Structure-activity relationships of carboline and carbazole derivatives as a novel class of ATP-competitive kinesin spindle protein inhibitors

Article abstract of DOI:10.1021/jm200448n

The kinesin spindle protein (KSP) is a mitotic kinesin involved in the establishment of a functional bipolar mitotic spindle during cell division. It is considered to be an attractive target for cancer chemotherapy with reduced side effects. Based on natu

Full text of DOI:10.1021/jm200448n

ARTICLE
pubs.acs.org/jmc
StructureꢀActivity Relationships of Carboline and Carbazole
Derivatives as a Novel Class of ATP-Competitive Kinesin Spindle
Protein Inhibitors
Tomoki Takeuchi,† Shinya Oishi,*,† Toshiaki Watanabe,† Hiroaki Ohno,† Jun-ichi Sawada,‡ Kenji Matsuno,‡
Akira Asai,‡ Naoya Asada,† Kazuo Kitaura,† and Nobutaka Fujii*,†
†Graduate School of Pharmaceutical Sciences, Kyoto University, Sakyo-ku, Kyoto 606-8501, Japan
‡Graduate School of Pharmaceutical Sciences, University of Shizuoka, Suruga-ku, Shizuoka 422-8526, Japan
S Supporting Information
b
ABSTRACT: The kinesin spindle protein (KSP) is a mitotic kinesin
involved in the establishment of a functional bipolar mitotic spindle
during cell division. It is considered to be an attractive target for cancer
chemotherapy with reduced side effects. Based on natural product
scaffold-derived fused indole-based inhibitors and known biphenyl-
type KSP inhibitors, various carboline and carbazole derivatives were
synthesized and biologically evaluated. β-Carboline and lactam-fused
carbazole derivatives exhibited remarkably potent KSP inhibitory
activity and mitotic arrest in prometaphase with formation of an
irregular monopolar spindle. The planar tri- and tetracyclic analogs
inhibited KSP ATPase in an ATP-competitive manner just like biphe-
nyl-type inhibitors.
’ INTRODUCTION
Many studies have been carried out on the development of
mitotic inhibitors. Most antimitotic agents (including taxanes
and various vinca alkaloids) inhibit the function of the β-tubulin
protein (a component of the microtubule), thereby showing
anticancer activity.1 However, the inhibition of microtubules
leads to severe peripheral neuropathy because microtubules are
also present in other nondividing cells (e.g., postmitotic neu-
rons) and possess other critical physiological functions in cellular
processes.2 Acquired drug resistance is another obstacle after the
long-term use of antimitotic agents. To overcome these limita-
tions, novel anticancer agents with fewer adverse effects against
other molecular targets are being developed.
The kinesin spindle protein (KSP; also known as Eg5) is a
Figure 1. Structures of the reported KSP inhibitors 1 and 2 and the
member of the kinesin-5 family. It plays an essential part in
potent KSP inhibitors with a 2,3-fused indole substructure.
the formation and maintenance of the bipolar spindle. The KSP
moves along antiparallel microtubules toward the plus end with
activity.7 Based on the common substructure of the known KSP
inhibitory terpendole E, 1, and HR22C16, 2 (Figure 1),8,9 the
ring-fused indoles were identified to be minimal scaffolds for KSP
inhibition. During the course of the study, it was also demon-
strated that the cytotoxic activity of β-carboline alkaloids harman
and harmine in part originated from KSP inhibitory activity.7
Although these natural products and natural product-related
energy from the hydrolysis of ATP, thereby separating centrosomes.3
Inhibition of the KSP leads to mitotic arrest in prometaphase with
irregular formation of the monopolar spindle and subsequent
apoptotic cell death.4 The KSP is abundantly expressed only in
proliferating human tissues and not presented in postmitotic
neurons in the human central nervous system, so the inhibitors
might work as therapeutic agents with better selectivity for the
treatment of malignant tumors.5,6
Recently, we reported that carbazoles 3aꢀc with a bulky alkyl
Received: April 14, 2011
Published: May 20, 2011
group at the 2- or 3-position exhibited potent KSP inhibitory
r
2011 American Chemical Society
4839
dx.doi.org/10.1021/jm200448n J. Med. Chem. 2011, 54, 4839–4846
|

Journal of Medicinal Chemistry
ARTICLE
Figure 2. Design of novel KSP inhibitors 9ꢀ13 with fused indole scaffolds such as carboline (A) and carbazoles (B, C).
Scheme 1. Synthesis of Carboline Derivativesa
compounds include a complex structure with cyclic and acyclic
accessory substituents, the identification of the core substructure
facilitated further structureꢀactivity relationship studies. Alter-
natively, KSP inhibitors with biaryl scaffolds such as 4ꢀ6 have
been reported (Figure 2).10 The biphenyls 5 and 6 bind to
the interface of helices R4 and R6 (which is remote from the
ATP binding site) to inhibit KSP activity in an ATP-competitive
manner.10b We assumed that carbazole derivatives 3 with KSP
inhibitory activity could be regarded as compounds with a
nitrogen atom bridged between the 2- and 20-carbons of biphe-
nyl. Restriction of the rotatable bond could configure the
coplanar conformation of biphenyl groups, as well as the appro-
priate orientations of the carbazole substituents, for binding to
KSP with subsequent reduced entropy loss. Accordingly, re-
ported herein is the design and synthesis of KSP inhibitors with
planar carbazole and carboline scaffolds. Enzymatic analysis of
a Reagents and conditions: (a) Pd(OAc)2, Xantphos, NaOt-Bu, toluene,
the potent derivatives is also described to investigate the poten-
tial binding and functional modes of action.
reflux; (b) Pd(OAc)2, PCy3 HBF4, K2CO3, t-BuCO2H, DMA, 130 °C;
3
(c) CuI, 1,10-phenanthroline, K2CO3, DMF, 110 °C.
Carboline derivatives 9aꢀd with a 7-trifluoromethyl group
were designed based on the 4-[4-(trifluoromethyl)phenyl]pyr-
idine 411 with moderate KSP inhibitory activity (Figure 2A).
β-Carboline 9b was designed to examine the restriction effect of
the rotatable bond between benzene and the pyridine ring of
compound 4, in which the 4-phenylpyridine framework
was included in the 6ꢀ5ꢀ6 tricyclic structure of β-carboline.
The arrangements of a pyridine nitrogen and a CF3 group of
γ-carboline 9c were presumably mimicked by the relative orienta-
tions of the pyridine nitrogen and the CF3 group of 4. Additional
R- and δ-carbolines 9a,d were the alternative regioisomers,
which were incompatible with the conformations of 4-phenyl-
pyridine derivative 4.
’ RESULTS AND DISCUSSION
Design of Potent KSP Inhibitors with a Ring-Fused Indole
Scaffold. Two possible orientations of the accessory groups on
the carbazole core were deliberated, when we focused on the
rotatable biaryl CꢀC bond of the 1-aryl-4-(trifluoromethyl)-
benzene scaffold in 4ꢀ6. Bidirectional orientations of 7 at the
carbazole 2- and 7-positions were designed from ring-closing via
one-nitrogen linkage between biphenyl 2- and 20-positions.
Reproduction of the outward-facing orientations of two acces-
sory groups in the biaryl compounds 4ꢀ6 could provide the 3,7-
disubstituted carbazoles 8.
4840
dx.doi.org/10.1021/jm200448n |J. Med. Chem. 2011, 54, 4839–4846

Journal of Medicinal Chemistry
ARTICLE
Scheme 2. Synthesis of Carbazole Derivatives with a
Lactam-Fused Structurea
Scheme 3. Synthesis of Carbazole Derivatives with a
Five-Membered Lactam Ring or Pyrrole-Fused Structurea
a Reagents and conditions: (a) 4-bromo-3-nitrobenzotrifluoride, Pd-
(OAc)2, XPhos, K2CO3, 1,4-dioxane, 60 °C; (b) 4-(dimethylamino)-
triphenylphosphine, o-DCB, 160 °C; (c) PCC, 1,2-DCE, 80 °C; (d)
4-(dimethylamino)triphenylphosphine, o-DCB, 140 °C.
For carbazoles 10a,b with a six-membered lactam, aryl bro-
mide 20a was initially converted to pinacolboronate 21a14 by a
palladium-catalyzed coupling reaction with bis(pinacolato)di-
boron.15 The subsequent SuzukiꢀMiyaura cross-coupling of
21a led to the 2-nitrobiphenyl derivative 22a (Scheme 2A).
The carbazoles 10a,b were obtained by triphenylphosphine-
mediated reductive cyclization of 22a.16 Carbazoles 11a,b with
a seven-membered lactam were also synthesized in a similar
procedure. The other lactam-fused carbazole 10c was obtained
by palladium-catalyzed N-arylation of lactam 20a and m-CF3-
substituted aniline 23 followed by oxidative biaryl coupling
(Scheme 2B).17 For the preparation of carbazole 10d, phenol
derivative 25 was initially converted to the corresponding
trifluoromethanesulfonate 26 (Scheme 2C).18 Next, organotin
compound 27 was obtained by treatment of 26 with bis-
(tributyltin) in the presence of a palladium catalyst and LiCl,19
and subsequent Stille coupling with aryl bromide provided
compound 28.20 The desired carbazole 10d was prepared using
identical conditions for the reductive cyclization.
The synthesis of carbazole 12 with a five-membered lactam
was undertaken via indole derivative 30, which was prepared
using SuzukiꢀMiyaura coupling from commercially available
boronic acid 29 (Scheme 3). Compound 30 was converted to
carbazole 31 by phosphine-mediated reductive cyclization. How-
ever, separation of the byproduct triphenylphosphine oxide from
the product 31 by repeated column chromatography failed.
When 4-(dimethylamino)triphenylphosphine was used for the
reducing reagent, the purification of carbazole 31 was readily
achieved by a single round of column chromatography. Unfortu-
nately, our efforts to obtain the desired carbazole 12 upon
treatment with various oxidative reagents resulted only in the
recovery or decomposition of 31. As an alternative, an isatin
derivative 32 was prepared by the oxidation of compound 30
using pyridinium chlorochromate (PCC).21 Interestingly, when
compound 32 was treated with 4-(dimethylamino)triphenyl-
phosphine, the carbonyl group at the isatin 3-position was reduced
a Reagents and conditions: (a) (pinB)2, PdCl2(dppf), KOAc, DMSO,
80 °C; (b) 4-bromo-3-nitrobenzotrifluoride, PdCl2(dppf), K2CO3, 1,4-
dioxane, 90 °C; (c) PPh3, o-DCB, 165 °C; (d) Pd2(dba)3 CHCl3,
3
DavePhos, NaOt-Bu, toluene, 100 °C; (e) Pd(OAc)2, O2, AcOH, 120 °C;
(f) Tf2O, pyridine, CH2Cl2, rt; (g) Pd(PPh3)4, LiCl, (Bu3Sn)2, 1,4-
dioxane, 100 °C; (h) 4-bromo-3-nitrobenzotrifluoride, Pd2(dba)3 CH-
3
Cl3, P(t-Bu)3 HBF4, CsF, toluene, 110 °C.
3
Dihydroquinolinone (GSK-1), 5, is also a highly potent KSP
inhibitor with a 4-trifluoromethylbiphenyl scaffold (Figure 2B),
which presumably binds in a twisted conformation of the
biphenyl part at the interface of the R4 and R6 helices of KSP.10b
Mimicking the possible planar conformation of biphenyl-type 5,
we designed four lactam-fused carbazole derivatives 10aꢀd.
Carbazole 10a or 10b was designed by bridging via an NH group
between dihydroquinolinone 5- or 7-positions and the 20-position
of 5, respectively. In carbazoles 10c,d with 2,3- and 3,4-fused
lactams, the orientations of the trifluoromethyl and lactam anilide
moieties in 5 were reproduced. To optimize the appropriate ring
systems of 10a,b for KSP inhibition, carbazoles 11a,b and 12 with
different ring-sized lactams were also evaluated. Carbazoles 13a,b
witha ureagroupatthe 2- or 3-position weresimilarlyinvestigated,
which are the constrained analogs of a biphenyl 6a (GSK-2) and
nonfluorinated congener 6b (Figure 2C).12
Synthesis of Carboline and Carbazole Derivatives for
Potential KSP Inhibitors. A series of carboline derivatives
9aꢀd were prepared by N-arylation using 2-, 3-, or 4-iodopyr-
idine (14, 16, 17)/2-bromo-5-(trifluoromethyl)aniline (15) or
3-amino-2-chloropyridine (18)/substituted iodobenzene (19)
and the subsequent palladium-catalyzed intramolecular CꢀH
arylation of N-aryl-2-haloaniline derivatives (Scheme 1).13
4841
dx.doi.org/10.1021/jm200448n |J. Med. Chem. 2011, 54, 4839–4846

Journal of Medicinal Chemistry
ARTICLE
Scheme 4. Synthesis of Carbazole Derivatives with a Urea
Groupa
Table 1. StructureꢀActivity Relationships of Carboline
Derivatives
a Reagents and conditions: (a) 4-bromo-3-nitrobenzotrifluoride, PdCl2-
(dppf), K2CO3, 1,4-dioxane, 80 °C; (b) PPh3, o-DCB, 165 °C; (c) 4 N
HCl in 1,4-dioxane, rt; (d) KOCN, AcOH, H2O, rt.
to the methylene group concomitantly with the formation of a
carbazole core structure to provide the desired carbazole 12.
The synthesis of compound 13a with a urea group at the
carbazole 2-position began from commercially available pinacol-
boronate ester 33 (Scheme 4). Compound 33 was converted to
the 2-nitrobiphenyl derivative 34 by SuzukiꢀMiyaura coupling,
which underwent reductive cyclization to afford Boc-protected
aminocarbazole 35. After deprotection, compound 36 was con-
verted to the expected carbazole 13a by using KOCN.22 Com-
pound 13b with a urea group at the carbazole 3-position was
prepared by the existing method.12
StructureꢀActivity Relationship Study of Carboline- and
Carbazole-Based KSP Inhibitors. The resulting carbolines and
carbazoles were evaluated for inhibition of KSP ATPase and
cytotoxicity toward HeLa cells. Among carboline derivatives
9aꢀd having a 7-trifluoromethyl group, no inhibitory effect
was observed by the R-carboline derivative 9a, whereas β- and
γ-carbolines 9b,c exerted enhanced KSP ATPase inhibitory
activity compared with the corresponding carbazole derivative
3a or biaryl-type derivative 4 [IC50(9b) = 0.052 μM; IC50(9c) =
0.095 μM] (Table 1). δ-Carboline 9d exhibited similar inhibitory
potency to the parent carbazole 3a. A positive correlation
between KSP ATPase inhibition and cytotoxicity toward HeLa
cells was observed among carboline-based KSP inhibitors 9aꢀd;
for example, the cytotoxicity of β-carboline 9b was the most
potent (IC50 = 0.81 μM). This investigation revealed that the
restriction of the rotatable bond in biaryl-type compounds using
fused indole scaffolds is a promising approach to design potent
KSP inhibitors.
a IC50 values were derived from the doseꢀresponse curves generated
from triplicate data points. b Inhibition of microtubule-activated KSP
ATPase activity. c Cytotoxic activity against HeLa cells after 72-h
exposure to each compound. d IC50 was ∼10 μM. e Not tested.
Table 2. StructureꢀActivity Relationships of Carbazole
Derivatives with a Lactam-Fused Structure
Carbazole derivatives 10aꢀd with a lactam-fused structure
were then investigated (Table 2). The accessory anilide group at
the carbazole 2-position (10a,b) improved KSP ATPase inhibi-
tory activity compared with 5. Interestingly, carbazole 10b with a
2,3-fused lactam ring showed remarkably potent cytotoxicity
(IC50 = 0.091 μM) in comparison with carbazole 10a fused at
the 1- and 2-positions, although 10b showed approximately
equipotent KSP ATPase inhibitory activity to 10a.23 An identical
modification at the carbazole 3-position (10c,d) was less effec-
tive, which showed similar or slightly greater potency compared
with the parent carbazole 3a. This may be due to the inappropri-
ate orientations of the lactam anilide moiety or the electron-
donating effect to the carbazole scaffold. The 2,3-fused lactam
10c did not show bioactivity in the cell-based assay, which was
in contrast with 10b, which had a similar fused ring systems.
a IC50 values were derived from the doseꢀresponse curves generated
from triplicate data points. b Inhibition of microtubule-activated KSP
ATPase activity. c Cytotoxic activity against HeLa cells after 72-h
exposure to each compound.
4842
dx.doi.org/10.1021/jm200448n |J. Med. Chem. 2011, 54, 4839–4846

Journal of Medicinal Chemistry
ARTICLE
Table 3. StructureꢀActivity Relationships of Carbazole
Table 4. StructureꢀActivity Relationships of Carbazole
Derivatives with a Urea Group
Derivatives with Five-, Six-, and Seven-Membered Lactam-
Fused or Pyrrole-Fused Structures
a IC50 values were derived from the doseꢀresponse curves generated
from triplicate data points. b Inhibition of microtubule-activated KSP
ATPase activity. c Cytotoxic activity against HeLa cells after 72-h
exposure to each compound.
a IC50 values were derived from the doseꢀresponse curves generated
from triplicate data points. b Inhibition of microtubule-activated KSP
ATPase activity. c Cytotoxic activity against HeLa cells after 72-h
exposure to each compound.
The outward orientation of the lactam NH group in 10c was
disadvantageous to the interaction with KSP and cellular effects.
The biological activities of urea-containing compounds were
also examined (Table 3).12 The urea group at the carbazole
2-position in 13a slightly improved KSP ATPase inhibitory
activity and cytotoxicity compared with the biphenyl-type in-
hibitors 6a,b. 3-Carbamidylcarbazole 13b was less potent than
the 2-substituted congener 13a, which was consistent with the
series of carbolines 9aꢀd and lactam-fused compounds 10aꢀd.
Subsequently, optimization of the lactam ringwas carried outfor
thehighly potentcarbazoles10a,b (Table 4). Carbazolederivatives
12 and 11b, which were fused at the carbazole 2,3-positions with a
five- and seven-membered lactam, respectively, exerted slightly less
bioactivity compared with the six-membered lactam 10b, suggest-
ing the great flexibility of the lactam carbonyl placement. Inhibitory
activity was also dependent upon the position of ring fusion.
Carbazole 11a, which was fused with a seven-membered lactam at
the 1- and 2-positions, showed less KSP inhibition and cytotoxicity
than the 2,3-fused congener 11b. This was consistent with the
bioactivity profiles of compounds 10a,b with a six-membered
lactam. The less potent activities of pyrrolocarbazole derivative
31 than 12 indicate that a lactam carbonyl group at this position is
an indispensable functional group for bioactivity.
Examinations of three series of carboline and carbazole deriva-
tives revealed that the appropriate restriction of the rotatable
linkage in biphenyl-type compounds provides favorable effects on
KSP ATPase inhibitory activity and cytotoxicity. Introduction of a
direct NH linkage in the original biphenyls 4ꢀ6 stabilized the
coplanar conformation and bidirectional arrangements of two
accessory groups to fabricate the potential interactions with the
enzyme. Although twisted bioactive conformations of two phenyl
rings in 5 for KSP were proposed,10b the potent constrained
analogs 10a,b suggest that the planar conformations could pro-
vide another favorable interaction for bioactivity. In particular,
modification at the carbazole 2-position with a polar group such as
a nitrogen atom (for β-carboline), lactam amide, or urea group
improved the inhibitory activities. The lactam-fused carbazole 10b
with a linear 6ꢀ6ꢀ5ꢀ6 fused ring system was the most potent
KSP inhibitor in this investigation.
Potent KSP inhibitors (10a,b and 13a) were evaluated for
inhibitory effects against the other motor proteins: centromere-
associated protein E (CENP-E), Kid, mitotic kinesin-like protein
1 (MKLP-1), and KIF-4 (see Supporting Information). The
compounds did not inhibit these kinesins even at 20 μM,
indicating that they are specific KSP inhibitors. Furthermore,
we investigated the effects of these compounds on cell cycle
progression (Table 5). All of these compounds triggered cell-
cycle arrest of HeLa cells to accumulate mitotic cells, and in
particular, lactam-fused carbazole 10b was effective in a low MI50
(50% mitotic index induction concentration). In the microscopic
analysis of the mitotic phenotype of the cells treated with highly
potent β-carboline 9b, carbazoles 10a,b, and 13a, the formation
of monopolar spindles (the typical phenotype of KSP inhibition)
was observed in all of the mitotically arrest cells (Figure 3 and the
Supporting Information). Although β-carbolines were reported
to inhibit topoisomerases I/II,24 the possible off-target effects
were not observed by treatment with 9b.
Mechanistic Studies of Carboline- and Carbazole-Based
KSP Inhibitors. It has been reported that most KSP inhibitors
(e.g., monastrol, S-trityl-L-cysteine) bind to the allosteric pocket
formed by helices R2 and R3 and loop L5.25,26 For example,
tetrahydrocarboline-based inhibitors such as 2 (which was a lead
compound for our carbazole-based inhibitors 3aꢀc) also bind to
the same allosteric pocket to show ATP-uncompetitive behavior.27
In contrast, the biphenyl-type inhibitors 5and6a (which work in an
ATP-competitive manner) bind at the interface of helices R4 and
R6 of the KSP ATPase domain (not at the ATP binding site).10b
4843
dx.doi.org/10.1021/jm200448n |J. Med. Chem. 2011, 54, 4839–4846

Journal of Medicinal Chemistry
ARTICLE
Table 5. Effects of KSP Inhibitors on Cell Cycle Progression
of HeLa Cells
Preliminary docking simulation study of these inhibitors
demonstrated that the adenine binding pocket at the ATP
binding site was too small to accommodate the potent lactam-
fused carbazole 10b (data not shown). In addition, reported
crystal structure analyses29 revealed that the structures of the
ATP binding site were highly conserved between KSP and
CENP-E. This suggested that possible binding of these inhibitors
to the nucleotide site would be contradictory to high KSP
selectivity. Accordingly, it is likely that these carboline and
carbazole derivatives bind to the interface of helices R4 and R6
in a similar binding mode to biphenyl-type inhibitors.
compound
MI50 (μM)a
5
0.29
3.7
6b
10a
10b
13a
1.1
0.13
1.3
a MI50 values indicate the concentration of 50% induction in mitotic
HeLa cells after 24-h exposure to each compound. Values were derived
from doseꢀresponse curves generated from triplicate data points.
’ CONCLUSIONS
We identified a novel class of small-molecule KSP inhibitors.
Studies of the structureꢀactivity relationship of carboline and
carbazole derivatives indicated that the coplanar conformations
of biphenyl groups favorably contribute to potent KSP inhibition.
In addition, hydrophilic groups (e.g., lactam amide group) at the
appropriate position are also indispensable functional groups.
The carbazole derivative 10b fused with a six-membered lactam
at the 2- and 3-positions exhibited the most potent KSP ATPase
inhibitory activity and the consequent cytotoxicity by effective
cell-cycle arrest at the M-phase. It can be concluded from the
biochemical studies and selectivity profiles for several kinesins
that the carbolines and carbazoles work as ATP-competitive KSP
inhibitors by presumably binding at the same site to the one for
the parent biphenyl-type inhibitors. These investigations suggest
that similar fused indole scaffolds may provide two distinctive
KSP inhibitors, which work through separate inhibitory modes:
tetrahydrocarboline-based ATP-uncompetitive inhibitors such as
2 and planar carboline/carbazole-based ATP-competitive inhi-
bitors such as 9b and 10b.
Figure 3. Mitotic phenotype of HeLa cells treated with fused indole
derivatives 9b (8.0 μM) and 10b (0.25 μM). Chromosomes are colored
blue and microtubules red.
’ EXPERIMENTAL SECTION
General. 1H NMR spectra were recorded using a JEOL AL-400 or a
JEOL ECA-500 spectrometer. Chemical shifts are reported in δ (ppm)
relative to Me4Si as an internal standard. 13C NMR spectra were
referenced to the residual CHCl3 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
(uncorrected). For flash chromatography, Wakogel C-300E (Wako)
was employed. For analytical HPLC, a Cosmosil 5C18-ARII column
(4.6 ꢁ 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 220 nm. The purity
of the compounds was determined by combustion analysis or HPLC
analysis (>95%).
KSP ATPase Assay. The microtubule-stimulated KSP ATPase
reaction was performed in a reaction buffer [20 mM PIPES-K (pH 6.8),
25 mM KCl, 2 mM MgCl2, 1 mM EGTA-KOH (pH 8.0)] containing
38 nM bacteria-expressed KSP motor domain (1ꢀ369) fused to a
histidine 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 (pH7.4), 0.04% (v/v) NP-40]. After preincu-
bation of 9.7 μL of the enzyme solution with 3.8 μL of each chemical
solution at 25 °C for 30 min, the ATPase reaction was initiated by the
addition of 1.5 μL of 0.3 mM ATP, followed by incubation at 25 °C for a
further 15 min. The reaction was terminated by the addition of 15 μL of
the Kinase-Glo Plus reagent (Promega). The ATP consumption in each
Figure 4. LineweaverꢀBurk plots of the kinetics of KSP ATPase.
ATPase activity of 80 nM KSP motor domain with 400 nM micro-
tubules as a function of ATP concentration in the absence (b) and
presence (O) of inhibitors were measured by steady-state KSP ATPase
assay. (A) β-carboline 9b (56 nM); (B) lactam-fused carbazole 10b
(32 nM). Each data point is the average of two independent experi-
ments. Enzyme velocities at the same inhibitor concentration were fitted
to the MichaelisꢀMenten equation to obtain Vmax
.
The carboline- and carbazole-based inhibitors 9ꢀ13 were designed
by a combination of two distinct scaffolds (ring-fused compounds
1ꢀ3and biaryl compounds 4ꢀ6) so the inhibitory mechanism and
binding site of KSP inhibitors 9ꢀ13 would be of interest.
To provide valuable insights into the binding mode of
inhibitory β-carboline 9b and lactam-fused carbazole 10b, KSP
ATPase activity was assessed in the absence or presence of
inhibitors when incubated with variable ATP concentrations
(Figure 4). LineweaverꢀBurk plots revealed consistent Vmax
values independent of the presence of the inhibitors. This
indicated that compounds 9b and 10b functionally compete
with ATP binding.28 It is of interest that these inhibitors 9b and
10b worked in a similar mode of inhibition to biphenyl-type 4ꢀ6
even though these belong to different types of scaffolds. This is
the first observation of β-carboline and carbazole scaffolds for
potent ATP-competitive KSP inhibitors.
4844
dx.doi.org/10.1021/jm200448n |J. Med. Chem. 2011, 54, 4839–4846

Journal of Medicinal Chemistry
ARTICLE
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 rates in
each condition were evaluated to determine IC50 values using Microsoft
Excel and GraphPad Prism software.
ATP concentration at a particular drug concentration: V = ([ATP] ꢁ
Vmax)/([ATP] þ Km).
’ ASSOCIATED CONTENT
S
Growth Inhibition Assay. HeLa cells were cultured in DMEM
medium supplemented with 10% (v/v) FCS and antibiotics at 37 °C in a
5% CO2 incubator. Growth inhibition assays using HeLa cells were
performed in 96-well plates (Greiner). HeLa cells were seeded at
5000 cells/well in 50 μL of DMEM and placed for 6 h. Chemical
compounds in DMSO were diluted 100-fold with the culture medium in
advance. Following the addition of 40 μL of the fresh culture medium,
30 μL of the chemical diluents were also added to the cell cultures. The
final volume of DMSO in the medium was equal to 0.25% (v/v). The
cells under chemical treatment were incubated for a further 72 h. The
wells in the plates were washed twice with the cultured medium without
phenol red. After 1 h incubation with 100 μL of the medium, the cell
culture in each well was supplemented with 20 μL of the MTS reagent
(Promega), followed by incubation for an additional 40 min. Absorbance
at 590 nm of each well was measured using a VersaMax plate reader
(Molecular Devices). At least three experiments were performed per
condition, and the averages and standard deviations of inhibition rates in
each condition were evaluated to determine IC50 values using Microsoft
Excel and GraphPad Prism software.
Supporting Information. Experimental procedures, char-
b
acterization, and bioassay data. This material is available free of
charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*Tel: þ81-75-753-4551. Fax: þ81-75-753-4570. E-mail: soishi@
pharm.kyoto-u.ac.jp (S.O.); nfujii@pharm.kyoto-u.ac.jp (N.F.).
’ ACKNOWLEDGMENT
This research was supported by the Targeted Protein Re-
search Program and Grants-in-Aid for Scientific Research from
the Ministry of Education, Culture, Sports, Science and Tech-
nology, Japan. T.W. is grateful for the JSPS Research Fellowships
for Young Scientists.
’ ABBREVIATIONS
Evaluation of Mitotic Arrest. HeLa cells were seeded at 105 cells/
well in 500 μL of the medium on Lab-Tek II 8-well CC2 glass chamber
slides (Nalge Nunc). After 6-h incubation at 37 °C in a 5% CO2
incubator, the culture medium was exchanged with 400 μL of fresh
medium containing different inhibitors or DMSO, and the cells were
placed at 37 °C in 5% CO2 for 24 h. The final volume of inhibitor
dissolved in DMSO in the medium was 0.4% (v/v). Before processing
for immunofluorescence, HeLa cells were washed once with PBS(ꢀ)
and then fixed with 3% paraformaldehyde/PBS(ꢀ) for 3 min and then
with MeOH at ꢀ20 °C for 10 min. Microtubules were stained with a
primary mouse monoclonal anti-R-tubulin antibody, DM1A, and a
secondary rabbit polyclonal AlexaFluor594-conjugated anti-mouse anti-
body (Molecular Probes). Mitotic chromosomes were stained with a
primary polyclonal rabbit anti-phospho-histone H3 (Ser10) antibody
(Upstate) followed by a secondary polyclonal AlexaFluor488-conju-
gated anti-rabbit antibody. DNA was counterstained with DAPI. Cells
were classified as “mitotic”, “interphase”, or “other” by visual inspection
to calculate the percentage of mitotic cells at each inhibitor concentra-
tion. At least 500 cells were examined per condition, and MI50 values
were evaluated using Microsoft Excel and GraphPad Prism software.
Fluorescent images were taken by a DP30BW CCD camera coupled to
an IX71 microscope (Olympus) and superimposed using Adobe Photo-
shop software.
CENP-E,centromere-associated protein E; KSP,kinesin spindle
protein; MKLP-1,mitotic kinesin-like protein 1
’ REFERENCES
(1) (a) Dutcher, J. P.; Novik, Y.; O’Boyle, K.; Marcoullis, G.; Secco,
C.; Wiernik, P. H. 20th-century advances in drug therapy in oncology -
part II. J. Clin. Pharmacol. 2000, 40, 1079–1092. (b) Jordan, M. A.;
Wilson, L. Microtubules as a target for anticancer drugs. Nat. Rev. Cancer
2004, 4, 253–265.
(2) (a) Rowinsky, E. K.; Chaudhry, V.; Cornblath, D. R.; Donehower,
R. C. Neurotoxicity of taxol. J. Natl. Cancer Inst. Monogr. 1993, 15, 107–
115. (b) Quasthoff, S.; Hartung, H. P. Chemotherapy-induced peripheral
neuropathy. J. Neurol. 2002, 249, 9–17.
(3) Sawin, K. E.; LeGuellec, K.; Philippe, M.; Mitchison, T. J. Mitotic
spindle organization by a plus-end-directed microtubule motor. Nature
1992, 359, 540–543.
(4) (a) Blangy, A.; Lane, H. A.; d’Herin, P.; Harper, M.; Kress, M.;
Nigg, E. A. Phosphorylation by p34cdc2 regulates spindle association of
human Eg5, a kinesin-related motor essential for bipolar spindle
formation in vivo. Cell 1995, 83, 1159–1169. (b) Walczak, C. E.; Vernos,
I.; Mitchison, T. J.; Karsenti, E.; Heald, R. A model for the proposed roles
of different microtubule-based motor proteins in establishing spindle
bipolarity. Curr. Biol. 1998, 8, 903–913. (c) Tao, W.; South, V. J.; Zhang,
Y.; Davide, J. P.; Farrell, L.; Kohl, N. E.; Sepp-Lorenzino, L.; Lobell, R. B.
Induction of apoptosis by an inhibitor of the mitotic kinesin KSP
requires both activation of the spindle assembly checkpoint and mitotic
slippage. Cancer Cell. 2005, 8, 49–59. (d) Marcus, A. I.; Peters, U.;
Thomas, S. L.; Garrett, S.; Zelnak, A.; Kapoor, T. M.; Giannakakou, P.
Mitotic kinesin inhibitors induce mitotic arrest and cell death in Taxol-
resistant and -sensitive cancer cells. J. Biol. Chem. 2005, 280, 11569–11577.
(5) Sakowicz, R.; Finer, J. T.; Beraud, C.; Crompton, A.; Lewis, E.;
Fritsch, A.; Lee, Y.; Mak, J.; Moody, R.; Turincio, R.; Chabala, J. C.;
Gonzales, P.; Roth, S.; Weitman, S.; Wood, K. W. Antitumor activity of a
kinesin inhibitor. Cancer Res. 2004, 64, 3276–3280.
Steady-State KSP ATPase Assay. An enzyme-coupled system
that regenerates ADP to ATP was used to measure ATP hydrolysis by
KSP, with NADH absorbance at 340 nm as an indirect measurement of
ATP turnover. To measure KSP ATPase activity in a reaction, the assay
buffer [2.5 mM KCl, 50 mM PIPES (pH 6.9), 1 mM MgCl2, 3 mM
potassium phophoenol pyruvate, 450 μM Na2 NADH, 1 mM dithio-
3
threitol, 8 μM taxol, 9 U/mL lactate dehydrogenase, 15 U/mL pyruvate
kinase, and 400 nM taxol-stabilized microtubules, and 80 nM bacteria-
expressed KSP motor domain; Km(ATP) = 15ꢀ35 μM] was supple-
mented with various concentrations of ATP along with KSP inhibitor
solution. The optical density of a reaction at 340 nm was measured every
5 s using a UV-2450 photometer (Shimadzu). The steady-state rate of
absorbance reduction was calculated using a linear least-squares fitting
method by Shimadzu UVProbe version 2.10 and Microsoft Excel. The
coupling activity of the ATP regeneration enzyme system was 100-fold
greater than the ATPase activity measured in these experiments. Enzyme
velocities were fitted to the MichaelisꢀMenten equation as a function of
(6) (a) Mayer, T. U.; Kapoor, T. M.; Haggarty, S. J.; King, R. W.;
Schreiber, S. L.; Mitchison, T. J. Small molecule inhibitor of mitotic
spindle bipolarity identified in a phenotype-based screen. Science 1999,
286, 971–974. (b) Sarli, V.; Giannis, A. Inhibitors of mitotic kinesins:
next-generation antimitotics. ChemMedChem 2006, 1, 293–298. (c)
Jackson, J. R.; Patrick, D. R.; Dar, M. M.; Huang, P. S. Targeted anti-
mitotic therapies: Can we improve on tubulin agents? Nat Rev. Cancer
4845
dx.doi.org/10.1021/jm200448n |J. Med. Chem. 2011, 54, 4839–4846

Journal of Medicinal Chemistry
ARTICLE
2007, 7, 107–117. (d) Matsuno, K.; Sawada, J.; Asai, A. Therapeutic
potential of mitotic kinesin inhibitors in cancer. Expert Opin. Ther. Pat.
2008, 18, 253–274. (e) Sarli, V.; Giannis, A. Targeting the kinesin
spindle protein: Basic principles and clinical implications. Clin. Cancer
Res. 2008, 14, 7583–7587.
(7) Oishi, S.; Watanabe, T.; Sawada, J.; Asai, A.; Ohno, H.; Fujii, N.
Kinesin spindle protein (KSP) inhibitors with 2,3-fused indole scaffolds.
J. Med. Chem. 2010, 53, 5054–5058.
(8) Nakazawa, J.; Yajima, J.; Usui, T.; Ueki, M.; Takatsuki, A.; Imoto,
M.; Toyoshima, Y.; Osada, H. A novel action of terpenedole E on the
motor activity of mitotic Kinesin Eg5. Chem. Biol. 2003, 10, 131–137.
(9) Hotha, S.; Yarrow, J. C.; Yang, J. G.; Garrett, S.; Renduchintala,
K. V.; Mayer, T. U.; Kapoor, T. M. HR22C16: A potent small-molecule
probe for the dynamics of cell division. Angew. Chem., Int. Ed. 2003,
42, 2379–2382.
(19) (a) Azizian, H.; Eaborn, C.; Pidcock, A. Synthesis of organo-
trialkylstannanes. The reaction between organic halides and hexaalk-
yldistannanes in the presence of palladium complexes. J. Organomet.
Chem. 1981, 215, 49–58. (b) Furuya, T.; Strom, A. E.; Ritter, T. Silver-
mediated fluorination of functionalized aryl stannanes. J. Am. Chem. Soc.
2009, 131, 1662–1663.
(20) Scott, W. J.; Crisp, G. T.; Stille, J. K. Palladium-catalyzed
coupling of vinyl triflates with organostannanes. A short synthesis of
Pleraplysillin-1. J. Am. Chem. Soc. 1984, 106, 4630–4632.
(21) Kumar, C. N. S. S. P.; Devi, C. L.; Rao, V. J.; Palaniappan, S. Use
of pyridinium chlorochromate and reusable polyaniline salt catalyst
combination for the oxidation of indoles. Synlett 2008, 13, 2023–2027.
(22) Wang, H.; Liu, L.; Wang, Y.; Peng, C.; Zhang, J.; Zhu, Q. An
efficient synthesis of 4-alkyl-2(1H)-quinazolinones and 4-alkyl-2-chlor-
oquinazolines from 1-(2-alkynylphenyl)ureas. Tetrahedron Lett. 2009,
50, 6841–6843.
(10) (a) Parrish, C. A.; Adams, N. D.; Auger, K. R.; Burgess, J. L.;
Carson, J. D.; Chaudhari, A. M.; Copeland, R. A.; Diamond, M. A.;
Donatelli, C. A.; Duffy, K. J.; Faucette, L. F.; Finer, J. T.; Huffman, W. F.;
Hugger, E. D.; Jackson, J. R.; Knight, S. D.; Luo, L.; Moore, M. L.;
Newlander, K. A.; Ridgers, L. H.; Sakowicz, R.; Shaw, A. N.; Sung,
C. M. M.; Sutton, D.; Wood, K. W.; Zhang, S. Y.; Zimmerman, M. N.;
Dhanak, D. Novel ATP-competitive kinesin spindle protein inhibitors.
J. Med. Chem. 2007, 50, 4939–4952. (b) Luo, L.; Parrish, C. A.; Nevins,
N.; McNulty, D. E.; Chaudhari, A. M.; Carson, J. D.; Sudakin, V.; Shaw,
A. N.; Lehr, R.; Zhao, H.; Sweitzer, S.; Lad, L.; Wood, K. W.; Sakowicz,
R.; Annan, R. S.; Huang, P. S.; Jackson, J. R.; Dhanak, D.; Copeland,
R. A.; Auger, K. R. ATP-competitive inhibitors of the mitotic kinesin
KSP that function via an allosteric mechanism. Nat. Chem. Biol. 2007,
33, 722–726. (c) Rickert, K. W.; Schaber, M.; Torrent, M.; Neilson,
L. A.; Tasber, E. S.; Garbaccio, R.; Coleman, P. J.; Harvey, D.; Zhang, Y.;
Yang, Y.; Marshall, G.; Lee, L.; Walsh, E. S.; Hamilton, K.; Buser, C. A.
Discovery and biochemical characterization of selective ATP competi-
tive inhibitors of the human mitotic kinesin KSP. Arch. Biochem. Biophys.
2008, 469, 220–231. (d) Matsuno, K.; Sawada, J.; Sugimoto, M.; Ogo,
N.; Asai, A. Bis(hetero)aryl derivatives as unique kinesin spindle protein
inhibitors. Bioorg. Med. Chem. Lett. 2009, 19, 1058–1061.
(23) The varied cell membrane permeability could be attributable to
the incompatible results between the in vitro KSP inhibition and
cytotoxicity.
(24) Funayama, Y.; Nishio, K.; Wakabayashi, K.; Nagao, M.; Shimoi,
K.; Ohira, T.; Hasegawa, S.; Saijo, N. Effects of β- and γ-carboline deri-
vatives of DNA topoisomerase activities. Mutat. Res. 1996, 349, 183–191.
(25) Yan, Y.; Sardana, V.; Xu, B.; Homnick, C.; Halczenko, W.;
Buser, C. A.; Schaber, M.; Hartman, G. D.; Huber, H. E.; Kuo, L. C.
Inhibition of a mitotic motor protein: Where, how, and conformational
consequences. J. Mol. Biol. 2004, 335, 547–554.
(26) (a) Kaan, H. Y. K.; Ulaganathan, V.; Hackney, D. D.; Kozielski,
F. An allosteric transition trapped in an intermediate state of a new
kinesin-inhibitor complex. Biochem. J. 2010, 425, 55–60. (b) Kim, E. D.;
Buckley, R.; Learman, S.; Richard, J.; Parke, C.; Worthylake, D. K.;
Wojcik, E. J.; Walker, R. A.; Kim, S. Allosteric drug discrimination is
coupled to mechanochemical changes in the kinesin-5 motor core.
J. Biol. Chem. 2010, 285, 18650–18661.
(27) Barsanti, P. A.; Ni, W. W. Z.; Duhl, D.; Brammeier, N.; Martin,
E.; Bussiere, D.; Walter, A. O. The discovery of tetrahydro-β-carbolines
as inhibitors of the kinesin Eg5. Bioorg. Med. Chem. Lett. 2010, 20, 157–
160.
(11) Asai, A.; Sawada, J.; Matsuno, K.; Ogo, N.; Nishigaki, J.; Kojima,
M. Japanese Patent JP200923986, 2009.
(28) Although the plots in the presence and absence of inhibitors
converged, the possible mixed mode of inhibition by binding to the
second site cannot be ruled out.
(29) Garcia-Saez, I.; Yen, T.; Wade, R. H.; Kozielski, F. Crystal
structure of the motor domain of the human kinetochore protein CENP-
E. J. Mol. Biol. 2004, 340, 1107–1116.
(12) The urea-containing carbazoles appeared in the previous patent
information; however, the biological data have not been reported:
Dhanak, D.; Knight, S. D.; Moore, M. L.; Newlander, K. A. PCT Int.
Appl. WO2006005063, 2006.
(13) Ackermann, L.; Althammer, A. Domino N-H/C-H bond activa-
tion: palladium-catalyzed synthesis of annulated heterocycles using
dichloro(hetero)arenes. Angew. Chem., Int. Ed. 2007, 46, 1627–1629.
(14) Owens, A. P.; Nadin, A.; Talbot, A. C.; Clarke, E. E.; Harrison,
T.; Lewis, H. D.; Reilly, M.; Wrigley, J. D. J.; Castro, J. L. High affinity,
bioavailable 3-amino-1,4-benzodiazepine-based γ-secretase inhibitors.
Bioorg. Med. Chem. Lett. 2003, 13, 4143–4145.
(15) (a) Ishiyama, T.; Murata, M.; Miyaura, N. Palladium(0)-
catalyzed cross-coupling reaction of alkoxydiboron with haloarenes: A
direct procedure for arylboronic esters. J. Org. Chem. 1995, 60, 7508–
7510. (b) Ishiyama, T.; Miyaura, N. Chemistry of group 13 element-
transition metal linkage: The platinum- and palladium-catalyzed reac-
tions of (alkoxo)diborons. J. Organomet. Chem. 2000, 611, 392–402.
(16) Freeman, A. W.; Urvoy, M.; Criswell, M. E. Triphenylpho-
sphine-mediated reductive cyclization of 2-nitrobiphenyls: A practical
and convenient synthesis of carbazoles. J. Org. Chem. 2005, 70, 5014–
5019.
(17) (a) Watanabe, T.; Ueda, S.; Inuki, S.; Oishi, S.; Fujii, N.; Ohno,
H. One-pot synthesis of carbazoles by palladium-catalyzed N-arylation
and oxidative coupling. Chem. Commun. 2007, 4516–4518. (b) Watanabe,
T.; Oishi, S.; Fujii, N.; Ohno, H. Palladium-catalyzed direct synthesis of
carbazoles via one-pot N-arylation and oxidative biaryl coupling: Synthesis
and mechanistic study. J. Org. Chem. 2009, 74, 4720–4726.
(18) Thompson, A. L. S.; Kabalka, G. W.; Akula, M. R.; Huffman,
J. W. The conversion of phenols to the corresponding aryl halides under
mild conditions. Synthesis 2005, 4, 547–550.
4846
dx.doi.org/10.1021/jm200448n |J. Med. Chem. 2011, 54, 4839–4846