Preparation of two sets of 5,6,7-trioxygenated dihydroflavonol derivatives as free radical scavengers and neuronal cell protectors to oxidative damage

Article abstract of DOI:10.1016/j.bmc.2009.03.032

An unusual class of 5,6,7-trioxygenated dihydroflavonols (3a-e and 4a-j) were designed and prepared. Their antioxidative properties were assessed by examining their capacities in several in vitro models, including superoxide anion and 1,1-diphenyl-2-picrylhydrazyl (DPPH) radical scavenging, rat liver homogenate lipid peroxidation inhibition, PC12 cells protection from oxidative damage, and xanthine oxidase inhibition. These dihydroflavonols displayed positive quenching abilities towards O₂^{{radical dot} -} and DPPH free radicals, in which the majority exhibited superior antioxidant properties to Vitamin C. cis-Configurated compound (±)-3e demonstrated remarkable inhibition to LPO with an IC₅₀ value of 1.9 ± 0.3 μM, which was apparently stronger than that of quercetin (IC₅₀ = 6.0 ± 0.4 μM). trans-Configurated dihydroflavonol (±)-4h exhibited significant protective effect on PC12 cells against oxidative damage with an EC₅₀ value of 41.5 ± 5.3 μM, more effective compared to that of quercetin (EC₅₀ = 81.8 ± 8.7 μM). The 6-OH-5,7-dimethoxy analogue (±)-3d showed significant inhibition of xanthine oxidase with an IC₅₀ value of 16.0 ± 0.8 μM, which is superior to that of allopurinol (IC₅₀ = 23.5 ± 2.0 μM). In addition to the hypothesized action mechanism of the bio-active compounds, 3D modeling was used to analyze the relationship between the minimized-energy structures and antioxidant activities.

Full text of DOI:10.1016/j.bmc.2009.03.032

Bioorganic & Medicinal Chemistry 17 (2009) 3414–3425
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Preparation of two sets of 5,6,7-trioxygenated dihydroflavonol derivatives
as free radical scavengers and neuronal cell protectors to oxidative damage
Jingxu Gong a,c, Kexin Huang a, Feng Wang b, Leixiang Yang b, Yubing Feng b, Haibo Li b, Xiaokun Li a,
,
*
a
Su Zeng b, Xiumei Wu a, Joachim Stöckigt b,d, Yu Zhao a,b, , Jia Qu
*
a Key Laboratory of Southern Zhejiang TCM R&D, Pharmacy School of Wenzhou Medical College, Chashan District, Wenzhou 325035, China
b College of Pharmaceutical Sciences, Zhejiang University, Yu Hang Tang Road 388, Hangzhou 310058, China
c Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Zu Chong Zhi Road 555, Pudong High-Tec Park, Shanghai 201203, China
d Lehrstuhl für Pharmazeutische Biologie, Institut für Pharmazie, Johannes-Gutenberg Universität Mainz, Staudinger Weg 5, D-55099 Mainz, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
An unusual class of 5,6,7-trioxygenated dihydroflavonols (3a–e and 4a–j) were designed and prepared.
Their antioxidative properties were assessed by examining their capacities in several in vitro models,
including superoxide anion and 1,1-diphenyl-2-picrylhydrazyl (DPPH) radical scavenging, rat liver
homogenate lipid peroxidation inhibition, PC12 cells protection from oxidative damage, and xanthine
oxidase inhibition. These dihydroflavonols displayed positive quenching abilities towards OÅ2ꢀ and DPPH
free radicals, in which the majority exhibited superior antioxidant properties to Vitamin C. cis-Configu-
Received 4 February 2009
Revised 15 March 2009
Accepted 18 March 2009
Available online 24 March 2009
This paper is dedicated to Professor Zhong
Jian JIA at Lanzhou University on the
occasion of her 80s birthday celebrations
rated compound ( )-3e demonstrated remarkable inhibition to LPO with an IC50 value of 1.9 0.3
which was apparently stronger than that of quercetin (IC50 = 6.0 0.4 M). trans-Configurated dihydrofl-
avonol ( )-4h exhibited significant protective effect on PC12 cells against oxidative damage with an EC50
value of 41.5 5.3 M, more effective compared to that of quercetin (EC50 = 81.8 8.7 M). The 6-OH-
5,7-dimethoxy analogue ( )-3d showed significant inhibition of xanthine oxidase with an IC50 value of
16.0 0.8 M, which is superior to that of allopurinol (IC50 = 23.5 2.0 M). In addition to the hypothe-
lM,
l
l
l
Keywords:
5,6,7-Trioxygenated dihydroflavonols
cis-Dihydroflavonoids
Antioxidant
Lipid peroxidation
Free radical scavenger
PC12 cells
l
l
sized action mechanism of the bio-active compounds, 3D modeling was used to analyze the relationship
between the minimized-energy structures and antioxidant activities.
Ó 2009 Elsevier Ltd. All rights reserved.
Xanthine oxidase inhibitor
1. Introduction
transfers electrons from hypoxanthine to uric acid, is coupled to
the reduction of oxygen into superoxide radicals (OÅ2ꢀ) or hydrogen
The reduction of molecular oxygen leads to the formation of
reactive oxygen species (ROS), including superoxide anion radi-
cals and hydroxyl radicals. ROS has physiological functions such
as bacterial ingestion. However, when there is an imbalance be-
tween cellular production of free radicals and the ability of cells
to withstand them, excessive ROS will cause oxidative damage to
cellular components. Oxidative damage may contribute to a vari-
ety of neurological diseases including stroke,1,2 Parkinson’s3 and
Alzheimer’s disease (AD).4,5 It also results in other pathologies,
including cardiac disease, atherosclerosis, hypertension, autoim-
mune rheumatic disease, cancer, viral disease (such as AIDS)
and aging.6–10
peroxide.11 Consequently, XO is considered to be associated with
the pathogenesis of oxidant-induced microvascular changes,
inflammation, aging, arteriosclerosis and designated gout.12,13 In
order to prevent and treat various diseases related to free radicals,
considerable efforts have thus been made in the discovery and
development of efficient synthetic or natural antioxidants acting
as neuronal protector and XO inhibitors.
Dihydroflavonols such as taxifolin (1), which belongs to the
flavonoids family, are an important class of secondary plant
metabolites and exhibit a wide spectrum of pharmacological
properties, including antioxidative, anti-fungal, hepato- and gas-
tro-protective, and antineoplastic activities.14–17 The potential
health benefits attributable to the antioxidant activity of these
compounds have received much attention recently. It is
thought that the antioxidant properties of these polyphenolic
compounds are the results of their high propensity to transfer
electrons, to chelate ferrous ions and to scavenge reactive oxygen
species.18–20 Most of the reported dihydroflavonols possess a
Xanthine oxidase (XO) is an enzyme involved in purine metab-
olism. There is evidence that the XO-catalysed reaction, which
* Corresponding authors. Tel./fax: +86 577 86699227 (X.L.); tel./fax: +86 571
88208449 (Y.Z.).
E-mail addresses: xiaokunli@163.net (X. Li), dryuzhao@126.com (Y. Zhao).
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.03.032

J. Gong et al. / Bioorg. Med. Chem. 17 (2009) 3414–3425
3415
5,7-dihydroxy-2,3-dihydroflavonol skeleton. The predominant
2. Results and discussion
2.1. Synthesis
2,3-dihydroflavonols were reported as 2,3-trans-configurated
ones. To our knowledge, little antioxidant activity data has been
reported for 2,3-cis-configurated dihydroflavonols.
It is noteworthy that another flavonoid, baicalein (2), has been
the recent subject of considerable attention. This 5,6,7-trioxygen-
ated flavone is capable of attenuating oxidative stress in various
in vitro models, suggesting that it may possess significant antiox-
idant activity.21 However, due to its adjacent trihydroxyl groups
located on the A ring, baicalein (2) is somewhat unstable. A few
derivatives were therefore produced by converting the hydroxyl
groups to the alkoxy groups on the A-ring, resulting in increased
activity on P-gp 170 inhibition.22 Fernández et al. analyzed inhi-
bition of aldose reductase by 33 baicalein derivatives with mod-
ifications on the A and B rings, leading to the conclusion that
compounds with either three hydroxyl or methoxyl functional-
ities on the A ring are effective inhibitors of aldose reductase.23
Furthermore, Morita et al. showed that the increased lipophilicity
of phenolic hydroxyls via alkylation may enhance the therapeutic
potential to X-ALD.24 Increasing the lipophilicity of drug mole-
cules is especially advantageous to the treatment of Alzheimer’s
Disease (AD), as this may enhance their ability to pass through
the blood–brain barrier.25 Thus multi-alkylation of baicalein
may not only improve the stability of the highly-oxidized parent
molecule, but also increase its chances of permeating cerebral
tissue. It is therefore reasonable to suggest that methylated ana-
logues of baicalein may constitute the pro-drug form of antioxi-
dant lead compounds, which may more easily target diseased
sites of the central nervous system whilst conferring protection
against oxidative damage.
Based on the aforementioned information, the combination of
the biologically effective 5,6,7-trioxygenated A ring moiety of
baicalein (2) with an dihydroflavonol skeleton of taxifolin (1)
may produce a new class of antioxidants, while introduction of
methoxy groups to replace the sensitive phenolic hydroxyls in
the A-ring of molecules confers much more stability to the
5,6,7-trioxygenated dihydroflavonols designed. However, exam-
ples of dihydroflavonols containing a tri-oxygenized pattern at
C-5, C-6 and C-7 positions in ring A are rarely found in nature
compared with dihydroflavonols di-oxygenated at C-5 and C-7.
Systematic comparisons of SAR principles on di-oxygenated and
trioxygenated dihydroflavonols are not available due to a lack
of information relating to the trioxygenated ones. To overcome
this shortfall, the paper presented here describes the design,
preparation and antioxidative evaluation of assembled 5,6,7-tri-
oxygenated dihydroflavonols (Fig. 1).
The two designed series of compounds include 5,7-dimethoxy-
6-hydroxy-dihydroflavonols 3 and 5,6,7-trimethoxy-dihydroflavo-
nols 4. The synthetic routes employed for the syntheses of 3 and
4 are illustrated in Scheme 1 and Scheme 2, respectively.
Commercially available 2,4,6-trihydroxy acetophenone 5 served
as starting material, and was selectively protected with Me2SO4 to
provide an excellent yield of dimethyl ether 6. Elbs oxidation of 2-
hydroxy-4,6-dimethoxyacetophenone 6 followed by methoxyme-
thylation of the resulting compound 7 in diisopropylethylamine
((i-Pr)2EtN) gave rise to 8, which was then treated with methoxy-
methyl chloride (MOMCl) in a CH2Cl2-aqueous NaOH solution in
the presence of tetrabutylammonium bromide (n-Bu4NBr), to af-
ford the key intermediate 9, as previously reported.26 Condensa-
tion of 9 with different benzaldehydes using ethanolic potassium
hydroxide resulted in the chalcone 10, which was further oxidized
with alkaline hydrogen peroxide to the corresponding chalcone
epoxides 11. Although Rf values of 10 and 11 were similar, the
reaction could be easily monitored by the variation of UV spectra
of the two compounds, in which the kmax changed from ca.
330 nm of the former (10) to ca. 280 nm in the case of the latter
(11). Treatment of the epoxide 11 with hydrochloric acid in meth-
anol furnished 5,7-dimethoxy-6-hydroxy-dihydroflavonols, ( )-3.
The 5,6,7-trimethoxy-dihydroflavonols 4 were synthesized as
shown in Scheme 2. 1,2,3-Trimethoxy-5-nitro-benzene 13 was
prepared by nitration of easily accessible 3,4,5-trimethoxybenzoic
acid 12 with HNO3/AcOH in 69% yield. Reduction of the nitro group
was accomplished in 85% yield using palladium charcoal in hydra-
zine-ethanol solution. The resulting 14 was transformed into its
diazonium salt, which was further hydrolyzed to yield 3,4,5-tri-
methoxy-phenol 15.27 The Friedel–Crafts reaction of 15 with acetic
anhydride (Ac2O) resulted in product 16 in a satisfactory yield,
which was then protected with MOMCl to produce the key inter-
mediate 17. In a similar way to Scheme 1, condensation of 17 with
different benzaldehydes followed by epoxidation smoothly led to
19, and the epoxidation of 18 to produce 19 could be determined
by the change of their UV spectra: the kmax altered from ca.
328 nm of 18 to ca. 285 nm of 19. Compound 19 were then suscep-
tible to cyclization with HCl/MeOH to procure target compounds
( )-4. This synthetic protocol afforded an neat approach and satis-
factory yield of 5,6,7-trimethoxylated dihydroflavonols.
Whereas epoxides 11 and 19 contain two centers of prochirality
at C(a) and C(b), and the method for cyclization is non-enantiose-
Two sets of a total of 15 new compounds as 5,6,7-trioxygenated
dihydroflavonols possessing diverse groups introduced into ring B
were synthesized. The antioxidant properties of these dihydrofl-
avonols were assessed by way of various experimental pharmaco-
logical models in vitro. The preliminary structure–activity
relationship was also analyzed and discussed.
lective, thus the purified compounds ( )-3 and ( )-4 might be
either (+)-2R,3R/(+)-2S,3R or (ꢀ)-2S,3S/(ꢀ)-2R,3S enantiomers.
The absolute stereochemistries of the chromatographic pure com-
pounds 3 and 4 were not determined yet; therefore the symbol of
( ) was adopted for naming these dihydroflavonols. In the 1H NMR
spectra of ( )-3a, ( )-3b, ( )-4a, ( )-4e–f, ( )-4h, a doublet signal of
R
R
R
R
OH
B
B
RO
RO
O
HO
O
HO
HO
O
O
OH
A
C
C
A
OH
OH
OR
O
OH
OH
O
Designed hybrids
Baicalein (2)
Taxifolin (1)
Figure 1. Structures of antioxidative flavonoids and designed hybridized molecules.

3416
J. Gong et al. / Bioorg. Med. Chem. 17 (2009) 3414–3425
OH
O
OMe
O
OMe
O
HO
a
b
c
HO
OH
MeO
OH
MeO
OH
6
5
7
OMe
O
OMe
2'
O
OMe
O
β
2
1
2
MOMO
MeO
MOMO
R
3
MOMO
d
e
α
4'
OH
MeO
OMOM
R
MeO
OMOM
5
5'
8
9
10
5'
2
1
R
4'
OMe
2'
O
2
β
8
1
2
MOMO
MeO
R
R
α
MeO
HO
O
O
3
2
f
g
R
2'
O
3
4'
6
OMOM
OH
5
5
OMe
5'
11
3
R2
R1
R2
R1
B
MeO
HO
O
MeO
HO
O
A
C
OH
OH
OMe O
OMe O
( )-3a. R1=H, R2=OH
( )-3b. R1=H, R2=OMe
( )-3c. R1, R2=OCH2O
( )-3d. R1=R2=OMe
( )-3e. R1=OMe , R2=OH
Scheme 1. Synthesis of 5,7-dimethoxy-6-hydroxy-dihydroflavonols 3a–e. Reagents and conditions: (a) Me2SO4, K2CO3, acetone, rt; 78%; (b) K2S2O8, KOH, H2O, rt; 44%; (c)
MOMCl, (i-Pr)2EtN, CHCl3, reflux; 75%; (d) MOMCl, NaOH, Bu4NBr, CH2Cl2, H2O, rt; 80%; (e) Benzaldehydes, KOH, EtOH, rt; (f) 30% H2O2, NaOH, MeOH, rt; (g) HCl, MeOH, THF,
55 °C.
H-2 appears at ca. 4.95 ppm with a coupling constant J = 12.0 Hz,
which indicated the relative configuration of these compounds is
trans. Compounds ( )-3c–e, ( )-4b–d, ( )-4g, ( )-4i–j were as-
signed a cis orientation according to their diagnostic coupling con-
stants of H-2 (ca. J2,3 = 2.0 Hz).28,29 Additionally, the coupling
constants of H-3 of the synthetic compounds are also in agreement
with above deduction in terms of their orientations. The 2,3-cis
dihydroflavonols are rarely found as natural products, however,
the formation of these cis-oriented natural flavonoids was substan-
tiated by both biosynthetic background and phytochemical evi-
dence.28–30 Marais et al. described that the acid-catalysed
cyclization of the chalcone epoxide will lead to the formation of
either (2R,3R)-2,3-trans or (2S,3R)-2,3-cis dihydroflavonols, while
the latter is thermodynamically less stable, and the epimeriza-
tion/racemization may occur to the cis epimer upon elevated tem-
perature and consequently yield another (2S,3S)-2,3-trans
epimer.31 However, in the presence of special chemical reagent,
the cis dihydroflavonols could be asymmetrically synthesized.31
There are several 2,3-cis dihydroflavonols obtained in this reported
investigation, which might be attributable either to the mild reac-
tion condition including the rather short reaction time, or to the
multi-oxygenated A ring which may stabilize the cation transition
state by a stronger
afforded the potency of stabilizing the cation, which make both
the -attack (lower attack) and the b-attack (upper attack) be-
r-effect. The adjacent aromatic B ring also
a
comes available. These factors, including the acid environment as
well as the polar protic solvent utilized in the reaction, made the
transition state much favor to SN1 characteristics leading to yield
both 2,3-trans and 2,3-cis products.31 Figure 2 hypothetically dem-
onstrated the formation procedure of ( )-2,3-trans and ( )-2,3-cis
compounds. It could be inferred that four kinds of conformers
could be formed due to the non-enantioselective epoxidation and
the following acidic cyclization procedures. Meanwhile, 2,3-cis
analogues were obtained only by SN1 pathway, whilst 2,3-trans
conformers could be produced by two pathways: one is directly
obtained by SN2 pathway, the other is from either a-attack of the
cation intermediate 2R or b-attack of the cation intermediate 2S
(Fig. 2). However, increasing reaction temperature or prolonging
reaction time could noticeably elevate the forming proportion of
2,3-trans to 2,3-cis conformers, which might be attributable to

J. Gong et al. / Bioorg. Med. Chem. 17 (2009) 3414–3425
3417
OMe
OMe
OMe
MeO
MeO
MeO
MeO
MeO
b
a
MeO
NO
2
COOH
NH
2
12
13
14
OMe
O
OMe
MeO
MeO
MeO
MeO
g
c, d
e, f
OH
OH
16
15
OMe
2'
O
OMe
O
β
2
5
1
MeO
MeO
R
MeO
3
h
i
α
4'
OMOM
2
R
MeO
OMOM
4
5'
R
3
R
18
17
3
R
4
2
5'
R
R
4'
OMe
2'
O
2
β
8
1
MeO
R
MeO
MeO
O
α
2
3
3
1
j
R
2'
O
4'
6
OMOM
2
MeO
R
OH
4
5
OMe
5
R
5'
R
3
O
19
4
3
3
R
R
4
4
2
1
R
2
1
R
R
R
B
MeO
O
MeO
O
O
R
R
A
C
MeO
OH
MeO
OH
OMe O
OMe
( )-4a. R1=R3=R4=H, R2=OH
( )-4b. R1=R3=R4=H, R2=OMe
( )-4c. R1,R2=OCH2O, R3=R4=H
( )-4d. R1=R2=OMe, R3=R4=H
( )-4g. R1=R2=R3=OMe, R4=H
( )-4i. R1=R2=R3=H, R4=OH
( )-4j. R1=R3=OMe, R2=OH, R4=H
( )-4e. R1=OMe, R2=OH, R3=R4=H
( )-4f. R1=OH, R2=OMe, R3=R4=H
( )-4h. R1=R2=OH, R3=R4=H
Scheme 2. Synthesis of 5,6,7-trimethoxy-dihydroflavonols 4a–j. Reagents and conditions: (a) HNO3, CH3COOH, rt; 69%; (b) 10% Pd/C, N2H4ꢁH2O, EtOH, reflux; 85%; (c) 40%
HBF4, NaNO2, THF, H2O, 0 °C; 80%; (d) H2SO4, Na2SO4, H2O, 70 °C; 70%; (e) Ac2O, BF3ꢁEt2O, CH2Cl2, reflux; 84%; (f) N(CH2CH2OH)3, H2O, rt; 86%; (g) MOMCl, NaOH, Bu4NBr,
CH2Cl2, H2O, rt; 85%; (h) Benzaldehydes, KOH, EtOH, rt; (i) 30% H2O2, NaOH, MeOH, rt; (j) HCl, MeOH, THF, 55 °C.
the reinforced epimerization or racemization of 2,3-trans conform-
ers to 2,3-cis analogues by the increased temperature or reaction
time.
test compounds were assessed by examining their capacity to
scavenge superoxide anion and 1,1-diphenyl-2-picrylhydrazyl
(DPPH) radicals, as well as their ability to inhibit rat liver
homogenate lipid peroxidation (LPO). Furthermore, the neuropro-
tective effects of the compounds prepared were performed on rat
pheochromocytoma (PC12) cells subjected to hydrogen
peroxide-induced damage. The inhibitory effects of the
synthesized dihydroflavonols on xanthine oxidase were also
measured.
2.2. Biological evaluation
The compounds synthesized were subjected to pharmacologi-
cal screening via various in vitro models relating to free radicals,
enzymes and cultured cell lines. The antioxidant properties of

3418
J. Gong et al. / Bioorg. Med. Chem. 17 (2009) 3414–3425
a
OMe
O
1
active O:
H
RO
3
b
2
H
R1
R2
MeO
OMOM
R=MOM or Me
R4
R3
H2O2, base
b: lower attack
a: upper attack
H+
OMe
O
H
OMe
O
H
O
RO
H
RO
H
R1
R2
R1
O
MeO
MeO
OMOM
OMOM
R4
H+
R4
2R,3S
R2
2S,3R
R3
SN1
R3
H+
H+
MOM
SN2
SN2
SN1
R3
R3
R4
R2
R1
MOM
R4
R2
R1
a
MeO
O
:
a
MeO
RO
O
:
H
OH
+
H
b
RO
b
H
+
H
OH
OMe
O
2S
2R
OMe
O
a: upper attack
a: upper attack
b: lower attack
b: lower attack
B
1
B
MeO
O
MeO
R'O
O
ring
2
3
epimerization or
racemization
4
R'O
OH
OH
OMe O
OMe
O
(+)-2R,3R-trans
(-)-2R,3S-cis
B
B
ring
MeO
O
ring
MeO
O
O
epimerization or
racemization
R'O
OH
R'O
OH
R'=H or Me
OMe O
OMe
(+)-2S,3R-cis
(-)-2S,3S-trans
Figure 2. Possible forming and transforming mechanisms of the 5,6,7-trioxygenated-2,3-dihydroflavonols, SN2 pathway produced merely 2,3-trans conformers, therefore
2,3-cis analogues were obtained only by SN1 pathway (Shaded circles represent the B rings of ( )-3 and ( )-4 with different substituents).
2.2.1. Superoxide anion and DPPH radical scavenging
those of the trans-orientated analogues. Furthermore, the mono-
oxygenated analogues at C-40 in the B ring such as ( )-3a, ( )-3b,
( )-4a and ( )-4b did not exhibit a detectable scavenging effect
of OÅ2ꢀ (Table 1). However, the other mono-oxygenated analogue
substituted at C-20 in the B ring, ( )-4i, showed a weak but distinct
Due to the presence of an unpaired electron, the superoxide an-
ion radical (OÅ2ꢀ) is a highly active type of reactive oxygen species
which can rapidly capture electrons from other molecules and lead
to the generation of a hydroxide radical, the most harmful free rad-
ical to human beings known to date. All of the 15 compounds syn-
thesized were subjected to OÅ2ꢀ scavenging experiments and the
results are shown in Table 1. The most active OÅ2ꢀ scavenger among
the synthesized series 3 was found to be ( )-3d, with an EC50 value
capacity to quench O2Åꢀ, with an EC50 value of 164.5 12.6
lM. In
addition, dihydroflavonols ( )-4b, ( )-4d and ( )-4g, which share
the same structural features of B ring being substituted purely by
one or more methoxy groups, did not show any OÅ2ꢀ scavenging
activity. It was observed that the majority of the designed hybrid
dihydroflavonols, except for ( )-3a, ( )-3b, ( )-4a, ( )-4b, ( )-4d
and ( )-4g, exhibited stronger scavenging ability towards OÅ2ꢀ rad-
of 50.0 6.4
tin (EC50 = 40.6 4.4
the strongest OÅ2ꢀ scavenger, with an EC50 value of 55.5 4.9
l
M, comparable to that of the positive control querce-
M). Among series 4, ( )-4j was found to be
M.
l
l
These results imply that the 40-OH substituent in the B ring has a
positive impact on the OÅ2ꢀ scavenging capacity of the cis-set of
compounds 4. In general, the cis-oriented 5,6,7-trimethoxy-
dihydroflavonols exhibited stronger OÅ2ꢀ quenching abilities than
icals than Vitamin C whose EC50 value is 255.6 28.2 lM (Table 1).
DPPH is a stable free radical existing in vitro, and bleaching of
DPPH absorption is representative of the capacity of test com-
pound to scavenge free aryl radicals with stable properties,

J. Gong et al. / Bioorg. Med. Chem. 17 (2009) 3414–3425
3419
Table 1
showed potent inhibitory effect (IC50 = 5.5 2.1
to that of the positive control quercetin (IC50 = 6.0 0.4 lM). As
l
M) comparable
Free radicals scavenging activities and inhibitory ability on rat liver homogenates
lipid peroxidation of compounds 3 and 4a
shown in Table 1, the multi-methoxylated derivatives ( )-4d and
( )-4g exhibited observable activities toward LPO inhibition alone,
while dihydroflavonols ( )-3d and ( )-4f displayed clear radical
scavenging activity but no inhibitory effects on LPO (Table 1).
The diverse results suggest that 5,6,7-trioxygenated dihydroflavo-
nols likely display different mechanisms of action towards O2Åꢀ
and DPPH free radicals scavenging compared to LPO inhibition.
Furthermore, it was observed that ( )-3e and ( )-4e inhibited
Compound
EC50
(
l
M)
IC50 of LPO (lM)
Superoxide anion
DPPH
b
( )-3a
( )-3b
( )-3c
( )-3d
( )-3e
( )-4c
( )-4d
( )-4e
( )-4f
( )-4g
( )-4h
( )-4i
( )-4j
Vitamin C
Quercetin
—
—
35.3 2.6
94.9 5.8
18.4 0.8
35.9 2.5
42.8 2.6
—
—
11.9 0.8
6.4 0.7
—
1.9 0.3
—
28.6 2.1
5.5 0.4
—
19.6 1.7
10.4 0.7
26.7 2.3
10.4 0.9
n.d.c
93.6 8.7
50.0 6.4
97.2 6.9
80.3 6.5
—
97.1 7.6
112.2 8.9
—
100.3 10.5
164.5 12.6
55.5 4.9
255.6 28.2
40.6 4.4
the lipid peroxidation in
a concentration-dependent manner
—
(Fig. 3), suggesting that ( )-3e and ( )-4e possess strong affinities
for lipid peroxide radical (LOO.). ( )-3e was found to be superior
to the positive control, quercetin, in all of the five concentrations
tested (Fig. 3). However, ( )-4e failed to exhibit better inhibition
27.8 2.5
123.8 9.6
—
7.3 1.2
82.4 7.0
54.1 4.9
154.5 27.9
5.5 1.2
of LPO at lower concentrations (<5.0
to quercetin against LPO at higher concentrations (6–100
l
M) while proving superior
M),
l
6.0 0.4
whilst remaining inferior to ( )-3e across the whole concentration
range. Nevertheless, though ( )-3e possesses two radical scaveng-
ing phenolic sites, ( )-4e should have an improved lipid-binding
affinity due to its higher hydrophobicity, conferred by the presence
of a higher number of methyl groups.
a
Data are expressed as the mean SD, n = 3.
’—’ indicates that EC50 or IC50 value is unable to be generated due to low
b
quenching or inhibition rates at test concentrations. ( )-4a and ( )-4b are omitted
herein due to no concrete EC50 and IC50 values could be generated.
c
n.d. = not detected.
The possible explanation for the best performance of ( )-3e ob-
served in the anti-LPO experiment was addressed (Supplementary
data 1). The free radicals may be trapped in both the delocalized A
ring (conjugated by the adjacent 4-oxo functionality which may
further stabilize the radical) and the delocalized B ring (Supple-
mentary data 1), based on the attacking probabilities of the free
radicals targeting the 5-OH (A-ring) or at 30-OH (B-ring). This re-
sulted in the significant inhibitory capacity of ( )-3e against LPO
mediated by the Fe2+–Vitamin C system.
A recent study showed that 5-methoxy flavonoids are easily
metabolized into their 5-demethylated derivatives, the latter pos-
sessing superior anti-inflammatory properties.32 This suggests that
the methyloxy functionalities at C-5 on the A ring of dihydroflavo-
nols reported here are also prone to demethylation and formation
of a 5-OH metabolite, which together with the 4-oxo group forms
an improved bi-ligand and should be beneficial to chelating ferrous
ions. In the Fe2+-mediated lipid peroxidation bioassay, the signifi-
cant activity of ( )-3e on LPO inhibition may be attributable to
the hypothesized metabolism of the 5-OMe becoming a 5-OH un-
der the experimental conditions of the LPO test (Supplementary
data 2). The exposed 5-hydroxyl metabolite subsequently chelates
Fe2+ with the help of the adjacent 4-oxo group, which forms a sta-
ble six-membered ring, thus preventing free radicals from binding
independent of any enzyme activity. The DPPH test results are
shown in Table 1. Though the 40-OH substituted compound ( )-
4a did not exhibit scavenging activity against DPPH, the other 40-
OH substituted analogues ( )-3a, ( )-3e, ( )-4e, ( )-4h and ( )-4j
displayed preferential quenching activity towards DPPH, compared
to OÅ2ꢀ. The cis-orientated derivatives of 4 possessing mono- or mul-
ti-methoxylated substituents alone at B ring such as ( )-4b, ( )-4d
and ( )-4g again failed to show free radical scavenging activity
with the DPPH model. This is in agreement with their deficiency
towards OÅ2ꢀ (Table 1). It is therefore suggested that the single vari-
ety of substituent of mono- or multi-methoxylated B ring is unfa-
vorable to compound’s radical scavenging. It is noteworthy that
compound ( )-4h, the only one possessing 30,40-ortho-diphenolic
hydroxyls at the B ring, predictably enhanced the quenching activ-
ity of DPPH radicals with an EC50 value of 7.3 1.2
tained the efficacy of the reference standard quercetin
(EC50 = 5.5 1.2 M). In a parallel comparison between the 6-OH
lM, almost at-
l
series compounds 3 and its 6-OMe analogue 4, compounds 3 were
predominantly found to scavenge free radicals, especially DPPH
radicals. This suggests that 5,7-dimethoxy-6-hydroxyl dihydrofl-
avonols may possess higher quenching efficacy towards free radi-
cals than their 5,6,7-trimethoxy substituted dihydroflavonol
analogues. Furthermore, the designed dihydroflanonols demon-
strated superior quenching ability towards DPPH radicals to that
100
( )3e
of Vitamin C (EC50 = 154.5 27.9 lM), with exception of ( )-4a,
( )-4b, ( )-4c, ( )-4d and ( )-4g (Table 1).
( )4e
80
quercetin
2.2.2. Inhibition of lipid peroxidation
60
40
20
0
Hydroxyl and hydroperoxyl radicals are able to attack polyun-
saturated fatty acids of phospholipids and other lipidic constitu-
ents of cell membrane, consequently resulting in lipid
peroxidation (LPO). LPO-induced cell membrane damage alters
its fluidity and ability to function correctly. Therefore, lipid perox-
idation inhibition is used as an index of membrane antioxidant
capacity. The results of LPO inhibition bioassay for compounds 3
and 4 are shown in Table 1. Though not every synthesized
dihydrofalvonol showed significant inhibitory effect on LPO, the
overall efficacy of active compounds was found to be satisfactory
(Table 1). Among them, the 30-methoxy-40-hydroxy compound
( )-3e demonstrated the most impressive inhibition against LPO
0.1
1
10
100
concentration (μM)
Figure 3. Concentration-dependent inhibitory activities of ( )-3e, ( )-4e and
quercetin against Fe2+-mediated peroxidation of rat liver homogenates: Ç = ( )-
3e, j = ( )-4e and 4 = quercetin. Values are means SD (n = 3).
(IC50 = 1.9 0.3 lM). Interestingly, its 6-OMe analogue ( )-4e also

3420
J. Gong et al. / Bioorg. Med. Chem. 17 (2009) 3414–3425
to the lipid membrane and polyunsaturated fatty acids of the
phospholipid bilayer (Supplementary data 2). This assumption
leads to the rationale that compound ( )-3e may also exhibit sig-
nificant activity in vivo. Due to the known association of ferrous
ions with the pathology of Alzheimer’s disease (AD),33 the synthe-
sized dihydroflavonols, especially those with clear LPO inhibiting
activity, may be developed as a new potential chemical entity for
the treatment of AD.
2.2.4. Xanthine oxidase inhibition
Xanthine oxidase (XO) is a cytosolic enzyme that is an impor-
tant source of oxygen free radicals. To explore the antioxidative
activities of the various dihydroflavonols designed, selected com-
pounds were subjected to XO inhibition assay. Compared to the
experimental results listed in Table 1 and Table 2, the test com-
pounds were found to behave differently in OÅ2ꢀ scavenging and
XO enzyme inhibiting. Although XO inhibition could be closely
associated with OÅ2ꢀ radicals, the XO inhibiting efficacies of the syn-
thesized dihydroflavonols were independent of their OÅ2ꢀ scaveng-
ing action, according to the lack of correlation observed between
these two assays (Table 1 and Table 2). It is worthy to point out
that the B-ring-ortho-phenolic hydroxyl dihydroflavonol ( )-4h,
possessing satisfactory antioxidative activity towards free radical
scavenging, LPO inhibition and PC12 cell protection, displayed XO
2.2.3. Protection on PC12 cells
Though a conclusive pathogeny of AD has not yet been eluci-
dated, recent studies suggest that oxidative stress is involved in
apoptotic mechanisms in which excessive production of reactive
oxygen species can lead to neuronal apoptosis, as seen for neuro-
degenerative disorders such as AD. Therefore, the therapeutic ef-
forts aiming at removing or preventing free radical formation are
beneficial to AD patients.34 Tissue culture of rat pheochromocy-
toma (PC12) cells is commonly used as a screening model for test-
ing the prevention of ROS-induced neuronal death.35 The
protective effect on neuronal cells against ROS hazard can be
conveniently evaluated by the cell viability via MTT assay. The
screening results in Table 2 show that 11 out of 15 synthesized
compounds showed clear protective effect towards PC12 cells at
the highest test concentration. Meanwhile, only dihydroflavonols
( )-3b and ( )-4h exhibited sufficient efficacy to enable the calcu-
lation of EC50 values. Compared to the reference substance querce-
enzyme inhibition with an IC50 value of 96.7 6.8
importantly, the best XO inhibitor among the test compounds,
( )-3d (IC50 = 16.0 0.8 M), showed an even better activity than
that of the positive drug, allopurinol (IC50 = 23.5 2.0 M). Since
lM. More
l
l
allopurinol is currently the front-line chemotherapy agent used
for the treatment of gout, the 5,6,7-trioxygenated dihydroflavonols
designed, in particular ( )-3d, thus deserves to be further devel-
oped as a new XO-inhibiting lead for XO-related diseases, including
gout.
Parallel comparison of the 6-OH dihydroflavonols (series 3) and
their 6-OMe analogues (series 4) shows that the former set (3) pos-
sesses stronger inhibitory activity towards the XO enzyme than the
latter (4) (Table 2). This result is consistent with those obtained in
the OÅ2ꢀ and DPPH scavenging, LPO inhibition and PC12 cell protec-
tion experiments.
tin (EC50 = 68.3 5.3
PC12 protective capacities, with EC50 values of 82.4 9.9
33.1 3.2 M, respectively.
l
M), ( )-3b and ( )-4h possessed remarkable
l
M and
l
It is interesting to note that the pair ( )-3d and ( )-4d seemed
to possess opposite antioxidant effects on the test models: ( )-4d
displayed effective activity only on LPO inhibition (its protective
effect on PC12 cell line was almost negligible) while its 6-OH ana-
logue ( )-3d showed a broader antioxidant spectrum except for
LPO inhibition. This is most likely caused by the different
mechanisms of action conferred by the minor variance of molecu-
lar structures; additional studies should shed light on this
phenomenon.
2.3. Energy-minimized analysis of selected 3D structures
In order to carry out preliminary exploration of the potential
structural-related mechanisms which confer the antioxidant prop-
erties of dihydroflavonols investigated, the energy-minimized 3D
models for selected compounds were established and analyzed. It
could be observed that both baicalein and quercetin, which possess
2,3-double bond moieties, possessed more planar A/C ring systems
than that of taxifolin (Fig. 4). The constructed 3D drawings of
2R,3R-4e has the similar stereochemistry to that of 2R,3R-taxifolin
described by Trouillas et al.36 Both 2R,3R-4e and 2S,3S-4e conform-
ers have, as described for 2R,3R-taxifolin,36 the axial H-2 and H-3
which results in the large coupling constants of these two protons
(Fig. 4 and Section 4). This conformation favors 3-OH’s forming the
hydrogen bond with the adjacent 4-oxo group (Fig. 4A and Fig. 4B).
In either stereochemistry, 2S,3S-trans and 2R,3R-trans, the steri-
cally smaller H-2 and H-3 of ( )-4e always occupies the axial con-
figuration positions (Fig. 4B). This configuration therefore allows
the 3-OH to form a hydrogen bond with C-4 carbonyl groups. How-
ever, by comparison of the enantiomers of ( )-3e possessing 2R,3S-
cis and 2S,3R-cis configurations, it could be observed that the pair
of molecules had to sacrifice the hydrogen bonding possibility be-
tween 4-oxo and 3-OH (for both conformers take axial 3-OHs) in
order to keep the larger C-2 aromatic substituent in an equatorial
orientation (Fig. 4C).
Scrutiny of the minimized-energy figures of the molecules sug-
gests that, in all cases, the B-ring did not demonstrate a co-planary
status with the A/C ring plane, as also noted for the 2,3-unsatu-
rated baicalein and quercetin. Furthermore, previous SAR calcula-
tions suggest that cis- or trans-configurations on the C ring do
not result in major differences in scavenging free radicals, but var-
iance could be found between these two configurations in the
experiments carried out in enzyme models.36 In the present study,
among the compounds tested, the cis-configurated ( )-3d was
found to be the best inhibitor of XO. Therefore, it may be that its
Table 2
Increased cell viability (at a gradient of test concentrations) and inhibition of XO (at
16 l
g/mL) by compounds 3a–e and 4a–d, 4g and 4ha
Compound
Increased percentage of PC12 cell viabilityb,c
g/mL 16 g/mL 32 g/mL
XO Inhibition rated,e
8
l
l
l
( )-3b
( )-3d
( )-4d
( )-4h
Quercetin
Allopurinol n.d.
19.5 1.2%
2.9 0.5%
ꢀ2.5 1.3%
40.5 4.1%
28.9 2.7%
37.2 3.5%
19.8 2.3%
1.9 1.2%
56.1 5.0%
42.6 5.6%
n.d.
53.6 5.9%
22.0 3.0%
2.4 1.1%
74.5 4.6%
61.7 7.2%
n.d.
21.1 1.6%
74.3 8.5%
46.6 5.2%
58.9 3.3%
n.d.f
94.0 5.9%
a
Data are expressed as the mean SD, n = 3.
The increased percentages of PC12 cell viability, respectively at 8, 16 and 32 lg/
b
mL by other compounds are: ꢀ3.8 0.4%, 13.4 0.9%, 19.1 0.9% for ( )-3a;
0.1 0.2%, 0.3 0.2%, 0.5 0.3% for ( )-3c; ꢀ0.2 0.2%, 0.5 0.4%, 1.6 0.6% for ( )-
3e; ꢀ1.4 0.4%, 5.1 0.8%, 17.1 1.2% for ( )-4a; 0.6 0.4%, 3.4 0.8%, 2.1 0.6%
for ( )-4b; 1.4 0.2%, 5.4 2.0%, 4.5 1.8% for ( )-4c; 2.1 0.6%, 9.2 2.3%,
15.1 1.2% for ( )-4g.
c
EC50 means the effective concentration of compound for 50% increase of the
PC12 cell viability. EC50 values for ( )-3b, ( )-4h and quercetin are 82.4 9.9
33.1 3.2 M and 68.3 5.3 M, respectively.
The inhibition rates of XO by other compounds at 16
lM,
l
l
d
l
g/mL are: 19.8 1.1% for
( )-3a; 12.5 0.9% for ( )-3c; 15.4 0.7% for ( )-3e; 0.2 0.1% for ( )-4a;
24.9 1.8% for ( )-4b; 0.8 0.4% for ( )-4c; 28.9 2.1% for ( )-4g.
e
IC50 denotes the concentration that inhibits the xanthine oxidase by 50%. IC50
values for ( )-3d, ( )-4h and allopurinol are 16.0 0.8
23.5 2.0 M, respectively.
n.d. = not detected.
lM, 96.7 6.8 lM and
l
f

J. Gong et al. / Bioorg. Med. Chem. 17 (2009) 3414–3425
3421
and their antioxidant properties were evaluated. The hybrids
showed differential suppression of OÅ2ꢀ and DPPH free radical for-
mation. A large portion of the designed target molecules exhibited
better quenching ability towards these radicals when compared to
Vitamin C. Comparison of the activities of the compounds, respec-
tively belonging to series 3 and 4, lead to the conclusion that 6-hy-
droxy substitution on the A ring is important to the antioxidant
activities. The dihydroflavonol owning an ortho-dihydroxy substi-
tution pattern on the B ring exhibited strong free radical scaveng-
ing activity, which might be attributable to its polyphenolic
character. Some of the active compounds exhibited neuroprotec-
tive properties and/or XO inhibitory activity in some cases superior
to those observed for the positive controls. The implication is that
5-O-methylated dihydroflavonols may serve as the suitable pro-
drug of their 5-hydroxyl metabolites. The present study may help
to further comprehend the mechanisms underlying the antioxida-
tive activity of dihydroflavonols and may therefore be useful for
further optimizing their pharmaceutical application. The pharma-
cological efficacy observed suggests that these 5,6,7-trioxygenated
dihydroflavonols hybridized by a taxifolin-like dihydroflavonol
with a baicalein-like 5,6,7-trioxygenated A ring skeleton are wor-
thy of further investigation including in vivo experiments to iden-
tify their potential in the application of treating ROS-related
diseases such as AD, gout, aging and cardiovascular disorders
resulting from oxidative damage. In addition, we afford series of
naturally rare cis-configurated dihydroflavones, not only enriched
the family of dihydroflavonoids, bust also broaden the pharmaco-
logical applicability of these cis-configurated flavonoids.
A
Baicalein
Quercetin
(2R,3R)-2,3-trans taxifolin
B
C
(2S,3S)-2,3-trans-4e
(2R,3R)-2,3-trans-4e
(2R,3S)-2,3-cis-3e
(2S,3R)-2,3-cis-3e
Figure 4. Minimized-energy analyses of 3D structures of the reference flavone,
flavonol and dihydroflavonols.
4. Experimental
4.1. Materials
specific B-ring orientation might afford a favorable chelating status
to selectively block the active site of the XO enzyme, in addition to
its ability to quench free radicals during purine metabolism.37
Furthermore, the reported SAR calculations suggest that taxifo-
lin possesses an inferior antioxidant activity to that of quercetin
which possesses a 2,3-double bond,36 since the latter may stabilize
the captured free radicals by more efficient electron delocalization.
Surprisingly, as seen from the LPO inhibition results, both the 6-OH
dihydroflavonol ( )-3e and the 6-OMe dihydroflavonol ( )-4e
exhibited their superior inhibitory efficacy against LPO compared
to quercetin, implying that the specific antioxidative mechanisms
of 5,6,7-trioxygenated dihydroflavonols differ from that of querce-
tin. Therefore, it is clear that introduction of an adjacent trioxygen-
ated moiety to the A ring can positively influence the activity of
dihydroflavonols as antioxidants.
In addition, from the finding that ( )-4h possesses higher PC12
cells protective activity than that of quercetin, it is further sur-
mised that the antioxidant efficiency of a molecule does not solely
depend on the amount of its free phenol groups or quantum prop-
erties. Additional inhibitory factors associated with enzymes and/
or receptors may also be involved in the antioxidative protection
of the cells, if they are not the major factors. Enhanced electron
density of the A ring in the synthesized compounds may also in-
duce free radical attack, which elevates the ability of these
highly-oxidized dihydroflavonols to consume free radicals and
thus execute satisfactory antioxidant activities.
Melting points were measured on a Perkin-Taike X-4 apparatus;
uncorrected. 1H NMR and 13C NMR spectra were recorded on a Bru-
ker AM-400 (400 MHz) instrument; chemical shifts d in ppm with
Me4Si as internal standard, coupling constants J in Hertz. ESI-MS
data were recorded on a Bruker Esquire 3000+ spectrometer and
EI-MS was performed on a Varian MAT-95 MS instrument. HRES-
I-MS data were recorded on a Micromass Q-Tof MS spectrometer.
All reactions were carried out in an oven or flame-dried glassware
with magnetic stirring. Solvents were purified by standard proce-
dures. Yields refer to chromatographically and spectroscopically
homogeneous materials unless otherwise stated. Thin-layer chro-
matography was performed on silica gel GF254; detection by UV
light (254 nm). Column chromatography (CC) was carried out on
silica gel (200ꢀ300 mesh; Qingdao Ocean Chemical Plant, China).
The optical density (OD) was measured spectrophotometrically
on an enzyme-linked immunosorbent assay reader (Bio-Tek Syn-
ergy HT, Bio-Tek Instruments Inc, Winooski, VT, USA). 1,1-Diphe-
nyl-2-picrylhydrazyl radical (DPPH), phenazine methosulfate
(PMS), nitroblue tetrazolium (NBT), hydrogen peroxide (H2O2), 3-
(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT),
thiobarbituric acid (TBA), and xanthine were purchased from Sig-
ma Chemical Co. (St. Louis, MO, USA). Tris base, Trixon-100 and
DMEM medium were obtained from Gibco (Grand Island, NY,
USA). Reduced form disodium salt (NADH) was supplied from
Amresco (Solon, OH, USA). Quercetin was prepared in our labora-
tory (HPLC purity of 99%). Vitamin C was purchased from Zhejiang
Xianju Pharmaceutical Co. Ltd (Tiantai, Zhejiang, China). All other
reagents were of the highest purity commercially available. The
rat pheochromocytoma (PC12) cell line was obtained from the
Shanghai Institute of Cell Biology, Chinese Academy of Sciences.
Sprague-Dawley rats were obtained from the Zhejiang Center of
Laboratory Animals, China. The use of animals was in accordance
3. Conclusions
Two series of hybridized compounds composed of both active
pharmacophores of 5,6,7-trioxygenated A ring (from baicalein)
and 2,3-dihydroflavonol moiety (from taxifolin) were synthesized

3422
J. Gong et al. / Bioorg. Med. Chem. 17 (2009) 3414–3425
with Guideline for the Care and Use of Laboratory Animals of Zhe-
jiang University.
trated HCl (0.3 mL) and MeOH (0.4 mL) dropwise. Then the mix-
ture was stirred at 35 °C for 15 min. After cooling, the reaction
mixture was concentrated in vacuo. The residue was poured into
ice-water (10.0 g) and extracted with EtOAc (3 ꢂ 10 mL). The or-
ganic phase was combined and dried over MgSO4, and the filtered
solvent was evaporated to give a yellow oil, which was subjected to
CC (Si gel 10.0 g, EtOAc/petroleum ether 1:2) to afford 42.2 mg of
3d (the major product). Yield: 56%. Yellowish solid. Rf (EtOAc/
petroleum ether 1:2) 0.12. 1H NMR (400 MHz, CD3COCD3) d 3.80
(s, 3H, OCH3), 3.82 (s, 3H, OCH3), 3.94 (s, 3H, OCH3), 3.98 (s, 3H,
OCH3), 4.57 (d, 1H, J = 2.4 Hz, H-3), 5.09 (d, 1H, J = 2.4 Hz, H-2),
6.37 (s, 1H, H-8), 6.92 (d, 1H, J = 8.4 Hz, H-50), 6.96 (dd, 1H,
J = 8.4, 1.6 Hz, H-60), 7.06 (d, 1H, J = 1.6 Hz, H-20); ESI-MS m/z:
377 [M+H]+. Anal. Calcd for C19H20O8: C, 60.64; H, 5.36. Found: C,
60.55; H, 5.18.
4.2. Synthesis38
4.2.1. Synthesis of Compounds 3a–e
4.2.1.1. ( )-trans-2,3-Dihydro-3,6-dihydroxy-5,7-dimethoxy-2-
(4-hydroxyphenyl)-4H-1-benzopyran-4-one [( )-3a]. To a solu-
tion of 11a (95.0 mg, 0.21 mmol, see Supplementary data 3) in
MeOH (3.0 mL) and THF (1.5 mL) was added, a mixture of concen-
trated HCl (0.2 mL) and MeOH (0.5 mL) dropwise. The mixture was
then stirred at 55 °C for 30 min. After cooling, the reaction mixture
was concentrated in vacuo. The residue was poured into ice-water
(10.0 g) and extracted with EtOAc (3 ꢂ 10 mL). The organic phase
was combined and dried over MgSO4, and the filtered solvent
was evaporated to give a yellow oil, which was subjected to CC
(Si gel 10.0 g, EtOAc/petroleum ether 1:2) to afford 51.0 mg of 3a
(the major product). Yield: 75%. Yellowish solid. Rf (EtOAc/petro-
leum ether 1:2) 0.11. 1H NMR (400 MHz, CD3COCD3) d 3.81 (s,
3H, OCH3), 3.87 (s, 3H, OCH3), 4.29 (d, 1H, J = 2.4 Hz, 3-OH), 4.44
(dd, 1H, J = 12.0, 2.4 Hz, H-3), 4.93 (d, 1H, J = 12.0 Hz, H-2), 6.36
(s, 1H, H-8), 6.86 (d, 2H, J = 8.4 Hz, H-30, H-50), 7.37 (d, 2H,
J = 8.4 Hz, H-20, H-60); ESI-MS m/z: 333 [M+H]+. Anal. Calcd for
C17H16O7: C, 61.44; H, 4.85. Found: C, 61.36; H, 4.79.
4.2.1.5. ( )-cis-2,3-Dihydro-3,6-dihydroxy-5,7-dimethoxy-2-
(3-hydroxy-4-methoxyphenyl)-4H-1-benzopyran-4-one[( )-3e].
To a solution of 11e (100.0 mg, 0.2 mmol, see Supplementary data
3) in MeOH (3.6 mL) and THF (1.8 mL) was added, a mixture of con-
centrated HCl (0.3 mL) and MeOH (0.4 mL) dropwise. The mixture
was then stirred at 30 °C for 15 min. After cooling, the reaction
mixture was concentrated in vacuo. The residue was poured into
ice-water (10.0 g) and extracted with EtOAc (3 ꢂ 10 mL). The or-
ganic phase was combined and dried over Na2SO4, and the filtered
solvent was evaporated to give a yellow oil, which was subjected to
CC (Si gel 10.0 g, EtOAc/petroleum ether 1:2) to afford 45.4 mg of
3e (the major product). Yield: 62%. Yellowish solid. Rf (EtOAc/
petroleum ether 1:2) 0.16. 1H NMR (400 MHz, CD3COCD3) d 3.86
(s, 3H, OCH3), 3.94 (s, 3H, OCH3), 3.98 (s, 3H, OCH3), 4.55 (d, 1H,
J = 2.4 Hz, H-3), 5.08 (d, 1H, J = 2.4 Hz, H-2), 6.37 (s, 1H, H-8),
6.81 (d, 1H, J = 8.0 Hz, H-50), 6.89 (dd, 1H, J = 8.0, 1.6 Hz, H-60),
7.05 (d, 1H, J = 1.6 Hz, H-20); ESI-MS m/z: 363 [M+H]+. Anal. Calcd
for C18H18O8: C, 59.67; H, 5.08. Found: C, 59.57; H, 4.92.
4.2.1.2. ( )-trans-2,3-Dihydro-3,6-dihydroxy-5,7-dimethoxy-2-
(4-methoxyphenyl)-4H-1-benzopyran-4-one [( )-3b]. To a solu-
tion of 11b (130.0 mg, 0.3 mmol, see Supplementary data 3) in
MeOH (4.0 mL) and THF (2.0 mL) was added, a mixture of concen-
trated HCl (0.2 mL) and MeOH (0.5 mL) dropwise. The mixture
was then stirred at 45 °C for 30 min. After cooling, the reaction mix-
ture was concentrated in vacuo. The residue was poured into ice-
water (10.0 g) and extracted with EtOAc (3 ꢂ 10 mL). The organic
phase was combined and dried over MgSO4, and the filtered solvent
was evaporated to give a yellow oil, which was subjected to CC (Si
gel 12.0 g, EtOAc/petroleum ether 1:3) to afford 68.2 mg of 3b.
Yield: 66%. Yellowish solid. Rf (EtOAc/petroleum ether 1:3) 0.15.
1H NMR (400 MHz, CDCl3) d 3.84 (s, 3H, OCH3), 3.91 (s, 3H,
OCH3), 3.98 (s, 3H, OCH3), 4.49 (d, 1H, J = 12.0 Hz, H-3), 4.99 (d,
1H, J = 12.0 Hz, H-2), 6.35 (s, 1H, H-8), 6.98 (d, 1H, J = 8.4 Hz, H-30,
H-50), 7.48 (d, 1H, J = 8.4 Hz, H-20, H-60); ESI-MS m/z: 347 [M+H]+;
HRESI-MS m/z: 345.0977 [MꢀH]+ (calcd for C18H17O7: 345.0969).
4.2.2. Synthesis of Compounds 4a–j
4.2.2.1. ( )-trans-2,3-Dihydro-3-hydroxy-5,6,7-trimethoxy-2-
(4-hydroxyphenyl)-4H-1-benzopyran-4-one [( )-4a]. To a solu-
tion of 19a (130.0 mg, 0.30 mmol, see Supplementary data 3) in
MeOH (4.0 mL) and THF (2.0 mL) was added, a mixture of concen-
trated HCl (0.3 mL) and MeOH (0.5 mL) dropwise. The mixture was
then stirred at 55 °C for 30 min. The reaction mixture was concen-
trated in vacuo. The residue was poured into ice-water (12.0 g) and
extracted with EtOAc (3 ꢂ 10 mL). The organic phase was com-
bined and dried over MgSO4, and the filtered solvent was evapo-
rated to give a yellow oil, which was subjected to CC (Si gel
12.0 g, EtOAc/petroleum ether 1:3) to afford 62.2 mg of 4a (the ma-
jor product). Yield: 60%. Yellowish solid. Rf (EtOAc/petroleum ether
1:3) 0.20. 1H NMR (400 MHz, CD3COCD3) d 3.70 (s, 3H, OCH3), 3.82
(s, 3H, OCH3), 3.97 (s, 3H, OCH3), 4.47 (d, 1H, J = 12.0 Hz, H-3), 4.95
(d, 1H, J = 12.0 Hz, H-2), 6.35 (s, 1H, H-8), 6.86 (d, 2H, J = 8.0 Hz, H-
30, H-50), 7.38 (d, 2H, J = 8.0 Hz, H-20, H-60); 13C NMR (100 MHz,
CD3COCD3) d 56.6 (OCH3), 61.2 (OCH3), 61.9 (OCH3), 73.6 (C-2),
84.2 (C-3), 97.2 (C-8), 107.5 (C-4a), 115.7 (C-30, C-50), 128.8 (C-
10), 130.1 (C-20, C-60), 138.3 (C-6), 154.4 (C-5), 158.8 (C-8a), 160.4
(C-7), 160.8 (C-40), 192.0 (C-4); ESI-MS m/z: 347 [M+H]+; HRESI-
MS m/z: 345.0977 [MꢀH]+ (calcd for C18H17O7: 345.0969).
4.2.1.3. ( )-cis-2,3-Dihydro-3,6-dihydroxy-5,7-dimethoxy-2-(1,3-
benzodioxol-5-yl)-4H-1-benzopyran-4-one [( )-3c]. To a solu-
tion of 11c (90.0 mg, 0.2 mmol, see Supplementary data 3) in
MeOH (3.0 mL) and THF (1.5 mL) was added, a mixture of concen-
trated HCl (0.3 mL) and MeOH (0.4 mL) dropwise. The mixture was
then stirred at 35 °C for 15 min. After cooling, the reaction mixture
was concentrated in vacuo. The residue was poured into ice-water
(10.0 g) and extracted with EtOAc (3 ꢂ 10 mL). The organic phase
was combined and dried over MgSO4, and the filtered solvent
was evaporated to give a yellow oil, which was subjected to CC
(Si gel 10.0 g, EtOAc/petroleum ether 1:3) to afford 41.9 mg of 3c
(the major product). Yield: 58%. Yellowish solid. Rf (EtOAc/petro-
leum ether 1:3) 0.25. 1H NMR (400 MHz, CDCl3) d 3.94 (s, 3H,
OCH3), 3.96 (s, 3H, OCH3), 4.57 (d, 1H, J = 2.0 Hz, H-3), 5.26 (d,
1H, J = 2.0 Hz, H-2), 5.95 (s, 2H, OCH2O), 6.35 (s, 1H, H-8), 6.81
(d, 1H, J = 8.0 Hz, H-50), 6.92 (dd, 1H, J = 8.0, 1.6 Hz, H-60), 7.01 (d,
1H, J = 1.6, H-20); ESI-MS m/z: 361 [M+H]+. HRESI-MS m/z:
359.0726 [MꢀH]+ (calcd for C18H15O8: 359.0761).
4.2.2.2. ( )-cis-2,3-Dihydro-3-hydroxy-5,6,7-trimethoxy-2-(4-
methoxyphenyl)-4H-1-benzopyran-4-one [( )-4b]. To a solu-
tion of 19b (161.0 mg, 0.40 mmol, see Supplementary data 3) in
MeOH (4.6 mL) and THF (2.3 mL) was added, a mixture of concen-
trated HCl (0.3 mL) and MeOH (0.5 mL) dropwise, ice was added to
the water bath to lower the reaction temperature under 35 °C. The
reaction mixture was stirred for 15 min and was concentrated in
vacuo. The residue was poured into ice-water (12.0 g) and ex-
4.2.1.4. ( )-cis-2,3-Dihydro-3,6-dihydroxy-5,7-dimethoxy-2-(3,4-
dimethoxyphenyl)-4H-1-benzopyran-4-one [( )-3d]. To a solu-
tion of 11d (93.0 mg, 0.2 mmol, see Supplementary data 3) in
MeOH (3.0 mL) and THF (1.5 mL) was added, a mixture of concen-

J. Gong et al. / Bioorg. Med. Chem. 17 (2009) 3414–3425
3423
tracted with EtOAc (3 ꢂ 10 mL). The organic phase was combined
and dried over MgSO4, and the filtered solvent was evaporated to
give a yellow oil, which was subjected to CC (Si gel 15.0 g,
EtOAc/petroleum ether 1:2) to afford 86.0 mg of 4b (the major
product). Yield: 60%. Yellowish solid. Rf (EtOAc/petroleum ether
1:2) 0.39. 1H NMR (400 MHz, CD3COCD3) d 3.80 (s, 3H, OCH3),
3.81 (s, 3H, OCH3), 3.94 (s, 3H, OCH3), 4.01 (s, 3H, OCH3), 4.57 (d,
1H, J = 2.4 Hz, H-3), 5.06 (d, 1H, J = 2.4 Hz, H-2), 6.38 (s, 1H, H-8),
6.93 (d, 2H, J = 8.0 Hz, H-30, H-50), 7.38 (d, 2H, J = 8.0 Hz, H-20, H-
60); ESI-MS m/z: 361 [M+H]+. Anal. Calcd for C19H20O7: C, 63.33;
H, 5.59. Found: C, 62.24; H, 5.51.
DMSO-d6) d 3.73 (s, 3H, OCH3), 3.84 (s, 3H, OCH3), 3.88 (s, 3H,
OCH3), 3.90 (s, 3H, OCH3), 4.38 (d, 1H, J = 12.0 Hz, H-3), 5.00 (d,
J = 12.0 Hz, H-2), 6.50 (s, 1H, H-8), 6.82 (d, J = 8.0 Hz, H-50), 6.93
(s, H-20), 6.98 (d, J = 8.0 Hz, H-60); ESI-MS m/z: 377 [M+H]+. Anal.
Calcd for C19H20O8: C, 60.63; H, 5.36. Found: C, 60.51; H, 5.32.
4.2.2.6. ( )-trans-2,3-Dihydro-3-hydroxy-5,6,7-trimethoxy-2-
(3-hydroxy-4-methoxyphenyl)-4H-1-benzopyran-4-one [( )-4f].
To a solution of 19f (163.0 mg, 0.35 mmol, see Supplementary data
3) in MeOH (5.4 mL) and THF (2.7 mL) was added, a mixture of con-
centrated HCl (0.4 mL) and MeOH (0.6 mL) dropwise. The mixture
was then stirred at 45 °C for 30 min. The reaction mixture was con-
centrated in vacuo. The residue was poured into ice-water (14.0 g)
and extracted with EtOAc (3 ꢂ 10 mL). The organic phase was com-
bined and dried over MgSO4, and the filtered solvent was evapo-
rated to give a yellow oil, which was subjected to CC (Si gel
15.0 g, EtOAc/petroleum ether 1:2) to afford 92.5 mg of 4f (the ma-
jor product). Yield: 70%. Yellowish solid. Rf (EtOAc/petroleum ether
1:2) 0.24. 1H NMR (400 MHz, CDCl3) d 3.83 (s, 3H, OCH3), 3.87 (s,
3H, OCH3), 3.92 (s, 3H, OCH3), 3.99 (s, 3H, OCH3), 4.46 (d, 1H,
J = 12.0 Hz, H-3), 4.95 (d, 1H, J = 12.0 Hz, H-2), 6.32 (s, 1H, H-8),
6.92 (d, 1H, J = 8.0 Hz, H-50), 7.03 (dd, 1H, J = 8.0, 1.6 Hz, H-60),
7.16 (d, 1H, J = 1.6 Hz, H-20); 13C NMR (100 MHz, CD3COCD3) d
56.0 (OCH3), 56.2 (OCH3), 61.3 (OCH3), 61.8 (OCH3), 72.8 (C-2),
83.3 (C-3), 96.4 (C-8), 106.2 (C-4a), 113.5 (C-50), 119.7 (C-60),
129.2 (C-10), 137.6 (C-6), 145.8 (C-30), 147.3 (C-40), 153.8 (C-5),
159.6 (C-8a), 160.3 (C-7), 191.2 (C-4); ESI-MS m/z: 377 [M+H]+.
Anal. Calcd for C19H20O8: C, 60.63; H, 5.36. Found: C, 60.58; H, 5.41.
4.2.2.3. ( )-cis-2,3-Dihydro-3-hydroxy-5,6,7-trimethoxy-2-(1,3-
benzodioxol-5-yl)-4H-1-benzopyran-4-one [( )-4c]. To a solu-
tion of 19c (84.0 mg, 0.20 mmol, see Supplementary data 3) in
MeOH (4.0 mL) and THF (2.0 mL) was added, a mixture of concen-
trated HCl (0.3 mL) and MeOH (0.5 mL) dropwise, ice was added to
the water bath to lower the reaction temperature under 32 °C. The
reaction mixture was stirred for 15 min and was concentrated in
vacuo. The residue was poured into ice-water (10.0 g) and ex-
tracted with EtOAc (3 ꢂ 10 mL). The organic phase was combined
and dried over MgSO4, and the filtered solvent was evaporated to
give a yellow oil, which was subjected to CC (Si gel 10.0 g,
EtOAc/petroleum ether 1:3) to afford 41.1 mg of 4c (the major
product). Yield: 55%. Yellowish solid. Rf (EtOAc/petroleum ether
1:3) 0.28. 1H NMR (400 MHz, CDCl3) d 3.86 (s, 3H, OCH3), 3.92 (s,
3H, OCH3), 3.97 (s, 3H, OCH3), 4.51 (br s, 1H, H-3), 5.09 (br s, H-
2), 5.90 (s, 2H, OCH2O), 6.32 (s, 1H, H-8), 6.82 (d, 1H, J = 8.0 Hz,
H-50), 6.88 (dd, 1H, J = 8.0, 1.2 Hz, H-60), 7.00 (d, 1H, J = 1.2 Hz, H-
20); ESI-MS m/z: 375 [M+H]+. Anal. Calcd for C19H18O8: C, 60.96;
H, 4.84. Found: C, 60.88; H, 4.86.
4.2.2.7. ( )-cis-2,3-Dihydro-3-hydroxy-5,6,7-trimethoxy-2-(3,4,5-
trimethoxyphenyl)-4H-1-benzopyran-4-one [( )-4g]. To a solu-
tion of 19g (162.0 mg, 0.35 mmol, see Supplementary data 3) in
MeOH (5.6 mL) and THF (2.8 mL) was added, a mixture of concen-
trated HCl (0.5 mL) and MeOH (0.5 mL) dropwise, ice was added to
the water bath to lower the reaction temperature under 35 °C. The
reaction mixture was kept stirring for 15 min and was concen-
trated in vacuo. The residue was poured into ice-water (12.0 g)
and extracted with EtOAc (3 ꢂ 10 mL). The organic phase was com-
bined and dried over MgSO4, and the filtered solvent was evapo-
rated to give a yellow oil, which was subjected to CC (Si gel
15.0 g, EtOAc/petroleum ether 1:2) to afford 52.5 mg of 4g (the ma-
jor product). Yield: 36%. Yellowish solid. Rf (EtOAc/petroleum ether
1:2) 0.36. 1H NMR (400 MHz, CD3COCD3) d 3.70 (s, 3H, OCH3), 3.81
(s, 3H, OCH3), 3.84 (s, 3H, OCH3), 3.94 (s, 6H, 2 ꢂ OCH3), 4.06 (s, 3H,
OCH3), 4.51 (d, 1H, J = 2.4 Hz, H-3), 5.15 (d, 1H, J = 2.4 Hz, H-2), 6.35
(s, 1H, H-8), 6.72 (s, 2H, H-20, H-60); ESI-MS m/z: 421 [M+H]+. Anal.
Calcd for C21H24O9: C, 60.00; H, 5.75. Found: C, 59.85; H, 5.67.
4.2.2.4. ( )-cis-2,3-Dihydro-3-hydroxy-5,6,7-trimethoxy-2-(3,4-
dimethoxyphenyl)-4H-1-benzopyran-4-one [( )-4d]. To a solu-
tion of 19d (130.0 mg, 0.30 mmol, see Supplementary data 3) in
MeOH (5.0 mL) and THF (2.5 mL) was added, a mixture of concen-
trated HCl (0.4 mL) and MeOH (0.5 mL) dropwise, ice was added to
the water bath to lower the reaction temperature under 32 °C. The
reaction mixture was stirred for 15 min and was concentrated in
vacuo. The residue was poured into ice-water (13.0 g) and ex-
tracted with EtOAc (3 ꢂ 10 mL). The organic phase was combined
and dried over MgSO4, and the filtered solvent was evaporated to
give a yellow oil, which was subjected to CC (Si gel 15.0 g,
EtOAc/petroleum ether 1:3) to afford 52.7 mg of 4c (the major
product). Yield: 45%. Yellowish solid. Rf (EtOAc/petroleum ether
1:3) 0.30. 1H NMR (400 MHz, CD3COCD3) d 3.80 (s, 3H, OCH3),
3.81 (s, 3H, OCH3), 3.82 (s, 3H, OCH3), 3.94 (s, 3H, OCH3), 4.02 (s,
3H, OCH3), 4.54 (d, 1H, J = 2.0 Hz, H-3), 5.09 (d, 1H, J = 2.0 Hz, H-
2), 6.36 (s, 1H, H-8), 6.92 (d, 1H, J = 8.0 Hz, H-50), 6.96 (dd, 1H,
J = 8.0, 1.6 Hz, H-60), 7.05 (d, 1H, J = 1.6 Hz, H-20); ESI-MS m/z:
391 [M+H]+. Anal. Calcd for C20H22O8: C, 61.53; H, 5.68. Found: C,
61.44; H, 5.59.
4.2.2.8. ( )-trans-2,3-Dihydro-3-hydroxy-5,6,7-trimethoxy-2-(3,4-
dihydroxyphenyl)-4H-1-benzopyran-4-one [( )-4h]. To a solu-
tion of 19h (150.0 mg, 0.30 mmol, see Supplementary data 3) in
MeOH (5.6 mL) and THF (2.8 mL) was added, a mixture of concen-
trated HCl (0.4 mL) and MeOH (0.6 mL) dropwise. The mixture was
then stirred at 55 °C for 30 min. The reaction mixture was concen-
trated in vacuo. The residue was poured into ice-water (13.0 g) and
extracted with EtOAc (3 ꢂ 10 mL). The organic phase was com-
bined and dried over MgSO4, and the filtered solvent was evapo-
rated to give a yellow oil, which was subjected to CC (Si gel
12.0 g, EtOAc/petroleum ether 2:3) to afford 58.3 mg of 4h (the
major product). Yield: 53%. Yellowish solid. Rf (EtOAc/petroleum
ether 1:2) 0.12. 1H NMR (400 MHz, DMSO-d6) d 3.73 (s, 3H,
OCH3), 3.84 (s, 3H, OCH3), 3.88 (s, 3H, OCH3), 4.38 (d, 1H,
J = 12.0 Hz, H-3), 5.00 (d, 1H, J = 12.0 Hz, H-2), 6.50 (s, 1H, H-8),
6.82 (d, 1H, J = 8.0 Hz, H-50), 6.93 (s, 1H, H-20), 6.99 (d, 1H,
J = 8.0 Hz, H-60); 13C NMR (100 MHz, CD3COCD3) d 56.6 (OCH3),
60.8 (OCH3), 61.4 (OCH3), 72.7 (C-2), 82.9 (C-3), 96.7 (C-8), 107.0
4.2.2.5. ( )-trans-2,3-Dihydro-3-hydroxy-5,6,7-trimethoxy-2-
(4-hydroxy-3-methoxyphenyl)-4H-1-benzopyran-4-one [( )-
4e]. To a solution of 19e (140.0 mg, 0.30 mmol, see Supplemen-
tary data 3) in MeOH (4.8 mL) and THF (2.4 mL) was added, a
mixture of concentrated HCl (0.3 mL) and MeOH (0.5 mL) drop-
wise. The mixture was then stirred at 45 °C for 30 min. The reac-
tion mixture was concentrated in vacuo. The residue was poured
into ice-water (12.0 g) and extracted with EtOAc (3 ꢂ 10 mL). The
organic phase was combined and dried over MgSO4, and the fil-
tered solvent was evaporated to give a yellow oil, which was
subjected to CC (Si gel 15.0 g, EtOAc/petroleum ether 1:2) to af-
ford 79.4 mg of 4e (the major product). Yield: 70%. Yellowish so-
lid. Rf (EtOAc/petroleum ether 1:3) 0.14. 1H NMR (400 MHz,

3424
J. Gong et al. / Bioorg. Med. Chem. 17 (2009) 3414–3425
(C-4a), 115.1(C-20), 115.3 (C-50), 119.3 (C-60), 128.1 (C-10), 136.9 (C-
6), 144.9 (C-40), 145.7 (C-30), 153.1 (C-5), 158.8 (C-8a), 159.1 (C-7),
191.1 (C-4); ESI-MS m/z: 363 [M+H]+; HRESI-MS m/z: 361.0930
[MꢀH]+ (calcd for C18H17O8: 361.0918).
of the stable radical DPPH radical.40 The effect of test compound on
free radical scavenging was reflected by the discoloration of DPPH
radical. In brief, reaction mixtures dissolved in methanol contain-
ing various concentrations of the test compounds dissolved in
DMSO and DPPH (0.4 mg/mL). The methanolic solution of DPPH
served as a control while quercetin and Vitamin C were used as ref-
erence free radical scavengers. The absorbance was measured at
517 nm after the mixture was incubated at 37 °C for 30 min.
4.2.2.9. ( )-cis-2,3-Dihydro-3-hydroxy-5,6,7-trimethoxy-2-(2-
hydroxyphenyl)-4H-1-benzopyran-4-one [( )-4i]. To a solution
of 19b (152.0 mg, 0.35 mmol, see Supplementary data 3) in
MeOH (4.6 mL) and THF (2.3 mL) was added, a mixture of con-
centrated HCl (0.4 mL) and MeOH (0.5 mL) dropwise, ice was
added to the water bath to lower the reaction temperature under
35 °C. The reaction mixture was kept stirring for 15 min and was
concentrated in vacuo. The residue was poured into ice-water
(12.0 g) and extracted with EtOAc (3 ꢂ 10 mL). The organic phase
was combined and dried over MgSO4, and the filtered solvent
was evaporated to give a yellow oil, which was subjected to CC
(Si gel 12.0 g, EtOAc/petroleum ether 1:3) to afford 61.8 mg of
4i (the major product). Yield: 53%. Yellowish solid. Rf (EtOAc/
petroleum ether 1:3) 0.22. 1H NMR (400 MHz, CDCl3) d 3.88 (s,
3H, OCH3), 3.94 (s, 3H, OCH3), 3.96 (s, 3H, OCH3), 4.72 (d, 1H,
J = 1.2 Hz, H-3), 5.24 (d, 1H, J = 1.2 Hz, H-2), 6.29 (s, 1H, H-8),
6.86–7.25 (m, 4 arom. H); ESI-MS m/z: 347 [M+H]+. Anal. Calcd
for C18H18O7: C, 62.42; H, 5.24. Found: C, 62.32; H, 5.18.
4.5. Inhibition of lipid peroxidation
The formation of malondialdehyde (MDA) was used as a mea-
sure of lipid peroxidation and was determined by the thiobarbitu-
ric acid (TBA) assay on rat liver homogenate in ice-cold PBS, using
colorimetry.41 The reaction mixture was composed of 200
l
L of
reaction solution, containing an aqueous solution of FeSO4
(4 M), Vitamin C (50 M) and rat liver homogenate, 4 L of syn-
thesized compounds (concentration from 0.1 to 100 M) was incu-
bated at 37 °C in capped tubes for 1 h, then 100 of
trichloroacetic acid (20%, v/v) was added to the mixture and re-
acted at room temperature for 30 min. Finally, 200 L of HCl
(0.1 M) and 100 L of TBA (1%, w/v) were added to each tube and
l
l
l
l
l
L
l
l
were incubated at 99 °C for 1 h. Centrifugation was carried out at
1000 rpm for 5 min, and the absorbance was measured at
532 nm. Quercetin was used as a standard.
4.2.2.10. ( )-cis-2,3-Dihydro-3-hydroxy-5,6,7-trimethoxy-2-(4-
hydroxy-3,5-dimethoxyphenyl)-4H-1-benzopyran-4-one [( )-4j].
To a solution of 19j (100.0 mg, 0.20 mmol, see Supplementary
data 3) in MeOH (7.0 mL) was added, a mixture of concentrated
HCl (0.25 mL) and MeOH (0.5 mL) dropwise. Ice was added to
the water bath to keep the reaction temperature under 35 °C.
The reaction mixture was stirred for 15 min and was concen-
trated in vacuo. The residue was poured into ice-water (10.0 g)
and extracted with EtOAc (3 ꢂ 10 mL). The organic phase was
combined and dried over MgSO4, and the filtered solvent was
evaporated to give a yellow oil, which was subjected to CC (Si
gel 10.0 g, EtOAc/petroleum ether 1:3) to afford 46.0 mg of 4j
(the major product). Yield: 56%. Yellowish solid. Rf (EtOAc/petro-
leum ether 1:3) 0.27. 1H NMR (400 MHz, CD3COCD3) d 3.85 (s,
3H, OCH3), 3.90 (s, 3H, OCH3), 3.92 (s, 6H, 2 ꢂ OCH3), 3.98 (s,
3H, OCH3), 4.47 (d, 1H, J = 1.2 Hz, H-3), 5.16 (d, J = 1.2 Hz, H-2),
6.31 (s, 1H, H-8), 6.68 (s, 2H, H-20, H-60); ESI-MS m/z: 407
[M+H]+. Anal. Calcd for C20H22O9: C, 59.11; H, 5.46. Found: C,
58.98; H, 5.41.
4.6. Protection of PC12 against H2O2 induced cell injury
The rat pheochromocytoma (PC12) cells were maintained at
37 °C in a humidified atmosphere containing 5% CO2. PC12 cells
were seeded into multiwall plates (96) at a density of 6 ꢂ 103 per
well in DMEM (high sucrose) medium, supplemented with 10%
heat-inactivated bovine calf serum, 100 units/mL penicillin, and
100 units/mL of streptomycin. All experiments were carried out
36 h after cells were seeded. The PC12 cells were preincubated
with samples which were dissolved in DMSO and diluted with
medium to the final concentrations of 32, 16, 8
tively. After 2 h, H2O2 (diluted by PBS, the final concentration
was 500 M) was added. Assays for cell viability were performed
lg/mL, respec-
l
4 h after H2O2 was added. Cell survivals were evaluated by two
methods: morphological observation with phase-contrast micro-
scope and MTT assay. A fresh solution of MTT (5 mg/mL) prepared
in NaCl (0.9%) solution was added to each well, and the plates were
incubated in a CO2 incubator at 37 °C for additional 3 h. The MTT
solution was aspirated off. 200 lL of DMSO was added to each well
4.3. Superoxide anion scavenging activity
to solubilize the formazan crystals in viable cells. The absorbance
was measured at 570 nm. Compared with the normal cells, the via-
bility of cells treated with drugs is calculated by OD (drug-treated)/
OD (normal cells) ꢂ 100%, while the negative control is calculated
as OD (DMSO, which is used to exclude the solvent effect)/OD (nor-
mal cells) ꢂ 100%. Thus the increased cell viability could be calcu-
lated as the following formula:
The superoxide anion scavenging activities of synthesized
compounds were assayed spectrophotometrically as reported
with a slight modification.39 Superoxide anion radicals were gen-
erated in a non-enzymic phenazine methosulfate-NADH system
by following of the reduction of nitroblue tetrazolium. In this as-
say, the superoxide anion radicals were measured in plates,
½ODðdrug-treatedÞ=ODðnormal cellsÞ ꢂ 100%ꢃ
ꢀ ½ODðDMSOÞ=ODðnormal cellsÞ ꢂ 100%ꢃ:
which contained 78
lM of NADH, 50 lM of nitroblue tetrazo-
lium, 5 M of phenazine methosulfate and the tested samples
l
with different concentrations in 16 mM Tris–HCl buffer at pH
8.0. The luminosity was monitored at 560 nm after 5 min of
the incubation at room temperature. The blank samples did not
contain phenazine methosulfate. Quercetin and Vitamin C were
used as reference inhibitors.
4.7. Inhibition of xanthine oxidase
The XO activity was evaluated by spectrophotometric measure-
ment of the formation of formazan.42 The reaction mixture in the
sample wells consisted of xanthine oxidase (obtained from rat li-
4.4. DPPH free radical scavenging activity
ver, 600
0.01 M, pH 8.75 (30
PMS (30 L, 100 M final concentration), Triton X-100 (10
0.4%) and the test compounds (30 L, at an original concentration
l
L, 540
l
M final concentration) in phosphate buffer
L), NBT (30 L, 100 M, final concentration),
L,
l
l
l
Quenching of free radicals by the synthesized compounds was
assayed spectrophotometrically at 517 nm against the absorbance
l
l
l
l

J. Gong et al. / Bioorg. Med. Chem. 17 (2009) 3414–3425
3425
17. Haraguchi, H.; Mochida, Y.; Sakai, S.; Masuda, H.; Tamura, Y.; Mizutani, K.;
Tanaka, O.; Chou, W. H. Biosci. Biotechnol. Biochem. 1996, 60, 945.
18. Montoro, P.; Braca, A.; Pizza, C.; De Tommasi, N. Food Chem. 2005, 92, 349.
19. Zhang, Q. -F.; Zhang, Z. -R.; Cheung, H. -Y. Food Chem. 2008. doi: 10.1016/
j.foodchem.2008.11.053.
of 16 lg/mL). After 2 h of incubation at 37 °C in water bath, the
absorbance was read at 550 nm. Xanthine was omitted in the blank
samples. Allopurinol was used as positive control for the assay.
20. Hernández, V.; Recio, M. C.; Máñez, S.; Giner, R. M.; Ríos, J.-L. Life Sci. 2007, 81,
480.
Acknowledgements
21. Lapchak, P. A.; Maher, P.; Schubert, D.; Zivin, J. A. Neuroscience 2007, 150, 585.
22. Lee, Y.; Yeo, H.; Liu, S. H.; Jiang, Z.; Savizky, R. M.; Austin, D. J.; Cheng, Y. C. J.
Med. Chem. 2004, 47, 5555.
23. Fernández, M.; Caballero, J.; Helguera, A. M.; Castro, E. A.; González, M. P.
Bioorg. Med. Chem. 2005, 13, 3269.
24. Morita, M.; Takahashi, I.; Kanai, M.; Okafuji, F.; Iwashima, M.; Hayashi, T.;
Watanabe, S.; Hamazaki, T.; Shimozawa, N.; Suzuki, Y.; Furuya, H.; Yamada, T.;
Imanaka, T. FEBS Lett. 2005, 579, 409.
25. Kavvadias, D.; Sand, P.; Youdim, K. A.; Qaiser, M. Z.; Rice-Evans, C.; Baur, R.;
Sigel, E.; Rausch, W. D.; Riederer, P.; Schreier, P. Br. J. Pharmacol. 2004, 142, 811.
26. Gong, J. X.; Feng, Y. B.; Wang, F.; Wang, Y. G.; Li, H. B.; Wu, Y. H.; Hao, X. J.; Wu,
X. M.; Bai, H.; Stöckigt, J.; Zhao, Y. Chin. Chem. Lett. 2006, 17, 449.
27. Duan, X. F.; Zhang, Z. B.; Duan, X. H. Chin. J. Org. Chem. 2003, 23, 353.
28. Prescott, A. G.; Stamford, N. P. J.; Wheeler, G.; Firmin, J. L. Phytochemistry 2002,
60, 589.
This work is partially supported by DAAD-CSC PPP Project (CSC
[2004] 3067) and intramural foundation from Wenzhou Medical
College. Y.Z. expresses his thanks to the Chinese Ministry of Educa-
tion as well as to Mr. Ka-shing Li for the ‘Cheung Kong Scholar
Chair Professorship’ at Zhejiang University. We thank Professor Da-
neel Ferreira and Professor Lihe Zhang (Beijing University) for their
kind comments and useful discussions on this research topic.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2009.03.032.
29. Markham, K. R.; Geiger, H. In The Flavonoids: Advances in Research Since 1986;
Harborne, J. B., Ed.; Chapman and Hall: London, 1994; p 441.
30. Islam, Md. T.; Tahara, S. Phytochemistry 2000, 54, 901.
31. Marais, J. P. J.; Ferreira, D.; Slade, D. Phytochemistry 2005, 66, 2145.
32. Li, S.; Pan, M.-H.; Lo, C.-Y.; Tan, D.; Wang, Y.; Shahidi, C.; Ho, C.-T. J. Funct. Foods
2009, 1, 2.
References and notes
1. Alexandrova, M. L.; Bochev, P. G. Free Radical Biol. Med. 2005, 39, 297.
2. Kossi, M. M. H. E.; Zakhary, M. M. Stroke 2000, 31, 1889.
3. Everse, J.; Coates, P. W. Free Radical Biol. Med. 2005, 38, 1296.
4. Perry, G.; Cash, A. D.; Smith, M. A. J. Biomed. Biotechnol. 2002, 2, 120.
5. Markesbery, W. R. Free Radical Biol. Med. 1997, 23, 134.
6. Lefer, D. J.; Granger, D. N. Am. J. Med. 2000, 109, 315.
7. Romero, J. C.; Reckelhoff, J. F. Hypertension 1999, 34, 943.
8. Sukkar, S. G.; Rossi, E. Autoimmun. Rev. 2004, 3, 199.
9. Toyokuni, S.; Okamoto, K.; Yodoi, J.; Hiai, H. FEBS Lett. 1995, 358, 1.
10. Schwarz, K. B. Free Radical Biol. Med. 1996, 21, 641.
11. Harris, C. M.; Sanders, S. A.; Massey, V. J. Biol. Chem. 1999, 274, 4561.
12. Hogg, N. Semin. Reprod. Endocrinol. 1998, 16, 241.
13. Pascual, E. Curr. Opin. Rheumatol. 2000, 12, 213.
14. Jovanovic, S. V.; Steenken, S.; Tosic, M.; Marjanovic, B.; Simic, M. G. J. Am. Chem.
Soc. 1994, 116, 4846.
15. Ramanathan, R.; Das, N. P.; Tan, C. H. Int. J. Oncol. 1993, 3, 115.
16. Matsumoto, T.; Tahara, S. Nippon Nogeik. Kaishi 2001, 75, 659.
33. Jayasena, T.; Grant, R. S.; Keerthisinghe, N.; Solaja, I.; Smythe, G. A. Neurosci.
Res. 2007, 57, 454.
34. Sultana, R.; Newman, S.; Mohmmad-Abdul, H.; Keller, J. N.; Butterfield, D. A.
Free Radical Res. 2004, 38, 449.
35. Luo, J.; Robinson, J. P.; Shi, R. Neurochem. Int. 2005, 47, 449.
36. Trouillas, P.; Fagnère, C.; Lazzaroni, R.; Calliste, C.; Marfak, A.; Duroux, J.-L. Food
Chem. 2004, 88, 571.
37. Wang, F.; Yang, L.; Huang, K.; Li, X.; Hao, X.; Stöckigt, J.; Zhao, Y. Nat. Prod. Res.
2007, 21, 196.
38. Detailed synthesis of the intermediates and their physical data are reported in
Supplementary data 3.
39. Robak, J.; Gryglewski, R. J. Biochem. Pharmacol. 1988, 37, 837.
40. Tapia, A.; Rodriguez, J.; Theoduloz, C.; Lopez, S.; Feresin, G. E.; Schmeda-
Hirschmann, G. J. Ethnopharmacol. 2004, 95, 155.
41. Lee, S. E.; Ju, E. M.; Kim, J. H. Exp. Mol. Med. 2002, 34, 100.
42. Majkic´-Singh, N.; Bogavac, L.; Kalimanovska, V.; Jelic´, Z.; Spasic´, S. Clin. Chim.
Acta 1987, 162, 29.