New chemical tools for investigating human mitotic kinesin Eg5

Article abstract of DOI:10.1016/j.bmc.2007.06.016

We have designed and synthesized a series of monastrol derivatives, an allosteric inhibitor of Eg5, a motor protein responsible for the formation and maintenance of the bipolar spindle in mitotic cells. Sterically demanding structural modifications have been introduced on the skeleton of the parent drug either via a multicomponent Biginelli reaction or a stepwise modification of monastrol. The ability of these compounds to inhibit Eg5 activity has been investigated using two in vitro steady-state ATPase assays (basal and microtubule-stimulated) as well as a cell-based assay. One compound in the series appeared more potent than monastrol by a fivefold factor. Three other compounds that were unable to inhibit Eg5 ATPase activity in vitro proved potent Eg5 inhibitors in the cell-based assay. The results obtained led to the identification of structure-activity relationships further used to design an affinity matrix that can be used for fast and efficient purification of Eg5 from crude lysate of eukaryotic cells.

Full text of DOI:10.1016/j.bmc.2007.06.016

Bioorganic & Medicinal Chemistry 15 (2007) 6474–6488
New chemical tools for investigating human mitotic kinesin Eg5
Emmanuel Klein,a Salvatore DeBonis,b Bernd Thiede,c Dimitrios A. Skoufias,b
Frank Kozielskib and Luc Lebeaua,*
aInstitut Gilbert-Laustriat, CNRS-Universite´ Louis Pasteur, Faculte´ de Pharmacie, 74 route du Rhin, 67401 Illkirch, France
bInstitut de Biologie Structurale, 41 rue Jules Horowitz, 38027 Grenoble Cedex 01, France
cUniversity of Oslo, The Biotechnology Centre of Oslo, Gaustadalleen 21, PO Box 1125 Blindern, 0317 Oslo, Norway
Received 22 December 2006; revised 24 May 2007; accepted 5 June 2007
Available online 10 June 2007
Abstract—We have designed and synthesized a series of monastrol derivatives, an allosteric inhibitor of Eg5, a motor protein
responsible for the formation and maintenance of the bipolar spindle in mitotic cells. Sterically demanding structural modifications
have been introduced on the skeleton of the parent drug either via a multicomponent Biginelli reaction or a stepwise modification of
monastrol. The ability of these compounds to inhibit Eg5 activity has been investigated using two in vitro steady-state ATPase
assays (basal and microtubule-stimulated) as well as a cell-based assay. One compound in the series appeared more potent than
monastrol by a fivefold factor. Three other compounds that were unable to inhibit Eg5 ATPase activity in vitro proved potent
Eg5 inhibitors in the cell-based assay. The results obtained led to the identification of structure–activity relationships further used
to design an affinity matrix that can be used for fast and efficient purification of Eg5 from crude lysate of eukaryotic cells.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
mitosis with a monoastral spindle phenotype. Human
Eg5 has been recently identified as one of the potential
targets of 4-(3-hydroxy-phenyl)-6-methyl-2-thioxo-
1,2,3,4-tetrahydro-pyrimidine-5-carboxylic acid ethyl es-
ter 1, further named monastrol9,10 (Scheme 1). That
racemic dihydropyrimidone (DHPM) is an allosteric
inhibitor of Eg511 and recent in vitro studies with cul-
tured neurons showed that it does not display any toxic-
ity, short term treatment even enhancing axonal
growth12,13 in contrast to anticancer drugs such as the
taxanes which are highly deleterious to axonal forma-
tion and growth. Whether these results will translate
into a lack of neurotoxicity in animals or even in hu-
mans is a key question which has not been answered
so far.
Kinesin motors are specialized enzymes that use ATP
hydrolysis to generate force and movement along their
cellular tracks, the microtubules (MTs).1 They are impli-
cated in cell division and therefore have generated inter-
est as potential protein targets in the treatment of
cancer.2–4 Indeed one of the major drawbacks associated
with anticancer drugs that perturb mitosis (e.g., vinca
alkaloids, taxanes, epothilones,. . .) is that they are all di-
rected against the same protein, tubulin, the microtubule
subunit which forms the mitotic spindle.5 As microtu-
bules are also involved in many other cellular processes
such as maintenance of organelles and cell shape, cell
motility, synaptic vesicles, and intracellular transport
phenomena, interference with their formation or depoly-
merization often leads to dose-limiting side effects like,
for example, peripheral neuropathy.6
DHPM is obtained via probably what is one of the first
described multicomponent reactions. Indeed the venera-
ble Biginelli dihydropyrimidine synthesis was discovered
by Pietro Biginelli who reported in 1893 on the acid-cat-
alyzed cyclocondensation reaction of ethyl acetoacetate,
benzaldehyde, and urea14 (Scheme 1). During more than
a century, only few chemists investigated the synthetic
utility and scope of the Biginelli reaction and less than
40 reports have appeared in the literature before 1999.
The discovery of the potential anticancer activity of
monastrol probably is not irrelevant to, or at least
One of these targets in the kinesin superfamily, Eg5, a
member of the kinesin-5 family, is responsible for the
formation and maintenance of the bipolar spindle.7,8
When inhibited by specific inhibitors, cells arrest in
Keywords: Antitumor agent; Inhibitor; Molecular motor; Eg5; Mon-
astrol; Structure–activity relationships.
*
Corresponding author. E-mail: lebeau@bioorga.u-strasbg.fr
0968-0896/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2007.06.016

E. Klein et al. / Bioorg. Med. Chem. 15 (2007) 6474–6488
6475
R2
R2
OH
OH
1 (monastrol) : R1=Et; R2,R3,R4=H.
2 : R1=(CH2CH2O)4-H; R2-R4,R3,R4=H.
3 : R1=Et; R2,R3=H; R4=(CH2CH2O)3-H.
4 : R1=Et; R2=OH; R3,R4=H.
R3
O
R3
HN
O
Cat.
5 : R1=Et; R2=(OCH2CH2)3-OH; R3,R4=H.
6 : R1=Et; R2=H; R3=NO2; R4=H.
7 : R1=Et; R2=H; R3=OH; R4=H.
CHO
OR1
OR1
NH2
S
NH
R4
O
S
N
R4
Scheme 1. The Biginelli condensation reaction and structure of monastrol 1 and analogs.
coincides with, an outbreak in the interest for the old
transformation as more than 350 articles have been pub-
lished on the Biginelli condensation since 1999.
4-alkoxy-3-hydroxybenzaldehyde, respectively. Simi-
larly, compounds 6 and 7 were obtained starting from
5-hydroxy-2-nitro-benzaldehyde and 2,5-dihydroxy-
benzaldehyde, respectively. However direct condensa-
tion of 2,5-dihydroxy-benzaldehyde with thiourea and
ethyl acetoacetate in the Biginelli reaction failed to yield
7 whatever the experimental conditions used. The target
compound was finally obtained via a two-step procedure
involving 2,5-diacetoxy-benzaldehyde in the Biginelli
condensation reaction to form 4-(2,5-diacetoxy-phe-
nyl)-6-methyl-2-thioxo-1,2,3,4-tetrahydro-pyrimidine-5-
carboxylic acid ethyl ester 8 that was subsequently
submitted to careful hydrolysis of acetates (Scheme 2).
Both compounds 7 and 8 were poorly stable explaining
the low yields obtained in these two transformations (see
Section 4). Compounds 9–16 were obtained via stepwise
transformations of monastrol (Scheme 3). Monoacetyla-
tion of monastrol 1 gave compound 9, and diacetylated
compound 10 was obtained as a side product. Hydroly-
sis of ethyl ester in 1 afforded acid 11 that was further
condensed with different alcohols and amines to yield
12, 13, 15, and 16. Saponification of thioacetate 13 led
to thiol 14. A first modification at C2 on the pyrimidine
ring was introduced by substituting thiourea for sele-
nourea in the Biginelli reaction to get compound 17
(Scheme 4). Adapting the alternate strategy for the syn-
thesis of dihydropyrimidones developed by Atwal,19 we
prepared as well S-alkylated compounds 18 and 19.
In the course of a research program focused on the
structural analysis of kinesins, especially Eg5, we were
confronted with the necessity to probe and purify kine-
sins from different sources as well as mutants. Thus we
synthesized a series of monastrol derivatives for a rapid
determination of the structure–activity relationship,
only scarce data being available in the literature.9,15,16
The results obtained from in vitro and cell-based assays
allowed us to design an affinity matrix for purification of
Eg5 recombinantly expressed in Escherichia coli and
from HeLa cells. Herein we describe the synthesis of
such monastrol derivatives, their potency to inhibit
Eg5 ATPase activity, and their mitotic arrest phenotype.
We report as well on the immobilization of a monastrol
derivative onto a polymer matrix. Finally a protocol for
the quick and efficient purification of full-length Eg5
from eukaryotic cells is discussed.
2. Results and discussion
2.1. Synthesis of monastrol derivatives
We coarsely identified three possible anchoring sites on
the monastrol backbone. The first one is the ester moiety
(C5), the second one is the allylic nitrogen atom (N1),
All the compounds prepared were analyzed for their
ability to inhibit Eg5 by using in vitro steady state ATP-
ase assays (basal and microtubule-stimulated activities)
as well as a cell-based assay leading to the induction
of monastrol spindles in HeLa cells (see Section 4).
The results obtained are reported in Table 1.
0
0
and the last one is the phenol ring (C4 and C6 ). A fourth
possibility for structural modification of the monastrol
backbone has been investigated consisting in alkylation
or replacement of sulfur at C2 on the pyrimidine ring.
The different monastrol derivatives were synthesized
either via a one-pot classical Biginelli procedure or by
a linear reaction sequence starting from monastrol.
Compound 2 was prepared applying the classical Bigi-
nelli condensation procedure14 with 3-hydroxybenzalde-
hyde, thiourea, and 2-{2-[2-(2-hydroxy-ethoxy)-ethoxy]-
ethoxy}-ethyl acetoacetate (Scheme 1). Compound 3
was synthesized using a similar protocol starting from
the corresponding N-substituted thiourea, 3-hydroxy-
benzaldehyde, and ethyl acetoacetate. The low yield
(7%) obtained for that reaction accounts for the poor
tolerance of the Biginelli reaction with regard to urea
N-substitution.17,18 Compounds 4 and 5 resulted from
the condensation of thiourea and ethyl acetoacetate with
3,4-dihydroxybenzaldehyde and the corresponding
OAc
OH
1-Ac2O
2-Biginelli
AcO
O
HO
HN
OEt
CHO
S
N
H
8
7
MeOH/NH3
Scheme 2. Synthesis of compounds 7 and 8.

6476
E. Klein et al. / Bioorg. Med. Chem. 15 (2007) 6474–6488
OAc
O
OH
OH
O
O
R
S
OH
2
SH
2
N
OEt
HN
O
HN
O
N
H
S
N
S
N
H
9: R=H
10: R=Ac
12
14
H
PPh3
DIAD
O
H
2
HCl
OH
O
HO
AcCl
Et3N
OH
O
H
O
KOH
AcS
2
1
S
2
PPh3
DIAD
HN
O
HN
OH
O
S
N
H
S
N
11
13
H
PPh3, DIAD
HO
O
OH
2
EtNH2•HCl
DCC, Et3N
OH
O
HO
OH
O
O
O
HN
O
O
NH
HN
N
H
S
N
H
N
S
S
N
H
15
16
H
Scheme 3. Synthesis of compounds 9–16.
than monastrol by a fourfold factor. Increasing the
length of the amide chain confirms that effect (com-
pound 23, see Scheme 5). However, introduction of a
hydrophilic end group (NH2, compound 24) at the
opposite of the chain has a reverse effect, presumably be-
cause of improved aqueous solubility. The same is ob-
served when the ethyl ester is replaced with a long
chain ester (compound 2). For esters with intermediate
chain length (compounds 12–14), the initial basal ATP-
ase activity of parent monastrol is preserved. It is likely
that the difference in potency between 2 (long chain es-
ter) and 12 (short chain ester) does not strictly result
from steric hindrance within the binding site of the pro-
tein. If so, compound 16 would exhibit a poor inhibition
score which is not the case (IC50 10 lM) even consider-
ing the dimeric structure of the compound. In the same
line of thought, the substitution of the free hydroxyl
group at the extremity of the diethylene glycol arm in
12 for the bulkier acetylsulfanyl moiety in 13 does not
result in a decrease of the activity, on the contrary. To
explore one step further the influence of substitution at
C5, we introduced a furyl residue as a planar sterically
demanding substituent at that position. The correspond-
ing compound 20 (Fig. 1) proved more potent than
monastrol by a fourfold factor. Finally, modifications
performed at C2 on the pyrimidine moiety gave con-
trasted results. Sulfur replacement for selenium (com-
pound 17, Scheme 4) provoked a twofold factor
decrease in the inhibition efficacy. A recent report by
Gartner et al. indicates that the same substitution for
oxygen is detrimental to activity as well.15 Concerning
S-alkylated compounds 18 and 19, no inhibition was ob-
served. That may be due either to steric or electronic (or
both) effects as both the size of the molecule and the
OH
O
OH
HN
OEt
H2N NH2
Se
Se
N
H
CH2OH
O
17
OH
O
NH
R
S
NH2
OEt
N
OEt
O
R
18: R=Et
19: R=Bn
S
N
H
Scheme 4. Synthesis of compounds 17–19.
2.2. Inhibition of basal Eg5 ATPase activity
We found that the compounds modified at the aromatic
0
0
ring in the C4 (compounds 4 and 5), C6 (compounds 6,
0
7, and 8), or O3 position (compounds 9 and 10) do not
inhibit Eg5 basal ATPase activity. Substitution at N1
(compound 3) was as well definitely detrimental to activ-
ity. More interesting are the compounds modified at the
carboxyl group in the C5 position. All of them are Eg5
inhibitors though a little less potent than monastrol in
general. Carboxylic acid 11 resulting from ethyl ester
hydrolysis in monastrol exhibits an IC50 value of
32 lM. Replacement of the ethyl ester in monastrol with
the corresponding ethyl amide leads as well to a decrease
in the basal ATPase activity, amide 15 being less active

E. Klein et al. / Bioorg. Med. Chem. 15 (2007) 6474–6488
6477
Table 1. Inhibition of basal and MT-stimulated Eg5 ATPase activity by different monastrol analogs (see Section 4 for details)
Inhibitor
Inhibition of basal
ATPase activity
IC50 (lM)
Inhibition of MT-
stimulated ATPase
activity IC50 (lM)
Monoastral spindles in cell-based assays
100 lM
50 lM
IC50 (lM)
Monastrol 1
6.1 1.5
23.0 9.0
n.i.
12.3 1.2
62.5 15.0
n.i.
94.4 3.4
16.8 7.0
13.6 2.1
n.d.
45.5 4.5
n.s.
n.s.
51.3
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
50.0
62.5
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
9.2
2
3
4
n.i.
n.i.
n.i.
n.i.
n.s.
n.s.
5
10.3 2.1
n.d.
n.d.
6
n.i.
n.i.
n.i.
n.i.
n.s.
7
MTs affected
56.2 10.0
49.7 11.7
39.3 3.2
n.s.
8
n.i.
n.i.
n.i.
n.i.
n.d.
9
89.6 3.7
83.5 4.4
15.5 2.7
50.5 4.8
32.5 8.5
15.2 1.3
17.8 4.1
11.4 3.8
33.2 5.7
n.s.
10
11
12
13
14
15
16
17
18
19
20
21
23
24
n.i.
n.i.
32.0 4.5
9.5 2.0
4.5 0.3
4.0 1.8
23.0 9.0
10.0 5.0
11.5 0.7
n.i.
72.2 17.0
21.0 10.0
11.5 2.0
5.5 2.8
82.3 20.2
24.0 0.4
23.0 7.0
n.i.
n.s.
n.s.
n.s.
n.s.
n.s.
n.d.
n.s.
n.s.
n.i.
1.5 0.1
n.i.
n.i.
3.0 0.1
n.i.
n.s.
n.d.
96.8 3.6
36.9 7.8
n.s.
n.d.
14.9 3.1
15.8 1.9
n.d.
n.d.
n.d.
39.0 15.0
22.0 3.0
69.1 2.0
18.3 2.0
n.s.
The percentage of mitotic cells showing monoastral spindles at drug concentrations 100 lM and 50 mM, and the calculated IC50 are indicated on the
right. n.i.: no inhibition at the higher concentration tested (500 lM). n.d.: not determined. n.s.: no significant induction of monoastral spindles in
treated cells (<25%).
OH
9
1) NaN3
2) PPh3
HCl
DCC
Et3N
O
O
O
Cl
Cl
H2N
N3
O
2
2
HN
N
H
N3
2
S
N
H
22
23
PPh3, H2O
O
O
O
N
O
OH
OH
O
O
O
O
O
HN
N
H
N
HN
N
NH2
2
2
H
H
S
N
H
S
N
24
25
H
Scheme 5. Synthesis of the monastrol-based affinity matrix.
electron density at the heterocycle are modified when
compared to parent monastrol.
active compounds do inhibit Eg5 microtubule-stimulated
ATPase activity though less efficiently than basal ATP-
ase activity. One exception is met with compound 14
presenting equivalent inhibition scores in both basal
and MT-stimulated assays and being twice as potent
as monastrol in the MT-stimulated assay (IC50
5.5 lM). The unexpectedly high inhibition of MT-stim-
ulated ATPase activity for compound 14 could result
from a covalent reaction of the free thiol group with
some cystein residue or disulfide bridge in the proximity
of the protein region interacting with microtubules, fur-
ther hindering or preventing stimulation of ATPase
activity by microtubules. Incubation under reductive
conditions (TCEP, 1 mM) however has no influence
on the activity of the compound and more sophisticated
That first set of results indicates there is no or little
restriction in the accessibility of the monastrol binding
site of Eg5 for dihydropyrimidinethiones with bulky
substituents in the C5 position. That conclusion is sup-
ported as well by recent results reported by Gartner
et al. who described interesting inhibition of Eg5 ATP-
ase activity by enastron and dimethylenastron (Fig. 1).15
2.3. Inhibition of microtubule-stimulated ATPase activity
Results are roughly parallel to those obtained in the
basal ATPase activity inhibition assay. The previously

6478
E. Klein et al. / Bioorg. Med. Chem. 15 (2007) 6474–6488
OH
O
OH
a
100
O
S
O
75
O
HN
N
OEt
50
25
0
S
N
H
N
H
20
21
OH
O
Control
6 μM
12 μM
25 μM
50 μM 100 μM
HN
R
S
N
H
Control
Monastrol
R
b
R = H : enastron (IC50 2 µM)
R = Me : dimethylenastron (IC50 0.2 µM)
Figure 1. Additional monastrol analogs referred to in the body text.
experiments are required to confirm or rule out that
hypothesis.
2.4. Cell-based assays
Compound 9
Compound 10
We looked at the ability of the different compounds to
induce monoastral spindle in asynchronous HeLa
cells.20 The compounds were evaluated at 50 and
100 lM, and the proportion of cells with the character-
istic monoastral spindle phenotype was quantified. The
higher the inhibition of ATPase activity, the more fre-
quent monoastral phenotype.
However, compound 13 which is more potent than mon-
astrol to inhibit both basal and microtubule-stimulated
Eg5 ATPase activities appears three times less active in
the cell-based assay. The same and even more striking
behavior is observed with compound 16 that does not
induce any monoastral spindle phenotype (11.4% mono-
astral spindles to be compared to 9.6% observed in the
absence of inhibitor) though it is a potent inhibitor of
basal as well as MT-stimulated Eg5 ATPase activities.
On the other hand, compounds 9 and 10 that do not in-
hibit at all Eg5 ATPase activity in the in vitro assays
proved to be among the most active compounds in the
cell-based assay and can be compared within that re-
spect to parent monastrol (Fig. 2). The IC50 of com-
pounds 9 and 10 were calculated to be 50.0 and
62.5 lM, respectively, and are close to that of monastrol
(51.3 lM). That result could be explained by a difference
in terms of drug delivery efficiency inside the cells during
the test. Indeed monastrol monoacetate 9 and diacetate
10 are more lipophilic than the other compounds in the
series and so are capable to better cross the cell mem-
brane. Once in the cytoplasm they are transformed into
parent monastrol by cellular esterases or chemical
hydrolysis. In other words, compounds 9 and 10 act as
prodrugs of monastrol. The fact that the phenol acetate
moiety is more sensitive to hydrolysis than N-acyl ureas
is consistent with the higher activity observed for mono-
acetate 9 when compared to diacetate 10, that one being
not fully hydrolyzed at the end of the 7 h incubation per-
iod with HeLa cells. To further support that hypothesis,
we preincubated compounds 9, 10, and N3-acetyl mon-
Figure 2. Induction of monoastral spindles by increasing concentra-
tions of monastrol and acetylated derivatives 9 and 10. HeLa cells were
exposed for 7 h to increasing concentrations of monastrol 1 (black
bars), compound 9 (gray bars), and compound 10 (open bars). (a) The
percentage of monoastral spindles was determined from the total
number of cells in M phase by staining fixed cells for b-tubulin (green)
and chromatin (red). (b) Representative spindles at 100 lM drug
concentration for compounds 1, 9, and 10. Control refers to absence of
drug.
astrol 21 (Fig. 1) in the presence and absence of medium
used in the cell-based assay (Dulbecco’s modified Ea-
gle’s medium supplemented with 10% fetal bovine ser-
um) and subsequently determined the inhibition of
basal Eg5 ATPase activity (Fig. 3). Compounds 9 and
10 do not inhibit Eg5 ATPase activity when preincu-
bated in the absence of medium, whereas low inhibition
is observed with compound 21. After incubation in cell-
based assay medium for 12 h, compounds 9, 10, and 21
do inhibit basal Eg5 activity with IC50 values of 3.9, 5.0,
and 3.8 lM, respectively. These results are in very good
agreement with the value obtained for monastrol
(6.1 lM) and are consistent with our hypothesis on a
preliminary enzymatic transformation of the com-
pounds into monastrol. Finally, ethyl acetate extraction
of cell-based assay medium incubated with diacylated
compound 10 (2 mM) for different periods of time and
thin layer chromatography analysis of the extracts
do confirm the quantitative transformation of the
compound into monastrol (by comparison with an
authentic sample). Half-life of diacetyl monastrol 10 in

E. Klein et al. / Bioorg. Med. Chem. 15 (2007) 6474–6488
6479
100
75
50
25
0
a
100
b
0
25
50
75
100
Inhibitor (µM)
75
50
25
0
Figure 3. Inhibition of basal Eg5 ATPase activity by acetylated
monastrol derivatives in the absence and presence of medium used in
cell-based assays. Compounds 8 (m), 9 (d), 10 (j), and 21 (Ç) in the
absence of medium, and the same compounds in the presence of
medium (n, s, h, and Æ, respectively).
these conditions was estimated to be less than 100 min.
Formation of the intermediate monoacetyl monastrol
control 0.5 µM 1 µM 5 µM 10 µM 20 µM 50 µM
9 was observed before final transformation into
monastrol.
c
Control
Compound 20
25 µM
More puzzling are the results obtained for compounds 7
and 8. These two compounds do not inhibit Eg5 ATPase
activities. However, in the cell-based assay microtubules
appeared to be affected by 50 lM of dihydroxy com-
pound 7, chromosomes being misaligned, and 8 proves
more potent than monastrol. The interpretation of that
is not clear as compound 8 is expected to be enzymati-
cally converted into 7 by cellular esterases (vide supra).
Moreover preincubation of 8 in the medium used in the
cell-based assay did not transform the compound into
an inhibitor of basal Eg5 ATPase activity (Fig. 3) what
is consistent with the results obtained with 7 in that as-
say. A possible explanation for these results would be a
chemical degradation of diacetyl compound 8 into an
active compound to be identified. That hypothesis could
be partly supported by the poor chemical stability of the
compound observed during its synthesis.
Nocodazole
332 nM
Compound 20
50 µM
Monastrol
100 µM
Compound 20
100 µM
2N 4N
2N 4N
Figure 4. Induction of monoastral spindles and mitotic arrest by
compound 20. (a) Representative spindles at 10 lM drug concentra-
tion for compound 20. Cells were fixed and stained for immuno-
fluorescence microscopy as in Figure 2. (b) Percentage of monoastral
spindles determined from the total number of mitotic cells after
exposure to compound 20. (c) Histograms obtained by flow cytometric
analysis of untreated cells and cells treated with nocodazole, monas-
trol, and three different concentrations of compound 20.
Last but not least, compound 20 appeared to be far
more potent than parent monastrol in inducing monoas-
tral spindles (Fig. 4a). Its IC50 value was determined to
be 9.2 lM versus 51.3 lM for monastrol (Table 1 and
Fig. 4b). We exposed HeLa cells for 24 h with com-
pound 20 and compared its ability to block cells in mito-
sis with that of nocodazole, a well-known mitotic
inhibitor targeting microtubules, as well as monastrol
(Fig. 4c). Cell populations were examined by two-
dimensional flow cytometry (cells were labeled with the
MPM2 monoclonal antibody recognizing phosphospec-
ific mitotic antigens, and propidium iodide for DNA
labeling). Compound 20 blocked cells with 4 N DNA
content at 25 lM. In addition, cells were arrested in
mitosis since 72% of the 4 N arrested cells were
MPM2 positive, whereas nocodazole treated cells were
82% positive. At 100 lM of compound 20, 77% of the
cells were mitotic whereas in the presence of 100 lM
of monastrol only 48% were mitotic. Thus compound
20 is far more potent in arresting cells in mitosis than
parent monastrol.
2.5. Synthesis of a monastrol affinity matrix and Eg5
purification procedure
Considering the results described above, an affinity ma-
trix was designed in order to improve our protocols for
purification of human full-length Eg5 from eukaryotic
cells (Scheme 5). Azidoamine 22 (prepared in 2 steps
from commercially available 1,2-bis(2-chloroethoxy)-
ethane) was condensed with carboxylic acid 11 to yield
monastrol derivative 23. A subsequent Staudinger reac-
tion afforded the reduced primary amino compound 24

6480
E. Klein et al. / Bioorg. Med. Chem. 15 (2007) 6474–6488
that was then reacted with NHS-activated Sepharose.
The affinity matrix thus immobilizes monastrol through
two amide bonds which guarantees hydrolytic stability
for safe and efficient recycling of the functionalized
support.
100
75
50
25
0
a
A: 40873.2500
8.11e4
A: 40874.07 2.61
The spacer arm between monastrol and the polymer re-
sin—three ethylene oxide units, approximately 1.1 nm
long—is expected to provide the immobilized dihydro-
pyrimidinethione with enough degree of freedom so as
to allow binding and immobilization of kinesin from
eluting cells homogenate.
40600
40800
41000
41200
41400
Mass (Da)
To test whether Eg5 can bind to and elute off the mon-
astrol affinity matrix, we first applied purified Eg52–386, a
monomeric recombinant Eg5 protein expressed in
E. coli,20 to the column. After extensive washing, we
eluted Eg52–386 off the column using excess amount of
(S)-trityl-L-cysteine (STLC), another potent Eg5 inhibi-
tor that shares the same binding pocket with monastrol
(data not shown).21 STLC was preferred to monastrol in
the elution step because of higher availability and lower
cost. As a second step, we purified recombinantly ex-
pressed Eg5 from an E. coli crude extract using the same
elution conditions. Mass spectrometry of the STLC re-
leased material revealed a single peak corresponding to
a peptide with molecular weight fitting to that predicted
for the expressed human Eg5 motor domain based on
amino acid sequence (Fig. 5a). A control experiment
with E. coli cells not expressing Eg5 and subsequent
analysis by Western blot indicated that no significant
material was retained on the column (Fig. 5b).
250
150
100
75
b
50
37
25
15
1
2
3
Figure 5. Purification of human Eg5 by chromatography on the
monastrol-based affinity matrix. (a) Analysis of the released peptides
following STLC elution by mass spectrometry shows a major peptide
with a molecular weight of 40,873 Da corresponding to the bacterially
expressed human Eg5 motor domain. (b) Western blot analysis of
E. coli cells in the absence and presence of the expression plasmid
coding for recombinant Eg51-386. Lane 1: protein standard; lane 2:
fraction eluted with STLC using E. coli cells (control); lane 3: fraction
Finally, we have extended the use of the monastrol affin-
ity matrix to the purification of the endogenous Eg5 pro-
tein from mammalian cells (HeLa) (Fig. 6). Following
high speed centrifugation, the supernatant obtained
from a lysate of mitotically arrested HeLa cells using
nocodazole was applied on the column. The latter was
washed extensively and elution of native Eg5 was per-
formed with STLC (see Section 4 for details). The eluted
fractions with Eg5 contained 15–35 lg of protein.
Immunoblot analysis using human Eg5 specific poly-
clonal antibody revealed that Eg5 was retained almost
quantitatively by the column and released by STLC
(Fig. 6b). An experiment performed under similar condi-
tions but using the control matrix (i.e., without immobi-
lized monastrol derivative) did not reveal any binding of
wild-type Eg5 (Fig. 6a). That gives confirmation of the
specificity of the interaction between Eg5 and the affinity
matrix which is strictly mediated by the immobilized
monastrol. The eluted samples were applied to a SDS–
PAGE for protein analysis. Several faint Coomassie
G-250 stained bands were detected and analyzed by pep-
tide mass fingerprinting and MALDI-TOF/TOF mass
spectrometry. Following that procedure, five different
proteins were identified (Table 2) of which human mito-
tic Eg5 was probably the full-length protein because the
21 peptides of the peptide mass fingerprint covered the
amino acid sequence range from 16 to 1033 (total
1057). Eg5 was identified by MS/MS of peptide 932SYL-
YPSTLVR941(Fig. 7). We also identified b-tubulin of the
ab-tubulin heterodimer, the building block of MTs, to
eluted with STLC using cells expressing recombinant Eg51–386
.
which Eg5 binds during mitosis to crosslink and slide
antiparallel MTs apart.8 It has previously been shown
that human Eg5 is in the ‘ATP-like’ conformation
(high-affinity for MTs), when either monastrol or an
analog is bound in the inhibitor-binding pocket, even
when MgADP is bound in the nucleotide-binding pock-
et.22,23 It is therefore likely that tubulin copurifies with
Eg5. Two additional cytoskeleton proteins identified
were actin, that—like tubulin—forms filaments and is
a major constituent of eukaryotic cells, and a-actinin,
that crosslinks and bundles actin filaments. Finally we
identified the heat shock cognate 71 kDa protein
(Hsc70), which is a constitutively expressed member of
the 70 kDa heat-shock protein family and functions as
a molecular chaperon. However, by using the control
matrix (i.e., without immobilized monastrol derivative)
we could show that actin and Hsc70 are simply abun-
dant housekeeping proteins that bind unspecifically to
the matrix.
3. Conclusion
A series of monastrol derivatives have been prepared by
introducing sterically demanding structural modifica-
tions on the skeleton of the parent drug. The ability of

E. Klein et al. / Bioorg. Med. Chem. 15 (2007) 6474–6488
6481
a
b
Wash Elution with NaCl
Wash
Elution with STLC
1
2
3
4
5
6
7
8
9
10 11 12
1
2
3
4
5
6
7 8 9 10 11 12 13
130
130
95
72
95
72
55
55
Figure 6. Western blot analysis with Eg5 specific polyclonal antibody of the different fractions obtained during the affinity chromatography
purification process (fraction size: 500 lL). The supernatant from HeLa cells arrested in mitotis with nocodazole was loaded onto the non-active
column (a) to eliminate proteins that bind non-specifically to the polymer matrix and the flow through was then loaded onto the monastrol-based
affinity matrix 25 (b). The columns were washed with lysis buffer, and proteins bound to the columns were eluted either with 2 M NaCl (a) or with
1 mM STLC (b). The eluted fractions from the monastrol affinity column were subsequently analyzed by mass spectrometry. (a) Lane 1: total lysate
from HeLa crude extract; lane 2: pellet after centrifugation; lane 3: supernatant after centrifugation; lane 4: flow-through; lanes 5–6: washes; lanes
7–12: fractions eluted with 2 M NaCl. (b) Lane 1: supernatant after centrifugation; lane 2: flow-through; lanes 3–7: washes; lanes 8–13: fractions
eluted with 1 mM STLC.
Table 2. Proteins identified from eluted fractions of the monastrol-based affinity column by mass spectrometry
Protein name
Accession No. MW (kDa) Mascot protein score SC (%) No. pep MS/MS sequences [m/z] (Mascot ions score)
Actin, cytoplasmic
a-Actinin-4
Heat shock cognate P11142
71 kDa protein
P60709
O43707
42
105
71
102
136
85
19
17
14
8
29AVFPSIVGRPR39 [1198.71 Da] (52)
194DGLAFNALIHR204 [1226.67 Da] (40)
160DAGTIAGLNVLR171 [1199.68 Da] (27)
37TTPSYVAFTDTER49 [1487.71 Da] (26)
93232SYLYPSTLVR941 [1198.69 Da] (50)
253LAVNMVPFPR262 [1143.65 Da] (21)
17
8
Kinesin Eg5
b-Tubulin
P52732
P07437
119
50
168
117
16
28
21
14
Protein name, Accession No., and theoretical molecular weight (MW) were derived from the Swiss-Prot entries. Sequence coverage (SC) and number
of peptides (No. pep) of the identified proteins are indicated. The Mascot protein score of the merged peptide mass fingerprints and MS/MS spectra,
and Mascot ion scores of individual fragmented peptides are shown.
cell-based assay, acting as prodrugs that are trans-
formed into monastrol inside the cellular compartment
before migrating to the nucleus. Eg5 tolerance for struc-
tural modifications at the C5 position on monastrol
backbone allowed optimization and furyl derivative 20
appears four to five times more active than monastrol.
The identification of structure–activity relationships re-
sulted in the design of an affinity matrix that has been
used for fast and efficient purification of Eg5 from crude
lysate of eukaryotic cells and mammalian cells. Mass
spectrometry analysis of the purified samples revealed
purification of Eg5 with b-tubulin. Two additional pro-
teins have been shown to bind non-specifically to the
column material.
8
6
4
2
b3
y1
b2
y6
y3
y2
b4
y7
y4
y8
0
200 400 600 800 1000 1200
Mass (m/z)
Figure 7. Identification of human mitotic Eg5 by MALDI-TOF/TOF.
The MALDI-TOF/TOF spectrum of the peptide with m/z 1198.60 Da
is presented. Kinesin Eg5 was unequivocally identified with the b-ions
b2–b4, and the y-ions y1–y4, and y6–y8 of the tryptic peptide
SYLYPSTLVR.
So the monastrol affinity matrix described in that work
can be used for the purification not only of the bacteri-
ally expressed Eg5 but also for the purification of the na-
tive human Eg5 protein from cellular extracts.
Furthermore the purification of the native molecule
should allow a number of investigations that are not rel-
evant with truncated Eg5 motor domain, that is, identi-
fication of possible Eg5 interacting proteins through the
cell cycle and studies on Eg5 post-translational modifi-
cations or oligomerization process. The problem of
monastrol specificity and monastrol-targets other than
these compounds to inhibit Eg5 activity in in vitro ATP-
ase assays and in cell-based assay was investigated.
Three compounds that were unable to inhibit Eg5 ATP-
ase activity in vitro proved potent Eg5 inhibitors in the

6482
E. Klein et al. / Bioorg. Med. Chem. 15 (2007) 6474–6488
Eg5 can be tentatively addressed as well with such an
affinity matrix.
J = 4.9, 5.1 Hz, 2H); 3.76–3.67 (m, 8H); 3.61 (t,
J = 4.9 Hz, 2H). 13C NMR (CDCl3, 50 MHz) d 179.8;
166.9; 133.4; 131.7; 129.0; 127.4; 72.6; 70.4; 70.2; 68.1;
61.7; 45.5. The N-alkyl-N0-benzoyl thiourea (0.57 g,
1.8 mmol) and sodium hydroxide (0.13 g, 3.2 mmol)
were stirred in water/acetonitrile (5 mL, 1:1) for 30 min
at room temperature. The reaction mixture was neutral-
ized by adding diluted HCl and evaporated under vac-
uum. The residue was suspended in CH2Cl2, washed
with water, dried over MgSO4, and evaporated to yield
1-{2-[2-(2-hydroxy-ethoxy)-ethoxy]-ethyl-thiourea
4. Experimental
4.1. Chemistry
All reactions were carried out with reagent grade sol-
vents. Commercially available reagents were used with-
out further purification. NHS-activated Sepharose 4
FastFlow was purchased from Amersham Biosciences.
Chemicals for Eg5 ATPase assays were from sources
indicated by Hackney and Jiang.24 1H NMR and 13C
NMR spectra were recorded on Bruker 200 and
300 MHz instruments, and chemical shifts are reported
in ppm downfield from TMS. IR spectra were recorded
1
(0.31 g, 82%) as a hygroscopic glassy solid. H NMR
(CD3OD/D2O 1:1, 200 MHz) d 3.67–3.57 (m, 12 H).
13C NMR (CD3OD/D2O 1:1, 50 MHz) d 159.6; 72.8;
70.7; 70.6; 69.8; 61.5; 40.9. The latter compound
(0.23 g, 1.1 mmol) was refluxed in ethanol (10 mL) with
ethyl acetoacetate (0.2 mL, 1.6 mmol), 3-hydroxybenzal-
dehyde (0.13 g, 1.1 mmol), and concentrated hydrochlo-
ric acid (0.4 mL) for 24 h. Ethyl acetate was added and
the resulting mixture was washed with brine, dried over
MgSO4, and reduced under vacuum. The residue was
purified by silica gel chromatography (EtOAc/EtOH
92:8) to yield compound 3 (33 mg, 7%) as a glassy solid.
1H NMR (CDCl3, 300 MHz) d 7.08 (t, J = 7.8 Hz, 1H);
6.76–6.67 (m, 3H); 5.27 (s, 1H); 4.16 (q, J = 7.2 Hz, 2H);
3.74–3.47 (m, 12H); 2.47 (s, 3H); 1.25 (t, J = 7.2 Hz,
3H). 13C NMR (CDCl3, 75 MHz) d 167.2; 157.5;
156.4; 143.6; 143.5; 129.6; 118.7; 114.8; 113.5; 113.4;
72.6; 70.3; 70.1; 69.6; 61.4; 60.2; 42.2; 29.7; 24.0; 14.3.
MS (m/z) 425.24 [M+H]+.
on
a
Perkin-Elmer-1600-FTIR spectrometer and
absorption values m are in cmꢀ1. Mass Spectra (MS)
were recorded on a Finnigan-4600 quadrupole instru-
ment at chemical ionization. Mass data are reported in
mass units (m/z).
4.1.1. Synthesis of 4-(3-hydroxy-phenyl)-6-methyl-2-thi-
oxo-1,2,3,4-tetrahydro-pyrimidine-5-carboxylic acid 2-{2-
[2-(2-hydroxy-ethoxy)-ethoxy]-ethoxy}-ethyl ester (2). 3-
Oxo-butyric acid 2-{2-[2-(2-hydroxy-ethoxy)-ethoxy]-
ethoxy}-ethyl ester (prepared by trans-esterification of
methyl acetoacetate with tetraethylene glycol over K10
montmorillonite25) (1.20 g, 4.3 mmol), thiourea (0.22 g,
2.9 mmol), and 3-hydroxybenzaldehyde (0.35 g,
2.9 mmol) were refluxed in i-PrOH (10 mL) with concen-
trated hydrochloric acid (0.5 mL) for 16 h. Isopropyl
alcohol was removed in vacuo, methylene chloride
(20 mL) was added, and the solution was washed twice
with HCl 5% (10 mL), dried over MgSO4, and reduced
under vacuum. The crude residue was purified by silica
gel chromatography (AcOEt/EtOH 100:0 to 90:10) to
yield compound 2 (0.79 g, 62%) as a hygroscopic glassy
4.1.3. Synthesis of 4-(3,4-dihydroxy-phenyl)-6-methyl-2-
thioxo-1,2,3,4-tetrahydro-pyrimidine-5-carboxylic acid
ethyl ester (4). Compound 4 (63 mg, 39%) was prepared
from 3,4-dihydroxybenzaldehyde following the same
procedure as described for 2. 1H NMR (CD3OD,
300 MHz)
d 6.73 (d, J = 2.0 Hz, 1H); 6.70 (d,
J = 8.7 Hz, 1H); 6.61 (dd, J= 2.0, 8.7 Hz, 1H); 5.17 (s,
1H); 4.08 (q, J = 7.0 Hz, 1H); 2.33 (s, 3H); 1.18 (t,
J = 7.0 Hz, 3H). 13C NMR (CD3OD, 50 MHz) d
175.6; 167.5; 146.4; 146.2; 145.1; 136.5; 119.4; 116.2;
114.9; 103.6; 61.2; 56.2; 17.6; 14.5. IR (neat) m 3312;
3182; 2991; 2923; 1687. MS (m/z) 308.98 [M+H]+.
1
solid. H NMR (CDCl3, 300 MHz) d 8.53 (s, 1H); 8.42
(s, 1H); 7.18 (t, J = 7.3 Hz, 1H); 6.91 (m, 1H); 6.80 (d,
J = 7.9 Hz, 1H); 6.72 (dd, J = 1.5, 7.9 Hz, 1H); 5.26 (d,
J = 2.6 Hz, 1H); 4.38–4.31 (m, 1H); 4.04–3.97 (m, 1H);
3.78–3.50 (m, 14H); 2.33 (s, 3H). 13C NMR (CDCl3,
50 MHz) d 173.5; 164.9; 156.0; 144.3; 143.8; 130.1;
118.1; 115.0; 114.9; 102.1; 72.3; 70.6; 70.5; 69.9; 69.8;
69.2; 62.7; 61.4; 55.9; 17.8. IR (neat) m 3253 (b); 2949;
2440 (b). MS (m/z) 441.26 [M+H]+.
4.1.4. Synthesis of 4-(3-hydroxy-4-{2-[2-(2-hydroxy-eth-
oxy)-ethoxy]-ethoxy}-phenyl)-6-methyl-2-thioxo-1,2,3,4-
tetrahydro-pyrimidine-5-carboxylic acid ethyl ester (5). 2-
[2-(2-Chloro-ethoxy)-ethoxy]-ethanol (2.9 mL, 20.0 mmol),
3,4-dihydroxybenzaldehyde (2.76 g, 20.0 mmol), potas-
sium carbonate (2.76 g, 20.0 mmol), and sodium iodide
(0.05 g, 0.3 mmol) were stirred in anhydrous DMF
(20 mL) at 80 ꢁC for 18 h. Methylene chloride (75 mL)
was added and the solution was washed twice with
NaOH 3 N (50 mL). The aqueous phase was acidified
with HCl 12 N, reduced in vacuo, and the residue was
triturated twice with EtOAc/MeOH (20 mL, 1:1). The
organic phase was reduced under vacuum and the resi-
due was purified by silica gel chromatography
(CH2Cl2/EtOAc/EtOH 50:50:5) to yield the expected 3-
hydroxy-4-{2-[2-(2-hydroxy-ethoxy)-ethoxy]-ethoxy}-
4.1.2. Synthesis of 3-{2-[2-(2-hydroxy-ethoxy)-ethoxy]-
ethyl}-4-(3-hydroxy-phenyl)-6-methyl-2-thioxo-1,2,3,4-
tetrahydro-pyrimidine-5-carboxylic acid ethyl ester (3). 2-
[2-(2-Amino-ethoxy)-ethoxy]ethanol26 (1.05 g, 7.0 mmol)
in anhydrous acetonitrile (5 mL) was added dropwise
to benzoyl isothiocyanate (0.95 mL, 7.0 mmol) in anhy-
drous acetonitrile (5 mL) at room temperature. The
reaction mixture was stirred for 30 min before the sol-
vent was removed in vacuo. The crude residue was puri-
fied by silica gel chromatography to yield 1-benzoyl-3-
{2-[2-(2-hydroxy-ethoxy)-ethoxy]-ethyl-thiourea (2.2 g,
1
1
48%) as a hygroscopic glassy solid. H NMR (CDCl3,
200 MHz) d 10.94 (br s, 1H); 9.15 (br s, 1H); 7.84 (dd,
J = 1.7, 6.8 Hz, 2H); 7.60–7.44 (m, 3H); 3.92 (td,
benzaldehyde (2.78 g, 52%) as a glassy solid. H NMR
(CDCl3, 200 MHz) d 9.80 (s, 1 H); 7.39 (d, J = 2.0 Hz,
1H); 7.34 (dd, J = 2.0, 8.3 Hz, 1H); 6.91 (d, J = 8.3 Hz,

E. Klein et al. / Bioorg. Med. Chem. 15 (2007) 6474–6488
6483
1H); 4.21–4.16 (m, 1H); 3.90–3.85 (m, 1H); 3.75–3.66
(m, 10H). 13C NMR (CDCl3, 50 MHz) d 191.1; 151.6;
147.2; 131.1; 123.5; 115.9; 112.1; 72.3; 70.6; 70.0; 69.0;
67.9. IR (neat) m 3498 (b); 2890 (b). MS (m/z) 271.04
[M+H]+. The latter compound (0.49 g, 1.8 mmol), ethyl
acetoacetate (0.35 mL, 2.7 mmol), thiourea (0.15 g,
2.0 mmol), and HCl 12 N (0.4 mL) were refluxed for
4 d in EtOH (10 mL). Alcohol was removed under re-
duced pressure, methylene chloride (50 mL) was added,
and the solution was washed twice with HCl 2 N. The
organic layer was dried over MgSO4, reduced under vac-
uum, and the residue was purified by silica gel chroma-
tography (CH2Cl2/EtOH 100:0 to 95:5). Compound 5
NaHCO3, water, and brine, and dried over MgSO4.
The solvent was removed in vacuo to yield pure 2,5-
diacetoxy-benzaldehyde (580 mg, 100%). 1H NMR
(CDCl3, 300 MHz) d 10.06 (s, 1H); 7.60 (d, J = 3.0 Hz,
1H); 7.35 (dd, J = 3.0, 8.7 Hz, 1H); 7.19 (d, J = 8.7 Hz,
1H); 2.37 (s, 3H); 2.30 (s, 3H). 13C NMR (CDCl3,
75 MHz) d 187.5; 169.0; 168.8; 148.9; 148.3; 128.6;
128.4; 124.5; 123.2; 20.9; 20.7. IR (neat) m 1759; 1692.
A mixture of the latter compound (448 mg, 2.0 mmol),
thiourea (153 mg, 2.0 mmol), ethyl acetoacetate
(230 lL, 3.0 mmol), and ytterbium(III) trifluorome-
thanesulfonate hydrate (63 mg, 0.1 mmol) in anhydrous
acetonitrile (10 mL) was stirred at 55–60 ꢁC for 2 h. The
solvent was removed under vacuum and the crude resi-
due was purified by silica gel chromatography (Et2O/
CH2Cl2 40:60) to yield 8 (132 mg, 17%) as a slightly yel-
1
(0.73 g, 91%) was obtained as a glassy solid. H NMR
(CDCl3, 300 MHz) d 6.85 (d, J = 2.2 Hz, 1H); 6.81 (d,
J = 5.0 Hz, 1H); 6.70 (dd, J= 2.2, 8.1 Hz, 1H); 5.27 (d,
J = 3.4 Hz, 1H); 4.14–4.05 (m, 4H); 3.86–3.62 (m,
12H); 2.34 (s, 3H); 1.17 (t, J = 7.2 Hz, 3H). 13C NMR
(CD3OD, 75 MHz) d 175.8; 167.3; 148.1; 147.7; 145.4;
138.4; 119.1; 115.2; 114.6; 103.3; 73.6; 71.5; 71.3; 70.7;
69.5; 62.1; 61.2; 56.1; 17.7; 14.5. IR (neat) m 3194 (b);
2930 (b). MS (m/z) 441.25 [M+H]+.
1
low solid. H NMR (CD3OD/CDCl3 1:1, 200 MHz) d
7.09 (d, J= 8.8 Hz, 1H); 6.99 (dd, J = 2.6, 8.8 Hz, 1H);
6.69 (d, J = 2.6 Hz, 1H); 5.42 (s, 1H); 4.08 (qd, J = 2.0,
7.0 Hz, 2H); 2.48 (s, 3H); 2.35 (s, 3H); 2.22 (s, 3H);
1.17 (t, J= 7.0 Hz, 3H). 13C NMR (CD3OD/CDCl3
1:1, 75 MHz) d 170.2; 167.6; 162.8; 160.6; 157.9; 148.8;
145.3; 134.9; 124.4; 122.2; 120.9: 101.4; 61.1; 36.7;
23.8; 21.2; 21.1; 14.4. IR (neat) m 1762; 1207; 1170. MS
(m/z) 393.27 [M+H]+.
4.1.5. Synthesis of 4-(3-hydroxy-6-nitro-phenyl)-6-
methyl-2-thioxo-1,2,3,4-tetrahydro-pyrimidine-5-carbox-
ylic acid ethyl ester (6). A mixture of 3-hydroxy-6-nitro-
benzaldehyde (127 mg, 0.76 mmol), thiourea (116 mg,
1.52 mmol), ethyl acetoacetate (97 lL, 0.76 mmol), and
4.1.8. Synthesis of 4-(3-acetoxy-phenyl)-6-methyl-2-thi-
oxo-1,2,3,4-tetrahydro-pyrimidine-5-carboxylic acid ethyl
ester (9). Acetic anhydride (0.12 mL, 1.3 mmol) was
added to a mixture of monastrol 1 (0.32 g, 1.1 mmol),
triethylamine (0.20 mL, 1.4 mmol), and 4-DMAP
(6 mg, 0.05 mmol) in anhydrous THF (5 mL). The
resulting solution was stirred at room temperature over-
night. THF was removed under vacuum, ethyl acetate
(10 mL) was added, and the solution was washed with
diluted hydrochloric acid. The organic layer was dried
over MgSO4, reduced in vacuo, and the residue was
purified by silica gel chromatography (Et2O/hexane
50:50 to 70:30) to yield 9 (0.31 g, 84%) as a slightly yel-
low powder. 1H NMR (CDCl3, 200 MHz) d 8.82 (s, 1H);
8.29 (s, 1H); 7.27 (t, J = 8.2 Hz, 1H); 7.12 (d, J = 7.8 Hz,
1H); 7.00–6.96 (m, 2H); 5.33 (d, J = 2.9Hz, 1H); 4.06 (q,
J = 7.1 Hz, 2H); 2.31 (s, 3H); 2.24 (s, 3H); 1.13 (t, J=
7.1 Hz, 3H). 13C NMR (CDCl3, 75 MHz) d 173.8;
169.2; 165.0; 150.6; 144.0; 143.5; 129.6; 124.0; 121.2;
119.9; 102.2; 60.3; 55.2; 21.0; 17.9; 13.9. IR (neat) m
3190 (b); 2986. MS (m/z) 335.26 [M+H]+. A fraction
eluting before compound 9 was identified as diacetylated
compound 10 (0.05 g, 12%). 1H NMR (CDCl3,
200 MHz) d 8.61 (s, 1H); 7.29 (t, J = 7.1 Hz, 1H); 7.14
(d, J = 5.9 Hz, 1H); 7.03–7.00 (m, 2H); 6.66 (s, 1H);
4.23 (q, J = 7.1 Hz, 2H); 2.78 (s, 3H); 2.36 (s, 3H);
2.27 (s, 3H); 1.28 (t, J = 7.1 Hz, 3H). 13C NMR (CDCl3,
50 MHz) d 178.0; 173.2; 169.2; 164.9; 150.7; 143.6;
140.4; 129.4; 123.7; 121.3; 119.7; 107.9; 60.9; 53.4;
27.6; 21.1; 17.3; 14.1. IR (neat) m 3270 (b); 2987; 1699.
MS (m/z) 377.20 [M+H]+.
ytterbium(III)
trifluoromethanesulfonate
hydrate
(48 mg, 0.08 mmol) was refluxed in acetonitrile (5 mL)
for 4 h before the solvent was removed under vacuum.
The crude residue was purified by silica gel chromatog-
raphy (Et2O/hexane 50:50 to 70:30) to yield 6 (166 mg,
64%) as
300 MHz)
a
d
yellow powder. 1H NMR (CD3OD,
7.94 (d, J= 9.0 Hz, 1H); 6.87 (d,
J = 2.7 Hz, 1H); 6.82 (dd, J = 2.7, 9.0 Hz, 1H); 6.04
(br s, 1H); 3.96 (q, J = 7.0 Hz, 2H); 2.43 (s, 3H); 1.00
(t, J= 7.0 Hz, 3H). 13C NMR (CDCl3, 50 MHz) d
176.3; 166.5; 164.5; 147.3; 141.5; 141.3; 128.8; 116.5;
116.2; 101.7; 68.8; 61.3; 17.5; 14.2. IR (neat) m 3195
(b); 2981. MS (m/z) 338.20 [M+H]+.
4.1.6. Synthesis of 4-(2,5-dihydroxy-phenyl)-6-methyl-2-
thioxo-1,2,3,4-tetrahydro-pyrimidine-5-carboxylic acid
ethyl ester (7). Compound 8 (76 mg, 0.2 mmol) was stir-
red for 2 h in methanolic ammonia (2 mL) at 0 ꢁC. The
solvent was removed under vacuum and the residue was
purified by silica gel chromatography (AcOEt) to yield 7
(9 mg, 15%) as a slightly yellow solid. 1H NMR
(CD3OD, 300 MHz) d 6.59 (d, J = 8.4 Hz, 1H); 6.47
(dd, J = 3.0, 8.4 Hz, 1H); 6.32 (d, J = 3.0 Hz, 1H);
5.55 (s, 1H); 4.15–4.04 (m; 2H); 2.47 (s, 3H); 1.19 (t,
J = 7.2 Hz, 3H). IR (neat) m 3324 (b); 3201.
4.1.7. Synthesis of 4-(2,5-diacetoxy-phenyl)-6-methyl-2-
thioxo-1,2,3,4-tetrahydro-pyrimidine-5-carboxylic acid
ethyl ester (8). Acetic anhydride (1.2 mL, 12.7 mmol)
was added dropwise to a suspension of 2,5-dihydroxy-
benzaldehyde (361 mg, 2.6 mmol) and potassium car-
bonate (722 mg, 5.2 mmol) in anhydrous diethyl ether
at 0 ꢁC. The mixture was stirred for 2 h at room temper-
ature, filtered, and the solid was washed with ethyl ace-
tate. The combined filtrate was washed with aqueous
4.1.9. Synthesis of 4-(3-hydroxy-phenyl)-6-methyl-2-thi-
oxo-1,2,3,4-tetrahydro-pyrimidine-5-carboxylic acid (11).
Monastrol 1 (0.18 g, 0.6 mmol) and potassium hydrox-
ide (0.10 g, 1.8 mmol) were stirred in water (2 mL) for
3 d. The solution was acidified with HCl 2 N and

6484
E. Klein et al. / Bioorg. Med. Chem. 15 (2007) 6474–6488
extracted with EtOAc. The organic layer was dried over
MgSO4, reduced under vacuum, and the residue was
purified over silica gel (EtOAc/CH2Cl2 50:50 to 100:0)
to yield compound 11 (0.13 g, 79%) as a white powder.
1H NMR (CD3OD, 300 MHz) d 7.12 (t, J = 7.5 Hz,
1H); 6.79–6.69 (m, 2H); 6.67 (dd, J = 5.6, 7.1 Hz, 1H);
5.27 (s, 1H); 2.34 (s, 3H). 13C NMR (CD3OD,
75 MHz) d 176.2; 169.4; 158.7; 146.0; 145.3; 130.6;
118.9; 115.8; 114.4; 103.7; 56.3; 17.7. IR (neat) m 3226
(b). MS (m/z) 265.10 [M+H]+.
4.1.12. Synthesis of 4-(3-hydroxy-phenyl)-6-methyl-2-thi-
oxo-1,2,3,4-tetrahydro-pyrimidine-5-carboxylic acid 2-(2-
mercapto-ethoxy)-ethyl ester (14). Compound 13 (21 mg,
0.05 mmol) was stirred at room temperature for 24 h in
i-PrOH (2 mL) with 3 drops of concentrated hydrochlo-
ric acid. The solvent was removed in vacuo and the res-
idue was filtered over silica gel (Et2O) to yield 14 (13 mg,
1
69%) as a glassy solid. H NMR (CD3OD, 200 MHz) d
7.13 (t, J = 7.7 Hz, 1H); 6.81–6.66 (m, 3H); 5.28 (s, 1H);
4.27–4.15 (m, 2H); 3.60 (t, J = 4.8 Hz, 2H); 3.49 (t,
J = 6.0 Hz, 2H); 2.58 (t, J = 6.5 Hz, 2H); 2.36 (s, 3H).
13C NMR (CD3OD, 50 MHz) d 176.3; 167.2; 158.8;
146.2; 146.1; 130.7; 118.9; 115.9; 114.6; 102.9; 74.0;
69.8; 64.4; 56.3; 24.7; 17.7. IR (neat) m 3242 (b); 1185.
MS (m/z) 335.14 [MꢀHS]+.
4.1.10. Synthesis of 4-(3-hydroxy-phenyl)-6-methyl-2-thi-
oxo-1,2,3,4-tetrahydro-pyrimidine-5-carboxylic acid 2-(2-
hydroxy-ethoxy)-ethyl ester (12). To a mixture of car-
boxylic acid 11 (57 mg, 0.22 mmol), diethylene glycol
(0.21 mL, 2.2 mmol), and triphenyl phosphine (117 mg,
0.44 mmol) in anhydrous THF (3 mL) was added DIAD
(64 lL, 0.33 mmol) at room temperature. The solution
was stirred for 2 h before the solvent was removed under
reduced pressure. The residue was purified by silica gel
chromatography (EtOAc/CH2Cl2 50:50 to 100:0) to
yield compound 12 (39 mg, 51%) as a white powder.
1H NMR (CD3OD, 300 MHz) d 7.13 (t, J= 7.9 Hz,
1H); 6.79 (d, J = 8.5 Hz, 1H); 6.77 (d, J = 2.6 Hz, 1H);
6.68 (dd, J = 2.6, 8.5 Hz, 1H); 5.28 (s, 1H); 4.26–4.12
(m, 2H); 3.64–3.61 (m, 4H); 3.52–3.45 (m, 2H); 2.34 (s,
3H). 13C NMR (CD3OD, 50 MHz) d 176.1; 167.2;
158.7; 146.1; 146.0; 130.7; 118.9; 115.8; 114.6; 102.9;
73.5; 70.0; 64.5; 62.2; 56.2; 17.8. IR (neat) m 3256 (b);
2952; 2447 (b). MS (m/z) 353.19 [M+H]+.
4.1.13. Synthesis of 4-(3-hydroxy-phenyl)-6-methyl-2-thi-
oxo-1,2,3,4-tetrahydro-pyrimidine-5-carboxylic acid eth-
ylamide (15). DCC (55 mg, 0.26 mmol) was added to a
mixture of acid 11 (35 mg, 0.13 mmol), ethylamine
hydrochloride (26 mg, 0.32 mmol), and triethylamine
(45 lL, 0.32 mmol) in anhydrous THF (2 mL). The
solution was stirred at room temperature overnight.
The precipitate was removed by filtration, the filtrate
was reduced in vacuo, and the residue was purified by
silica gel chromatography (EtOAc/CH2Cl2 70:30 to
100:0) to yield amide 15 (38 mg, 96%) as a white solid.
1H NMR (CDCl3/CD3OD 1:1, 200 MHz) d 7.13 (t,
J = 8.8 Hz, 1H); 6.75–6.69 (m, 3H); 5.20 (s, 1H);
3.17–3.06 (m, 2H); 2.02 (s, 3H); 0.95 (t, J= 7.3 Hz,
3H). 13C NMR (CDCl3/CD3OD 1:1, 50 MHz) d
174.4; 167.7; 157.9; 143.7; 133.8; 130.4; 118.2; 115.9;
113.8; 108.2; 57.0; 34.8; 16.6; 14.4. MS (m/z) 292.15
[M+H]+.
4.1.11. Synthesis of 4-(3-hydroxy-phenyl)-6-methyl-2-thi-
oxo-1,2,3,4-tetrahydro-pyrimidine-5-carboxylic acid 2-(2-
acetylsulfanyl-ethoxy)-ethyl ester (13). Potassium thioac-
etate (0.39 g, 3.4 mmol) and 2-[2-hydroxy-ethoxy)-ethyl
methanesulfonate]26 (0.33 g, 1.8 mmol) in anhydrous
DMF (5 mL) were stirred at 80 ꢁC for 4 h. Diethyl ether
(25 mL) was added and the resulting solution was
washed with water. The organic layer was dried over
MgSO4, reduced under vacuum, and the residue was
purified by silica gel chromatography (Et2O/hexane
75:25) to yield expected S-2-(2-hydroxy-ethoxy)-ethyl
ethanethioate (0.26 g, 89%). 1H NMR (CDCl3,
4.1.14. Synthesis of 1,5-di-[4-(3-hydroxy-phenyl)-6-
methyl-2-thioxo-1,2,3,4-tetrahydro-pyrimidine-5-carboxy]-
3-oxapentane (16). DIAD (60 lL, 0.30 mmol) was
added dropwise at room temperature to a mixture of
diethylene glycol (10 lL, 0.10 mmol), acid 11 (49 mg,
0.18 mmol), and triphenyl phosphine (112 mg,
0.43 mmol) in anhydrous THF (2 mL). The solution
was stirred for 1 h, reduced in vacuo, and the residue
was purified by silica gel chromatography (EtOAc) to
300 MHz)
d 3.66 (t, J = 5.0 Hz, 2H); 3.58 (t,
1
J = 6.3 Hz, 2H); 3.53 (t, J = 5.0 Hz, 2H); 3.06 (t,
J = 6.3 Hz, 2H); 2.31 (s, 3H). 13C NMR (CDCl3,
75 MHz) d 195.4; 71.9; 69.5; 61.5; 30.4; 28.7. MS (m/z)
164.99 [M+H]+. DIAD (77 lL, 0.37 mmol) was added
dropwise at room temperature to a mixture of the latter
compound (38 mg, 0.23 mmol), acid 11 (50 mg,
0.19 mmol), and triphenyl phosphine (106 mg,
0.40 mmol) in anhydrous THF (2 mL). The solution
was stirred for 1 h, reduced in vacuo, and the residue
was purified by silica gel chromatography (Et2O/hexane
70:30) to yield compound 13 (32 mg, 41%) as a white so-
yield compound 16 (49 mg, 87%) as a white solid. H
NMR (CD3OD, 200 MHz) d 7.10 (t, J = 7.8 Hz, 2H);
6.79–6.64 (m, 6H); 5.26 (s, 2H); 4.17–4.07 (m, 4H);
3.51 (t, J = 4.5 Hz, 4H); 2.33 (s, 6H). 13C NMR
(CD3OD, 50 MHz) d 176.1; 167.2; 158.8; 146.1; 146.0;
130.7; 119.0; 115.9; 114.6; 102.9; 70.0; 64.6; 56.3; 17.8.
IR (neat) m 3239 (b); 2443 (b). MS (m/z) 599.45
[M+H]+.
4.1.15. Synthesis of 4-(3-hydroxy-phenyl)-6-methyl-2-sel-
enoxo-1,2,3,4-tetrahydro-pyrimidine-5-carboxylic
acid
1
lid. H NMR (CDCl3, 200 MHz) d 8.40 (s, 1H); 8.07 (s,
ethyl ester (17). A mixture of finely ground selenourea
(246 mg, 2.00 mmol), 3-hydroxybenzaldehyde (105 mg,
0.85 mmol), and magnesium chloride hexahydrate
(75 mg, 0.37 mmol) was stirred with ethyl acetoacetate
(120 lL, 0.94 mmol) at 80 ꢁC for 3 h. The reaction mix-
ture was purified by silica gel chromatography (Et2O/
hexane 67:33) to yield 17 (221 mg, 76%) as a white so-
1H); 7.13 (t, J = 8.1 Hz, 1H); 6.82–6.72 (m, 3H); 5.32 (t,
J = 2.6 Hz, 1H); 4.22–4.15 (m, 2H); 3.64–3.60 (m, 2H);
3.53 (t, J = 6.5 Hz, 2H); 3.05 (t, J = 6.5 Hz, 2H); 2.34
(s, 3H); 2.32 (s, 3H). 13C NMR (CDCl3, 50 MHz) d
196.6; 174.0; 165.3; 156.4; 143.9; 143.8; 130.1; 118.6;
115.5; 113.8; 102.3; 69.5; 68.7; 63.3; 55.6; 30.6; 28.7;
18.3. MS (m/z) 411.27 [M+H]+.
1
lid. H NMR (CDCl3, 200 MHz) d 8.84 (sb, 2H); 7.09

E. Klein et al. / Bioorg. Med. Chem. 15 (2007) 6474–6488
6485
(t, J = 7.8 Hz, 1H); 6.85–6.70 (m, 3H); 5.32 (s, 1H); 4.07
(q, J = 6.6 Hz, 2H); 2.29 (s, 3H); 1.14 (t, J = 6.6 Hz,
3H). 13C NMR (CDCl3, 50 MHz) d 169.8 (t, J=
189.6 Hz); 165.8; 156.2; 143.2; 142.5; 130.0; 118.8;
115.7; 113.9; 103.2; 60.8; 55.7; 56.3; 18.1; 14.0. IR
(neat) m 3287; 3159; 3106; 2981; 1663. MS (m/z)
340.98 [M+H]+.
3.6 Hz, 1H); 3.94 (s, 1H); 2.28 (s, 3H). 13C NMR
(CDCl3, 75 MHz) enol form d 189.4; 176.0; 145.9;
145.8; 115.5; 112.4; 96.0; 24.4; diketone form d 201.3;
82.1; 150.4; 147.1; 118.5; 112.7; 54.2; 30.5. IR (neat) m
3130; 1586. MS (m/z) 153.07 [M+H]+. The latter com-
pound (643 mg, 4.2 mmol) was stirred with thiourea
(355 mg, 4.7 mmol), 3-hydroxybenzaldehyde (516 mg,
4.2 mmol), and ytterbium(III) trifluoromethanesulfo-
nate hydrate (288 mg, 0.5 mmol) in refluxing anhydrous
THF (10 mL) for 10 h. The solvent was removed under
vacuum and the residue was purified by silica gel chro-
matography (Et2O/hexane 50:50 to 100:0) to yield 20
4.1.16. Synthesis of 4-(3-hydroxy-phenyl)-6-methyl-2-
(ethylthio)-1,4-dihydro-pyrimidine-5-carboxylic acid ethyl
ester (18). Sodium hydrogenocarbonate (328 mg,
3.91 mmol)
and
2-ethylisothiourea27
(105 mg,
1
1.01 mmol) were added to ethyl 2-(3-hydroxybenzylid-
ene)-3-oxobutanoate28 (182 mg, 0.78 mmol) in anhy-
drous DMF (2 mL). The resulting mixture was stirred
for 3 h at 80 ꢁC, and volatile was removed under re-
duced pressure. The crude residue was purified by silica
gel chromatography (Et2O/hexane 70:30) to yield 18
(103 mg, 41%) as a white solid. 1H NMR (CD3OD,
(1.22 g, 93%) as a yellow powder. H NMR (DMSO-
d6, 200 MHz) d 10.27 (br s, 1H); 9.57 (d, J = 3.2 Hz,
1H); 9.43 (s, 1H); 7.92 (br s, 1H); 7.14 (d, J = 3.6 Hz,
1H); 7.06 (d, J = 7.8 Hz, 1H); 6.67–6.56 (m, 4H); 5.33
(d, J = 3.2 Hz, 1H); 1.90 (s, 3H). 13C NMR (CDCl3,
75 MHz) d 180.6; 174.0; 157.4; 152.6; 146.9; 144.2;
140.4; 129.5; 118.1; 116.8; 114.7; 113.2; 112.5; 109.5;
55.0; 17.1. IR (neat) m 3181 (b); 1614; 1590; 1563;
1212. MS (m/z) 315.15 [M+H]+.
300 MHz)
d 7.09 (t, J = 7.8 Hz, 1H); 6.77 (d,
J = 8.1 Hz, 1H); 6.74 (d, J = 1.8 Hz, 1H); 6.64 (dd,
J = 2.1, 8.1 Hz, 1H); 5.46 (s, 1H); 4.09 (q, J = 6.9 Hz,
2H); 3.10–3.04 (m, 1H); 2.86–2.79 (m, 1H); 2.30 (s,
3H); 1.21 (t, J = 6.9 Hz, 3H); 1.20 (t, J = 7.2 Hz, 3H).
13C NMR (CD3OD/CDCl3 1:1, 50 MHz) d 168.3;
157.7; 153.2; 146.9; 140.0; 130.0; 118.9; 114.9; 114.4;
101.5; 60.7; 58.7; 25.7; 19.1; 14.8; 14.6. IR (neat) m
3253; 2974; 1692. MS (m/z) 321.36 [M+H]+.
4.1.19. Synthesis of 2-acetyl-4-(3-hydroxy-phenyl)-6-
methyl-2-thioxo-1,2,3,4-tetrahydro-pyrimidine-5-carbox-
ylic acid ethyl ester (21). Diacetylated compound 10
(52 mg, 0.14 mmol) in methyl alcohol/water (5 mL,
4:1) was stirred for 1 h with aqueous NaHCO3 saturated
solution (1 mL) at room temperature. The pH of the
solution was brought to 7 with diluted HCl before sol-
vent was removed under vacuum. The residue was sus-
pended in ethyl acetate and washed with water, dried
over MgSO4, reduced in vacuo, and purified by silica
gel chromatography (Et2O/hexane 50:50) to yield 21
(36 mg, 78%) as a slightly yellow powder. 1H NMR
(CDCl3, 300 MHz) d 8.62 (s, 1H); 7.15 (t, J = 7.8 Hz,
1H); 6.84 (d, J = 7.8 Hz, 1H); 6.79 (s, 1H); 6.74 (d,
J = 7.8 Hz, 1H); 6.64 (s, 1H); 4.22 (q, J = 7.0 Hz, 2H);
2.78 (s, 3H); 2.36 (s, 3H); 1.28 (t, J= 7.0 Hz, 3H). 13C
NMR (CDCl3, 50 MHz) d 178.3; 173.6; 165.3; 156.1;
143.5; 140.4; 130.1; 118.9; 115.3; 113.6; 108.4; 61.3;
53.8; 27.9; 17.6; 14.3. IR (neat) m 3270 (b); 2982. MS
(m/z) 335.36 [M+H]+.
4.1.17. Synthesis of 4-(3-hydroxy-phenyl)-6-methyl-2-
(benzylthio)-1,4-dihydro-pyrimidine-5-carboxylic
acid
ethyl ester (19). Compound 19 (188 mg, 69%) was pre-
pared from 2-benzylisothiourea29 following the same
procedure as for 18. 1H NMR (CDCl3, 300 MHz) d
7.36–7.22 (m, 5H); 7.20 (t, J = 7.8 Hz, 1H); 6.89 (dd,
J = 0.6, 7.2 Hz, 1H); 6.78 (m, 1H); 6.77 (d,
J = 7.8 Hz, 1H); 5.65 (sb, 1H); 4.43 (d, J = 13.8 Hz,
2H); 4.14 (q, J = 6.6 Hz, 2H); 2.36 (s, 3H); 1.22 (t,
J = 7.2 Hz, 3H). 13C NMR (CDCl3, 50 MHz) d 167.1;
156.7; 154.0; 145.9; 137.4; 133.3; 130.0; 129.4; 129.0;
127.7; 119.1; 115.2; 114.4; 102.3; 60.5; 58.0; 30.1; 20.1;
14.6. IR (neat) m 3237; 2926; 1684. MS (m/z) 383.41
[M+H]+.
4.1.20. Synthesis of 2-[2-(2-azido-ethoxy)-ethoxy]-ethyl-
amine (22). 1,2-Bis(2-chloroethoxy)-ethane (10.6 g,
56.7 mmol), sodium azide (12.0 g, 184.6 mmol), and so-
dium iodide (1.2 g, 8.0 mmol) were stirred in refluxing
acetonitrile (60 mL) for 7 d. The solvent was removed
under vacuum and diethyl ether (100 mL) was added.
The resulting suspension was successively washed with
water, aqueous Na2S2O5, water, and brine. The organic
layer was dried over MgSO4 and reduced in vacuo to
yield analytically pure 1,2-bis(2-azidoethoxy)-ethane
(11.2 g, 99%). 1H NMR (CDCl3, 300 MHz) d 3.71–
3.68 (m, 8H); 3.39 (t, J = 5.1 Hz, 4H). 13C NMR
(CDCl3, 75 MHz) d 71.2; 70.6; 51.1. MS (m/z) 200.78
[M+H]+. Triphenyl phosphine (9.09 g, 34.6 mmol) in
diethyl ether (50 mL) was added dropwise over a 3-h
4.1.18. Synthesis of 5-furanoyl-4-(3-hydroxy-phenyl)-6-
methyl-2-thioxo-1,2,3,4-tetrahydro-pyrimidine
(20).
Ethyl furoate (5.16 g, 40.0 mmol) in anhydrous acetone
(2.9 mL, 40.0 mmol) was added to a suspension of t-
BuOK (9.00 g, 80.0 mmol) in anhydrous toluene
(75 mL) at 0 ꢁC. The mixture was stirred for 6 h at room
temperature before acetic acid (4.60 mL, 80.0 mmol)
and water (30 mL) were added. The organic layer was
washed with water, brine, dried over MgSO4, and
reduced in vacuo. Purification by silica gel chromatogra-
phy (Et2O/hexane 10:90) afforded 1-(furan-2-yl)butane-
1,3-dione (3.85 g, 63%) as a viscous oil that slowly
crystallized at room temperature. Both conjugated enol
and diketone forms are observed by NMR in a 80/20
ratio. 1H NMR (CDCl3, 300 MHz) enol form d 7.55
(dd, J = 0.9, 1.8 Hz, 1H); 7.13 (dd, J = 0.9, 3.6 Hz,
1H); 6.52 (dd, J = 1.8, 3.6 Hz, 1H); 6.05 (s, 1H); 2.12
(s, 3H); diketone form d 7.59 (dd, J = 0.9, 1.8 Hz, 1H);
7.25 (dd, J = 0.6, 3.6 Hz, 1H); 6.55 (dd, J = 1.8,
period
to
1,2-bis(2-azidoethoxy)-ethane
(6.94 g,
34.6 mmol) in Et2O/THF/HCl 1 N (100 mL, 5:1:5) at
room temperature. The reaction mixture was vigorously
stirred for 12 h before the organic layer was washed with
HCl 4 N. The aqueous layer was then washed twice with

6486
E. Klein et al. / Bioorg. Med. Chem. 15 (2007) 6474–6488
diethyl ether and brought to pH 14 by adding sodium
hydroxide pellets. The resulting aqueous solution was
washed twice with CH2Cl2, the organic layers were
pulled, dried over MgSO4, and reduced under vacuum
to yield analytically pure compound 22 (6.01 g, 99%).
1H NMR (CDCl3, 300 MHz) d 3.62–3.57 (m, 6H); 3.49
(t, J = 5.1 Hz, 2H); 3.35 (t, J = 5.1 Hz, 2H); 2.84 (t,
J = 5.4 Hz, 2H); 2.32 (s, 2H). 13C NMR (CDCl3,
75 MHz) d 73.6; 70.9; 70.6; 70.4; 51.0; 42.0. MS (m/z)
174.96 [M+H]+.
The control (non-derivatized) chromatography matrix
was prepared differently. NHS-activated Sepharose 4
Fast Flow (16–23 lmol/mL, 1 mL) was washed with
HCl 1 mM (15 mL) at 4 ꢁC and suspended in 100 mM
ethanolamine (10 mL). The suspension was gently sha-
ken for 1 h, extensively washed with ultra pure water
and stored in water/ethyl alcohol (8:2) until use.
4.2. ATPase assays
Measurements of ATPase rates in the absence and pres-
ence of MTs were performed in duplicate at room tem-
perature. Steady-state basal and MT-activated ATPase
rates were measured with Eg52–386 using the pyruvate ki-
nase/lactate dehydrogenase-linked assay.24 To test the
inhibition of MT-stimulated Eg5 ATPase activity, Eg5
was used at 40 nM in the presence of 1 lM MTs. The as-
says were performed in the presence of DMSO (2.5%).
The maximal inhibitor concentrations tested were
500 lM. The data were treated using Kaleidagraph 3.0
(Synergy Software). The inhibition of the basal Eg5
ATPase activity (in the absence of MTs) was measured
in the presence of 1.0 lM Eg5. Paralleled experiments
were carried out with acetylated monastrol derivatives
8, 9, 10, and 21 with prior incubation of the drugs
(2 mM) for 12 h at 37 ꢁC in Dulbecco’s modified Eagle’s
medium (GIBCO, BRL) supplemented with 10% fetal
bovine serum (Hyclone) and subsequent measurement
of the basal Eg5 ATPase activity.
4.1.21. Synthesis of 1-azido-5-[4-(3-hydroxy-phenyl)-6-
methyl-2-thioxo-1,2,3,4-tetrahydro-pyrimidine-5-carbox-
amido]-3-oxa-pentane (23). DCC (252 mg, 1.22 mmol)
was added to a mixture of acid 11 (265 mg, 1.00 mmol),
azidoamine 22 (188 mg, 1.08 mmol), and triethylamine
(200 lL, 1.43 mmol) in anhydrous THF (10 mL). The
solution was stirred at room temperature overnight.
The precipitate was removed by filtration, the filtrate
was reduced under vacuum, and the residue was puri-
fied by silica gel chromatography (EtOAc/CH2Cl2
80:20 to 100:0) to yield compound 23 (260 mg, 62%)
as a white solid. 1H NMR (CDCl3/CD3OD 1:1,
300 MHz) d 7.13 (t, J = 8.4 Hz, 1H); 6.73–6.71 (m,
3H); 5.24 (s, 1H); 3.61 (t, J = 4.8 Hz, 4H); 3.55–3.34
(m, 6H); 3.35 (t, J = 5.1 Hz, 2H); 2.07 (s, 3H). 13C
NMR (CDCl3/CD3OD 1:1, 50 MHz) d 174.0; 167.6;
157.7; 143.5; 134.7; 130.4; 118.2; 115.8; 113.7; 107.6;
70.6; 70.2; 70.1; 69.7; 56.7; 50.8; 39.6; 16.7. IR (neat)
m 3262 (b); 2924; 2360 (b); 2108. MS (m/z) 422.24
[M+H]+.
4.3. Cell-based assays
4.1.22. Synthesis of 1-amino-5-[4-(3-hydroxy-phenyl)-6-
methyl-2-thioxo-1,2,3,4-tetrahydro-pyrimidine-5-carbox-
amido]-3-oxa-pentane (24). Azide 23 (210 mg,
0.50 mmol) and triphenyl phosphine (198 mg, 0.75
mmol) were stirred with water (250 lL) in THF
(5 mL) at room temperature overnight. THF was re-
moved under reduced pressure, water (10 mL) was
added, and the solution was washed twice with CH2Cl2.
The aqueous layer was lyophilized to yield analytically
pure amine 24 (159 mg, 81%). 1H NMR (CD3OD,
300 MHz) d 6.91 (t, J = 8.1 Hz, 1H); 6.52–6.45 (m,
3H); 5.05 (s, 1H); 3.30–3.02 (m, 10H); 2.53 (t,
J = 5.4 Hz, 2H); 1.83 (s, 3H). 13C NMR (CDCl3,
75 MHz) d 175.6; 169.2; 159.3; 145.2; 135.0; 130.9;
118.5; 116.3; 114.7; 108.9; 73.1; 71.2; 71.1; 70.3; 57.6;
41.9; 40.4; 16.8. IR (neat) m 3257 (b); 2926; 2360. MS
(m/z) 396.21 [M+H]+.
HeLa cells were grown on Dulbecco’s modified Eagle’s
medium supplemented with 10% fetal bovine serum
and maintained in a humid incubator at 37 ꢁC in 5%
CO2. Cells were left to adhere for at least 24 h on
poly-D-lysine-coated glass coverslips before the addition
of the drugs. Following 7-h incubation with drugs, cells
were fixed with 1% paraformaldehyde-PBS at 37 ꢁC for
3 min followed by 5-min incubation in 100% methanol
at ꢀ20 ꢁC. Coverslips were then washed with PBS and
cells were stained with anti-b-tubulin monoclonal anti-
bodies for 1 h and then with an FITC-conjugated goat
anti-mouse secondary antibody (Jakson ImmunoRe-
search Laboratories, West Grove, PA USA) for 30 min
and counterstained with propidium iodide. Images were
collected with a MRC-600 Laser Scanning Confocal
apparatus (BioRAD Laboratories) coupled to a Nikon
Optiphot microscope.
4.1.23. Synthesis of monastrol affinity matrix (25). NHS-
activated Sepharose 4 Fast Flow (16–23 lmol/mL,
6 mL) was washed with HCl 1 mM (75 mL) at 4 ꢁC
and suspended in water (10 mL). Amine 24 (56 mg,
0.14 mmol) was added, the suspension was brought to
pH 7 with Et3N and gently shaken overnight at room
temperature. The resin was filtered, washed successively
with water (10 mL), 1 mM ethanolamine (20 mL, pH 8),
50 mM Tris buffer (10 mL, pH 8), and 70 mM AcOH
(10 mL, pH 4). Tris buffer and AcOH washings were re-
peated twice and, finally, the resin was extensively
washed with ultra pure water, and stored in water/ethyl
alcohol (8:2) until use.
Cells were analyzed by two-dimensional flow cytome-
try using MPM-2 monoclonal antibody recognizing
mitosis specific phosphoepitopes,30 and propidium
iodide, a marker of DNA content. Cells were fixed,
incubated with MPM-2 antibodies, and labeled with
FITC-conjugated goat anti-mouse IgG secondary anti-
bodies and propidium iodide as described previously.31
Data were collected using a FACScan flow cytometer
(Becton Dickinson, San Jose, CA), with propidium
iodide in the first, and MPM-2 in the second dimen-
sion, using Cellquest software, and for each sample
10,000 events were collected and aggregated cells were
gated out.

E. Klein et al. / Bioorg. Med. Chem. 15 (2007) 6474–6488
6487
4.4. Protocol for Eg5 purification by affinity
chromatography
Swiss-Prot database, setting mass accuracy to 50 ppm
(MALDI-TOF/TOF) for peptide mass fingerprinting
and 0.4 Da for MS/MS, allowing for up to one missed
cleavage site, pyro-Glu formation at N-terminal Gln,
oxidation of methionine, N-terminal acetylation of the
protein, and modification of cysteines by acrylamide.
The MS/MS spectra were further manually confirmed
considering the criteria for MALDI-TOF/TOF de-
scribed elsewhere.33
HeLa cells were plated in 10 · 15-cm plates and blocked
in mitosis with 100 ng/mL nocodazole (SIGMA). All
subsequent steps were performed at 4 ꢁC. Mitotic cells
were collected by shake off, washed with PBS, resus-
pended in lysis buffer (20 mM PIPES, 1 mM EGTA,
1 mM MgCl2, 0.2 M NaCl, 1 mM DTT), and then lysed
by homogenization. Following centrifugation at 100 kg
for 1 h, the supernatant was collected (3 mL; protein
content: 2.0 mg/mL) and loaded onto a column condi-
tioned with the control (non-derivatized) chromatogra-
phy matrix (1 mL). The flow-through was directly
loaded onto a second column packed with the affinity
matrix 25 (Sepharose-monastrol beads; 1 mL; flow-
through: 3 mL; protein content: 1.2 mg/mL). Both col-
umns were washed with 5 mL of lysis buffer. Material
from the non-active first column was eluted with 2 M
NaCl, whereas bound Eg5 was released from the second
column with 5 mL of lysis buffer supplemented with
1 mM STLC. The eluates were collected as 500 lL frac-
tions. The eluted fractions with Eg5 contained 30–70 lg/
mL of protein. Equal volumes from each fraction were
taken and SDS gel samples were prepared in Lammli
buffer and loaded in 8% SDS-gel electrophoresis fol-
lowed by transfer to nitrocellulose sheets using a semi-
dry transfer apparatus. Blots were incubated with
rabbit anti-Eg5 antibodies (directed against the specific
C-terminal part of the human protein) used at 1/1000
dilution and then exposed to horseradish peroxidase-
conjugated goat anti-rabbit IgG, diluted 1/5000, and
developed by echochemiluminescence (ECL, Amer-
sham). A similar procedure was applied using the con-
trol resin.
Acknowledgments
This work has been funded by grants from ARC (Asso-
ciation pour la Recherche sur le Cancer, numbers 5197
´
`
& 3973) and MENRT (Ministere de l’Education Natio-
nale, de la Recherche et de la Technologie). We thank
Medhi Boulifa for assistance in the preparation of com-
pounds 18 and 19, Anne Blangy for providing us with
the Eg5 antibody, David Lemaire for mass spectrometry
analysis, and Rose-Laure Indurato for excellent techni-
cal assistance during the cell-based assays.
References and notes
1. Vale, R. D.; Reese, T. S.; Sheetz, M. P. Cell 1985, 42, 39–
50.
2. Coleman, P. J.; Fraley, M. E. Expert Opin. Ther. Pat.
2004, 14, 1659–1667.
3. Duhl, D. M.; Renhowe, P. A. Curr. Opin. Drug Discov.
Dev. 2005, 8, 431–436.
4. Bergnes, G.; Brejc, K.; Belmont, L. Curr. Top. Med.
Chem. 2005, 5, 127–145.
5. Pellegrini, F.; Budman, D. R. Cancer Invest. 2005, 23,
264–273.
6. Quasthoff, S.; Hartung, H. P. J. Neurol. 2002, 249, 9–17.
7. Sawin, K. E.; Mitchison, T. J. Proc. Natl. Acad. Sci.
U.S.A. 1995, 92, 4289–4293.
8. Blangy, A.; Lane, H. A.; d’Herin, P.; Harper, M.; Kress,
M.; Nigg, E. A. Cell 1995, 83, 1159–1169.
9. Mayer, T. U.; Kapoor, T. M.; Haggarty, S. J.; King, R.
W.; Schreiber, S. L.; Mitchison, T. J. Science 1999, 286,
971–974.
10. Kapoor, T. M.; Mayer, T. U.; Coughlin, M. L.; Mitch-
ison, T. J. J. Cell Biol. 2000, 150, 975–988.
11. Maliga, Z.; Kapoor, T. M.; Mitchison, T. J. Chem. Biol.
2002, 9, 989–996.
12. Haque, S. A.; Hasaka, T. P.; Brooks, A. D.; Lobanov, P.
V.; Baas, P. W. Cell Motil. Cytoskel. 2004, 58, 10–16.
13. Yoon, S. Y.; Choi, J. E.; Huh, J. W.; Hwang, O.; Lee, H.
S.; Hong, H. N.; Kim, D. Cell Motil. Cytoskel. 2005, 60,
181–190.
14. Biginelli, P. Gazz. Chim. Ital. 1893, 23, 360–413.
15. Gartner, M.; Sunder-Plassmann, N.; Seiler, J.; Utz, M.;
Vernos, I.; Surrey, T.; Giannis, A. ChemBioChem 2005, 6,
1–6.
16. Maliga, Z.; Mitchison, T. J. BMC Chem. Biol. 2006, 6, 2.
17. Kappe, C. O. Tetrahedron 1993, 49, 6937–6963.
18. Kappe, C. O. Acc. Chem. Res. 2000, 33, 879–888.
19. Atwal, K. S.; Rovnyak, G. C.; Schwartz, J. J. Org. Chem.
1989, 54, 5898–5907.
20. DeBonis, S.; Simorre, J. P.; Crevel, I.; Lebeau, L.;
Skoufias, D. A.; Blangy, A.; Ebel, C.; Gans, P.; Cross,
R.; Hackney, D. D.; Wade, R. H.; Kozielski, F. Biochem-
istry 2003, 42, 338–349.
4.5. Protein identification by mass spectrometry
Proteins from the eluted fractions were further separated
by 10% SDS–PAGE with a MiniProtean 3 cell (Bio-
Rad, Munich, Germany), stained with Coomassie
G-250, and identified by peptide mass fingerprinting
and tandem mass spectrometry (MS/MS). The gel bands
were excised with a scalpel for in-gel digestion with
0.1 lg of trypsin in 20 lL of 50 mM ammonium hydro-
genocarbonate, pH 7.8. An ULTRAFLEX II (Bruker
Daltonics, Bremen, Germany) MALDI-TOF/TOF mass
spectrometer was used with a mass accuracy of 50 ppm
after external calibration. The samples were analyzed in
the MS mode (for generation of peptide mass finger-
prints) as well as in the TOF/TOF mode (for fragmenta-
tion analysis of the 1–2 most intense peaks). The matrix
used was a-cyano-4-hydroxycinnamic acid (20 mg/mL)
in 0.3% aqueous trifluoroacetic acid/acetonitrile (2:1).
The samples were applied to a stainless steel sample
holder and introduced into the mass spectrometer after
drying. Mass spectra were transformed into peak lists
by using FlexAnalysis v.2.4 (Bruker Daltonics, Bremen,
Germany). The peak lists of the MS and MS/MS spectra
were merged by BioTools v.3.0 (Bruker Daltonics,
Bremen, Germany). Proteins were identified using an
in-house copy of the protein identification software pro-
gram Mascot.32 Mass spectra were searched against the

6488
E. Klein et al. / Bioorg. Med. Chem. 15 (2007) 6474–6488
21. DeBonis, S.; Skoufias, D. A.; Lebeau, L.; Lopez, R.;
Robin, G.; Margolis, R. L.; Wade, R. H.; Kozielski, F.
Mol. Cancer Ther. 2004, 3, 1079–1090.
22. Yan, Y. W.; Sardana, V.; Xu, B.; Homnick, C.; Hal-
czenko, W.; Buser, C. A.; Schaber, M.; Hartman, G. D.;
Huber, H. E.; Kuo, L. C. J. Mol. Biol. 2004, 335, 547–554.
23. Garcia-Saez, I.; DeBonis, S.; Lopez, R.; Trucco, F.;
27. Sprague, J. M.; Johnson, T. B. J. Am. Chem. Soc. 1937,
59, 1837–1840.
28. Sweet, F.; Fissekis, J. D. J. Am. Chem. Soc. 1973, 95,
8741–8749.
29. Donleavy, J. J. J. Am. Chem. Soc. 1936, 58, 1004–
1005.
30. Davis, F. M.; Tsao, T. Y.; Fowler, S. K.; Rao, P. N. Proc.
Natl. Acad. Sci. U.S.A. 1983, 80, 2926–2930.
31. Andreassen, P. R.; Skoufias, D. A.; Margolis, R. L.
Methods Mol. Biol. 2004, 281, 213–225.
32. Perkins, D. N.; Pappin, D. J.; Creasy, D. M.; Cottrell, J. S.
Electrophoresis 1999, 20, 3551–3567.
33. Schmidt, F.; Krah, A.; Schmid, M.; Jungblut, P. R.;
Thiede, B. Rapid Commun. Mass Spectrom. 2006, 20, 933–
936.
´
Rousseau, B.; Thuery, P.; Kozielski, F. J. Biol. Chem.
2007, 282, 9740–9747.
24. Hackney, D. D.; Jiang, W. Methods Mol. Biol. 2001, 164,
65–71.
25. Braibante, M. E. F.; Braibante, H. S.; Missio, L.;
Andricopulo, A. Synthesis 1994, 898–900.
26. Lebeau, L.; Oudet, P.; Mioskowski, C. Helv. Chim. Acta
1991, 74, 1697–1706.