Free radical scavenging and antioxidative activity of caffeic acid amide and ester analogues: Structure-activity relationship

Article abstract of DOI:10.1021/jf010830b

The structure-activity relationships of synthetic caffeic acid amide and ester analogues as potential antioxidants and free radical scavengers have been investigated. The 2,2-diphenyl-1-picrylhydrazyl radical (DPPH·) scavenging activity of the test compounds was N-trans-caffeoyl-L-cysteine methyl ester (5) > N-trans-caffeoyldopamine (4) > N-trans-caffeoyltyramine (3) > N-trans-caffeoyl-β-phenethylamine (2) > Trolox C (8) > caffeic acid phenethyl ester (1) > caffeic acid (6) > ferulic acid (7). This established that the radical scavenging activity of the compounds increased with increasing numbers of hydroxyl groups or catechol moieties and also with the presence of other hydrogen-donating groups (-NH, -SH). The antioxidative activity of the compounds was also investigated in an emulsified linoleic acid oxidation system accelerated by 2,2′-azobis(2-amidinopropane) dihydrochloride. The order was 1 > 2 > 4 > 3 ≥ 5 > 6 > 8 > 7. Therefore, in the emulsion system, the antioxidative activity of the test compounds depends not only on the hydroxyl groups or catechol rings but also on the partition coefficient (log P) or hydrophobicity of the compounds. This supports the concept that hydrophobic antioxidants tend to exhibit better antioxidative activity in an emulsion system.

Full text of DOI:10.1021/jf010830b

468 J. Agric. Food Chem. 2002, 50, 468−472
Free Radical Scavenging and Antioxidative Activity of Caffeic
Acid Amide and Ester Analogues: Structure−Activity
Relationship
‡
SOPHEAK SON AND BETTY A. LEWIS*,†
Department of Food Science and Division of Nutritional Sciences, Savage Hall, Cornell University,
Ithaca, New York 14853
The structure-activity relationships of synthetic caffeic acid amide and ester analogues as potential
antioxidants and free radical scavengers have been investigated. The 2,2-diphenyl-1-picrylhydrazyl
radical (DPPH‚) scavenging activity of the test compounds was N-trans-caffeoyl-L-cysteine methyl
ester (5) > N-trans-caffeoyldopamine (4) > N-trans-caffeoyltyramine (3) > N-trans-caffeoyl-â-
phenethylamine (2) > Trolox C (8) > caffeic acid phenethyl ester (1) > caffeic acid (6) > ferulic acid
(7). This established that the radical scavenging activity of the compounds increased with increasing
numbers of hydroxyl groups or catechol moieties and also with the presence of other hydrogen-
donating groups (-NH, -SH). The antioxidative activity of the compounds was also investigated in
an emulsified linoleic acid oxidation system accelerated by 2,2′-azobis(2-amidinopropane) dihydro-
chloride. The order was 1 > 2 > 4 > 3 g 5 > 6 > 8 > 7. Therefore, in the emulsion system, the
antioxidative activity of the test compounds depends not only on the hydroxyl groups or catechol
rings but also on the partition coefficient (log P) or hydrophobicity of the compounds. This supports
the concept that hydrophobic antioxidants tend to exhibit better antioxidative activity in an emulsion
system.
KEYWORDS: Antioxidants; antioxidative activity; free radical scavenging activity; caffeic acid analogues;
ferulic acid; partition coefficient; structure-activity relationship
INTRODUCTION
on specific enzymes that induce free radical and lipid hydro-
peroxide formation (11, 12). Therefore, their antioxidative
actions could prevent oxidative rancidity in foods and oxidative
damages in vivo, relating to diseases such as cancer, diabetes,
and cardiovascular, Alzheimer’s, and Parkinson’s diseases.
The structural feature responsible for the antioxidative and
free radical scavenging activity of caffeic acid is the ortho-
dihydroxyl functionality in the catechol ring. The presence of
the electron-donating hydroxyl group at the ortho-position also
lowers the O-H bond dissociation enthalpy and increases the
rate of H-atom transfer to peroxyl radicals (13). The unsaturated
2,3-double bond of the side chain also maximizes the stabiliza-
tion of the phenolic radical (14). The antioxidative activity of
caffeic acid analogues, however, depends on several other
factors such as the electron-donating and withdrawing substit-
uents on the catechol ring, the number of hydroxyl groups or
catechol moieties, the involvement of other H-donating groups
(-NH, -SH), the chemical stability, and the hydrophobicity
or partition coefficient (log P) of the compounds.
Caffeic acid and its analogues are widely distributed in the
plant kingdom and are found in coffee beans, olives, propolis,
fruits, and vegetables. They are usually found as various simple
derivatives including amides, esters, sugar esters, and glycosides,
or in rather more complex forms such as rosmarinic acid (dimer),
lithospermic acid (trimer), verbascoside (heterosidic ester and
glycoside of dihydroxyphenethylethanol and caffeic acid), and
the flavonoid-linked derivatives (1).
The physiological functionality of caffeic acid and its
analogues has attracted much attention and has been studied
by many research groups in recent years. The compounds are
known to have antibacterial (2), antiviral (3), antiinflammatory
(4), antiatherosclerotic (5), antioxidative (6, 7), antiproliferative
(8), immunostimulatory (9), and neuroprotective properties (10).
These properties are associated with either their properties as
antioxidants and enzyme inhibitors or their binding activity with
specific receptors. Caffeic acid and its analogues are potential
natural antioxidants with multiple mechanisms involving free
radical scavenging, metal ion chelation, and inhibitory actions
To compare the antiradical and antioxidative activity of
caffeic acid amide and ester analogues, and to better understand
the effects of the physicochemical parameters mentioned above,
we synthesized a series of caffeic acid amide and ester analogues
and conducted a structure-activity relationship study of them
(Figure 1). Furthermore, caffeic acid amide analogues have not
* Corresponding author (telephone 607-255-2621; fax 607-255-1033;
e-mail ba14@cornell.edu).
† Division of Nutritional Sciences.
‡ Department of Food Science.
10.1021/jf010830b CCC: $22.00 © 2002 American Chemical Society
Published on Web 12/21/2001

Caffeic Acid Radical Scavenging and Antioxidative Activity
J. Agric. Food Chem., Vol. 50, No. 3, 2002 469
Figure 2. Radical scavenging action of antioxidants (ArOH).
General Synthetic Procedure for Caffeic Acid Amide Analogues.
The amides were synthesized from caffeic acid and the corresponding
amines (either in the free base or hydrochloride form) using benzo-
triazol-1-yloxy-tris(dimethylamino)phosphonium hexafluorophosphate
(BOP) as coupling reagent (22, 23). Briefly, 10 mmol of caffeic acid
was dissolved in 20 mL of dimethylformamide and 1.4-2.8 mL (10-
20 mmol) of triethylamine. The solution was cooled in an ice-water
bath and 10 mmol of the amine was added followed by 10 mmol of
BOP dissolved in 20 mL of dichloromethane. The reaction mixture
was stirred at 0 °C for 30 min and then stirred at room temperature for
2 h. Dichloromethane was removed under reduced pressure, and the
residual solution was diluted with 150 mL of water. The crude product
was then extracted with ethyl acetate, washed successively with 1 N
HCl, water, 1 M NaHCO3, and water, then dried over MgSO4, and the
solvent was evaporated. The residue was chromatographed on a silica
gel column using a mixture of ethyl acetate/n-hexane (1:1 or 2:1) as
eluents. Yields were between 45 and 76%.
Figure 1. Structures of caffeic acid analogues and reference compounds.
1, Caffeic acid phenethyl ester; 2, N-trans-caffeoyl-â-phenethylamine; 3,
N-trans-caffeoyltyramine; 4, N-trans-caffeoyldopamine; 5, N-trans-caffeoyl-
L-cysteine methyl ester; 6, caffeic acid; 7, ferulic acid; 8, Trolox C.
been studied to any extent. Their efficiency as radical scavengers
was evaluated by their activity toward a stable free radical, 2,2-
diphenyl-1-picrylhydrazyl (DPPH) (15, 16). Their potency as
antioxidants was evaluated using a Tween-emulsified linoleic
acid oxidation system induced by 2,2′-azobis(2-amidinopropane)
dihydrochloride (17, 18) and measured by the ferric thiocyanate
assay (19, 20). Trolox C, and caffeic and ferulic acids served
as reference compounds (Figure 1).
N-trans-Caffeoylphenethylamine. Mp 138-140 °C; yield 76%. IR
ν
max (cm-1): 3490-3300, 3100-3000, 1660, 1625-1600, 1515, 1200.
1H NMR: δ 2.83 (t, J ) 7.8, 2H, CH2-Ar), 3.51 (dd, J ) 5.9, 2H,
-NH-CH2), 6.43 (d, J ) 15.6, 1H, -CHdCHa-), 7.41 (d, J ) 15.6,
1H, -CHbdCH-), caffeic acid ring: 6.81 (d, J ) 8.3, 1H, H-5), 6.9
(dd, J ) 2.0, J ) 7.81, 1H, H-6), 7.06 (d, J ) 2.0, 1H, H-2),
phenethylamine ring: δ 7.2 2-7.26 (set of signals, 5H, H-2′, H-6′,
H-3′, H-5′, H-4′), phenolic hydroxyl groups: 8.3 (br, 2H, 2 × OH),
amide N-H group: 7.3 (s, 1H, N-H).
MATERIALS AND METHODS
N-trans-Caffeoyltyramine. Mp 206-208 °C; yield 45%. IR νmax
1
(cm-1): 3490-3200, 3100-3000, 1655, 1625-1600, 1515, 1210. H
Chemicals. Caffeic acid, ferulic acid, Trolox C, (2-bromoethyl)-
benzene, â-phenethylamine, tyramine, 3-hydroxytyramine hydrochloride
(dopamine hydrochloride), L-cysteine methyl ester hydrochloride,
hexamethylphosphoramide, 2,2-diphenyl-1-picrylhydrazyl (DPPH‚),
2,2′-azobis(2-amidinopropane) dihydrochloride, linoleic acid, polyoxy-
ethylenesorbitan monolaureate (Tween 20), and ferrous chloride tet-
rahydrate were purchased from Aldrich-Sigma Chemical Co. Sodium
dihydrogen phosphate monohydrate, anhydrous sodium hydrogen
phosphate, and ammonium thiocyanate were purchased from Fischer
Scientific, Inc. All other reagents and solvents were of analytical,
spectrometric, or HPLC grade.
NMR: δ 2.72 (t, J ) 7.3, 2H, CH2-Ar), 3.47 (dd, J ) 5.8, 2H, -NH-
CH2-), 6.43 (d, J ) 15.6, 1H, -CHdCHa-), 7.40 (d, J ) 15.6, 1H,
-CHbdCH-), caffeic acid ring: 6.72 (d, J ) 2.9, 1H, H-2), 6.81 (d,
J ) 8.3, 1H, H-5), 6.89 (dd, J ) 2.9, J ) 8.0, 1H, H-6), tyramine ring:
7.02 (d, J ) 8.3, 2H, H-2′, H-6′), 7.04 (d, J ) 8.3, 2H, H-3′, H-5′),
phenolic hydroxyl groups: 8.2 (br, 3H, 3 × OH), amide N-H group:
7.35 (s, 1H, N-H).
N-trans-Caffeoyldopamine. Mp 171-173 °C; yield 57%. IR νmax
1
(cm-1): 3490-3200, 3100-3000, 1650, 1625-1540, 1500, 1200. H
NMR: δ 2.67 (t, J ) 7.3, 2H, CH2-Ar), 3.47 (dd, J ) 6.3, 2H, -NH-
CH2-), 6.43 (d, J ) 15.6, 1H, -CHdCHa-), 7.4 (d, J ) 15.6, 1H,
-CHbdCH-), caffeic acid ring: 6.8 (d, J ) 8.3, 1H, H-5), 6.9 (dd, J
) 2.9, J ) 8.0, 1H, H-6), 7.05 (d, J ) 2.0, 1H, H-2), dopamine ring:
6.54 (dd, J ) 1.95, J ) 7.8, 1H, H-6′), 6.7 (d, J ) 7.8, 1 H, H-5′),
6.72 (s, 1H, H-2′), phenolic hydroxyl groups: 8.1 (br, 4H, 4 × OH),
amide N-H group: 7.32 (s, 1H, N-H).
N-trans-Caffeoyl-L-cysteine methyl ester. Mp 155-157 °C; yield
45%. IR νmax (cm-1): 3500-3200, 3100-3000, 2555, 1725, 1650,
1640-1580, 1500, 1200, 1100. 1H NMR: 1.94 (t, J ) 8.0, 1H, -SH),
2.99 (t, J ) 6.3, 2H, -CH2-SH), 3.7 (s, 3H, -O-CH3), 4.82 (m, 1H,
-CH (attached to -CH2-SH, -COOCH3, and -NH-CO-), 6.6 (d,
J ) 15.6, 1H, -CHdCHa-), 7.44 (d, J ) 15.6, 1H, -CHbdCH-),
caffeic acid ring: 6.83 (d, J ) 8.3, 1H, H-5), 6.94 (dd, J ) 1.9, J )
6.3, 1H, H-6), 7.08 (d, J ) 1.9, 1H, H-2), phenolic hydroxyl groups:
8.2 (br s, 2H, 2 × OH), amide N-H group: 7.55 (d, J ) 7.3, 1H,
N-H).
Apparatus. Synthesized compounds were purified on a silica gel H
(32-63 mesh) (Selecto Scientific) column and identified by TLC, UV,
NMR, and X-ray diffraction analysis. Melting points were determined
on a Fisher-Johns melting apparatus and are uncorrected. Thin-layer
chromatography (TLC) was performed on precoated silica gel F254 plates
(Merck) using a 254-nm UV lamp (model UVG-54) or/and iodine vapor
to visualize the compounds. IR spectra were recorded on a Perkin-
Elmer 683 infrared spectrophotometer using Nujol as mulling agent or
performed neat; only the most significant absorption bands are reported
1
(νmax, cm-1). H NMR data were acquired at room temperature on a
Varian VXR-400S operating at 400 MHz. Acetone-d6 was used as
solvent; chemical shifts are expressed in δ (parts per million) values
relative to tetramethylsilane (TMS) as internal reference; coupling
constants (J) are given in Hertz. A Spectronic Genesys 8 UV/VIS
spectrophotometer was used in the DPPH‚ and ferric thiocyanate assays.
Synthesis of Caffeic Acid Phenethyl Ester. The compound was
synthesized by base-catalyzed alkylation of caffeic acid salt with (2-
bromoethyl)benzene in hexamethylphosphoramide (21). Recrystalliza-
tion of the product from ether/n-hexane gave the final product as a
pale-yellow powder; mp 124.5-126 °C, yield ca. 70%.
Determination of Radical Scavenging Activity. 2,2-Diphenyl-1-
picrylhydrazyl (DPPH‚) was used as a stable radical (Figure 2). DPPH‚
in ethanol (500 µM, 2 mL) was added to 2 mL of the test compounds
at different concentrations in ethanol. The final concentrations of the

470 J. Agric. Food Chem., Vol. 50, No. 3, 2002
Son and Lewis
the three methods are based on the number of atomic contributions,
the number of molecules used in the database, and the types of atoms
in the molecules. The correlation coefficients between the calculated
and the observed values of the molecules used in these calculations
were in the range of 0.926-0.931. The average log P values (( SD)
of our test compounds were as follows: 1 (3.493 ( 0.127), 2 (2.797
( 0.175), 3 (2.447 ( 0.232), 4 (2.093 ( 0.295), 5 (0.787 ( 0.081), 6
(1.303 ( 0.240), 7 (1.580 ( 0.144), and 8 (3.200 ( 0.026).
Statistical Analysis. Correlation and regression analyses of the
radical scavenging activity versus the concentrations of antioxidants
were carried out using the regression program in StatView (SAS
Institute Inc., Cary, NC). IC50 values were obtained from the slope
equations (Y ) a + bX). Multiple comparisons were by one-way
analysis of variance (ANOVA), followed by Fisher’s PLSD test using
the same software package. Statistical significance was accepted if the
null hypothesis was rejected at the P < 0.05 level. The data in Figure
4 were plotted using SigmaPlot 2000 (SPSS Science, Chicago, IL).
RESULTS AND DISCUSSION
Free Radical Scavenging Activity. DPPH‚ assay is the
simplest method to measure the ability of antioxidants to
intercept free radicals. The scavenging effects of caffeic acid
amide and ester analogues are shown in Table 1. Their
scavenging activity of DPPH‚ radicals decreased in the following
order: 5 > 4 > 3 > 2 > 8 > 1 > 6 > 7 with all values
significantly different at P < 0.05. This sequence indicates that
the DPPH‚ radical scavenging activity of the test compounds
is due to their hydrogen-donating ability. Increasing the number
of hydroxyl or catechol groups, as seen from the sequence 4 >
3 > 2 > 1 > 6 > 7, increases radical scavenging activity. The
presence of other H-donating groups (sulfhydryl, amide) in the
molecule also accelerates this activity. Compound 5 with an
-SH group as an additional H-donating group in the molecule
is the most active of the test compounds. The difference in
radical scavenging activity between 2 and 1 (with 2 being more
active) may be due to the larger effect of the amide bond on
the stability and radical scavenging activity of the molecule than
the ester bond.
Antioxidative Activity. The in vitro model using AAPH-
induced lipid peroxidation of Tween-emulsified linoleic acid is
a common method used to measure the antioxidative activity
of synthetic and natural antioxidants. In this model assay, the
oxidation is carried out under conditions relatively similar to
the in vivo system, and the oxidation rate is proportional to the
concentration of AAPH. The alkylperoxyl radicals produced
from AAPH, which later on cause lipid peroxidation, are very
similar to the radicals formed in biological systems. Factors such
as compositional and interfacial phenomena relating to solubility
and hydrophobicity of components, and partitioning properties
of antioxidants between lipid and aqueous phases, can also be
included in this assay. The high initial velocity of the reaction
in this assay also makes it easy to estimate the extent to which
oxidation is delayed in the presence of antioxidants (18).
Figure 3. Inhibitory effects of antioxidants on linoleic acid peroxidation
induced by 2,2′-azobis(2-amidinopropane) dihydrochloride and measured
by the ferric thiocyanate assay. R−N)N−R is the 2,2′-azobis(2-ami-
dinopropane) dihydrochloride; R ) −(CH ) C(NH ) ) NH‚HCl; LH )
3 2
2
linoleic acid; LOOH ) linoleic acid hydroperoxide; ArOH ) antioxidants.
test compounds in the reaction mixtures were 12.5, 25, 37.5, and 50
µM. Each mixture was then shaken vigorously and held for 30 min at
room temperature and in the dark. The decrease in absorbance of DPPH‚
at 517 nm was measured. Ethanol was used as a blank solution. DPPH‚
solution (2 mL) in ethanol (2 mL) served as the control. Special care
was taken to minimize the loss of free radical activity of DPPH‚ stock
solution as recommended by Blois (15). All tests were performed in
triplicate. The radical scavenging activity of the samples (antioxidants)
was expressed in terms of IC50 (concentration in µM required for a
50% decrease in absorbance of DPPH‚ radical) and as % Inhibition of
DPPH‚ absorbance (% Inhibition ) [(A control - A test)/A control)] × 100),
where A control is the absorbance of the control (DPPH‚ solution without
test sample) and A
is the absorbance of the test sample (DPPH‚
test
solution plus antioxidant). A plot of absorbance vs concentration was
made to establish the standard curve and to calculate IC50. Caffeic acid,
ferulic acid, and Trolox C were used as reference compounds.
Determination of Antioxidative Activity. The antioxidative activity
was evaluated by using 2,2′-azobis(2-amidinopropane) dihydrochloride
(AAPH)-induced lipid peroxidation of a Tween-emulsified linoleic acid
system and measured by the ferric thiocyanate assay (Figure 3). Briefly,
0.2 mL of distilled water, 0.5 mL of 0.2 M phosphate buffer, pH 7.0
(prepared from stock solutions of NaH2PO4 and NaHPO4, 0.2 M each),
and 0.5 mL of 0.25% Tween-20 (in buffer solution) were mixed with
0.5 mL of 2.5% (w/v) linoleic acid in ethanol. The mixture was then
stirred for 1 min. The peroxidation was initiated by the addition of
AAPH solution (0.1 M, 50 µL). The ethanolic solution of antioxidant
(100 µM, 0.5 mL) was then added, and the reaction was carried out at
37 °C for 675 min in the dark. The degree of inhibition of oxidation
was measured by the ferric thiocyanate method for each interval of
75, 150, 225, 300, and 375 min. To 0.1 mL of peroxidation reaction
mixture at each interval, 0.1 mL of 30% ammonium thiocyanate and
0.1 mL of 2 × 10-2 M freshly prepared FeCl2 (in 3.5% aqueous HCl)
were added. Precisely 3 min after addition, the absorbance of the red
complex [Fe(SCN)]2+ (24) was measured at 500 nm. The control for
the assay was prepared in the same manner by mixing all of the
chemicals and reagents except the test compound. Caffeic acid, ferulic
acid, and Trolox C served as reference compounds. All tests were
performed in triplicate.
The inhibitory effects on lipid peroxidation or the antioxi-
dative activity of the compounds are shown in Figure 4. We
defined the antioxidative activity of the compounds as the ability
to delay lipid peroxidation, i.e., the time required to reach
maximum absorbance of 4. The stronger the antioxidant the
longer the period of time to reach this maximal absorbance.
The inhibitory activity of the compounds in decreasing order
was 1 > 2 > 4 > 3 g 5 > 6 > 8 > 7 with all values
significantly different at P < 0.05 except 3 and 5. Caffeic acid
amide and ester analogues all exhibited stronger activity than
Trolox C, caffeic, and ferulic acid. Compounds 1 and 2 were
highly active compounds. The results show that the antioxidative
Determination of Partition Coefficient (Log P). The logarithm of
the partition coefficient for n-octanol/water was computed using CS
ChemPropPro software, an add-on program to ChemDraw Ultra
(CambridgeSoft). The program performs the calculations in three
different methods using least-squares analysis. The differences between

Caffeic Acid Radical Scavenging and Antioxidative Activity
J. Agric. Food Chem., Vol. 50, No. 3, 2002 471
Figure 4. Antioxidative activity of the compounds on 2,2′-azobis(2-amidinopropane) dihydrochloride-induced lipid peroxidation of a Tween-emulsified
linoleic acid system, measured by the ferric thiocyanate assay.
Table 1. Scavenging Activity of Antioxidants for DPPH‚ Radicala; Data
Are Shown as IC50 (µM)b and % Inhibition at 50 µM of Antioxidantsc
converted, through subsequent reactions, to a ketodiene inter-
mediate and finally to the stable Trolox C quinone (27, 28). In
addition, Trolox C has only one hydroxyl group.
compound
IC50 (µM)
inhibition, % (± SD)
In conclusion, our results clearly show that the synthetic
amide and ester analogues of caffeic acid exhibit stronger
antioxidative activity in the two assays studied, as compared to
that of the parent compounds caffeic and ferulic acids, and to
the synthetic antioxidant Trolox C. The free radical scavenging
activity of the compounds increased significantly with the
numbers of hydroxyl groups or catechol moieties and also with
the presence of other hydrogen-donating groups (-SH and
-NH-CO) in the molecule. Compounds 5 and 4 are the most
active compounds in the DPPH radical scavenging assay. The
antioxidative activity of the compounds in the AAPH-induced
lipid peroxidation of linoleic acid emulsion system, however,
depends not only on the hydroxyl groups or catechol rings but
also on the partition coefficient (log P) or hydrophobicity of
the compounds. It is clear that hydrophobic antioxidants tend
to exhibit better antioxidative activity in this system. Compounds
1 and 2 are the most active compounds in this emulsion system.
1
2
3
4
5
6
7
8
51.80
43.07
41.62
24.22
23.36
54.31
57.21
46.39
47.81 ± 0.15a
55.63 ± 0.49b
58.89 ± 2.16c
91.33 ± 0.59d
94.02 ± 0.32e
f
44.37 ± 0.09
26.01 ± 0.62g
53.58 ± 0.32h
a The final concentration of DPPH ethanolic solution was 2.5 × 10-4 M. b The
IC50 (µM) values were calculated from the slope equations of the dose−response
curves (r2 ) 0.946 − 0.992). c Values are means ± standard deviation (SD) of
three determinations. Values with different superscripts are significantly different
(P < 0.05).
activity in the emulsion system depends not only on the number
of hydroxyl groups or catechol rings but also on the solubility,
hydrophobicity (or partition coefficient, log P), and stability of
the compounds. These results confirm the paradoxical behavior
of antioxidants in lipid emulsions. On the basis of log P values
and the results shown in Figure 4, it is apparent that within the
series of caffeic acid amide and ester analogues, the highly active
compounds 1 and 2 have high log P values. However, when
comparing two compounds of comparable lipophilicity, the
structural factors predominate in determining the activity of a
specific compound. For example, compound 4, although having
lower log P value but with an additional hydroxyl group in the
structure, exhibited stronger antioxidative activity than 3. The
same explanation holds for 6 compared to 7. The fact that 5
does not have the highest potential activity in the emulsion
system, as it does in the radical scavenging assay, may be due
to (a) its low affinity for partitioning between the lipid and
aqueous phases (lower value of log P), (b) its ability to form
the hydroperoxide-adducts, or (c) the rapid oxidation of the
sulfhydryl (-SH) group to the disulfide (-S-S-) group (25,
26). Trolox C shows stronger activity at the initial stage of
peroxidation, but then it rapidly loses activity. This may be
explained by its high partitioning affinity (high value of log P)
in the emulsion system and by its ability to undergo one-electron
oxidation to its corresponding phenoxyl radical which is then
LITERATURE CITED
(1) Macheix, J. J.; Fleuriet, A.; Billot, J. Hydroxycinnamic acids.
In Fruit Phenolics; CRC Press: Boca Raton, FL, 1990; pp 20-
34.
(2) Setzer, W. N.; Setzer, M. C.; Bates, R. B.; Nakkiew, P.; Jackes,
B. R.; Chen, L.; McFerrin, M. B.; Meehan, E. J. Antibacterial
hydroxycinnamic esters from Piper caninum from Paluma,
Northqueensland, Australia. The crystal and molecular structure
of (+)-bornyl coumarate. Planta Med. 1999, 65, 747-749.
(3) King, P. J.; Ma, G.; Miao, W.; Jia, Q.; McDougall, B. R.;
Reinecke, M. G.; Cornell, C.; Kuan, J.; Kim, T. R.; Robinson,
W. E., Jr. Structure-activity relationships: analogues of the
caffeoylquinic and dicaffeoyltartaric acids as potent inhibitors
of human immunodeficiency virus type 1 integrase and replica-
tion. J. Med. Chem. 1999, 42, 497-509.
(4) Sud’ina, G. F.; Mirzoeva, O. K.; Pushkareva, M. A.; Korshunova,
G. A.; Sumbatyan, N. V.; Varfolomeev, S. D. Caffeic acid
phenethyl ester as a lipoxygenase inhibitor with antioxidant
properties. FEBS Lett. 1993, 329, 21-24.
(5) Nardini, M.; D’Aquino, M.; Tomassi, G.; Gentill, V.; Di Felice,
M.; Scaccini, C. Inhibition of human low-density lipoprotein
oxidation by caffeic acid and other hydroxycinnamic acid
derivatives. Free Radical Biol. Med. 1995, 19, 541-552.

472 J. Agric. Food Chem., Vol. 50, No. 3, 2002
Son and Lewis
(6) Chen, J. H.; Ho, C.-T. Antioxidant activities of caffeic acid and
its related hydroxycinnamic acid compounds. J. Agric. Food
Chem. 1997, 45, 2374-2378.
(7) Ley, J. P.; Bertram, H.-J. Synthesis of polyhydroxylated aromatic
mandelic acid amides and their antioxidative potential. Tetra-
hedron 2001, 57, 1277-1282.
(8) Chen, J.-H.; Shao, Y.; Huang, M.-T.; Chin, C.-K.; Ho, C.-T.
Inhibitory effect of caffeic acid phenethyl ester on human
leukemia HL-60 cells. Cancer Lett. 1996, 108, 211-214.
(9) Lin, L.-C.; Kuo, Y.-C.; Chou, J.-C. Immunomodulatory principles
of Dichrocephala bicolor. J. Nat. Prod. 1999, 62, 405-408.
(10) Kim, S. R.; Kim, Y. C. Neuroprotective phenylpropanoid ester
of rhamnose isolated from roots of Scrophularia buergeriana.
Phytochemistry 2000, 54, 503-509.
(11) Sugiura, M.; Naito, Y.; Yamaura, Y.; Fukaya, C.; Yokoyama,
K. Inhibitory activities and inhibition specificities of caffeic acid
derivatives and related compounds toward 5-lipoxygenase. Chem.
Pharm. Bull. 1989, 37, 1039-1043.
(12) Michaluart, P.; Masferrer, J. L.; Carothers, A. M.; Subbhara-
maiah, K.; Zweifel, B. S.; Koboldt, C.; Mestre, J. R.; Grunberger,
D.; Sacks, P. G.; Tanabe, T.; Dannenberg, A. J. Inhibitory effects
of caffeic acid phenethyl ester on the activity and expression of
cyclooxygenase-2 in human oral epithelial cells and in a rat
model of inflammation. Cancer Res. 1999, 59, 2347-2352.
(13) Lucarini, M.; Pedulli, G. F. Bond dissociation enthalpy of
R-tocopherol and other phenolic antioxidants. J. Org. Chem.
1994, 59, 5063-5070.
(14) Rice-Evans, C. A.; Miller, N. J.; Paganga, G. Structure-
antioxidant activity relationships of flavonoids and phenolic
acids. Free Radical Biol. Med. 1996, 20, 933-956.
(15) Blois, M. S. Antioxidant determinations by the use of a stable
free radical. Nature 1958, 181, 1199-1200.
ous dispersion of linoleic acid. J. Agric. Food Chem. 2000, 48,
1129-1134.
(19) Shanta, N. C.; Decker, E. A. Rapid, sensitive, iron-based
spectrophotometric methods for determination of peroxide values
of food lipids. J. AOAC Int. 1994, 77, 421-424.
(20) Mihaljevic, B.; Katusin-Razem, B.; Razem, D. The reevaluation
of the ferric thiocyanate assay for lipid hydroperoxides with
special considerations of the mechanistic aspects of the response.
Free Radical Biol. Med. 1996, 21, 53-63.
(21) Son, S.; Lobowsky, E. B.; Lewis, B. A. Caffeic acid phenethyl
ester (CAPE): synthesis and X-ray crystallographic analysis.
Chem. Pharm. Bull. 2001, 49, 236-238.
(22) Dudash, J., Jr.; Jiang, J.; Mayer, S. C.; Joullie, M. M. Compara-
tive study of selected coupling reagents in dipeptide synthesis.
Synth. Commun. 1993, 23, 349-356.
(23) Czarnocki, Z.; Matuszewska, M. P.; Matuszewska, I. Highly
efficient synthesis of fatty acid dopamides. Org. Prep. Proced.
Int. 1998, 30, 699-719.
(24) Ramette, R. W. Formation of monothiocyanatoiron (III):
a
photometric equilibrium study. J. Chem. Edu. 1963, 40, 71-
72.
(25) Hill, J. W.; Coy, R. B.; Lewandowski, P. E. Oxidation of cysteine
to cystine using hydrogen peroxide. J. Chem. Edu. 1990, 67,
172.
(26) Shanlin, F. U.; Gebicki, S.; Jessup, W.; Gebicki, J. M.; Dean,
R. T. Biological fate of amino acid, peptide and protein
hydroperoxides. Biochem. J. 1995, 311, 821-827.
(27) Thomas, M. J.; Bielski, B. H. J. Oxidation and reaction of Trolox
C, a tocopherol analogue, in aqueous solution. A pulse-radiolysis
study. J. Am. Chem. Soc. 1989, 111, 3315-3319.
(28) Delicado, E. N.; Ferrer, A. S.; Carmona, F. G. A kinetic study
of the one-electron oxidation of Trolox C by the hydroperoxidase
activity of lipoxygenase. Biochim. Biophys. Acta 1997, 1335,
127-134.
(16) Brand-Williams, W.; Cuvelier, M. E.; Berset, C. Use of a free
radical method to evaluate antioxidant activity. Food Sci.
Technol. 1995, 28, 25-30.
(17) Niki, E. Free radical initiators as source of water- or lipid-soluble
peroxyl radical. Methods Enzymol. 1990, 186, 100-108.
(18) Liegeois, C.; Lermusieau, G.; Collin, S. Measuring antioxidant
efficiency of wort, malt, and hops against the 2,2′-azobis(2-
amidinopropane) dihydrochloride-induced oxidation of an aque-
Received for review June 26, 2001. Revised manuscript received
November 14, 2001. Accepted November 15, 2001.
JF010830B