1878
S. SAITO and J. KAWABATA
control. A reduction in absorbance of 0.228 by the
positive control was regarded as corresponding to the
consumption of two molecules of DPPH radical. All
experiments were performed in triplicate.
Molecular orbital calculations. The electron density
and energy of LUMO were calculated by the AM1
method using the MOPAC 2000 program in the
Chem3D package (CambridgeSoft, Cambridge, MA).
Isolation of oxidation products of 1. To a solution of
compound 1 (67 mg, 0.40 mmol) in methanol (1.6 liters)
was added DPPH radical (0.95 g, 2.4 mmol, 6.0 equiv.),
and the mixture was stirred for 1 h at room temperature.
A solution of sodium dithionite (0.40 g, 2.4 mmol, 6.0
equiv.) in water (24 ml) was added to the reaction
mixture, and this was stirred for 30 min to reduce
unstable o-quinones to their catechol forms. The result-
ing mixture was concentrated under reduced pressure,
and the residue was dissolved in acetone and filtered.
The filtrate was concentrated under reduced pressure.
The residue was subjected to silica gel column chroma-
tography and eluted first with chloroform to remove
DPPH radical and its reduced hydrazine, and then with
methanol. The resulting methanol eluate was concen-
trated under reduced pressure. The crude product was
further purified by preparative HPLC (column, ODS-
Prep, 20 ꢀ 250 mm, GL-Sciences, Tokyo, Japan; mobile
phase, 40% acetonitrile containing 0.1% formic acid;
flow rate, 5.0 ml/min; detection, UV 254 nm) to afford 4
(0.9 mg, 2.1%, tR 12.4 min)7) and 13 (7.0 mg, 8.8%, tR
15.0 min). 13: a yellow oil. 1H-NMR (methanol-d4):
3.29 (3H, s, 20-OCH3), 3.53 (3H, s, 20-OCH3), 3.73
(3H, s, 70-OCH3), 3.76 (3H, s, 7-OCH3), 5.81 (1H, d,
J ¼ 9:8 Hz, H-50), 7.11 (1H, s, H-2), 7.73 (1H, d, J ¼
9:8 Hz, H-60). 13C-NMR (methanol-d4): 52.2 (7-OCH3),
52.5 (20-OCH3), 52.8 (70-OCH3), 53.9 (20-OCH3), 65.3
(C-10), 93.4 (C-30), 116.9 (C-2), 119.0 (C-20), 119.6
(C-5), 123.4 (C-1), 124.4 (C-50), 137.7 (C-6), 145.8
(C-3), 148.1 (C-4), 150.0 (C-60), 168.1 (C-7), 171.5
(C-70), 196.5 (C-40). HMBC correlation peaks: H-2/C-4,
C-6, H-50/C-10, C-30, H-60/C-6, C-20, C-40, 20-OCH3/
C-20, 7-OCH3/C-7, 70-OCH3/C-70. HR-FAB-MS m=z
ðM ꢁ HÞꢁ: calcd. for C18H17O10, 393.0822; found,
393.0827.
Results and Discussion
Compound 13, together with the C-2 methanol adduct
(4), was isolated as an oxidation product from the
reaction mixture of 1 and DPPH radical in methanol. Its
molecular formula was identified as C18H18O10 from the
HR-FAB-MS result, indicating that 13 was a dimeric
1
product of 1. The H-NMR spectrum showed a pair of
doublet signals at ꢀ 5.81 (H-50) and ꢀ 7.73 (H-60), with a
larger coupling constant (J ¼ 9:8 Hz) than that found in
4 (JH-5/H-6 ¼ 8:6 Hz). H-60 showed JCH HMBC corre-
3
lations with an acetal carbon at ꢀ 119.0 (C-20) and a
carbonyl carbon at ꢀ 196.5 (C-40), indicating that 13 was
a quinone acetal-like product. Additionally, HMBC
correlations between C-20 and two methoxy groups
suggested the formation of dimethyl acetal at C-20. H-50
was correlated with two sp3 carbons of C-10 and C-30.
Moreover, the HMBC spectrum showed correlations of
H-2/C-6 and H-60/C-6. These results suggest that
the two protocatechuoyl residues were linked at C-6
and C-10. The other interunit C-C linkage was between a
quaternary benzene carbon assignable to C-5 (ꢀ 119.6)
and a downfield-shifted quaternary oxycarbon of C-30 (ꢀ
1
93.4). On the basis of H and 13C-NMR, HMQC, and
HMBC spectral data, the structure of 13 is proposed to
be as depicted in Fig. 1.
From a reaction mixture of 4 and DPPH radical in
methanol, compound 14 was isolated by the procedure
described for 13. The negative FAB-MS of 14 exhibited
a ½M ꢁ Hꢂꢁ peak at m=z 423, indicating that the
molecular weight of 14 was 30 mass units larger than
that of 13. The 1H-NMR spectrum of 14 was very
similar to that of 13, except for the disappearance of a
singlet signal of H-2 and the presence of the signal of an
additional methoxy group. Furthermore, the 2D NMR
spectra were consistent with those of 13. Hence it was
concluded that the structure of 14 was the 2-methoxy
derivative of 13.
A plausible formation mechanism for 13 is outlined in
Scheme 1. First, 5 undergoes nucleophilic attack by a
methanol molecule at C-20 to yield dimethyl acetal (10).
Since the resulting dimethyl acetal (10) does not have a
proton at C-20, regeneration of a catechol structure
cannot occur. Instead, ketonization of C-40 generates an
Isolation of compound 14. Compound 14 was isolated
from a reaction mixture of 4 and DPPH radical by the
method described for 13 (15% from 4, tR 17.2 min).
Compound 14: a yellow oil. 1H-NMR (methanol-d4):
3.29 (3H, s, 20-OCH3), 3.53 (3H, s, 20-OCH3), 3.73
(3H, s, 7-OCH3), 3.76 (3H, s, 70-OCH3), 3.80 (3H, s,
2-OCH3), 5.81 (1H, d, J ¼ 9:8 Hz, H-50), 7.73 (1H, d,
J ¼ 9:8 Hz, H-60). 13C-NMR (methanol-d4): 52.1 (7-
OCH3), 52.4 (20-OCH3), 53.0 (70-OCH3), 53.8 (20-
OCH3), 62.1 (2-OCH3), 65.5 (C-10), 93.0 (C-30), 116.3
(C-1), 118.8 (C-5), 135.7 (C-3), 139.4 (C-6), 147.8
(C-4), 148.1 (C-2), 119.0 (C-20), 124.4 (C-50), 150.3
(C-60), 167.2 (C-7), 171.2 (C-70), 196.3 (C-40). HMBC
correlation peaks: H-50/C-10, C-30, H-60/C-6, C-20,
C-40, 2-OCH3/C-2, 20-OCH3/C-20, 7-OCH3/C-7, 70-
OCH3/C-70. HR-FAB-MS m=z ðM ꢁ HÞꢁ: calcd. for
C19H19O11, 423.0927; found, 423.0951.
MeO OMe
OH
OH
2'
4
3'
HO
3
5
O
4'
5'
2
1'
1
6
6'
CO2Me
CO2Me
Fig. 1. Key HMBC Correlations of 13.