Acid mediated hydrolysis of blueberry anthocyanins

Article abstract of DOI:10.1248/cpb.49.114

Acid mediated hydrolysis of anthocyanins was studied using capillary zone electrophoresis (CZE). A commercially available wild blueberry (Bilberry) extract was dissolved in different concentrations of TFA (0.1, 1, 3, 9%), then was subjected to thermodecom

Full text of DOI:10.1248/cpb.49.114

114
Notes
Chem. Pharm. Bull. 49(1) 114—117 (2001)
Vol. 49, No. 1
Acid Mediated Hydrolysis of Blueberry Anthocyanins
c
Takashi ICHIYANAGI, ,a Kikuo OIKAWA,a Chigusa TATEYAMA,b and Tetsuya KONISHI
*
Department of Hygiene Chemistrya and Department of Radiochemistry-Biophysics,c Niigata College of Pharmacy, Niigata
950–2081, Japan and Niigata Woman’s College,b Niigata 950–0806, Japan.
Received July 10, 2000; accepted October 5, 2000
Acid mediated hydrolysis of anthocyanins was studied using capillary zone electrophoresis (CZE). A com-
mercially available wild blueberry (Bilberry) extract was dissolved in different concentrations of TFA (0.1, 1, 3,
9%), then was subjected to thermodecomposition reaction at 95 °C. After the reaction, the samples were ana-
lyzed by CZE. The hydrolysis rate of each anthocyanin and the formation of the aglycon were determined by the
change in the peak pattern of the anthocyanins in the electropherogram. Each anthocyanin peak decreased time
dependently in a first order kinetic fashion. It was revealed that the hydrolysis rate of each anthocyanin was de-
termined primarily by the type of conjugated sugar and not by the aglycon structure. The rate constant of antho-
cyanin hydrolysis was in the following order, arabinosideϾgalactosideϾglucoside without regard to the aglycon
structure. The kinetic behavior of this anthocyanin hydrolysis together with the CZE mobility allowed us to iden-
tify an unknown CZE peak as delphinidin 3-O-b-arabinoside. At low TFA concentration, significant decomposi-
tion of the anthocyanidin nucleus occurred, but the glycoside hydrolysis predominated at high TFA concentra-
tion. It was further revealed that the aglycon released reacted successively to form polymeric products at higher
TFA conditions.
Key words blueberry anthocyanin; capillary electrophoresis; anthocyanin decomposition; glycoside hydrolysis
lution was diluted 10 times with 1% TFA, then 100 ml of the solution was
taken in a sample tube, dried up in vacuo and finally dissolved in 1 ml of 1%
TFA. After passing through a 0.2 mmf membrane filter, the sample solutions
were subjected to CZE. Triplicate determinations were run for each sample.
The CZE conditions were as reported previously.9)
Anthocyanins are the reddish pigments widely distributed
in colored fruits and vegetables, such as eggplant, grapes and
blueberry. They are usually present as the glycosides having
arabinose, glucose or galactose attached to the anthocyanidin
nucleus. The phytoceutical significance of anthocyanins has
been discussed in relation to a wide range of physiological Results
functions such as vision improvement,1) antioxidant,2—5) anti-
Hydrolysis of each anthocyanin in the Bilberry extract was
cancer6) and so on. Few studies have been done on the struc- determined in 1% TFA using the CZE method. Twelve antho-
ture–radical scavenging activity relationship among antho- cyanin peaks were identified as shown in Fig. 2 from the
cyanin aglycons and the glycosides.7)
comparison of the mobility of standard anthocyanins as de-
It is known that the flavylium cation form of the antho- scribed in our previous report.9) When the TFA solution of
cyanidin nucleus is unstable under conditions such as neutral the Bilberry extract was heated in a sealed ampule filled with
and alkaline pH, high temperature and light illumination.8) Argon at 95 °C, the CZE peaks of each anthocyanin de-
From the point of view of using anthocyanins as a functional creased with reaction time, concomitantly with an increase of
food material, it is important to know the stability of antho- aglycon peak (Fig. 2). The time course changes of antho-
cyanins as a mixture.
cyanin and the aglycon peak are shown in Fig. 3 together
Since there are few studies on the degradation property of with the 525 nm absorbance (A525 nm) change of the reaction
anthocyanins, we precisely examined here the acid mediated solution that was determined as the indication of anthocyani-
reaction of anthocyanins by capillary zone electrophoresis din nucleus. It was revealed that the aglycon formation
(CZE) using commercially available wild berry (Bilberry)
extract as the sample of anthocyanin mixture. The results re-
vealed that the hydrolytic rate of anthocyanin glycoside was
not dependent on the aglycon structure, but dependent on the
type of conjugated sugar.
Materials and Methods
Reagents All the reagents including Na-borate (NaBO4), trans-1,2-di-
aminocyclohexane N,N,NЈ,NЈ-tetra acetic acid monohydrate (CyDTA) and
trifluoroacetic acid (TFA) were obtained from Wako Pure Chemical Indus-
tries, Co. Ltd., Japan. CLEAN 99 K200 was from Clean Chemical Co. Ltd.,
Japan.
Bilberon 25, the concentrated extract of Bilberry (Vaccinium myrtillus L.,
BILBERRY), was a kind gift from Tokiwa Phytochemicals Co. Ltd.
Methods. Acid Mediated Hydrolysis of Anthocyanin Bilberon 25
was dissolved in different concentrations of TFA (0.1, 1, 3, 9%) to make the
concentration to 6.47ϫ10Ϫ3 mol delphinidin eq/l (calculated using eϭ30906
at 525 nm absorbance). An aliquot was taken into an ampule, sealed with
Argon, then was subjected to hydrolytic reaction at 95 °C for a defined pe-
riod. After the reaction, each sample was analyzed by CZE to determine the
hydrolytic degradation rate of each anthocyanin.
Analysis of Anthocyanins by CZE An aliquot of the above reaction so- Fig. 1. Structure of Anthocyanins
To whom correspondence should be addressed. e-mail: kouji@niigata-pharm.ac.jp
© 2001 Pharmaceutical Society of Japan

January 2001
115
Fig. 2. Change of Anthocyanin Peaks in the CZE Electrophoretogram after TFA Mediated Hydrolysis Reaction
(A) Before hydrolysis reaction, (B) 30 min reactions, (C) 60 min reactions, (D) 90 min reactions. 1, malvidin 3-O-b-D-glucoside; 2, peonidin 3-O-b-D-glucoside; 3, malvidin 3-O-
b-D-galactoside; 4, petunidin 3-O-b-D-glucoside; 5, petunidin 3-O-b-D-galactoside; 6, cyanidin 3-O-b-D-glucoside; 7, delphinidin 3-O-b-D-glucoside; 8, unknown 1; 9, cyanidin 3-
O-b-D-galactoside; 10, delphinidin 3-O-b-D-galactoside; 11, cyanidin 3-O-b-D-arabinoside; 12, unknown 2.
Fig. 3. Time Course Changes of Anthocyanin Disappearance, Aglycon
Formation and Bleaching of A525 nm at 95 °C in 9% TFA
a) Bleaching of A525 nm of the reaction solution determined by visible absorption
spectroscopy. b) Anthocyanin disappearance and aglycon formation determined by
CZE. Symbols: ᭿, A525 nm; ᭺, anthocyanin; ᭹, aglycon.
showed biphasic kinetics, that is, the aglycon peak decreased
after the formation reached the maximum after prolonged re-
action period. The higher the TFA concentration was, the
shorter the time reaching the maximum became (data not
shown).
Fig. 4. Effect of Conjugated Sugar on the Rate of Anthocyanin Hydrolysis
Values are meanϮS.D. of three independent reactions. Symbols: ᭺, cyanidin 3-O-b-
glucoside; ᭹, cyanidin 3-O-b-galactoside; ᮀ, cyanidin 3-O-b-arabinoside.
When the changes of CZE peak of each anthocyanin were

116
Vol. 49, No. 1
Fig. 6. TFA Concentration Dependence of Anthocyanin Hydrolysis and
Bleaching of A525 nm
Symbols: ᭺, bleaching at A525 nm; ᭹, anthocyanin glycoside.
of anthocyanin glycoside and the bleaching of A525 nm (degra-
dation of anthocyanidin nucleus) are shown in Fig. 6. The
hydrolysis rate constant of anthocyanin increased linearly
with TFA concentration whereas the rate constant for A525 nm
Fig. 5. Effect of Aglycon Structure on the Rate of Anthocyanin Hydroly-
sis
Values are meanϮS.D. of three independent reactions. Symbols: ᮀ, malvidin 3-O-b- bleaching decreased with TFA concentration.
glucoside; ᭿, peonidin 3-O-b-glucoside; ᭝, petunidin 3-O-b-glucoside; ᭺, cyanidin 3-
O-b-glucoside; ᭹, delphinidin 3-O-b-glucoside.
Discussion
The present experiments revealed that Anthocyanin hy-
Table 1. Rate Constants of Anthocyanin Hydrolysis in TFA Solution at
95 °C
drolysis in TFA solution is determined primarily by the con-
jugated sugar type and not by the aglycon structure. The hy-
drolysis rate constants for the glycosides were in the follow-
ing order, arabinosideϾgalactosideϾglucoside without re-
gard to the aglycon structure. Bemisser studied the acid-cat-
alyzed hydrolysis properties of model glycosides of small
molecules such as ethanol and benzyl alcohol, and reported
similar differences in the hydrolytic behavior of the glyco-
sides such that galactoside was more unstable than glucoside
when the glycosides with the same aglycon were com-
pared.10) Thus, he concluded that the difference in the confor-
mational strain in the sugar moiety was the cause of the ki-
netic difference of the glycoside hydrolysis. However, his
data also showed that the aglycon structure significantly af-
Anthocyanin
0.1% TFA
1% TFA
3% TFA
9% TFA
Mv 3-Glc
Mv 3-Gal
Pn 3-Glc
Pt 3-Glc
7.6
17.9
6.7
11.4
21.7
5.2
12.2
26.6
11.3
25.6
21.6
30.2
9.3
26.2
6.9
21.5
27.0
18.5
27.8
37.3
28.9
43.5
59.4
29.4
42.2
33.2
62.7
47.2
52.5
42.4
53.4
37.2
38.7
64.3
117.9
41.9
61.1
43.6
122.7
18.7
25.9
12.5
21.9
33.7
15.0
30.5
26.8
34.8
Pt 3-Gal
Cy 3-Glc
Cy 3-Gal
Cy 3-Ara
Dp 3-Glc
Dp 3-Gal
Unknown 1
Unknown 2
First order rate constants of anthocyanin degradation at different TFA concentration fected the hydrolysis behavior of the glycosides as much as
(minϪ1ϫ10Ϫ3).
the conjugated sugar. Therefore, it is not yet conclusive that
the same mechanism is involved in the anthocyanin hydroly-
sis, but it is quite interesting that the degradation of antho-
cyanin is governed primarily by the conjugated sugar struc-
ture rather than the aglycon structure.
plotted against the reaction period, the disappearance rate of
the anthocyanin peak was first order (Fig. 4). The kinetic
constants thus obtained increased with increasing TFA con-
centration. When the kinetic constants for the anthocyanins
were compared, it was revealed that the anthocyanins having
the same aglycon structure showed different kinetic constants
depending on their conjugated sugar type. On the other hand,
the anthocyanins having the same conjugated sugar showed
similar kinetic constants without regard to their aglycon
structure (Figs. 4 and 5). Hence the hydrolysis rate of antho-
cyanins does not depend on the aglycon structure, but on the
type of conjugated sugar. The hydrolysis rate constant was
found in the following order, arabinosideϾgalactosideϾglu-
coside in all anthocyanins. The hydrolysis rate constants of
each anthocyanin are summarized in Table 1.
The present study also identified an unknown anthocyanin
(peak 12) from our previous CZE study. We have previously
proposed the aglycon structure of peak 12 to be delphinidin
from its electrophoretic mobility behavior.9) In the present
study, the degradation rate constant of peak 12 was identical
with those for arabinoside, thus it is concluded that the
unidentified peak 12 is delphinidin 3-O-b-D-arabinoside.
From the comparison of the kinetic profiles of anthocyanin
disappearance, the aglycon formation, and the bleaching of
A525 nm at different TFA concentrations, the degradation path
of anthocyanin was suggested to be different at different pH
conditions. At low TFA concentration, the bleaching of
A525 nm of the reaction solution occurred more markedly than
The TFA concentration dependence of both the hydrolysis

January 2001
117
at high TFA and the rate constant was comparable to those of
Recently, it was reported that polyphenols such as pro-
the glycoside hydrolysis and aglycon formation (see Fig. 6). cyanidin, condensed products of anthocyanidin and catechin,
At TFA concentrations higher than 3%, however, the A525 nm is formed in red wine during aging and they have strong an-
bleaching rate decreased and inversely the glycoside hydroly- tioxidant activity.11) If we take account of the solubility and
sis and the aglycon formation were accelerated, thus the time color properties of the new product, in addition to the TLC
to reach the maximum aglycon formation was shortened. separation pattern indicating the product has a higher molec-
These results indicate that at a low concentration of TFA, the ular weight than the aglycon, we may suggest that the same
direct bleaching of the anthocyanidin nucleus due to decom- type of anthocyanidin polymer was formed in the acidic hy-
position predominated because of the unstable nature of the drolysate mentioned above. Further studies are under way to
flavilium cation. At high TFA concentration, however, the hy- clarify the structure and the antioxidant property of the red-
drolysis reaction became marked to release the aglycon being dish precipitate.
transformed successively into another product, probably, the
polymerized product. Indeed, a red shift in the lmax was ob-
served for the product compared to the original anthocyanin
Acknowledgment We thank to M. Uchida of Tokiwa Phytochemicals
Co. Ltd., in Chiba, Japan for providing “Bilberon 25”.
solution (lmax 555 nm and 505 nm, respectively, in acidic
References
1) Mercier A., Perdriel G., Rozier J., Cheraleraud J., Bull. Soc. Ophtal-
mol. Fr., 65, 1049—1053 (1965).
2) Tamura H., Yamagami A., J. Agric. Food Chem., 42, 1612—1615
(1994).
3) Igarashi K., Abe S., Satoh J., Agric. Biol. Chem., 54, 171—175 (1990).
4) Kayamori F., Igarashi K., Biosci. Biotech. Biochem., 58, 570—571
(1994).
5) Teresa Satue-Gracia M., Heinonen M., Frankel E. N., J. Agric. Food
Chem., 45, 3362—3367 (1997).
6) Bomser J., Madhavi D. L., Singletary K., Smith M. A. L., Planta Med.,
42, 212—216 (1996).
7) Tsuda T., Shiga K., Ohshima K., Kawakishi S., Osawa T., Biochem.
Pharmacol., 52, 1033—1039 (1996).
8) Brouillard R., “Flavonoids and Flower Color. In The Flavonoids,” ed.
by Harvorne J. B., Chapman and Hall, London, 1988, pp. 525—538.
9) Ichiyanagi T., Tateyama C., Oikawa K., Konishi T., Biol. Pharm. Bull.,
23, 492—497 (2000).
methanol). Since the secondary product shift absorbs 525 nm
light significantly, the observed rate of A525 nm bleaching
could be decreased under the conditions that the reactions
(hydrolysis and chromophore decomposition) compete with
each other, that is, at high TFA concentration as shown in
Fig. 6.
Although the secondary product(s) was suggested to be
formed in the reaction mixture as discussed above, we could
not detect any additional peak generation in the electrophore-
togram. This is because the product was insoluble to the re-
action solution since we observed a reddish precipitate at the
bottom of the reaction ampules whenever the aglycon peak
disappeared in the electrophoretogram. This was further con-
firmed by the TLC separation of the reactants dissolved in
5% TFA-methanol solution, showing the time dependent in- 10) Bemisser J. N., “Acid-catalyzed Hydrolysis of Glycosides,” Advan.
Carbohydrate Chem., 1967, pp. 25—108.
11) Sato M., Suzuki Y., 2nd ICoff abstract, 1999, p. 146.
crease of the third component concomitantly with the de-
crease of both anthocyanin and the aglycon spots (data not
shown).