Beruflich Dokumente
Kultur Dokumente
net
Current Medicinal Chemistry, 2014, 21, 2691-2701
2691
Department of Organic Chemistry, Key Laboratory of Chemical Biology (Ministry of Education), School of Pharmaceutical Sciences, Shandong University, Jinan 250012, P. R. China; 2College of Animal Science & Veterinary Medicine,
Shandong Agricultural University, Taian 271018, P. R. China
Abstract: Kinesin spindle protein (KSP) plays an essential role in centrosome separation and formation of the bipolar
mitotic spindle. Its exclusive involvement in the mitotic spindle of proliferating cells presents an opportunity for
developing new anticancer agents with reduced side effects relative to antimitotics that target tubulin. Small molecule
KSP inhibitors have demonstrated their potential as novel antimitotic agents. Several KSP inhibitors have progressed into
clinical trials and many others are in preclinical developments. Recently, KSP inhibitors of wide structural diversity have
appeared in literatures. This review will summarize the developments of KSP inhibitors based on the five-membered heterocycle scaffolds in recent 10 years. These small molecule KSP inhibitors were classified as dihydropyrazoles, dihydropyrroles, thiophenes, dihydrothiadiazoles, thiazoles and fused pyrroles, their structure-activity relationships were discussed.
Keywords: Antitumor agents, ATP competitive inhibitors, ATP uncompetitive inhibitors, cancer, Eg5, heterocycles, kinesin,
kinesin spindle protein, KSP, KSP inhibitors, structure-activity relationships.
INTRODUCTION
Mitotic kinesins are enzymes essential for assembly and
function of the mitotic spindle. These microtubule-based
motor proteins travel along microtubules reaching into every
corner of the cell. They are regarded as biological machines
that transduce chemical energy into mechanical forces and
motion [1, 2]. Kinesins utilize the energy from ATP hydrolysis to power their movement unidirectionally along microtubules and to transport molecular cargo to specific destinations. Mitotic kinesins play essential roles during all phases
of mitosis. During mitosis, kinesins organize microtubules
into the bipolar structure that is the mitotic spindle. They
mediate movement of chromosomes along spindle microtubules, as well as structural changes in the mitotic spindle
associated with specific phases of mitosis [3, 4]. To date,
there are at least 12 kinesins involved in mitosis and meiosis
which are responsible for spindle and chromosomal movement. Among them, Kinesin spindle protein (KSP, HsEg5) is
a slow, plus end-directed motor of the kinesin-5 subfamily
[5-7]. KSP is mainly consisted of three domains: the motor
domain, the coiled-coil domain and the C-terminal domain.
The motor domain starts with the residue Val21 and ends
with the residue Lys357 and before the motor domain there
is a 20 residue long stretch. A short linker (Asn358Pro363)
connects the motor to a predicted interrupted coiled-coil
*Address correspondence to these authors at the Department of Organic
Chemistry, Key Laboratory of Chemical Biology (Ministry of Education),
School of Pharmaceutical Sciences, Shandong University, Jinan 250012, P.
R. China; Tel: +86-531-88382006; Fax: +86-531-88382548;
E-mail: liuzhaop@sdu.edu.cn; and College of Animal Science & Veterinary
Medicine, Shandong Agricultural University, Taian 271018, P. R. China;
Tel: +86-538-8233881; Fax: +86-538-8241419; E-mail: wrzh@sdau.edu.cn
1875-533X/14 $58.00+.00
Zhao et al.
H2N
O
S
N
F
S
N
N
NH2
OH
F
ARRY-520
Phase II for plasma cell leukemia
O
N
MK-0731
Phase I for advanced solid tumor
H
N
H
N
O
N
N
F3C
N
S
N
O
NHEt
HN
NH2
N
H
Ispinesib
Phase II for non-small-cell lung cancer
O
Cl
H
N
N
AZD4877
Phase II for bladder cancer
N
N
Cl
S
O
EMD534085
Phase I for lymphoma
LY2523355
Phase II for metastatic breast cancer
O
F
Cl
Cl
N
O
O
NH2
ARQ 621
Phase I for metastatic solid tumors
NH2
SB-743921
Phase II for Hodgkin's disease
their selectivity and binding modes with the KSP were discussed.
1. 4,5-DIHYDROPYRAZOLES
3,5-Diaryl-4,5-dihydropyrazoles were identified as potent
and cell-active KSP inhibitors through the optimization of
high-throughput screening (HTS) hits [27]. In a series of 3-(2chlorophenyl)-5-(3-hydroxyphenyl)-4,5-dihydropyrazoles, the
N1-acetyl analogue 1 (Fig. 3) exhibited the best KSP binding
ability with an IC50 of 450 nM. Increasing the size of the acyl
group from acetyl to propionyl or isobutyryl reduced potency
(~2.5-fold), but the pivaloyl analogue restored the potency
(460 nM). The replacement of the acetyl group with a 4aminobutanoyl or a benzoyl group had a detrimental effect on
the potency. N1-Methylation or the addition of an oxygen
spacer to the acetyl group by formation of a methyl carbamate
resulted in a loss of measurable activity. Moving the hydroxyl
group from 3-position to 2- or 4-position on the C5 phenyl
ring decreased the potency. Replacing the 3-hydroxyl group
with methoxy, cyano, amino, or acetamido groups also reduced the potency, possibly because there was a hydrogenbonding interaction between the phenolic hydroxyl group and
carbonyl oxygen of Glu118. Additionally, the alkyl replacements for the 5-aryl group led to loss of activity, since the aryl
ring might form a hydrophobic interaction between Pro137
and Arg119, which was beneficial to the potency [27].
Among the N1-acetyl-3-aryl-5-phenyl-4,5-dihydropyrazoles, the 3-(2-chlorophenyl)-substituted analogue inhibited
Fig. (2). Docking of MK-0731 into the allosteric binding site of KSP using SYBYL-X 1.3 (PDB: 3CJO).
Cl
3
2 N
F
4
5
H2N
OH
OH
N
N 1
COCH3
( )n
N
COR
COCH3
NHEt
O
6: n = 3; 7: n = 4
R
H2N
11: R =
H2N
NAc
F
F
F
F
12: R =
13: R =
F
N
N
CH3
NMe2
O
(S)-8
N
O
14: R =
S
CH3
15
R
COCH3
N
CH3
N
N
16
17: X = F, R =
O
18: X = F, R = Me2N
19: X = Me, R = Me2N
NAc
9
10
11
12
13
14
15
16
17
18
19
(R)-19
0.9
1.4
2.8
1.8
1.0
1.8
0.82
1.6
3.8
2.1
2.0
0.2
MDR ratio
88
4.4
2.5
1.7
3.4
1.1
5.2
1.0
1.6
Zhao et al.
were seen for the MDR ratio, wherein amines with reduced
basicity, and tertiary amines tended to have improved potency against the Pgp-overexpressing cell line with lower
MDR ratios [30]. The X-ray structure analysis showed that
both aromatic rings in 16 formed hydrophobic interactions
with the nonpolar pocket of KSP, the tethered amine at C4
occupied the unoccupied region above the plane of the core
dihydropyrazole scaffold, and the C3 acetyl group projected
into the solvent exposed area of the enzyme [31].
Further modifications by removing the ring strain in 16
gave a racemate 17 that maintained excellent intrinsic potency (3.8 nM) while suffering a 16-fold loss in cellular potency (78 nM). Modulation of the amine to a tertiary heterocyclic amine or thiomorpholine dioxide afforded compounds
with diminished cellular potency as compared to 17. Potency
in cells was regained when the acetylpiperazine was replaced
with various tertiary amines such as morpholine (53 nM), 3fluoroazetidine (40 nM), pyrrolidine (15 nM), and bridged
morpholine (11 nM). The dimethylamine analogue 18 exhibited excellent biochemical potency of 2.1 nM and cellular
potency of 9.4 nM. Replacement of the 5-F in 17 with 5-Cl
or 5-Br improved the cellular activity (up to twofold). Deletion of the 5-F substituent in 17 afforded a sixfold decrease
in intrinsic potency concomitant with a fivefold decrease in
cell potency. Incorporation of a trifluoromethyl group was
also deleterious with respect to potency. However, changing
the 5-position fluorine to a methyl group afforded compound
19 which gave a twofold and threefold improvement in intrinsic (2.0 nM) and cellular (27 nM) potency as compared to
17. Especially, the enantiomer (R)-19 exhibited excellent
biological activities. (R)-19 was eightfold more potent than
dihydropyrazolobenzoxazine 16 in KSP inhibition (0.2 nM).
It was particularly potent in cells (3.2 nM) showing approximately equivalent potency to 16 (5.0 nM). (R)-19 also
demonstrated good aqueous solubility (> 4 mg/mL) at moderate pH of 4, showed good potency in Pgp-overexpressing
cells (MDR ratio of 1.6), and exhibited full mitotic arrest in a
mouse xenograft model of cancer with low plasma exposure
(EC99 = 67 nM) [31].
2. DIHYDROPYRROLES
To mimic the 3,5-diaryl-4,5-dihydropyrazole pharmacophore, a series of 2,4-diaryl-2,5-dihydropyrroles were developed as potent KSP inhibitors. Introduction of basic amide and urea moieties at N1 position of the 2,4-diaryl-2,5dihydropyrroles could improve the KSP binding potency and
aqueous solubility, while combination of C2 phenol group
with neutral N1 side chain could minimize the hERG binding
affinity to improve their selectivity [32]. The 4-piperidinyl
urea 20 (Fig. 4) was the most potent among the urea series
and had a good aqueous solubility (> 10 mg/mL pH = 5).
The crystal structure of compound 20 bound to the allosteric
site of KSP revealed that the N1 piperidinyl group projected
into the polar solvent exposed region (above Trp127 and
Tyr211) away from the hydrophobic pocket inside and reduced the energy for desolvation. The N-methyl group directed into a hydrophobic region below the dihydropyrrole
scaffold that was benefit to the potency. However, compound 20 had a low IKr potassium channel hERG binding
IC50 value. High hERG binding affinity could cause alteration of cardiac ventricular repolarization, prolongation of QT
4
R
Zhao et al.
R2
OH
2
N
N1
R1
CH3
NH2
OH
20: R = H; 21: R = Ac
H
N
R
KSP IC50
(nM)
Cell EC50
(nM)
hERG IC50
(M)
2.6
50
3.6
2.0
1.2
0.5
2.9
1.3
7.0
4.0
6.8
69
12
8.6
3.3
3.0
3.1
2.6
22
9.1
1.3
>10
2.4
3.5
8.5
5.6
18
7.8
33
15
20
21
22
23
24
25
26
27
28
29
OH
CH3
28: R = Me; 29: R = CH2CH2OH
H2N
H2N
HO
F
R
N
N
CH3
N
H3C
CH3
30-33
R
H3C
CH3
34
KSP IC50
MDR ratio
(nM)
pKa
O
3
5
hERG IC50
(M)
30
7.4
345.1
9.8
19.2
31
6.2
21.2
8.8
14.6
32
Me
CH2CH2F
5.0
4.5
7.6
13.8
33
cyclopropyl
5.9
1.2
7.5
15.2
20.5
MK-0731
2.2
4.5
7.6
34
2.2
1200
10.3
7.1
35
5.2
5.0
7.0
15.9
Consistent with its favorable pKa, MK-0731 also had significant activity in Pgp-overexpressing cells, with potencies
of 4.2 and 18.9 nM in the parental and Pgp-overexpressing
cell lines, respectively, resulting in an MDR ratio of 4.5. It
had the ability to induce a mitotic block with an IC50 of 19
nM in cells that were refractory to paclitaxel due to Pgp
overexpression. In addition, it was >20000-fold selective for
KSP over a panel of eight structurally and functionally related kinesins. The inhibition of KSP by MK-0731 was not
competitive with either ATP or microtubules. Unlike taxanes
and the epothilones, MK-0731 had no effect on microtubule
polymerization in vitro at 20 M. It induced apoptosis in
A2780 cells with an EC50 of 2.7 nM. In mouse xenograft
assays, MK-0731 induced dose-dependent mitotic arrest in
tumors and inhibited tumor growth comparable to paclitaxel.
Installation of an aminopropyl side chain to the C2 position of the dihydropyrrole scaffold resulted in a potent, water
soluble KSP inhibitor (34), but the aminopropyl group induced susceptibility to cellular efflux by P-glycoprotein
(Pgp). Substituting the primary amine in 34 with an ethyl, a
fluoroethyl, or a difluoroethyl group decreased the KSP
binding potency (5-fold). In addition, the ethylated analogue
remained a good substrate for Pgp efflux (MDR ratio > 135),
the fluoroethyl compound had a significant improvement in
the MDR ratio (= 32), while the difluoroethyl analogue had a
good MDR ratio of almost unity (= 3). Trifluoroethylation of
the primary amino group continued the trend of reduced
MDR ratio (= 1), but also suffered a substantial reduction in
KSP potency (55-fold). However, by carefully modulating
the basicity of the amino group through introducing a
difluoromethyl substituent at the -position of the primary
amine group led to compound 35 that maintained potency
against KSP and had greatly improved efficacy in a Pgpoverexpressing cell line [36]. The discovery that cellular
efflux by Pgp could be overcome by carefully modulating
the basicity of an amine might be of general use to medicinal
chemists attempting to transform leading compounds into
cancer cell- or CNS-penetrant drugs.
3. THIOPHENES
Ultra high-throughput screening of the Kalypsys compound collection identified the thiophene scaffold for the
development of novel KSP inhibitors [37]. The analogue that
had no substituents at the thiophene C4 and C5 positions was
inactive (IC50 > 100 M) in microtubule-stimulated KSP
ATPase activity assay. Some improvement was observed
with dimethyl substitution at the thiophene C4 and C5 positions, but more dramatic effects were observed with fused
aliphatic 5- to 7-membered rings appended onto the thiophene. In particular, the cyclohexyl analogue 36 (Fig. 6),
with activity in the ATPase assay of 1.7 M and the MPM-2
cytoblot assay of 0.48 M, became a scaffold for further
optimization. Replacements of the 2-thiophene in the amide
portion of 36 with the 3-thiophene, thiazole, methyl thiazole,
furan, and phenyl led to a log or more loss of activity. Removing the amide carbonyl of 36 also led to a log loss of
activity. The co-crystal structure of KSP with 36, Mg2+, and
ADP revealed that the cyclohexylthiophene core of 36 occupied the same hydrophobic pocket as the difluorophenyl
group in the 1,2,4-trisubstituted dihydropyrroles (e.g. 28).
One ethyl of the diethylamide group projected into solvent
and the other in the pocket formed by Tyr211, Leu214, and
Glu215. The pendant 2-thiophene amide of 36 filled the
4. DIHYDROTHIADIAZOLES
LY2523355 (Fig. 1), also known as KF89617 or Litronesib, was a dihydrothiadiazole analogue developed by Kyowa
Hakko Kirin [17, 38]. Litronesib is an ATP-noncompetitive,
allosteric, reversible KSP inhibitor (IC50 < 0.1 M) that has
no observed effect on microtubule formation. It was effective
against MV-4-11, HCT-116, A549 and SK-OV-3 cell lines
(GI50 < 10 M) [39, 40]. Litronesib has been investigated in
clinical trials for the treatment of multiple cancers by Eli
Lilly and Company [23]. The main toxicities of it were neutropenia and leucopenia. Grade 4 neutropenia was observed,
but all recovered to above 500/mm2 within 7 days. Japanese
patients at 4 or 5 mg/m2/day required G-CSF support, and
F
S
1
S
7
6
O
2
5
4
O
1
S
NH
NH
3N
H3C
N(Et)2
36
O
37
N(Et)2
38
5
N
6
H2N
Zhao et al.
nM; 2-F: 300 nM; 4-Br: 2400 nM; 4-Cl: 290 nM), although
multiple substitutions did not lead to further improvement
(2,6-F: 280 nM; 2,4-F: 330 nM; 2,4,6-F: 230 nM). Substitutions on the pyridyl ring in 40 reduced the potency (2-OH:
1400 nM; 2-MeO: 1800 nM; 3-Cl: 7000 nM). Replacement
of the 4-pyridyl in 40 with 3-pyridyl or substituted phenyl
ring also led to reduced potency (IC50: 8304400 nM).
CH3
1
S
N
1
2
3
N
N
39
40
N
41
3
R1
R2
CF3
1
N
H
42: R1 = Et, R2 = COCH2CH2NH2
43: R1 = Cl, R2 = CH2CH2NH2HCl
OH
X
R
N
H
44: X = N
45: R = NHCONH2, X = C
42
43
44
45
the carbazole (45) was more potent than that on the position. Carbazole derivatives fused with a five- to sevenmembered lactam ring at the ,-positions exhibited similar
activity, suggesting the great flexibility of the lactam carbonyl placement [50]. Insertion of an ether linkage between
the carbazole core and the CF3 group, and replacing the CF3
group with carbonyl groups could not improve the potency.
-Substituted carbazoles showed no or less KSP inhibitory
activity, and the t-Bu- or CF3- group at the - or - position
of carbazole core was the most favorable accessory group
[51].
CONCLUSION
Anti-mitotic chemotherapeutics, such as the taxanes and
vinca alkaloids, represent one of the main classes of effective
cancer therapies, and are broadly used against a wide range
of cancer types. However, mechanism related toxicities and
acquired resistance have stimulated considerable interest in
developing anti-mitotics that target mechanisms other than
direct microtubule inhibition. Kinesin spindle protein (KSP)
plays an essential role in centrosome separation and
formation of the bipolar mitotic spindle. Its exclusive
involvement in the mitotic spindle of proliferating cells
presents an opportunity for developing new anticancer agents
with reduced side effects relative to antimitotics that target
tubulin. A large number of KSP inhibitors based on many
distinct chemical scaffolds are currently in development.
This review summarized the recent developments of KSP
inhibitors based on the five-membered heterocycle scaffolds.
Dihydropyrroles and dihydropyrazoles represent the most
common chemical scaffolds that target an allosteric pocket in
the motor domain of KSP. These inhibitors do not inhibit
any other kinesins and their high specificity is based on a
particularly long loop L5 region in the motor domain, the
length of which is unique for KSP. They inhibit or slow
down ADP release, but do not interfere with ATP binding
during the ATP hydrolysis cycle. As discussed above, introducing suitable substituents and carefully modulating the
basicity of the dihydropyrroles and dihydropyrazoles could
minimize the hERG binding affinity to improve their selectivity, reduce their susceptibility to cellular efflux by Pglycoprotein (Pgp), and lead to potent KSP inhibitors with
good aqueous solubility and highly favorable pharmacokinetics. These modification strategies might be of general use
to medicinal chemists attempting to transform leading compounds into anticancer drugs. Compared with the allosteric
KSP inhibitors, the ATP competitive KSP inhibitors with
five-membered heterocycle scaffolds are less studied. The
thiazole ATP competitive KSP inhibitors may represent dis-
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
Zhao et al.
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[40]
[41]
[42]
[43]
[44]
[46]
[47]
[48]
[49]
[50]
[51]
Angeles, A.R.; Yang, L.; Dai, C.; Brunskill, A.; Basso, A.D.; Siddiqui, M.A. Synthesis, determination of absolute configuration, and
biological evaluation of spiro-fused thiadiazoline inhibitors of
kinesin spindle protein (KSP). Tetrahedron Lett., 2010, 51(48),
6236-6239.
Rickert, K.W.; Schaber, M.; Torrent, M.; Neilson, L.A.; Tasbet,
E.S.; Carbaccio, 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 competitive inhibitors of the human mitotic kinesin KSP. Arch. Biochem. Biophys., 2008, 469(2), 220-231.
Sheth, P.R.; Shipps Jr, G.W.; Seghezzi, W.; Smith, C.K.;
Chuang, C.C.; Sanden, D.; Basso, A.D.; Vilenchik, L.; Gray, K.;
Annis, D.A.; Nickbarg, E.; Ma, Y.; Lahue, B.; Herbst, R.; Le,
H.V. Novel benzimidazole inhibitors bind to a unique site in the
kinesin spindle protein motor domain. Biochemistry, 2010,
49(38), 8350-8358.
Barsanti, P.A.; Wang, W.; Ni, Z.J.; Duhl, D.; Brammeier, N.; Martin, E.; Bussiere, D.; Walter, A.O. The discovery of tetrahydro-carbolines as inhibitors of the kinesin Eg5. Bioorg. Med. Chem.
Lett., 2010, 20(1), 157-160.
Liu, F.; Yu, L.Q.; Jiang, C.; Yang, L.; Wu, W.T.; You, Q.D. Discovery of tetrahydro--carbolines as inhibitors of the mitotic kinesin KSP. Bioorg. Med. Chem. Lett., 2010, 18(12), 4167-4177.
Takeuchi, T.; Oishi, S.; Watanabe, T.; Ohno, H.; Sawada, J.; Matsuno, K.; Asai, A.; Asada, N.; Kitaura, K.; Fujii, N. Structureactivity relationships of carboline and carbazole derivatives as a
novel class of ATP-competitive kinesin spindle protein inhibitors.
J. Med. Chem., 2011, 54(13), 4839-4846.
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(13), 5054-5058.