The natural diterpene tonantzitlolone A and its synthetic enantiomer inhibit cell proliferation and kinesin-5 function

Article abstract of DOI:10.1016/j.ejmech.2016.02.022

Tonantzitlolone A, a diterpene isolated from the Mexican plant Stillingia sanguinolenta, shows cytostatic activity. Both the natural product tonantzitlolone A and its synthetic enantiomer induce monoastral spindle formation in cell experiments which indicates inhibitory activity on kinesin-5 mitotic motor molecules. These inhibitory effects on kinesin-5 could be verified in in vitro single-molecule motility assays, where both tonantzitlolones interfered with kinesin-5 binding to its cellular interaction partner microtubules in a concentration-dependent manner, yet with a larger effect of the synthetic enantiomer. In contrast to kinesin-5 inhibition, both tonantzitlolone A enantiomers did not affect conventional kinesin-1 function; hence tonantzitlolones are not unspecific kinesin inhibitors. The observed stronger inhibitory effect of the synthetic enantiomer demonstrates the possibility to enhance the overall moderate anti-proliferative effect of the lead compound tonantzitlolon A by chemical modification.

Full text of DOI:10.1016/j.ejmech.2016.02.022

European Journal of Medicinal Chemistry 112 (2016) 164e170
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Short communication
The natural diterpene tonantzitlolone A and its synthetic enantiomer
inhibit cell proliferation and kinesin-5 function
,
1
b
c
c, 2
Tobias J. Pfeffer a , Florenz Sasse , Christoph F. Schmidt , Stefan Lakamper
,
€
Andreas Kirschning d, Tim Scholz a
,
*
a Molecular and Cell Physiology, Hannover Medical School (MHH), Carl-Neuberg-Str. 1, D-30625 Hannover, Germany
b Chemical Biology, Helmholtz Centre for Infection Research (HZI), Inhoffenstraße 7, D-38124 Braunschweig, Germany
c
€
€
Drittes Physikalisches Institut, Georg-August University Gottingen, Friedrich-Hund-Platz 1, D-37077 Gottingen, Germany
d Institute of Organic Chemistry and Center of Biomolecular Drug Research (BMWZ), Leibniz University Hannover, Schneiderberg 1B, D-30167 Hannover,
Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Tonantzitlolone A, a diterpene isolated from the Mexican plant Stillingia sanguinolenta, shows cytostatic
activity. Both the natural product tonantzitlolone A and its synthetic enantiomer induce monoastral
spindle formation in cell experiments which indicates inhibitory activity on kinesin-5 mitotic motor
molecules. These inhibitory effects on kinesin-5 could be verified in in vitro single-molecule motility
assays, where both tonantzitlolones interfered with kinesin-5 binding to its cellular interaction partner
microtubules in a concentration-dependent manner, yet with a larger effect of the synthetic enantiomer.
In contrast to kinesin-5 inhibition, both tonantzitlolone A enantiomers did not affect conventional
kinesin-1 function; hence tonantzitlolones are not unspecific kinesin inhibitors. The observed stronger
inhibitory effect of the synthetic enantiomer demonstrates the possibility to enhance the overall mod-
erate anti-proliferative effect of the lead compound tonantzitlolon A by chemical modification.
© 2016 Elsevier Masson SAS. All rights reserved.
Received 13 November 2015
Received in revised form
5 February 2016
Accepted 7 February 2016
Available online 8 February 2016
Keywords:
Inhibitor
Cytostatic
Diterpene
Kinesin spindle protein
Monoaster
Cancer
1. Introduction
Indian tribes of Navajos and Creek used the closely related
S. sylvatica in a similar fashion [3,4]. So far, only initial reports
The endemic Mexican medical plant Stillingia sanguinolenta is
particularly rich of unusual cyclic organic compounds [1,2]. For
example, the two diterpenes tonantzitlolone A (1) and B (2; OAc at
C-40 of side chain) are secondary metabolites from S. sanguinolenta.
Although diterpene cyclases commonly provide diterpenes with
14-membered rings, the backbone of tonantzitlolone A and B is
based on a rare 15-membered flexibilane skeleton (3) (Fig. 1) [1,2].
It is known that plants of the genus Stillingia, which covers about 30
species mainly in the Americas, served natives for various medici-
nal purposes. For example, the roots of S. sanguinolenta are used in
poultices after childbirth, and northern Mexican natives recom-
mend infusions of leaves for the treatment of pulmonary ailments.
appeared on the biological activity of these tonatzitlolones [5,6].
Here we disclose our preliminary findings on the biological anti-
proliferative activity of tonantzitlolone A (1) and its synthetic
enantiomer (ent)-tonantzitlolone A (4) as inhibitors of the mitotic
motor protein kinesin-5 using a single-molecule fluorescence
approach.
2. Results and discussion
Tonantzitlolone A (1) used in these studies was isolated from
Stillingia sanguinolenta as reported previously [1] and described in
the experimental section while its enantiomer (4) was obtained by
total synthesis [5,7,8]. As judged by HPLC and 1H NMR spectroscopy
the purities were determined to be >95%, respectively. For a first
biological evaluation both compounds were subjected to in vitro
biological testing of their anti-proliferative activity on different
mammalian cell lines. This biological evaluation indicated moder-
* Corresponding author.
E-mail address: scholz.tim@mh-hannover.de (T. Scholz).
Present address: Dept. of Cardiology and Angiology, Hannover Medical School,
1
Carl-Neuberg-Str. 1, D-30625 Hannover, Germany.
2
ate low-mM activity. The results from these tests are given as values
Present address: Institute for Mechanical Systems, D-MAVT, ETH Zurich, Tan-
nenstrasse 3, CH-8092 Zurich, Switzerland.
for the half-maximal cell growth inhibitory concentration (GI50
,
http://dx.doi.org/10.1016/j.ejmech.2016.02.022
0223-5234/© 2016 Elsevier Masson SAS. All rights reserved.

T.J. Pfeffer et al. / European Journal of Medicinal Chemistry 112 (2016) 164e170
165
Fig. 1. Structures of tonantzitlolone A (1) and B (2), (ent)-tonantzitlolone A (4) and schematic representation of the flexibilane backbone (3).
Table 1).
Although both tonantzitlolone A enantiomers demonstrated an
anti-proliferative impact on cell lines with GI50 values ranging from
~5 to 50 M, from those cell experiments the origin of this cyto-
static effect of tonantzitlolone A (1) and its enantiomer (4) and their
cellular targets remained unclear. Very recently, the natural com-
pound tonantzitlolone A (1) has been reported to activate PKC-
theta and the heat shock factor 1 (HSF1) transcription factor in
synthetic enantiomer 4 on kinesin-5 function we used a single-
molecule Total Internal Refection Fluorescence (TIRF) microscopy
approach [17,18] (Fig. 3A) comparable to previous studies on
kinesin-5 inhibition by monastrol [19]. We made use of a chimeric
kinesin-5 motor construct as a model system for kinesin-5 function,
in which the catalytic motor domain of kinesin-5 was fused to the
truncated neck coiled coil of kinesin-1. This dimeric construct was
reported to be very suitable for such studies as it was found to be
fully functional and capable of long processive runs along micro-
tubules, yet without the complex tetrameric structure of kinesin-5
and its associated tail-mediated self-inhibition [19,20]. In contrast
to other inhibitor tests like kinesin-5 ATPase measurements or cell-
based screening assays, single-molecule experiments can be per-
formed with very little material in extremely small volumes of less
m
low-mM concentrations (~1e5 mM) and predominantly in renal
carcinoma cells, thus starving these cells of glucose and causing cell
death [6]. However, in addition to the anti-proliferative activity of
tonantzitlolone A (1) we observed that when PtK2 potoroo kidney
cells were incubated for 18 h with 21.5
mM of 1 dissolved in
methanol, ~20% of mitotic cells were arrested in mitosis and
showed an unphysiological monoastral half spindle (Fig. 2B)
instead of a normal bipolar spindle apparatus (Fig. 2A). In contrast,
phenotypes of interphase cells were not affected by tonantzitlolone
A (1), and cells incubated with only the solvent behaved regularly
(Fig. 2). Such monoastral phenotype has been associated before
with the inhibition of the molecular motor kinesin-5 (also known
as Kif11, Eg5 or kinesin spindle protein) [9]. Thus, the observation of
monoastral half-spindle formation suggests the motor protein
kinesin-5 as one cellular target of tonantzitlolone A.
Members of the kinesin superfamily of motor proteins convert
the chemical energy derived from ATP hydrolysis into directed
movement along their cytoskeletal microtubule tracks. Microtu-
bules are key structural elements of the spindle, and kinesin-5 is a
mitotic motor protein which is required to generate the spindle's
bipolar architecture thus playing an essential role in proper spindle
morphogenesis [10,11]. This implies that kinesin-5 is also important
for tumor formation. In mitosis, homotetrameric plus-end directed
kinesin-5 molecules cross-link and simultaneously move anti-
parallel interpolar spindle microtubules [12e14], which results in
outward-directed relative microtubule sliding and spindle pole
separation [15]. While tubulin-derived microtubules are needed for
multiple essential processes also in post-mitotic cells, kinesin mo-
tors that function almost exclusively during mitosis have recently
emerged as a drugable target class in cancer chemotherapy [16],
making small molecule inhibitors of kinesin-5 rational alternatives
to tubulin targeting anti-cancer drugs.
than 5 ml. This can be important when only very limited sample
sizes e.g. from total syntheses are available. In such single-molecule
motility experiments, individual fluorescently labeled motor mol-
ecules are visualized and tracked while they proceed under ATP
consumption along immobilized microtubules. From such experi-
ments, functional parameters such as speed of movement or
kinesin-5 microtubule binding frequency can be determined, pa-
rameters which are not accessible in cell screening or ATPase
measurements. To visualize directed motion of individual kinesin
molecules along microtubules, so-called kymographs, i.e. time plots
of fluorescent kinesin positions on a microtubule, are often used
(Fig. 3B). While static molecules appear as vertical traces, trajec-
tories of kinesin molecules moving directed along the microtubule
deviate towards one microtubule end to which the kinesin mole-
cule moves. The velocity can be extracted from the slope of the
respective trajectory.
Both enantiomers of tonantzitlolone A inhibit kinesin-5 mole-
cules in a dose-dependent manner (Fig. 4, Fig. S1, Tables S1 and S2).
While kinesin-5 velocity (Fig. 4A) and the fraction of motile
kinesin-5 molecules (Fig. S1A) were unaffected, the attachment
frequency of kinesin-5 to microtubules (Fig. 4B) as well as the total
number of kinesin molecules bound per micrometer microtubule
length (Fig. S1B) were significantly reduced with increasing in-
hibitor concentrations. The attachment frequency represents the
binding affinity of kinesin to microtubules and is described by the
number of kinesin molecules which landed and started moving on
a microtubule per micrometer microtubule length per second time
[21]. Tonantzitlolone A reduced the attachment of kinesin-5 mol-
ecules to microtubules with a half maximal inhibitory effect at IC50
To directly test for possible inhibitory effects of 1 and its
~147
mM. In comparison, the synthetic enantiomer 4 showed a
Table 1
stronger inhibitory effect than the natural compound 1. The pres-
ence of (ent)-tonantzitlolone A had a more than three-fold stronger
inhibitory effect on kinesin-5 binding to microtubules with a half
Half-maximal anti-proliferative concentrations GI50 [mM] of 1 and 4 with cultured
mammalian cell lines. Values shown are means of two determinations in parallel; U-
937 (human histiocytic lymphoma), L-929 (mouse fibroblasts), PtK2 (potoroo kidney
cells); n.d., not determined.
maximal inhibitory effect at IC50 ~44.5 mM.
Compound
U-937
L-929
PtK2
While the attachment frequency of kinesin-5 to microtubules is
related to the attachment rate constant kon, the total number of
kinesin-5 molecules on microtubules, which additionally includes
1
4
45 15
4.5 1.9
>80
43
n.d.
6
37
6

166
T.J. Pfeffer et al. / European Journal of Medicinal Chemistry 112 (2016) 164e170
Fig. 2. Effect of tonantzitlolone A on mitotic cells. Microtubules are shown in green, DAPI labeled DNA is shown in blue. (A) PtK2 potoroo kidney cells with a mitotic cell in the centre
which formed a regular bipolar spindle apparatus. (B) After incubation of PtK2 potoroo cells with 21.5 M tonantzitlolone A for 18 h in ~20% of mitotic cells monoastral half-spindles
could be observed. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
m
stationary as well as moving kinesin-5 molecules present already at
the beginning of recordings, is related to the ratio of kon and the
dissociation rate constant koff (Fig. 5A). Furthermore, koff directly
influences the run length of kinesin molecules as faster dissociation
leads to shorter traveling distances. Under the experimental con-
ditions used in this study, however, the duration of processive runs
of kinesin-5 molecules usually exceeded the already extended
observation times (e.g. Fig. 3B) both in the absence and the pres-
ence of inhibitors. Hence, koff could not have been increased
dramatically by tonantzitlolone A or (ent)-tonantzitlolone A. As
mentioned above, the total number of kinesin-5 molecules per
micrometer microtubule length, i.e. the ratio kon/koff, was signifi-
cantly reduced with increasing concentrations of tonantzitlolone A
(1) or (ent)-tonantzitlolone A (4) (Fig. S1B). With IC50 values of ~175
immediate release of the kinesin motor molecule from the micro-
tubule and prevent initiation of processive movement.
Current models for allosteric kinesin-5 inhibition by small
molecules involve competition with the energy substrate ATP
[23e25] and disturbance of microtubule binding and microtubule-
stimulated ATPase activity as a result of kinesin conformational
changes in response to inhibitor binding [26e28]. Many reported
kinesin-5 inhibitors such as monastrol, ispinesib and S-trityl-L-
cysteine target a surface-exposed allosteric inhibitor binding
pocket, which is remote from the nucleotide binding site and
mainly comprises loop-5 of kinesin-5 molecules and adjacent res-
idues of helix-3 [28e32]. Loop-5 is a key structural element found
in all kinesins but is longest in the mitotic kinesin-5, and it was
suggested to function there as an on/off switch which can interact
with the nucleotide binding site [33,34]. Although both enantio-
mers of tonantzitlolone A are chemically identical with respect to
their 15-membered flexibilane backbone and functional groups, the
stronger inhibition of kinesin-5 by (ent)-tonantzitlolone A (4)
suggests stronger binding to kinesin-5 and/or higher inhibition
efficiency. Such stereochemical aspects have been reported before
to have an impact on the potency of kinesin-5 inhibitors
[9,29,32,35]. Hydrophobic interactions of the inhibitor with the
core of the kinesin-5 motor were suggested to be essential for
binding affinity and biological activity, and increased binding af-
finity and therefore increased biological activity was explained by
additional binding of the inhibitor to a cooperative minor binding
pocket [29]. The higher inhibition potency of (ent)-tonantzitlolone
A (4) could be a hint that the absolute structure of the synthetic
enantiomer 4 might fit better into the allosteric inhibitor binding
pocket of kinesin-5. Rare cases have been reported where both
enantiomers exhibited similar biological activities. Wiskostatin, a
small molecule that inhibits activation of Arp2/3 by N-WASP and
therefore actin-polymerization is such an example. The (S)-isomer
and ~54 mM, respectively, the concentration-dependent decrease
showed a behaviour almost similar to the attachment frequency.
The observed reduction of the total number of kinesin molecules
per micrometer microtubule length could therefore be a result of a
respective reduction in kon. Together with the observation that no
decrease in the run length, i.e. koff, could be observed in the pres-
ence of both tonantzitlolones, this suggests a possible model in
which tonantzitlolone A (1) and (ent)-tonantzitlolone A (4) reduce
initial attachment of kinesin-5 molecules to microtubules while
dissociation is not enhanced (Fig. 5B).
An undiminished velocity of individual kinesin-5 molecules in
the presence of the well-characterized kinesin-5 inhibitor monas-
trol has been reported before in combination with an increase in
kinesin detachment from microtubules [19]. In multi-molecule
kinesin motility assays, on the other hand, increasing concentra-
tions of monastrol caused a (sometimes very abrupt) decrease in
microtubule gliding velocity with IC50 values of ~14e240 mM [9,19],
which has also been observed for the inhibitor Terpendole E [22].
For an individual kinesin-5 molecule, microtubule-stimulated
ATPase might be unaffected by Tonantzitlolone A inhibitors as
single-molecule velocity remained unchanged even at high inhib-
itor concentrations. In ensemble ATPase measurements, however,
an increasing number of non-active kinesin molecules clogged by
increasing inhibitor concentrations would feign an overall decrease
in ATPase activity. The observation that the attachment frequency
of kinesin-5 molecules decreased with increasing inhibitor con-
centrations suggests that tonantzitlolone A and its synthetic
enantiomer lock kinesin-5 molecules in an ADP-containing or ADP-
like conformation with low microtubule affinity. This might cause
shows an IC50 value of 4.35
3.44 M [36].
mM and (R)-wiskostatin an IC50 value of
m
For a possible future therapeutic use of tonantzitlolone A deri-
vates it is necessary to screen for unwanted side effects on other
kinesin family motor proteins such as neuronal kinesin-1. To test if
tonantzitlolone A and (ent)-tonantzitlolone A (4) selectively inhibit
kinesin-5, the motile behaviour of kinesin-1 molecules in the
presence of these inhibitors was measured (Fig. 4, open symbols).
Both velocity and microtubule attachment frequency of kinesin-1
molecules remained unchanged in the presence of 1 and 4, and

T.J. Pfeffer et al. / European Journal of Medicinal Chemistry 112 (2016) 164e170
167
Fig. 3. In vitro kinesin-1 and kinesin-5 single-molecule motility assays (A) Experimental arrangement (not to scale): GFP-tagged (green) kinesin molecules land on and move along
fluorescently Cy5-labeled microtubules (alpha-tubulin red and beta-tubulin dark blue) immobilized on a glass surface of a flow chamber by anti-tubulin antibodies (orange).
Evanescent field illumination (blue rays) allows single molecule detection. (B) Kymographs, time plots of fluorescent kinesin positions on a microtubule, of GFP-tagged kinesin-1
(left) and GFP-tagged kinesin-5 molecules (right, two examples) moving along immobilized microtubules in the presence of 1 mM ATP. The velocity of a molecule can be extracted
from the slope of the respective trajectory. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
immobile kinesin-1 molecules did not occur more often (Fig. S1A).
This suggests that (i) these tonantzitlolones inhibit kinesin-5 but
not kinesin-1 function and therefore are not general inhibitors for
kinesin motors and that (ii) these tonantzitlolones do not act as
kinesin-5 inhibitors via binding to the microtubule track which
would affect both kinesin-1 [18] and kinesin-5. However, it was
reported previously that another mitotic kinesin family member,
class-7 kinesin centromere-associated protein-E (CENP-E), can be
curbed by loop-5 directed allosteric inhibition [37], yet with a
different cellular phenotype of failing metaphase chromosome
alignment. Therefore, the impact of tonantzitlolone A on other
kinesin family members such as kinesin-7 needs to be tested,
although kinesin-7 inhibition would also result in the aspired
mitotic arrest of tumor cells.
of tumor cells [38] or are eliminated from the cells by the activity of
the multidrug resistance (MDR) [35]. Since single-molecule in vitro
as well as cell growth experiments showed kinesin-5 inhibition,
both tonantzitlolones seem to be membrane permeable and are
active in cells.
3. Conclusions
The mitotic molecular motor protein kinesin-5 was reported to
be a rational target of drug discovery in cancer chemotherapy. In
addition to tonantzitlolone A's recently reported cytotoxic effects
based on PKC-theta and HSF1 transcription factor activation [6], our
results from cell biological and single-molecule motility studies
show that the natural compound tonantzitlolone A (1) and its
synthetic enantiomer (ent)-tonantzitlolone A (4) are novel mem-
brane permeable inhibitors of kinesin-5 with anti-proliferative
activity. Kinesin-5 inhibition by the latter one was stronger, while
Sometimes, drugs characterized in in vitro inhibition assays do
not - or only to a small extent - show biological activity in cell-
based experiments because they cannot pass the cell membrane

168
T.J. Pfeffer et al. / European Journal of Medicinal Chemistry 112 (2016) 164e170
Fig. 5. Functional parameters of individual kinesin-5 molecules (A) and a possible
mechanism for tonantzitlolone inhibitory effects on kinesin-5 (B). Tonantzitlolone A
(1) and (ent)-tonantzitlolone A (4) inhibit initial attachment of kinesin-5 molecules to
microtubules while dissociation is not enhanced and speed of movement not reduced.
Fig. 4. Tonantzitlolone A (1, square symbols) and (ent)-tonantzitlolone A (4, diamond
symbols) effects on kinesin-1 (open symbols) and kinesin-5 (solid symbols) motile
properties. Error bars represent SEM of individual experiments. Detailed values are
given in Tables S1 and S2. (A) Single-molecule velocity of kinesin-1 and kinesin-5
molecules did not change with increasing concentrations of both tonantzitlolones.
Value levels found in solvent DMSO controls but in the absence of inhibitors are
depicted as horizontal dashed lines both for kinesin-1 and kinesin-5. (B) The attach-
ment frequency of kinesin-5 molecules, i.e. the number of kinesin molecules which
landed on a microtubule and started moving per micrometer microtubule stretch and
second, decreased exponentially (light and dark gray fits) with increasing concentra-
tions of both tonantzitlolone A enantiomers while kinesin-1 was not affected. The
attachment frequency of kinesin-1 observed in solvent DMSO controls but in the
absence of inhibitors is depicted as horizontal dashed line.
herbarium specimen is deposited at the Monterrey Institute of
Technology and Higher Education (ITESM) Herbarium (ID 8482).
800 g air dried S. sanguinolenta roots were extracted over night
with a combination of petroleum ether, methyl tert-butyl ether, and
methanol. After evaporation of the solvent, 35 g extract were dis-
solved in methanol and kept at ꢀ20 ꢁC over night. The soluble part
(24 g) was separated into 10 fractions by column chromatography
using mixes of petroleum ether, ethyl acetate, and methanol with
increasing polarity. Fractions 1, 2, 4, 8, 9 and 10 were discarded,
while fraction 3 was further separated by HPLC (RP 8; 8 ꢂ 250 mm)
to yield 1 mg of tonantzitlolone A (methanol/H2O ¼ 9:1, Rt 8.3 min).
S. sanguinolenta aerial parts were treated in the same manner as
described above yielding 31 g of extract from 800 g starting ma-
terial. This extract was separated by column chromatography with
mixtures comprising petroleum ether, methyl tert-butyl ether, and
methanol to furnish eight fractions. Fractions 1, 7 and 8 were dis-
carded. HPLC of fraction 2 separated 15 mg of tonantzitlolone A (Rf
0.35) from other S. sanguinolenta compounds such as cembrene A
and kauranes as additionally evaluated by thin layer chromatog-
raphy (petroleum ether/methyl tert-butyl ether ¼ 9:1) as published
both enantiomeric diterpenes 1 and 4 did not inhibit neuronal
kinesin-1, another member of the kinesin superfamily. The exact
mechanism by which these tonantzitlolones inhibit kinesin-5 as
well as their binding site within kinesin-5 molecules need to be
characterized in future studies. To increase the overall moderate
inhibitor potency of this novel natural-compound based kinesin-5
inhibitors diversely functionalized tonantzitlolone derivatives
could be produced by semi- and total synthesis and characterized
by crystallographic and in silico inhibitor docking [29,30,39] and
functional in vitro and cell activity studies.
previously [1]. The optical rotation for purified tonantzitlolone A
20
was [
a
]
¼ þ134ꢁ (CHCl3; c 0.25), and spectroscopic data (see
D
supporting information) were as reported previously [1]: IR nmax
(KBr) cmꢀ1: 3381, 1741, 1684; Electron ionization mass spectrom-
etry (EIMS) (probe, 70 eV) m/z (relative intensities): 464.2774 (2)
[M]þ (calculated for C26H40O7: 464.2774), 446 [M ꢀ H2O]þ (6), 350,
[M-RCOOH]þ (2), 332 [350-H2O]þ (1), 304, [332-CO]þ (1), 245 (5),
180 (6), 97 [RCO]þ (100). Tonantzitlolone A (1), stored in the
refrigerator at 4 ꢁC, is stable for several years.
4. Experimental
4.1. Tonantzitlolone preparation and characterization
Tonantzitlolone A (1) was extracted from the endemic Mexican
medical plant Stillingia sanguinolenta as reported previously [1,7].
S. sanguinolenta plant material was collected in Valle Alto, Mon-
terrey N.L., Mexico in May 1990 and March 1992 by J. Jakupovic. A
The synthetic enantiomer (ent)-tonantzitlolone A (4) was pro-
duced by total synthesis and purified by column chromatography

T.J. Pfeffer et al. / European Journal of Medicinal Chemistry 112 (2016) 164e170
169
(hexane/ethyl acetate ¼ 2:1) as previously described in detail
For analysis individual fluorescently labeled molecules were
located and tracked basically as described previously using the
computer program ImageJ (W.S. Rasband, NIH, Bethesda, MD), the
[5,7,8] (see also supporting information). Optical rotation for (ent)-
20
tonantzitlolone A was determined to be [
a
]
¼ ꢀ119ꢁ (c 0.06,
D
CHCl3). Apart from this difference all spectroscopic data (NMR, IR,
MS) (see also supporting information) were in full accordance with
those of the plant extracted authentic material (see above) [5,7].
plug-in
Multiple
Kymograph
[21]
and
the
macro
ListSelectionCoordinates.
Conflict of interest
4.2. Cellular proliferation assays
The authors declare no conflict of interest.
3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
(MTT) was used to measure growth and viability of cells which are
Acknowledgement
capable of reducing it to a violet formazan product. 60
mL of serial
dilutions of the test compounds were added to 120 L aliquots of a
cell suspension (50,000/mL) in 96-well microplates. Blank and
solvent controls were incubated under identical conditions for 5 d.
20 mL MTT in phosphate buffered saline (PBS) were added to give a
final concentration of 0.5 mg/ml. After 2 h, the precipitate of for-
m
We are grateful to AnalytiCon Discovery (Potsdam, Germany) for
an authentic sample of tonantzitlolone A. The total synthesis of
(ent)-tonantzitlolone A was achieved by Dr. C. Jasper for which we
are thankful. We thank P. Uta, B. Hinkelmann and T. Beier for
excellent technical assistance and J. Pliquet, D.J. Manstein and B.
Brenner for helpful discussions.
mazan crystals was centrifuged, and the supernatant discarded.
The precipitate was washed with PBS (100
mL) and dissolved in
isopropanol containing 0.4% hydrochloric acid (100
mL). The
Appendix A. Supplementary data
microplates were gently shaken for 20 min to ensure a complete
dissolution of the formazan and finally measured at 595 nm using
an ELISA plate reader. All experiments were carried out in two
parallel experiments. Activity values were calculated as the mean
with respect the controls set to 100%.
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2016.02.022.
References
€
[1] G. Drager, F. Jeske, E. Kunst, E.G. Lopez, H.V. Sanchez, F. Tsichritzis,
4.3. Fluorescence staining
A. Kirschning, J. Jakupovic, Tonantzitlolone and other diterpenes from Stil-
lingia sanguinolenta, Eur. J. Org. Chem. (2007) 5020e5026.
PtK2 cells (ATCC CCL-56) were grown on glass coverslips (13 mm
diameter) in four-well plates. Exponentially growing cells were
incubated with the compounds for 18 h. Cells were fixed with cold
(ꢀ20 ꢁC) acetone-methanol (1:1) for 10 min. For labeling of mi-
crotubules, cells were first incubated with a mouse antibody
[2] T. Busch, A. Kirschning, Recent advances in the total synthesis of pharma-
ceutically relevant diterpenes, Nat. Prod. Rep. 25 (2008) 318e341.
[3] C.F. Millspaugh, American Medicinal Plants, Dover Publications Inc, New York,
1974, p. 811.
[4] D.E. Moerman, Medicinal Plants of Native America, University of Michigan,
Museum of Antropology, Ann Arbor, 1986, p. 910.
[5] C. Jasper, A. Adibekian, T. Busch, M. Quitschalle, R. Wittenberg, A. Kirschning,
Total synthesis of cyclic diterpene tonantzitlolone based on a highly stereo-
selective substrate-controlled aldol reaction and ring-closing metathesis,
Chem. Eur. J. 12 (2006) 8719e8734.
against a-tubulin (1:500; Sigma), then with a secondary goat anti-
mouse IgG antibody conjugated with Alexa Fluor 488 (1:200; Mo-
lecular Probes) at 37 ꢁC for 45 min. The nuclei and chromosomes
[6] C. Sourbier, B.T. Scroggins, P.Z. Mannes, P.J. Liao, K. Siems, D. Wolf, J.A. Beutler,
W.M. Linehan, L. Neckers, Tonantzitlolone cytotoxicity toward renal cancer
cells is PKCtheta- and HSF1-dependent, Oncotarget 6 (2015) 29963e29974.
[7] C. Jasper, R. Wittenberg, M. Quitschalle, J. Jakupovic, A. Kirschning, Total
synthesis and elucidation of the absolute configuration of the diterpene
tonantzitlolone, Org. Lett. 7 (2005) 479e482.
were stained with DAPI (1 mg/ml). The cells were washed with PBS
between different incubations. The coverslips were mounted using
Prolong Antifade (Molecular Probes), and viewed with a Zeiss
Axiophot fluorescence microscope using appropriate filter sets.
€
[8] R. Wittenberg, C. Beier, G. Drager, G. Jas, C. Jasper, H. Monenschein,
4.4. Single-molecule TIRF microscopy
A. Kirschning, Towards the total synthesis of tonantzitlolone e preparation of
key fragments and the complete carbon backbone, Tetrahedron Lett. 45
(2004) 4457e4460.
Motility assays of single eGFP-tagged rat kinesin-1 [18] or
kinesin-5/kin-1 chimera molecules [19] along immobilized
paclitaxel-stabilized Cy5-labeled microtubules (MTs) [17,18] were
performed at 23 ꢁC in buffer BRB12 (12 mM PIPES, 1 mM MgCl2,
[9] T.U. Mayer, T.M. Kapoor, S.J. Haggarty, R.W. King, S.L. Schreiber, T.J. Mitchison,
Small molecule inhibitor of mitotic spindle bipolarity identified in
phenotype-based screen, Science 286 (1999) 971e974.
a
[10] K.E. Sawin, K. LeGuellec, M. Philippe, T.J. Mitchison, Mitotic spindle organi-
zation by
a plus-end-directed microtubule motor, Nature 359 (1992)
1 mM EGTA, 10 mg/ml glucose, pH 6.8) supplemented with 10 mM
540e543.
paclitaxel and 1 mM ATP in the presence of different inhibitor
concentrations. MTs and kinesin molecules were excited by a HeNe
laser (633 nm, 35 mW, Coherent, CA, USA) and a 473 nm DPSS laser
(MBL 473, 100 mW, HB-Laserkomponenten, Schwaebisch Gmünd,
Germany), respectively, and their positions detected using a self-
made objective type total internal reflection fluorescence (TIRF)
microscope with single-fluorophore sensitivity [18]. To minimize
photo damage of proteins or bleaching of fluorophores the assay
buffers were supplemented with an oxygen scavenger system
(10 mg/ml glucose, 50 U/ml glucose oxidase, 7600 U/ml catalase,
[11] A. Castillo, H.C. Morse 3rd, V.L. Godfrey, R. Naeem, M.J. Justice, Overexpression
of Eg5 causes genomic instability and tumor formation in mice, Cancer Res. 67
(2007) 10138e10147.
[12] L.C. Kapitein, B.H. Kwok, J.S. Weinger, C.F. Schmidt, T.M. Kapoor, E.J. Peterman,
Microtubule cross-linking triggers the directional motility of kinesin-5, J. Cell
Biol. 182 (2008) 421e428.
[13] L.C. Kapitein, E.J. Peterman, B.H. Kwok, J.H. Kim, T.M. Kapoor, C.F. Schmidt, The
bipolar mitotic kinesin Eg5 moves on both microtubules that it crosslinks,
Nature 435 (2005) 114e118.
[14] M.T. Valentine, P.M. Fordyce, T.C. Krzysiak, S.P. Gilbert, S.M. Block, Individual
dimers of the mitotic kinesin motor Eg5 step processively and support sub-
stantial loads in vitro, Nat. Cell Biol. 8 (2006) 470e476.
[15] B.H. Kwok, J.G. Yang, T.M. Kapoor, The rate of bipolar spindle assembly de-
pends on the microtubule-gliding velocity of the mitotic kinesin Eg5, Curr.
Biol. 14 (2004) 1783e1788.
[16] R. Sakowicz, J.T. Finer, C. Beraud, A. Crompton, E. Lewis, A. Fritsch, Y. Lee,
J. Mak, R. Moody, R. Turincio, et al., Antitumor activity of a kinesin inhibitor,
Cancer Res. 64 (2004) 3276e3280.
[17] M.H. Hinrichs, A. Jalal, B. Brenner, E. Mandelkow, S. Kumar, T. Scholz, Tau
protein diffuses along the microtubule lattice, J. Biol. Chem. 287 (2012)
38559e38568.
and 10 mM DTT). Additionally, 50
ms laser pulses interrupted by 450
ms pauses were used to excite the fluorescently labeled samples.
After incubation for 7 min with different concentrations of poten-
tial inhibitors or their solvent DMSO movement of fluorescently
labeled kinesin-1 molecules was recorded for time periods of 60 s
with 5 frames per second while for the slower kinesin-5 movement
time periods of 600 s and frame rates of 0.5 per second were used.
[18] A. Rump, T. Scholz, C. Thiel, F.K. Hartmann, P. Uta, M.H. Hinrichs, M.H. Taft,

170
T.J. Pfeffer et al. / European Journal of Medicinal Chemistry 112 (2016) 164e170
G. Tsiavaliaris, Myosin-1C associates with microtubules and stabilizes the
inhibitors: an important cooperative 'minor binding pocket' which increases
the binding affinity significantly, J. Mol. Model 13 (2007) 987e992.
[30] H.Y. Kaan, J. Weiss, D. Menger, V. Ulaganathan, K. Tkocz, C. Laggner,
F. Popowycz, B. Joseph, F. Kozielski, Structure-activity relationship and
multidrug resistance study of new S-trityl-L-cysteine derivatives as inhibitors
of Eg5, J. Med. Chem. 54 (2011) 1576e1586.
[31] E.D. Kim, R. Buckley, S. Learman, J. Richard, C. Parke, D.K. Worthylake,
E.J. Wojcik, R.A. Walker, S. Kim, Allosteric drug discrimination is coupled to
mechanochemical changes in the kinesin-5 motor core, J. Biol. Chem. 285
(2010) 18650e18661.
[32] I. Garcia-Saez, S. DeBonis, R. Lopez, F. Trucco, B. Rousseau, P. Thuery,
F. Kozielski, Structure of human Eg5 in complex with a new monastrol-based
inhibitor bound in the R configuration, J. Biol. Chem. 282 (2007) 9740e9747.
[33] T.D. Harrington, N. Naber, A.G. Larson, R. Cooke, S.E. Rice, E. Pate, Analysis of
the interaction of the Eg5 Loop5 with the nucleotide site, J. Theor. Biol. 289
(2011) 107e115.
[34] J.S. Waitzman, A.G. Larson, J.C. Cochran, N. Naber, R. Cooke, F. Jon Kull, E. Pate,
S.E. Rice, The loop 5 element structurally and kinetically coordinates dimers of
the human kinesin-5, Eg5, Biophys. J. 101 (2011) 2760e2769.
mitotic spindle during cell division, J. Cell Sci. 124 (2011) 2521e2528.
€
[19] S. Lakamper, C. Thiede, A. Düselder, S. Reiter, M.J. Korneev, L.C. Kapitein,
E.J. Peterman, C.F. Schmidt, The effect of monastrol on the processive motility
of a dimeric kinesin-5 head/kinesin-1 stalk chimera, J. Mol. Biol. 399 (2010)
1e8.
€
[20] A. Düselder, C. Thiede, C.F. Schmidt, S. Lakamper, Neck-linker length depen-
dence of processive Kinesin-5 motility, J. Mol. Biol. 423 (2012) 159e168.
[21] A. Seitz, H. Kojima, K. Oiwa, E.M. Mandelkow, Y.H. Song, E. Mandelkow, Sin-
gle-molecule investigation of the interference between kinesin, tau and
MAP2c, Embo J. 21 (2002) 4896e4905.
[22] J. Nakazawa, J. Yajima, T. Usui, M. Ueki, A. Takatsuki, M. Imoto, Y.Y. Toyoshima,
H. Osada, A novel action of terpendole E on the motor activity of mitotic
Kinesin Eg5, Chem. Biol. 10 (2003) 131e137.
[23] T. Takeuchi, S. Oishi, T. Watanabe, H. Ohno, J. Sawada, K. Matsuno, A. Asai,
N. Asada, K. Kitaura, N. Fujii, Structure-activity relationships of carboline and
carbazole derivatives as a novel class of ATP-competitive kinesin spindle
protein inhibitors, J. Med. Chem. 54 (2011) 4839e4846.
[24] L. Luo, C.A. Parrish, N. Nevins, D.E. McNulty, A.M. Chaudhari, J.D. Carson,
V. Sudakin, A.N. Shaw, R. Lehr, H. Zhao, et al., ATP-competitive inhibitors of
the mitotic kinesin KSP that function via an allosteric mechanism, Nat. Chem.
Biol. 3 (2007) 722e726.
[35] H.Y. Kaan, V. Ulaganathan, O. Rath, H. Prokopcova, D. Dallinger, C.O. Kappe,
F. Kozielski, Structural basis for inhibition of Eg5 by dihydropyrimidines:
stereoselectivity of antimitotic inhibitors enastron, dimethylenastron and
fluorastrol, J. Med. Chem. 53 (2010) 5676e5683.
[25] C.A. Parrish, N.D. Adams, K.R. Auger, J.L. Burgess, J.D. Carson, A.M. Chaudhari,
R.A. Copeland, M.A. Diamond, C.A. Donatelli, K.J. Duffy, et al., Novel ATP-
competitive kinesin spindle protein inhibitors, J. Med. Chem. 50 (2007)
4939e4952.
[36] J.R. Peterson, L.C. Bickford, D. Morgan, A.S. Kim, O. Ouerfelli, M.W. Kirschner,
M.K. Rosen, Chemical inhibition of N-WASP by stabilization of
a native
autoinhibited conformation, Nat. Struct. Mol. Biol. 11 (2004) 747e755.
[26] Z. Maliga, T.M. Kapoor, T.J. Mitchison, Evidence that monastrol is an allosteric
inhibitor of the mitotic kinesin Eg5, Chem. Biol. 9 (2002) 989e996.
[27] L. Luo, J.D. Carson, K.S. Molnar, S.J. Tuske, S.J. Coales, Y. Hamuro, C.M. Sung,
V. Sudakin, K.R. Auger, D. Dhanak, et al., Conformation-dependent ligand
regulation of ATP hydrolysis by human KSP: activation of basal hydrolysis and
inhibition of microtubule-stimulated hydrolysis by a single, small molecule
modulator, J. Am. Chem. Soc. 130 (2008) 7584e7591.
[37] K.W. Wood, L. Lad, L. Luo, X. Qian, S.D. Knight, N. Nevins, K. Brejc, D. Sutton,
A.G. Gilmartin, P.R. Chua, et al., Antitumor activity of an allosteric inhibitor of
centromere-associated protein-E, Proc. Natl. Acad. Sci. U. S. A. 107 (2010)
5839e5844.
[38] S. Oishi, T. Watanabe, J. Sawada, A. Asai, H. Ohno, N. Fujii, Kinesin spindle
protein (KSP) inhibitors with 2,3-fused indole scaffolds, J. Med. Chem. 53
(2010) 5054e5058.
[28] H.Y. Kaan, J. Major, K. Tkocz, F. Kozielski, S.S. Rosenfeld, “Snapshots” of
ispinesib-induced conformational changes in the mitotic kinesin Eg5, J. Biol.
Chem. 288 (2013) 18588e18598.
[39] S. Nagarajan, D.A. Skoufias, F. Kozielski, A.N. Pae, Receptor-ligand interaction-
based virtual screening for novel Eg5/kinesin spindle protein inhibitors,
J. Med. Chem. 55 (2012) 2561e2573.
[29] C. Jiang, Y. Chen, X. Wang, Q. You, Docking studies on kinesin spindle protein