596 J. Agric. Food Chem., Vol. 51, No. 3, 2003
Bratt et al.
product was recrystallized from acetone/water to give the title compound
as yellow crystals (0.172 g, 40%).
Table 1. Concentrations of the Individual Avenanthramides Used in
the Combination Study (Micromolar)a
The NMR data for N-[4-hydroxy-(E)-cinnamoyl]anthranilic acid (1p),
N-[4-hydroxy-3-methoxy-(E)-cinnamoyl]anthranilic acid (1f), N-[4′-
hydroxy-(E)-cinnamoyl]-5-hydroxyanthranilic acid (2p), N-[3′,4′-di-
hydroxy-(E)-cinnamoyl]-5-hydroxyanthranilic acid (2c), and N-[4′-
hydroxy-3′-methoxy-(E)-cinnamoyl]-5-hydroxyanthranilic acid (2f) were
in good agreement with literature data (7).
c (−)b c (0) c (+) f (−) f (0) f (+) s (−) s (0) s (+)
1
2
40
30
65
65
90
100
80
50
160
130
240
210
40
40
70
70
100
100
a1 corresponds to anthranilic acid and 2 to 5-hydroxyanthranilic acid; p, c, f,
N-[3,4-Dihydroxy-(E)-cinnamoyl]anthranilic Acid (1c). 1H NMR
(DMSO-d6) δ 11.24 (s, COOH, 1H), 9.53 (s, OH, 1H), 9.13 (s, OH,
1H), 8.58 (dd, J ) 8.4, 1.1 Hz, 1H), 7.99 (ddd, J ) 7.9, 1.7, 0.4 Hz,
1H), 7.59 (ddd, J ) 7.4, 7.4, 1.7 Hz, 1H), 7.44 (d, J ) 15.5 Hz, 1H),
7.15 (ddd, J ) 7.9, 7.3, 1.2 Hz, 1H), 7.08 (d, J ) 2.1 Hz, 1H), 7.00
(ddd, J ) 8.1, 2.2, 0.4.Hz, 1H), 6.77 (d, J ) 8.2 Hz, 1H), 6.50 (d, J )
15.5 Hz, 1H); 13C NMR (DMSO) δ 169.5, 164.1, 148.0, 145.6, 141.9,
134.0, 131.1, 125.9, 122.6, 121.2, 120.3, 118.4, 116.6, 115.8, 114.6;
IR νmax cm-1 3369, 1636, 1221; UV (MeOH) λmax 339; mp 221-230
°C. Anal. Calcd for C16H13NO5‚0.5H2O: C, 62.34; H, 4.57; N, 4.54.
Found: C, 61.84; H, 4.47; N, 4.23.
N-[4-Hydroxy-3,5-dimethoxy-(E)-cinnamoyl]anthranilic Acid (1s). 1H
NMR (DMSO-d6) δ 11.29 (s, COOH, 1H), 8.90 (s, OH, 1H), 8.63 (ddd,
J ) 8.5, 1.2, 0.4 Hz, 1H), 8.0 (ddd, J ) 8.0, 1.7, 0.4 Hz, 1H), 7.60
(ddd, J ) 7.7, 7.2, 1.8 Hz, 1H), 7.53 (d, J ) 15.4 Hz, 1H), 7.15 (ddd,
J ) 7.9, 7.2, 1.2 Hz, 1H), 7.04 (s, 2H), 6.76 (d, J ) 15.4 Hz, 1H),
3.82 (s, 6H); 13C NMR (DMSO) δ 169.4, 164.2, 148.0, 142.4, 141.1,
137.9, 134.0, 131.1, 124.7, 122.6, 120.3, 119.2, 116.5, 106.1, 56.1; IR
νmax cm-1 3401, 1609, 1268; UV (MeOH) λmax 232, 344; mp 199-
200 °C. Anal. Calcd for C18H17NO6‚H2O: C, 59.83; H, 5.30; N, 3.88.
Found: C, 59.54; H, 5.20; N, 3.45.
N-[4′-Hydroxy-3′,5′-dimethoxy-(E)-cinnamoyl]-5-hydroxyanthra-
nilic Acid (2s). 1H NMR (DMSO-d6) δ 10.84 (s, COOH, 1H), 9.59 (s,
OH, 1H), 8.86 (s, OH, 1H), 8.38 (d, J ) 9.0 Hz, 1H), 7.47 (d, J )
15.4 Hz, 1H), 7.37 (d, J ) 3.0 Hz, 1H), 7.02 (dd, J ) 9.0, 3.0 Hz,
1H), 7.00 (s, 2H), 6.72 (d, J ) 15.4 Hz, 1H), 3.81 (s, 6H); 13C NMR
(DMSO) δ 169.1, 163.7, 152.4, 148.0, 141.5, 137.7, 133.0, 124.9, 122.4,
120.9, 119.5, 118.3, 116.5, 106.0, 56.1; IR νmax cm-1 3371, 1608, 1221;
UV (MeOH) λmax 226, 346; mp 206 °C. Anal. Calcd for C18H17NO7‚
0.5 H2O: C, 58.69; H, 4.93; N, 3.80. Found: C, 58.44; H, 4.87; N,
3.76.
Antioxidant Activity. DPPH. A modified method of Brand-Williams
et al. (15) was used. One hundred microliters of the avenanthramide
or cinnamic acid solution (0.50 mM in methanol) and 900 µL of DPPH
solution in methanol (0.076 mM) were mixed in a tube (molar ratio )
1:1.4) at the same time as the scanning session was initialized,
immediately vortexed, transferred to a cuvette, and placed in the
spectrophotometer. The absorbance (A) at 517 nm was recorded during
20 min. The small spontaneous decrease of absorbance of the DPPH
solution was investigated by time (t). The linear equation obtained (A
) -0.0087t + 0.7704; R2 ) 0.9713) was used to calculate a start
absorbance for each sample. This start value was used to calculate the
decrease of absorbance (∆A) at 2 min for all samples. DL-R-Tocopherol
was run as a reference. DL-R-Tocopherol and caffeic and sinapic acids
were also mixed with DPPH in an additional molar ratio (1:5.6). All
samples were run in duplicates. The lowest absorbance obtained, using
pyrogallol as an antioxidant, was 0.01.
and s correspond to p-coumaric, caffeic, ferulic, and sinapic acid, respectively.
b (−) low concentration; (0) mean concentration; (+) high concentration.
solution was added the inhibitor in ethanol (100 µL, 3.2 mM; 40 µM
final concentration) by syringe. After 15 min of stirring/equilibration,
a thermostated solution of 2,2′-azobis(2,4-dimethylvaleronitrile) (AMVN)
in chlorobenzene (0.5 mL, 22.4 mM) was added. Five microliter
samples were withdrawn every 10 min (after interruption of the stirring
for 30 s) and injected onto a 150 × 3.9 mm i.d., 5 µm Resolve silica
90 Å column (Waters) eluted with hexane/ethanol (90:10) with a flow
rate of 1.0 mL/min. After sampling, stirring was immediately resumed.
The reaction was followed for at least 70 min. The formation of
conjugated diene hydroperoxides (retention times of 2.8-3.2 min) was
monitored at 234 nm, and the concentrations were determined by
integration using an experimentally determined response factor [a
weighed amount of linoleic acid containing a trace of linoleic acid
hydroperoxide was allowed to react in CDCl3 with bis[4-(dimethy-
lamino)phenyl]telluride and the conversion to the corresponding
1
telluroxide was determined by integration in the H NMR spectrum].
The data obtained during the first 30 min were adjusted to a linear
graph, and the slope was used as a measurement of antioxidant activity.
Oat Avenanthramide Extraction and Analysis. Groats and hulls
(separated by hand) (5 g) were extracted and analyzed in triplicate
according to the method described by Bryngelsson et al. (17). Methanol
extracts were subjected to HPLC analysis on a 125 × 4 mm i.d., 5 µm
reversed phase C-18 column (HP ODS Hypersil) using a mobile phase
consisting of two solvents: A, 0.01 M phosphate buffer (pH 2.8) and
acetonitrile (95:5, v/v); B, acetonitrile. Samples were run with a linear
gradient over 60 min from 0 to 40% B in A. The components were
detected at 340 nm. To identify the avenanthramides in the HPLC
chromatograms, retention times and UV spectra of the peaks were
compared with those of synthetic standards.
Statistical Analysis. Results from analyses of individual avenan-
thramides and cinnamic acids were statistically evaluated by Tukey’s
pairwise comparison (R ) 0.05). Results from the combination study
were statistically evaluated by fractional factorial fit. All analyses were
conducted using the software Minitab release 11.12 (Minitab Inc., State
College, PA).
RESULTS
Synthesis. A modified version of the method of Mayama et
al. (13) was used for the preparation of avenanthramides (Figure
2). 2-Methylbenzoxazin-4-ones derived from anthranilic (1) or
5-hydroxyanthranilic acid (2) (O-acetylated) were obtained by
treatment with acetic anhydride. Subsequent acid-catalyzed aldol
reactions with substituted benzaldehydes gave the protected
benzoxazinones, and final treatment with phosphate-promoted
hydrolysis resulted in the avenanthramides. All compounds were
crystalline with colors varying from pale yellow to brown. A
novel feature of the modified procedure is the mild phosphate-
induced deprotection and ring-opening step. This treatment gave
no side reactions, and the crude product was obtained in ∼90%
yield (as judged from NMR data). However, optimal conditions
for recrystallization were difficult to find, and only 50% of the
pure product was obtained. The syntheses of 1c, 1s, and 2s are
reported for the first time.
Combination Effects. A combination study was performed in the
DPPH system described above. The two avenanthramides with the
weakest activity (1p and 2p) were excluded, and combination effects
of the remaining six avenanthramides were investigated according to
a fractional factorial design with 32 experiments and 6 central points.
The experiments were divided into two blocks that were analyzed on
two separate days. The design allowed all main effects and two factor
interactions to be independently estimated. The high (+) and low (-)
concentrations of the six avenanthramides were chosen to obtain a
similar antioxidant activity contribution after 20 min from each
compound at each level (Table 1). Narrow intervals of the concentra-
tions were chosen to get linear relations between concentration and
effect.
Structure)Antioxidant Activity Relationships. There are
several different methods described in the literature for the
evaluation of antioxidant activity (18 and references cited
therein). In the present study, two commonly used methods,
Linoleic Acid. A slightly modified method of Braughler et al. (16)
was used. Linoleic acid in chlorobenzene (7.5 mL, 36.2 mM) was stirred
(1100 rpm) at 42 °C in a 20 mL thermostated reaction vessel. To this