Color and antioxidant properties of cyanidin-based anthocyanin pigments

Article abstract of DOI:10.1021/jf0204811

A series of cyanidin-based anthocyanin pigments was investigated to determine the effect of structural variation on a number of chemical and physical properties: CIELAB color coordinates, visual detection thresholds, hydration constants (pK_H), and in vitro antioxidant activities (ORAC). In addition to individual isolated compounds, purified total pigment isolates from blackberry, elderberry, black carrot, red cabbage, and sweet potato were also examined. Acylation with cinnamic acids shifted color tonality (hue angle) to purple, and markedly increased pK_H and antioxidant activity, but lowered the visual detection threshold. Glycosidic substitution at the 5 position moved tonalities toward purple and decreased pK_H, and tended to lower the ORAC value, but raised the visual detection threshold. Increasing the number of sugar substituents at the 3 position also affected all of these parameters, however, the extent was not predictable. Antioxidant levels of purified anthocyanin extracts were much higher than expected from anthocyanin content indicating synergistic effect of anthocyanin mixtures.

Full text of DOI:10.1021/jf0204811

6172 J. Agric. Food Chem. 2002, 50, 6172−6181
Color and Antioxidant Properties of Cyanidin-Based
Anthocyanin Pigments
§
§
,§
FLORIAN C. STINTZING, ANGELA S. STINTZING, REINHOLD CARLE,*
BALZ FREI,‡ AND RONALD E. WROLSTAD†
Department of Food Science and Technology, Oregon State University, 100 Wiegand Hall,
Corvallis, Oregon 97331-6602; Oregon State University, Weniger Hall, Corvallis, Oregon 97331-6512;
and Institute of Food Technology, Section Plant Foodstuff Technology, Hohenheim University,
Garbenstrasse 25, 70599 Stuttgart, Germany
A series of cyanidin-based anthocyanin pigments was investigated to determine the effect of structural
variation on a number of chemical and physical properties: CIELAB color coordinates, visual detection
thresholds, hydration constants (pK ), and in vitro antioxidant activities (ORAC). In addition to individual
H
isolated compounds, purified total pigment isolates from blackberry, elderberry, black carrot, red
cabbage, and sweet potato were also examined. Acylation with cinnamic acids shifted color tonality
(
hue angle) to purple, and markedly increased pK
detection threshold. Glycosidic substitution at the 5 position moved tonalities toward purple and
decreased pK , and tended to lower the ORAC value, but raised the visual detection threshold.
H
and antioxidant activity, but lowered the visual
H
Increasing the number of sugar substituents at the 3 position also affected all of these parameters,
however, the extent was not predictable. Antioxidant levels of purified anthocyanin extracts were
much higher than expected from anthocyanin content indicating synergistic effect of anthocyanin
mixtures.
KEYWORDS: Colorants; anthocyanins; cyanidin; CIELAB; visual detection threshold; hydration constant;
ORAC; antioxidant; blackberry; elderberry; black carrot; red cabbage; sweet potato
INTRODUCTION
published regarding the antioxidant potencies of purified an-
thocyanin pigments and their mixtures. Wang et al. (14)
measured the oxygen radical absorbing capacities (ORAC) of
several purified anthocyanins and demonstrated that variation
in B-ring structure and glycosylation can have a marked effect
on antioxidant properties.
In addition to their roles as secondary metabolites in the
pigmentation of many flowers, fruits, vegetables, and grains (1),
anthocyanins have gained increasing interest as functional
compounds for coloring food (2), and as potent agents against
oxidative stress (3). Their success as natural alternatives for
artificial dyes mainly depends on their economic feasibility, their
chemical, biochemical, and physical stability during processing,
and their appearance at food pH (4). Varying pH levels induces
structural transformations (Figure 1) which affect both color
quality and intensity. The effect of B-ring substitution on color
appearance is long established, and there is considerable
information on the effect of acylation and glycosidic substitution
on pigment stability and a number of color properties (1).
Furthermore, there is increasing epidemiological evidence that
fruit- and vegetable-based diets are likely to improve the
antioxidant status of human beings (5, 6). A number of
investigators have shown a high correlation between total
anthocyanins, total phenolics, and the in vitro antioxidant
activities of fruit and vegetable extracts (7-13). Little has been
Our objective was to further examine the effect of structural
variation through acylation with cinnamic acids, as well as
position and extent of glycosidic substitution on antioxidant
activities and several color properties. We limited our investiga-
tion to cyanidin-based pigments and selected a range of natural
colorants and juice concentrates that could be sources for
pigments varying in sugar substitution patterns and acylation.
These included blackberry and elderberry juice concentrates,
and black carrot, sweet potato, and red cabbage extracts.
Individual pigments were purified from these sources by HPLC
and solid-phase extraction. Color parameters included lightness
(L*), chromatic tonality (hue angle, h°), and metric chroma (C*)
measured in the CIELAB scale. The color activity concept
developed by Hofmann (15) for Maillard reaction products and
subsequently used for anthocyanins (16) was also applied.
Because of the practical significance in food applications of
the equilibrium reaction between the flavylium cation and the
hemiketal form (Figure 1), the pKH was also determined. The
ORAC assay was used for measuring antioxidant properties.
*
To whom correspondence should be addressed (telephone +49-711-
4
59-2314; fax +49-711-459-4110; e-mail carle@uni-hohenheim.de).
§
Hohenheim University.
‡
Linus Pauling Institute.
†
Oregon State University.
1
0.1021/jf0204811 CCC: $22.00 © 2002 American Chemical Society
Published on Web 09/07/2002

Color and Antioxidant Properties of Cyanidin Pigments
J. Agric. Food Chem., Vol. 50, No. 21, 2002 6173
Figure 1. pH-Dependent structural transformations of anthocyanins (modified after (28)).
This assay has the advantage of directly assessing the ability
of a compound to quench free radicals close to physiological
pH.
(v/v) first yielded a colorless and then a red fraction. The red
anthocyanin eluate was collected, and it was repeatedly dissolved in
methanol and concentrated in vacuo, until complete removal of residual
acid. The obtained anthocyanin mixtures were diluted with purified
water containing 0.01% HCl for the isolation of different cyanidin
pigments by semi-prep HPLC. For color studies, the anthocyanin
mixture was repeatedly dissolved in methanol and concentrated in
vacuo, until complete removal of residual acid and water.
MATERIALS AND METHODS
Pigment Material. Elderberry and evergreen blackberry juice
concentrates were from IFF Flavors North America (Salem, OR), and
black carrot concentrate was provided by GNT USA, Inc. (Tarrytown,
NY). Red cabbage and purple sweet potato concentrates were supplied
by San-Ei Gen F. F. I., Inc. (New York, NY).
Solvents and Reagents. All solvents and reagents were of analytical
or HPLC grade. Phosphoric, formic, and acetic acids, methanol, and
acetonitrile were from Merck (Darmstadt, Germany). Hydrochloric acid
was from Baker Chemical Co. (Phillipsburg, NJ). â-Phycoerythrin was
from Sigma Chemical Co. (St. Louis, MO). 2,2′-Azobis(2-amidino-
propane) dihydrochloride (AAPH) was obtained from Wako Chemicals
USA Inc. (Richmond, VA). 6-Hydroxy-2,5,7,8-tetramethylchroman-
Pigment Isolation. Individual cyanidin derivatives (Figure 2) were
isolated from the prepurified samples by semipreparative HPLC
applying different gradient systems (see below). Solvents were evapo-
rated in vacuo at 30 °C, and purities were checked using HPLC systems
I-IV (see below) and monitoring spectra of the peaks at 280, 320,
and 520 nm. The ratio of peak areas of the isolated pigment and total
peak area at 280 nm was taken for purity determination (min. 97%).
The purified pigments were further concentrated in vacuo at 30 °C.
Acid and water were removed by repeated addition of methanol. Pure
pigments were dried in a desiccator and stored at -70 °C in a screw-
top test tube until further analyses.
2
-carboxylic acid (Trolox) and 2,4,6-tri(2-pyridyl)-s-triazine (TPTZ)
were purchased from Fluka (Buchs, Switzerland).
Anthocyanin Semi-Purification. Fruit or vegetable concentrates
were diluted with purified water at ratios (w/w) of 1/5 for blackberry,
High-Performance Liquid Chromatography (HPLC). Analytical
HPLC of Anthocyanins: Apparatus. A high-performance liquid chro-
matograph Perkin-Elmer Series 400, equipped with a 50 µL injection
loop and a Hewlett-Packard 1040A photodiode array detector was used.
Data analysis was performed with Hewlett-Packard HPLC-2D Chem-
Station software.
1
/20 for elderberry, 1/30 for black carrot, and 1/50 for red cabbage
and sweet potato prior to semi-purification on high-load C18-Sep-Pak
cartridges (Waters Assoc., Milford, MA), previously activated with
methanol followed by aqueous 0.01% HCl (17). Using this method,
samples (1 mL) were desalted with three volumes of acidified water
before removal of nonanthocyanin phenolics by rinsing with 3 vol of
ethyl acetate. Elution with methanol containing 0.01% aqueous HCl
Columns, Mobile Phases, and HPLC Conditions. Solvents and
samples were filtered through a 0.45-µm Millipore filter type HA
(Millipore Corp., Bedford, MA). Chromatographic analysis was carried
out using an analytical scale (5-µm, 250 mm × 4.6 mm i.d.) Prodigy

6174 J. Agric. Food Chem., Vol. 50, No. 21, 2002
Stintzing et al.
Figure 2. Chemical structures of cyanidin derivatives investigated. Structures for pigments are given as previously reported (21, 22, 26). Pigment
assignments, sources, and mass data are given in Table 1.
C
18 reversed-phase column (Phenomenex, Torrance, CA), fitted with a
5% acetonitrile, 10% glacial acetic acid, and 1% phosphoric acid (85%)
in water (v/v/v). Separation of anthocyanins was achieved within 25
min. The first 5 min were performed isocratically with 100% A,
followed by a linear gradient from 100% A to 80% A in 15 min and
then to 60% A in 5 min. The same gradient was used for the purity
check of cyanidin 3-glucoside.
guard column C18 ODS (4 mm × 3.0 mm i.d.). Simultaneous monitoring
was performed at 280, 320, and 520 nm at a flow rate of 1 mL/min.
System I (Blackberry and Elderberry Extracts, Cyanidin 3-Gluco-
side). As described before (18), mobile phase B consisted of 100%
HPLC-grade acetonitrile, whereas mobile phase A was a mixture of

Color and Antioxidant Properties of Cyanidin Pigments
J. Agric. Food Chem., Vol. 50, No. 21, 2002 6175
System II (Black Carrot Extract and Pigments). Mobile phase B
consisted of 90/10 HPLC-grade acetonitrile/purified water (v/v) and
mobile phase A was 10% aqueous formic acid. Separation of antho-
cyanins was achieved within 25 min, starting with 93% A to 85% A in
were calculated as ratio of total anthocyanin content of the extract
determined as cyanidin 3-glucoside (see below) and the relative peak
area of the respective anthocyanin at 520 nm. Color contribution was
expressed as percentage of the color activity value in relation to the
dilution factor of the extract (15).
1
8 min, and then to 50% A in 7 min. The same system was used for
purity control of cyanidin 3-xylosyl-galactoside, cyanidin 3-xylosyl-
glucosyl-galactoside, and its sinapoyl and feruloyl derivatives, respec-
tively.
System III (Red Cabbage Extract, Cyanidin 3-Sophoroside-5-
Glucoside, and Its Diacylated DeriVatiVe). Mobile phase B consisted
of 90/10 HPLC-grade acetonitrile/purified water (v/v), and mobile phase
A was 10% aqueous formic acid. Separation of anthocyanins was
achieved within 50 min. Gradient steps were from 92% A to 89% A,
then to 85% A in 15 min, followed by an isocratic elution for 10 min,
and finally to 67% A in 15 min. For purity control of cyanidin
Color Analyses. Color parameters (CIELAB) were recorded with a
ColorQuest HunterLab spectrocolorimeter equipped with HunterLab
Universal Software Version 3.0 (HunterLab, Hunter Associates Labo-
ratories Inc., Reston, VA). Purified anthocyanin extracts and purified
pigments were dissolved in three different buffers (pH 0.45, 3.5, 6.0),
and the absorbance was normalized to 0.5 after 30 min. Buffered
pigment solutions were filled in a 0.25 cm path length optical glass
cell (Hellma, M u¨ llheim, Germany) and CIE L*, a*, and b* values were
determined immediately in the total transmission mode in triplicate
2
2 1/2
using Illuminant C and 10° observer angle. Chroma (a* + b* ) and
hue angle (arctan b*/a*) were calculated from CIELAB a* and b*
coordinates.
3-sophoroside-5-glucoside and its diacylated derivative substituted with
coumaric and sinapic acids, the gradient was 92% A to 89% A in 10
min, then to 67% A in 15 min, and finally an isocratic elution for 5
min.
Determination of pH. For pH measurements, a Corning pH-meter
3
40 (Corning Inc., Corning, NY) equipped with a Corning electrode
System IV (Sweet Potato Extract). Mobile phase B consisted of
“
3-in-1” Combo w/RJ with temperature correction was used under
9
0/10 HPLC-grade acetonitrile/purified water (v/v), and mobile phase
continuous stirring of the solution.
A was 10% aqueous formic acid. Separation of anthocyanins was
achieved within 60 min. The first 10 min were performed isocratically
with 92% A, followed by a gradient to 87% within 18 min, and then
to 85% A in 12 min, and finally to 67% A in 20 min.
Determination of Hydration Constants. An aqueous hydrochloric
acid solution of pH 0.38 was prepared. Purified pigments or extracts
were taken up in this solution to reach an initial absorbance of 0.4-
0.8 at λmax. Solutions were equilibrated for 30 min at 24 ( 1 °C in the
dark. Through addition of 8 N NaOH under constant stirring, the initial
pH was set to 0.45 ( 0.01 and the absorbance was measured after 15
min of pH equilibration. Different concentrations of aqueous NaOH
(8 N, 4 N, 2 N, 0.5 N, 0.125 N, and 0.03125 N) were used for titration.
After each pH change, the absorbance was measured in a 1 cm path-
length quartz cell (Hellma, M u¨ llheim, Germany) using a Shimadzu 160
UV spectrophotometer (Kyoto, Japan) with a temperature-controlled
cell holder (TCC-240A). Titration curves were obtained by plotting
absorbance values versus pH from 0.45 to 6 at 500 nm. A curve fit
was performed using Microsoft Excel 97, and the resulting polynomial
term of sixth power was processed with Maple V (Release 5, Waterloo
SemipreparatiVe HPLC of Anthocyanins: Apparatus. A semiprepara-
tive Dynamax Rainin model SD-300 high-performance liquid chro-
matograph was used, and it was equipped with a 1 mL injection loop
and a Hewlett-Packard 1040A photodiode array detector. Data were
analyzed with Hewlett-Packard HPLC-2D ChemStation software.
Columns, Mobile Phases, and HPLC Conditions. Solvents were
filtered through a 0.45 µm Millipore filter type HA (Millipore Corp.,
Bedford, MA) before use. Semipreparative HPLC was performed using
a semipreparative scale (5 µm, 250 mm × 21.4 mm i.d.) Microsorb
C
18 reversed-phase column (Rainin Instrument Co., Inc., Emeryville,
CA), fitted with a guard module C18 ODS (50 mm × 21.4 mm i.d.).
Data were collected at 520 nm at a flow rate of 12 mL/min.
System V (Cyanidin 3-Glucoside). Mobile phase B was 100% HPLC-
grade acetonitrile, and phase A consisted of 1% aqueous formic acid.
A linear gradient was applied from 80% A to 75% A in 10 min.
System VI (Black Carrot Pigments). Mobile phase B was 100%
HPLC-grade acetonitrile, and mobile phase A consisted of 10% aqueous
formic acid. A linear gradient was followed from 88% A to 83% A in
Maple Inc.) to calculate the pH of the hydration constant. The pK
H
values were obtained from the mean of duplicate measurements.
Total Anthocyanin Content. The monomeric anthocyanin content
was determined using a pH-differential method (19). Spectral measure-
ments were performed on a Shimadzu 160 UV spectrophotometer
(Kyoto, Japan) at 510 and 700 nm with 1 cm path length disposable
1
3 min.
cuvettes. Anthocyanin contents were calculated using the molar
System VII (Cyanidin 3-Sophoroside-5-Glucoside and Its Diacylated
-1
-1
absorptivity of cyanidin 3-glucoside (26 900 L cm mg ) and a
DeriVatiVe). Mobile phase B was 100% HPLC-grade acetonitrile, and
mobile phase A consisted of 10% aqueous formic acid. A linear gradient
was used from 90% A to 77% A in 3 min followed by isocratic elution
for 9 min.
Electrospray Mass Spectrometry (MS). Pigment identity was
checked using electrospray MS with a Perkin-Elmer SCIEX API III+
mass spectrometer equipped with an ion spray interface (ISV) 4700,
orifice voltage of 120 eV) operated in the positive-ion mode. Purified
pigments were dissolved in 0.1% aqueous TFA/100% MeOH (1/1, v/v)
for direct injection.
Assessment of Visual Detection Threshold. Purified anthocyanin
extracts or purified pigments were dissolved in 1 mL of pH 3.5 buffer
and equilibrated for 30 min prior to stepwise dilution. Minimal pigment
concentration at which a difference between two solutions containing
purified water and the anthocyanin solution can still be visually
discerned is defined as the visual detection threshold (15). The dilution
factor was determined by diluting purified pigment extracts with buffer
until virtually no color was perceived. Visual detection thresholds were
obtained in a triangle test by duplicate determination as described earlier
-1
molecular weight of 449.2 g mol
.
Antioxidant Capacity. Purified extracts and isolated pigments (4.7-
.6 mM) were dissolved in methanol. Solutions were subsequently
9
diluted with buffer (1/10, v/v) prior to analysis.
Oxygen Radical Absorbing Capacity (ORAC) Assay. The procedure
reported by Cao et al. (20) was used and adapted for a Cytofluor 4000
microplate fluorometer (PerSeptive Biosystems, Framingham, MA). In
the final reaction mixture, â-phycoerythrin acted as target for the peroxyl
radicals generated by AAPH. The decrease of phycoerythrin fluores-
cence in the absence (control) and presence of antioxidants was
monitored for 2 h at 2 min intervals at excitation and emission
wavelengths of 485 and 585 nm, respectively. Trolox was used as a
standard, and radical absorbing capacities were calculated using the
area under the control curve and the quenching curve of â-phycoerythrin
in the presence of the respective anthocyanin solution. Results were
expressed as µmol Trolox equivalents per µmol pigment. For the
extracts, pigment concentrations were calculated on the basis of cyanidin
3-glucoside as described above. The relation between ORAC values
and anthocyanin concentration was evaluated by using linear regression
analysis (Microsoft Excel 97) with three data points for each of the
pigments. Using the least-squares method, a straight line (ORAC [µM
Trolox equivalents] ) m × anthocyanin concentration [µM] + b) was
found as the best fit for the data points with a correlation coefficient
above 0.98. Standard errors of intercept (b) and slope (m) were obtained
from regression analyses (14).
(15, 16). The samples (2 mL total volume) were placed in plastic
cuvettes (1 cm path length) and inspected in a light booth (Macbeth
Color Identification Lamp “Executive” Type B BX 324, Macbeth Corp.,
Newburgh, NY) against a white background in diffuse daylight mode
(Northsky daylight, D65). The color activity value is defined as the ratio
of pigment concentration to the detection threshold of the individual
pigment (15). Concentrations of individual pigments in the extracts

6176 J. Agric. Food Chem., Vol. 50, No. 21, 2002
Stintzing et al.
Table 1. Pigment Assignments and Molecular Masses of Isolated Anthocyanins from Different Sources
number
anthocyanin
source
m/z [M]+
1
2
3
4
5
6
7
cyanidin 3-glucoside
cyanidin 3-xylosyl-galactoside
cyanidin 3-xylosyl-glucosyl-galactoside
cyanidin 3-sinapoyl-xylosyl-glucosyl-galactoside
cyanidin 3-feruloyl-xylosyl-glucosyl-galactoside
cyanidin 3-sophoroside-5-glucoside
blackberry
449.2
581.2
743.4
949.4
919.6
773.6
1125.8
black carrot
black carrot
black carrot
black carrot
red cabbage
red cabbage
cyanidin 3-coumaroyl-sinapoyl-sophoroside-5-glucoside
Table 2. Visual Detection Thresholds and Color Contributions of Individual Anthocyanins in Pigment Extracts
VDTb
[mg/L]
% peak area
at 520 nm
% color
contribution
source
DFa
anthocyanin
blackberry
elderberry
1596
2422
cyanidin 3-glucoside
cyanidin 3-glucoside
cyanidin 3-xylosyl-galactoside
cyanidin 3-xylosyl-glucosyl-galactoside
cyanidin 3-sinapoyl-xylosyl-glucosyl-galactoside
cyanidin 3-feruloyl-xylosyl-glucosyl-galactoside
cyanidin 3-sophoroside-5-glucoside
1.3
1.3
0.9
2.4
0.9
0.4
3.6
3.6
2.0
82.6
28.6
44.1
14.9
27.5
13.5
1.8
52.0
18.0
36.1
4.5
21.2
22.6
0.6
black carrot
black carrot
black carrot
black carrot
sweet potato
red cabbage
red cabbage
19600
19600
19600
19600
25003
26147
26147
cyanidin 3-sophoroside-5-glucoside
cyanidin 3-coumaroyl-sinapoyl-sophoroside-5-glucoside
17.6
25.2
7.0
18.2
a
DF: Dilution factor. b VDT: Visual detection threshold.
RESULTS AND DISCUSSION
trates allowed assessment of their relative color strengths.
Extracts mainly composed of acylated anthocyanins (red cab-
bage and sweet potato) required 8- to 16-fold higher dilution
factors than blackberry and elderberry. This observation could
be substantiated by the examination of the purified pigments.
Visual detection thresholds of anthocyanins acylated with
cinnamic acids were generally lower, indicating a higher
tinctorial strength. Acylation of cyanidin 3-xylosyl-glucosyl-
galactoside with sinapic (4) and ferulic (5) acid resulted in a
Characterization of Anthocyanin Material. The purified
anthocyanins examined in this investigation and their sources
are listed in Table 1. Pigment purities were checked as described
above, and their identities were confirmed by chromatographic
and mass spectrometric analyses. Cyanidin 3-glucoside (1), the
most commonly occurring anthocyanin pigment in nature, was
isolated from commercial evergreen blackberry (Rubus lacin-
iatus Willd.) extract. Additional pigments previously identified
in this extract (18) included cyanidin 3-rutinoside (2.4%),
cyanidin substituted with xylose (7.7%), cyanidin 3-malonyl-
glucoside (3.6%), and cyanidin 3-dioxalyl-glucoside (3.0%).
Black carrot (Daucus carota L. ssp. satiVus var. atrorubens
Alef.) extract served as the source for cyanidin 3-xylosyl-
galactoside (2; 44.1% of total peak area), cyanidin 3-xylosyl-
glucosyl-galactoside (3; 14.9%), cyanidin 3-xylosyl-glucosyl-
galactoside acylated with sinapic acid (4; 27.5%), or ferulic acid
3
-fold and 6-fold increase, respectively. This is consistent with
other reports of a hyperchromic effect through cinnamic acid
substitution (28, 29). In contrast, Degenhardt et al. (16) found
that nonacylated wine pigments showed increased tinctorial
strength when compared to the respective acylated compounds.
These pigments were acylated monoglycosides, however, whereas
we investigated acylated diglycosides. Dangles et al. (28)
reported that two sugars are necessary as spacers to allow folding
of the acyl moiety leading to higher chromophore integrity.
Diacylation of cyanidin 3-sophoroside-5-glucoside with cou-
maric and sinapic acids (7) did not even reduce the threshold
as much as monoacylation of pigment 3. This was surprising,
as diacylation would be expected to be more effective in
protecting both faces of the flavylium cation from nucleophilic
attack (30). Possibly, 5-glycosylation hinders free rotation of
the aromatic acyl moiety, thereby reducing the shielding effect
on the second face of the flavylium nucleus. In line with another
study (16), 3,5-diglycosides had lower color strength than
(5; 13.5%). In contrast to the report of Gl a¨ ssgen et al. (21),
hydroxybenzoyl- and coumaroyl-derivatives were not detected.
Red cabbage (Brassica oleracea L.) extract was a source for
cyanidin 3-sophoroside-5-glucoside (6) and a derivative sub-
stituted with sinapic and coumaric acids (7). Fifteen pigments
were separated in this extract, with a predominance of acylated
(81.1%) over nonacylated (18.9%) anthocyanins. This pigment
profile was in accord with previous reports (22-24). Sweet
potato (Ipomoea batatas [L.] Lam.) pigments consisting of
peonidin- and cyanidin-3-sophoroside-5-glycosylated structures
3
-glycosides. When comparing cyanidin mono-, di-, and tri-
(25, 26) was considered in this study as an alternative to red
glycosides, diglycosylation with xylose and galactose reduced
the visual detection threshold, whereas additional substitution
with glucose increased the threshold. It is evident that the
number and position of glycosidic substitution(s), as well as
the type of sugar, can affect color strength.
cabbage without negative flavor. Its complex pattern comprising
more than 15 anthocyanins was characterized by 90.2% acylated
and 9.8% nonacylated structures. Consistent with Inami et al.
(27), elderberry (Sambucus nigra L.) extract contained exclu-
sively nonacylated anthocyanins: cyanidin 3-sambubioside
(
(
54.7%), cyanidin 3-glucoside (28.6%), cyanidin 3,5-diglucoside
12.9%), and cyanidin 3-sambubioside-5-glucoside (3.8%).
Visual Detection Threshold. The methodology of Degen-
Interestingly, the color contribution of individual anthocyanins
is, in almost all cases, less than their respective peak area
percentage (Table 2). This was particularly true for nonacylated
pigments. Although cyanidin 3-glucoside represented the major
compound of blackberry extract, its color impact was only about
50%. The same was true for black carrot, for which 100% of
hardt et al. (16) was used to determine visual detection
thresholds and the color contribution of individual anthocyanins
to the total extract (Table 2). Dilution factors for the concen-

Color and Antioxidant Properties of Cyanidin Pigments
J. Agric. Food Chem., Vol. 50, No. 21, 2002 6177
Table 3. Colorimetric and Spectral Characteristics of Isolated Anthocyanins
anthocyanin
L*
C*
h°
λmax
pH 0.45
92.9
92.7
92.6
91.7
91.9
93.2
91.6
cyanidin 3-glucoside
cyanidin 3-xylosyl-galactoside
cyanidin 3-xylosyl-glucosyl-galactoside
cyanidin 3-sinapoyl-xylosyl-glucosyl-galactoside
cyanidin 3-feruloyl-xylosyl-glucosyl-galactoside
cyanidin 3-sophoroside-5-glucoside
8.9
9.3
9.1
10.5
10.0
8.7
11.9
5.0
5.8
351.1
355.2
0.3
510
513
513
526
525
511
532
cyanidin 3-coumaroyl-sinapoyl-sophoroside-5-glucoside
11.2
339.7
pH 3.50
92.5
92.3
92.1
91.5
91.6
92.6
91.4
cyanidin 3-glucoside
cyanidin 3-xylosyl-galactoside
cyanidin 3-xylosyl-glucosyl-galactoside
cyanidin 3-sinapoyl-xylosyl-glucosyl-galactoside
cyanidin 3-feruloyl-xylosyl-glucosyl-galactoside
cyanidin 3-sophoroside-5-glucoside
9.1
9.3
9.1
10.4
10.0
8.4
8.0
2.6
3.9
348.8
353.7
355.7
333.4
512
515
515
529
527
514
536
cyanidin 3-coumaroyl-sinapoyl-sophoroside-5-glucoside
10.9
pH 6.00
88.9
89.6
89.8
89.9
90.1
92.2
88.6
cyanidin 3-glucosidea
cyanidin 3-xylosyl-galactoside
cyanidin 3-xylosyl-glucosyl-galactoside
cyanidin 3-sinapoyl-xylosyl-glucosyl-galactoside
cyanidin 3-feruloyl-xylosyl-glucosyl-galactoside
4.1
6.4
5.1
8.3
7.8
4.9
7.8
304.1
341.3
340.7
313.3
316.1
321.5
268.3
550
541
542
547
549
545
587
a
a
a
cyanidin 3-sophoroside-5-glucoside
cyanidin 3-coumaroyl-sinapoyl-sophoroside-5-glucoside
a
Data given do not represent a stable appearance of the solution because degradation proceeded quickly.
the peak area contributed to only about 85% of the overall
coloring strength. Because coelution of anthocyanins was
excluded, this discrepancy can be explained only by interactions
between the individual anthocyanins requiring higher dilution
factors for the extracts. While the visual detection thresholds
were determined for isolated pigments, extract dilution factors
were assessed with intermolecular arrangements between dif-
ferent anthocyanin structures being possible (31). For red
cabbage, the color contribution of two pigments accounting for
about 43% of the peak area was only 25%. From this it can be
deduced that other pigments, in particular the acylated antho-
cyanins, constituted the major color fraction. It is worth noting
that cyanidin 3-feruloyl-xylosyl-glucosyl-galactoside (5) from
black carrot accounted for only half of the peak area compared
to the sinapoyl-derivative (4), but equally contributed to color
appearance (Table 2). When applying the color activity concept
to anthocyanins, it should be remembered that quantification
of the flavylium cation is carried out at low pH, whereas
thresholds are determined at the pH of the food application.
Transition of the flavylium ions to colorless structures will occur
to a varying extent at higher pH values depending on the nature
of the pigment investigated (32, Figure 1).
Table 4. Colorimetric and Spectral Characteristics of Purified
Anthocyanin Extracts
anthocyanin extract
L*
C*
h°
λmax
pH 0.45
pH 3.50
pH 6.00
blackberry
elderberry
black carrot
red cabbage
sweet potato
92.1
92.0
91.8
92.0
91.8
9.0
9.2
10.3
10.7
11.2
10.6
4.7
358.3
348.9
348.2
510
510
520
520
520
blackberry
elderberry
black carrot
red cabbage
sweet potato
90.9
90.9
91.2
91.1
91.1
8.5
8.5
10.3
11.0
11.3
17.4
10.9
356.2
340.4
342.3
510
520
520
530
530
blackberrya
elderberrya
black carrot
red cabbage
sweet potato
88.78
88.33
88.84
89.20
89.32
5.74
30.9
17.4
325.1
304.4
321.4
530
540
550
550
550
4.63
7.01
8.36
9.23
a
Data given do not represent a stable appearance of the solution because
degradation proceeded quickly.
Color Attributes of Anthocyanin Solutions. In determining
CIELAB coordinates, pigments and extracts were diluted to a
common absorbance value of 0.5, which accounted for the
similarity in lightness (L*) for the different samples (Tables 3
and 4). Because chroma was only slightly affected by different
substitution patterns, it was of particular interest how antho-
cyanin structure affected hue angle. The latter is expressed on
a 360° color wheel where 0°/360° ) purplish-red, 90° ) yellow,
angle, shifting the color from red to purplish-red and purple.
Substitution of cyanidin 3-xylosyl-glucosyl-galactoside with
sinapic acid had a greater effect than with ferulic acid. The
diacylated pigment from red cabbage exhibited a pronounced
shift in hue angle compared to the nonacylated parent compound
of 21° at pH 0.45 and of 22° at pH 3.5. At pH 6, the diacylated
cyanidin derivative even displayed a blue appearance. Glycosidic
substitution at 3 or 5 position decreased hue angle. Glycosylation
at the 3 position with a di- or trisaccharide reduced the hue
angle slightly by 4 to 7°. The color indices for nonacylated
pigments at pH 6.0 need to be regarded as transient because
these pigments rapidly degraded at that pH.
1
80° ) bluish-green, and 270° ) blue. Although it is well
established that methoxylation and hydroxylation of the B-ring
has a predictable effect on color and wavelength of maximum
absorption (29), this investigation examined the effect of
acylation with cinnamic acids and glycosidic substitution on
these parameters. At pH 0.45, color quality of the pigments can
be completely ascribed to the flavylium ion (33) allowing
assessment of the impact of structural modifications (Table 3).
Acylation with cinnamic acids had a pronounced effect on hue
The color properties of the extracts (Table 4) are in accord
with the values determined for the purified pigments. For
extracts predominantly consisting of nonacylated anthocyanins

6
178 J. Agric. Food Chem., Vol. 50, No. 21, 2002
Stintzing et al.
ref
H
Table 5. Hydration Constants (pK ) of Individual Anthocyanins and Pigment Extracts
pKH
literature data
anthocyanin
cyanidin 3-glucoside
3.01 ± 0.04
3.13 ± 0.02
3.26 ± 0.06
3.79 ± 0.08
4.42 ± 0.20
2.32 ± 0.01
3.12 ± 0.01
3.02 ± 0.06
3.41 ± 0.01
3.37 ± 0.05
4.41 ± 0.05
(38)
(34)
(34)
(34)
cyanidin 3-xylosyl-galactoside
cyanidin 3-xylosyl-glucosyl-galactoside
cyanidin 3-sinapoyl-xylosyl-glucosyl-galactoside
cyanidin 3-feruloyl-xylosyl-glucosyl-galactoside
cyanidin 3-sophoroside-5-glucoside
cyanidin 3-coumaroyl-sinapoyl-sophoroside-5-glucoside
anthocyanin source
-
blackberry
elderberry
black carrot
red cabbage
3.07 ± 0.02
2.96 ± 0.02
3.67 ± 0.06
2.69 ± 0.00
3.15 ± 0.02
2.99 ± 0.06
(35)
sweet potato
H
Table 6. Polynomial Fits for pK -Determination of Isolated Anthocyanins and Pigment Extracts
coefficientsa
ab
bb
cb
d b
eb
f b
gb
r c
anthocyanin
cyanidin 3-glucoside
−0.0022
0.0285
0.0176
0.0662
−0.0233
−0.0006
0.0250
−0.0013
0.0009
−0.1154
−0.0519
−0.3426
0.1122
0.0247
−0.1323
0.0148
0.0007
−0.0008
0.0138
0.1541
0.2372
0.0370
0.0528
0.1357
−0.0304
0.7972
−0.3163
−0.1643
0.2966
−0.0744
−0.0309
−0.0146
−0.0684
−0.5110
−0.6930
−0.1349
−0.1874
0.0565
0.2398
−0.8751
0.5013
0.3695
−0.2921
0.1599
0.0934
0.0572
0.1531
0.7534
0.8969
0.2129
0.2835
−0.2507
−0.3277
0.3573
−0.4724
−0.4092
0.0492
−0.1671
−0.1107
−0.0875
−0.1594
−0.5633
−0.5498
−0.1656
−0.2096
1.1516
1.2018
0.9446
1.0519
1.1186
0.8703
0.3981
0.3366
0.3807
0.3056
0.6902
0.6653
0.3817
0.5152
0.99
0.99
0.99
0.99
0.99
0.99
0.99
0.99
0.99
0.99
0.99
0.99
0.99
0.99
−
0.0015
−0.0047
.002
−0.00007
cyanidin 3-xylosyl-galactoside
0
cyanidin 3-xylosyl-glucosyl-galactoside
cyanidin 3-sinapoyl-xylosyl-glucosyl-galactoside
cyanidin 3-feruloyl-xylosyl-glucosyl-galactoside
cyanidin 3-sophoroside-5-glucoside
−0.0017
0.00004
0.00009
−
−0.00003
.00005
0.0011
.0022
0.0002
.0003
0.0005
0
−0.0013
−0.0211
−0.0373
−0.0046
−0.0067
0
cyanidin 3-coumaroyl-sinapoyl-sophoroside-5-glucoside
0
anthocyanin extract
blackberry
−0.0028
0.0012
−0.0007
0.0012
−0.0004
0.0388
0.0173
0.0077
0.0147
0.0058
0.0071
−0.0143
−0.0192
−0.0147
0.0005
−0.1953
−0.0839
−0.0168
−0.0563
−0.0223
−0.0286
0.1062
0.1458
0.1082
0.0223
0.4272
0.1575
−0.0547
0.0427
0.0006
0.0163
−0.3498
−0.5072
−0.3788
−0.1510
−0.4224
−0.1122
0.2044
0.0724
0.0999
0.0747
0.4694
0.7783
0.6096
0.3141
0.0699
−0.0831
−0.3047
−0.2485
−0.1917
−0.1818
−0.3242
−0.5877
−0.4985
−0.3202
0.9915
0.9545
0.9952
1.2166
0.9386
0.9902
0.8758
0.9561
0.8461
0.8118
0.99
0.99
0.99
0.99
0.99
0.99
0.99
0.99
0.99
0.99
−
elderberry
−
black carrot
red cabbage
sweet potato
−0.0005
0.0007
0.0009
0.0008
0.0002
−
a
6
5
4
3
2
b
Coefficients for the polynomial term y ) ax + bx + cx + dx + ex + fx + g, with y for the absorbance and x for pH. The two coefficients given are for the
c
duplicates. Regression coefficient.
such as blackberry and elderberry, hue angles ranged between
transformed into the colorless hemiketal corresponding to a 50%
decline in absorbance. Because absorbance was measured at
500 nm, interference with the blue quinoidal base could be
excluded (34). This method permitted determination of hydration
constants of both purified extracts and isolated pigments (Table
5). The best fit for the titration curves was obtained with a
polynomial term of sixth power (Table 6) from which the pHs
of the hydration constants were calculated using Maple V.
Assuming an equilibrium between the flavylium cation and the
hemiketal structures, the pKH can be described by the following
equation (35, 36):
5
and 17°, corresponding to red color and increased (i.e., tonality
moved yellower) with higher pH. Red cabbage and sweet potato
extracts were composed primarily of highly acylated anthocya-
nins and were characterized by h° values indicative of purple
tonalities. Black carrot extract consisted of nearly equivalent
proportions of nonacylated and acylated anthocyanins and
displayed intermediate h° values. Although red cabbage and
sweet potato extracts had similar tonalities at pH 0.45 and pH
3
.5, there was a difference of 17° at pH 6.0. It is worth noting
that hue angle and values of maximum absorbance did not
necessarily correspond. This phenomenon is due to the consid-
erable variation in absorption curve shape throughout the extent
of the visible spectrum.
chemiketal
pK ) -log
+ pH
H
cflaVyliumcation
Hydration Equilibrium Constants. pKH values were deter-
mined by titration starting at pH 0.45 where 100% of the
pigment would be present as the flavylium cation (33). The pKH
was reached when 50% of the red flavylium form had been
By comparing the values for nonacylated 3-glycosides with
cyanidin-3-sophoroside-5-glucoside (Table 5), it is clear that
substitution at the 5 position markedly lowered the hydration

Color and Antioxidant Properties of Cyanidin Pigments
J. Agric. Food Chem., Vol. 50, No. 21, 2002 6179
Table 7. Peroxyl Radical Absorbing Capacities for Pigment Extracts and Isolated Anthocyanins
coefficientsa
m (slope)b
b (intercept)b
r c
anthocyanin extract
blackberry
elderberry
black carrot
red cabbage
sweet potato
13.868 ± 0.093
9.220 ± 0.225
9.154 ± 0.033
9.781 ± 0.425
10.011 ± 0.398
−0.232 ± 0.172
0.977 ± 0.895
0.745 ± 0.099
−0.581 ± 1.422
2.704 ± 1.082
0.999
0.999
0.999
0.999
0.999
anthocyanin
cyanidin 3-glucoside
cyanidin 3-xylosyl-galactoside
cyanidin 3-xylosyl-glucosyl-galactoside
cyanidin 3-sinapoyl-xylosyl-glucosyl-galactoside
cyanidin 3-feruloyl-xylosyl-glucosyl-galactoside
cyanidin 3-sophoroside-5-glucoside
2.257 ± 0.135
2.255 ± 0.049
1.479 ± 0.123
2.020 ± 0.160
3.035 ± 0.035
1.581 ± 0.061
2.852 ± 0.162
2.732 ± 1.220
1.119 ± 0.536
2.321 ± 3.159
−0.057 ± 1.599
0.556 ± 0.272
2.167 ± 1.012
1.279 ± 1.857
0.996
0.999
0.996
0.997
0.999
0.999
0.998
cyanidin 3-coumaroyl-sinapoyl-sophoroside-5-glucoside
a
The coefficients are given for the term ORAC [µM Trolox equivalent] ) m × anthocyanin concentration [µM] + b. b The slopes (m) are significantly different from zero
c
(
p < 0.05) whereas the intercepts (b) do not significantly differ from zero (p > 0.05). Therefore, the antioxidant capacity is directly expressed by the slope (m). Multiple
correlation coefficient.
constant. Others have reported similar decreases when compar-
anthocyanin structure is dependent on the acidity of the medium
(Figure 1) with the pseudobase, cis- and trans-chalcones, and
quinoidal bases predominating at pH 7 (32). The chalcones and
quinoidal bases display double bonds in conjugation with a keto
group, a structural element defined by Bors et al. (44) to be
essential for efficient antioxidant action. Anthocyanin pigments
are known to be effective quenchers of free radicals, and a
number of investigators have reported high linear correlations
between total anthocyanin pigment content and ORAC values
(10-13, 45). Wang et al. (14) studied the effect of glycosylation
and variation in anthocyanin B-ring structure on ORAC values.
B-ring structure had a marked effect on antioxidant activity with
ortho-hydroxylation and methoxylation substantially increasing
antioxidant activity. Anthocyanidins had higher ORAC values
than their glycosides, which is to be expected because the
aglycons are very unstable and highly reactive. We took a very
similar experimental approach, focusing on the effect of
cinnamic acid acylation and glycosidic substitution patterns.
Considering this good correlation (r > 0.99) between ORAC
and anthocyanin content in this study (Table 7), higher pigment
concentrations went along with higher antioxidant capacities.
As assessed by regression analysis, this dose-dependent relation-
ing 3-glucosides with their corresponding 3,5-diglucosides (37-
4
0). As described previously, glycosylation at the 5 position
results in decreased electron-density of the pyrilium ring which
favors nucleophilic attack by water, thereby enhancing hemiketal
formation (41, 42). A higher number of sugar residues at the 3
position increased the hydration constant to a small degree.
Acylation with cinnamic acids substantially raised the hydration
constant. The ferulic acid moiety of cyanidin 3-xylosyl-glucosyl-
galactoside increased the hydration constant more than the
sinapoyl residue. Acylation of cyanidin 3-sophoroside-5-glu-
coside with sinapic and coumaric acids only partially compen-
sated for the destabilizing effect of 5-substitution.
By comparing the hydration constants of individual antho-
cyanins with those of their respective sources, it is evident that
cyanidin 3-glucoside was largely responsible for the value of
blackberry extract (Table 5). Although blackberry contains
minor acylated pigments (18), their concentration was probably
too low to have a measurable impact. Black carrot and
blackberry extracts contain exclusively 3-substituted cyanidin
pigments. The markedly higher hydration constant for black
carrot can be attributed to the substantial degree of cinnamic
acid acylation (41.0%). Whereas red cabbage has a high
proportion of acylated anthocyanins (81.1%), its hydration
constant is much lower than those for black carrot, blackberry,
and elderberry. This can most likely be attributed to glycosidic
substitution at the 5 position. Although sweet potato anthocya-
nins are also characterized by 3,5-substitution, the higher degree
of acylation (90.2%) compensates for 5-glucosylation giving it
an intermediate position between red cabbage and black carrot.
The hydration constants obtained in this work are in good
agreement with those found by others (35, 38). The pKH values
reported by Redus et al. (34) showed the same trend, but differed
in absolute numbers (Table 5). Hydration constants have very
practical ramifications because higher values will translate into
substantially greater color intensity at the pH of most food
systems.
2
ship could be adequately described by a straight line (R > 0.99,
data not shown). Our results (Table 7) also show that the
intercepts (b) were not significantly different from zero (p >
0.05), thus antioxidant activity could be directly expressed by
the slope (m). Whereas cyanidin 3-glucoside and cyanidin
3-xylosyl-galactoside had equivalent values, addition of a third
sugar to form cyanidin 3-xylosyl-glucosyl-galactoside reduced
the antioxidant level. Cyanidin 3-sophoroside-5-glucoside and
cyanidin 3-xylosyl-glucosyl-galactoside had similar values,
therefore, one cannot conclude whether the reduction is due to
5-glycosylation or the presence of a third sugar residue.
However, Terahara et al. (46) found that cyanidin 3-sophoroside-
5-glucoside had lower antioxidant activity than the correspond-
ing 3-monoglucoside using the â-carotene bleaching assay. The
more striking finding is the impact of cinnamic acid acylation:
esterification of cyanidin 3-xylosyl-glucosyl-galactoside with
sinapic acid distinctly increased the ORAC value, whereas
ferulic acylation doubled the antioxidant activity. Diacylation
of cyanidin 3-sophoroside-5-glucoside with coumaric and si-
napic acids also markedly increased the ORAC value. Tamura
and Yamagami (47) reported that acylation of malvidin-3-
glucoside with p-coumaric acid increased antioxidant activity
Peroxyl-Radical Absorbing Capacity. The ORAC assay has
received widespread acceptance as a measure of antioxidant
capacity, as it is believed to mimic the ability of biological
compounds to trap free radicals (43). Another advantage of the
ORAC assay is that it is conducted close to physiological pH
at 7.4, which is appropriate when evaluating pharmacological
properties. The matter of pH is even more important, because

6180 J. Agric. Food Chem., Vol. 50, No. 21, 2002
Stintzing et al.
as measured by linoleic acid oxidation. The only pigment studied
by Wang et al. (14) in common with those of the present study
was cyanidin 3-glucoside. We obtained a value of 2.26 µmol
Trolox equivalents per µmol anthocyanin, whereas Wang and
co-workers (14), dissolving anthocyanins in dimethyl sulfoxide
and performing the assay at pH 7.0, found 3.49. This might
indicate that different solvent systems (solvent type and pH)
yield varying ORAC values for the same commodity investi-
gated.
(4) Wrolstad, R. E. Anthocyanins. In Natural Food Colorants:
Science and Technology; Lauro, G. J., Francis, F. J., Eds.; Marcel
Dekker Inc.: New York, 2000; pp 237-252.
(
5) Ames, B. M.; Shigena, M. K.; Hagen, T. M. Oxidants, antioxi-
dants and the degenerative diseases of aging. Proc. Natl. Acad.
Sci. U.S.A. 1993, 90, 7915-7922.
6) Steinmetz, K. A.; Potter, J. D. Vegetables, fruit, and cancer
prevention: a review. J. Am. Diet. Assoc. 1996, 96, 1027-1039.
7) Wang, H.; Cao, G.; Prior, R. L. Total antioxidant capacity of
fruits. J. Agric. Food Chem. 1996, 44, 701-705.
(
(
Surprisingly, the pigment extracts had ORAC values 3- to
(8) Vinson, J. A.; Hao, Y.; Su, X.; Zubik, L. Phenol antioxidant
quantity and quality in foods. J. Agric. Food Chem. 1998, 46,
3630-3634.
6-fold higher than those of the individual pigments (Table 7).
One possible explanation is a synergistic effect arising from
anthocyanin mixtures. Furthermore, principles found for the
purified pigments were not reflected in the antioxidant potencies
of the extracts. In part, this can be explained by the fact that
calculation of ORAC values for the extracts was based on
cyanidin 3-glucoside concentration.
(
9) Abuja, P. M.; Murkovic, M.; Pfannhauser, W. Antioxidant and
prooxidant activities of elderberry (Sambucus nigra) extract in
low-density lipoprotein oxidation. J. Agric. Food Chem. 1998,
4
6, 4091-4096.
(
10) Kalt, W.; Forney, C. F.; Martin, A.; Prior, R. L. Antioxidant
capacity, vitamin C, phenolics, and anthocyanins after fresh
storage of small fruits. J. Agric. Food Chem. 1999, 47, 4638-
This study was undertaken to relate color quality and
antioxidant properties to structural functions of cyanidin deriva-
tives. Tonality was influenced by the site of glucosylation as
well as type and degree of hydroxycinnamic acid substitution.
Increasing the number of sugar substituents at the 3 position
can also affect all of these parameters, however, the extent was
not predictable. Although we could confirm an increase in color
strength through acylation, it was also proved that slightly
different substitution of the acyl moiety affected chromophore
protection and color appearance. Diacylation does not neces-
sarily result in improved color strength, as shown in the case
of 5-glycosylation. Besides their coloring function, anthocyanins
are also believed to display an array of pharmacological
properties (3). As was demonstrated in this study, not only
concentration, but also distinct structural elements of antho-
cyanins may influence the antioxidant properties in vitro.
Whereas 5-glycosylation led to a decrease, acylation raised
antioxidant potency. Because anthocyanin mixtures exhibited
stronger capacities than expected by their qualitative and
quantitative pigment compositions, synergistic effects were
suspected. These assumptions need to be verified in vivo.
However, learning more about their chemical properties may
help in understanding principles influencing anthocyanin bio-
availability, tissue retention, and mechanisms of antioxidant
action.
4
644.
(11) Wang, S. Y.; Lin, H.-S. Antioxidant activity in fruits and leaves
of blackberry, raspberry, and strawberry varies with cultivar and
developmental stage. J. Agric. Food Chem. 2000, 48, 140-146.
(12) Ehlenfeldt, M. K.; Prior, R. L. Oxygen radical absorbance
capacity (ORAC) and phenolic and anthocyanin concentrations
in fruit and leaf tissues of highbush blueberry. J. Agric. Food
Chem. 2001, 49, 2222-2227.
13) Moyer, R. A.; Hummer, K. E.; Finn, C. E.; Frei, B.; Wrolstad,
R. E. Anthocyanins, phenolics and antioxidant capacity in diverse
small fruits: Vaccinium, Rubus, and Ribes. J. Agric. Food Chem.
(
2
002, 50, 519-525.
(
14) Wang, H.; Cao, G.; Prior, R. L. Oxygen radical absorbing
capacity of anthocyanins. J. Agric. Food Chem. 1997, 45, 304-
309.
(15) Hofmann, T. Studies on the influence of the solvent on the
contribution of single Maillard reaction products to the total color
of browned pentose/alanine solutions - A quantitative correlation
using the color activity concept. J. Agric. Food Chem. 1998,
4
6, 3912-3917.
(
16) Degenhardt, A.; Hofmann, S.; Knapp, H.; Winterhalter, P.
Preparative isolation of anthocyanins by high-speed counter-
current chromatography and application of the color activity
concept to red wine. J. Agric. Food Chem. 2000, 48, 5812-
5818.
(17) Rodriguez-Saona, L. E.; Wrolstad, R. E. Anthocyanins. Extrac-
tion, isolation and purification of anthocyanins. In Current
Protocols in Food Analytical Chemistry; Wrolstad, R. E., Ed.;
John Wiley & Sons: New York, 2001; Unit F1.1, pp 1-11.
ACKNOWLEDGMENT
(
18) Stintzing, F. C.; Stintzing, A. S.; Carle, R.; Wrolstad, R. E. A
novel zwitterionic anthocyanin from evergreen blackberry (Rubus
laciniatus Willd.). J. Agric. Food Chem. 2002, 50, 396-399.
We thank Robert W. Durst, Department of Food Science and
Technology, Oregon State University (Corvallis, OR) for
technical advice, and Brian L. Arbogast, Environmental Health
Science Center, Department of Chemistry, Oregon State Uni-
versity for performing the MS analyses. We are grateful to
Deborah J. Hobbs, Linus Pauling Institute, Oregon State
University for performing the ORAC analyses. IFF Flavors
North America (Salem, OR), San-Ei Gen F.F.I. Inc. (New York,
NY), and GNT Inc. (Tarrytown, NY) are acknowledged for
providing the pigment material.
(19) Giusti, M. M.; Wrolstad, R. E. Characterization and measurement
of anthocyanins by UV-visible spectroscopy. In Current
Protocols in Food Analytical Chemistry; Wrolstad, R. E., Ed.;
John Wiley & Sons: New York, 2001; Unit F1.3., pp 1-13.
(
20) Cao, G.; Alessio, H. M.; Cutler, R. G. Oxygen-radical absorbance
capacity assay for antioxidants. Free Radical Biol. Med. 1993,
1
4, 303-311.
(
21) Gl a¨ ssgen, W. E.; Seitz, H. U.; Metzger, J. W. High-performance
liquid chromatography/electrospray mass spectrometry and tan-
dem mass spectrometry of anthocyanins from plant tissues and
cell cultures of Daucus carota L. Biol. Mass. Spec. 1992, 21,
LITERATURE CITED
2
71-277.
(
(
(
1) Mazza, G.; Miniati, E. Anthocyanins in Fruits, Vegetables, and
(22) Idaka, E.; Yamakita, H.; Ogawa, T.; Kondo, T.; Yamamoto, M.;
Goto, T. Structure of three diacylated anthocyanins isolated from
red cabbage Brassica oleracea. Chem. Lett. 1987, 6, 1213-1216.
(23) Giusti, M. M.; Rodriguez-Saona, L. E.; Griffin, D.; Wrolstad,
R. E. Electrospray and tandem mass spectroscopy as tools for
anthocyanin characterization. J. Agric. Food Chem. 1999, 47,
4657-4664.
Grains. CRC Press: Boca Raton, FL, 1993.
2) Downham, A.; Collins, P. Colouring our foods in the last and
next millennium. Int. J. Food Sci. Technol. 2000, 35, 5-22.
3) Rice-Evans, C. A.; Miller, J. A.; Paganga, G. Structure-
antioxidant activity relationships of flavonoids and phenolic
acids. Free Radical Biol. Med. 1996, 20, 933-956.

Color and Antioxidant Properties of Cyanidin Pigments
J. Agric. Food Chem., Vol. 50, No. 21, 2002 6181
(
(
(
24) Dyrby, M.; Westergaard, N.; Stapelfeldt, H. Light and heat
sensitivity of red cabbage extract in soft drink model systems.
Food Chem. 2001, 72, 431-437.
(39) Brouillard, R.; Delaporte, B. Chemistry of anthocyanin pigments.
2. Kinetic and thermodynamic studies of proton transfer,
hydration, and tautomeric reactions of malvidin 3-glucoside. J.
Am. Chem. Soc. 1977, 99, 8461-8468.
25) Bassa, I. A.; Francis, F. J. Stability of anthocyanins from sweet
potatoes in a model beverage. J. Food Sci. 1987, 52, 1753-
(
40) Figueiredo, P.; Elhabiri, M.; Toki, K.; Saito, N.; Brouillard, R.
Anthocyanin intramolecular interactions. A new mathematical
approach to account for the remarkable colorant properties of
the pigments extracted from Matthiola incana. J. Am. Chem.
Soc. 1996, 118, 4788-4793.
1
754.
26) Terahara, N.; Shimidzu, T.; Kato, Y.; Nakamura, M.; Maitani,
T.; Yamaguchi, M.-A.; Goda, Y. Six diacylated anthocyanins
from the storage root of purple sweet potato, Ipomoea batatas.
Biosci. Biotechnol. Biochem. 1999, 63, 1420-1424.
(
(
(
41) Dangles, O.; El-Hajji, H. Synthesis of 3-methoxy- and 3-(beta-
D-glucopyranosyloxy)flavylium ions. Influence of the flavylium
substitution pattern on the reactivity of anthocyanins in aqueous
solutions. HelV. Chim. Acta 1994, 77, 1595-1610.
42) Figueiredo, P.; Elhabiri, M.; Toki, K.; Saito, N.; Dangles, O.;
Brouillard, R. New aspects of anthocyanin complexation. In-
tramolecular copigmentation as a means for colour loss? Phy-
tochemistry 1996, 41, 301-308.
(
27) Inami, O.; Tamura, I.; Kikuzaki, H.; Nakatani, N. Stability of
anthocyanins of Sambucus canadensis and Sambucus nigra. J.
Agric. Food Chem. 1996, 44, 3090-3096.
(
28) Dangles, O.; Saito, N.; Brouillard, R. Kinetic and thermodynamic
control of flavylium hydration in the pelargonidin-cinnamic acid
complexation. Origin of the extraordinary flower color diversity
of Pharbitis nil. J. Am. Chem. Soc. 1993, 115, 3125-3132.
29) Jackman, R. L.; Smith, J. L. Anthocyanins and betalains. In
Natural Food Colorants; Hendry, G. F., Houghton, J. D., Eds.;
Blackie Academic & Professional: London, England, 1996; pp
(
43) Prior, R. L.; Cao, G. In vivo antioxidant capacity: comparison
of different analytical methods. Free Radical Res. Med. 1999,
2
7, 1173-1181.
2
44-309.
(
30) Goto, T.; Kondo, T. Structure and molecular stacking of
anthocyanins - Flower color variation. Angew. Chem., Int. Ed.
English 1991, 30, 17-33.
(44) Bors, W.; Heller, W.; Michel, C.; Saran, M. Flavonoids as
antioxidants: Determination of radical-scavenging efficiencies.
Methods Enzymol. 1990, 186, 343-355.
(45) Prior, R. L.; Cao, G.; Martin, A.; Sofic, E.; McEwen, J.; O’Brien,
C.; Lischner, N.; Ehlenfeldt, M.; Kalt, W.; Krewer, G.; Mainland,
C. M. Antioxidant capacity as influenced by total phenolic and
anthocyanin content, maturity, and variety of Vaccinium species.
J. Agric. Food Chem. 1998, 46, 2686-2693.
(
31) Hoshino, T.; Matsumoto, U.; Goto, T.; Harada, N. Evidence for
the self-association of anthocyanins IV: PMR spectroscopic
evidence for the vertical stacking of anthocyanin molecules.
Tetrahedron Lett. 1982, 23, 433-436.
(
32) Brouillard, R.; Dangles, O. Flavonoids and flower colour. In The
FlaVonoids: AdVances in Research since 1986; Harborne, J. B.,
Ed.; Chapman & Hall: London, England, 1994; pp 465-588.
33) Timberlake, C. F. Anthocyanins-occurrence, extraction and
chemistry. Food Chem. 1980, 5, 69-80.
34) Redus, M.; Baker, D. C.; Dougall, D. K. Rate and equilibrium
constants for the dehydration and deprotonation reactions of some
monoacylated and glycosylated cyanidin derivatives. J. Agric.
Food Chem. 1999, 47, 3449-3459.
(
46) Terahara, N.; Callebaut, A.; Ohba, R.; Nagata, T.; Ohnishi-
Kameyama, M.; Suzuki, M. Acylated anthocyanidin 3-sophoro-
side-5-glucosides from Ajuga reptans flowers and the corre-
sponding cell cultures. Phytochemistry 2001, 58, 493-500.
47) Tamura, H.; Yamagami, A. Antioxidative activity of monoacy-
lated anthocyanins isolated from Muscat Bailey A grape. J. Agric.
Food Chem. 1994, 42, 1612-1615.
(
(
(
(
35) Havl ´ı kov a´ , L.; M ´ı kov a´ , K. Transformations of elderberry an-
thocyanins. Z. Lebensm.-Unters. Forsch. 1987, 184, 289-293.
36) Dougall, D. K.; Baker, D. C.; Gakh, E.; Redus, M. Biosynthesis
and stability of monoacylated anthocyanins. Food Technol. 1997,
Received for review April 24, 2002. Revised manuscript received July
25, 2002. Accepted July 25, 2002. The financial support by a grant from
the DAAD-German Academic Exchange Service (DAAD-Doktoran-
denstipendium im Rahmen des gemeinsamen Hochschulsonderpro-
gramms III von Bund und L a1 ndern) and by fruit - International Fruit
Foundation, Heidelberg-Schlierbach, Germany, is gratefully acknowl-
edged. This is technical paper 11863 from the Oregon Agricultural
Experiment Station.
(
1
1, 69-71.
(
37) Sondheimer, E. On the relation between spectral changes and
pH of the anthocyanin pelargonidin 3-monoglucoside. J. Am.
Chem. Soc. 1953, 75, 1507-1508.
(
38) Timberlake, C. F.; Bridle, P. Flavylium salts, anthocyanidins and
anthocyanins. 1. Structural transformations in acid solutions. J.
Sci. Food Agric. 1967, 18, 473-478.
JF0204811