at room temperature was added 10%
Pd/C (35.8 mg, 0.03 mmol). The mixture was heated under reflux
4.1.5. 6-[(3-Hydroxyphenyl)amino]-3,4-dihydroquinolin-2(1H)-
one (11j)
A
suspension of 6-[(3-methoxyphenyl)amino]-3,4-dihydro-
quinolin-2(1H)-one 11h (500 mg, 1.86 mmol) in dry CH2Cl2

T. Takeuchi et al. / Bioorg. Med. Chem. 22 (2014) 3171–3179
3177
(5 mL) was cooled to ꢃ78 °C and then BBr3 (7.45 mL of 1 M solu-
4.1.8. 7-{[3-(Trifluoromethyl)phenyl]amino}quinolin-2(1H)-
one (12c)
AcOH (1.0 mL) was added to a flask containing 7-{[3-(trifluoro-
methyl)phenyl]amino)}-3,4-dihydroquinolin-2(1H)-one 12a
tion in CH2Cl2, 7.45 mmol) was added. After the mixture was stir-
red for 30 min at ꢃ78 °C, the stirring was continued for additional
18 h at room temperature. The reaction was quenched by addition
of water (12 mL), and CH2Cl2 was removed under reduced pres-
sure. The aqueous solution was neutralized by addition of aqueous
NaOH, and then extracted three times with EtOAc. The organic
layer was washed with brine, and dried over Na2SO4. The organic
solvent was removed under reduced pressure and the crude resi-
due was purified by flash chromatography with n-hexane/EtOAc
(1:3) to afford compound 11j (232 mg, 49% yield): brown needle
crystal; mp 213–214 °C; IR (neat) cmꢃ1: 1651 (C@O), 3222 (NH),
3321 (NH); 1H NMR (500 MHz, DMSO-d6) d 2.41 (t, J = 7.4 Hz, 2H;
CH2), 2.81 (t, J = 7.4 Hz, 2H; CH2), 6.15 (d, J = 8.0 Hz, 1H; Ar), 6.38
(d, J = 8.0 Hz, 1H; Ar), 6.39 (s, 1H; Ar), 6.74 (d, J = 8.0 Hz, 1H; Ar),
6.85 (d, J = 8.0 Hz, 1H; Ar), 6.89 (s, 1H; Ar), 6.94 (t, J = 8.0 Hz, 1H;
Ar), 7.77 (s, 1H; NH), 9.07 (s, 1H; OH), 9.89 (s, 1H; NH); 13C NMR
(125 MHz, DMSO-d6) d 25.1, 30.5, 102.2, 106.0, 106.6, 115.6,
117.4, 118.2, 124.4, 129.7, 131.7, 137.7, 145.8, 158.1, 169.7; Anal.
Calcd for C15H14N2O2: C, 70.85; H, 5.55; N, 11.02. Found: C,
71.11; H, 5.59; N, 10.89.
(80.0 mg, 0.26 mmol) and Pd(OAc)2 (22.4 mg, 0.10 mmol) and an
oxygen balloon was connected to the reaction vessel. After stirring
for 2 h at 115 °C, the reaction mixture was cooled to room temper-
ature and concentrated in vacuo. The residue was purified by flash
chromatography with n-hexane/EtOAc (1:3 to 1:5) to afford 12c
(34.1 mg, 43% yield): white solid; mp 177–179 °C; IR (neat)
cmꢃ1: 1655 (C@O), 3452 (NH); 1H NMR (500 MHz, DMSO-d6) d
6.24 (d, J = 9.2 Hz, 1H; C=CH), 6.88 (dd, J = 8.0, 2.3 Hz, 1H; Ar),
7.06 (d, J = 2.3 Hz, 1H; Ar), 7.23 (d, J = 8.0 Hz, 1H; Ar), 7.41 (s, 1H;
Ar), 7.45 (d, J = 8.0 Hz, 1H; Ar), 7.50–7.53 (m, 2H; Ar), 7.75 (d,
J = 9.2 Hz, 1H; C=CH), 8.99 (s, 1H; NH), 11.53 (s, 1H; NH); 13C
NMR (125 MHz, DMSO-d6) d 99.7, 112.3, 113.1, 113.5, 116.7,
117.7, 121.0, 124.1 (q), 129.0, 130.0 (q), 130.4, 139.8, 140.5,
142.9, 144.5, 162.2; HRMS (FAB): m/z calcd for C16H11F3N2O
[M+H]+ 305.0902; found: 305.0905.
4.1.9. 6-{[3-(Trifluoromethyl)phenyl]amino}-3,4-
dihydroquinoline-2(1H)-thione (12k)
To a stirred solution of 6-{[3-(trifluoromethyl)phenyl]amino}-
3,4-dihydroquinolin-2(1H)-one 5a (50.0 mg, 0.16 mmol) in toluene
(1.0 mL) was added Lawesson’s reagent (33.0 mg, 0.08 mmol)
under an argon atmosphere. After stirring for 30 min at 110 °C,
the reaction mixture was cooled to room temperature and concen-
trated in vacuo. The residue was purified by flash chromatography
with n-hexane/EtOAc (3:1) to afford 12k (51.4 mg, 100% yield):
yellow solid; mp 195–196 °C; IR (neat) cmꢃ1: 1499 (C@S), 3361
(NH); 1H NMR (500 MHz, DMSO-d6) d 2.77 (t, J = 8.0 Hz, 2H; CH2),
2.90 (t, J = 8.0 Hz, 2H; CH2), 6.97 (s, 1H; Ar), 6.98 (d, J = 8.0 Hz,
1H; Ar), 7.04 (d, J = 8.0 Hz, 1H; Ar), 7.06 (d, J = 8.0 Hz, 1H; Ar),
7.20 (s, 1H; Ar), 7.27 (d, J = 8.0 Hz, 1H; Ar), 7.41 (d, J = 8.0 Hz, 1H;
Ar), 8.49 (s, 1H; NH), 12.15 (s, 1H; NH); 13C NMR (125 MHz,
DMSO-d6) d 24.0, 38.9, 111.4, 114.9, 117.0, 117.3, 118.3, 118.5,
124.3 (q), 126.6, 130.0 (q), 130.3, 131.3, 138.3, 144.9, 197.3; HRMS
(FAB): m/z calcd for C16H13F3N2S (M+) 322.0752; found: 322.0758.
4.1.6. 6-[(3-Aminophenyl)amino]-3,4-dihydroquinolin-2(1H)-
one (11l)
To a stirred solution of 6-[(3-nitrophenyl)amino]-3,4-dihydro-
quinolin-2(1H)-one 11k (62.0 mg, 0.22 mmol) in AcOH (2.2 mL)
at room temperature was added zinc powder (102 mg, 1.56 mmol)
portionwise. After being stirred at room temperature for 1 h, the
reaction mixture was filtered through a pad of Celite and concen-
trated under vacuum. The residue was diluted with EtOAc, and
the whole was washed with saturated NaHCO3, brine, and dried
over Na2SO4. The organic solvent was removed under reduced
pressure and the crude residue was purified by flash chromatogra-
phy with n-hexane/EtOAc (1:8) to afford compound 11l (41.6 mg,
75% yield): pale yellow solid; mp 196–198 °C; IR (neat) cmꢃ1
:
1662 (C@O), 3221 (NH), 3344 (NH); 1H NMR (500 MHz, DMSO-
d6) d 2.40 (t, J = 7.4 Hz, 2H; CH2), 2.80 (t, J = 7.4 Hz, 2H; CH2), 4.93
(br, 2H; NH2), 6.00 (d, J = 8.0 Hz, 1H; Ar), 6.15 (d, J = 8.0 Hz, 1H;
Ar), 6.24 (s, 1H; Ar), 6.72 (d, J = 8.0 Hz, 1H; Ar), 6.80 (d, J = 8.0 Hz,
1H; Ar), 6.83 (d, J = 8.0 Hz, 1H; Ar), 6.86 (s, 1H; Ar), 7.58 (s, 1H;
NH), 9.87 (s, 1H; NH); 13C NMR (125 MHz, DMSO-d6) d 25.2, 30.5,
101.4, 104.5, 105.7, 115.5, 116.9, 117.7, 124.3, 129.3, 131.2,
138.3, 145.0, 149.2, 169.6; HRMS (FAB): calcd for C15H15N3O (M+)
253.1215; found: 253.1213.
4.1.10. 7-{[4-(Trifluoromethyl)pyridin-2-yl]amino}-2H-
benzo[b][1,4]thiazin-3(4H)-one (13)
Following the general procedure for 11i, compound 13 (48.5 mg,
12% yield) was synthesized from 7-bromo-2H-benzo[b][1,4]thiazin-
3(4H)-one and 2-amino-4-(trifluoromethyl)pyridine: white solid;
mp 217–218 °C; IR (neat) cmꢃ1: 1688 (C@O), 3194 (NH); 1H NMR
(500 MHz, DMSO-d6) d 3.45 (s, 2H; CH2), 6.92 (d, J = 8.6 Hz, 1H;
Ar), 6.99 (d, J = 5.2 Hz, 1H; Ar), 7.03 (s, 1H; Ar), 7.33 (dd, J = 8.6,
1.7 Hz, 1H; Ar), 7.81 (d, J = 1.7 Hz, 1H; Ar), 8.37 (d, J = 5.2 Hz, 1H;
Ar), 9.43 (s, 1H; NH), 10.44 (s, 1H; NH); 13C NMR (125 MHz,
DMSO-d6) d 29.1, 106.2, 108.6, 117.1, 117.5, 117.7, 119.3, 123.0
(q), 131.7, 135.9, 137.6 (q), 149.3, 156.1, 164.8; HRMS (FAB): m/z
calcd for C14H10F3N3OS (M+) 325.0497; found: 325.0497.
4.1.7. 3-[(2-Oxo-1,2,3,4-tetrahydroquinolin-6-yl)amino]benzoic
acid (11n)
To a solution of methyl 3-[(2-oxo-1,2,3,4-tetrahydroquinolin-6-
yl)amino]benzoate 11m (400 mg, 1.35 mmol) in 5.2 mL of MeOH/
H2O (3:1 v/v) was added LiOHꢀH2O (170 mg, 4.05 mmol) at 0 °C,
then the solution was warmed to 50 °C. After 1 h, the reaction mix-
ture was acidified to below pH 2 using 1 M HCl, then EtOAc and
brine were added to the mixture. The organic extracts were
washed with brine and dried over Na2SO4. The organic solvent
was removed under reduced pressure to afford compound 11n
4.1.11. 7-{[4-(Trifluoromethyl)pyridin-2-yl]amino}-1,3,4,5-
tetrahydro-2H-benzo[b]azepin-2-one (14)
Following the general procedure for 11i, compound 14
(16.7 mg, 18% yield) was synthesized from 7-bromo-1,3,4,5-tetra-
(366 mg, 96% yield): white solid; mp 259–261 °C; IR (neat) cmꢃ1
:
1656 (C@O), 1684 (C@O), 3203 (NH), 3326 (NH); 1H NMR
(500 MHz, DMSO-d6) d 2.43 (t, J = 7.4 Hz, 2H; CH2), 2.85 (t,
J = 7.4 Hz, 2H; CH2), 6.81 (d, J = 8.0 Hz, 1H; Ar), 6.93 (d, J = 8.0 Hz,
1H; Ar), 6.94 (s, 1H; Ar), 7.17 (d, J = 8.0 Hz, 1H; Ar), 7.27–7.31 (m,
2H; Ar), 7.53 (s, 1H; Ar), 8.12 (s, 1H; NH), 9.98 (s, 1H; NH), 12.78
(br, 1H; CO2H); 13C NMR (125 MHz, DMSO-d6) d 25.1, 30.4, 115.5,
115.8, 118.1, 118.9, 119.0, 119.2, 124.6, 129.3, 131.7, 132.6,
136.9, 145.2, 167.6, 169.8; HRMS (FAB): m/z calcd for C16H14N2O3
(M+) 282.1004; found: 282.1011.
hydro-2H-benzo[b]azepin-2-one
and
2-amino-4-(trifluoro-
methyl)pyridine: pale yellow solid; mp 185–187 °C; IR (neat)
cmꢃ1: 1688 (C@O), 2936 (NH); 1H NMR (500 MHz, DMSO-d6) d
2.09–2.16 (m, 4H; CH2 ꢂ 2), 2.67 (t, J = 6.9 Hz, 2H; CH2), 6.92 (d,
J = 8.0 Hz, 1H; Ar), 6.98 (d, J = 5.2 Hz, 1H; Ar), 7.07 (s, 1H; Ar),
7.53 (d, J = 8.0 Hz, 1H; Ar), 7.55 (s, 1H; Ar), 8.37 (d, J = 5.2 Hz, 1H;
Ar), 9.36 (s, 1H; NH), 9.42 (s, 1H; NH); 13C NMR (125 MHz,
DMSO-d6) d 27.8, 30.1, 32.7, 106.0, 108.4, 117.4, 119.6, 121.9,

3178
T. Takeuchi et al. / Bioorg. Med. Chem. 22 (2014) 3171–3179
122.0, 123.0 (q), 132.7, 134.1, 137.4, 137.5 (q), 149.3, 156.3, 173.1;
HRMS (FAB): m/z calcd for C16H14F3N3O (M+) 321.1089; found:
321.1092.
rates in each condition were evaluated to determine IC50 values
using the GraphPad Prism software.
4.3. Growth inhibition assay
4.1.12. 6-[3-(Trifluoromethyl)phenoxy]-3,4-dihydroquinolin-
2(1H)-one (17a)
A549, HCT-116 and MCF-7 cells were cultured in Dulbecco’s
modified Eagle’s medium (DMEM, Sigma), McCoy’s 5A medium
(GIBCO) and Eagle’s minimal essential medium (EMEM, Wako),
respectively, supplemented with 10% (v/v) fetal bovine serum at
37 °C in a 5% CO2-incubator. Growth inhibition assays using these
cells were performed in 96-well plates (BD Falcon). A549, HCT-116
and MCF-7 cells were seeded at 500, 5000 and 5000 cells/well in
To a suspension of KOt-Bu (60.5 mg, 0.54 mmol) in DMF
(2.1 mL) was added 6-hydroxy-3,4-dihydroquinolin-2(1H)-one 15
(80.0 mg, 0.49 mmol) at 0 °C and the reaction mixture was stirred
at this temperature for 15 min. Bis(3-trifluoromethylphenyl)iodo-
nium tetrafluoroborate 16 (259 mg, 0.51 mmol) was added in one
portion and the reaction mixture was stirred at 40 °C for 1 h, then
quenched with H2O at 0 °C and extracted into CHCl3. The organic
layer was washed with brine, and dried over Na2SO4. The organic
solvent was removed under reduced pressure and the crude resi-
due was purified by flash chromatography with n-hexane/EtOAc
(1:1) to afford compound 17a (39.9 mg, 27% yield): white solid;
mp 158–159 °C; IR (neat) cmꢃ1: 1681 (C@O), 3203 (NH); 1H NMR
(500 MHz, DMSO-d6) d 2.44 (t, J = 7.4 Hz, 2H; CH2), 2.87 (t,
J = 7.4 Hz, 2H; CH2), 6.91 (s, 2H; Ar), 6.99 (s, 1H; Ar), 7.20–7.23
(m, 2H; Ar), 7.42 (d, J = 8.0 Hz, 1H; Ar), 7.58 (t, J = 8.0 Hz, 1H; Ar),
10.12 (s, 1H; NH); 13C NMR (125 MHz, DMSO-d6) d 24.7, 30.0,
113.6, 116.3, 118.7, 119.1, 119.5, 121.0, 123.7 (q), 125.7, 130.6
(q), 131.2, 135.3, 149.6, 158.4, 169.9; HRMS (FAB): m/z calcd for
50
Chemical compounds in DMSO were diluted 250-fold with the cul-
ture medium in advance. Following the addition of 40 L of the
fresh culture medium to the cell cultures, 30 L of the chemical dil-
lL of culture media, respectively, and were cultured for 6 h.
l
l
uents were also added. The final volume of DMSO in the medium
was equal to 0.1% (v/v). The cells under chemical treatment were
incubated for further 72 h. The wells in the plates were washed
twice with the cultured medium without phenol-red. After 1 h
incubation with 100
lL of the medium, the cell culture in each well
was supplemented with 20
lL of the MTS reagent (Promega), fol-
lowed by incubation for additional 40 min. Absorbance at 490 nm
of each well was measured using a Wallac 1420 ARVO SX multila-
bel counter (Perkin Elmer). At least three experiments were per-
formed per condition and the averages and standard deviations
of inhibition rates in each condition were evaluated to determine
IC50 values using the GraphPad Prism software.
C
16H12F3NO2 (M+) 307.0820; found: 307.0826.
4.1.13. 6-{[3-(Trifluoromethyl)phenyl]thio}-3,4-
dihydroquinolin-2(1H)-one (20a)
2-Propanol (1.0 mL) was added to a flask containing 6-mer-
capto-3,4-dihydroquinolin-2(1H)-one 18 (50.0 mg, 0.28 mmol), 3-
4.4. Thermodynamic solubility in aqueous solution
trifluoromethyliodobenzene 19a (40.4
lL, 0.28 mmol), CuI
(32.0 mg, 0.17 mmol), ethylene glycol (31.2
l
L, 0.56 mmol) and
An equal volume of a mixture of 1/15 M phosphate buffer (pH
7.4, 0.5 mL) and EtOH (0.5 mL), or 1/15 M phosphate buffer (pH
7.4, 1.0 mL) was added to a compound in a vial. The suspension
was then shaken for 48 h at 25 °C, and undissolved material was
separated by filtration. m-Cresol was added as an internal standard
(final concentration: 0.05 mg/mL) and the mixture was diluted in
DMF and injected onto the HPLC column. The peak area ratio of
the sample to the standard was recorded by UV detection at
254 nm. The concentration of the sample solution was calculated
using a previously determined calibration curve, corrected for the
dilution factor of the sample.
K2CO3 (96.7 mg, 0.70 mmol) under an argon atmosphere. The mix-
ture was stirred at 80 °C for 13 h. After cooling, the reaction mix-
ture was diluted with EtOAc, and filtered through a pad of Celite.
The filtrate was concentrated in vacuo. Crude material was purified
by flash chromatography with n-hexane/EtOAc (3:2) to afford com-
pound 20a (6.2 mg, 7% yield): white solid; mp 102–104 °C; IR
(neat) cmꢃ1: 1676 (C@O), 3387 (NH); 1H NMR (500 MHz, DMSO-
d6) d 2.53 (t, J = 7.4 Hz, 2H; CH2), 2.96 (t, J = 7.4 Hz, 2H; CH2), 7.00
(d, J = 8.0 Hz, 1H; Ar), 7.40 (d, J = 8.0 Hz, 1H; Ar), 7.42–7.45 (m,
2H; Ar), 7.49 (s, 1H; Ar), 7.58–7.59 (m, 2H; Ar), 10.34 (s, 1H;
NH); 13C NMR (125 MHz, DMSO-d6) d 24.4, 30.0, 116.4, 122.4,
122.6, 123.0, 123.7 (q), 125.4, 129.9 (q), 130.3, 131.0, 133.5,
133.7, 139.5, 140.1, 170.1; HRMS (FAB): m/z calcd for C16H12F3NOS
(M+) 323.0592; found: 323.0591.
4.5. Molecular modeling
Docking calculations for compound 11d were performed using a
similar protocol in our previous research16 based on the crystal
structure of the KSP-inhibitor complex (PDB ID: 3ZCW).27 The pro-
tonation states of the amino acid residues of KSP and the direction
of the hydrogen atoms involved in the hydrogen bonds were
assigned using the Protonate3D algorithm28 implemented in
4.2. KSP ATPase assay
The microtubules-stimulated KSP ATPase reaction was per-
formed in a reaction buffer [20 mM PIPES-KOH (pH 6.8), 25 mM
KCl, 2 mM MgCl2, 1 mM EGTA-KOH (pH 8.0)] containing 38 nM of
the bacteria-expressed KSP motor domain (1–369) fused to histi-
dine-tag at the carboxyl-terminus and 350 nM microtubules in
96-well half-area plates (Corning). Each chemical compound in
DMSO at different concentrations was diluted 12.5-fold with the
chemical dilution buffer [10 mM Tris-OAc (pH 7.4), 0.04% (v/v)
MOE.29 The
a4/a6 allosteric site was chosen from the binding sites
detected by the MOE-SiteFinder module. Docking pose generation
was performed applying pharmacophore restraint to form
hydrogen bonds with Asn271 and Leu292. The 100 initial docked
candidate poses were optimized by the MMFF94x forcefield30
and the pose with the lowest binding energy (Ebind) estimated by
the MM/GBVI method31 was adopted as the predicted binding
mode.
NP-40]. After pre-incubation of 9.7
3.8 L of each chemical solution at 25 °C for 30 min, the ATPase
reaction was initiated by the addition of 1.5 L of 0.3 mM ATP solu-
tion, and followed by incubation at 25 °C for further 15 min. The
reaction was terminated by the addition of 15 L of the Kinase-
lL of the enzyme solution with
l
l
Acknowledgements
l
Glo Plus reagent (Promega). The ATP consumption in each reaction
was measured as the luciferase-derived luminescence by ARVO
Light (PerkinElmer). At least three experiments were performed
per condition and the averages and standard deviations of inhibition
This work was supported by Grants-in-Aid for Scientific
Research, and Platform for Drug Discovery, Informatics, and Struc-
tural Life Science from MEXT, Japan. T.T. and R.M. are grateful for
JSPS Research Fellowships for Young Scientists.

T. Takeuchi et al. / Bioorg. Med. Chem. 22 (2014) 3171–3179
3179
14. Oishi, S.; Watanabe, T.; Sawada, J.; Asai, A.; Ohno, H.; Fujii, N. J. Med. Chem.
2010, 53, 5054.
Supplementary data
15. Takeuchi, T.; Oishi, S.; Watanabe, T.; Ohno, H.; Sawada, J.; Matsuno, K.; Asai, A.;
Asada, N.; Kitaura, K.; Fujii, N. J. Med. Chem. 2011, 54, 4839.
16. Takeuchi, T.; Oishi, S.; Kaneda, M.; Ohno, H.; Nakamura, S.; Nakanishi, I.;
Yamane, M.; Sawada, J.; Asai, A.; Fujii, N. ACS Med. Chem. Lett. doi: http://
dx.doi.org/10.1021/ml500016j.
17. Yan, Y.; Sardana, V.; Xu, B.; Homnick, C.; Halczenko, W.; Buser, C. A.; Schaber,
M.; Hartman, G. D.; Huber, H. E.; Kuo, L. C. J. Mol. Biol. 2004, 335, 547.
18. Kaan, H. Y. K.; Ulaganathan, V.; Hackney, D. D.; Kozielski, F. Biochem. J. 2010,
425, 55.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2014.04.008.
References and notes
1. Miki, H.; Okada, Y.; Hirokawa, N. Trends Cell Biol. 2005, 15, 467.
2. Wordeman, L. Semin. Cell Dev. Biol. 2010, 21, 260.
3. Hirokawa, N.; Noda, Y.; Tanaka, Y.; Niwa, S. Nat. Rev. Mol. Cell Biol. 2009, 10,
682.
19. Barsanti, P. A.; Wang, W.; Ni, Z.; Duhl, D.; Brammeier, N.; Martin, E.; Bussiere,
D.; Walter, A. O. Bioorg. Med. Chem. Lett. 2010, 20, 157.
20. For a recent review, see: Ishikawa, M.; Hashimoto, Y. J. Med. Chem. 2011, 54, 1539.
21. Anderson, K. W.; Tundel, R. E.; Ikawa, T.; Altman, R. A.; Buchwald, S. L. Angew.
Chem., Int. Ed. 2006, 45, 6523.
4. Kashina, A. S.; Baskin, R. J.; Cole, D. G.; Wedaman, K. P.; Saxton, W. M.; Scholey,
J. M. Nature 1996, 379, 270.
5. Sawin, K. E.; LeGuellec, K.; Philippe, M.; Mitchison, T. J. Nature 1992, 359, 540.
6. Blangy, A.; Lane, H. A.; d’Herin, P.; Harper, M.; Kress, M.; Nigg, E. A. Cell 1995,
83, 1159.
7. Walczak, C. E.; Vernos, I.; Mitchison, T. J.; Karsenti, E.; Heald, R. Curr. Biol. 1998,
8, 903.
8. Mayer, T. U.; Kapoor, T. M.; Haggarty, S. J.; King, R. W.; Schreiber, S. L.;
Mitchison, T. J. Science 1999, 286, 971.
22. Avdeef, A.; Testa, B. Cell. Mol. Life Sci. 2002, 59, 1681.
23. Jalalian, N.; Ishikawa, E. E.; Silva, L. F., Jr.; Olofsson, B. Org. Lett. 2011, 13, 1552.
24. Kwong, F. Y.; Buchwald, S. L. Org. Lett. 2002, 4, 3517.
25. Cava, M. P.; Levinson, M. I. Tetrahedron 1985, 41, 5061.
26. The incompatible results between the in vitro KSP inhibition and cytotoxicity
could be attributed to the varied cell membrane permeability. The less
solubility of compound 13 in aqueous solution may be disadvantageous to the
in vivo studies.
9. Tao, W.; South, V. J.; Zhang, Y.; Davide, J. P.; Farrell, L.; Kohl, N. E.; Sepp-
Lorenzino, L.; Lobell, R. B. Cancer Cell 2005, 8, 49.
27. Ulaganathan, V.; Talapatra, S. K.; Rath, O.; Pannifer, A.; Hackney, D. D.;
Kozielski, F. J. Am. Chem. Soc. 2013, 135, 2263.
28. Labute, P. Proteins 2009, 75, 187.
29. MOE ver 2010.10, Chemical Computing Group Inc., Montreal, Canada.
30. Halgren, T. A.; Nachbar, R. B. J. Comput. Chem. 1996, 17, 587.
31. Labute, P. J. Comput. Chem. 2008, 29, 1693.
10. For a review, see: Jackson, J. R.; Patrick, D. R.; Dar, M. M.; Huang, P. S. Nat. Rev.
Cancer 2007, 7, 107.
11. For a review, see: Matsuno, K.; Sawada, J.; Asai, A. Expert Opin. Ther. Patent
2008, 18, 253.
12. For a review, see: Sarli, V.; Giannis, A. Clin. Cancer Res. 2008, 14, 7583.
13. For a review, see: Rath, O.; Kozielski, F. Nat. Rev. Cancer 2012, 12, 527.