Synthesis and biological evaluation of l-cysteine derivatives as mitotic kinesin Eg5 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2007.04.101

Inhibition of Eg5 represents a novel approach for the treatment of cancer. Here, we report the synthesis and structure-activity relationship of S-trityl-l-cysteine (STLC) derivatives as Eg5 inhibitors. Some of these derivatives such as 4f demonstrated enh

Full text of DOI:10.1016/j.bmcl.2007.04.101

Bioorganic & Medicinal Chemistry Letters 17 (2007) 3921–3924
Synthesis and biological evaluation of L-cysteine derivatives
as mitotic kinesin Eg5 inhibitors
Naohisa Ogo,a Shinya Oishi,b Kenji Matsuno,a Jun-ichi Sawada,a
Nobutaka Fujiib and Akira Asaia,*
aCenter for Drug Discovery, School of Pharmaceutical Sciences, University of Shizuoka,
52-1 Yada, Suruga-ku, Shizuoka 422-8526, Japan
bGraduate School of Pharmaceutical Sciences, Kyoto University, Sakyo-ku, Kyoto 606-8501, Japan
Received 7 March 2007; revised 26 April 2007; accepted 30 April 2007
Available online 3 May 2007
Abstract—Inhibition of Eg5 represents a novel approach for the treatment of cancer. Here, we report the synthesis and structure-
activity relationship of S-trityl-L-cysteine (STLC) derivatives as Eg5 inhibitors. Some of these derivatives such as 4f demonstrated
enhanced inhibitory activity against Eg5 and induced mitotic arrest with characteristic monoastral spindles in HeLa cells.
Ó 2007 Elsevier Ltd. All rights reserved.
Antimitotic agents such as taxol, epothilone, and vinca
alkaloids have found clinical utility as cancer chemo-
therapeutic agents.1 These agents bind to the b-tubulin
protein a component of the mitotic spindle microtu-
bules. Although microtubules play important roles
throughout the cell cycle of cancer cells, disruption of
microtubule dynamics also produces undesirable side ef-
fects, such as toxicity in non-dividing cells like periphe-
ral neurons, leading to peripheral neuropathy in
patients.2
of Eg5 inhibitors have been found and some of them ap-
peared as the therapeutic potential for anticancer
drugs.7
In order to find a new structural class of kinesin Eg5
inhibitors, we screened our chemical library using
in vitro ATPase assay measuring the test compound’s
ability to prevent the ATP hydrolysis by Eg5.8 As the re-
sults of screening efforts, we identified L-cysteine deriva-
tives 1 and 2 (Fig. 1) as kinesin Eg5 inhibitors. Although
compound 1 (S-trityl-L-cysteine, STLC) was known as
an Eg5 inhibitor,9,10 there were no reports on struc-
ture-activity relationships (SARs) for the related ana-
logues. Therefore, we decided to make synthetic efforts
on its derivatizations to find potent and selective inhib-
itors of Eg5. In this paper, we describe the synthesis of
L-cysteine derivatives and their inhibitory activity
against Eg5 as well as their ability to inhibit cell division.
The microtubule associated kinesin Eg5 also known as
kinesin spindle protein (KSP), a member of the kine-
sin-5 family, plays an important role in the early stages
of mitosis. It is responsible for the formation and main-
tenance of the bipolar spindle.3 Inhibition of Eg5 leads
to cell cycle arrest during mitosis and causes cells with
a monopolar spindle, so called monoasters.4 Because
Eg5 is not expressed in postmitotic neurons and is likely
to act only in dividing cells, its inhibitors might provide
better specificity than microtubule inhibitors in the
treatment of human malignancies.5 In 1999, Monastrol,
the first small-molecule inhibitor of Eg5, was discovered
in a phenotype-based screening.6 Since then, a number
Keywords: S-Trityl-L-cysteine; Mitotic kinesin; Eg5; Inhibitor;
Anticancer.
*
Corresponding author. Tel./fax: +81 54 264 5231; e-mail: aasai@u-
shizuoka-ken.ac.jp
Figure 1. Eg5 inhibitors: Monastrol, cysteine derivatives (1), and (2).
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.04.101

3922
N. Ogo et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3921–3924
Table 1. SAR of thiol substituent of L-cysteine
against Eg5 and an anti-proliferation activity in HeLa
cells (Table 1).11 Surprisingly, all examination of substit-
uents on the thiol group in the L-cysteine resulted in loss
of potency in both assays (3a–i). Based on these initial
results, we expected that cysteine derivatives containing
triarylmethyl group would have a greater potential for
high inhibition levels of the Eg5. Next we focused on
modifications for the triarylmethyl, amino, and carboxyl
moieties in STLC.
Compound
R
ATPase
a
IC50 (lM)
Cytotoxicity
a
IC50 (lM)
3a (L-cysteine)
3b
H
CH3
>63
>63
>63
>63
>63
>63
>S3
>63
>63
1.80
>50
>50
>50
>50
>50
>50
>50
>50
>50
3.29
3c
3d
t-Bu
The synthetic procedures for the target compounds are
summarized in Schemes 1 and 2. Compound 4f was eas-
ily prepared by treating L-cysteine with commercially
available triarylmethyl chloride 5.12 Compound 4f was
converted to the methyl ester 4i with SOCl2 in MeOH,
and subsequently acylation by AcCl led to compound
4j. Compounds 4g and 4h were synthesized as follows;
triphenylmethanols 8 and 9 were prepared by treating
corresponding aryl bromide 6 or 7 with n-BuLi at
ꢀ78 °C in THF to form aryl lithium which upon treat-
ment with benzophenone, and condensation with L-cys-
teine in the presence of TFA provided target compounds
4g and 4h, respectively.
CH2Ph
3e
CH2-40-anisole
CH2COOH
CH2CH2NH2
CH-9H-fluorene
CHPh2
3f
3g
3h
3i
1 (STLC)
CPh3
a IC50 values were derived from dose–response curves generated from
triplicate data points.
As expected, triarylmethyl group in L-cysteine deriva-
tives was essential for high inhibition level of both in
Eg5 ATPase and cytotoxic activities and close correla-
tion for inhibitory activities between Eg5 ATPase and
cytotoxicity was observed across almost all compounds
(Tables 1 and 2). In general, protection of the amino ter-
minal in 1 resulted in loss of both activities (4a–d, 4j).
Modification of the carboxylic acid terminal in 1 to
the primary amide as in 2 or the methyl ester as in 4e
yielded little loss of Eg5 ATPase activity. On the other
hand, cytotoxicity of 2 and 4e was slightly improved
compared to 1, presumably reflecting different cell per-
meability properties. There is a possibility that the major
cellular uptake pathway of STLC is amino acid trans-
porter such as System L. System L and other amino acid
transporters require free amino and carboxyl moiety as
substrate.13 However, we showed that the cytotoxicity
of 2, 4e, and 4i was improved compared to 1. These re-
sults suggest that hydrophobicity of STLC derivatives is
more important for incorporation.
Scheme 1. Reagents and conditions: (a) DMF, rt; (b) SOC12, MeOH,
0 °C then reflux; (c) AcC1, pyridine, CH2C12, rt.
Drastic changes both in Eg5 ATPase and cytotoxic
activities were observed by incorporation of substituents
onto the phenyl ring (1 versus 4f–h, 4e versus 4i).
Remarkably, compound 4f showed >10-fold potent
inhibition both in Eg5 ATPase and cytotoxic activities
as compared to 1. However, incorporation of 3,4-dime-
thoxy substituents onto the phenyl ring resulted in loss
of potency (4g). Among 4-MeO analogues (4f, 4i, and
4j), the same SAR for L-cysteine moiety above men-
tioned was obtained. That is, modification of carboxylic
acid terminal to methyl ester yielded little loss of Eg5
ATPase activity (4f versus 4i), and acetyl protection of
amino terminal resulted in activity losses of the both
(4i versus 4j). Introduction of trifluoromethyl group
provided compound 4h with sevenfold increase in Eg5
ATPase and sixfold increase in cytotoxic activity over
1. We then chose three compounds (4f, 4h, and 4i) and
evaluated for inhibition of four other structurally and
Scheme 2. Reagents and conditions: (a) n-BuLi, THF, ꢀ78 °C; then
benzophenone, THF; (b) L-cysteine, TFA, rt.
During our first attempts to develop a SAR and to iden-
tify the key features that would drive the inhibition, we
focused on modifying the thiol substituents in the
L-cysteine. At first, we confirmed the lack of potency
of commercially available triphenylmethane and triphe-
nylmethanethiol
against
Eg5
(ATPase
assay
IC50 > 63 lM). Then, we purchased compounds (3a–i)
from several suppliers and evaluated for their activity

N. Ogo et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3921–3924
Table 2. SAR of side chain (R1, R2) and phenyl substituent (R3)
3923
R1
R2
R3
ATPase IC50 (lM)
Cytotoxicity IC50 (lM)
Mitotic arrest MI50 (lM)
a
a
b
Compound
1 (STLC)
2
H
OH
H
1.8
3.1
3.29
1.78
>50
11.9
>50
16.5
1.72
0.21
49.7
0.53
0.32
>50
1.4
H
NH2
OH
H
NT
NT
NT
NT
NT
NT
0.15
NT
0.34
NT
NT
4a
4b
4c
4d
4e
4f
4g
4h
4i
Ac
Fmocc
Boc
Trityl
H
H
>63
55.0
>63
>63
5.8
OH
OH
H
H
OH
H
OMe
OH
OH
H
H
4-OMe
3,4-Dimethoxy
4-CF3
4-OMe
4-OMe
0.15
14.8
0.22
0.68
>63
H
H
OH
OMe
OMe
H
Ac
4j
a IC50 values were derived from dose–response curves generated from triplicate data points.
b MI50 values were derived from dose–response curves generated from triplicate data points; NT, not tested.
c Fmoc; 9-fluorenylmethoxycarbonyl.
Table 3. ATPase inhibitory activity of selected compoundsa
we further characterized the cellular activity of com-
pounds 4f and 4h. Cells with suppressed Eg5 function,
for example by RNAi or small molecule inhibitors, dis-
play a characteristic monoastral phenotype and arrest
in mitosis. HeLa cells treated with compounds 4f or 4h
for 18 h, corresponding to approximately one cell cycle,
showed an increase in the mitotic index relative to un-
treated cells. The MI50 values (concentration of 50%
induction in M phase in HeLa cells at 20 h) of both of
compounds (4f and 4h) were lower than that of STLC
(Table 2).15 We next examined the mitotic phenotype of
the cells treated with compound 4f.16 As shown in the
fluorescent microscopy images of Figure 2, compound
4f induced the formation of monopolar mitotic spindles
in HeLa cells after 20-h treatment. Multiple monopolar
mitotic spindles but no bipolar mitotic spindles were ob-
served (Fig. 2). Several DAPI-stained chromosomes
(blue) are detected in the focal plane with microtubules
of the mitotic spindle (red). As expected, the monoastral
phenotype was observed in cells treated with 4f.
Compound
Eg5
CENP-E
Kid
MKLP-1
KIF-4
1 (STLC)
89
98
0
0
0
0
0
0
0
0
2
10
7
4
5
3
4
4f
4h
4i
79
100
5
a The inhibition values were determined by ATPase assay and are
indicated as a percentage of the solvent only control. Values are the
means of three independent measurements. The final concentration
of the compounds in the assay was 20 lM.
In conclusion, we have identified specific and potent cell-
permeable inhibitors of the mitotic kinesin Eg5. Struc-
tural optimization of the screening hits 1 and 2 led to
the sub-micromolar inhibitors 4f, 4h, and 4i. The most
potent compound 4f was >10-fold more potent than ini-
tial compound 1, both in Eg5 ATPase and cytotoxic
activities. Future efforts will focus on further improve-
ment in potency for this compound series.
Figure 2. HeLa cells with normal spindle (B) and with monoasters
after treatment with compound 4f (1 lM) (A and C).
Acknowledgments
functionally related mitotic kinesins (CENP-E, kid,
MKLP1, KIF4).14 All these kinesins were not inhibited
by the tested compounds including 1 at 20 lM, respec-
tively (Table 3). These results indicate that the selected
derivatives are specific Eg5 inhibitors as STLC is.
We thank Ms. Keiko Fukamoto for excellent technical
assistance. This work was supported by the Drug Dis-
covery Program of the Pharma Valley Center and by
Grant-in-Aid for Scientific Research from Ministry of
Education, Culture, Sports, Science, and Technology
of Japan.
In order to confirm that the cytotoxicity displayed by the
STLC derivatives was a consequence of Eg5 inhibition,

3924
N. Ogo et al. / Bioorg. Med. Chem. Lett. 17 (2007) 3921–3924
10. Skoufias, D. A.; Debonis, S.; Saoudi, Y.; Lebeau, L.;
References and notes
Crevel, I.; Cross, R.; Wade, R. H.; Hackney, D.; Koziel-
ski, F. J. Biol. Chem. 2006, 281, 17559.
1. Dutcher, J. P.; Novik, Y.; O’Boyle, K.; Marcoullis, G.;
Secco, C.; Wiernik, P. H. J. Clin. Pharmacol. 2000, 40,
1079.
2. Quasthoff, S.; Hartung, H. P. J. Neurol. 2002, 249, 9.
3. Blangy, A.; Lane, H. A.; d’Herin, P.; Harper, M.; Kress,
M.; Nigg, E. A. Cell 1995, 83, 1159.
11. Cytotoxic activity against HeLa cells was examined using
CellTiter96 MTS reagent (Promega) after exposure for
48 h to STLC derivatives. Each growth inhibitory con-
centration was reported as the average of at least three
independent determinations.
12. Rudolph, J.; Theis, H.; Hanke, R.; Endermann, R.;
Johannsen, L.; Geschke, F. U. J. Med. Chem. 2001, 44, 619.
13. Uchino, H.; Kanai, Y.; Kim, D. K.; Wempe, M. F.;
Chairoungdua, A.; Morimoto, E.; Anders, M. W.; Endou,
H. Mol. Pharmacol. 2002, 61, 729.
14. Other kinesins’ inhibitory activity was assessed by measur-
ing the amount of ATP resulting from in vitro reaction using
Kinase-Glo Plus luminescent kinase assay (Promega). The
in vitro ATPase assay was performed as described as Eg5.
15. Mitotic arrest was measured by assessing the mitosis-
specific phosphorylation of Histone H3 using anti-
phospho-Histone H3 (Ser10) antibody (Upstate). The
positive cells to the antibody were counted under the
microscope while total cell number was measured by
assessing the DAPI-stained cells.
16. HeLa cells treated with STLC derivatives were fixed by the
conventional paraformaldehyde/methanol method, and
stained immunologically. Mitotic chromatin was stained
by anti-phospho-Histon H3 (Ser10) antibody (Upstate).
Microtubules and DNA/chromatins were stained by mono-
clonal anti-a-tubulin (DM1A) antibody (Sigma) and DAPI,
respectively. Fluorescent images were observed under
Olympus IX71 microscope with DP30BW CCD camera.
4. Walczak, C. E.; Vernos, I.; Mitchison, T. J.; Karsenti, E.;
Heald, R. Curr. Biol. 1998, 8, 903.
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. Cancer Res. 2004, 64,
3276.
6. Mayer, T. U.; Kapoor, T. M.; Haggarty, S. J.; King, R.
W.; Schreiber, S. L.; Mitchison, T. L. Science 1999, 289,
971.
7. (a) Bergnes, G.; Brejc, K.; Belmont, L. Curr. Top. Med.
Chem. 2005, 5, 127; (b) Duhl, D. M.; Renhowe, P. A.
Curr. Opin. Drug Discov. Dev. 2005, 8, 431.
8. Assays for Eg5 ATPase activity were carried out in 96-well
half-area plates. Our typical reaction contained 38 nM of
recombinant Eg5, 30 lM ATP, and 350 nM taxol-stabi-
lized microtubules in 15 ll of the reaction buffer with
20 lM of small molecules. The microtubule-activated Eg5
ATPase activity was evaluated using Kinase-Glo lumines-
cent kinase assay (Promega).
9. Debonis, S.; Skoufias, D. A.; Lebeau, L.; Lopez, R.;
Robin, G.; Margolis, R. L.; Wade, R. H.; Kozielski, F.
Mol. Cancer Ther. 2004, 3, 1079.