Binding of filamentous actin and winding into fibrillar aggregates by the polyphenolic C-glucosidic ellagitannin vescalagin

Article abstract of DOI:10.1002/anie.201006712

Winding it up: The plant polyphenolic metabolite vescalagin fulfills all the requirements for use as an antiactin agent in cellular biological investigations. Despite its high hydrophilicity, it rapidly enters cells and disturbs the organization of the actin cytoskeleton in a dose-dependent reversible manner by binding fibrillar actin and forcing the actin filaments (left) to wind themselves into ball-like fibrillar aggregates (right). Copyright

Full text of DOI:10.1002/anie.201006712

DOI: 10.1002/anie.201006712
Polyphenolic Natural Products
Binding of Filamentous Actin and Winding into Fibrillar Aggregates by
the Polyphenolic C-Glucosidic Ellagitannin Vescalagin**
Stꢀphane Quideau,* Cꢀline Douat-Casassus, Daniela Melanie Delannoy Lꢁpez,
Carmelo Di Primo, Stefan Chassaing, Rꢀmi Jacquet, Frꢀdꢀric Saltel, and Elisabeth Genot*
Herein we describe the effects of the C-glucosidic ellagitannin
vescalagin (1, Figure 1a), a polyphenolic natural product
readily available from fagaceous woody plant sources, such as
oak (Quercus sp.),[1] on actin, one of the most abundant
structural proteins in eukaryotic cells. Monomeric globular
actin (G-actin) subunits assemble through an adenosine
triphosphate (ATP) dependent process into polymeric fibril-
lar actin (F-actin) filaments, which are further ordered into
three-dimensional architectures to establish the functional
actin cytoskeleton.[2] A dynamic equilibrium between the
G-actin and F-actin states continuously ensures the adapta-
tion of the actin cytoskeleton during its various roles in
controlling cell shape, cytokinesis, motility, adhesion, and
gene expression.[3]
Natural products have been key players in probing the
role of actin by perturbing the assembly and/or disassembly of
actin filaments.[4] The bicyclic heptapeptide fungal metabolite
phalloidin and the cyclodepsipeptide sponge or microbial
metabolites jasplakinolide and chondramide C bind to F-actin
and inhibit depolymerization by strengthening actin–actin
Figure 1. a) Structures of the four oak-derived C-glucosidic ellagitan-
nins used in the antiactin in cellulo assay. b–d) Fluorescent images
showing F-actin organization in untreated BAE cells (b), in BAE cells
exposed to vescalagin (1, 50 mm) for 1 h (c), and in BAE cells exposed
to cytochalasin D (4 mm) for 1 h (d). Scale bar: 5 mm. CT=control.
[*] Prof. S. Quideau, Dr. C. Douat-Casassus, D. M. Delannoy Lꢀpez,
Dr. S. Chassaing, R. Jacquet
Universitꢁ de Bordeaux
Institut des Sciences Molꢁculaires (CNRS-UMR 5255)
Institut Europꢁen de Chimie et Biologie (IECB)
2 rue Robert Escarpit, 33607 Pessac Cedex (France)
Fax: (+33)5-4000-2215
interactions, whereas the marine macrolides known as
latrunculins bind to G-actin and prevent actin-filament
formation.[5] Other polyketide-derived natural products,
such as the fungal metabolites named cytochalasins, bind to
F-actin (and to G-actin) and thereby block both the assembly
and disassembly of individual actin monomers.[5a] Herein we
report that the polyphenolic plant metabolite vescalagin (1)
not only interacts with actin filaments but also winds them
into fibrillar aggregates in vitro and in cellulo.
Vescalagin (1) belongs to a particular group of ellagitan-
nins[6] that comprises a unique series of highly hydrosoluble
C-glucosidic variants. In these compounds, the usual ellagi-
tannin glucopyranose core is replaced by an open-chain
E-mail: s.quideau@iecb.u-bordeaux.fr
Dr. C. Di Primo
IECB/INSERM U869
2 rue Robert Escarpit, 33607 Pessac Cedex (France)
Dr. F. Saltel,[+] Dr. E. Genot[+]
IECB/INSERM U1053
2 rue Robert Escarpit, 33607 Pessac Cedex (France)
Fax: (+33)5-4000-3076
E-mail: e.genot@iecb.u-bordeaux.fr
[+] These authors contributed equally.
[**] Financial support from the Agence Nationale de la Recherche (ANR-
06-BLAN-0139 and ANR-06-BLAN-0362), the Conseil Regional
d’Aquitaine (grant 20071301020), La Ligue Contre le Cancer
(Comitꢁ Dordogne 2006 and 2007), and the Institut Europꢁen de
Chimie et Biologie (2007 Internal Seed Grant Program) is gratefully
acknowledged. D.M.D.L. acknowledges the support of Fundayacu-
cho and the French Embassy in Caracas, Venezuela, for her doctoral
research assistantship. We thank A. Nube, A. Labrousse, and I.
Maridonneau-Parini for preliminary experiments, R. Boujemaa-
Paterski and L. Blanchoin for reagents and precious advice on actin
polymerization assays, and A. Grelard for NMR spectroscopic
experiments.
glucose moiety, which is rarely encountered in nature and
[7]
À
results from the establishment of a C aryl glucosidic bond.
Our interest in studying these C-glucosidic ellagitannins stems
from the premise that the highly preorganized medium-sized-
ring-containing multiple-phenol array featured by such nat-
ural products should be structurally well-suited to interfere
with the construction of protein-made cellular architectures,
such as actin filaments or microtubules.[5b] Some ellagitannins,
which interfere with bone resorption, are known to affect
osteoclastic actin rings.[8] We thus first examined the effect of
a selection of four C-glucosidic ellagitannins on the actin
cytoskeleton in living cells. These four compounds are the two
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201006712.
Angew. Chem. Int. Ed. 2011, 50, 5099 –5104
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5099

Communications
most abundant C-glucosidic ellagitannins found in oak heart-
Live imaging carried out on BAE cells expressing actin–
GFP and treated with 1 (100 mm) showed immediate alter-
ation of actin–GFP distribution at cell margins (see Fig-
ure S7). Destabilization of the stress fibers was visualized by
the progressive loss of filamentous staining (see Movie 4)
concomitantly with cell retraction (see Movies 1–3). Notice-
ably, thick F-actin bundles were maintained. A fluorescence
recovery after photobleaching (FRAP) assay on actin–GFP-
expressing BAE cells in the absence or presence of 1
(Figure 2; see also Movies 5 and 6) revealed that 1 increases
wood, vescalagin (1) and its C1 epimer castalagin (2), and
their two minor congeners, vescalin (3) and castalin (4), both
of which lack the hexahydroxydiphenoyl (HHDP) unit at the
4- and 6-positions of the glucose core (Figure 1a).[1,9a]
We used bovine aortic endothelial (BAE) cells, a well-
characterized type of primary cell.[10] Any of the four
ellagitannins used at a concentration of 50 mm rapidly
provoked the disappearance of the internal stress-fiber net-
work observed in control cells (see Figure 1b,c and Figure S1
in the Supporting Information). Cytochalasin D, known to
inhibit F-actin polymerization,[5b,11] was also effective; how-
ever, the resulting actin perturbation appeared distinct from
that induced by ellagitannins 1–4 (Figure 1d). Focal adhe-
sions, which anchor stress fibers to the matrix through
integrins,[12] underwent dissolution, suggestive of alterations
in cell adhesion (see Figure S2). Overall, the changes in actin
configuration elicited by 1–4 were markedly different from
those induced by other natural products known to target
actin.[5a,b] We focused thereafter on vescalagin (1), because it
is the easiest of the four compounds to modify by selective
chemical means,[9,13] and it is available in large quantities by
extraction from its main natural sources.[9a]
The vescalagin-induced F-actin-disrupting effect seen in
BAE cells was also observed in fibroblast cells (BHK-21),
which also express b-actin as the main actin isoform, as well as
in smooth muscle cells (A7r5), which, in contrast, predomi-
nantly express a-actin (see Figure S3). Despite subtle differ-
ences, the impact of 1 on the actin cytoskeleton appeared
neither cell- nor actin-isoform-specific, which suggests that 1
can affect all types of mammalian cells. Furthermore, like for
cytochalasin D, no alteration of the microtubule network
could be detected upon the treatment of cells with 1 (see
Figure S4). Remarkably, all cytoskeletal alterations could be
completely reversed within 1 hour by washing 1 from the cells
with medium. Phenotype recovery occurred at a rate similar
to that observed for cytochalasin D[14] (see Figure S5).
The vescalagin-induced dissolution of stress fibers
affected cellular morphology and, eventually, viability.
Within the range of concentrations and incubation times
examined, cells changed their well-spread morphology to a
more retracted morphology (see Movies 1–3 in the Support-
ing Information). Mitosis, still observed when 1 was used at a
concentration of 50 mm (see Movie 1), became impaired at
100 mm (see Movie 2 and Figure S6). Staining with propidium
iodide (PI) revealed no cytotoxicity after 24 hours at 50 mm;
however, the presence of apoptotic nuclei, which indicate
irreversible commitment to cell death, was eventually
detected when cells were exposed to 1 at 100 mm. Functional
consequences for cells translated into impaired wound-repair
capacity (see Figure S6). The efficient serum-induced stim-
ulation of the healing of mechanically injured endothelium
was decreased by about half in the presence of 1 at a
concentration of 20 mm. The denudated area remained
virtually unpopulated for 6 hours after exposure to 1 at a
concentration of 100 mm; these conditions are still compatible
with the maintenance of cell viability (see Figure S6). These
data illustrate the rapid (see Movies 1–3) and sustained
effects of 1 on cells.
Figure 2. a,b) Live imaging of BAE cells expressing actin–GFP were
subjected to FRAP in the absence (a) or presence (b) of vescalagin (1,
100 mm). The boxed region (200ꢂ100 pixel square) was photobleached
(before (t=0 s), immediately after (t=38 s), and after photobleaching
(t=98 s)). c) Normalized fluorescence intensity in the boxed region
for the entire duration of the FRAP experiment. Fluorescence recovery
(starting at the empty circle or square) recorded over time reveals the
rates of actin turnover within this area. d) Quantitation of the results.
The immobile fraction was calculated from the difference between the
pre- and post-photobleaching intensities (n=6). FI=fluorescence
intensity; IF=immobile fraction.
the immobile fraction of actin trapped in F-actin bundles. We
conclude that 1 affects both thin and thick actin fibers.
However, thick F-actin bundles made of packed actin
filaments are less vulnerable to the action of 1 than the
single-filament dendritic meshwork at the cell periphery.
The rapidity with which 1 induces drastic perturbations
within cells suggests that 1 is capable of crossing the plasma
membrane. To investigate this behavior further, we prepared
a fluorescent vescalagin derivative by chemical synthesis and
tracked it in cellulo. After exploring various possibilities for
the design of such a derivative with an appropriate fluoro-
phore conveniently attached to the natural product, we
settled on a vescalagin derivative equipped with a fluores-
5100
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 5099 –5104

cein-terminated 13-atom-long linker[15] by using fluorescein
isothiocyanate (FITC) as the starting fluorophore. The choice
of such a relatively long linker was made to prevent
quenching of the fluorescein fluorescence through intramo-
lecular interactions with the electronic-rich polyphenolic
vescalagin-derived unit.[16] The synthesis of this vescalagin–
FITC conjugate 6 commenced with the preparation of the
fluorescein-terminated linker 5 on a solid support (see the
Supporting Information for details). With this thiol-function-
alized fluorescein derivative in hand, we took advantage of
the known chemoselectivity expressed at the vescalagin
hydroxylated C1 position under acid-catalyzed nucleophilic
substitution reaction conditions[9,13] to forge the vescalagin–
FITC conjugate 6 in 30% yield after purification by reversed-
phase HPLC (Figure 3a).
When added to living BAE cells, 6 displayed antiactin
effects similar to those observed for 1 (Figure 1c). This
fluorescent vescalagin derivative colocalized with F-actin in
the remaining internal thick stress fibers and aggregates
(Figure 3b); it was thus established that this compound had
passed the plasma membrane and reached its target. To better
visualize this target, we fixed the BAE cells to stabilize the
actin cytoskeleton. After permeabilization, simultaneous
staining with 6 and fluorescent Alexa546–phalloidin revealed
that, in the absence of the actin-destabilization effect, 6
highlighted the entire cytoskeleton (Figure 3c), similarly to
phalloidin; this result shows that 6 binds all actin fibers in
cellulo.
To explore whether or not the interaction between
vescalagin (1) and actin was direct, we set up an in vitro
assay based on actin polymerization from a solution of Ca2+
–
actin–ATP monomers. Spontaneous polymerization occurred
when Ca2+ was replaced by the Mg2+ ions present in the F-
actin buffer.[17a] High-speed centrifugation of the samples
enabled separation of the neoformed polymers from the
monomers to discriminate the binding of 1 to either F-actin or
G-actin, or both. To visualize these presumptive molecular
associations, we again used the vescalagin fluoroprobe 6 and a
fluorescent Alexa633–phalloidin conjugate to stain F-actin.
Actin polymerization was allowed to reach the steady state
under permissive and nonpermissive conditions (i.e., in the
presence or absence of Mg2+-containing F-actin buffer) and
was continued for 30 min in the presence of either 6 or
Alexa633–phalloidin. The samples were then centrifuged.
Figure 4a shows the resulting colored pellets consisting of
insoluble actin stained with the fluorescent compound and
indicates the expected binding of phalloidin to F-actin, as well
as that of 6 to that insoluble actin material. Thus, the
vescalagin–FITC conjugate 6 directly binds F-actin.
In the same way, we quantified the amount of F-actin
(pellet) relative to the amount of G-actin (supernatant) by
SDS-PAGE after the centrifugation of samples treated with
either 1 (100 mm), cytochalasin D (4 mm), or phalloidin (2 mm).
Polymerization performed in the presence of cytochalasin D
yielded a G-actin/F-actin ratio similar to that observed for the
control (Figure 4b), as expected from the ability of cytocha-
lasin D to block actin assembly and disassembly. However,
vescalagin-treated samples were characterized by a marked
depletion of G-actin, similar to that observed in the case of
samples treated with phalloidin. Thus, 1 promotes the actin
filament state, presumably by binding to F-actin and thereby
displacing the equilibrium in favor of F-actin, in a dose-
dependent manner (Figure 4c). The same effect was observed
with 6 (Figure 4d).
Analysis based on surface plasmon resonance (SPR)[13]
confirmed that 1 does not bind G-actin. The vescalagin–biotin
conjugate 9 was synthesized by first treating 1 with octane-1,8-
dithiol to furnish the sulfhydryl thioether 1-deoxyvescalagin
derivative 7 (Figure 5a).[13] Thiol 7 was then coupled to the
biotinylated maleimide linker 8.[18a] This coupling reaction
was performed in deuterated dimethyl sulfoxide (DMSO) to
enable its monitoring by 1H NMR spectroscopy.[18b] The
reaction was complete after 7 hours at room temperature
and afforded pure 9, which was precipitated from the reaction
mixture in 95% yield upon the addition of Et2O/MeOH (3:1,
v/v; see the Supporting Information for details). This
vescalagin–biotin conjugate was then readily immobilized in
one step on streptavidin-coated sensor chips in preparation
for a series of SPR analyses with G-actin and F-actin
(Figure 5b). No detectable binding occurred when a 2 mm
solution of G-actin was injected onto the vescalagin-coated
surface; however, the SPR response strongly increased when
F-actin (prepared from the 2 mm solution of G-actin)[19] was
injected. These results showed that vescalagin only binds
Figure 3. a) Preparation of the fluorescent vescalagin–FITC conjugate
6. b) Fluorescent image of live BAE cells upon the addition of 6
(50 mm, 30 min), after fixation and staining with Alexa546–phalloidin,
showing F-actin disturbance similar to that observed with 1 (scale bar:
2 mm). The zoomed view of the boxed area highlights remaining stress
fibers (*) and actin aggregates (*). c,d) Simultaneous staining of actin
filaments in lamellipodium (c) and stress fibers (d) in fixed BAE cells
with Alexa546–phalloidin and 6 revealed a very similar pattern (scale
bars: 7 mm). TFA=trifluoroacetic acid.
Angew. Chem. Int. Ed. 2011, 50, 5099 –5104
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5101

Communications
Figure 4. a) Actin polymerization at its steady state under both permis-
sive and nonpermissive conditions was continued (30 min) in the
presence of either Alexa633–phalloidin or 6 (vescalagin–FITC). After
high-speed centrifugation, F-actin stained with Alexa633–phalloidin
(blue) or with 6 (orange) was detected; however, no F-actin was
detected with FITC alone. b) The fractions were examined for actin
content by SDS-PAGE followed by Coomassie blue staining. Treatment
with cytochalasin D yielded G-actin and F-actin in quantities similar to
those obtained in the control experiment (CT), whereas samples
treated with 1 or phalloidin showed a depletion of G-actin. c) This
effect is dose-dependent and d) not altered by the presence of the
FITC-bearing unit in 6.
Figure 5. a) Synthesis of the biotinylated vescalagin conjugate 9.
b) SPR analysis of the binding of F-actin (top: black line), G-actin
(bottom: dark-gray line), BSA (gray with open circles), and streptavidin
(gray with open squares) to vescalagin. Compound 9 (217 RU) was
immobilized on a streptavidin-coated sensor chip. Protein solutions
(2 mm in running buffer) were injected across the surface at a rate of
20 mLminÀ1 for 147 s [time elapsed between the start of injection (first
arrow) and the start of the dissociation phase (second arrow)] at
258C. The dissociation phase was recorded for 300 s. RU=resonance
units.
filamentous actin. The slow dissociation phase observed
indicates that the vescalagin/F-actin complexes are highly
stable. The absence of significant SPR responses upon the
injection of 2 mm solutions of bovine serum albumin (BSA) or
streptavidin further supports the specific nature of the
interaction between vescalagin and F-actin.
physically cross-linked F-actin (Figure 6b). A merged image
of actin (red) and vescalagin (green) fluorescence revealed
the uniform colocalization of actin aggregates with 6.
How vescalagin affects actin polymerization was visual-
ized by confocal microscopy. Actin polymerization was
initiated with Alexa568–actin monomers in an F-actin
buffer. Live imaging showed free actin filaments undergoing
changes of shape on a scale of seconds; they underwent
elongation in all directions. When the steady state had been
reached (Figure 6a), the vescalagin–FITC conjugate 6 was
added. Instantly, actin filaments were observed to agglutinate
into irregular modules, which suggested that vescalagin
The stabilization of filaments by phalloidin did not
prevent this effect (results not shown). Therefore, the
vescalagin-induced collapse of actin filaments is not mediated
by actin depolymerization. When conditions under which
actin remains monomeric were used (i.e., the CaCl2-contain-
ing G-actin buffer was used),[20] the distribution of the
vescalagin–FITC conjugate appeared diffuse. This observa-
tion confirmed that vescalagin had no effect on G-actin
(Figure 6c). Furthermore, when Ca2+–actin was converted
5102
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 5099 –5104

products with a privileged capacity for binding to actin.
Vescalagin has all the requisites to be utilized as an antiactin
agent in cellular biological investigations in its natural form 1
or fluorescent version 6. Despite its high hydrophilicity,[22]
vescalagin rapidly enters cells. Its dose-dependent effects on
the actin cytoskeleton rely on its interaction with F-actin
without any perturbation of the microtubule network. At
variance with phalloidin, its effects on the actin cytoskeleton
were found to be fully reversible (within 1 h of treatment and
at concentrations up to 100 mm).
Our results are consistent with a mode of action through
which the binding of vescalagin occurs along the length of the
actin filament, probably at the protein–protein interface of
the so-polymerized actin supramolecular association. The two
analogous polyhydroxylated arene motifs (i.e., the nonahy-
droxyterphenoyl (NHTP) and HHDP units; see Figure 1a)
are probably the key structural features that enable vescala-
gin to wind F-actin into fibrillar aggregates by engaging it
through multiple intra- and/or intermolecular contacts. Fur-
thermore, since phalloidin retains its actin-binding capacity
for F-actin decorated with the vescalagin–FITC conjugate 6,
as well as for vescalagin-induced actin aggregates, we
conclude that vescalagin and phalloidin do not bind at the
same site(s). Future studies will be devoted to the elucidation
of the vescalagin–actin interaction at the molecular level and
to the comparative evaluation of the antiactin effects of the
other C-glucosidic ellagitannins initially screened in this
study. Moreover, the SPR method we developed, which uses
a biotinylated vescalagin conjugate to discriminate the bind-
ing of vescalagin to F- and G-actin, confirms the utility of the
technique for the observation of specific polyphenol–protein
interactions.[13] The implications of these antiactin oak-
derived C-glucosidic ellagitannins on human health will
form the basis of our future investigations.[23]
Figure 6. a) Live imaging of Alexa568–actin filaments grown at the
steady state. b) Actin filaments had collapsed into irregular modules in
less than 1 min following the addition of vescalagin–FITC (6; 100 mm).
Fluorescence of Alexa568–actin (red, left; 6 can be seen in the green-
channel insert) and uniform colocalization (yellow, right; the insert
shows a z-cut section) of actin aggregates with 6. c) Live imaging in
the red and green channels of Alexa568–actin under conditions under
which actin remains monomeric, following the addition of 6. d) Left:
actin polymerization initiated for Alexa568–actin monomers in the
presence of 6 by the addition of F-actin buffer; small and sporadic
clusters of aggregated F-actin (red; 6 can be seen in the green-channel
insert). Right: colocalization (yellow) of actin and 6 (the insert shows
a z-cut section) as observed at the onset of the experiment (t=1 min).
e) Live imaging of Alexa488–actin filaments grown at the steady state
after overnight exposure to jasplakinolide or vescalagin (50 mm). Scale
bars: 5 mm.
into Mg2+–actin by the use of conditions under which actin
remains monomeric (i.e., in the presence of ethylene glycol
tetraacetic acid),[20] 6 was unable to aggregate actin. This
result rules out any effect of the actin conformational change
upon Ca2+/Mg2+ exchange prior to polymerization.[17b] How-
ever, when 6 was added to G-actin, actin aggregation was
induced upon the initiation of actin polymerization by the
addition of the standard F-actin buffer. Sporadic clusters of
aggregated F-actin were observed in multiple foci all over the
coverslip. These clusters grew in size over time to form the
aggregates shown in Figure 6d. Similar results were obtained
with 1 (see Movie 7). This effect appears to be distinct from
those induced by other actin-stabilizing drugs, such as
jasplakinolide (Figure 6e) and phalloidin (similar results to
those observed with jasplakinolide). Three-dimensional anal-
ysis of these clusters revealed a fibrillar arrangement of actin,
but a random organization of the fibrils within the clusters
(see Figure S8). Collectively, these findings demonstrate that
1 becomes capable of binding actin only when polymerized or
undergoing polymerization into filaments. Thus, 1 presents
two functions: it binds actin filaments and winds them into
fibrillar aggregates.
The actin-filament-aggregation effect of 1 did not prevent
actin polymerization and furthermore decreased the pool of
G-actin. By extension, in cellulo, the spontaneous induction
of disorganized aggregates of F-actin by 1 would be expected
to circumvent regulated actin-filament elongation at filament
ends and thus lead to a cellular environment in which there is
insufficient polymerization-competent G-actin to maintain
normal stress-fiber turnover. Alterations of cellular G-actin
levels are known to affect the synthesis of actin and of other
actin-regulatory proteins.[21]
Received: October 26, 2010
Revised: December 23, 2010
Published online: April 27, 2011
Keywords: actin · aggregation · natural products · polyphenols ·
.
surface plasmon resonance
[1] a) W. Mayer, W. Gabler, A. Riester, H. Korger, Justus Liebigs
Ann. Chem. 1967, 707, 177 – 181; b) W. Mayer, H. Seitz, J. C.
Jochims, K. Schauerte, G. Schilling, Justus Liebigs Ann. Chem.
1971, 751, 60 – 68; c) G.-i. Nonaka, T. Sakai, T. Tanaka, K.
Mihashi, I. Nishioka, Chem. Pharm. Bull. 1990, 38, 2151 – 2156.
[2] C. G. Dos Remedios, D. Chhabra, M. Kekic, I. V. Dedova, M.
Tsubakihara, D. A. Berry, N. J. Nosworthy, Physiol. Rev. 2003, 2,
433 – 473.
[3] C. D. Nobes, A. Hall, Cell 1995, 81, 53 – 62.
[4] I. Spector, F. Braet, N. R. Shochet, M. R. Bubb, Microsc. Res.
Tech. 1999, 47, 18 – 37.
[5] a) J. A. Cooper, J. Cell Biol. 1987, 105, 1473 – 1478; b) T. Usui,
Biosci. Biotechnol. Biochem. 2007, 71, 300 – 308; c) H. Wald-
mann, T. S. Hu, S. Renner, S. Menninger, R. Tannert, T. Oda,
H. D. Arndt, Angew. Chem. 2008, 120, 6573 – 6577; Angew.
Chem. Int. Ed. 2008, 47, 6473 – 6477; d) R. Tannert, L. G. Milroy,
B. Ellinger, T. S. Hu, H. D. Arndt, H. Waldmann, J. Am. Chem.
Soc. 2010, 132, 3063 – 3077.
In conclusion, this study demonstrates that polyphenolic
C-glucosidic ellagitannins constitute another pool of natural
Angew. Chem. Int. Ed. 2011, 50, 5099 –5104
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5103

Communications
[6] a) Chemistry and Biology of Ellagitannins: An Underestimated
[13] C. Douat-Casassus, S. Chassaing, C. Di Primo, S. Quideau,
ChemBioChem 2009, 10, 2321 – 2324.
[14] I. Foissner, G. O. Wasteneys, Plant Cell Physiol. 2007, 48, 585 –
597.
[15] M. S. Gonꢂalves, Chem. Rev. 2009, 109, 190 – 212.
[16] S. Soares, N. Mateus, V. de Freitas, J. Agric. Food Chem. 2007, 55,
6726 – 6735.
[17] a) J. D. Pardee, J. A. Spudich, J. Cell. Biol. 1982, 93, 648 – 654;
b) M. F. Carlier, J. Biol. Chem. 1991, 266, 1 – 4.
[18] a) J. F. W. Keana, M. D. Ogan, Y. Lu, M. Beer, J. Varkeyt, J. Am.
Chem. Soc. 1986, 108, 7957 – 7963; b) R. J. Pounder, M. J.
Stanford, P. Brooks, S. P. Richards, A. P. Dove, Chem.
Commun. 2008, 5158 – 5159.
Class of Bioactive Plant Polyphenols (Ed.: S. Quideau), World
Scientific, Singapore, 2009; b) S. Quideau, K. S. Feldman, Chem.
Rev. 1996, 96, 475 – 503; c) E. Haslam, Y. Cai, Nat. Prod. Rep.
1994, 11, 41 – 66.
[7] S. Quideau, M. Jourdes, D. Lefeuvre, P. Pardon, C. Saucier, P.-L.
Teissedre, Y. Glories in Recent Advances in Polyphenol
Research, Vol. 2 (Eds.: C. Santos-Buelga, M. T. Escribano-
Bailon, V. Lattanzio), Wiley-Blackwell, Oxford, 2010, pp. 81 –
137.
[8] a) O. Hoffmann, J. Nguemo, D. Mꢀhlgassner, A. Liebhart, S.
Quideau, Bone 2006, 39, S9 – S10; b) E. K. Park, M. S. Kim, S. H.
Lee, K. H. Kim, J. Y. Park, T. H. Kim, I. S. Lee, J. T. Woo, J. C.
Jung, H. I. Shin, J. Y. Choi, S. Y. Kim, Biochem. Biophys. Res.
Commun. 2004, 325, 1472 – 1480; c) T. Hatano, L. Han, S.
Taniguchi, T. Okuda, Y. Kiso, T. Tanaka, T. Yoshida, Chem.
Pharm. Bull. 1995, 43, 2033 – 2035.
[19] For a recent example of SPR analysis involving the use of a
solution of F-actin, see: R. J. Barfoot, K. H. Sheikh, B. R. G.
Johnson, J. Colyer, R. E. Miles, L. J. C. Jeuken, R. J. Bushby,
S. D. Evans, Langmuir 2008, 24, 6827 – 6836.
[9] a) S. Quideau, M. Jourdes, D. Lefeuvre, D. Montaudon, C.
Saucier, Y. Glories, P. Pardon, P. Pourquier, Chem. Eur. J. 2005,
11, 6503 – 6513; b) S. Quideau, M. Jourdes, C. Saucier, Y. Glories,
P. Pardon, C. Baudry, Angew. Chem. 2003, 115, 6194 – 6196;
Angew. Chem. Int. Ed. 2003, 42, 6012 – 6014.
[10] C. Varon, F. Tatin, V. Moreau, E. Van Obberghen-Schilling, S.
Fernandez-Sauze, E. Reuzeau, I. Kramer, E. Gꢁnot, Mol. Cell.
Biol. 2006, 26, 3582 – 3594.
[20] L. Blanchoin, T. D. Pollard, J. Biol. Chem. 1998, 273, 25106 –
25111.
[21] A. Sotiropoulos, D. Gineitis, J. Copeland, R. Treisman, Cell 1999,
98, 159 – 169.
[22] The octan-1-ol/water partitioning coefficient Kow of vescalagin is
0.1: C. M. Spencer, Y. Cai, R. Martin, T. H. Lilley, E. Haslam,
J. Chem. Soc. Perkin Trans. 2 1990, 651 – 660.
[23] S. Quideau, E. Genot, F. Saltel, C. Douat-Casassus, D. M.
Delannoy Lꢃpez, EP11305186, 2011.
[11] J. F. Casella, M. D. Flanagan, S. Lin, Nature 1981, 293, 302 – 305.
[12] C. Le Clainche, M.-F. Carlier, Physiol. Rev. 2008, 88, 489 – 513.
5104
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 5099 –5104