10-(2-oxo-2-Phenylethylidene)-10H-anthracen-9-ones as highly active antimicrotubule agents: Synthesis, antiproliferative activity, and inhibition of tubulin polymerization

Article abstract of DOI:10.1021/jm801338r

A series of 10-(2-oxo-2-phenylethylidene)-10H-anthracen-9-ones were synthesized and evaluated for interactions with tubulin and for antiproliferative activity against a panel of human and rodent tumor cell lines. The 4-methoxy analogue 17b was most potent

Full text of DOI:10.1021/jm801338r

1284
J. Med. Chem. 2009, 52, 1284–1294
10-(2-oxo-2-Phenylethylidene)-10H-anthracen-9-ones as Highly Active Antimicrotubule Agents:
Synthesis, Antiproliferative Activity, and Inhibition of Tubulin Polymerization
Helge Prinz,*,† Peter Schmidt,‡ Konrad J. Bo¨hm,§ Silke Baasner,‡ Klaus Mu¨ller,† Eberhard Unger,§ Matthias Gerlach,‡ and
Eckhard G. Gu¨nther‡
Institute of Pharmaceutical and Medicinal Chemistry, Westphalian Wilhelms-UniVersity, Hittorfstrasse 58-62, D-48149 Mu¨nster, Germany,
Leibniz Institute for Age ResearchsFritz Lipmann Institute (FLI), Beutenbergstrasse 11, D-07745 Jena, Germany, ÆternaZentaris GmbH,
Weismu¨llerstrasse 50, D-60314 Frankfurt, Germany
ReceiVed October 23, 2008
A series of 10-(2-oxo-2-phenylethylidene)-10H-anthracen-9-ones were synthesized and evaluated for
interactions with tubulin and for antiproliferative activity against a panel of human and rodent tumor cell
lines. The 4-methoxy analogue 17b was most potent, displaying IC50 values ranging from 40 to 80 nM,
including multidrug resistant phenotypes, and had excellent activity as an inhibitor of tubulin polymerization
(IC50 ) 0.52 µM). Concentration-dependent flow cytometric studies showed that KB/HeLa cells treated
with 17b were arrested in the G2/M phases of the cell cycle (EC50 ) 90 nM). In competition experiments,
17b strongly displaced [3H]-colchicine from its binding site in the tubulin. The results obtained demonstrate
that the antiproliferative activity is related to the inhibition of tubulin polymerization.
Introduction
nesulfonic acid esters 11 and 12 (Chart 2).15-17 These compounds
strongly inhibited tumor cell growth, tubulin polymerization,
and caused also significant arrest of mitosis. This previous work
prompted us to extend our studies to structurally related
anthracenone-based enone systems in which the anthracenone
moiety is linked directly to the termini of an enone group.
Several previous investigations on benzophenones, chalcones,
and related enones have demonstrated their potent antimitotic
activities.18,19 In this report, we describe the synthesis and
biological evaluation of 10-(2-oxo-2-phenylethylidene)-10H-
anthracen-9-ones as novel tubulin polymerization inhibitors.
Several of these compounds inhibited the growth of various
tumor cell lines, acted in a cell-cycle-dependent manner, and
were found to be potent inhibitors of tubulin polymerization.
Antitubulin activity of the most active compound is comparable
or superior to those of the reference compounds, such as
nocodazole, podophyllotoxin, and colchicine.
The mitotic spindle, constituted by microtubules generated
by polymerization of Rꢀ-tubulin dimers, is considered one of
the most important targets to treat many types of malignancies.1,2
A still expanding family of antimitotic drugs displaying a wide
structural heterogeneity have been identified to act on tubulin.
Many of these agents work by inhibiting polymerization of
tubulin into microtubules, others by stabilizing the microtubule
structure.3-6 The Vinca bis-indole alkaloids vinblastine (1a,
Chart 1) and vincristine (1b) as well as the taxanes are well
established to treat a broad range of leukemias and lymphomas
as well as many types of solid tumors. It was mainly the
discovery of these compounds that has stimulated intense efforts
aimed at the development of further microtubule-targeting drugs.
Colchicine (2) has not been successfully used as an anticancer
agent, due primarily to its narrow therapeutic window, but
played an important role in deciphering the properties and
functions of tubulin and microtubules. Many natural products,
such as combretastatin A-47 (3) or the epothilones8 as well as
some synthetic molecules including sulfonamide E-70109 (4),
4-anilinoquinazoline10 (5), acridinyl-9-carboxamide D-8231811
(6), indazole-based TH-48212 (7), or propenenitrile CC-507913
(8) (Chart 1), to name just a few, are known to mediate
antiproliferating activities through binding to tubulin. In recent
years, substantial progress has been made in the identification
of small-molecular colchicine-site binders derived from natural
sources or by screening compound libraries in combination with
traditional medicinal chemistry.3,5,14 However, no representative
of this class has yet been introduced into cancer chemotherapy.
Chemistry
The general method for the synthesis of the novel 10-
substituted anthracenones is outlined in Scheme 1. For the
synthesis of compounds 17b, 17d, 17f, 17g, and 17j as well as
for 17l and 17m, the acid chloride 16 was synthesized from 13
and then reacted with suitable aromatic compounds in a
Friedel-Crafts acylation reaction (Scheme 1). A drawback is
that the reaction is limited to reactive arenes and therefore lacks
variability. Therefore, the aldol condensation reaction of 13 with
phenylglyoxal hydrates was thought to be an alternative
method.20 However, the synthesis of the desired 10-(2-oxo-2-
phenylethylidene)-10H-anthracen-9-ones required a careful choice
of the reaction conditions. Condensation reaction of 10H-
anthracen-9-one 13 with aromatic aldehydes is generally ap-
plicable21 and was thought to be the method of choice. Initial
attempts using gaseous hydrogen chloride22,23 or pyridine/
piperidine24 for the condensation reaction unexpectedly failed.
For this reason, we investigated different reaction conditions.
Performing the aldol-type reaction in ethanol/piperidine as
described for 14, afforded 10-(1-hydroxy-2-oxo-2-phenyl-ethyl)-
10H-anthracen-9-ones 19a-c (Scheme 2). Although these
compounds proved to be ineffective in the biological assays
We have recently reported the potent in vitro antitumor
activity of 10-[(3-hydroxy-4-methoxy-benzylidene)]-9(10H)-
anthracenone (9), 9-[(4-hydroxy-3,5-dimethoxy-benzylidene]-
naphtho[2,3-b]thiophen-4(9H)-one (10) and 4-methoxybenze-
* To whom correspondence should be addressed. Phone: +49 251-
8332195. Fax: +49 251-8332144. E-mail: prinzh@uni-muenster.de.
†
Institute of Pharmaceutical and Medicinal Chemistry, Westphalian
Wilhelms-University.
‡
ÆternaZentaris.
Leibniz Institute for Age ResearchsFritz Lipmann Institute.
§
10.1021/jm801338r CCC: $40.75
2009 American Chemical Society
Published on Web 02/16/2009

10-(2-oxo-2-Phenylethylidene)-10H-anthracen-9-ones
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 5 1285
Chart 1. Examples of Tubulin Interacting Agents
Chart 2. Tubulin Interacting Benzylidenes 9 and 10 and Sulfonates 11 and 12
(see below), they were valuable intermediates for further
syntheses. In this connection, we additionally obtained 23 when
13 reacted with the aliphatic glyoxyl dimethylacetal under the
same conditions. This is remarkable, as 13 is known to react
only with more reactive aliphatic aldehydes. At the end,
synthesis of further 10-(2-oxo-2-phenylethylidene)-10H-anthra-
cen-9-ones (17a, 17c, 17h) was successfully accomplished by
condensation reaction of 13 with appropriately substituted
phenylglyoxal hydrates20 in boiling acetic acid anhydride. The
starting phenylglyoxal hydrates were obtained from commercial
sources. For the preparation of compounds 17e, 17i, and 17k
(Scheme 1), ether groups of 17b, 17d, and 17g had to be cleaved
by BBr3 in dichloromethane or AlCl3 in DCE. As an important
aspect, the use of phenylglyoxal derivatives enhanced the
structural variability and especially gave access to constitutional
isomers, which we suggested would not easily be obtained by
means of Friedel-Crafts acylation reaction (17b versus 17c,
17g versus 17h). For SAR studies and in order to investigate
the relevance of the enone CdC-double bond, we were
interested in the synthesis of 21. However, attempts to alkylate15,25
13 at C-10 were unsuccessful (Scheme 3). Interestingly, in
contrast to the previously described C-10 alkylation of 9(10H)an-
thracenone,15 reaction of 13 with R-bromoacetophenone in the
presence of potassium carbonate in acetone afforded O-alkylated
enol 22 instead of C-alkylation. Keto-enol tautomerism in the
anthrone series has been described,26 and the formation of
compound 22 is suggested to be due to a base-catalyzed
enolization of 13. Furthermore, to investigate the relevance of
the anthracenone ketone function for activity, 17b was selec-
tively reduced using sodium borohydride to afford 10-hydroxy-
substituted anthracenylidene 20 (Scheme 4). Esters 18a and 18b
were initially prepared under conditions of the Friedel-Crafts
acylation reaction as starting materials for the synthesis of
hydroxy-methoxy aryl ketones. Unfortunately, preparation via
Fries rearrangement reaction27 of the esters by catalysis of Lewis
acids failed.
Biological Results and Discussion
In Vitro Cell Growth Inhibition Assay. In an initial screen,
the compounds synthesized were evaluated for antiproliferative
activity against the human chronic myelogenous leukemia cell
line K562,28 which is widely used for potential antitumor
compounds. Cell proliferation was determined directly by
counting the cells with a hemocytometer after 48 h of treatment.
Table 1 summarizes the biological data for inhibition of K562
cell growth and inhibition of tubulin polymerization (see below).
Data obtained with several reference compounds are also
presented. In the 10-(2-oxo-2-phenylethylidene)-10H-anthracen-
9-one series, five compounds (17b, 17e, 17h, 17j, and 17l)
showed IC50 values in the submicromolar (lower than 1 µM)
range. Compound 17b was the most potent and displayed strong
antiproliferative activity with an IC50 value of lower than

1286 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 5
Scheme 1. R1-R5 are defined in Table 1a
Prinz et al.
a Reagents and conditions: (a) OHC-COOH ·H2O, ethanol/piperidine, 90 °C, N2; (b) CH3OH, H2SO4, reflux; (c) SOCl2, CH2Cl2, reflux, 2 h; (d) 16, DCE,
AlCl3, 0 °C, 2 h; for 17j toluene instead of DCE; (e) BBr3, CH2Cl2, -78 °C to rt, N2; (f) AlCl3, DCE, 0 °C.
0.1 µM. Replacement of the 4-methoxy group of 17b (K562;
IC50 0.07 µM) with a 4-methyl group (17j) gave a 9-fold loss
of potency. Additionally, a marked loss in potency was observed
for the isomeric 17c, indicating that a methoxy group placed at
the para-position of the terminal phenyl ring is important, while
shifting it to the meta-position is detrimental to potency.
Bismethoxy compound 17d was only a moderate inhibitor
(K562; IC50 1.3 µM), whereas cleavage of the ether groups (17e)
resulted in a nearly 8-fold increase in potency. Interestingly,
introduction of a 3,4,5-trimethoxyphenyl group in 17h, a well
defined pharmacophore for the inhibition of tubulin polymeri-
zation found in colchicine, combretastatin A-4, and podophyl-
lotoxin, caused an approximately 10-fold drop in potency in
comparison with 17b. This is in agreement with similar
observations of our previous reports.15-17 Moreover, constitu-
tional isomer 17g was examined for comparison with 17h, and
activity was again strongly reduced (17h, K562; IC50 ) 0.7 µM
versus 17g, IC50 ) 11 µM). Because electronic properties of
these compounds are similar, loss of activity is presumably due
primarily to steric limitations around the carbonyl function near
the phenyl ring. Also, for 17i (K562; IC50 > 30 µM), bearing
a phenolic group neighboring the enone carbonyl function and
participating in an intramolecular hydrogen-bonding, activity
was found to be more or less eliminated as compared to 17g.
The unsubstituted compound 17a displayed moderate antipro-
liferative activity with an IC50 value of 3.6 µM. Potency was
found to be almost identical with the phenolic 17k (K562; IC50
) 3.43 µM), which is more polar than 17b. Moreover, 10-(1-
hydroxy-2-oxo-2-phenyl-ethyl) substituted anthracenones
19a-19c revealed a dramatic loss of antiproliferative activity.
This is obviously closely related to the loss of the double bond
at C-10 and saturation. Therefore we constitute that conforma-
tional flexibility is not well tolerated in this position. These
results confirmed earlier observations, which had been made in
the benzylidene-9(10H)anthracenone series.15 Replacement of
the phenyl ring of 17a and 17b with a bioisosteric thiophene in
17l and 17m, respectively, resulted in about a 10-fold reduction
in potency for the methoxy-substituted 17l (K562; IC50 ) 0.69
µM) and merely weak activity for the unsubstituted 17m (IC50
5.2 µM).
Moderate antiproliferative activity was found for the phenolic
esters 18a and 18b, which displayed IC50 values in the range
of 1-2 µM. Methyl ester 15 was also examined and found to
be less potent than 18a and 18b with an IC50 of 6 µM. Reduction
of the anthracenone keto group in 17b to a secondary alcohol
group yielded 20, which had a 3-fold reduced potency as

10-(2-oxo-2-Phenylethylidene)-10H-anthracen-9-ones
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 5 1287
Scheme 2. R1-R3 are defined in Table 1a
a Reagents and conditions: (a) 3-methoxyphenyl-, 3,4,5-trimethoxyphenyl-, or phenylglyoxal hydrate, (CH3CO)2O, reflux; (b) 3-methoxyphenyl-, 3,4,5-
trimethoxyphenyl-, or phenylglyoxal hydrate, C2H5OH, piperidine, reflux, 6 h; (c) glyoxal dimethyl acetal (60% aq soln), C2H5OH, piperidine, reflux, 6 h.
Scheme 3a
Scheme 4a
a Reagents and conditions: (a) NaBH4, ethanol/THF, rt.
cell lines derived from human tumors was measured by cellular
metabolic activity using the XTTa assay.29 Of the tested
compounds, 17b displayed high overall potencies, with IC50
values in the range of 60-80 nM toward several proliferating
cell lines (Table 2). Thus 17b showed activities comparable to
those of colchicine. Compound 17e was also confirmed to be
active with IC50 values ranging from 0.16 to 0.43 µM. Even
17d was found to be effective, although potency was found to
be substantially reduced. Furthermore, compounds 17b, 17d,
and 17e were not active against cell cycle arrested RKO cells
(human colon adenocarcinoma) with ectopic inducible expres-
a Reagents and conditions: (a) 2-bromo-4′-methoxyacetophenone, K2CO3,
acetone, reflux.
compared to 17b. This finding demonstrated the importance of
the intact anthracenone carbonyl functional group for potency.
Also, 10-hydroxy-substituted anthracenylidene derivative 20 was
among the most potent compounds, which gave rise to further
derivatization within this series. However, partial oxidation of
20 under the conditions of the cellular assay can not be excluded
at that point in time.
Effect on Growth of Different Tumor Cell Lines. To further
characterize the tumor cell growth inhibitory profile of the
compounds, the effect of the highly active p-methoxy analogue
17b and analogues 17d and 17e against a panel of five tumor
sion of cyclin-dependent kinase inhibitor p27kip1 30 By contrast,
.
growth of proliferating RKOp27-cells (not induced) was strongly
a Abbreviations: aq, aqueous; ADR, doxorubicin; CSI, colchicine site
inhibitor; DCE, 1,2-dichloroethane; ITP, inhibition of tubulin polymerization;
Kip, kinase inhibitor protein; MDR, multidrug resistant; MRP, multidrug
resistance protein; MTP, microtubule protein; ND, not determined; soln,
solution; VCR, vincristine; XTT, 2,3-bis-(2-methoxy-4-nitro-5-sulfophenyl)-
2H-tetrazolium-5-carboxanilide.

1288 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 5
Prinz et al.
Table 1. Inhibitory Effect of 10-(2-oxo-2-Phenylethylidene)-10H-anthracen-9-ones and Related Compounds on the Growth of K562 Cells and Tubulin
Polymerization
a
compd
R1
R2
R3
R4
R5
K562 IC50 [µM]
ITPb IC50 [µM]
15
6.0
ND
5.0
0.52
3.3
6
2.6
0.9
>10
3.7
>10
1.1
0.9
2.3
10
17a
17b
17c
17d
17e
17f
17g
17h
17i
17j
17k
17l
17m
18a
18b
19a
19b
19c
20
H
H
H
H
H
OCH3
OCH3
H
OH
H
H
OCH3
H
OCH3
OCH3
H
H
H
OCH3
OCH3
OH
H
OCH3
H
OCH3
OH
OCH3
OCH3
OCH3
OCH3
CH3
H
H
H
H
H
H
H
OCH3
H
H
H
H
H
H
H
H
H
H
H
H
H
3.60
0.07
2.60
1.30
0.16
ND
11
0.70
>30
0.64
3.43
0.69
5.21
1.9
1.64
27
18
21
0.21
10
H
OCH3
OCH3
OCH3
H
H
OH
H
H
H
H
OCH3
OCH3
H
H
H
H
H
H
H
OCH3
10
10
>10
>10
>10
1.20
ND
1.4
0.76
0.35
0.13
ND
H
OCH3
OCH3
22
colchicine
nocodazole
podophyllotoxin
vinblastine sulfate
adriamycin
0.02
ND
ND
0.001
0.01
a IC50 values are the means of at least three independent determinations (SD < 10%). b ITP ) inhibition of tubulin polymerization; IC50 values were
determined after 30 min at 37 °C and represent the concentration for 50% inhibition of the maximum tubulin assembly rate.
Table 2. Antiproliferative Activity of 17b, 17d, and 17e against
Table 3. Antiproliferative Activity of 17b and 17d, Paclitaxel,
Nocodazole, and Vindesine against Tumor Cell Lines with Different
Resistance Phenotypes (XTT Assay)a
Different Tumor Cell Linesa
ICa [µM]
50
ICa [µM]
50
RKOp27kip
LT12 LT12 MDR L1210 L1210 VCR P388 P388 ADR
not
17b
17d
0.05
0.37
0.05
0.44
0.40
0.05
0.26
0.05 0.05
0.97 0.97
0.06 >5
0.06 0.07
0.02 >5
0.04
0.90
0.04
0.07
0.01
0.05
1.1
>5
0.05
1.10
compd
17b
17d
KB/HeLa SKOV3 SF268 NCI-H460 induced induced
0.06
1.08
0.27
0.01
0.14
0.03
0.06
ND
0.16
0.01
0.17
0.05
0.06
2.7
0.22
0.01
0.30
0.05
0.07
1.4
0.43
0.01
0.15
0.07
0.08
0.51
0.24
0.01
0.11
0.02
>9
>9
paclitaxel 0.006
nocodazole 0.30
vindesine 0.001
17e
>9
paclitaxel
nocodazole
colchicine
>3
a All experiments were performed at least in two replicates (n ) 2), and
IC50 data were calculated from dose-response curves by nonlinear
regression analysis.
>10
>10
a All experiments were performed at least in two replicates (n ) 2), and
IC50 data were calculated from dose-response curves by nonlinear
regression analysis.
cellular pumps, such as the 170 kDa P-glycoprotein (P-gp),33
encoded by the mdr1 gene and the 180 kDa MDR protein
(MRP).34 Important aspects concerning the key mechanisms of
antimicrotubule drug resistance have recently been reviewed.31,35,36
Because colchicine, nocodazole, and paclitaxel are substrates
for the P-gp efflux pump, we compared them with representative
compounds 17b and 17d on the proliferation of tumor cell lines
with different resistance phenotypes in an XTT-based assay.
On the whole, as documented by the IC50 data (Table 3), 17b
and even 17d were effective against the wild type cell lines
and retained high activity in cell lines with various MDR
phenotypes. This behavior was distinct from reference com-
pounds like paclitaxel and vindesine, which expressed only low
inhibited by 17b, with an IC50 value of 0.08 µM and, somewhat
less potent, by 17e (IC50 ) 0.24 µM) and 17d (IC50 ) 0.51
µM), indicating activity in cells, which were not arrested in G1-
phase. Tubulin-targeting agents are used extensively for the
treatment of many human malignancies, leading frequently to
multiple drug resistance (MDR) in patients and a loss of efficacy
over time. As a major problem, cancer cells do not respond to
treatment or develop a broad spectrum resistance to several
anticancer drugs, among them antitubulin agents.31,32 Multiple
drug resistance is multifaceted phenomenon and mediated,
among other factors, by overexpression of transmembrane

10-(2-oxo-2-Phenylethylidene)-10H-anthracen-9-ones
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 5 1289
Table 4. Cell Cycle Analysis of KB/HeLa Cells Treated with 17b and
Reference Compounds Vincristine, Colchicine, Paclitaxel, and
nocodazole
17b colchicine nocodazole paclitaxel vincristine
ECa [nM]
90
14
91
49
2.4
50
a EC50 values were determined from dose-response cell cycle analysis
experiments and represent the concentration for 50% cells arrested in G2/M
phase after 24 h. All experiments were performed at least in two replicates
(n ) 2), and EC50 data were calculated from dose-response curves by
nonlinear regression analysis (GraphPad Prism).
activity in LT12MDR, L1210VCR, and P388ADR cell lines
(IC50 data ranging from 0.26 to >5 µM) than to the wild type
cell lines. We therefore conclude that the 10-(2-oxo-2-phenyl-
ethylidene)-10H-anthracen-9-ones investigated here are not
substrates of the P-gp efflux pump. However, the effect of
selective pressure of 17b on the nonresistant cell lines eventually
leading to an MDR status has not been studied yet.
Effect on Cell-Cycle Progression. By targeting the mitotic
spindle, microtubule inhibitors arrest the cell cycle during the
metaphase phase. As a consequence, mitosis is blocked at the
transition from metaphase to anaphase. To gain further insight
into the mode of action, the most active compound 17b was
assayed for its effects on cell cycle using an established KB/
HeLa (human cervical epitheloid carcinoma) cell-based assay
system. For a thorough comparison of 17b with known G2/M
cell cycle inhibitors, subconfluent KB/HeLa cells were exposed
to test compounds and the percentage of cells in G2/M phase
after 24 h was plotted against different concentrations of the
compounds. The concentration for 50% cells arrested in G2/M
phase by 17b was found to be in the range of 90 nM (Table 4),
thus being as active as nocodazole (EC50 ) 91 nM). For
comparison, the recently described anthracenone derivative 9
showed an EC50 value of 0.2 µM in a contemporaneous
experiment. In summary, the effect of 17b on cell cycle
progression correlated well with its strong antiproliferative and
antitubulin activity and is similar to that observed for the
majority of antimitotic agents. However, in this assay, the
reference compounds were comparable or superior to 17b, with
the order of activity vincristine > colchicine > paclitaxel >
nocodazole ∼ 17b.
Figure 1. (A) Inhibition of in vitro polymerization of tubulin (in total
1.2 mg/mL protein; ∼85% tubulin plus ∼15% microtubule-associated
proteins) at 37 °C by various concentrations of 17b (IC50 ) 0.52 µM).
The steady-state tubulin assembly level in the absence of inhibitor was
set 100%. (B) IC50 values were determined by sigmoidal fitting the
plot of the steady state levels of tubulin assembly (taken 30 min after
the temperature had been switched from 0 to 37 °C) against drug
concentration and represent the concentration for 50% inhibition of
the maximum tubulin polymerization level.
17b to be 2.1-fold more potent than 17j as a tubulin polymer-
ization inhibitor but found it to be 9.1 times more potent than
17j in the growth inhibitory assay. This is probably due to the
fact that the tumor cell growth assay and the cell-free tubulin
polymerization assay differ in a number of important issues,
such as tubulin concentration present within the cells, the
presence of different types of microtubule-associated proteins
as well as the possible effects of regulatory proteins expressed
in the cell but being absent in the tubulin assay. Another aspect
could be differences in cell permeability of the various
compounds. Moreover, compound 20 (ITP; IC50 ) 1.20 µM)
was slightly more potent than colchicine. Similar to the cell
growth inhibition assay, this documented that tubulin polym-
erization inhibiting properties can also be found with 10-
hydroxy-substituted anthracenylidenes as exemplified by 20. The
moderate to weak inhibitors of tumor cell growth showed only
poor inhibitory effects on tubulin polymerization.
Microtubule-destabilizing drugs often reveal specific tubulin
binding sites. Therefore, we determined if 17b interacted directly
with tubulin by binding to either the colchicine or paclitaxel-
binding domains on tubulin using a competition-binding scintil-
lation proximity assay.37 We found that 17b inhibited [3H]colch-
icine binding to biotinylated tubulin (Figure 2) with an IC50
value of 0.40 µM versus colchicine (IC50 ) 1.06 µM). This
result was consistent with the growth inhibitory and tubulin
polymerization inhibitory activity. However, 17b did not
compete with [3H]paclitaxel (IC50 > 100 µM). No stabilization
of the colchicine binding was observed, as it is documented for
Vinca site binders.38,39 Therefore, we conclude that binding to
tubulin at the colchicine-binding site is likely and that the
In Vitro Tubulin Polymerization Assays. To investigate
whether the antiproliferative activities of the novel compounds
were related to an interaction with tubulin, 21 analogues were
assayed for their inhibitory effects on tubulin polymerization
using assay conditions as described previously.16 If an inhibitor
binds to or interferes with tubulin, the assembly steady-state
level is decreased, as exemplified for 17b (Figure 1). The results
obtained with the test agents are summarized in Table 1. For
comparison, the potent antimitotic compounds colchicine,
podophyllotoxin, nocodazole, and vinblastine were also exam-
ined. Compound 17b was again the most potent (ITP; IC50
0.52 µM), having nearly twice the potency of 17f (ITP; IC50
)
)
0.9 µM), 17j (ITP; IC50 ) 1.1 µM), and 17k (ITP; IC50 ) 0.9
µM). Nevertheless, these four compounds proved to be excep-
tionally strong inhibitors of tubulin polymerization (IC50 e 1.1
µM), as did the reference drugs with the exception of colchicine
(ITP; IC50 ) 1.4 µM). Seven other compounds had IC50 values
ranging from 1.2-6 µM and were relatively potent tubulin
polymerization inhibitors. In general, compounds having IC50
values in the range of g10 µM were regarded to have no
appreciable activity as an inhibitor of tubulin polymerization.
Obviously, there is more variability in the IC50 data for inhibition
of K562 cell growth (Table 1) than in the IC50 values for
inhibition of tubulin polymerization. For example, we found

1290 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 5
Prinz et al.
in 17b, bind to the same region as the colchicine C ring,42 we
hypothesize that the anthracenone system might function as a
hydrophobic center, similar to the colchicine A ring. The
anthracenone carbonyl oxygen probably has importance as a
hydrogen bond acceptor. Reduction of this carbonyl group
resulted in a decreased inhibition of tubulin polymerization. The
novel compounds 17a-17m embody an enone group as a linker
between two aryl groups. The carbonyl function in the linker
between the terminal aryl ring and the anthracenone part might
provide additional binding features. Obviously, this carbonyl
group in 17b could be accommodated by the colchicine binding
site, which is supported by very similar data obtained from the
[3H]colchicine binding assay for both 17b and 9 (9, 0.37 µM;
17b, 0.40 µM). However, it has to be pointed out that this
structural modification gave no substantially greater potency of
17b relative to 9 for the inhibition of tubulin polymerization
(9, 0.67 µM; 17b 0.52 µM). Notably, while the tubulin activities
differ only slightly, we found 17b to be 3.5-fold less active
than 9 as an inhibitor of K562 tumor cell growth. We found,
that the CdC double-bond linking the anthracenone system and
the terminal aryl ring is also critical for activity. Saturation and
hydroxylation of the alkene (19a-c) resulted in a significant
loss of activity, indicating the importance of a more restricted
conformational flexibility in the linker region. This is also
confirmed by strongly decreased activities for C-10-benzylated
analogues of 9, as documented in our our earlier work.15
Figure 2. [3H] Colchicine competition binding assay of 17b and
colchicine. Radiolabeled colchicine or paclitaxel, unlabeled compound,
and biotin-labeled tubulin were incubated together for 2 h at 37 °C.
capacity to interact with the mitotic spindle contributes to the
antiproliferative activity of 17b. A large number of synthetic
and natural compounds with diverse structures have been shown
to bind at the colchicine site, one of the major binding sites on
tubulin, and to inhibit tubulin assembly. The diffraction map
of a close structural analogue of colchicine bound to Rꢀ-tubulin
was reported in 2004.40 On the basis of these important findings,
computer simulation models have recently been employed to
explain empirical structure-activity relationships of colchicine
site inhibitors (CSIs) of tubulin polymerization.41,42 The most
active analogue 17b contains some potential pharmacophore
fragments such as the oxygen atom of the anthracenone carbonyl
group as well as the oxygen atom and the methyl group of the
methoxy substituent. In terms of the common pharmacophore41
and consistent with our experimental data, the methoxy group
in 17b as well as the -hydroxy/-methoxy groups in 9, 10 (Chart
2) are substitutents of choice in the terminal aryl ring and are
critical for activity. As seen with several other CSIs,41 the carbon
atom of the methoxy group in 17b might probably function as
a hydrophobic center. Both the methoxy group and the terminal
aryl ring are essential features for activity.41,42 A phenolic
hydroxy group being flanked by one or two methoxy groups
(9, 10, Chart 2) is another important determinant for activity
within the anthracenone derived CSIs,15,16 with the hydroxy
group possibly hydrogen-bonding to a suitable tubulin residue
in the binding site. It has recently been hypothesized that the
terminal isovanillinyl ring in 9 possibly binds to the same region
of the protein as the pseudoaromatic tropone ring C in
colchicine.42 Interestingly, as documented by our earlier work,
the isovanillinyl partial structure could not be successfully
replaced by trimethoxy (a prominent pharmacophore present
in many CSIs) with retention of high activity. A similar
observation has been made with trimethoxy analogue 17h,
displaying an approximately 7-fold drop in potency in com-
parison with 17b for the inhibition of tubulin polymerization.
According to the published data,41 the trimethoxyphenyl ring
in colchicine, podophyllotoxin, CA-4, and steganacin constitutes
an important hydrophobic center. It has been demonstrated by
X-ray diffraction studies that the trimethoxyphenyl groups for
a close colchicine analogue and for podophyllotoxin are located
in the ꢀ-tubulin structure near the amino acid residue Cys239.40
On the basis of our data, the significantly larger hydrophobic
trimethoxy moiety in 17h relative to the methoxy in 17b is in
general not well tolerated in the terminal aryl ring in the
anthracenone based CSIs. On condition that the terminal
isovanillinyl ring in 9, and thus the methoxy-substituted ring
Conclusion
As an extension of our previously published work, we have
presented synthetic inhibitors of tubulin polymerization based
on a 10-(2-oxo-2-phenylethylidene)-10H-anthracen-9-one mo-
lecular skeleton with 17b being a novel and potent antimitotic
agent. Similar to many other inhibitors of tubulin polymerization,
selected compounds were efficacious in inhibiting tumor cell
proliferation with IC50 values at the submicromolar level. The
best results for inhibition of antiproliferative activity were
obtained with the para-methoxy derivative 17b. In accordance
with our previous findings, the 4-methoxy substitution pattern
in the terminal phenyl ring is again critical for both strong
inhibition of tumor cell proliferation and inhibition of tubulin
polymerization. It is obvious that one of the molecular targets
of the investigated compounds is the tubulin system. Compound
17b is an excellent inhibitor of tubulin polymerization and most
likely interacts with tubulin at the colchicine site. However, with
the insertion of an additional carbonyl group in the linker, we
obtained no substantially greater potency of 17b relative to the
recently described, structurally related 9 for the inhibition of
tubulin polymerization. Both compounds were found to be
nearly equipotent in the [3H]colchicine binding assay. The
induction of G2/M arrest was demonstrated for 17b, being
typical for agents that inhibit tubulin polymerization. Notably,
no growth inhibitory effect was found in cell-cycle arrested cells
(RKOp27). Whereas the effectiveness of numerous clinically
anticancer drugs is limited by the fact that they are substrates
for the efflux pumps Pgp170 and MRP, 17b was found to be
active toward parental tumor cell lines and multidrug-resistant
cell lines. Because of their attractive in vitro antitumor profile,
we believe that compounds of this structural class are attractive
for further structural modifications and that our findings may
provide useful information for the design of novel antitumor
agents. Investigations on the role of the 10-(2-oxo-2-phenyl-
ethylidene)-10H-anthracen-9-ones for antimitotic activity are in

10-(2-oxo-2-Phenylethylidene)-10H-anthracen-9-ones
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 5 1291
1647 cm-1. 1H NMR (CDCl3, 400 MHz, 300 K) δ 8.32 (dd, 1H, J
) 7.82 Hz, J ) 1.18 Hz), 8.23 (dd, 1H, J ) 7.72 Hz, J ) 1.18
Hz), 7.99 (d, 1H, J ) 8.41 Hz), 7.85, 6.82 (d, 2H, J ) 9 Hz),
7.71-7.67 (m, 1H), 7.61-7.56 (m, 2H), 7.43-7.39 (m, 1H),
7.30-7.26 (m, 1H), 7.18 (s, 1H), 3.81 (s, 3H)). MS m/z 340 (100,
M+). Anal. (C23H16O3).
progress, and results from other modifications will be reported
in due course.
Experimental Section
Melting points were determined with a Kofler melting point
apparatus and are uncorrected. Spectra were obtained as follows:
1H NMR spectra were recorded with a Varian Mercury 400 plus
(400 MHz) spectrometer, using tetramethylsilane as an internal
standard. Fourier-transform IR spectra were recorded on a Bio-
Rad laboratories Typ FTS 135 spectrometer, and analysis was
performed with WIN-IR Foundation software. Elemental analyses
were performed at the Mu¨nster Microanalysis Laboratory, using a
Heraeus CHN-O Rapid microanalyzer, and all values were within
(0.4% of the calculated composition. Mass spectra (EI) were
recorded in the EI mode using a MAT GCQ Finnigan instrument.
For HRMS (ESI-nanospray mode) analysis of 17k, a LTQ Orbitrap
XL (Thermo Fisher Scientific, Bremen) spectrometer was used. All
organic solvents were appropriately dried or purified prior to use.
Mixtures were stirred magnetically. Solvents were usually removed
by rotary evaporation under reduced pressure. R-Keto aldehyde
hydrates (phenylglyoxals) were obtained from commercial sources.
Analytical TLC used to monitor reactions was done on Merck silica
60 F254 alumina coated plates (E. Merck, Darmstadt, Germany).
Silica gel column chromatography was performed using Acros
60-200 mesh silica gel.
(10-oxo-10H-Anthracen-9-ylidene) Acetic Acid (14). A stirred
mixture of 10H-anthracen-9-one 13 (10 g, 51.5 mmol) and glyoxylic
acid monohydrate (9.2 g, 100 mmol) in ethanol (150 mL) was
heated under reflux (N2 atmosphere, oil bath, 90 °C). While raising
the temperature, a few drops of piperidine were added at such a
rate that a clear solution was formed. After 4 h (TLC control,
CH2Cl2), the mixture was cooled to room temperature, poured into
H2O (400 mL)/6 M HCl (10 mL), and extracted with CH2Cl2 (5 ×
50 mL). The combined organic extracts were washed with water
and dried over Na2SO4. Then, the drying agent was removed by
filtration and the solvent was evaporated in vacuo to a small volume
(30-40 mL). The carboxylic acid crystallized in an ice bath. The
crystals were filtered off and washed with cold CH2Cl2 (100 mL)
to remove traces of 13 (TLC control, CH2Cl2). Concentration of
the mother liquor gave a small second pale-yellow crop (7.60 g,
59%).
(10-oxo-10H-Anthracen-9-ylidene) Acetic Acid Methyl Ester
(15). The carboxylic acid 14 (2.0 g, 7.99 mmol) was added to a
solution of methanol (100 mL) and conc H2SO4 (1 mL). The mixture
was heated under reflux with stirring for 24 h. After complete
conversion (TLC control, CH2Cl2), the mixture was cooled to room
temperature, poured into water (300 mL), and extracted with CH2Cl2
(3 × 50 mL). The combined organic layers were washed with water,
dried (Na2SO4), and concentrated. The residue was then purified
by silica gel chromatography (CH2Cl2) to afford the pure ester as
a white solid (1.5 g, 71%). 1H NMR (CDCl3, 400 MHz, 300 K) δ
8.27-8.24 (m, 2H), 7.85-7.80 (m, 2H), 7.66-7.62 (m, 1H),
7.60-7.54 (m, 3H), 6.66 (s, 1H), 3.79 (s, 3H).
(10-oxo-10H-Anthracen-9-ylidene) Acetyl Chloride (16). Thio-
nyl chloride (30 mL) was added to 14 (2.0 g, 7.99 mmol), followed
by a catalytic amount of DMF (2 drops). The mixture was heated
to reflux for 2 h and then allowed to cool to room temperature.
Excess thionyl chloride was removed in vacuo. After treating the
residue with n-hexane (2 × 30 mL) and evaporation, the crude
acid chloride was obtained as a yellow solid (2.0 g, 93%) and used
without further purification. 1H NMR (CDCl3, 400 MHz, 300 K) δ
8.26-8.24 (m, 2H), 7.97-7.95 (m, 1H), 7.85-7.83 (m, 1H),
7.69-7.61 (m, 4H), 6.82 (s, 1H).
10-[2-(3-Methoxyphenyl)-2-oxoethylidene]-10H-anthracen-9-
one (17c). 3-Methoxyphenylglyoxal hydrate (0.82 g 5.15 mmol)
was added to a solution of 13 (1.00 g, 5.15 mmol) in 50 mL of
acetic anhydride, and the mixture was heated to reflux. The solution
became orange-red. When the reaction was complete (TLC control,
ethyl acetate/petroleum ether 3/7), the mixture was poured into water
(100 mL), stirred for 20 min, and then extracted with CH2Cl2 (3 ×
50 mL). The combined organic layers were washed with water,
dried over Na2SO4, and the solvent was then evaporated under
reduced pressure. The residue was purified by silica gel chroma-
tography (ethyl acetate/petroleum ether 3/7) to afford 17c as a pale-
yellow powder (0.38 g, 22%). 1H NMR (CDCl3, 400 MHz, 300 K)
δ 8.32 (dd, 1H, J ) 7.83 Hz, J ) 1.17 Hz), 8.23 (dd, 1H, J ) 7.82
Hz, J ) 0.78 Hz), 7.99 (d, 1H, J ) 7.82), 7.72-7.68 (m, 1H),
7.62-7.58 (m, 1H), 7.54 (dd, 1H, J ) 7.83 Hz, J ) 0.78 Hz),
7.46-7.40 (m, 3H), 7.31-7.28 (m, 1H), 7.26-7.24 (m, 1H), 7.20
(s, 1H), 7.07-7.04 (m, 1H), 3.78 (s, 3H).
10-[2-(3,4-Dimethoxyphenyl)-2-oxoethylidene]-10H-anthracen-
9-one (17d). Anhydrous AlCl3 (1.07 g, 8.00 mmol) was added in
one portion to a suspension of crude 16 (2.14 g, 8 mmol) in DCE
(30 mL), and the solution was cooled to 0 °C. After stirring the
dark-red solution for 10 min, a solution of 1,2-dimethoxybenzene
(1.11 g, 8 mmol) in 5 mL DCE was added dropwise. After stirring
for 2 h at 0 °C (TLC control, CH2Cl2), the reaction mixture was
poured into water (300 mL)/6 N HCl (50 mL), stirred for 10 min,
and then extracted with CH2Cl2 (4 × 50 mL). The combined organic
layers were washed with water, dried (Na2SO4), and concentrated.
Thereafter, the residue was purified by silica gel chromatography
(ethyl acetate/petroleum ether 7/3) to afford 17d as yellow needles
1
(0.97 g, 33%). H NMR (CDCl3, 400 MHz, 300 K) δ 8.34-8.32
(m, 1H), 8.25-8.23 (m, 1H), 8.00-7.98 (d, 1H, J ) 8.00 Hz),
7.72-7.68 (m, 1H), 7.62-7.60 (m, 1H), 7.58-7.55 (m, 1H), 7.50
(d, 1H, J ) 2.0 Hz), 7.44-7.39 (m, 2H), 7.31-7.27 (m, 1H), 7.17
(s, 1H), 6.71 (d, 1H, J ) 8.2 Hz), 3.88 (s, 3H), 3.87 (s, 3H)).
10-[2-(3,4-Dihydroxyphenyl)-2-oxoethylidene]-10H-anthracen-
9-one (17e). 10-[2-(3,4-Dimethoxy-phenyl)-2-oxo-ethylidene]-10H-
anthracen-9-one 17d (0.34 g, 1.00 mmol) was dissolved in CH2Cl2
(10 mL). A solution of boron tribromide (5 mL, 5.00 mmol) in
CH2Cl2 (20 mL) was added dropwise with stirring at -78 °C (N2
atmosphere). Stirring was continued overnight while the mixture
was allowed to warm to room temperature. Then the reaction was
terminated by addition of water (300 mL), and the product was
subsequently extracted with ethyl acetate. The extract was dried
with Na2SO4, and the solvent was then removed in vacuo. The
residue was then chromatographed (ethyl acetate/petroleum ether
1
7/3) to afford a yellow powder (0.10 g, 30%). H NMR (DMSO-
d6, 400 MHz, 300 K) δ 10.01 (s, 1H), 9.43 (s, 1H), 8.26 (d, 1H, J
) 7.83 Hz), 8.18 (dd, 1H, J ) 7.83 Hz, J ) 1.17 Hz), 8.14-8.11
(m, 1H), 7.82-7.78 (m, 1H), 7.68-7.64 (m, 1H), 7.65 (s, 1H),
7.54-7.45 (m, 3H), 7.36-7.34 (m, 2H), 6.77 (d, J ) 9.00 Hz).
10-[2-(2,4-Dimethoxyphenyl)-2-oxoethylidene]-10H-anthracen-
9-one (17f). See Supporting Information for details.
10-[2-oxo-2-(2,3,4-Trimethoxyphenyl)ethylidene]-10H-anthra-
cen-9-one (17g). See Supporting Information for details.
10-[2-oxo-2-(3,4,5-Trimethoxyphenyl)ethylidene]-10H-anthra-
cen-9-one (17h). See Supporting Information for details.
10-[2-(2-Hydroxy-3,4-dimethoxyphenyl)-2-oxoethylidene]-
10H-anthracen-9-one (17i). AlCl3 was added in one portion to a
solution of 10-[2,3,4-trimethoxy-phenyl)-2-oxo-ethylidene]-9(10H)-
anthracenone (17h, 0.40 g, 1.00 mmol) in DCE (25 mL) at 0 °C.
When the reaction was completed (TLC control), the mixture was
poured into water (400 mL)/6 N HCl (50 mL) and extracted with
CH2Cl2 (3 × 30 mL). The combined organic layers were washed
with water, dried over Na2SO4, and the solvent was then evaporated
under reduced pressure. The residue was purified by silica gel
10-(2-oxo-2-Phenylethylidene)-10H-anthracen-9-one (17a).
See Supporting Information for details.
10-[2-(4-Methoxyphenyl)-2-oxoethylidene]-10H-anthracen-9-
one (17b). The title compound was prepared from 16 (2.14 g, 8.00
mmol) and methoxybenzene (0.87 mL, 8.00 mmol) in a similar
manner as described for the preparation of 17d.
Purification by silica gel chromatography (CH2Cl2) afforded 17b
as a pale-yellow powder (0.82 g, 30%): mp 120-122 °C; FTIR

1292 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 5
Prinz et al.
chromatography (ethyl acetate/petroleum ether 1:1) to afford 17i
as a pale-yellow powder (0.17 g, 42%). 1H NMR (CDCl3) δ 12.30
(s, 1H), 8.34-8.32 (m, 1H), 8.27-8.24 (m, 1H), 7.98 (d, 1H, J )
7.82 Hz), 7.73-7.68 (m, 1H), 7.63-7.59 (m, 1H), 7.49-7.45 (m,
1H); 7.40-7.37 (m, 1H), 7.26 (d, 1H, J ) 9.00 Hz), 7.14 (s, 1H),
6.30 (d, 1H, J ) 9.00 Hz), 3.92 (s, 3H), 3.87 (s, 3H).
medium (Gibco) containing 10% fetal bovine serum (FBS, Bio-
chrom KG), streptomycin (0.1 mg/mL), penicillin G (100 units/
mL), and L-glutamine (30 mg/L) at 37 °C in 5% CO2. Cell line
SF268 (central nerval system carcinoma) were obtained from NCI.
The human tumor cell lines KB/HeLa (cervical carcinoma), SKOV3
(ovarian adenocarcinoma), NCI-H460, and L1210 (murine leuke-
mia) were obtained from ATCC. The RKO human colon adeno-
carcinoma cells containing an ecdysone-inducible expression vector
of p27kip1 were described recently.43 The L1210VCR cell line
(expression of MDR-1, MRP-1 negative) was described recently.44
Rat LT12 cells, the LT12/MDR subline (ectopic expression of
human MDR-1 protein) as well as P388 cells and the P388/Adr
subline (selected with doxorubicine, elevated levels of MDR-1
protein) were provided by Dr. Nooter (University Hospital, Rot-
terdam, The Netherlands).
10-[2-oxo-2-p-tolyl-ethylidene]-10H-anthracen-9-one (17j). See
Supporting Information for details.
10-[2-(4-Hydroxy-phenyl)-2-oxo-ethylidene]-10H-anthracen-
9-one (17k). See Supporting Information for details.
10-[2-(4-Methoxythiophen-2-yl)-2-oxoethylidene]-10H-anthra-
cen-9-one (17l). See Supporting Information for details.
10-[(2-oxo-2-Thiophen-2-yl-ethylidene)]-10H-anthracen-9-
one (17m). See Supporting Information for details.
[(10-oxo-(10H)-anthracen-9-ylidene)]-acetic Acid 2-Methoxy-
phenyl Ester (18a). See Supporting Information for details.
[(10-oxo-(10H)-Anthracen-9-ylidene)]-aceticAcid2,6-Dimethoxy-
phenyl Ester (18b). See Supporting Information for details.
10-(1-Hydroxy-2-oxo-2-phenylethyl)-10H-anthracen-9-one (19a).
A stirred mixture of 13 (1.0 g, 5.15 mmol) and phenylglyoxal
hydrate (0.69 g, 5.15 mmol) in ethanol (50 mL) was heated under
reflux (N2 atmosphere, oil bath, 90 °C). While raising the temper-
ature, a few drops of piperidine were added at such a rate that a
clear solution was formed. After 4 h (TLC control), the mixture
was cooled to room temperature, and the solvent was evaporated
in vacuo. Purification of the residue by silica gel chromatography
(ethyl acetate/petroleum ether 7/3) afforded 19a as a pale-yellow
powder (0.30 g, 18%); mp 164-165 °C; FTIR 1659 cm-1. 1H NMR
(CDCl3, 400 MHz, 300 K) δ 8.20-8.18 (m, 1H), 8.07-8.05 (m,
1H), 7.72-7.70 (m, 2H), 7.66-7.65 (m, 1H), 7.62-7.58 (m, 2H),
7.48-7.42 (m, 3H), 7.40-7.36 (m, 1H), 7.23-7.22 (m, 1H),
7.01-6.99 (m, 1H), 5.42 (dd, 1H, J ) 7.04 Hz, J ) 2.35 Hz), 4.72
(d, 1H, J ) 2.34 Hz), 3.54 (d, 1H, J ) 7.04 Hz). Anal. (C22H16O3).
10-(1-Hydroxy-2-(4-methoxyphenyl)-2-oxoethyl)-10H-anthra-
cen-9-one (19b). See Supporting Information for details.
Assay of Cell Growth. First, K562 cells were plated at 2 × 105
cells/mL in 48 well dishes (Costar, Cambridge, MA). Then untreated
control wells were assigned a value of 100%. Drugs were made
soluble in DMSO/methanol 1:1, and control wells received equal
volumes (0.5%) of vehicle alone. Drugs were dissolved in methanol/
DMSO 1:1. To each well were added 5 µL of drug and the final
volume in the well was 500 µL. Cell numbers were counted with
a Neubauer counting chamber (improved, double grid) after
treatment with chemicals for 48 h. Each assay condition was
prepared in triplicate, and the experiments were carried out three
times. IC50 values were obtained by nonlinear regression (GraphPad
Prism) and represent the concentration at which cell growth was
inhibited by 50%. The adjusted cell number was calculated as a
percentage of the control, which was the number of cells in wells
without the addition of compound.
XTT Assay. The XTT assay was used to determine proliferation
by quantification of cellular metabolic activity.29 IC50 values were
obtained by nonlinear regression (GraphPad Prism). The various
tumor cell lines were seeded in microtiter plates in a density of 1
× 103-8 × 104 cells/well in 125 µL for logarithmic cell growth
and were incubated with different concentrations of cytotoxic agents
for 48 h.
10-(1-Hydroxy-2-(3,4,5-trimethoxyphenyl)-2-oxoethyl)-10H-
anthracen-9-one (19c). See Supporting Information for details.
2-[(10-Hydroxy-10H-anthracen-9-ylidene)]-1-(4-methoxyphe-
nyl)-ethanone (20). Fine powdered sodium borohydride (0.23 g,
6.00 mmol) was added in one portion to a suspension of 17b (0.34
g, 1 mmol) in a mixture of THF (20 mL) and ethanol (5 mL) at
room temperature. When the reaction was completed (TLC control,
CH2Cl2), water (3 mL) was added and the solution was then
extracted with CH2Cl2 (50 mL). The extract was dried with Na2SO4,
and the solvent was then removed in vacuo. The residue was
chromatographed on a column of SiO2 (ethyl acetate/petroleum ether
7/3) to afford 20 as a yellow powder (98 mg, 29%); mp 170 °C;
In RKOp27Kip1 cells, expression of p27Kip1 was induced by 3
µM ponasterone A in 24 h, leading to an arrest of these cells in the
G1 phase of the division cycle. Cell cycle specific substances, such
as tubulin inhibitors, were only cytotoxic if the cells were not
arrested and the cell cycle was in progress. The assay was performed
in 96-well plates. The cell count of induced cells was about three
times higher than that of noninduced cells. RKO cells with/without
p27Kip1 expression (2 × 104 cells/well induced; 6 × 103 cells/well
not induced, in 125 µL) were treated with the test compound for
48 h at 37 °C. The controls were untreated cells (( induction). On
day 1, the cells were plated (( ponasterone A) and incubated at
37 °C for 24 h. On day 2, the test substance was added (control
DMSO) and incubation at 37 °C was continued for another 48 h.
Thereafter, a standard XTT assay was carried out.
Flow Cytometric Analysis of Cell-Cycle Status. For a con-
centration-dependent cell cycle analysis, subconfluent KB/HeLa
cells were exposed to test compounds for 24 h at 37 °C, detached,
and collected. After fixation with 70% ethanol, the DNA was
simultaneously stained by propidium iodide and digested with
RNase. The DNA content of cells was determined with a FACS
CaliburTM cytometer (Beckton Dickinson, Heidelberg, Ger-
many). The number of cells in G2/M phase was calculated by
cell cycle analysis software (Mod Fit LT; VERITY). IC50 values
were calculated by nonlinear regression (GraphPad Prism).
Isolation of MTP and In Vitro MT Assembly Assay by
Turbidimetric Measurement. Microtubule protein (MTP) con-
sisting of 80-90% tubulin and 10-20% microtubule associated
proteins (MAPs) was isolated from porcine brain by two cycles
of temperature-dependent disassembly (0 °C)/reassembly (37 °C)
according to the method described by Shelanski et al.45
Throughout the preparation, a buffer containing 20 mM PIPES
(1,4-piperazine diethane sulfonic acid, pH 6.8), 80 mM NaCl,
0.5 mM MgCl2, 1 mM EGTA (ethylene bis(oxyethylenenitrilo)
tetraacetic acid), and 1 mM DTT was used. To increase the yield,
1
FTIR 3421, 1638, cm-1. H NMR (CDCl3, 400 MHz, 300 K) δ
7.95, 6.84 (d, 2H, J ) 5.08 Hz), 7.81-7.79 (m, 1H), 7.72-7.70
(m, 1H), 7.62 (d, 1H, J ) 7.44 Hz), 7.45-7.41 (m, 3H), 7.29-7.26
(m, 1H), 7.09-7.05 (m, 1H), 6.95 (s, 1H), 5.66 (s, 1H), 3.82 (s,
3H). MS m/z 342 (14.49, M+), 325 (100). Anal. (C23H18O3).
2-(Anthracen-9-yloxy)-1-(4-methoxyphenyl)-ethanone (22). 9-
(10H)-Anthracenone (13, 1 g, 5.15 mmol) and dry K2CO3 (7.00 g,
50.75 mmol) were suspended in absolute acetone (50 mL) under
N2. 2-Bromo-4′-methoxyacetophenone (1.19 g, 5.2 mmol) was
added, and the mixture was refluxed under nitrogen until the
reaction was completed (TLC control, CH2Cl2). The reaction
mixture was then cooled and poured into water (400 mL), acidified
with 6 N HCl, and extracted with CH2Cl2 (3 × 30 mL). The
combined CH2Cl2 extracts were washed, dried over Na2SO4, and
then evaporated. Purification by silica gel chromatography (CH2Cl2/
hexane 8/2) afforded 22 as a pale-yellow powder (0.25 g, 14%).
1H NMR (CDCl3, 400 MHz, 300 K) δ 8.35-8.33 (s, 2H), 8.28 (s,
1H), 8.03-8.00 (s, 2H), 7.96 (d, 2H, J ) 9.0 Hz), 7.49-7.47 (m,
4H), 6.95 (d, 2H, J ) 9.0 Hz), 5.48 (s, 2H), 3.87 (s, 3H);
10-(1-Hydroxy-2,2-dimethoxy-ethyl)-10H-anthracen-9-one (23).
See Supporting Information for details.
Biological Assay Methods. Cells and Culture Conditions.
Human chronic myelogenous K562 leukemia cells were obtained
from DSMZ, Braunschweig, Germany, and cultured in RPMI 1640

10-(2-oxo-2-Phenylethylidene)-10H-anthracen-9-ones
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 5 1293
(9) Yoshimatsu, K.; Yamaguchi, A.; Yoshino, H.; Koyanagi, N.; Kitoh,
K. Mechanism of action of E7010, an orally active sulfonamide
antitumor agent: inhibition of mitosis by binding to the colchicine
site of tubulin. Cancer. Res. 1997, 57, 3208–3213.
(10) Sirisoma, N.; Kasibhatla, S.; Pervin, A.; Zhang, H.; Jiang, S.;
Willardsen, J. A.; Anderson, M. B.; Baichwal, V.; Mather, G. G.;
Jessing, K.; Hussain, R.; Hoang, K.; Pleiman, C. M.; Tseng, B.; Drewe,
J.; Cai, S. X. Discovery of 2-chloro-N-(4-methoxyphenyl)-N-meth-
ylquinazolin-4-amine (EP128265, MPI-0441138) as a potent inducer
of apoptosis with high in vivo activity. J. Med. Chem. 2008, 51, 4771–
4779.
(11) Baasner, S.; Emig, P.; Gerlach, M.; Mu¨ller, G.; Paulini, K.; Schmidt,
P.; Burger, A. M.; Fiebig, H.-H.; Gu¨nther, E. G. D-82318sA novel,
synthetic, low molecular weight tubulin inhibitor with potent in vivo
antitumor activity. EORTC-NCI-AACR Meeting, Frankfurt, Germany,
2002, Poster 112.
(12) Meng, F.; Cai, X.; Duan, J.; Matteucci, M. G.; Hart, C. P. A novel
class of tubulin inhibitors that exhibit potent antiproliferation and in
vitro vessel-disrupting activity. Cancer Chemother. Pharmacol. 2008,
61, 953–963.
(13) Zhang, L. H.; Wu, L.; Raymon, H. K.; Chen, R. S.; Corral, L.; Shirley,
M. A.; Narla, R. K.; Gamez, J.; Muller, G. W.; Stirling, D. I.; Bartlett,
J. B.; Schafer, P. H.; Payvandi, F. The synthetic compound CC-5079
is a potent inhibitor of tubulin polymerization and tumor necrosis
factor-alpha production with antitumor activity. Cancer. Res. 2006,
66, 951–959.
(14) Hadfield, J. A.; Ducki, S.; Hirst, N.; McGown, A. T. Tubulin and
microtubules as targets for anticancer drugs. Prog. Cell Cycle Res.
2003, 5, 309–25.
(15) Prinz, H.; Ishii, Y.; Hirano, T.; Stoiber, T.; Camacho Gomez, J. A.;
Schmidt, P.; Du¨ssmann, H.; Burger, A. M.; Prehn, J. H.; Gu¨nther,
E. G.; Unger, E.; Umezawa, K. Novel benzylidene-9(10H)-anthra-
cenones as highly active antimicrotubule agents. Synthesis, antipro-
liferative activity, and inhibition of tubulin polymerization. J. Med.
Chem. 2003, 46, 3382–3394.
(16) Zuse, A.; Schmidt, P.; Baasner, S.; Bo¨hm, K. J.; Mu¨ller, K.; Gerlach,
M.; Gu¨nther, E. G.; Unger, E.; Prinz, H. 9-Benzylidene-naphtho[2,3-
b]thiophen-4-ones as novel antimicrotubule agents-synthesis, antipro-
liferative activity, and inhibition of tubulin polymerization. J. Med.
Chem. 2006, 49, 7816–7825.
the reassembly steps were performed in the presence of
glycerol.46 The protein concentration was determined by the
Lowry procedure using bovine serum albumin as a standard.
Satisfactory purity of the MTP was reached after the second
cycle of assembly and disassembly. Microtubules to be used for
determining antimicrotubule effector activities were prepared by
in vitro self-assembly of MTP in presence of GTP (guanosin-
5′-triphosphate) and magnesium ions. The MTP concentration
for assembly was 1.2 mg/mL (determined by the Lowry
procedure using bovine serum albumin as a standard). The
assembly measurements were done directly in cuvettes. To start
microtubule formation, the stock MTP solutions were diluted
with the preparation buffer to 1.2 mg/mL, GTP was added to
0.6 mM (final concentration), and the samples were transferred
into spectrophotometer (Cary 4E, Varian Inc.), equipped with a
temperature-controlled multichannel cuvette holder, adjusted to
37 °C. Turbidity was recorded over 30 min at 360 nm.
The tubulin effectors were added from stock solutions in DMSO.
The final DMSO concentration was 1%. Control measurements were
made with DMSO, only. To quantify the drug activity, the turbidity
signal after 30 min (plateau level, representing the assembly/
disassembly steady state) was compared with that of the control
samples. The IC50 value is defined as the drug concentration that
causes a 50% inhibition in relation to the assembly level without
the drug.
3H Colchicine Competition-Binding Assay.37 3H-Colchicine
was diluted and biotin-labeled tubulin (T333, Cytoskeleton, Denver,
CO) was reconstituted according to the manufacturers protocol. The
diluted compounds and the 3H-colchicine were transferred to a 96-
well isoplate (PE-Wallac, Boston, MA), and buffer and the
reconstituted biotin-labeled tubulin were added. After incubation,
streptavidin-coated yttrium SPA beads (Amersham Pharmacia
Biotech, Piscataway, NJ) were added and the bound radioactivity
was determined using a MicroBeta Trilux Microplate scintillation
counter (PE-Wallac Boston, MA). IC50 values were obtained by
nonlinear regression (GraphPad Prism).
(17) Zuse, A.; Schmidt, P.; Baasner, S.; Bo¨hm, K. J.; Mu¨ller, K.; Gerlach,
M.; Gu¨nther, E. G.; Unger, E.; Prinz, H. Sulfonate derivatives of
naphtho[2,3-b]thiophen-4(9H)-one and 9(10H)-anthracenone as highly
active antimicrotubule agents. Synthesis, antiproliferative activity, and
inhibition of tubulin polymerization. J. Med. Chem. 2007, 50, 6059–
6066.
(18) Lawrence, N. J.; McGown, A. T. The chemistry and biology of
antimitotic chalcones and related enone systems. Curr. Pharm. Des.
2005, 11, 1679–1693.
Acknowledgment. We thank Cornelia Lang, Britta Kroen-
ert, Sonja Seidel, Katrin Zimmermann, Magnus Bro¨cker,
Marcel Overmann, and Angelika Zinner for excellent techni-
cal assistance.
Supporting Information Available: IR, 1H NMR, and MS data
of new compounds 17a-17m, 18a-18b, 19a-19c, 20, 22, and
23, cell cycle analysis data of KB/HeLa cells treated with 17b and
reference compounds, table of elemental analysis results of all target
compounds. This material is available free of charge via the Internet
at http://pubs.acs.org.
(19) Liou, J. P.; Chang, J. Y.; Chang, C. W.; Chang, C. Y.; Mahindroo,
N.; Kuo, F. M.; Hsieh, H. P. Synthesis and structure-activity
relationships of 3-aminobenzophenones as antimitotic agents. J. Med.
Chem. 2004, 47, 2897–2905.
(20) Afsah, E. M.; Etman, H. A.; Hamama, W. S.; Sayed-Ahmed, A. F. A
study on the reaction of 1,3-indandione with Schiff bases: synthesis
of new 1,3-indandiones with expected psychopharmacological and
anticoagulant activit. Boll. Chim. Farm. 1998, 137, 244–248.
(21) El-Shafie, S. M. M. Synthesis of 10-arylidene-9-anthrones and related
compounds. Indian J. Chem. 1978, 16B, 828–830.
References
(22) Bergmann, E. D.; Rabinovitz, M.; Glily, S. A ring enlargement in the
9,10-dihydroanthracene series. Tetrahedron Suppl. 1966, 8, 141–148.
(1) Honore, S.; Pasquier, E.; Braguer, D. Understanding microtubule
dynamics for improved cancer therapy. Cell. Mol. Life Sci. 2005, 62,
3039–3056.
¨
(23) Weitz, E. Uber einige Anthronabko¨mmlinge. Liebigs Ann. Chem. 1919,
418, 29–35.
(2) Pellegrini, F.; Budman, D. R. Review: tubulin function, action of
antitubulin drugs, and new drug development. Cancer InVest. 2005,
23, 264–273.
(24) Rappoport, Z.; Apeloig, Y.; Greenblatt, J. Vinylic cations from
solvolysis. 29. Solvolysis of 9-(R-bromoarylidene)anthrones as a probe
to the reactivity-selectivity relationship in solvolysis reactions. J. Am.
Chem. Soc. 1980, 102, 3837–3848.
(3) Hearn, B. R.; Shaw, S. J.; Myles, D. C. Microtubule targeting agents.
(25) Mu¨ller, K.; Gu¨rster, D.; Piwek, S.; Wiegrebe, W. Antipsoriatic
anthrones with modulated redox properties. 1. Novel 10-substituted
1,8-dihydroxy-9(10H)-anthracenones as inhibitors of 5-lipoxygenase.
J. Med. Chem. 1993, 36, 4099–4107.
Compr. Med. Chem. II 2007, 7, 81–110.
(4) Jordan, M. A. Mechanism of action of antitumor drugs that interact
with microtubules and tubulin. Curr. Med. Chem. Anticancer Agents
2002, 2, 1–17.
(5) Li, Q.; Sham, H. L. Discovery and development of antimitotic agents
that inhibit tubulin polymerisation for the treatment of cancer. Expert
Opin. Ther. Patents 2002, 12, 1663–1702.
(6) Mahindroo, N.; Liou, J. P.; Chang, J. Y.; Hsieh, H. P. Antitubulin
agents for the treatment of cancersa medicinal chemistry update.
Expert Opin. Ther. Patents 2006, 16, 647–691.
(7) Pettit, G. R.; Singh, S. B.; Hamel, E.; Lin, C. M.; Alberts, D. S.; Garcia-
Kendall, D. Isolation and structure of the strong cell growth and tubulin
inhibitor combretastatin A-4. Experientia 1989, 45, 209–211.
(26) Baba, H.; Takemura, T. Spectrophotometric investigations of the
tautomeric reaction between anthrone and anthranol. I. The keto-enol
equilibrium. Tetrahedron 1968, 24, 4779–4791.
´
(27) Martin, R. Etudes sur la re´action de fries. XIV. Me´canismes engendre´s
dans les re´actions acido-catalyse´es. IV. Pre´paration des acyl-5 gaiacols.
Bull. Soc. Chim. Fr. 1977, 901–905.
(28) Lozzio, C. B.; Lozzio, B. B. Human chronic myelogenous leukemia
cell-line with positive philadelphia chromosome. Blood 1975, 45, 321–
334.
(29) Scudiero, D. A.; Shoemaker, R. H.; Paull, K. D.; Monks, A.; Tierney,
S.; Nofziger, T. H.; Currens, M. J.; Seniff, D.; Boyd, M. R. Evaluation
of a soluble tetrazolium/formazane assay for cell growth and drug
(8) Goodin, S.; Kane, M. P.; Rubin, E. H. Epothilones: mechanism of
action and biologic activity. J. Clin. Oncol. 2004, 22, 2015–2025.

1294 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 5
Prinz et al.
sensitivity in culture using human and other tumor cell lines. Cancer
Res. 1988, 48, 4827–4833.
(30) Schmidt, M.; Lu, Y.; Parant, J. M.; Lozano, G.; Bacher, G.; Beckers,
T.; Fan, Z. Differential roles of p21Waf1 and p27Kip1 in modulating
chemosensitivity and their possible application in drug discovery
studies. Mol. Pharmacol. 2001, 60, 900–906.
(31) Dumontet, C.; Jaffrezou, J. P.; Tsuchiya, E.; Duran, G. E.; Chen, G.;
Derry, W. B.; Wilson, L.; Jordan, M. A.; Sikic, B. I. Resistance to
microtubule-targeted cytotoxins in a K562 leukemia cell variant
associated with altered tubulin expression and polymerization. Bull.
Cancer 2004, 91, 81–112.
(32) Dumontet, C.; Sikic, B. I. Mechanism of action of and resistance to
antitubulin agents: microtubule dynamics, drug transport, and cell
death. J. Clin. Oncol. 1999, 3, 1061–1070.
(33) Fardel, O.; Lecureur, V.; Guillouzo, A. The P-glycoprotein multidrug
transporter. Gen. Pharmacol. 1996, 27, 1283–1291.
(34) Cole, S. P.; Deeley, R. G. Multidrug resistance mediated by the ATP-
binding cassette transporter protein MRP. BioEssays 1998, 20, 931–
940.
(35) Kavallaris, M.; Annereau, J. P.; Barret, J. M. Potential mechanisms
of resistance to microtubule inhibitors. Semin. Oncol. 2008, 35, 22–
27.
(36) Verrills, N. M.; Kavallaris, M. Improving the targeting of tubulin-
binding agents: lessons from drug resistance studies. Curr. Pharm.
Des. 2005, 11, 1719–1733.
(39) Bai, R. L.; Pettit, G. R.; Hamel, E. Binding of dolastatin 10 to tubulin
at a distinct site for peptide antimitotic agents near the exchangeable
nucleotide and vinca alkaloid sites. J. Biol. Chem. 1990, 265, 17141–
17149.
(40) Ravelli, R. B.; Gigant, B.; Curmi, P. A.; Jourdain, I.; Lachkar, S.;
Sobel, A.; Knossow, M. Insight into tubulin regulation from a complex
with colchicine and a stathmin-like domain. Nature 2004, 428, 198–
202.
(41) Nguyen, T. L.; McGrath, C.; Hermone, A. R.; Burnett, J. C.;
Zaharevitz, D. W.; Day, B. W.; Wipf, P.; Hamel, E.; Gussio, R. A
common pharmacophore for a diverse set of colchicine site inhibitors
using a structure-based approach. J. Med. Chem. 2005, 48, 6107–
6116.
(42) Zefirova, O. N.; Diikov, A. G.; Zyk, N. V.; Zefirov, N. S. Ligands of
the colchicine site of tubulin: a common pharmacophore and new
structural classes. Russ. Chem. Bull. Int. Ed. 2007, 56, 680–688.
(43) Schmidt, M.; Lu, Y.; Liu, B.; Fang, M.; Mendelsohn, J.; Fan, Z.
Differential modulation of paclitaxel-mediated apoptosis by p21waf1
and p27kip1. Oncogene 2000, 19, 2423–2429.
(44) Bacher, G.; Nickel, B.; Emig, P.; Vanhoefer, U.; Seeber, S.; Shandra,
A.; Klenner, T.; Beckers, T. D-24851, a novel synthetic microtubule
inhibitor, exerts curative antitumoral activity in vivo, shows efficiacy
toward multidrug-resistant tumor cells, and lacks neurotoxicity. Cancer
Res. 2001, 61, 392–399.
(45) Shelanski, M. L.; Gaskin, F.; Cantor, C. R. Microtubule assembly in
the absence of added nucleotides. Proc. Nat. Acad. Sci. U.S.A. 1973,
70, 765–768.
(37) Tahir, S. K.; Kovar, P.; Rosenberg, S.; Ng, S.-C. A rapid colchicine
competition-binding scintillation proximity assay using biotin-labeled
colchicine. Biotechniques 2000, 29, 156–160.
(38) Bai, R.; Roach, M. C.; Jayaram, S. K.; Barkoczy, J.; Pettit, G. R.;
Luduena, R. F.; Hamel, E. Differential effects of active isomers,
segments, and analogs of dolastatin 10 on ligand interactions with
tubulin. Correlation with cytotoxicity. Biochem. Pharmacol. 1993, 45,
1503–1515.
(46) Vater, W.; Bo¨hm, K. J.; Unger, E. A simple method to obtain brain
microtubule protein poor in microtubule-associated proteins. Acta
Histochem. Suppl. 1986, 33, 123–129.
JM801338R