Mechanistic studies of hydrogen-peroxide-mediated anthocyanin oxidation

Article abstract of DOI:10.1016/j.tet.2018.09.012

The oxidation of cyanidin-3-O-glucoside by hydrogen peroxide was investigated in a range of solvents. The reaction products had chemical structures identical to those formed by the reaction of this compound with the alkylperoxyl radical 2,2?-azobis(2,4-dimethyl)valeronitrile. A plausible oxidation mechanism was proposed based on the obtained reaction products, and this mechanism was confirmed by HPLC–MS experiments using ¹⁸O-labeled reagents. Further, the reaction conditions were found to influence both the reaction rate and the products formed during the transformation, which validated the proposed mechanism.

Full text of DOI:10.1016/j.tet.2018.09.012

Accepted Manuscript
Mechanistic studies of hydrogen-peroxide-mediated anthocyanin oxidation
Ryuya Satake, Emiko Yanase
PII:
S0040-4020(18)31067-6
10.1016/j.tet.2018.09.012
TET 29784
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 14 July 2018
Revised Date: 31 August 2018
Accepted Date: 4 September 2018
Please cite this article as: Satake R, Yanase E, Mechanistic studies of hydrogen-peroxide-mediated
anthocyanin oxidation, Tetrahedron (2018), doi: 10.1016/j.tet.2018.09.012.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical Abstract
Mechanistic studies of hydrogen-peroxide-
mediated anthocyanin oxidation
Leave this area blank for abstract info.
Ryuya Satake and Emiko Yanase*
Faculty of Applied Biological Sciences, Gifu University, 1-1 Yanagido, Gifu 501-1193, Japan

ACCEPTED MANUSCRIPT
Tetrahedron
journal homepage: www.elsevier.com
Mechanistic studies of hydrogen-peroxide-mediated anthocyanin oxidation
Ryuya Satake and Emiko Yanase
Faculty of Applied Biological Sciences, Gifu University, 1-1 Yanagido, Gifu 501-1193, Japan
ARTICLE INFO
ABSTRACT
Article history:
Received
Received in revised form
Accepted
The oxidation of cyanidin-3-O-glucoside by hydrogen peroxide was investigated in a range of
solvents. The reaction products had chemical structures identical to those formed by the reaction
of this compound with the alkylperoxyl radical 2,2ʹ-azobis(2,4-dimethyl)valeronitrile. A
plausible oxidation mechanism was proposed based on the obtained reaction products, and this
mechanism was confirmed by HPLC–MS experiments using 18O-labeled reagents. Further, the
reaction conditions were found to influence both the reaction rate and the products formed
during the transformation, which validated the proposed mechanism.
Available online
Keywords:
Anthocyanin
Hydrogen peroxide
Oxidation
2009 Elsevier Ltd. All rights reserved
.
Oxidation mechanism
———
Corresponding author. Tel./fax: +81-58-293-2914; e-mail: e-yanase@gifu-u.ac.jp

2
Tetrahedron
ACCEPTED MANUSCRIPT

50 mM phosphate buffer (pH 6.8), and the reaction mixture was
ACCEPTED MANUSCRIPT
analyzed using HPLC. Although the amount of compound 1 was
observed to decrease, no product peaks were detected. The same
result was obtained in phosphate buffer without H2O2.
Anthocyanins are known to be less stable under neutral and basic
conditions, and the obtained results suggest that the reaction of 1
with H2O2 cannot be evaluated under these conditions. Therefore,
the reaction of 1 with H2O2 was performed under acidic
conditions (pH 2.4), in which anthocyanins are expected to be
stable. Under these conditions, the HPLC signal corresponding to
1 decreased significantly, and the appearance of three new peaks
was observed. Additionally, the area of these peaks increased
when an 80% aqueous solution of acetonitrile (MeCN) was used
instead of water (Fig. 2A).
1. Introduction
Anthocyanins, which are widely distributed in the plant
kingdom, are responsible for the red, blue, and violet colors of
the flowers and fruits of various plants. Structurally,
anthocyanins are the sugar-containing analogs of anthocyanidin
aglycones, which contain
a characteristic flavylium (2-
phenylchromenium) cation. In addition, as a type of polyphenol,
anthocyanins are known to exhibit antioxidant properties.1–3
Reactive oxygen species such as hydrogen peroxide (H2O2),
hydroxyl radicals (·OH), and superoxide anion radicals are
widely present in the human body. The excessive generation of
reactive oxygen species is believed to cause damage to proteins,
lipids, carbohydrates, and DNA. One of the key roles of
anthocyanins in plants is the protection against oxidative damage.
For example, it has been reported that the concentration of
anthocyanins in plant tissues is enhanced under oxidative
conditions to protect the plants from oxidation.4,5 In the case of
the human body, the generation of reactive oxygen species is
believed to be a factor in the development of various diseases,
including cancer and heart disease. It has been reported that the
presence of dietary polyphenols such as anthocyanins can defend
cells from oxidative damage.6,7
The obtained compounds were isolated by preparative HPLC
and analyzed by NMR spectroscopy and MS. As a result, the
reaction products were identified as protocatechuic acid (3, 9%
yield), which was derived from the B-ring, and compound 2 and
its aglycone 2a (57% yield), which were produced by opening
the C-ring (Fig. 1). The partial conversion of compound 2 to its
aglycone 2a during the concentration process was attributed to
the facile hydrolysis of the glucose ester under acidic conditions.
These three compounds were identical to those reported
previously in the reaction of 1 with AMVN (Fig. 2B).14 In
contrast, the reaction of 1 with H2O2 in ethanol (EtOH) produced
compound 4 as the main product. The formation of this product
occurred through a rearrangement reaction accompanied by ring
contraction, and this product was also obtained in the reaction of
Although various reports have discussed the oxidation
reactions of anthocyanins,8–13 the details of their radical
scavenging mechanism have not yet been revealed. In our
previous study, we investigated the reaction of cyanidin-3-O-
glucoside (1) with the alkylperoxyl radical 2,2ʹ-azobis(2,4-
dimethyl)valeronitrile (AMVN) and proposed an oxidation
mechanism based on the chemical structures of the obtained
products.14 In this study, we investigated the reaction of 1 with
H2O2 and hydroxyl radicals in detail and identified the oxidation
products (Fig. 1). In addition, the reaction of 1 with H2O2 was
examined in a range of solvents and the effect of the reaction
conditions on the reactivity of this system was determined.
Finally, we proposed and confirmed a plausible mechanism for
the oxidation of 1 under H2O2 conditions.
A:H2O2
2
3
2a
B:AMVN
3
5
2
1 with AMVN in EtOH (Fig. 1).
Fig. 2. HPLC chromatograms of the products formed in the reaction of 1
with (A) H2O2 in H2O and (B) AMVN in MeCN/H2O.
It has been reported that the alkyl radical generated by heating
AMVN reacts with oxygen to yield an alkyl peroxide radical,15
indicating that the reaction with AMVN does not involve the
direct reaction of an alkyl radical. Interestingly, when 1 was
treated with H2O2 in an 80% aqueous solution of MeCN,
compound 5, which was produced in 15% yield in the reaction
with AMVN,14 was not observed. It should be noted that the
difference between the chemical structures of compounds 5 and 2
is the position of oxidation, as compound 5 is formed by
oxidation at the 4-position of 1, whereas 2 is produced by
oxidation at the 3-position. Considering the reaction mechanism,
Fig. 1. Oxidation of 1 in various solvents under H2O2 conditions.
2. Results and discussion
2.1. Reaction of 1 with H2O2
Anthocyanins are known to exhibit radical scavenging ability,
as mentioned in Section 1. To clarify the radical scavenging
mechanism, the reaction of 1 with H2O2, a reactive oxygen
species, was investigated. Compound 1 was treated with H2O2 in

4
Tetrahedron
ACCEPTED MANUSCRIPT
oxidation at the 4-position of compound 1 requires either the
of protonation, competition exists between H O and the reaction
2 2
loss of the aromaticity of the C-ring or ring opening owing to the
nucleophilic addition of water at the 2-position. Conversely, it is
assumed that oxidation at the 3-position of 1 occurs when the
aromaticity of the C-ring is maintained. Thus, compound 5 was
not formed in the presence of H2O2 because the rate of oxidation
was higher than that of the nucleophilic addition of water to the
2-position owing to the greater reactivity of H2O2. This result is
consistent with compound 1 being consumed faster in the
presence of H2O2 than in the AMVN system.
solvent. The pKa values (measured in water) for the conjugate
acids of the various solvents are EtOH (−2.4) < DMSO (−1.8) <
H2O (−1.7).18–20 However, the inverse order was observed for the
reaction rate. As solvent protonation is difficult for solvents with
low conjugate acid pKa values, the protonation of H2O2 is
expected to be promoted. In fact, a high reaction rate was
observed when the reaction was conducted in EtOH, which is the
solvent with the lowest conjugate acid pKa value. In addition, this
rate decreased upon the addition of water. Furthermore, little
difference in reactivity was observed between DMSO and H2O,
which have similar conjugate acid pKa values. These results
suggest a direct relationship between the pKa value and solvent
reactivity in this system. However, although the conjugate acid
pKa value of MeCN was the lowest among the various solvents
examined (i.e., −10),20 the reaction rate in MeCN/H2O was the
lowest observed and inconsistent results were obtained. It should
be noted that owing to the insolubility of compound 1 in MeCN,
20% water was added. Under these conditions, the reactivity was
likely altered by the addition of water, which affected solvent
protonation. As the pKa value of the conjugate acid of MeCN is
significantly smaller than that of water, the protonation of H2O2
2.2. Reaction of 1 with hydroxyl radicals
Among reactive oxygen species, hydroxyl radicals exhibit the
highest oxidative ability. They are produced from H2O2 under
UV irradiation conditions or by the Fenton reaction. To
investigate differences from the reaction under H2O2 conditions,
1 was reacted with hydroxyl radicals produced by the Fenton
reaction. First, to confirm the production of hydroxyl radicals, a
qualitative analysis was performed using the ESR spin-trapping
technique. Following the addition of H2O2 to
FeSO4·7H2O/diethylenetriaminepentaacetic acid (DTPA)
aqueous solution containing 2-(5,5-dimethyl-2-oxo-2λ5-
[1,3,2]dioxaphosphinan-2-yl)-2-methyl-3,4-dihydro-2H-pyrrole
1-oxide (CYPMPO), an ESR signal corresponding to the
CYPMPO–hydroxyl radical adduct (a spin adduct of the
hydroxyl radical) was detected. Although the quantities were
slightly lower than that in water, hydroxyl radicals were also
observed in 80% MeCN and 80% EtOH aqueous solutions under
the same conditions, confirming the generation of hydroxyl
radicals under these conditions. In contrast, no signal was
detected in the presence of 1 under the same conditions (Fig. S1).
This result suggests that the generated hydroxyl radicals were
trapped by 1.
a
+
is favored over MeCN, indicating that the generation of H3O2
species is affected by the addition of water.
(A)
Next, compound 1 was treated under the Fenton reaction
conditions in acidic water (pH 2.4), and the solution was
analyzed using HPLC. Under these conditions, the HPLC signal
corresponding to 1 decreased and several new peaks appeared.
The areas of these peaks increased when an 80% aqueous
solution of MeCN was used instead of water. However, the
chromatographic pattern observed for this reaction solution was
the same as that observed for the reaction under H2O2 conditions,
as described in Section 2.1. Thus, both reactions are assumed to
produce the same products. This result suggests that the products
generated by the treatment of 1 under Fenton reaction conditions
were the same as those obtained by the reaction of 1 with H2O2,
not with hydroxyl radicals.
(B)
2.3. Effect of solvent on the reaction of 1 with H2O2
To determine the effect of different solvents on the reactivity
of 1 with H2O2, the quantity of 1 in the reaction mixture was
monitored by HPLC. The transformations were conducted under
acidic conditions to eliminate the effects of degradation reactions
not caused by H2O2. As shown in Fig. 3A, the greatest decrease
in the concentration of 1 was observed when the reaction was
performed in EtOH, with no HPLC signals corresponding to 1
observed after 2 h. In contrast, the lowest reactivity was observed
in 80% aqueous MeCN, with ~30% of 1 remaining after 4 h. The
reactivity of 1 in the various solvents followed the order EtOH >
EtOH/H2O > DMSO = DMF = H2O > MeCN/H2O. The oxidation
of 1 by H2O2 under acidic conditions was presumed to involve a
bimolecular nucleophilic substitution reaction (SN2 reaction) that
Fig. 3. Influence of reaction solvent on reactivity and product yield. (A)
Reduction in the content of compound 1. (B) Product yields of compounds 2,
3, 4, and 6 in various reaction solvents. The yields of compounds 2 and 4
include their aglycone yields.
+
occurs via the formation of H3O2 , which is generated by the
protonation of H2O2.18 This species forms easily in acidic H2O2
solutions and is extremely reactive, indicating that the reaction is
dominated by diffusion control.16,17 Furthermore, in the context

ACCEPTED MANUSCRIPT
Fig. 4. Hypothetical oxidation mechanism for compound 1.
reaction products were analyzed using negative ion mode UPLC–
ESI-MS/MS (Fig. S2).
2.4. Effect of solvent on the reaction products
The effect of the solvent on the product selectivity of the
reaction between H2O2 and 1 was examined using EtOH,
EtOH/H2O, 80% aqueous MeCN, DMSO, and DMF (Fig. 3B).
Upon addition of H2O2, compound 2 was obtained as the major
product when aqueous MeCN, DMSO, and DMF were used,
whereas compound 4 was obtained as the major product in
addition to trace amounts of compound 6 when EtOH was used.
In the EtOH/H2O solvent system, the production of compounds 3
and 4 was inhibited as the water ratio increased, whereas the
formation of compound 2 was promoted.
Compound 2A, obtained by treatment of 1 with H2O2 in
MeCN/H218O, gave a pseudo molecular ion peak ([M−H]−) at m/z
321, which is 2 Da higher than the original signal and confirms
the incorporation of the 18O atom. To identify the position of the
18O oxygen atom, MS2 analysis at m/z 321 was performed. In this
case, two major fragment ions were observed at m/z 155 and 183.
By comparison with the MS2 data of compound 2, the fragment
ion peak at m/z 155 was identified as 18O-protocatechuic acid,
suggesting that the carbonyl oxygen atom at the 2-position of the
C-ring was derived from H218O (Fig. 4). Anthocyanins are known
to exist as mixtures of several structures in chemical equilibrium
because of reversible hydration at the 2-position. The production
of compound 2A, which has 18O at the 2-position of the C-ring,
was presumed to occur through this hydration reaction.
In contrast, upon treatment of 1 with H218O2 in MeCN/H2O,
compound 2B with a pseudo molecular ion peak ([M−H]−) at m/z
321 was observed. In this case, the MS2 analysis suggested that
the carbonyl oxygen atom at the 3-position of the C-ring was
derived from H218O2. Furthermore, treatment of compound 1 with
H218O2 in EtOH yielded compound 5 bearing an 18O-label on the
carbonyl moiety at the 4-position of the C-ring. Oxidation of
tautomeric species I occurs at the 3-position owing to electron
donation from the phenolic hydroxyl group at the 5- or 7-position
of the A-ring, resulting in the formation of species II (Fig. 4). In
MeCN/H2O, species II is converted to 2 by cleavage of the C2–
C3 bond, driven by aromatization of the A-ring. The production
of 2B, in which the carbonyl oxygen at the 3-position is labeled,
supports this mechanism. In EtOH, oxidation of species I, which
is formed by the nucleophilic addition of EtOH at the 2-position,
generates species III in the same manner. Species III is converted
to 4 by a rearrangement reaction.
With the exception of the water-based systems, a total yield of
60–75% was obtained for compounds 2, 3, and 4. However, upon
increasing the water content in the EtOH/H2O system, this yield
decreased sharply. We expect this to be due to the protic nature
and high polarity of water, which likely complicates the reaction
by promoting nucleophilic addition to the product and the
reaction intermediate.
Additionally, the generation of ethyl protocatechuate (6) from
the B-ring was observed in EtOH, although the amount formed
decreased with increasing water content. The formation of trace
amounts of protocatechuic acid (3) was also observed under all
solvent conditions examined. Initially, we assumed that
compound 3 was generated via the hydrolysis of compound 2, as
this product was formed in aqueous solution at pH 4. However,
compound 3 was generated when the reaction mixture obtained
by reacting 1 with H2O2 in the presence of water (and containing
compounds 2, 3, and 4 in low yields) was treated under strongly
oxidizing conditions (data not shown). This result suggests that
compound 3 was generated from 1 via a more complex route.
2.5. Confirmation of the proposed oxidation mechanism
A mechanism for the oxidation of 1 by H2O2 was then
proposed based on the chemical structures of the reaction
products, as outlined in Fig. 4. To confirm the proposed
mechanism, 1 was oxidized using H218O or H218O2 and the
3. Conclusion
Our investigation of the reaction of 1 with H2O2 revealed that
the reaction products were identical to those formed in the
reaction with AMVN.
A potential mechanism for this
transformation, proposed based on the chemical structures of the

6
Tetrahedron
ACCEPTED MANUSCRIPT
obtained products, was confirmed by UPLC-MS/MS
To an aqueous solution of FeSO4·7H2O/DTPA (1 mM/1 mM,
experiments using 18O-labeled reagents. This mechanism was
further validated by observing the influence of various reaction
conditions on the reaction rate and product distribution.
100 µL) was added water, 80% MeCN/H2O, or 80% EtOH/H2O
(100 µL) and CYPMPO aqueous solution (310 mM, 10 µL). The
spectra were recorded in a 25 ꢀL glass capillary 15 min after
mixing. Typically, the instrumental settings were as follows:
center field, 326.4 mT; sweep width, ±7.5 mT; field modulation
frequency, 100 kHz; field modulation width, 0.2 mT; amplitude,
500; sweep time, 1 min; time constant, 0.03 s; microwave
frequency, 9163.310 MHz; and microwave power, 0.998 mW.
The ESR spectrum of manganese held in the ESR cavity was
used as an internal standard.
As H2O2 is a reactive oxygen species that is present in many
biological systems, it is likely that the degradation products
formed by reactions between anthocyanins and H2O2 are also
present in anthocyanin-containing foods. However, the potential
benefits of these products to human health remain under debate.
In addition, as the contents of these products in various plant
species are expected to increase post-harvest, they could
potentially be used to determine the freshness of anthocyanin-
containing fruits and vegetables. Furthermore, although
anthocyanins are important plant-derived pigments, they are
rather unstable. We therefore expect that the results presented
herein will aid in the development of novel methods for
improving the stability of the anthocyanin structure. Further
studies on anthocyanin stability are now in progress.
4.4. Comparison of the reactivity of 1 in various solvents
To a solution of compound 1 (1 mg) in H2O (200 ꢀL, 10 mM)
was added aqueous H2O2 (100 ꢀL), which was prepared from
30% aqueous H2O2 by diluting 400 times with H2O. The resulting
solution was stirred at 40 °C and HPLC analyses were performed
after 0, 1, 2, and 4 h. The HPLC analysis conditions were as
follows: COSMOSIL 5C18-MS-II (4.6 mm id × 150 mm; Nacalai
Tesque Inc., Kyoto, Japan); mobile phase, 0.5% TFA/12%
aqueous MeCN; flow rate, 1.0 mL/min; system temperature, 35
°C; and detection wavelength, 520 nm. The reaction was
monitored using the peak area of compound 1. The above
reaction was repeated using DMSO, DMF, EtOH, and
MeCN/H2O as the solvent instead of H2O.
4. Experimental section
4.1. General
Cyanidin-3-O-glucoside (1) was isolated from a black bean
extract, which was a kind gift from Nagara Science Co. (Gifu,
Japan), and recrystallized from a mixture of 5–10% HCl in
methanol (MeOH).21,22 H2O2 solution (30% (w/w) in H2O, 9.8 M)
and all organic solvents were purchased from Wako Pure
Chemical Industries, Ltd. (Osaka, Japan).
4.5. Confirmation of the oxidation mechanism using H218O or
H218O2
To confirm the proposed oxidation mechanism, the reaction
was performed according to the procedure described in Section
4.4 using H218O or H218O2 instead of H2O or H2O2, respectively.
All reaction mixtures were analyzed by UPLC–TOF-MS. For the
analysis of compound 2, the mobile phase was composed of 0.5%
formic acid/10% MeCN in H2O, whereas for the analysis of
compound 4, 0.5% formic acid/15% MeCN in H2O was used.
The flow rate was set at 0.2 mL/min and the ESI interface was
operated in negative ion mode. The MS parameters for the
analysis were as follows: capillary potential, 0.6 kV; sampling
cone, 13 V; desolvation temperature, 500 °C; source temperature,
150 °C; desolvation gas flow, 1000 L Ar/h; and cone gas flow, 50
L Ar/h. The mass spectrometer was operated in MS/MS mode
with a collision cell energy of 13 V.
ESR analysis was performed using a JEOL JES-FA100
spectrometer (Tokyo, Japan). The spin-trapping reagent
CYPMPO was purchased from Shidai Systems (Saitama, Japan).
NMR spectra were recorded on a JEOL ECA-500 or a JEOL
ECA-600 instrument (Tokyo, Japan) using TMS as an internal
standard. UPLC–ESI-MS was performed using a Waters Xevo
G2 QTOF mass spectrometer (Waters, Milford, MA, USA)
equipped with a C18 analytical column (ACQUITY UPLC BEH
C18 column, 2.1 mm id × 100 mm; Waters, Milford, MA, USA).
The mobile phase was composed of 0.5 vol% acetic acid, 10
vol% aqueous MeCN, or 15 vol% aqueous MeCN. All analyses
were conducted at 35 °C, and the flow rate was set at 0.2
mL/min. HPLC analyses were performed using a JASCO PU-
2089 intelligent pump equipped with a JASCO MD-2010 PDA
detector and a JASCO CO-2065 column oven (Tokyo, Japan). A
COSMOSIL 5C18-MS-II column (4.6 mm id × 150 mm; Nacalai
Tesque Inc., Kyoto, Japan) and a NB-ODS-9 column (10 mm id
× 250 mm; Nagara Science Co., Ltd., Gifu, Japan) were used for
analytical and preparative HPLC, respectively.
Acknowledgement
This work was partially supported by Nagara Science Co.,
Ltd.
Declarations of interest: none
4.2. Reaction of 1 with H2O2 in 80% aqueous MeCN
Supplementary data
Compound 1 (30.0 mg) was dissolved in 80% aqueous MeCN
(6 mL) prior to the addition of aqueous H2O2 (24.5 mM, 3 mL)
that had been diluted 400 times using 80% aqueous MeCN. The
resulting solution was stirred at 40 °C for 8 h. Subsequently, the
reaction mixture was cooled to room temperature, MeCN was
removed under reduced pressure, and the resulting residue was
subjected to column chromatography using HP20SS resin
(Mitsubishi Chemical Co., Tokyo, Japan). Elution was achieved
using MeOH after washing with two column volumes of water to
remove H2O2. The sample was purified by HPLC (column: NB-
ODS-9, 10 mm id × 250 mm) using MeCN/H2O (10:90 v/v)
containing 0.5% trifluoroacetic acid (TFA) to yield compounds 2
(57%) and 3 (9%). The chemical structure of compound 2 was
confirmed by comparison with literature NMR data.14
Supplementary data associated with this article can be found
in the online version, at http://dx.doi.org/
References
1. Harborne JB, Ed. The Flavonoids: Advances in Research Since
1986; Chapman & Hall: London, UK, 1994.
2. Harborne JB, Williams CA. Nat. Prod. Rep. 1998;15:631–652.
3. Castañeda-Ovando A, Pacheco-Hernández ML, Páez-Hernández
ME, Rodríguez JA, Galán-Vidal CA. Food Chem. 2009;113:859–
871.
4. Zhang K-M, Yu H-J, Shi K, Zhou Y-H, Yu J-Q, Xia X-J. Plant
Sci. 2010;179:202–208.
5. Bi X, Zhang J, Chen C, Zhang D, Li P, Ma F. Food Chem.
2014;152:205–209.
4.3. ESR spectra

6.
McGhie TK, Walton MC. Mol. Nutr. Food RAes.C20C07;E51P:7T02–E71D3. MANUSCRIPT
7. Nakamura S. J. Nippon Med. Sch. 2013;9:164–169.
8. Karrer P, Widmer R, Helfenstein A, Hürliman W, Nievergelt O,
Monsasrrat-Thoms P. Helv. Chim. Acta 1927;10:729–757.
9. Karrer P, Widmer R. Helv. Chim. Acta 1927;10:758–763.
10. Jurd L. J. Org. Chem. 1964;29:2602–2605.
11. Jurd L. Recent Adv. Phytochem. 1972;5:135–164.
12. Tsuda T, Watanabe M, Ohshima K, Norinobu S, Choi S-W,
Kawakishi S, Osawa T. J. Agric. Food Chem. 1994;42:2407–
2410.
13. Es-Safi N-E, Meudec E, Bouchut C, Fulcrand H, Ducrot P-H,
Herbette G, Cheynier V. J. Agric. Food Chem. 2008;56:4584–
4591.
14. Kamiya H, Yanase E, Nakatsuka S. Food Chem. 2014;155:221–
226.
15. Sang S, Tian S, Wang H, Stark RE, Rosen RT, Yang CS, Ho C-T.
Bioorg. Med. Chem. 2003;11:3371–3378.
16. Leere Øiestad ÅM, Petersen AC, Bakken V, Vedde J, Uggerud E.
Angew. Chem. Int. Ed. 2001;40:1305–1309.
17. Lindinger W, Albritton DL, Howard CJ, Fehsenfeld FC, Ferguson
EE. J. Chem. Phys. 1975;63:5220–5222.
18. Bagno A, Scorrano G. J. Am. Chem. Soc. 1988;110:4577–4582.
19. Evans pKa Table.
http://evans.rc.fas.harvard.edu/pdf/evans_pKa_table.pdf (access
date 23 Jan. 2018).
20. Pine SH, Hendrickson JB, Cram DJ, Hammond GS. Organic
Chemistry, 4th Ed.; McGraw-Hill: Tokyo, Japan, 1980; pp 98–102.
21. Lee JH, Kang NS, Shin S-O, Shin S-H, Lim S-G, Suh D-Y, Beak
I-Y, Park K-Y, Ha TJ. Food Chem. 2009;112:226–231.
22. Valls J, Millán S, Martí MP, Borràs E, Arola L. J. Chromatogr. A
2009;1216:7143–7172.

ACCEPTED MANUSCRIPT
Highlights
The hydrogen-peroxide-mediated oxidation of cyanidin-3-O-glucoside was
investigated.
The effect of solvent on the reaction rate and product distribution was examined.
A plausible mechanism for oxidation under hydrogen peroxide conditions was
proposed.
The reaction mechanism was validated by UPLC-MS/MS using 18O-labeled reagents.