Mitotic arrest induced by a novel family of DNA topoisomerase II inhibitors

Article abstract of DOI:10.1021/jm100155y

A small structure-focused library of propargylic enol ethers was prepared by means of a modular and efficient chemodifferentiating organocatalyzed multicomponent reaction. The most active compound (GI₅₀ 0.25 μM) against solid tumor cells was selected as lead. Cell cycle analysis and immunoblotting demonstrated arrest at the metaphase, pointing out human topoisomerase II as plausible molecular target. In vitro assays were carried out, showing that the lead behaves as a catalytic inhibitor of the enzyme.

Full text of DOI:10.1021/jm100155y

J. Med. Chem. 2010, 53, 3835–3839 3835
DOI: 10.1021/jm100155y
Mitotic Arrest Induced by a Novel Family of DNA Topoisomerase II Inhibitors
†,
†
§
§
ꢀ
Fernando Garcı
Leticia G. Leon, Carla Rı
os-Luci,† David Tejedor,‡, Eduardo Perez-Roth, Juan C. Montero, Atanasio Pandiella,
ꢀ
´
,†,
ꢀ
ꢀ
´
a-Tellado,‡, and Jose M. Padron*
†
´
BioLab, Instituto Universitario de Bio-Orgaꢀnica “Antonio Gonzaꢀlez” (IUBO-AG), Universidad de La Laguna, C/Astrofısico Francisco
´
‡
´
Saꢀnchez 2, 38206 La Laguna, Spain, Instituto de Productos Naturales y Agrobiologıa, Consejo Superior de Investigaciones Cientıficas (CSIC),
§
C/Astrofısico Francisco Sanchez 3, 38206 La Laguna, Spain, Centro de Investigacion del Cancer, IBMCC/CSIC;Universidad de Salamanca,
´
Salamanca, Spain, and Instituto Canario de Investigacioꢀn del Caꢀncer (ICIC), http://www.icic.es
ꢀ
ꢀ
ꢀ
Received October 27, 2009
A small structure-focused library of propargylic enol ethers was prepared by means of a modular and
efficient chemodifferentiating organocatalyzed multicomponent reaction. The most active compound
(GI50 0.25 μM) against solid tumor cells was selected as lead. Cell cycle analysis and immunoblotting
demonstrated arrest at the metaphase, pointing out human topoisomerase II as plausible molecular
target. In vitro assays were carried out, showing that the lead behaves as a catalytic inhibitor of
the enzyme.
Scheme 1. Derivatization of Propargylic Enol Ether 1a
Introduction
Cancer is a genetic disease characterized by uncontrolled
accumulation of tumor cells causedby increasedproliferation,
decreased cell death, or both. Cell cycle is the major regulatory
mechanism of cell proliferation, which is controlled by several
proteins that represent attractive drug targets for therapeutic
intervention.1 Targeting one of these proteins, TOP2,a has
demonstrated clinical benefit for patients.2 TOP2 is a nuclear
enzyme essential for cell division and possesses a wide array of
biological functions mainly related with topological problems
that arise in double-stranded DNA.3 In particular, TOP2 is
required for metaphase-anaphase transition,4 and it has been
proposed as a metaphase checkpoint.5
a Reagents: (a) Et3N (cat.), 80%; (b) TFA, 72%; (c) methyl propio-
Within our program directed at the discovery of new
antitumor agents, we have developed a fast and modular
methodology for the synthesis of novel densely functionalized
molecules incorporating three chemically differentiated con-
stitutive blocks in an ordered and defined manner.6 Hence, we
prepared a small and structure-biased library of propargylic
enol ethers to search for a lead. Herein we report on our
preliminary results on this searching, which successfully
afforded an unprecedented class of TOP2 inhibitors.
late, Et3N, 98%; (d) (i) LiOH H2O, THF:H2O, 0 °C, 30 min; (ii) HCl
3
(aq), 92%; (e) NaOH, H2O, 88% (two steps). See Supporting Informa-
tion for specific identity of compounds 1a-y.
conjugated alkynes (Scheme 1).7 Compounds 1a-u were
obtained by the direct reaction of the appropriate aldehyde
with the corresponding acetylide in average yield over 80%.
In addition to propiolates, the method tolerates ethynyl
p-tolyl sulfone. The enol ester group of 1 was exchanged in
a two-step sequence to create a new functional-diversity
point on the molecule (dissimilar ester groups). Derivatives
1v-y were synthesized from the available propargylic alco-
hols 2 and were designed for structure-activity relationship
studies. Product 1 was selectively hydrolyzed with LiOH
to the propargylic acid 3, which in turn was transformed
into the sodium carboxylate 4, providing an additional
functional-diversity point on the molecule.
Cytotoxicity. In a preliminary screening, the antiprolifera-
tive activity of the library was determined in the human solid
tumor cell lines SW1573 (nonsmall cell lung) and HBL-100
(breast). Analysis of the GI50 values allowed us to obtain
some information on the lipophilic requirements involved
in tumor cell growth inhibition providing the following
structure-activity relationships (Figure 1). An ester or a
Results and Discussion
Chemistry. The library was built in parallel using an
efficient manifold based on triethylamine catalyzed chemo-
differentiating ABB0 3CRs involving aldehydes and terminal
*To whom correspondence should be addressed. Phone: þ34 922 316
502 ext. 6126. Fax: þ34 922 318 571. E-mail: jmpadron@ull.es. Address:
ꢀ
ꢀ
Instituto Universitario de Bio-Organica “Antonio Gonzalez” Universi-
ꢀ
´
dad de La Laguna C/Astrofısico Francisco Sanchez 2, 38206 La Laguna,
Spain
a Abbreviations: TOP2, human DNA topoisomerase II; GI50, 50%
growth inhibition; SAR, structure-activity relationship; PARP, poly-
(ADP-ribose) polymerase-1; Chk2, checkpoint kinase 2; CI, catalytic
inhibitor; kDNA, intact kinetoplast DNA; Nck, nicked; Lin, linearized;
SC, supercoiled; Rel, relaxed.
r
2010 American Chemical Society
Published on Web 04/20/2010
pubs.acs.org/jmc

3836 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9
Leoꢀn et al.
G2/M Transition. We next looked at protein expression at
the G2/M phase transition to discern between G2 arrest and
M arrest. The G2 cell cycle arrest is thought to function as a
defense mechanism (the so-called G2 checkpoint) in response
to DNA damage, which allows cells to repair DNA before
progressing into mitosis.8 To determine whether compound
1 was able to induce DNA damage to cells, we investigated
two key proteins involved in the response to DNA damage:
Chk2 and histone H2AX. Upon DNA damage, Chk2 and
H2AX are phosphorylated. In cells treated with 1, there was
no evidence of phosphorylation of neither Chk2 nor H2AX
(Figure 3a), denoting the absence of harm to DNA. These
results show that the G2 checkpoint is not activated by the
exposure of cells to 1, allowing cells to enter mitosis. This
point was further confirmed with the expression of cyclin B1,
which is required for mitotic initiation. Immunoblotting of
HeLa cells treated with 1 showed that compound did not
affect cyclin B1 levels (Figure 3b), thus indicating cell cycle
progression into mitosis.
Time-dependent phosphorylation of the biological mar-
kers histone H3 and BubR1 also support that treated cells
entered mitosis. In mammalian cells, phosphorylation of
histone H3 on Ser10 (pS10-H3) starts in late G2 phase, it is
completed in late prophase, and it is maintained through
metaphase. Dephosphorylation of pS10-H3 begins in ana-
phase and ends at early telophase.9 Thus, high levels of pS10-
H3 are only observed during mitosis.10 Exposure of HeLa
cells to 1 increased the levels of pS10-H3 in a time dependent
manner (Figure 3b), as indicated by Western blotting ana-
lyses of cell extracts with an antibody that specifically
recognizes the phosphorylated form. This result indicates
that treated cells were not able to progress into anaphase.
Further evidence of this result was derived from the study
of BubR1 activation. The activation of BubR1 (pBubR1)
by phosphorylation is crucial for mitotic timing. pBubR1
inhibits anaphase promoting complex/cyclosome11 and is
concomitant with a delayed progression into anaphase.
Consistently with our previous findings for pS10-H3, we
observed that treatment of HeLa cells with 1 increased the
amount of pBubR1, as indicated by phosphorylation-in-
duced retardation of the electrophoretic mobility of this
protein. Altogether, immunobloting results show that 1 does
not induce damage to DNA, allowing cells to enter mitosis
until the cell cycle is arrested in metaphase.
Figure 1. Summary of SAR.
Table 1. Antiproliferative Activity (GI50) of Lead 1 (Z = CO2Me,
R = nBu)a
GI50
solid tumor cell lines
HBL-100
HeLa
0.25 ( 0.10
2.8 ( 0.24
0.24 ( 0.07
3.8 ( 1.3
3.1 ( 0.9
1.9 ( 0.6
1.3 ( 0.3
34 ( 1.3
SW1573
T-47D
WiDr
nontumor cell lines
Hs27
hTERT-HPNE
MCF-10A
a Values in μM ( standard deviation. Means of three to five experi-
ments. See Supporting Information for specific cytotoxicity data of
compounds 1a-y.
tosyl group at the alkyne end (Z) is essential for the activity.
In this particular context, the bulkiness of the alkoxy group is
inversely related to the biological effect. Thus, the antipro-
liferative activities are in the order Me>Et>tBu>menthyl.
Additional groups like hydrogen (1v-w), carboxylic acid (3),
or carboxylate (4) produce a loss in either activity or potency.
Regarding the alkyl side chain (R), compounds with linear
substituents are more active than the corresponding acyclic
branched analogues, and these are followed by the cyclic side
chains, which are less active. Another important observation
is obtained from the length of the alkyl side chain (R). There
is no direct relationship between the number of carbon atoms
on the side chain and the growth inhibition. The best result
is obtained when R is n-butyl. On the basis of the in vitro
profile of the synthesized compounds, derivative 1 (Z =
CO2Me, R = nBu) was selected as a lead compound for
further biological evaluation.
Lead 1 was evaluated against an extended panel of human
solid tumor cell lines. Table 1 shows the GI50 values obtained
in this study, which were in the range 0.24-3.8 μM. For
comparative purposes, compound 1 was tested in a set of
three immortalized nontumor cell lines. The GI50 values were
in the range 1.3-34 μM and are listed in Table 1. Remark-
ably, lead 1 is more active against cells that divide faster,
while it is less potent against cells with a much slower
dividing profile such as MCF-10A.
A recent work using iRNA TOP2-depleted cells showed
that the decatenation activity of TOP2 appeared to be
essential for metaphase-anaphase transition.12 Indeed, fail-
ure to initiate TOP2 decatenation resulted in metaphase
arrest independent of DNA damage.5 Taking into account
all these considerations, the results obtained from the cell
cycle and protein expression studies pointed out TOP2 as a
plausible molecular target. Therefore, we next looked to the
ability of 1 to inhibit TOP2.
Cell Cycle. To get insights into the mechanism of action,
we analyzed whether 1 affected cell cycle progression by
using flow cytometry. In all cell lines tested, 1 induced cell
cycle arrest in G2/M phase and in a dose dependent manner
(Figure 2). This increase in G2/M phase is concomitant with
a decrease in the G0/G1 and the S stages. Incubation with 1
did not induce significant cell lysis, as there was no large
increase of cells with a sub-G1 DNA content, indicative of
lack of apoptosis. Accordingly, as expected, Annexin V
binding assays and PARP proteolysis experiments do not
show any evidence of apoptosis.
TOP2 Assays. Drugs targeting TOP2 are divided into two
broad classes: the so-called TOP2 poisons and the TOP2 CIs.
Whereas TOP2 poisons produce DNA double-strand breaks
through stabilization of the TOP2-DNA covalent com-
plexes, TOP2 CIs are a heterogeneous group of compounds
interfering with the catalytic cycle. To test whether compound
1 was activeagainstTOP2, weperformed three specificinvitro
assays with commercially available purified enzyme and a set
of small molecules of circular DNA as substrates. The use of
thesesmallDNA moleculesgreatly facilitatesthe visualization
of the reactions by agarose gel electrophoresis.

Brief Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9 3837
Figure 2. Cell cycle phase distribution of untreated cells (control) and cells treated with compound 1 for 24 h at three drug doses. Values are
means of at least two experiments.
Figure 3. Immunobloting of protein extracts from HeLa cells exposed to 1 at 7.5 μM after 3, 6, 12, and 24 h of drug treatment. Cþ (positive
control).
We first tested the capacity of lead 1 to inhibit TOP2-
mediated kDNA decatenation, the so-called decatenation
assay. This assay is considered to be the most specific one to
detect TOP2 activity.13 kDNA cannot enter a typical agarose
gel during electrophoresis unless DNA minicircles (Nck, SC
and Rel) are previously released by the action of TOP2. In
addition to 1, the TOP2 CI ellipticine and the TOP2 poison
VP-16 were also tested for comparison purposes. In Figure
4a, complete decatenation was seen in the positive control
(TOP2þ) and VP-16 lanes, while ellipticine and 1 inhibited
TOP2 decatenation activity as seen by comparison with the
sample that lacked the enzyme (TOP2).
In a second experiment, the SC DNA relaxation assay, we
measured the TOP2-dependent relaxation of negatively SC
plasmid pHOT1. Commercially available pHOT1 contains
small amounts of two topoisomers: a relaxed covalently
closed form (Rel) and a nicked single-strand break form
(Nck). In an agarose gel electrophoresis under controlled
conditions, the order of migration of these forms is, from
fastest to slowest, SC, Nck, and Rel. As shown in Figure 4b,
the effect of lead 1 showed a pattern that resembled that of
the control without the enzyme (TOP2-). In contrast, VP-16
did not inhibit the enzyme, resembling the positive control
(TOP2þ) lane, appearing as a set of variably relaxed topo-
isomers that migrate more slowly than the SC form.
Both the decatenation and the relaxation assays show
unequivocally that 1 is a TOP2 inhibitor. Although in both
experiments there is a difference in activity profile between 1
and VP-16, it had to be established whether compound 1
could poison the TOP2 reaction by increasing the steady
state of the cleavage complex, as it has been established for
VP-16.14 In fact, as shown previously, VP-16 has little effect
on the overall TOP2 activity in spite of being a strong TOP2
poison. Therefore, we run the cleavage complex assay with
pHOT1 as substrate. In this experiment, 1 was added once
the relaxation reaction had already started and the electro-
phoresis was run under the presence of an excess of ethidium
bromide to fully positively supercoil all the topoisomers (i.e.,
make them run as a sole band during the electrophoresis) to
clearly unmask the linear form (Lin) of the DNA. This form

3838 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9
Leoꢀn et al.
Figure 4. Agarose gel electrophoresis of TOP2 assays. (a) kDNA decatenation; (b) SC DNA relaxation; (c) cleavage complex.
is the result of the cut by the enzyme, and although it is
normally rather transient and difficult to detect because the
enzyme quickly reseals it, it can be seen with potent inhibitors
of the ligation step of the catalytic cycle of TOP2, as is the
case of VP-16. Figure 4c shows the presence of the Lin form
in the VP-16 lane, while incubation with 1 did not lead to the
Lin form.
The result of the cleavage complex assay confirmed that 1
is a TOP2 inhibitor that is not a poison, thus behaving as a
true TOP2 CI. Additional support for this evidence was
obtained from the immunoblotting experiments (Figure 3).
Unlike TOP2 CIs, TOP2 poisons produce DNA damage and
cell cycle arrest in G2, which were not observed when cells
were exposed to 1. Like almost any drug, we cannot discard
that compound 1 may have additional biological targets.
However, all the experimental data presented herein suggests
that the most relevant cellular target is TOP2.
51.1, 52.8, 70.5, 78.5, 83.0, 99.1, 153.1, 159.8, 167.5. IR (CHCl3,
cm-1) 2957.1, 2870.8, 2241.4, 1721.4, 1628.5, 1436.8, 1260.4,
1134.5. MS, m/z (relative intensities) 254 (Mþ, 0.8), 153 (58),
121 (24), 93 (100), 91 (22), 79 (56), 77 (26), 59 (45), 55 (26). Anal.
(C13H18O5) C, H.
Acknowledgment. Financial support cofinanced by the
EU-FEDER: the Spanish MICIIN (CTQ2008-06806-C02-
01/BQU, CTQ2008-06806-C02-02/BQU, and BFU2006-
01813/BMC), MSC (RTICC RD06/0020/1046 and RD06/
0020/0041); Canary Islands’ ACIISI (PI 2007/021) and
FUNCIS (REDESFAC PI 01/06 and 35/06). L.G.L and
E.P.R.: Spanish MSC-FIS Sara Borrell contracts. J.M.P.:
ꢀ
Spanish MEC-FSE Ramon y Cajal contract.
Supporting Information Available: Experimental procedures
for synthesis and biological tests. Characterization of com-
pounds 1a-y, 3, and 4. Antiproliferative activity of all com-
pounds. Cell cycle histograms and Annexin V plots. This
material is available free of charge via the Internet at http://
pubs.acs.org.
Conclusion
In summary, we have identified a novel human TOP2 CI
from a small and structure-focused library of propargylic enol
ethers. These simple, linear, and densely functionalized frag-
ments are assembled in a fast, modular, and efficient manner
by a chemodifferentiating organocatalyzed ABB0 3CR based
manifold using commercially available aldehydes and alkyl
propiolates. It is noteworthy that the synthesis of fragment
lead 1 can be performed in a multigram scale without effi-
ciency loss. Ongoing lead optimization studies will shed light
on the mechanism of TOP2 inhibition and will be reported in
due time.
References
(1) Murray, A. W. Recycling the cell cycle: cyclins revisited. Cell 2004,
116, 221–234.
(2) Nitiss, J. L. Targeting DNA topoisomerase II in cancer chemo-
therapy. Nat. Rev. Cancer 2009, 9, 338–350.
(3) Nitiss, J. L. DNA topoisomerase II and its growing repertoire of
biological functions. Nat. Rev. Cancer 2009, 9, 327–337.
(4) Shamu, C. E.; Murray, A. W. Sister chromatid separation in frog
egg extracts requires DNA topoisomerase II activity during ana-
phase. J. Cell Biol. 1992, 117, 921–934.
(5) Skoufias, D. A.; Lacroix, F. B.; Andreassen, P. R.; Wilson, L.;
Margolis, R. L. Inhibition of DNA decatenation, but not DNA
damage, arrests cells at metaphase. Mol. Cell 2004, 15, 977–990.
Experimental Section
ꢀ
ꢀ
(6) (a) Tejedor, D.; Lopez-Tosco, S.; Cruz-Acosta, F.; Mendez-Abt,
G.; Garcıa-Tellado, F. Acetylides from alkyl propiolates as build-
ing blocks for C3 homologation. Angew. Chem., Int. Ed. 2009, 48,
´
The purity of final compounds was assessed by elemental
analyses and found to be >95% in all cases.
ꢀ
ꢀ
2090–2098. (b) Tejedor, D.; Gonzalez-Cruz, D.; Santos-Exposito, A.;
Marrero-Tellado, J. J.; de Armas, P.; García-Tellado, F. Multicomponent
Domino Processes Based on the Organocatalytic Generation of Con-
jugated Acetylides. Efficient Synthetic Manifolds for Diversity-
Oriented Molecular Construction. Chem.;Eur. J. 2005, 11, 3502–
3510.
Methyl 4-((E)-2-(Methoxycarbonyl)vinyloxy)oct-2-ynoate (1).
Methyl propiolate (2.00 mmol) and n-pentanal (1.1 mmol) were
dissolved in 10 mL of CH2Cl2. After the mixture was cooled to
0 °C, triethylamine (1.0 mmol) was added and the reaction
mixture was allowed to stir for 30 min. The solvent and excess
reagents were then removed under reduced pressure. Product
1 was isolated by flash column chromatography (silica gel,
(7) (a) Tejedor, D.; Garcıa-Tellado, F. Chemo-differentiating ABB’
´
multicomponent reactions. Privileged building blocks. Chem.
Soc. Rev. 2007, 36, 484–491. (b) Tejedor, D.; García-Tellado, F.;
Marrero-Tellado, J. J.; de Armas, P. Efficient Domino Process
Based on the Catalytic Generation of Non-Metalated, Con-
jugated Acetylides in the Presence of Aldehydes or Activated
Ketones. Chem.;Eur. J. 2003, 9, 3122–3131. (c) de Armas, P.;
García-Tellado, F.; Marrero-Tellado, J. J.; Tejedor, D.; Maestro,
1
n-hexane/EtOAc 90/10) as colorless oil, 83% yield. H NMR
(CDCl3, 400 MHz): δ 0.89 (t, 3H, J=7.4 Hz), 1.30-1.37 (m, 2H),
1.39-1.47 (m, 2H), 1.81-1.91 (m, 2H), 3.67 (s, 3H), 3.75 (s, 3H),
4.60 (t, 1H, J=6.4 Hz), 5.34 (d, 1H, J=12.5 Hz), 7.51 (d, 1H, J=
12.5 Hz). 13C NMR (CDCl3, 100 MHz): δ 13.7, 22.0, 26.8, 34.2,
ꢀ
M. A.; Gonzalez-Platas, J. Alkynoates as a Source of Reactive

Brief Article
Alkylinides for Aldehyde Addition Reaction. Org. Lett. 2001, 3,
1905–1908.
(8) Taylor, W. R.; Stark, G. R. Regulation of the G2/M transition by
p53. Oncogene 2001, 20, 1803–1815.
(9) Hans, F.; Dimitrov, S. Histone H3 phosphorylation and cell
division. Oncogene 2001, 20, 3021–3027.
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 9 3839
BUBR1, BUB3, CDC20, and MAD2. J. Cell Biol. 2001, 154,
925–936.
(12) Coelho, P. A.; Queiroz-Machado, J.; Carmo, A. M.; Moutinho-
Pereira, S.; Maiato, H.; Sunkel, C. E. Dual role of topoisomerase II
in centromere resolution and aurora B activity. PLoS Biol. 2008, 6,
1758–1777.
(10) Crosio, C.; Fimia, G. M.; Loury, R.; Kimura, M.; Okano, Y.;
Zhou, H.; Sen, S.; Allis, C. D.; Sassone-Corsi, P. Mitotic phos-
phorylation of histone H3: spatio-temporal regulation by mamma-
lian Aurora kinases. Mol. Cell. Biol. 2002, 22, 874–885.
(11) Sudakin, V.; Chan, G. K.; Yen, T. J. Checkpoint inhibition
of the APC/C in HeLa cells is mediated by a complex of
(13) Muller, M. T.; Helal, K.; Soisson, S.; Spitzner, J. R. A rapid and
quantitative microtiter assay for eukaryotic topoisomerase II.
Nucleic Acids Res. 1989, 17, 9499.
(14) Bromberg, K. D.; Burgin, A. B.; Osheroff, N. A two-drug model
for etoposide action against human topoisomerase IIR. J. Biol.
Chem. 2003, 278, 7406–7412.