Development and validation of oxygen radical absorbance capacity assay for lipophilic antioxidants using randomly methylated β-cyclodextrin as the solubility enhancer

Article abstract of DOI:10.1021/jf0113732

We recently reported the improved oxygen radical absorbance capacity (ORAC) assay using fluorescein (FL) as the fluorescent probe. The current ORAC_FL assay is limited in hydrophilic antioxidant due to the aqueous environment of the assay. Lipop

Full text of DOI:10.1021/jf0113732

J. Agric. Food Chem. 2002, 50, 1815−1821 1815
Development and Validation of Oxygen Radical Absorbance
Capacity Assay for Lipophilic Antioxidants Using Randomly
Methylated â-Cyclodextrin as the Solubility Enhancer
DEJIAN HUANG, BOXIN OU,* MAUREEN HAMPSCH-WOODILL,
JUDITH A. FLANAGAN, AND ELIZABETH K. DEEMER
Brunswick Laboratories, 6 Thacher Lane, Wareham, Massachusetts 02571
We recently reported the improved oxygen radical absorbance capacity (ORAC) assay using
fluorescein (FL) as the fluorescent probe. The current ORACFL assay is limited in hydrophilic antioxidant
due to the aqueous environment of the assay. Lipophilic antioxidants mainly include the vitamin E
family and carotenoids, which play a critical role in biological defense systems. In this paper, we
expanded the current ORACFL assay to lipophilic antioxidants. Randomly methylated â-cyclodextrin
(RMCD) was introduced as the water solubility enhancer for lipophilic antioxidants. Seven percent
RMCD (w/v) in a 50% acetone-H2O mixture was found to sufficiently solubilize vitamin E compounds
and other lipophilic phenolic antioxidants in 75 mM phosphate buffer (pH 7.4). This newly developed
ORAC assay (abbbreviated ORACFL-LIPO) was validated through linearity, precision, accuracy, and
ruggedness. The validation results demonstrate that the ORACFL-LIPO assay is reliable and robust.
For the first time, by using 6-hydroxy-2,5,7,8-tetramethyl-2-carboxylic acid as a standard (1.0), the
ORAC values of R-tocopherol, (+)-γ-tocopherol, (+)-δ-tocopherol, R-tocopherol acetate, tocotrienols,
2,6-di-tert-butyl-4-methylphenol, and γ-oryzanol were determined to be 0.5 ( 0.02, 0.74 ( 0.03, 1.36
( 0.14, 0.00, 0.91 ( 0.04, 0.16 ( 0.01, and 3.00 ( 0.26, respectively. The structural information of
oxidized R-tocopherol obtained by liquid chromatography/mass spectrometry reveals that the
mechanism for the reaction between the vitamin E and the peroxyl radical follows the hydrogen atom
transfer mechanism, which is in agreement with the notion that vitamin E is the chain-breaking
antioxidant.
KEYWORDS: ORAC; cyclodextrin; LC/MS; lipophilic antioxidants; vitamin E; chain breaking; hydrogen
atom transfer
INTRODUCTION
(FL) assay from our laboratory; all of these methods are
conducted in an aqueous system, as such, none of which are
suitable for lipophilic antioxidants. Without knowing the actual
effectiveness of the lipophilic antioxidants, consumers can be
exposed to unsafe concentrations or ineffective dosages. We
overcame this obstacle by introducing randomly methylated
â-cyclodextrin (RMCD) as a molecular host to enhance the
solubility of lipophilic antioxidants in aqueous solution. Specif-
ically, cyclodextrins (CDs) are cyclic (R-1,4)-linked oligosac-
charides of R-D-gluco-pyranose containing a relatively hydro-
phobic (fatlike) central cavity and hydrophilic (waterlike) outer
surface. This property of CD has made it increasingly popular
as a vehicle for enhancing the solubility of fat soluble
compounds in an aqueous environment in pharmaceutical and
food industries (3, 4). In this paper, we will report the
development and validation of a new ORAC assay specific for
lipophilic antioxidant activity. To our knowledge, it is the first
time that vitamin E and other common lipophilic phenolic
antioxidants were measured using the ORAC assay.
Antioxidants can be physically classified by their solubility
into two groups (1) (i) hydrophilic antioxidants, such as vitamin
C and the majority of polyphenolic compounds, and (ii)
lipophilic antioxidants, mainly including vitamin E and carot-
enoids. Similar to hydrophilic antioxidants, lipophilic antioxi-
dants play an important role in a wide spectrum of biochemical
and physiological processes. Of primary interest is their optimal
antioxidant activity in vitro and in vivo. Unlike hydrophilic
antioxidants, which do not accumulate in the body and are
excreted in the urine, lipophilic antioxidants penetrate the
lipoprotein cell membrane more easily and therefore reach a
higher level of bioavailability (2). Although lipophilic antioxi-
dants are highly bioavailable, it is not a trivial task to accurately
measure their antioxidant activity in vitro. There are a number
of methods available for measuring antioxidant activity such
as the oxygen radical absorbance capacity (ORAC) fluorescein
* To whom correspondence should be addressed. Fax: (508)295-6615.
E-mail: bou@brunswicklabs.com.
10.1021/jf0113732 CCC: $22.00 © 2002 American Chemical Society
Published on Web 02/08/2002

1816 J. Agric. Food Chem., Vol. 50, No. 7, 2002
Huang et al.
MATERIALS AND METHOD
Chemicals and Apparatus. RMCD was purchased from Cyclolab
R&D Ltd. (Budapest, Hungary). FL and 6-hydroxy-2,5,7,8-tetramethyl-
2-carboxylic acid (Trolox) were purchased from Aldrich (Milwaukee,
WI). 2,2′-Azobis (2-amidino-propane) dihydrochloride (AAPH) was
obtained from Wako Chemicals USA (Richmond, VA). γ-Oryzanol
was purchased from TCI America (Portland, OR). Nutriene (toco-
trienols) was obtained from Eastman Chemicals Company (Kingsport,
TN). 4-Difluoro-5-(4-phenyl-1,3-butadienyl)-4-bora-3a,4a- diaza-s-in-
dacene-3-undecanoic acid (BODIPY 581/591 C11) was purchased from
Molecular Probes, Inc. (Eugene, OR). All other standards were
commercially available form Sigma or Aldrich. All ORAC analyses
were performed on a COBAS FARA II analyzer (Roche Diagnostic
System Inc., Branchburg, NJ) using an excitation wavelength of 493
nm and an emission filter of 515 nm. The photobleaching experiment
of BODIPY 581/591 C11 was carried out in a Bio-Tek fluorescence
microplate reader FL600 (Bio-Tek, Winooski, VT). The identification
of oxidized Trolox and R-tocopherol induced by AAPH was performed
by a HP 1100 high-performance liquid chromatography (HPLC) system
(Hewlett-Parkard, Palo Alto, CA) coupled with a Finigan LCQ ion trap
mass spectroscopic detector (ThermoFinigan, San Jose, CA).
Figure 1. FL fluorescence decay curves induced by AAPH in the presence
of R-tocopherol at different concentrations.
Sample Preparation. Approximately 0.5 g of sample was dissolved
in 20 mL of acetone. An aliquot of sample solution was appropriately
diluted with 7% RMCD solvent (w/v) made in a 50% acetone-water
mixture (v/v) and was shaken for 1 h at room temperature on an orbital
shaker at 400 rpm. The sample solution was ready for analysis after
further dilution with 7% RMCD solvent.
Automatic ORAC Assay. The automated ORAC assay was carried
out on a COBAS FARA II spectrofluorometric centrifugal analyzer.
The procedure was based on a previous report of Ou and co-workers
(5). With the exception of samples and Trolox standards, which were
made in 7% RMCD solvent, all other reagents were prepared at 75
mM phosphate buffer (pH 7.4). In the final assay mixture (0.4 mL
total volume), FL (6.3 × 10-8 M) was used as a target of free radical
attack and AAPH (1.28 × 10-2 M) was used as a peroxyl radical
generator. Seven percent RMCD was used as the blank, and Trolox
(12.5, 25, 50, and 100 µM) was used as the control standard. The
analyzer was programmed to record the fluorescence of FL every minute
after the addition of AAPH. All measurements were expressed relative
to the initial reading. Final results were calculated using the differences
of areas under the FL decay curves between the blank and a sample.
These results were expressed as micromoles Trolox equivalent (TE).
Competitive Reaction of Trolox and R-Tocopherol with AAPH.
A total of 0.5 mL of 0.5 mM Trolox and 0.5 mL of 0.5 mM
R-tocopherol (prepared in 7% RMCD) were mixed in a 1.5 mL HPLC
sample vial followed by the addition of 20 µL of 640 mM AAPH; the
reaction solution was incubated at 37 °C in an autosampler. At every
10 min interval, 5 µL of reaction solution was injected into a Zorbax
C18 column (Hewlett-Packard, Palo Alto, CA) (2.1 mm × 150 mm, 3
µm) until the reaction was completed. The mobile phase was 70%
methanol with a flow rate of 0.3 mL/min, and the UV detector was set
at 280 nm.
Characterization of Oxidized Products of Trolox Induced by
AAPH. A mixture of 0.2 mL of 2.0 mM Trolox and 0.8 mL of 640
mM AAPH was incubated at 37 °C for 30 min. The reaction products
were separated by a Zorbax C18 column (2.1 mm × 150 mm, 3 µm)
at 37 °C with a mobile phase of 70% methanol at a flow rate of 0.3
mL/min, and the UV detector was set at 280 nm. The oxidized products
were characterized using a Finnigan LCQ ion trap mass spectrometer
equipped with an API chamber and an electrospray ionization (ESI)
source. The ionization was negative ion mode, and Aux gas and Sheath
gas were set to 72 and 14 units, respectively. An ionization reagent of
1.5 mM ammonium hydroxide was added at a rate of 0.05 mL/min
through a Tee device using a secondary HPLC pump before the API
chamber. Trolox was used as a standard for calibrating the system.
Figure 2. Regression of net AUC of R-tocopherol on different concentra-
tions of R-tocopherol. The net AUC ) AUC
− AUCblank; the AUC
sample
was calculated by the equation previously described by Ou et al. (see 5).
deionized water 3 times. The oxidized products of R-tocopherol in the
methylene chloride phase were then analyzed by liquid chromatography/
mass spectroscopy (LC/MS). With the exception of 100% acetonitrile
as the mobile phase, the LC/MS conditions are the same as above.
Photobleaching Experiment of BODIPY 581/591 C11. BODIPY
581/591 C11 was dissolved in a 1:9 butyronitrile and octane mixture
(v/v) to give a 7.9 × 10-7 M solution; 200 µL of the BODIPY solution
was then added into a 96 well polyproplene plate. The photostability
of BODIPY was evaluated by monitoring the change of fluorescence
intensity over a 1 h period. The fluorescence reading was taken every
minute by a Bio-Tek fluorescence microplate reader FL600 (Bio-Tek,
Winooski, VT) equipped with an excitation filter for 545/40 nm and
an emission filter for 620/40 nm.
RESULTS
Method Validation. (a) Linearity. The correlation between
the net area under the curve (AUC) for the antioxidant and its
concentration was evaluated using four pure compounds. Figure
1 illustrates the FL fluorescence decay curves in the presence
of R-tocopherol and AAPH. Figure 2 depicts the linear response
between the concentration and the net AUC for R-tocopherol.
Table 1 summarizes the net AUCs corresponding to the different
concentrations and the linear coefficient (r2) for the pure
compounds.
Characterization of Oxidized Products of R-Tocopherol Induced
by AAPH. A mixture of 0.5 mL of 20 mM R-tocopherol (prepared in
7% RMCD) and 0.5 mL of 640 mM AAPH was incubated at 37 °C
for 30 min and was extracted with 10 mL of methylene chloride. The
methylene chloride phase was separated and washed with 5 mL of
(b) Limit of Quantitation (LOQ) and Limit of Detection
(LOD). The LOQ is the lowest concentration on the calibration
curve, while the LOD is the lowest amount of antioxidant that

ORAC Assay for Lipophilic Antioxidants
J. Agric. Food Chem., Vol. 50, No. 7, 2002 1817
Table 1. Net AUC vs Concentrationa
compd
concn (µM)
net area
r2
γ-oryzanol
25
12.5
6.25
3.125
100
50
25
75
50
25
12.5
6.25
200
100
50
28.94
15.87
8.51
0.9979
4.32
γ-tocopherol
δ-tocopherol
28.45
14.83
7.78
36.11
27.52
15.56
8.34
0.9971
0.9668
4.46
R-tocopherol
40.67
19.89
10.45
6.07
0.9990
Figure 3. Ruggedness of the ORACFL-LIPO assay. An amount of 12.5
µM γ-oryzanol was analyzed using two COBAS FARA II analyzers for
55 days.
25
a Regression equation is expressed as Y (net area) ) kX (concentration) +
intercept.
Table 3. Relative ORAC Values of Pure Compounds (n g 4)
Table 2. Precision and Accuracy of the ORACFL-LIPO Assay
compds
ORAC
1.0
0.50 ± 0.02
0.00
0.74 ± 0.03
1.36 ± 0.14
0.91 ± 0.04
1.02 ± 0.11
0.41 ± 0.01
0.16 ± 0.01
3.00 ± 0.26
vitamin E
QC1
40
QC2
80
QC3
160
Trolox
R-tocopherol
nominal concn (µM)
R-tocopherol acetate
(+)-γ-tocopherol
(+)-δ-tocopherol
nutriene (tocotrienols)
2,2,5,7,8-pentamethyl-6-chromanol
2-tert-butyl-4-methylphenol
BHT
Run 1
intramean (µM)
41.69
2.87
8.79
104.25
4
84.48
2.12
2.52
105.60
4
175.40
5.67
3.23
109.62
4
a
SD
b
% RSD
c
% REC
N
γ-oryzanol
Run 2
intramean (µM)
42.74
2.65
6.20
106.86
4
92.01
5.12
5.57
115.01
4
171.24
10.86
6.34
107.03
4
SD
% RSD
% REC
N
Run 3
intramean
SD
% RSD
% REC
N
39.0425
5.46
13.99
97.60
4
85.8525
6.54
7.62
107.31
4
167.21
2.61
1.56
104.50
4
pooled runs
38.16
3.66
intermean (µM)
SD
87.45
6.89
171.28
6.38
% RSD
% REC
N
9.66
95.39
12
5.24
109.31
12
3.71
107.05
12
a Standard deviation. b Relative standard deviation. c Recovery.
Figure 4. Kinetic curves for Trolox and R-tocopherol in the presence of
AAPH. The reaction mixture contains 0.25 mM Trolox, 0.25 mM
R-tocopherol, and 12.8 mM AAPH in RMCD (7% w/v) solution. The mixture
was incubated at 37 °C in a HPLC autosampler, and 5 µL mixture was
analyzed by HPLC at every 10 min interval until the reaction was
completed.
can be detected. In our experiment, the LOQ and LOD were
determined to be 12.5 and 5.0 µM, respectively.
(c) Precision and Accuracy. Table 2 summarizes the precision
and accuracy of the ORAC assay using R-tocopherol as a
candidate compound. The precision, which is expressed as
relative standard deviation (%RSD) for all quality control
concentrations, was within (15%. The accuracy of the method
varies from 97.60 to 115.01% within individual batches and
from 95.39 to 107.05% between all of the batches.
highest value of 3.0, while vitamin E acetate shows no
antioxidant activity.
Competitive Reaction between Trolox and R-Tocopherol
with AAPH. The competitive reaction between Trolox and
R-tocopherol with AAPH was examined, and the kinetic curves
for Trolox and R-tocopherol were illustrated in Figure 4. It is
evident that the reaction rate of Trolox with AAPH is much
faster than that of R-tocopherol with AAPH.
(d) Ruggedness. To determine the reproducibility of the
method, a ruggedness experiment was performed. Using two
COBAS FARA II analyzers, 12.5 µM γ-oryzanol was analyzed
for 50 days. Results are shown in Figure 3.
ORAC Values for Pure Lipophilic Antioxidants. Table 3
lists the results for common lipophilic phenolic antioxidants,
which were referenced as the TEs. γ-Oryzanol possesses the
Mechanistic Studies. The mechanisms for peroxidation of
Trolox and R-tocopherol can be elucidated based on their

1818 J. Agric. Food Chem., Vol. 50, No. 7, 2002
Huang et al.
Figure 5. (a) HPLC chromatogram for the Trolox oxidation profile induced by AAPH. The mixture contained 0.4 mM Trolox and 512 mM AAPH in 75
mM phosphate buffer (pH 7.4) and was incubated at 37 °C for 30 min. The reaction products were separated by a Zorbax C18 column (2.1 mm × 150
mm, 3 µm) at 37 °C with a mobile phase of 70% methanol at a flow rate of 0.3 mL/min, and the UV detector was set at 280 nm. The oxidized products
were characterized by using a Finnigan LCQ ion trap mass spectrometer equipped with an API chamber and an ESI source. (b) HPLC chromatogram
for the R-tocopherol oxidation profile induced by AAPH. A mixture of 0.5 mL of 20 mM R-tocopherol (prepared in 7% RMCD) and 0.5 mL of 640 mM
AAPH was incubated at 37 °C for 30 min and was extracted with 10 mL of methylene chloride. The methylene chloride phase was separated and
washed with 5 mL of DI water 3 times. The oxidized products of R-tocopherol in the methylene chloride phase were then analyzed by LC/MS. With the
exception of 100% acetonitrile as the mobile phase, the LC/MS conditions are the same as a.
Figure 7. Regression of net AUC of Trolox on different concentrations of
Trolox in the presence/absence of RMCD (7% w/v). Each data point
represents an average of six separate measurements.
Figure 6. Effect of RMCD concentration (w/v) on the net AUC of 50 µM
R-tocopherol.
oxidized products obtained by LC/MS. As shown in Figure
5a,b, both Trolox and R-tocopherol were oxidized into two
major products with 16 and 32 mass units increase. Figure 8
illustrates the proposed structures for oxidized products and the
oxidation mechanism.
a group of naturally occurring cage molecules, which are built
up from R-D-glucose units. Depending on the number of glucose
moieties in the ring (6, 7, or 8), they are named R-, â-, and
γ-cyclodextrin. CDs are doughnut-shaped and can bind a wide
variety of organic “guest” compounds inside their apolar cavities
in aqueous solution. The main driving force for this binding is
hydrophobic interactions. Recently, Szente et al. studied the
solubilization of carotene and fatty acids in aqueous solution,
and their results indicated that the solubilizing power of CD
derivatives are methylated-â-cyclodextrin . hydroxypropylated-
â-cyclodextrin ) branched-â-cyclodextrin (4). The use of 10-
40% methylated-â-cyclodextrin results in the 1000-fold en-
hancements of the aqueous solubility of lipophilic compounds.
In light of Szente’s work, we employed RMCD in the ORAC
DISCUSSIONS
The ORAC assay was originally developed by Cao and co-
workers (6) and was significantly improved by Ou et al. (5).
However, the improved ORAC assay does not address the issue
of lipophilic antioxidants because the assay is performed in
aqueous solution. The application of CD for solubilization of
lipophilic compounds has been extensively studied, particularly
with the carotene/fatty acids-CD interaction (3, 4, 7). CDs are

ORAC Assay for Lipophilic Antioxidants
J. Agric. Food Chem., Vol. 50, No. 7, 2002 1819
effect on the phenol group derived from the long (C16H31)
aliphatic tail of R-tocopherol, an electron-donating group, may
be the cause of the lower ORAC value. A remotely related
analogue is that the higher alkanoic acid has lower acidity than
that of acetic acid, because the alkyl groups manifest a small
but significant electron donation to the carboxyl carbon (8). We
further measured the ORAC value for R-tocopherol acetate, a
popular ingredient in vitamin E supplements. Our result
indicated that R-tocopherol acetate does not possess any
antioxidant capacity under current experimental conditions. This
result provides additional evidence to support our conclusion
that the phenol group is an essential group for radical trapping
antioxidant activity.
Mechanistic Studies on the Vitamin E Antioxidants. We
examined the oxidized products of R-tocopherol and Trolox with
AAPH by LC/MS and identified two oxidized products for each
compound. Figure 5a,b shows the oxidation profiles for Trolox
and R-tocopherol induced by AAPH obtained by HPLC,
respectively. Figure 8 illustrates the proposed reaction mech-
anisms for Trolox and R-tocopherol in the presence of AAPH.
As shown, the reaction was initiated by the formation of
phenoxyl radical I due to a hydrogen atom being abstracted
from the phenol group by the peroxyl radical. Phenoxyl radical
I can further undergo intramolecular arrangement to form
intermediate II, a tertiary carbyl radical. In the presence of O2,
intermediate II is peroxidized to yield intermediate III, a peroxyl
radical that may abstract a hydrogen from the water molecule
to yield product 2 and a highly reactive hydroxyl radical (HO•).
Meanwhile, the intermediate II can couple with the generated
HO• to produce product 1.
Structure-Activity Relationship. Table 3 summarizes the
ORAC values for some common lipophilic antioxidants. The
ORAC values for the three tocopherols are significantly
different. The ORAC values of R-, γ-, and δ-tocopherol are
0.50, 0.74, and 1.36, respectively. The difference in ORAC value
can be attributed to the steric effect of the ortho methyl groups.
The number of methyl groups ortho to the phenol group of
tocopherol decreases from two (R), one (γ), and zero (δ). A
similar trend was observed for 2-tert-butyl-4-methylphenol (0.41
( 0.01) and 2,6-di-tert-butyl-4-methylphenol (BHT) (0.16 (
0.01). Apparently, the steric factor has a significant impact on
the antioxidant activity; the less steric hindrance results in higher
ORAC values. Our findings, however, are in contrast with the
theoretical calculations by Wright and co-workers who calcu-
lated the O-H bond dissociation energies (BDE) for R- (75.78
kcal/mol), γ- (79.57 kcal/mol), and δ-tocopherol (81.43 kcal/
mol) (9). On the basis of the BDE, they suggested that
R-tocopherol has the highest antioxidant activity among the
three. In fact, the ORAC value is a kinetic parameter while BDE
relates to thermodynamics of a reaction. Therefore, BDE and
ORAC values may not necessarily have the same trend. From
our results, steric hindrance plays a significant negative role in
antioxidant activity. γ-Oryzanol was found to possess a much
higher ORAC value than that of any vitamin E antioxidant. In
support of our result, a recent study by Godber and co-worker
found that γ-oryzanol has a much higher antioxidant activity
than tocopherols in protecting a cholesterol oxidation system
accelerated by AAPH (10). The significantly greater ORAC
value of γ-oryzanol can be attributed to the electronic contribu-
tion of an ortho methoxyl group and the larger π-conjugation
system, involving 11 atoms, as compared to vitamin E with only
8 such atoms. A recent report of Mulder and co-workers on
coenzyme Q10H2 reveals that the hydrogen atom abstraction is
surprisingly easy from intramolecularly hydrogen-bonded meth-
Figure 8. Proposed oxidation mechanism for Trolox and R-tocopherol in
the presence of AAPH.
assay for lipophilic antioxidants. As demonstrated in the
following sections, RMCD was determined to be an ideal
solubility enhancer for the studied antioxidants, and for the first
time, the antioxidant values for tocopherols, tocotrienols, and
other common lipophilic antioxidants were obtained using the
ORAC assay.
Effect of RMCD Concentration on the Solubility of
Vitamin E in Aqueous Solution. Figure 6 shows the net AUC
of 50 µM R-tocopherol at different RMCD concentrations (w/
v) in 75 mM phosphate buffer solution (pH 7.4). The net AUC
of R-tocopherol increases with the increase of RMCD concen-
tration and reaches a plateau after 4% RMCD. The plateau
indicates that R-tocopherol is completely soluble in 75 mM
phosphate buffer solution. This conclusion was further con-
firmed by the HPLC studies (data not shown). Taking into
account the concentration variations, 7% RMCD in 50% acetone
solution was chosen for sample preparation.
Effect of RMCD on the ORAC Value of Trolox. Lipophilic
antioxidants, such as the vitamin E family and oryzanol, consist
of a long aliphatic tail (g16 carbons) and a hydrophilic phenol
group on the head. The tail can fit into a CD cavity allowing
the hydrophilic phenol group to remain in aqueous solution.
Therefore, the reactivity of the headgroup is not inhibited by
CD complexation. RMCD itself consists of hydroxyl and
methoxyl functional groups and is doughnut-shaped with an
open moiety; therefore, RMCD does not possess any antioxidant
activity nor does it prevent the complexed antioxidant molecule
from functioning as antioxidant. To confirm this, the ORAC
values of Trolox were examined in the presence (or absence)
of 7% RMCD. Figure 7 shows the linear curves of the net AUC
against the Trolox concentrations in phosphate buffer and 7%
RMCD solution, respectively. It becomes obvious that the two
curves are almost superimposed, suggesting that RMCD is inert
in the ORAC assay.
Difference in ORAC Value between R-Tocopherol and
Trolox. Because of the structural similarity, Trolox is expected
to possess a similar ORAC value to that of R-tocopherol.
However, our results reveal that the ORAC value of R-toco-
pherol is about 50% less than that of Trolox. The difference in
ORAC value prompted us to investigate the reaction kinetics
and mechanisms of Trolox/R-tocopherol with AAPH. Figure
4 shows the competitive reaction kinetic curves between Trolox
and R-tocopherol in the presence of AAPH. As shown, the rate
of Trolox with AAPH tends to be much faster than that of
R-tocopherol with AAPH. This observation is in agreement with
their ORAC values. At this point, we suggest that the inductive

1820 J. Agric. Food Chem., Vol. 50, No. 7, 2002
Huang et al.
Figure 11. Possible photobleaching mechanism for BODIPY 581/591 after
it was exposed to excitation light.
CONCLUSION
Using RMCD as the solubility enhancer, we successfully
developed a validated ORAC assay for lipophilic antioxidants
(ORACFL-LIPO). The ORACFL-LIPO method was determined to
be robust, reliable, and sensitive. We have also demonstrated
that the phenol group is the key functional group for antioxidant
activity, and the steric hindrance around the phenol group
decreases ORAC values of tocopherols and other phenolic
lipophilic antioxidants. The stereoelectronic effect derived from
the long (C16H31) aliphatic tail of R-tocopherol also causes the
lower ORAC value than that of Trolox. The oxidized Trolox
and R-tocopherol have been identified using LC/MS, and the
reaction mechanism was determined to follow a classic hydrogen
atom transfer (HAT) mechanism. It is necessary to point out
that the ORACFL-LIPO assay does not measure the antioxidant
activity of carotenoids and polyunsaturated fatty acids, since
chemically carotenoids and fatty acids are not the chain-breaking
antioxidants. Instead, they may act as the singlet oxygen
scavengers and therefore follow a different reaction mechanism.
Figure 9. Structures of studied lipophilic antioxidants.
ACKNOWLEDGMENT
The authors thank Dr. Ronald L. Prior of the USDA, ARS,
Arkansas Children’s Nutrition Center for his critical comments
and helpful suggestions.
LITERATURE CITED
(1) Halliwell, B. Antioxidants in human health and diseases. Annu.
ReV. Nutr. 1996, 16, 33-50.
Figure 10. Fluorescence decay curve of BODIPY 581/591 C11 under
microplate reader conditions (Bio-Tek FL 600). Solvent ) butyronitrile/
octane (1:9); temperature ) 41 °C; concentration of BODIPY ) 7.9 ×
10-7 M.
(2) Burton, G. W.; Ingold, K. U. Vitamin E: application of the
principles of physical organic chemistry to the exploration of
its structure and function. Acc. Chem. Res. 1986, 19, 194-201.
(3) Pfitzner, I.; Francz, P. I.; Biesalski, H. K. Carotenoid: methyl-
beta-cyclodextrin formulations: an improved method for supple-
mentation of cultured cells. BBA Gen. Subjects 2000, 1474, 163-
168.
(4) Szente, L.; Mikuni, K.; Hashimoto, H.; Szejtli, J. Stabilization
and solubilization of lipophilic natural colorants with cyclodex-
trins. J. Inclusion Phenom. Mol. Recognit. Chem. 1998, 32, 81-
89.
(5) Ou, B.; Hampsh-Woodill, M.; Prior, R. Development and
validation of an improved Oxygen Radical Absorption Capacity
assay using fluorescein as a fluorescent Probe. J. Agric. Food
Chem. 2001, 49, 4619-4626.
(6) Cao, G.; Alessio, H. M.; Cutler, R. G. Oxygen-radical absorbance
capacity assay for antioxidants Free Radical Biol. Med. 1993,
14, 303-311.
(7) Szente, L.; Szejtli, J.; Szeman, J.; Kato, L. Fatty-acid cyclodextrin
complexes - properties and applications J. Inclusion Phenom.
Mol. Recognit. Chem. 1993, 16, 339-354.
(8) Streitwieser, A.; Heathcock, C. H. Carboxylic Acids. In Intro-
duction to Organic Chemistry; Macmillan Publishing Co., Inc.:
New York, 1976; p 429.
(9) Wright, J. S.; Carpenter, D. J.; McKay, D. J.; Ingold, K. U.
Theoretical calculation of substituent effects on the O-H bond
strength of phenolic antioxidants related to vitamin E. J. Am.
Chem. Soc. 1997, 119, 4245-4252.
(10) Xu, Z.; Hua, N.; Godber, J. S. Antioxidant activity of tocopherols,
tocotrienols, and γ-oryzanol components from rice bran against
cholesterol oxidation accelerated by 2,2′-azobis(2-methylpropi-
oxyphenols (11). γ-Oryzanol possesses such a structural feature,
and thus, it exhibits strong antioxidant activity. Figure 9 lists
the molecular structures of lipophilic antioxidants studied in this
paper.
Comparison of the ORACFL-LIPO Assay with Other Re-
lated Methods. So far, only a few methods have been developed
to measure lipophilic antioxidant activity. For example, Pulido
et al. used a modified ferric-reducing/antioxidant power assay
(FRAP) to measure the antioxidant activity of carotenoids (12);
Acosta et al. modified the Trolox equivalent antioxidant capacity
assay (TEAC) to measure the antioxidant activity of lipophilic
vitamins (13). The drawbacks of using FRAP and TEAC for
antioxidant activity measurement have been extensively dis-
cussed in our previous paper (5). Recently, Naguib developed
a fluorometric method for lipophilic antioxidant activity, in
which BODIPY 581/591 or BODIPY 665/676 was utilized as
the fluorescent probe (14). Aldini et al. further modified
Naguib’s method to measure the lipid compartment of plasma
(15). However, as shown in Figure 10, the BODIBY dye was
found to be photobleached after it was exposed to excitation
light. The photobleaching is likely caused by the cis/trans
isomerization of the olefinic double bond in the presence of
light (Figure 11). Therefore, the BODIBY-based method is not
suitable for quantitation of antioxidant capacity.

ORAC Assay for Lipophilic Antioxidants
J. Agric. Food Chem., Vol. 50, No. 7, 2002 1821
onamidine) dihydrochloride. J Agric. Food Chem. 2001, 49,
2077-2081.
(11) de Heer, M. I.; Mulder, P.; Korth, H.-G.; Ingold, K. U.; Lusztyk,
J. Hydrogen atom abstraction kinetics from intramolecularly
hydrogen bonded ubiquinol-10 and other (poly)methoxy phenols
J. Am. Chem. Soc. 2000, 122, 2355-2360.
(12) Pulido, R.; Bravo, L.; Saura-Calixo, F. Antioxidant activity of
dietary polyphenols as determined by a modified ferric reducing/
antioxidant power assay. J. Agric. Food Chem. 2000, 48, 3396-
3402.
(13) Arnao, M. B.; Cano, A.; Acosta, M. The hydrophilic and
lipophilic contribution to total antioxidant activity. Food Chem.
2001, 73, 239-244.
(14) Naguib, Y. M. A. Antioxidant activities of astaxanthin and related
carotenoids. J. Agric. Food Chem. 2000, 48, 1150-1154.
(15) Aldini, G.; Yeum, K.-J.; Russell, R. M.; Krinsky, N. I. A method
to measure the oxidizability of both the aqueous and lipid
compartments of plasma Free Radical Biol. Med. 2001, 31,
1043-1050.
Received for review October 16, 2001. Revised manuscript received
December 31, 2001. Accepted January 4, 2002.
JF0113732