902
A. Fujimoto, T. Masuda / Food Chemistry 132 (2012) 901–906
bioactivities. In this study, the antioxidation product from rosmari-
nic acid in a lipid oxidation system was investigated. To clarify the
antioxidation product from rosmarinic acid, a useful and generally
applicable method to identify the produced unstable quinone deriv-
ative was newly developed using an odourless thiol reagent.
Tesque); solvent, 1% acetic acid in H2O/CH3OH = 9:1, flow rate,
9.5 ml minꢀ1; detection, 280 nm, to give fraction 1 (14 mg) at the
retention time (9 min) and fraction 2 (82 mg) at the retention time
(11 min). Fraction 1 was purified again by preparative HPLC under
the following new conditions, column, Cosmosil 5C18-AR-II
(20 ꢁ 250 mm); solvent, 1% acetic acid in H2O/CH3OH = 85:15, flow
rate, 9.5 ml minꢀ1; detection, 280 nm, to give compound 4 (8 mg).
Fraction 2 (82 mg) was purified by a recycle preparative HPLC
under the following new conditions, column, Cosmosil 5C18-AR-
II (20 ꢁ 250 mm); solvent, 1% acetic acid in H2O/CH3OH = 70:30,
flow rate, 9.5 ml minꢀ1; detection, 280 nm, recycle 5 (5 mg) and
6 (27 mg).
2. Materials and methods
2.1. Chemicals and instruments
Rosmarinic acid (1) was purchased from Aldrich Japan (Tokyo,
Japan). 2,20-Azobis(isobutyronitrile) (AIBN) and 1-dodecanethiol
were obtained from Tokyo Kasei (Tokyo, Japan). Ethyl linoleate
was purchased from Kanto Chemicals (Tokyo, Japan) and used after
purification by silica gel (silica 60, Merck, Darmstat, Germany)
chromatography developed with 2.5% ethyl acetate in hexane. All
solvents and other reagents were obtained from Nacalai Tesque
(Kyoto, Japan). The NMR spectra were measured using an
ECS-400 spectrometer (JEOL, Tokyo, Japan) with the manufacturer-
supplied pulse sequences [1H, 13C, correlated spectroscopy (HH-
COSY) and nuclear Overhauser effect spectroscopy (NOESY)]. The
mass spectra were measured with XEVO Qtof spectrometers
(Waters Japan, Tokyo, Japan) in the negative ESI mode. A LC-
20ATvp low pressure gradient system (Shimadzu, Kyoto, Japan)
equipped with a photodiode array detector (SPD-M20AVP, Shima-
dzu) and a DGU-12A degasser was employed for the analytical
HPLC. A PU-980 pump equipped with a UV-975 detector (JASCO,
Tokyo, Japan) was used for analysis of ethyl linoleate hydroperox-
ides. A LC-6AD recycling system (Shimadzu) equipped with a UV-
970 detector (JASCO) was used for preparative HPLC.
4. HR–ESI-MS (m/z) [MꢀH]ꢀ calcd for C30H39O8S, 559.2366;
found, 559.2387; 1HNMR (400 MHz, CD3OD) d 5.16 (1H, brd,
J = 7.3 Hz, H2), 3.00(1H, brd, J = 13.7 Hz, H3a), 2.99 (1H, dd,
J = 13.7 and 7.3 Hz, H3b), 6.75 (1H, brs, H20), 6.68 (1H, d,
J = 8.2 Hz, H50), 6.65 (1H, brd, J = 8.2 Hz, H6’), 6.33 (1H, d,
J = 16.0 Hz, H200), 8.43 (1H, d, J = 16.0 Hz, H300), 6.83 (1H, d,
J = 8.8 Hz, H5000), 7.23 (1H, d, J = 8.8 Hz, H6000), 2.69 (2H, t,
J = 6.8 Hz, 2000-SCH2), 1.1–1.4 (20H, m), 0.88 (3H, t, J = 6.4 Hz,
CH3); Important correlations observed in the NOESY, H200-H6000,
13
H300-2000-SCH2; CNMR (100 MHz, CD3OD) d 38.2 (C3), 130.7 (C10),
117.5 (C20), 146.0 (C30), 145.2 (C40), 116.3 (C50), 121.9 (C60), 168.5
(C100), 116.7 (C200), 146.1 (C300), 120.0 (C1000), 129.7 (C2000), 148.1
(C3000), 148.6 (C4000), 117.0 (C5000), 123.0 (C6000).
5. HR-ESI-MS (m/z) [MꢀH]ꢀ calcd for C30H39O8S, 559.2366; found,
559.2355; 1HNMR [400 MHz, CD3OD-acetone-d6(2:1, v/v) d 5.16 (1H,
dd, J = 9.6 and 3.4 Hz, H2), 3.41 (1H, dd, J = 14.0 and 3.4 Hz, H3a),
3.19 (1H, dd, J = 14.0 and 9.6 Hz, H3b), 6.64 (1H, s, H20), 6.66 (1H, s,
H50), 6.10 (1H, d, J = 16.0 Hz, H200), 7.39 (1H, d, J = 16.0 Hz, H300), 6.92
(1H, d, J = 1.6 Hz, H2000), 6.67 (1H, d, J = 8.0 Hz, H5000), 6.83 (1H, dd,
J = 8.0 and 1.6 Hz, H2000), 2.68 (2H, t, J = 7.6 Hz, 60-SCH2), 0.78 (3H, t,
J = 7.2 Hz, 60-SC11H22-CH3); 13CNMR (100 MHz, CD3OD) d 37.1 (C3),
122.4 (C10), 121.5 (C20), 147.3 (C30), 145.0 (C40), 115.1 (C50), 133.1
(C60), 168.6 (C100), 115.1 (C200), 147.9 (C300), 127.7 (C1000), 114.8 (C2000),
146.8 (C3000), 149.6 (C4000), 116.4 (C5000), 123.0 (C6000), Signals for C1 and
C2 were not identified for their low intensity.
2.2. HPLC Detection of the antioxidation products from rosmarinic acid
against ethyl linoleate oxidation
Rosmarinic acid (0.4 mg) in CH3CN solution (4 ml) and AIBN
(158 mg) were added to 61 mg of ethyl linoleate in a straight vial
(75 mm height, 40 mm diameter). The control vial was also pre-
pared in a similar manner, but without addition of rosmarinic acid.
Both solutions were well stirred and then incubated at 37 °C for 6 h
with shaking (100 times minꢀ1) by a water bath shaker. Twenty
microlitres of the reaction solution were injected into the analyti-
cal HPLC system under the following conditions, column, Cosmosil
5C18-AR-II (4.6 ꢁ 250 mm, Nacalai Tesque); solvent system, 1%
acetic acid in H2O (solvent A) and CH3CN (solvent B); elution, linear
gradient from 5% of solvent B to 100% of solvent B for 40 min, and
then isocratic mode of 100% solvent B for 10 min; flow rate,
1.0 ml minꢀ1; detection, 280 nm. At intervals for both solutions, a
6. HR-ESI-MS (m/z) [MꢀH]ꢀ calcd for C30H39O8S, 559.2366;
found, 559.2366; 1HNMR (400 MHz, CD3OD) d 5.18 (1H, brd,
J = 9.2 Hz, H2), 3.09 (1H, d, J = 13.6 Hz, H3a), 2.97 (1H, dd, J = 13.6
and 9.2 Hz, H3b), 6.69 (1H, brs, H20), 6.78 (1H, brs, H60), 6.25 (1H,
d, J = 16.0 Hz, H200), 7.54 (1H, d, J = 16.0 Hz, H300), 7.03 (1H, brs,
H2000), 6.76 (1H, d, J = 8.4 Hz, H5000), 6.93 (1H, brd, J = 8.4 Hz, H6000),
2.73 (2H,t, J = 7.6 Hz, 50-SCH2), 1.1–1.5 (20H, m), 0.88 (3H, t,
J = 7.6 Hz, 50-SC11H22-CH3); Important correlations observed in the
NOESY, H60- 50-SCH2; 13CNMR (100 MHz, CD3OD) d 174.5 (C1),
75.3 (C2), 31.8 (C3), 129.7 (C10), 116.7 (C20), 146.0 (C30), 144.8
(C40), 125.4 (C50), 122.4 (C60), 168.5 (C100), 115.2 (C200), 147.5 (C300),
127.7 (C1000), 114.7 (C2000), 146.8 (C3000), 149.7 (C4000), 116.4 (C5000),
123.1 (C6000), 34.8 (50-SCH2), 14.5 (CH3).
20
ll-aliquot was removed from the solution and diluted with
Assignment of 13C signals of 4–6 were based on the reported
data for rosmarinic acid (Kikuzaiki & Nakatani, 1989).
380
ll of methanol. Ten microlitres of the diluted solution was in-
jected into the HPLC to analyse the ethyl linoleate hydroperoxides
using a ODS-A column (4.6 ꢁ 150 mm, YMC, Tokyo, Japan) eluted
with CH3CN/H2O (9:1, v/v) at 1.0 ml minꢀ1; with detection at
234 nm.
2.4. HPLC analysis of the thiol adduct from the antioxidation product of
rosmarinic acid in a lipid oxidation system
2.3. Preparation procedure for thiol adducts (4–6)
One ml of the reaction solution was taken from the antioxidation
vial of rosmarinic acid against ethyl linoleate oxidation to a new vial
1 h after the reaction started. To the new vial was added 1-dode-
To a solution of rosmarinic acid (150 mg) in CH3CN (40 ml) was
added DPPH (2,2-diphenyl-1-picrylhydrazyl, 574 mg) at 23 °C with
stirring. After the reaction mixture was allowed to stand for
10 min, 1-dodecanethiol (253 mg) was added to the mixture. The
mixture was allowed to stand for 1 h at 23 °C and then evaporated
to dryness in vacuo to give a residue (1 g). A part of the residue
(0.4 g) was separated by preparative HPLC under the following
conditions, column, Cosmosil 5C18-AR-II (20 ꢁ 250 mm, Nacalai
canethiol (0.6
(3.5 l) at 23 °C. After standing for 1 h at 23 °C, 20
acid was added to acidify the solution and then 10
l
l) in CH3CN (10
l
l) and subsequently triethylamine
l of phosphoric
l of the solution
l
l
l
was analysed by HPLC under the following conditions: column, Cos-
mosil 5C18-AR-II (4.6 ꢁ 250 mm); solvent system, 1% acetic acid in
H2O (solvent A) and CH3CN (solvent B); elution, linear gradient from
5% of solvent B to 100% of solvent B for 40 min, and then isocratic