Real-Time Analysis of Polyphenol–Protein Interactions by Surface Plasmon Resonance Using Surface-Bound Polyphenols

Article abstract of DOI:10.1002/chem.202005187

A selection of bioactive polyphenols of different structural classes, such as the ellagitannins vescalagin and vescalin, the flavanoids catechin, epicatechin, epigallocatechin gallate (EGCG), and procyanidin B2, and the stilbenoids resveratrol and piceatannol, were chemically modified to bear a biotin unit for enabling their immobilization on streptavidin-coated sensor chips. These sensor chips were used to evaluate in real time by surface plasmon resonance (SPR) the interactions of three different surface-bound polyphenolic ligands per sensor chip with various protein analytes, including human DNA topoisomerase IIα, flavonoid leucoanthocyanidin dioxygenase, B-cell lymphoma 2 apoptosis regulator protein, and bovine serum albumin. The types and levels of SPR responses unveiled major differences in the association, or lack thereof, and dissociation between a given protein analyte and different polyphenolic ligands. Thus, this multi-analysis SPR technique is a valuable methodology to rapidly screen and qualitatively compare various polyphenol–protein interactions.

Full text of DOI:10.1002/chem.202005187

Full Paper
doi.org/10.1002/chem.202005187
Chemistry—A European Journal
&
Surface Plasmon Resonance
Real-Time Analysis of Polyphenol–Protein Interactions by Surface
Plasmon Resonance Using Surface-Bound Polyphenols
Daniela MØlanie Delannoy López,[a] Dong Tien Tran,[a] Guillaume Viault,[a] Sofiane Dairi,[a]
Philippe Alexandre Peixoto,[a] Yoan Capello,[a] Latitia Minder,[b] Laurent PouysØgu,[a]
Elisabeth GØnot,[c] Carmelo Di Primo,[d] Denis Deffieux,*[a] and StØphane Quideau*[a, e]
Abstract: A selection of bioactive polyphenols of different
structural classes, such as the ellagitannins vescalagin and
vescalin, the flavanoids catechin, epicatechin, epigallocate-
chin gallate (EGCG), and procyanidin B2, and the stilbenoids
resveratrol and piceatannol, were chemically modified to
bear a biotin unit for enabling their immobilization on strep-
tavidin-coated sensor chips. These sensor chips were used
to evaluate in real time by surface plasmon resonance (SPR)
the interactions of three different surface-bound polyphenol-
ic ligands per sensor chip with various protein analytes, in-
cluding human DNA topoisomerase IIa, flavonoid leucoan-
thocyanidin dioxygenase, B-cell lymphoma 2 apoptosis regu-
lator protein, and bovine serum albumin. The types and
levels of SPR responses unveiled major differences in the as-
sociation, or lack thereof, and dissociation between a given
protein analyte and different polyphenolic ligands. Thus, this
multi-analysis SPR technique is a valuable methodology to
rapidly screen and qualitatively compare various polyphe-
nol–protein interactions.
Introduction
products they contain have not yet been fully exploited by the
pharmaceutical industry. Among the reasons for this mere rele-
gation of plant polyphenols to natural antioxidant agents are
mainly concerns about their possible toxicity, poor bioavailabil-
ity, and lack of specificity in interacting with proteins. If certain
polyphenols indeed only act as nonspecific protein precipitat-
ing agents (i.e., tanning action),[1g] which are usually character-
ized by relatively low binding affinities (i.e., dissociation con-
stants KD in the micromolar to millimolar range), other poly-
phenols can exhibit much higher and protein-specific affini-
ties.[1,2] This is, among numerous examples, the case for the
tight binding of the soy isoflavone genistein to the estrogen
receptor alpha,[2j] for the inhibition of an ATP synthase by the
stilbenoid resveratrol,[2k] for the submicro- to nanomolar bind-
ing of the tea flavanoid epigallocatechin gallate (EGCG) to
1) the T cell-expressed ZAP-70 tyrosine kinase,[2i] 2) the metasta-
sis-associated 67 kDa laminin receptor,[2o] and 3) the pro-angio-
genic vascular endothelial growth factor,[2c] as well as for the
inhibition of the DNA topoisomerase IIa[2d,n,r] or for the pertur-
bation of actin cytoskeletal dynamics by the oak C-glucosidic
ellagitannins vescalagin and vescalin.[2a,e]
Plant polyphenolic compounds, which are abundant in fruits
and vegetables, and present in many plant-derived foodstuffs
and beverages, are usually acclaimed for their antioxidant ac-
tivity, and yet they can also play biologically relevant roles by
interacting with cellular proteins.[1] Today, plant polyphenolic
extracts are essentially used in the production of food supple-
ments, parapharmaceuticals, and cosmetics, but the natural
[a] Dr. D. M. Delannoy López, Dr. D. T. Tran, Dr. G. Viault, Dr. S. Dairi,
Dr. P. A. Peixoto, Y. Capello, Prof. L. PouysØgu, Dr. D. Deffieux,
Prof. S. Quideau
Univ. Bordeaux, ISM (CNRS-UMR 5255)
351 cours de la LibØration, 33405 Talence Cedex (France)
E-mail: denis.deffieux@u-bordeaux.fr
stephane.quideau@u-bordeaux.fr
[b] L. Minder
INSERM, CNRS, IECB (US001, UMS 3033)
Univ. Bordeaux
2 rue Robert Escarpit, 33607 Pessac Cedex (France)
[c] Dr. E. GØnot
Centre de Recherche Cardio-Thoracique de Bordeaux (INSERM U1045)
Univ. Bordeaux
Investigations on polyphenol–protein interactions are com-
monly performed in aqueous solutions by using various analyt-
ical techniques, such as those based on NMR spectroscopy,[3a–f]
including saturation transfer difference spectroscopy,[3e,f] iso-
thermal titration calorimetry (ITC),[3f–m] mass spectrometry
(MS),[3d,n–s] FTIR spectroscopy,[3k,t] UV/Vis absorption and fluores-
cence spectroscopy,[3k,4a–h,o] circular dichroism (CD),[3d,k,4c,d,f,g,i,j]
small-angle X-ray scattering (SAXS),[4j–l] dynamic light scattering
(DLS),[4a,d,i–k] and flow nephelometry.[4d,m,n] These techniques
enable one to determine the binding sites and strengths, as
2 rue Robert Escarpit, 33607 Pessac Cedex (France)
[d] Dr. C. Di Primo
INSERM, CNRS (U1212, UMR 5320), IECB
Univ. Bordeaux
2 rue Robert Escarpit, 33607 Pessac Cedex (France)
[e] Prof. S. Quideau
Institut Universitaire de France
1 rue Descartes, 75231 Paris Cedex 05 (France)
Supporting information and the ORCID identification number(s) for the
author(s) of this article can be found under:
https://doi.org/10.1002/chem.202005187.
Chem. Eur. J. 2021, 27, 5498 –5508
5498
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005187
Chemistry—A European Journal
well as the general thermodynamic parameters of the interac-
tions, and to decipher their possible modus operandi and the
nature of the resulting complexes. However, specific bindings
of polyphenols, especially those of higher molecular mass
(e.g., flavanolic proanthocyanidins, pyrogallolic gallotannins,
and ellagitannins) to proteins are unfortunately often difficult
to discriminate from nonspecific bindings because certain pro-
teins can become rapidly heavily coated with these highly hy-
droxylated aromatic compounds (tanning action) during the
course of most analyses. Considering this problematic issue,
we envisioned taking advantage of surface plasmon resonance
(SPR), which enables the detection of molecular interactions in
real time with high sensitivity using very small quantities of
molecular partners in microfluidic devices.[5] This refractometric
optical measurement technique, which is based on the modu-
lation of the resonance of an incoming light wave with an ex-
cited surface plasmon wave from a metallic layer (e.g., gold)
on a glass sensor surface, the refractive index of which is af-
fected by the accumulation of mass through adsorption of
molecules from a circulating aqueous solution,[5] has found
many applications in the analysis of bio(macro)molecular inter-
actions.[5a,6] One of the interacting partners, referred to as the
ligand, is immobilized on the sensor surface, and the other
partner in solution, referred to as the analyte, is eluted over
the surface. In classical SPR experiments involving proteins and
small molecules, the simplest method is to make the protein
the immobilized ligand, since commercial sensor chips are con-
veniently provided with already functionalized surfaces that fa-
cilitate the covalent immobilization of proteins.[7] However,
using polyphenolic molecules as analytes, which are therefore
used in excess in the microfluidic mobile phase, would render
the discrimination between specific and nonspecific interac-
tions with proteins difficult, due to the aforementioned ten-
dency of some polyphenols to invariably stick to most proteins
through multiple binding interaction.
Figure 1. Selected plant polyphenols of different representative structural
classes. Blue circles indicate the carbon sites of attachment of biotin-termi-
nated units to these polyphenolic molecules.
Results and Discussion
Synthesis of SPR-ready biotinylated polyphenolic probes
The selected plant polyphenols belong to different structural
classes and are representatives of the differences in chemical
composition, size, morphology, and conformational freedom
encountered for this large family of natural products.[1g] The
polyhydroxylated terarylic C-glucosidic ellagitannins vescalagin
(1) and vescalin (2) are globular, propellerlike and rather large
and rigid compounds. The di- and triphenolic flavanols cate-
chin (3), epicatechin (4), and epigallocatechin gallate (EGCG, 5)
are small yet rather flexible stereoisomeric molecules, and their
oligomeric proanthocyanidin variants, here represented by the
dimeric procyanidin B2 (6), are large elongated, threadlike mol-
ecules. The diphenolic stilbenes resveratrol (7) and its catechol-
ic variant piceatannol (8) are small, flat, and quite rigid com-
pounds.
This is the reason why the SPR method we developed is in-
stead based on the immobilization of the polyphenol molecule
on the surface of the sensor chip.[2e,8] In such a reverse SPR ex-
periment, the protein becomes the analyte and can be sup-
plied over the sensor chip surface bearing the immobilized
polyphenolic ligand in a continuous-flow mode at low micro-
fluidic concentrations, which could enable the observation of
specific higher-affinity interactions to be emphasized and that
of nonspecific lower-affinity interactions to be limited. Our first
few implementations of this reverse SPR technique met these
expectations by enabling us to unveil the high affinity of the
ellagitannin vescalin for human DNA topoisomerase IIa[8a] and
the preferential interaction of its congener vescalagin with fila-
mentous actin over that with globular actin.[2e] Herein, we
report the results of our work aimed at generalizing the imple-
mentation of this reverse SPR technique for studying polyphe-
nol–protein interactions using several polyphenols of four dif-
ferent structural types (i.e., ellagitannins, flavanols, proantho-
cyanidins, and stilbenoids; Figure 1) and several proteins with
different structural criteria.
Studying the interactions of these polyphenols with proteins
by our reverse SPR method necessitates relying on chemical
synthesis for modifying their structure with a view to their im-
mobilization on the sensor-chip surface. We chose to use
streptavidin-coated sensor chips for immobilizing biotinylated
derivatives of our selected polyphenols by taking advantage of
the strong noncovalent interaction between streptavidin and
biotin (KD ꢀ10À15 m), as shown in Figure 2.[5a] First, a linker had
to be introduced on these polyphenols at specific sites, which
were chosen to limit as much as possible any interference with
the binding to proteins. Therefore, the phenolic hydroxyl
groups were left intact because the binding to proteins is usu-
ally sealed through hydrogen bonding with these groups.[1g]
Instead, only reactive carbon centers served to install the linker
(Figure 1), which was equipped with either a terminal sulfhydr-
yl group or a terminal carboxyl group for subsequent coupling
with either the biotinylated maleimide derivative 9 or the bio-
tinylated polyethylene glycol (PEG) derivative 10 (Figure 2).
This mode of immobilization based on the biotin–streptavi-
din interaction proved to be advantageous over the disulfide-
Chem. Eur. J. 2021, 27, 5498 –5508
www.chemeurj.org
5499
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005187
Chemistry—A European Journal
Scheme 1. Synthesis of the C1-deoxy vescalagin- and vescalin-C1-thioether
sulfhydryl derivatives 1a and 2a.[2e,8a]
Figure 2. Schematic illustrations of the design and synthesis of polyphenol–
biotin conjugates and their binding to streptavidin-coated sensor chips.
catechin (3) can be easily functionalized by oxidation with 2,3-
dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) in the presence
of a wide range of oxygen-based nucleophiles.[14] However,
since sulfhydryl-based nucleophiles are sensitive to oxidation
to disulfides with DDQ, a direct oxidative C4 thioetherification
could not be envisaged. To circumvent this problem, octane-
1,8-dithiol was introduced through nucleophilic substitution of
an allyl ether derivative of 3, which was obtained by oxidative
C4 etherification.[14d] Thus, the phenolic and secondary hydrox-
yl groups of catechin (3) were first protected by silylation with
tert-butyl(chloro)dimethylsilane (TBDMSCl) to furnish 3a, as
previously described.[15] The resulting persilylated catechin de-
rivative was then treated with DDQ in the presence of allyl al-
cohol to furnish allyl ether 3b in 85% yield as a single stereo-
isomer, in agreement with previous related works (Sche-
me 2).[14b,d,e] The nucleophilic substitution reaction of 3b with
octane-1,8-dithiol was promoted by the use of the Lewis acid
BF3·Et2O[14b,d] at low temperature in CH2Cl2 to cleanly afford the
thioether 3c. The stereochemistry at its C3 and C4 centers was
bond exchange-based mode previously used for the immobili-
zation of vescalin (2),[8a] as it is performed in only one step
from the biotinylated polyphenolic conjugates and allows for
better stability over time in comparison to the fragile disulfide
covalent bond. Eight of these nine different sensor-surface-
bound polyphenolic systems were then utilized to examine
their interactions with proteins of varying structures, sizes and
functions: 1) the a isoform of human DNA topoisomerase II
(Top2a, 175 kDa), a nuclear enzyme targeted in anticancer che-
motherapies and inhibited by ellagitannins,[2d,n,r,9] 2) bovine
serum albumin (BSA, 66 kDa), a loosely structured globular pro-
tein serving as a key transporter (e.g., for fatty acids) in the cir-
culatory system and the principal protein model used in nu-
merous studies on polyphenol–protein interactions,[2t,3i,10]
3) myoglobin (Mb, 17 kDa), here used as a model for small and
tight globular metalloproteins, 4) streptavidin (54 kDa), the
biotin-binding tetrameric protein, 5) type I collagen (ca.
300 kDa), a triple-helical fibrillar structural protein, 6) leucoan-
thocyanidin dioxygenase (LDOX, 43 kDa), an enzyme involved
in the biosynthesis of flavonoids,[11] and 7) the B-cell lympho-
ma 2 (Bcl-2) apoptosis regulator protein (27 kDa), an anti-apop-
totic protein known to be downregulated by resveratrol and
thought to be inhibited through direct interactions with other
polyphenols.[12]
1
confirmed to be cis by H NMR analysis, again in agreement
with previous related works.[14b,d] The desilylation of 3c was
not as trivial as expected. In fact, our choice of using silyl pro-
tecting groups instead of benzyl groups, which are classically
used for flavonoid derivatization,[14,16] was based on the risk of
poisoning the metal catalyst with the sulfur-containing groups
of 3c during hydrogenolytic cleavage of the benzyl ether
groups. However, several attempts to cleave the TBDMS ether
groups by using tetra-n-butylammonium fluoride (TBAF),
3HF·Et3N, or KF in the presence of 18-crown-6 were all fruitless,
giving only partially desilylated products, even after extended
reaction times. Fortunately, the removal of all TBDMS groups
could be achieved with (HF)x·pyridine in THF, which furnished
the desired catechin-C4-thioether sulfhydryl derivative 3d in a
moderate but sufficient yield of 37% (Scheme 2).
For vescalagin (1) and its simpler congener vescalin (2), the
syntheses of their corresponding SPR-ready probes started
with the incorporation of a sulfhydryl group mounted onto an
appropriate linker. Octane-1,8-dithiol was chosen for this pur-
pose, and its installation was performed in only one step with-
out any prior protection of 1 or 2 by taking advantage of the
remarkable chemo- and stereoselective reactivity expressed at
their C1 center.[2n,13] As previously reported,[2e,8a] the desired thi-
oether sulfhydryl C1-deoxy derivatives 1a and 2a were both
obtained by acid-catalyzed nucleophilic substitution reactions
in good yields (Scheme 1).
For the derivatization of catechin (3) and epicatechin (4)
through the A ring, the greater inherent nucleophilic character
of their C8 centers[17] was exploited to install a propanoic acid
tether, by following a five-step reaction sequence that we pre-
viously described.[18] The resulting carboxylic acids 3 f and
4a[18] were initially intended to be converted to thioester deriv-
The analogous derivatization of catechin (3) and epicatechin
(4) was envisioned by installing a sulfhydryl-terminated linker
at their C4 or C8 centers (Figure 1). The C4 benzylic position of
Chem. Eur. J. 2021, 27, 5498 –5508
www.chemeurj.org
5500
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005187
Chemistry—A European Journal
for the total synthesis of 5[19] to meet our objective. Inspired
by Chan and Li’s enantioselective synthesis of (À)-5,[19a] we
converted 3,4,5-tribenzyloxycinnamyl alcohol and 3,5-dibenzyl-
oxyphenol to the pentabenzylated epigallocatechin 5a, which
was obtained as a mixture of enantiomers (the major isomer
was the pentabenzylated (À)-(2R,3R)-cis-epigallocatechin, ee=
42%) in seven steps and 31% overall yield from the starting
cinnamyl alcohol.[19a,20] This poorly enantioenriched 5a was for-
mylated at its C8 center by a Vilsmeier–Haack reaction
(Scheme 3), followed by a short methanolysis to cleave the for-
miate group concomitantly installed on the secondary hydrox-
yl group at C3.
Scheme 2. Synthesis of the catechin-C4-thioether sulfhydryl derivative 3d
and of the catechin- and epicatechin-C8-propanamido sulfhydryl derivatives
3h and 4c.
atives by using octane-1,8-dithiol under standard Steglich-type
esterification conditions in the presence of 1-ethyl-3-(3-dimeth-
ylaminopropyl) carbodiimide (EDCI). However, a d-lactonization
involving the carboxyl group and the phenolic hydroxyl group
at the C7 center of 3 f (see 3g in Scheme 2) prevailed over the
expected thioesterification. An attempt to proceed with a lac-
tone-opening transesterification using octane-1,8-dithiol was
also inoperative. Performing this reaction under basic condi-
tions to enhance the nucleophilicity of octane-1,8-dithiol as
thiolate anion was not envisaged because of the risk of gener-
ating phenolate anions, which are prone to cause autoxidative
degradation of such polyphenolic substances. Therefore, we
decided to use another aliphatic-type sulfhydryl linker unit
equipped with a primary amino group, and we opted for cys-
teamine. The carboxylic acids 3 f and 4a were then each al-
lowed to react with the EDCI coupling agent in CH3CN at 508C
until complete formation of the corresponding lactones 3g
and 4b, as monitored by TLC. Cysteamine was added directly
to the reaction mixtures and rapidly promoted opening of
these lactonic flavanoids to give rise to the formation of the
desired catechin- and epicatechin-C8-propanamido sulfhydryl
derivatives 3h and 4c, which were isolated in 62 and 36%
yield, respectively (Scheme 2).
Scheme 3. Synthesis of the epigallocatechin gallate-C8-propanoic acid deriv-
ative 5e.
The resulting aldehyde 5b was thus obtained in 70% yield,
and was then converted to the a,b-unsaturated ester 5c in
77% yield thanks to a Doebner–Knoevenagel reaction with
monobenzyl malonate in the presence of 4-(dimethylamino)-
pyridine (DMAP) and piperidine catalysts in DMF.[21] At this
stage we separated the major enantiomer from our working
mixture, and this purification was performed by semiprepara-
tive chiral HPLC to furnish enantiopure (2R,3R)-cis-epigallocate-
chin derivative (À)-5c in 50% yield from the mixture 5b (see
the Supporting Information for details). This compound was
then galloylated at its C3 position in high yield with 3,4,5-tri-
benzyloxybenzoic acid under Steglich-type conditions by using
EDCI hydrochloride as coupling agent,[19d] and the resulting
gallate 5d was finally subjected to classical Pd-catalyzed hy-
drogenation conditions to promote the reduction of its olefinic
bond, as well as the removal of its nine benzyl groups, to
quantitatively afford the desired C8-propanoic acid epigalloca-
techin gallate derivative (À)-5e (Scheme 3). The same lactoni-
zation as for 3 f and 4a (Scheme 2) was followed by the addi-
tion of cysteamine to generate the expected analogous sulf-
hydryl derivative, but complications arose during the purifica-
The modification of epigallocatechin gallate (5) was next
considered by adopting the same strategy as for the C8 deriva-
tization of (epi)catechins 3 and 4. Starting from commercially
available (À)-5 was not viewed as a convenient option, be-
cause this natural flavanol enantiomer is rather expensive and
not fully adapted to our derivatization protocol. Therefore, a
complete chemical synthesis of the required derivative of 5
was carried out by adapting some of the methods developed
Chem. Eur. J. 2021, 27, 5498 –5508
www.chemeurj.org
5501
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005187
Chemistry—A European Journal
tion of this coupling product. Therefore, we opted for another
solution that involved the amidation of 5e with the amino-
PEG/biotin derivative 10 (see Figure 2 and Scheme 6).
The derivatization of procyanidin-B2 (6) was also chosen to
take place at its available A-ring C8 center. Again, a complete
chemical synthesis of the required derivative of 6 was carried
out (Scheme 4). The procyanidin core was elaborated by fol-
found to be particularly sensitive to oxidation to the disulfide,
was attempted, and it was instead directly engaged in a conju-
gate nucleophilic addition reaction with the maleimide/biotin
derivative 9 (see Figure 2 and Scheme 6).
Finally, resveratrol (7) and its natural catecholic analogue, pi-
ceatannol (8), were also equipped with a propanamido sulf-
hydryl moiety at their aromatic C2 centers (Scheme 5). Without
Scheme 5. Synthesis of resveratrol- and piceatannol-C2-propanamido sulf-
hydryl derivatives 7c and 8b.
any protection of its phenolic hydroxyl groups, 7 was regiose-
lectively formylated by a Vilsmeier–Haack reaction in high
yield, as previously reported.[22] The resulting aldehyde 7a was
this time treated with Meldrum’s acid in the presence of tri-
ethylammonium formiate to furnish, after acidic treatment, the
desired lactone 7b in 51% yield.[23] This lactone was subjected
to our SIBX-mediated phenolic ortho-hydroxylation conditions
to generate the corresponding piceatannol derivative 8a in
high yield.[24] Lactones 7b and 8a were then each allowed to
react with cysteamine to deliver the expected resveratrol- and
piceatannol-C2-propanamido sulfhydryl derivatives 7c and 8b
in reasonable yields (Scheme 5).
Scheme 4. Synthesis of the procyanidin B2-C8-propanamido sulfhydryl deriv-
ative 6c.
lowing the procedures described by Suzuki, Ohmori, and co-
workers.[14b,16c] The pentabenzylated epicatechin 4e[18] was first
acetoxylated at its benzylic C4 center by oxidation with DDQ
in a mixture of acetic acid and CH2Cl2 to give the b-acetoxy de-
rivative 4 f in 41% yield as the sole diastereomer.[14b] The Lewis
acid activation of 4 f with BF3·OEt2 in the presence of a three-
fold excess of 4e[16c] led, after 23 h at low temperature, to the
formation of the expected flavonoid dimer 6a as the sole ster-
eodimer in 80% yield. Full NMR analysis confirmed both the
C4-C8 regioselectivity and the b orientation of the interflavan
bond, in agreement with the literature data.[14a,16d–f] Moreover,
the examination of the NMR spectra of 6a in [D6]acetone and
complementary NOESY data indicated that this dimer is a mix-
ture of two rotamers I and II in a ratio of approximately 85:15
(see the Supporting Information).[14a,16d–f] With this protected
procyanidin B2 6a in hand, we installed a propanoic acid
tether at its available C8 center by following a sequence of re-
actions similar to that used for the derivatization of epigalloca-
techin gallate (5), that is, Vilsmeier–Haack formylation, Doeb-
ner–Knoevenagel olefination, and hydrogenation; see
Scheme 4 and the Supporting Information for experimental de-
tails. The resulting modified procyanidin B2 6b was then
lactonized under Steglich-type conditions in DMF and then al-
lowed to react with cysteamine (Scheme 4). In this case, no pu-
rification of the expected sulfhydryl derivative 6c, which was
All of the sulfhydryl derivatives 1a, 2a, 3d, 3h, 4c, 6c, 7c,
and 8b were each rapidly coupled by Michael addition to the
maleimide/biotin unit 9[2e,25] to give the required SPR-ready
biotinylated polyphenolic probes in good to high yields (see
Figure 2 and Scheme 6). The low solubility of the polyphenolic
sulfhydryl derivatives in standard organic solvents led us to
perform these coupling reactions in [D6]DMSO, which conve-
niently enabled us to monitor the progress of the reactions by
1H NMR analysis. The resulting products were precipitated
from the [D6]DMSO solution by addition of mixtures of Et2O
and CHCl3, and then, if deemed necessary, further purified by
semipreparative HPLC, to afford the biotinylated polyphenolic
probes 1b, 2b, 3e, 3i, 4d, 6d, 7d, and 8c in yields ranging
from 67 to 100% (Scheme 6 and see the Supporting Informa-
tion for details). As mentioned above, the biotinylated EGCG
probe was instead finally built by lactonizing the carboxylic
acid derivative 5e (see Scheme 3), directly followed by open-
ing of this lactone by using the amino-PEG/biotin derivative 10
(see Figure 2 and Scheme 6). The low reactivity of the lactone
Chem. Eur. J. 2021, 27, 5498 –5508
www.chemeurj.org
5502
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005187
Chemistry—A European Journal
given protein with a given polyphenol could be measured in
such a way, for example, in the cases of the nanomolar binding
of EGCG with the metastasis-associated laminin receptor,[2o]
and of the submicromolar binding of EGCG to the signal trans-
ducer and activator of transcription-1 (STAT1) protein[26e] and
to the protein phosphatase-1.[26f] In most cases, however, the
discrimination between specific and nonspecific polyphenol–
protein interactions was less evident,[26a–d,h,i] likely due to the
aforementioned tendency of certain polyphenols to form pre-
cipitating complexes with some proteins (tanning action), and
even more so when solutions of polyphenolic mixtures (e.g.,
tannin extracts) are used as analytes.[26i]
Far from pretending that our reverse SPR technique consti-
tutes a panacea for the analysis of polyphenol–protein interac-
tions, it offers an alternative that enables a rapid qualitative
and comparative screening of the interactions of a selection of
proteins with several adequately modified (and pure) polyphe-
nols. Our biotinylated polyphenolic probes were immobilized
on streptavidin-coated sensor surfaces (see Figure 2). Each
sensor chip was divided into four separate flow cells, one of
which was kept blank to serve as a control surface, whereas
the three other flow cells were used to immobilize three differ-
ent polyphenolic probes. These three probes per sensor chip
were selected on the basis of the structural class of the poly-
phenol they bear and the type of proteins to be tested, so we
decided to leave the piceatannol-bearing probe 8c aside from
this experimental development. The following four sensor
chips A–D were prepared: A (1b, 2b, 3e), B (3e, 3i, 4d), C
(1b, 3i, 7d), and D (5 f, 6d, 7d). Thus, this multi-analysis SPR
setup enabled us to study in real time the behavior of a given
protein towards three different polyphenols in a single experi-
ment run, and to screen the interaction of other proteins with
the same three polyphenols in subsequent runs. Equimolar
quantities of each polyphenolic probe were immobilized over
the three different flow cells of the same sensor chip. Briefly,
500 mm stock solutions of the biotinylated polyphenolic probes
were prepared by dissolving them in DMSO, and then diluted
to 25–50 nm concentrations with the SPR aqueous running
buffer (50 mm sodium phosphate, pH 7, 150 mm NaCl, 0.005%
of surfactant Tween-20). These ligand solutions were then in-
jected at a flow rate of 5 mLminÀ1 over the flow cell surface
until the desired level of immobilization, which is expressed in
resonance units (RU), was obtained (see the Supporting Infor-
mation for details). The SPR protein binding assays were next
performed (vide infra) by injecting the protein solutions in in-
creasing order of concentration, without any regeneration of
the sensor surfaces between injections (i.e., single-cycle kinet-
ics).[27] This modus operandi offers the advantage of allowing
the detection of interactions at different protein concentra-
tions, without any risk of damaging the sensor surface by
using a potentially ineffective regenerating agent. Thus, each
of the protein solutions of increasing concentration was inject-
ed over a period of about 3 min (association phase), which
was followed by a period of about 10 min to observe the dis-
sociation of the polyphenol–protein complex. At the end of
these runs, if no remaining complex was observed, the sensor
chip surface was simply rinsed by three injections of running
Scheme 6. Final step(s) of the elaboration of nine SPR-ready biotinylated
polyphenolic probes.
and tedious purification of the amide product 5 f enabled us
to generate this biotinylated EGCG probe in only 15% yield
(Scheme 6), but in largely sufficient quantities to carry out sev-
eral SPR experiments.
Real-time SPR analysis of polyphenol-protein interactions
The classical manner of running SPR experiments by immobiliz-
ing proteins on sensor surfaces has been implemented in a
few studies aimed at examining the interaction(s) of a given
protein with a particular polyphenol or a selection thereof.[2o,26]
In some rare cases, the specific high-affinity interaction of a
Chem. Eur. J. 2021, 27, 5498 –5508
www.chemeurj.org
5503
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005187
Chemistry—A European Journal
buffer, and a solution of a different protein could be injected.
However, in cases in which some remaining polyphenol–pro-
tein complex was still observed, two or three injection pulses
of 20 mL of a 0.05% solution of sodium dodecyl sulfate (SDS)
were carried out at a flow rate of 20 mLminÀ1 to regenerate
the sensor surface (see the Supporting Information for details).
SDS was the best surface regenerant we could find for these
polyphenol–protein complexes, but we observed that some-
how it altered the functionalized surfaces. Despite our efforts,
it was very difficult, if not impossible, to obtain perfectly super-
imposed sensorgrams for repeated binding cycles (e.g., see
Figure 3). However, the SPR responses followed the same
trend and therefore do not alter the conclusions drawn from
this work.
sensorgrams showed that, at best, the level of SPR responses
very slowly decreases during the dissociation phase of the SPR
experiment, and hence indicate that Top2a remains associated
with these two ellagitannins. These observations are in sharp
contrast with those made in the case of the catechin-bearing
probe 3e, for which no interaction was detected with Top2a
(see Figure 3c).
These first results were already qualitatively significant, as
they demonstrate that this reverse SPR method is capable of
discriminating the interactions, or lack thereof, between poly-
phenols of different structural classes and a given protein. The
SPR responses observed for both Top2a–vescalin and Top2a–
vescalagin interactions do not allow one to determine unam-
biguously which of these two ellagitannins is the best Top2a
ligand. Nevertheless, these responses are in accordance with
our previous results on the inhibition of Top2a-mediated deca-
tenation of kinetoplast DNA by these ellagitannins[2d,n] and on
the identification of vescalagin as a preferential catalytic inhibi-
tor of the a-isoform of Top2 both in vitro and in cellulo.[2d] The
relatively low level of SPR signals observed for the vescal(ag)in
probe systems relative to a 1:1 model of interaction with
Top2a suggests either that only a small amount of the protein
could approach the immobilized polyphenols 1b and 2b, per-
haps due to some steric impediment of the access of the pro-
tein analyte in the microfluidic mobile phase, and/or that the
binding mode is complex. Therefore, the recorded sensor-
grams do not fit a 1:1 interaction model, enjoining us from cal-
culating KD values for these ellagitannin–Top2a binding sys-
tems.
The significant differences in interaction behaviors between
the vescal(ag)in probes 1b/2b and the catechin probe 3e to-
wards Top2a could be simply attributed to the fact that the
catechin unit of 3e has a much lower number of hydroxyl
groups and aromatic rings, which confer combined hydrophilic
and hydrophobic characters to polyphenols in their interac-
tions with proteins. This possibility would suggest that the in-
teractions observed by SPR for the vescal(ag)in probes 1b/2b
would be of a nonspecific nature and should thus also be ob-
served regardless of the tertiary structure or size of the protein
involved in such interactions. To exclude such a possibility,
sensor chip A was subjected to successive injections of solu-
tions of proteins of different sizes and structures. No SPR re-
sponse was observed after the injection of solutions of BSA,
myoglobin, streptavidin, or type I collagen in the same concen-
tration range (see the Supporting Information, Figure S1).
These absences of SPR signals indicate that these proteins do
not interact with the three polyphenols mounted on this first
series of probes, and that the differences in behavior observed
between these proteins, including Top2a, and the polyphenol-
ic probes 1b, 2b, and 3e are not simply due to their number
of phenolic groups, but to some privileged polyphenol–protein
binding relationships (or lack thereof).
Figure 3. Chip-A sensorgrams recorded on injections of Top2a over a) the
vescalagin-bearing probe 1b, b) the vescalin-bearing probe 2b, and c) the
catechin-bearing probe 3e.
The first protein binding assays were performed with sensor
chip A and solutions of human DNA topoisomerase 2a (Top2a)
by using a commercial aqueous stock solution, which was di-
luted to the desired 6.25, 12.5, and 25 nm concentrations with
the running buffer. The recorded sensorgrams are shown in
Figure 3. The SPR responses increased in a dose-dependent
manner when Top2a solutions were injected over immobilized
vescalagin- and vescalin-bearing probes 1b and 2b (see Fig-
ure 3a and b). After each injection of the Top2a solutions, the
At higher concentrations (i.e., 125, 250, 500 nm), type I colla-
gen started to behave more like Top2a (see Figure 3a and b)
in a dose-dependent manner with rather rapid associations
and slow dissociations with/from the vescalagin- and vescalin-
bearing probes 1b and 2b (Figure 4a). Again, no major differ-
Chem. Eur. J. 2021, 27, 5498 –5508
www.chemeurj.org
5504
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005187
Chemistry—A European Journal
Figure 4. Chip-A sensorgrams recorded on injections of a) type I collagen
and b) streptavidin at 125–500 nm concentration.
ence in the level of SPR responses with 1b and 2b was ob-
served, and still no distinctive signal was observed with the
catechin-bearing probe 3e. BSA and myoglobin still did not
show any evidence of binding to these polyphenolic probes
(see the Supporting Information, Figure S2), but very weak SPR
responses with fast dissociations could be detected for strepta-
vidin interacting with the vescalin-bearing probe 2b and with
the catechin-bearing probe 3e (Figure 4b).
Figure 5. a) Chip-B sensorgrams recorded on injections of LDOX at 200–
800 nm concentrations. b) Chip-C and c) chip-D sensorgrams recorded on in-
jections of Bcl-2 at 250–1000 nm concentrations.
Thus, at these higher concentrations, the elongated fibrillar
type I collagen protein showed a clear preference for binding
to the ellagitannins vescal(ag)ins (see Figure 4a), whereas the
globular homotetrameric streptavidin protein could appear to
slightly prefer to bind to the flavanol catechin (see Figure 4b),
although such an interpretation of the very weak signals ob-
served remains subject to caution. Notwithstanding these con-
cerns, the observations made again demonstrate the value of
this SPR technique to rapidly examine and qualitatively distin-
guish the interactions of different polyphenols with different
proteins.
ed,[18,28] whereas the same probe gave very weak or no SPR
binding signals with Top2a, BSA, collagen, myoglobin, and
streptavidin (see Figures 3 and 4, as well as the Supporting In-
formation, Figures S1 and S2).
The sensor chips C and D, which are equipped with probes
bearing polyphenols of different structural types (i.e., the ella-
gitannin vescalagin, the stilbenoid resveratrol, the flavanoids
catechin and EGCG, and the dimeric procyanidin B2), were
used to examine interactions with Bcl-2. The SPR responses ob-
tained on injecting solutions of Bcl-2 at concentrations of 250,
500, and 1000 nm (see Figures 5b and c) revealed that Bcl-2
binds to the vescalagin moiety of probe 1b in a manner simi-
lar to that of the interactions of 1b with other proteins, which
are characterized by rather rapid associations and slow dissoci-
ations. The small increase of the SPR signal between Bcl-2 con-
centrations of 500 and 1000 nm would indicate that this inter-
action reaches a level of saturation around 1000 nm (Fig-
ure 5b).
Sensor chip B, which is equipped with the C4-linked cate-
chin probe 3e and the C8-linked (epi)catechin probes 3i and
4d, was used to examine interactions with the LDOX flavonoid
enzyme in solutions at concentrations of 200, 400, and 800 nm
(Figure 5a). The resulting SPR responses increased in a dose-
dependent manner, which clearly revealed binding of LDOX to
these immobilized flavanols in agreement with the results of
our previous work on affinity chromatography and affinity-
based chemoproteomic capture of LDOX by these flava-
nols.[18,28]
The SPR responses obtained with the resveratrol probe 7d
increased in a dose-dependent manner and are characterized
by both very fast association and dissociation phases. Such a
touch-and-go interaction between Bcl-2 and resveratrol was
similarly observed on both sensor chips C and D (Figure 5b
and c). The SPR monitoring of the interaction between Bcl-2
and the C8-linked catechin probe 3i was more erratic when
comparing signals obtained at 500 and 1000 nm, but again
All three interactions are again globally characterized by
rather rapid associations and slow dissociations. These last re-
sults further confirm that this SPR technique can be used to
unveil structure–protein binding relationships with our poly-
phenolic probes, since the interaction of the LDOX flavonoid
enzyme with the catechin probe 3e was expectedly detect-
Chem. Eur. J. 2021, 27, 5498 –5508
www.chemeurj.org
5505
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005187
Chemistry—A European Journal
this interaction is characterized by very fast association and
dissociation phases (Figure 5b). Finally, as compared with res-
veratrol, the two larger and more highly hydroxylated polyphe-
nols EGCG and procyanidin B2 exhibited a behavior more simi-
lar to that of vescalagin in their interaction with Bcl-2 with a
rather fast association phase and a slow dissociation phase.
This is even more the case for the EGCG probe 5 f, for which
the interaction with Bcl-2 appears to reach saturation around
500 nm, in contrast to the case of the procyanidin B2 probe
6d, for which the SPR response seems to be still dose-depen-
dent between Bcl-2 concentrations of 500 and 1000 nm (Fig-
ure 5c). It is also noteworthy that the injections of BSA on
these sensor chips C and D again revealed no significant bind-
ing with these five polyphenolic probes in this same range of
concentrations for BSA (see the Supporting Information, Fig-
ure S3).
tions to this toolbox of immobilizable ligands will certainly find
numerous and valuable applications, not only for screening
polyphenol–protein interactions by SPR, but also for identify-
ing cellular proteins targeted by bioactive polyphenols or
plant proteins involved in their biosynthesis through the im-
plementation of affinity chromatographic techniques or more
modern affinity-based mass spectrometry-aided (chemo)pro-
teomic protocols.[28]
Acknowledgements
This work was supported by institutional financial supports
from the Institut Universitaire de France (IUF), the University of
Bordeaux and the CNRS, and research grants from the Agence
Nationale de la Recherche (ANR-17-CE11-0033-01 PRC POLY-
PHENOLPROT and ANR-16-CE91-0006-01 PRCI HIGHERFLAVO).
The authors wish to thank Fundayacucho and the French Em-
bassy in Caracas, Venezuela, for D.M.D.L.’s doctoral research as-
sistantship, the government of Vietnam for D.T.T.’s doctoral re-
search assistantship, the University of Bordeaux for G.V.’s post-
doctoral fellowship (ATER), the Erasmus Mundus Action 2
GreenIT for S.D.’s post-doctoral fellowship, and the ANR for
Y.C.’s doctoral research assistantship. The authors also grateful-
ly acknowledge the contributions of Thierry Dakhli (IECB), Sab-
rina Rousseau (IECB) and HØl›ne CarriØ (S.Q.’s research team)
for the production of the LDOX protein, and the verification of
its activity prior to the SPR experiments. This work has benefit-
ed from the facilities and expertise of the Structural Biophysi-
co-Chemistry platform at IECB, CNRS UMS3033, INSERM US001,
and the University of Bordeaux.
Conclusion
The series of biotinylated polyphenolic probes we have synthe-
sized using polyphenols of different structural classes [i.e., the
ellagitannins vescalagin (1) and vescalin (2), the flavanols cate-
chin (3), epicatechin (4) and epigallocatechin gallate (5), the di-
meric flavanoid procyanidin B2 (6) and the stilbenoids resvera-
trol (7) and piceatannol (8)] constitutes a useful toolbox of SPR
ligands to examine qualitatively their interactions with various
proteins in real time. Even though the recorded sensorgrams
did not allow us to determine the dissociation equilibrium con-
stants for these polyphenol–protein complexes, our multi-anal-
ysis reverse SPR setup enabled the rapid and convenient com-
parative analysis of the interaction behavior (i.e., type and
level of binding) between a given protein analyte and different
immobilized polyphenolic ligands, as well as between a given
polyphenolic ligand and different protein analytes. Thus, privi-
leged and dose-dependent interactions between the two ella-
gitannins vescalagin (1) and vescalin (2) and human DNA top-
oisomerase IIa (Top2a) were revealed by SPR, in accordance
with previous results on their inhibitory effect on Top2a.[2d,n] At
higher concentrations of proteins (>100 nm), type I collagen
also binds to the same ellagitannins, but streptavidin, myoglo-
bin, and BSA do not significantly interact either with these ella-
gitannins or with catechin (3). However, this flavanol 3 and its
epimer epicatechin (4) expectedly gave SPR responses in a
dose-dependent manner on interacting with the flavonoid
enzyme LDOX. The SPR responses of the interactions between
the small globular Bcl-2 apoptosis regulator protein and five
different polyphenolic ligands [i.e., the ellagitannin vescalagin
(1), the flavonoids catechin (3), epigallocatechin gallate (5),
procyanidin B2 (6) and the stilbenoid resveratrol (7)] revealed
different behaviors, such as touch-and-go interactions with the
smaller polyphenols 3 and 7, and interactions characterized by
rather rapid associations and slower dissociations with the
larger and more highly hydroxylated polyphenols 1, 5, and 6.
Of particular note is that BSA, the protein standard in studies
of polyphenol–protein interactions, gave no significant SPR re-
sponse with all of the polyphenolic ligands tested herein.
These biotinylated polyphenolic entities and any future addi-
Conflict of interest
The authors declare no conflict of interest.
Keywords: analytical methods · immobilization · polyphenols ·
proteins · surface plasmon resonance
[1] a) R. PØrez-Gregorio, S. Soares, N. Mateus, V. de Freitas, Molecules 2020,
25, 3443; b) Q. Zhou, H. Pan, J. Li, Cancers 2019, 11, 1565; c) A. B. San-
thakumar, M. Battino, J. M. Alvarez-Suarez, Food Chem. Toxicol. 2018,
113, 49–65; d) K. Goszcz, G. G. Duthie, D. Stewart, S. J. Leslie, I. L.
Megson, British J. Pharmacol. 2017, 174, 1209–1225; e) M. Agrawal,
Drug Metab. Rev. 2015, 47, 420–430; f) T. Jayasena, A. Poljak, G. Smythe,
N. Braidy, G. Münch, P. Sachdev, Ageing Res. Rev. 2013, 12, 867–883;
g) S. Quideau, D. Deffieux, C. Douat-Casassus, L. PouysØgu, Angew.
Chem. Int. Ed. 2011, 50, 586–621; Angew. Chem. 2011, 123, 610–646;
h) M. Asensi, A. Ortega, S. Mena, F. Feddi, J. M. Estrela, Crit. Rev. Clin.
Lab. Sci. 2011, 48, 197–216; i) A. Crozier, I. B. Jaganath, M. N. Clifford,
Nat. Prod. Rep. 2009, 26, 1001–1043; j) E. Middleton, Jr., C. Kandaswami,
T. C. Theoharides, Pharmacol. Rev. 2000, 52, 673–751; k) M. Zhu, J. D.
Philipson, P. M. Greengrass, N. E. Bowery, Y. Cai, Phytochemistry 1997, 44,
441–447; l) E. Haslam, J. Nat. Prod. 1996, 59, 205–215.
[2] For a selection of articles on high-affinity protein-specific interactions
with plant polyphenols, see: a) D. Georgess, P. Spuul, C. Le Clainche, D.
Le Nihouannen, I. Fremaux, T. Dakhli, D. M. Delannoy López, D. Deffieux,
P. Jurdic, S. Quideau, E. GØnot, Eur. J. Cell Biol. 2018, 97, 533–545; b) S.-
W. Chao, M.-Y. Su, L.-C. Chiou, L.-C. Chen, C.-I. Chang, W.-J. Huang, J.
Nat. Prod. 2015, 78, 1969–1976; c) C. W. A. Moyle, A. B. Cerezo, M. S.
Winterbone, W. J. Hollands, Y. Alexeev, P. W. Needs, P. A. Kroon, Mol.
Chem. Eur. J. 2021, 27, 5498 –5508
www.chemeurj.org
5506
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005187
Chemistry—A European Journal
Nutr. Food Res. 2015, 59, 401–412; d) C. Auzanneau, D. Montaudon, R.
Jacquet, L. PouysØgu, D. Deffieux, S. Puyo, A. Elkaoukabi-Chaibi, F. De
Giorgi, F. Ichas, S. Quideau, P. Pourquier, Mol. Pharmacol. 2012, 82, 134–
141; e) S. Quideau, C. Douat-Casassus, D. M. Delannoy López, C. Di Pri-
mo, S. Chassaing, R. Jacquet, F. Saltel, E. GØnot, Angew. Chem. Int. Ed.
2011, 50, 5099–5104; Angew. Chem. 2011, 123, 5205–5210; f) R. L.
Wiseman, Y. Zhang, K. P. K. Lee, H. P. Harding, C. M. Haynes, J. Price, F. Si-
cheri, D. Ron, Mol. Cell 2010, 38, 291–304; g) T. Kumamoto, M. Fujii, D.-
X. Hou, Cancer Lett. 2009, 275, 17–26; h) T. Kumamoto, M. Fujii, D.-X.
Hou, Mol. Cell. Biochem. 2009, 332, 33–41; i) J.-H. Shim, H. S. Choi, A. Pu-
gliese, S.-Y. Lee, J.-I. Chae, B. Y. Choi, A. M. Bode, Z. Dong, J. Biol. Chem.
2008, 283, 28370–28379; j) E. J. Yearley, E. A. Zhurova, V. V. Zhurov, A. A.
Pinkerton, J. Am. Chem. Soc. 2007, 129, 15013–15021; k) J. R. Gledhill,
M. G. Montgomery, A. G. W. Leslie, J. E. Walker, Proc. Natl. Acad. Sci. USA
2007, 104, 13632–13637; l) J.-Y. Jo, E. Gonzalez de Mejia, M. A. Lila, J.
Agric. Food Chem. 2006, 54, 2083–2087; m) D. J. Maloney, J.-Z. Deng,
S. R. Starck, Z. Gao, S. M. Hecht, J. Am. Chem. Soc. 2005, 127, 4140–
4141; n) S. Quideau, M. Jourdes, D. Lefeuvre, D. Montaudon, C. Saucier,
Y. Glories, P. Pardon, P. Pourquier, Chem. Eur. J. 2005, 11, 6503–6513;
o) H. Tachibana, K. Koga, Y. Fujimura, K. Yamada, Nat. Struct. Mol. Biol.
2004, 11, 380–381; p) J. Zheng, V. D. Ramirez, British J. Pharmacol.
2000, 130, 1115–1123; q) G. M. Polya, B. H. Wang, L. Y. Foo, Phytochemis-
try 1995, 38, 307–314; r) Y. Kashiwada, G. Nonaka, I. Nishioka, K. J. Lee,
I. Bori, Y. Fukushima, K. F. Bastow, K. H. Lee, J. Pharm. Sci. 1993, 82, 487–
492; s) Y. Kashiwada, G.-i. Nonaka, I. Nishioka, L. B. Ballas, J. B. Jiang,
W. P. Janzen, K.-H. Lee, Bioorg. Med. Chem. Lett. 1992, 2, 239–244;
t) A. E. Hagerman, L. G. Butler, J. Biol. Chem. 1981, 256, 4494–4497.
[3] For a selection of articles on NMR-, ITC-, MS- and FTIR-based analytical
techniques applied to the study of polyphenol-protein interactions
over the years, see: a) N. J. Murray, M. P. Williamson, T. H. Lilley, E.
Haslam, Eur. J. Biochem. 1994, 219, 923–935; b) N. J. Baxter, T. H. Lilley,
E. Haslam, M. P. Williamson, Biochemistry 1997, 36, 5566–5577; c) A. J.
Charlton, E. Haslam, M. P. Williamson, J. Am. Chem. Soc. 2002, 124,
9899–9905; d) C. Simon, K. Barathieu, M. Laguerre, J.-M. Schmitter, E.
Fouquet, I. Pianet, E. J. Dufourc, Biochemistry 2003, 42, 10385–10395;
e) R. R. Reddy, B. V. N. Phani Kumar, G. Shanmugam, B. Madhan, A. B.
Mandal, J. Phys. Chem. B 2015, 119, 14076–14085; f) S. Soares, E. Bran-
d¼o, I. García-EstØvez, F. Fonseca, C. Guerreiro, F. Ferreira da Silva, N.
Mateus, D. Deffieux, S. Quideau, V. de Freitas, J. Agric. Food Chem. 2019,
67, 9579–9590; g) J. P. McManus, K. G. Davis, J. E. Beart, S. H. Gaffney,
T. H. Lilley, E. Haslam, J. Chem. Soc. Perkin Trans. 2 1985, 1429–1438;
h) R. A. Frazier, A. Papadopoulou, I. Mueller-Harvey, D. Kissoon, R. J.
Green, J. Agric. Food Chem. 2003, 51, 5189–5195; i) E. R. Deaville, R. J.
Green, I. Mueller-Harvey, I. Willoughby, R. Frazier, J. Agric. Food Chem.
2007, 55, 4554–4561; j) S. V. E. Prigent, A. G. J. Voragen, G. A. van Ko-
ningsveld, A. Baron, C. M. G. C. Renard, H. Gruppen, J. Dairy Sci. 2009,
92, 5843–5853; k) S. Pal, C. Saha, M. Hossain, S. K. Dey, G. S. Kumar,
PLoS One 2012, 7, e43321; l) M. Karonen, M. Oraviita, I. Mueller-Harvey,
J.-P. Salminen, R. J. Green, J. Agric. Food Chem. 2015, 63, 10647–10654;
m) M. Karonen, M. Oraviita, I. Mueller-Harvey, J.-P. Salminen, R. J. Green,
J. Agric. Food Chem. 2019, 67, 12730–12740; n) S. VergØ, T. Richard, S.
Moreau, S. Richelme-David, J. Vercauteren, J.-C. PromØ, J.-P. Monti, Tetra-
hedron Lett. 2002, 43, 2363–2366; o) F. Canon, R. Ballivian, F. Chirot, R.
Antoine, P. Sarni-Manchado, J. Lemoine, P. Dugourd, J. Am. Chem. Soc.
2011, 133, 7847–7852; p) F. Canon, A. R. Milosavljevic, G. van der Rest,
M. RØfrØgiers, L. Nahon, P. Sarni-Manchado, V. Cheynier, A. Giuliani,
Angew. Chem. Int. Ed. 2013, 52, 8377–8381; Angew. Chem. 2013, 125,
8535–8539; q) M. R. Perez-Gregorio, N. Mateus, V. de Freitas, J. Agric.
Food Chem. 2014, 62, 10038–10045; r) B. Plet, A. Delcambre, S.
Chaignepain, J.-M. Schmitter, Tetrahedron 2015, 71, 3007–3011; s) M. T.
Engstrçm, X. Sun, M. P. Suber, M. Li, J.-P. Salminen, A. E. Hagerman, J.
Agric. Food Chem. 2016, 64, 8994–9003; t) A. Edelmann, B. Lendl, J. Am.
Chem. Soc. 2002, 124, 14741–14747.
1596–1603; e) Q. Li, Q. Wei, E. Yuan, J. Yang, Z. Ning, Int. J. Food Sci.
Technol. 2014, 49, 1063–1069; f) J.-h. Shi, J. Wang, Y.-Y. Zhu, J. Chen, Lu-
minescence 2014, 29, 522–530; g) S. Sekowski, M. Ionov, N. Abdulladja-
nova, R. Makhmudov, S. Mavlyanov, K. Milowska, M. Bryszewska, M. Za-
maraeva, Colloids Surf. B 2017, 155, 159–165; h) T.-N. Nguyen, E. Du-
breucq, V. Perrier, Q.-H. Tran, C. Charpentier, C. Charnay, F. Terki, C. Jay-
Allemand, L. P. R. Bidel, Food Chem. 2020, 318, 126482; i) C. Pascal, C.
Poncet-Legrand, A. Imberty, C. Gautier, P. Sarni-Manchado, V. Cheynier,
A. Vernhet, J. Agric. Food Chem. 2007, 55, 4895–4901; j) E. Jçbstl, J.
O’Connell, J. P. A. Fairclough, M. P. Williamson, Biomacromolecules 2004,
5, 942–949; k) C. Pascal, C. Poncet-Legrand, B. Cabane, A. Vernhet, J.
Agric. Food Chem. 2008, 56, 6724–6732; l) C. Shi, H. Tang, J. Xiao, F. Cui,
K. Yang, J. Li, Q. Zhao, Q. Huang, Y. Li, J. Agric. Food Chem. 2017, 65,
656–665; m) V. de Freitas, N. Mateus, J. Agric. Food Chem. 2001, 49,
940–945; n) E. Carvalho, N. Mateus, V. de Freitas, Anal. Chim. Acta 2004,
513, 97–101; o) S. Das, S. Sarmah, Z. Hazarika, M. Ataur Rohman, P. Sar-
khel, A. Nath Jha, A. Singha Roy, Phys. Chem. Chem. Phys. 2020, 22,
2212–2228.
[5] a) J. Homola, Chem. Rev. 2008, 108, 462–493; b) R. L. Rich, D. G. Myszka,
J. Mol. Recognit. 2009, 23, 1–64; c) D.-S. Wang, S.-K. Fan, Sensors 2016,
16, 1175; d) L. Joss, T. A. Morton, M. L. Doyle, D. G. Myszka, Anal. Bio-
chem. 1998, 261, 203–210.
[6] a) A. Renodon-Corni›re, L. H. Jensen, J. L. Nitiss, P. B. Jensen, M. Sehest-
ed, Biochemistry 2002, 41, 13395–13402; b) E. A. Smith, W. D. Thomas,
L. L. Kiessling, R. M. Corn, J. Am. Chem. Soc. 2003, 125, 6140–6148; c) T.
Neumann, H.-D. Junker, K. Schmidt, R. Sekul, Curr. Top. Med. Chem.
2007, 7, 1630–1642; d) C. Di Primo, J. Mol. Recognit. 2008, 21, 37–45;
e) C. Masante, C. Jaubert, W. Palau, J. Plissonneau, L. Besnard, M. Ven-
tura, C. Di Primo, Cell. Mol. Life Sci. 2015, 72, 3375–3385; f) E. Dausse, A.
BarrØ, A. AimØ, A. Groppi, A. Rico, C. Ainali, G. Salgado, W. Palau, E. Da-
guerre, M. Nikolski, J.-J. ToulmØ, C. Di Primo, Biosens. Bioelectron. 2016,
80, 418–425; g) J. Visentin, L. Couzi, C. Dromer, M. Neau-Cransac, G.
Guidicelli, V. Veniard, K. Nubret-le Coniat, P. Merville, C. Di Primo, J.-L.
Taupin, Biosens. Bioelectron. 2018, 117, 191–200; h) H. Duan, M. Dono-
van, F. Hernandez, C. Di Primo, E. Garanger, X. Schultze, S. Lecomman-
doux, Angew. Chem. Int. Ed. 2020, 59, 13591–13596; Angew. Chem.
2020, 132, 13693–13698.
[7] For a selection of articles on SPR analysis of small molecules binding to
immobilized proteins, see: a) B. Johnsson, S. Lçfås, G. Lindquist, Anal.
Biochem. 1991, 198, 268–277; b) P. Stenlund, Š. Frostell-Karlsson, O. P.
Karlsson, Anal. Biochem. 2006, 353, 217–225; c) A. T. Neffe, M. Bilang, B.
Meyer, Org. Biomol. Chem. 2006, 4, 3259–3267; d) I. Navratilova, M. Dio-
szegi, D. G. Myszka, Anal. Biochem. 2006, 355, 132–139; e) B. Nguyen,
F. A. Tanious, W. D. Wilson, Methods 2007, 42, 150–161; f) M. C. Jecklin,
S. Schauer, C. E. Dumelin, R. Zenobi, J. Mol. Recognit. 2009, 22, 319–
329; g) T. Vo, A. Paul, A. Kumar, D. W. Boykin, W. D. Wilson, Methods
2019, 167, 15–27.
[8] a) C. Douat-Casassus, S. Chassaing, C. Di Primo, S. Quideau, ChemBio-
Chem 2009, 10, 2321–2324; b) A. Watrelot, D. T. Tran, T. Buffeteau, D.
Deffieux, C. Le Bourvellec, S. Quideau, C. Renard, Appl. Surf. Sci. 2016,
371, 512–518.
[9] a) J. J. Champoux, Annu. Rev. Biochem. 2001, 70, 369–413; b) Y. Pommi-
er, E. Leo, H. Zhang, C. Marchand, Chem. Biol. 2010, 17, 421–433.
[10] a) B. X. Huang, H.-Y. Kim, C. Dass, J. Am. Soc. Mass Spectrom. 2004, 15,
1237–1247; b) J. A. Hamilton, S. Era, S. P. Bhamidipati, R. G. Reed, Proc.
Natl. Acad. Sci. USA 1991, 88, 2051–2054.
[11] a) J. J. Turnbull, W. J. Sobey, R. T. Aplin, A. Hassan, J. L. Firmin, C. J. Scho-
field, A. G. Prescott, Chem. Commun. 2000, 2473–2474; b) J.-I. Nakajima,
Y. Tanaka, M. Yamazaki, K. Saito, J. Biol. Chem. 2001, 276, 25797–25803.
[12] a) P. Ferrer, M. Asensi, S. Priego, M. Benlloch, S. Mena, A. Ortega, E. Ob-
rador, J. M. Esteve, J. M. Estrela, J. Biol. Chem. 2007, 282, 2880–2890;
b) J. Ge, Y. Liu, Q. Li, X. Guo, L. Gu, Z. G. Ma, Y. P. Zhu, Biomed. Environ.
Sci. 2013, 26, 902–911; c) S. Verma, A. Singh, A. Kumari, C. Tyagi, S.
Goyal, S. Jamal, A. Grover, J. Recept. Signal Transduction 2017, 37, 391–
400.
[4] For a selection of articles on UV/Vis absorption-, fluorescence-, CD-,
SAXS-, DLS- and nephelometry-based analytical techniques applied to
the study of polyphenol–protein interactions, see: a) C. Poncet-Legrand,
A. Edelmann, J.-L. Putaux, D. Cartalade, P. Sarni-Manchado, A. Vernhet,
Food Hydrocolloids 2006, 20, 687–697; b) S. Soares, N. Mateus, V. de
Freitas, J. Agric. Food Chem. 2007, 55, 6726–6735; c) L. Liang, H. A.
Tajmir-Riahi, M. Subirade, Biomacromolecules 2008, 9, 50–56; d) M.
Liang, R. Liu, W. Qi, R. Su, Y. Yu, L. Wang, Z. He, Food Chem. 2013, 138,
[13] S. Quideau, M. Jourdes, C. Saucier, Y. Glories, P. Pardon, C. Baudry,
Angew. Chem. Int. Ed. 2003, 42, 6012–6014; Angew. Chem. 2003, 115,
6194–6196.
[14] a) W. Tückmantel, A. P. Kozikowski, L. J. Romanczyk, J. Am. Chem. Soc.
1999, 121, 12073–12081; b) K. Ohmori, N. Ushimaru, K. Suzuki, Tetrahe-
dron Lett. 2002, 43, 7753–7756; c) A. Saito, N. Nakajima, A. Tanaka, M.
Chem. Eur. J. 2021, 27, 5498 –5508
www.chemeurj.org
5507
ꢀ 2021 Wiley-VCH GmbH

Full Paper
doi.org/10.1002/chem.202005187
Chemistry—A European Journal
Ubukata, Tetrahedron 2002, 58, 7829–7837; d) C. J. Hayes, B. P. Whittak-
er, S. A. Watson, A. M. Grabowska, J. Org. Chem. 2006, 71, 9701–9712;
e) J. A. Steenkamp, C. H. L. Mouton, D. Ferreira, Tetrahedron 1991, 47,
6705–6716.
politano, M. d’Ischia, Chem. Res. Toxicol. 2008, 21, 2407–2413; c) R. Ber-
nini, G. Fabrizi, L. PouysØgu, D. Deffieux, S. Quideau, Curr. Org. Synth.
2012, 9, 650–669; d) ”Stabilized 2-Iodoxybenzoic Acid (SIBX)”: S. Qui-
deau, A. ChØnedØ in Encyclopedia of Reagents for Organic Synthesis,
Wiley, Chichester, 2012.
[15] A. H. De Groot, R. A. Dommisse, G. L. Lemi›re, Tetrahedron 2000, 56,
1541–1549.
[25] R. J. Pounder, M. J. Stanford, P. Brooks, S. P. Richards, A. P. Dove, Chem.
Commun. 2008, 5158–5159.
[16] a) A. Saito, N. Nakajima, A. Tanaka, M. Ubukata, Biosci. Biotechnol. Bio-
chem. 2002, 66, 1764–1767; b) A. P. Kozikowski, W. Tückmantel, G.
Bottcher, L. J. Romanczyk, J. Org. Chem. 2003, 68, 1641–1658; c) K.
Ohmori, N. Ushimaru, K. Suzuki, Proc. Natl. Acad. Sci. USA 2004, 101,
12002–12007; d) A. Saito, N. Nakajima, N. Matsuura, A. Tanaka, M. Ubu-
kata, Heterocycles 2004, 62, 479–489; e) I. Tarascou, K. Barathieu, Y.
AndrØ, I. Pianet, E. J. Dufourc, E. Fouquet, Eur. J. Org. Chem. 2006,
5367–5377; f) F. Viton, C. Landreau, D. Rustidge, F. Robert, G. William-
son, D. Barron, Eur. J. Org. Chem. 2008, 6069–6078.
[17] a) T. Okajima, J. Mol. Struct. THEOCHEM 2001, 536, 73–82; b) J. A. Del-
cour, D. Ferreira, D. G. Roux, J. Chem. Soc. Perkin Trans. 1 1983, 1711–
1717.
[18] C. Chalumeau, D. Deffieux, S. Chaignepain, S. Quideau, ChemBioChem
2011, 12, 1193–1197.
[26] a) G. GyØmµnt, . Zajµcz, B. BØcsi, C. Ragunath, N. Ramasubbu, F. Erdçdi,
G. Batta, L. Kandra, Biochim. Biophys. Acta 2009, 1794, 291–296;
b) J. R. L. Guerreiro, V. de Freitas, D. S. Sutherland, M. G. F. Sales, Procedia
Eng. 2012, 47, 498–501; c) K. Hara, M. Ohara, I. Hayashi, T. Hino, R. Nish-
imura, Y. Iwasaki, T. Ogawa, Y. Ohyama, M. Sugiyama, H. Amano, Eur. J.
Oral Sci. 2012, 120, 132–139; d) Y. Zhang, S. Shi, J. Guo, Q. You, D.
Feng, J. Chromatogr. A 2013, 1293, 92–99; e) M. Maldini, S. Di Micco, P.
Montoro, E. Darra, S. Mariotto, G. Bifulco, C. Pizza, S. Piacente, Phyto-
chemistry 2013, 86, 64–71; f) A. Kiss, B. BØcsi, B. Kolozsvµri, I. Komµromi,
K. E. KçvØr, F. Erdçdi, FEBS J. 2013, 280, 612–626; g) J. R. L. Guerreiro,
V. E. Bochenkov, K. Runaher, H. Aslan, M. Dong, J. J. Enghild, V. de Frei-
tas, M. G. Ferreira Sales, D. S. Sutherland, ACS Sens. 2016, 1, 258–264;
h) J. R. L. Guerreiro, N. Teixeira, V. de Freitas, M. G. F. Sales, D. S. Suther-
land, Food Chem. 2017, 233, 457–466; i) J. Gombau, P. Nadal, N. Canela,
S. Gómez-Alonso, E. García-Romero, P. Smith, I. Hermosín-GutiØrrez, J. M.
Canals, F. Zamora, Food Chem. 2019, 275, 397–406.
[19] a) L. Li, T. H. Chan, Org. Lett. 2001, 3, 739–741; b) T.-J. Ding, X.-L. Wang,
X.-P. Cao, Chin. J. Chem. 2006, 24, 1618–1624; c) K. Krohn, I. Ahmed, M.
John, Synthesis 2009, 779–786; d) K. Ohmori, T. Yano, K. Suzuki, Org.
Biomol. Chem. 2010, 8, 2693–2696; e) H. Tanaka, A. Chino, T. Takahashi,
Tetrahedron Lett. 2012, 53, 2493–2495.
[27] R. Karlsson, P. S. Katsamba, H. Nordin, E. Pol, D. G. Myszka, Anal. Bio-
chem. 2006, 349, 136–147.
[20] D. T. Tran, Doctoral thesis, University of Bordeaux (France), 2015, Order
N8 2015BORD0443.
[28] a) H. CarriØ, D. T. Tran, S. Rousseau, S. Chaignepain, J.-M. Schmitter, D.
Deffieux, S. Quideau, Chem. Commun. 2014, 50, 9387–9389.
[21] B. List, A. Doehring, M. T. Hechavarria Fonseca, A. Job, R. Rios Torres, Tet-
rahedron 2006, 62, 476–482.
[22] B.-F. Ruan, X. Lu, J.-F. Tang, Y. Wei, X.-L. Wang, Y.-B. Zhang, L.-S. Wang,
H.-L. Zhu, Bioorg. Med. Chem. 2011, 19, 2688–2695.
[23] G. Tóth, K. E. KçvØr, Synth. Commun. 1995, 25, 3067–3074.
[24] a) A. Ozanne, L. PouysØgu, D. Depernet, B. FranÅois, S. Quideau, Org.
Lett. 2003, 5, 2903–2906; b) M. De Lucia, L. Panzella, A. Pezzella, A. Na-
Manuscript received: December 3, 2020
Accepted manuscript online: January 14, 2021
Version of record online: February 24, 2021
Chem. Eur. J. 2021, 27, 5498 –5508
www.chemeurj.org
5508
ꢀ 2021 Wiley-VCH GmbH