Feruloylacetone as the model compound of half-curcumin: Synthesis and antioxidant properties

Article abstract of DOI:10.1016/j.ejmech.2011.01.039

In order to clarify the contribution of phenolic and enolic hydroxyl group to the antioxidant capacity of feruloylacetone, a model compound of half-curcumin, 6-(p-hydroxy-m-methoxyphenyl)-5-hexene-2,4-dione (FT), 6-(p-benzyloxy-m-methoxyphenyl)-5-hexene-2,4-dione (BMFT), 6-(m,p- dihydroxyphenyl)-5-hexene-2,4-dione (DDFT), 6-(p-hydroxy-m-methoxyphenyl)hexane- 2,4-dione (DHFT), 6-(p-hydroxy-m-methoxyphenyl)-5-hexene-2,4-diol (THFT), and ethyl 2-(p-hydroxy-m-methoxybenzylidene)-3-oxobutanoate (EOFT) were synthesized. The radical-scavenging abilities of these compounds were tested by trapping 2,2′-azinobis(3-ethylbenzothiazoline-6-sulfonate) cationic radical (ABTS⁺), 2,2′-diphenyl-1-picrylhydrazyl (DPPH), and galvinoxyl radicals. The reductive capacities were screened by quenching singlet oxygen and by inhibiting the oxidation of linoleic acid. They were also employed to inhibit the oxidation of DNA mediated by hydroxyl radical and 2,2′-azobis(2-amidinopropane hydrochloride) (AAPH). In addition, they were applied to protect erythrocytes against AAPH- and hemin-induced hemolysis. The obtained results revealed that the antioxidant capacity of half-curcumin was derived from the phenolic-OH and the conjugated linkage between phenolic and enolic-OH. The enolic-OH itself cannot trap radicals.

Full text of DOI:10.1016/j.ejmech.2011.01.039

European Journal of Medicinal Chemistry 46 (2011) 1198e1206
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Feruloylacetone as the model compound of half-curcumin:
Synthesis and antioxidant properties
*
Jian-Ying Feng, Zai-Qun Liu
Department of Organic Chemistry, College of Chemistry, Jilin University, No. 2519 Jiefang Road, Changchun 130021, China
a r t i c l e i n f o
a b s t r a c t
Article history:
In order to clarify the contribution of phenolic and enolic hydroxyl group to the antioxidant capacity of fer-
uloylacetone, a model compound of half-curcumin, 6-(p-hydroxy-m-methoxyphenyl)-5-hexene-2,4-dione
(FT), 6-(p-benzyloxy-m-methoxyphenyl)-5-hexene-2,4-dione (BMFT), 6-(m,p-dihydroxyphenyl)-5-hexene-
2,4-dione (DDFT), 6-(p-hydroxy-m-methoxyphenyl)hexane-2,4-dione (DHFT), 6-(p-hydroxy-m-methox-
yphenyl)-5-hexene-2,4-diol (THFT), and ethyl 2-(p-hydroxy-m-methoxybenzylidene)-3-oxobutanoate
(EOFT) were synthesized. The radical-scavenging abilities of these compounds were tested by trapping
2,20-azinobis(3-ethylbenzothiazoline-6-sulfonate) cationic radical (ABTSþꢀ), 2,20-diphenyl-1-picrylhydrazyl
(DPPH), and galvinoxyl radicals. The reductive capacities were screened by quenching singlet oxygen and by
inhibiting the oxidation of linoleic acid. They were also employed to inhibit the oxidation of DNA mediated by
hydroxyl radical and 2,20-azobis(2-amidinopropane hydrochloride) (AAPH). In addition, they were applied to
protect erythrocytes against AAPH- and hemin-induced hemolysis. The obtained results revealed that the
antioxidantcapacityofhalf-curcuminwasderivedfromthephenolic-OHandtheconjugatedlinkagebetween
phenolic and enolic-OH. The enolic-OH itself cannot trap radicals.
Received 7 August 2010
Received in revised form
20 January 2011
Accepted 25 January 2011
Available online 1 February 2011
Keywords:
Feruloylacetone
Half-curcumin
Antioxidant
Free radical
Erythrocyte
DNA
Ó 2011 Elsevier Masson SAS. All rights reserved.
1. Introduction
2. Chemistry
Curcumin (1,7-bis(p-hydroxy-m-methoxyphenyl)-1,6-heptdiene-
3,5-dione) is the major component in Curcuma longa. Curcumin and
its in vivo metabolites can prevent inflammatory factor, retard
nerval, cardiovascular, and pulmonary degenerations, and even
inhibit the occurrence of cancer [1e4]. Recently, much research
attention focuses on the synthesis of curcumin-related compounds
and the evaluation of their bioactivities [5]. The enolic tautomer
of curcumin was believed to mainly contribute to the bioactivities
and photophysical and photochemical properties [6,7]. The anti-
inflammatory activity of curcumin was enhanced if the structural
feature of diketone in curcumin was replaced by cycloketone [8].
Meanwhile, dehydrozingerone whose structure was shown in
Scheme 1 was regarded to be the model compound of half-curcumin
[9]. The ability of phenolic-OH in dehydrozingerone to trap radicals
was estimated in the experimental system of the autoxidation of
linoleic acid (LH) [10]. Dehydrozingerone is also usually applied in
the design of the antitumor drugs [11,12].
Some investigations revealed that hydrogen atom in the enolic-
OH other than the methylene of curcumin can be abstracted by
radicals in polar solvents [13,14]. Thus, two vanillylidene groups,
enoleketo tautomerism, and the conjugated system in the carbon
chain play an important role in the bioavailability of curcumin [15,16].
Furthermore, in order to clarify the relationship between the struc-
ture of half-curcumin and the antioxidant property, 6-(p-hydroxy-m-
methoxyphenyl)-5-hexene-2,4-dione (feruloylacetone, FT, structure
shown in Scheme 1) involves enoleketo tautomerism, the conjugated
system, and only one vanillylidene group together with its structural
analogs were prepared following Scheme 2. The phenolic-OH
in FT was etherified by benzyl group to afford 6-(p-benzyloxy-
m-methoxyphenyl)-5-hexene-2,4-dione (BMFT). The antioxidant
capacity of BMFT was only contributed from the enolic-OH. The
demethylation of FT produced 6-(m,p-dihydroxyphenyl)-5-hexene-
2,4-dione (DDFT) that contained two phenolic-OHs. The C]C in FT
was reduced to generate 6-(p-hydroxy-m-methoxyphenyl)hexane-
2,4-dione (DHFT). The conjugated linkage between phenolic and
enolic-OH was cut off, and the antioxidant capacity of DHFT was
individually contributed from phenolic and enolic-OH. Two C]O in
FT were reduced to form 6-(p-hydroxy-m-methoxyphenyl)-5-hex-
ene-2,4-diol (THFT). The antioxidant capacity of THFT was only
* Corresponding author. Tel.: þ86 431 88499174; fax: þ86 431 88499159.
E-mail address: zaiqun-liu@jlu.edu.cn (Z.-Q. Liu).
0223-5234/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2011.01.039

J.-Y. Feng, Z.-Q. Liu / European Journal of Medicinal Chemistry 46 (2011) 1198e1206
1199
Scheme 1. Structures of dehydrozingerone and tautomers of feruloylacetone.
derived from the phenolic-OH. Finally, ethyl 2-(p-hydroxy-m-
methoxybenzylidene)-3-oxobutanoate (EOFT) was prepared by
vanillin and ethyl acetoacetate in order to detect the influence of ester
group on the antioxidant capacity.
ReN]NeR, R ¼ eCMe2C(]NH)NH2)-induced oxidation of methyl
linoleate. In biological experimental systems, half-curcumin deriva-
tives were applied to inhibit AAPH, Cu2þ/glutathione (GSH), and
hydroxyl radical (ꢀOH)-induced oxidations of DNA. These compounds
were also used to inhibit AAPH- and hemin-induced hemolysis of
erythrocytes. Therefore, the contribution of phenolic and enolic-OH
to the antioxidant capacity of half-curcumin will be clarified by using
chemical and biological experimental systems.
3. Pharmacology
Curcumin and its derivatives are lipophilic antioxidants, but the
insolubility in water does not affect the bioavailability in vivo [2].
The antioxidant activities of curcumin derivatives were evaluated
in chemical and biological experimental systems [17], which were
still applied in this work. In the chemical experimental systems,
half-curcumin derivatives were applied to trap 2,20-azinobis(3-
ethylbenzothiazoline-6-sulfonate) cationic radical (ABTSþꢀ), 2,20-
diphenyl-1-picrylhydrazyl (DPPH) and galvinoxyl radicals, and to
quench singlet oxygen (1O2). These half-curcumin derivatives were
employed to retard the autoxidation of linoleic acid (LH), and
to inhibit 2,20-azobis(2-amidinopropane hydrochloride) (AAPH,
4. Results
4.1. Scavenging free radicals and quenching 1O2
The antioxidant abilities of half-curcumin derivatives can be
directly characterized by interacting with ABTSþꢀ, DPPH and galvi-
noxyl radicals. Mixing the solutions of half-curcumin derivatives
with these radicals decreases the absorbance of radical solutionwith
the reaction period increasing. The decay rates of the concentrations
m
n
n
n
n
Scheme 2. Synthetic routine for half-curcumin derivatives in this work.

1200
J.-Y. Feng, Z.-Q. Liu / European Journal of Medicinal Chemistry 46 (2011) 1198e1206
formation of LOOꢀ. FT and THFT can also retard the decay of the
absorbance of -carotene. In particular, DDFT is the most efficient
antioxidant to keep the absorbance of -carotene almost invariable.
However, the decays of -carotene in the presence of BMFT, EOFT
Table 1
Decay rates of the concentration of free radicals in the presence of half-curcumin
b
derivatives.a
b
Free radicals
d[radical]/dt (
m
M minꢂ1
)
b
and DHFT approach to that of the blank experiment, indicating that
the abilities of BMFT, EOFT and DHFT to protect LH are very low.
The decomposition of AAPH generates radical that combines
with oxygen to form a water-soluble peroxyl radical (ROOꢀ), and
ROOꢀ is a radical-initiator to oxidize LH. Thus, AAPH-induced
oxidation of LH is an in vitro experimental system usually employed
to mimic polyunsaturated fatty acid (PUFA) undergoing oxidation.
Measuring the formation of peroxide of LH (LOOH) [18] or the
depletion of LH [19] can follow the oxidative process of LH. Because
methyl ester of LH is readily evaporated in GC, methyl linoleate
instead of LH is employed if GC is used to follow the oxidation [20].
The panel B and C in Fig. 2 illustrate the decays of the concentration
of methyl linoleate in the presence of half-curcumin derivatives. In
the blank experiment, the concentration of methyl linoleate
decreases continuously, indicating that AAPH exhausts methyl
linoleate successively. With BMFT and EOFT added, the decreases of
the concentration of methyl linoleate are close to that of the blank
experiment, indicating that BMFT and EOFT cannot hinder the
oxidation of methyl linoleate. As panel C of Fig. 2 shows, the addi-
tions of DDFT, FT, DHFT and THFT retard the decay of the concen-
tration of methyl linoleate for a period, and then, the concentration
of methyl linoleate decreases as in the blank experiment. The inhi-
bition periods (tinh) generated by DDFT, FT, DHFT and THFT are
shown as the cross-point of the tangents of the inhibitive and the
oxidative periods, and listed in Table 2. DDFTand FTgenerate longer
tinh than DHFT and THFT, revealing that DDFT and FT possess higher
antioxidant abilities to inhibit AAPH-induced oxidation of methyl
linoleate than DHFT and THFT.
FT
6.04
22.29
0.13
DDFT
THFT
DHFT
EOFT
BMFT
ABTSþ
DPPH
7.56
41.21
0.08
6.26
8.57
0.13
3.85
16.95
0.06
1.78
7.54
0.02
e
e
e
ꢀ
Galvinoxyl
a
The final concentrations of half-curcumin derivatives were 50.0
m
M in the test of
trapping ABTSþ and DPPH radicals, and 25.0
radical.
mM in the test of trapping galvinoxyl
ꢀ
of these radicals can be calculated by the decay rates of the absor-
bance multiplying the extinction coefficient ( ) of these radicals, and
3
are listed inTable1. Itcanbe foundthattheadditionof half-curcumin
derivatives lead to fast decrease of DPPH radical, and slow decrease
of galvinoxyl radical. BMFT cannot react with these radicals, while
DDFT exhibits relative high activity to trap these radicals.
1O2 is formed in the mixture of histidine, NaClO and H2O2, and
can decrease the absorbance of 4-nitroso-N,N-dimethylaniline
(NDMA) at 440 nm (A440). The absorbance of NDMA was 1.013 at the
beginning of the reaction at 30 ꢁC, and was 0.458 after 40 min
since 1O2 generated in the reaction mixture decolorizes NDMA. In
the presence of 250 mM EOFTthe absorbance of NDMA just decreases
to 0.993 after 40 min, implying that EOFT is able to preserve NDMA
against 1O2-induced decolorization by quenching 1O2. Hence, the
percentage of 1O2 quenched by EOFT can be calculated as (0.993 ꢂ
0.458)/(1.013 ꢂ 0.458) ꢃ 100 ¼ 96.4. Fig. 1 illustrates the relation-
ships between the quenched percentages of 1O2 and the
concentrations of half-curcumin derivatives employed. It can be
found that THFT cannot quench 1O2, and the quenched percent-
ages of 1O2 increase with the concentrations of FT and DHFT. On
the other hand, a maximum percentage of quenched 1O2 is found
4.3. Effects of half-curcumin derivatives on Cu2þ/GSH-, OH- and
ꢀ
when 150
m
M BMFT is applied. EOFT and DDFT quench 1O2
concentration-dependently. The entire sequence of half-curcumin
AAPH-induced oxidation of DNA
derivatives to quench 1O2 is EOFT w DDFT > BMFT > FT w DHFT.
The intracellular GSH and Cu(II) oxidize DNA to form carbonyl
species [21], which can be measured at 535 nm after reacting with
thiobarbituric acid (TBA) [22]. So, carbonyl species deriving from the
oxidation of DNA are called TBA reactive substance (TBARS) as well,
and are indexed to characterize the oxidative extent of DNA. As an in
vitro experimental system to mimic DNA destroyed by Cu2þ/GSH,
the absorbance of TBARS in the blank experiment (Aref) is assigned
as 100% when DNA is mixed with Cu2þ/GSH for 2 h. Meanwhile, the
absorbance of TBARS in the presence of half-curcumin derivatives
(Adetect) is compared with Aref to express the effects of half-curcumin
derivatives on Cu2þ/GSH-induced oxidation of DNA. The left
columns in Fig. 3 show Adetect/Aref ꢃ 100 in the presence of half-
curcumin derivatives. In the presence of half-curcumin derivatives
all the percentage of TBARS are higher than that in the blank
experiment, indicating that the additions of half-curcumin deriva-
tives improve Cu2þ/GSH-induced oxidation of DNA, and generate
much more TBARS. The half-curcumin derivatives behave as
prooxidants in this case. Especially, the prooxidant ability of DDFT is
much higher than that of FT. The prooxidant abilities of other half-
curcumin derivatives are higher than FT, and lower than DDFT.
ꢀOH is conveniently prepared by mixing tetrachlorohydroquinone
(TCHQ) with H2O2 [23]. The attack of ꢀOH to ribosyl moiety destroys
4.2. Protecting linoleic acid and methyl linoleate
LH and
b-carotene can form water-soluble emulsion in the
presence of Triton X-100 or Tween. The oxygen dissolved in water
oxidizes LH to form peroxyl radical of LH (LOOꢀ) that can bleach
b
-carotene. As shown as the blank experiment in Fig. 2, the decay of
the absorbance at 460 nm indicates that LOOꢀ in the autoxidation of
LH bleaches -carotene continuously. The addition of 50.0 M half-
curcumin derivatives retards the consumption of -carotene, indi-
b
m
b
cating that half-curcumin derivatives protect LH by inhibiting the
ꢀ
DNA and form TBARS [24], thus, OH-induced oxidation of DNA is
capable of an in vitro experimental system to screen the antioxidant
ability. TBARS generated in the mixture of DNA, TCHQ and H2O2 is
measured after 30 min, and the absorbance of TBARS in the blank
experiment is assigned as 100%. The absorbance of TBARS in
the presence of half-curcumin derivatives is compared with that in
the blank experiment, and shown as the right column in Fig. 3. The
Fig. 1. Percentages of 1O2 quenched by various concentrations of half-curcumin
derivatives at 30 ꢁC for 40 min.

J.-Y. Feng, Z.-Q. Liu / European Journal of Medicinal Chemistry 46 (2011) 1198e1206
1201
Fig. 2. The decay of the absorbance of b-carotene (panel A), and the decay of the concentration of methyl linoleate (panel B and C) in the presence of half-curcumin derivatives.
percentages of TBARS formed in the presence of half-curcumin
derivatives are lower than that in the blank experiment, so, half-
curcumin derivatives are active to protect DNA. Especially, BMFT, FT,
and DHFT possess relative high efficacy to protect DNA, while the
abilities of DDFT, EOFT, and THFTare lower than that of BMFT, FT, and
DHFT.
endogenous antioxidants, leading to a lag time (tlag). Adding exoge-
nous antioxidants prolongs the tlag, and the difference of tlag in the
presence and absence of the additional antioxidants is designed
as inhibition period (tinh) representing the ability of the antioxidant
to protect erythrocytes [28]. The tinh generated by adding different
concentrations of half-curcumin derivatives are measured and
The peroxyl radical (ROOꢀ) deriving from AAPH abstracts an H
atom from the C-40 atom of DNA, leading to strand breaks [25], and
generating TBARS eventually [26]. The formation of TBARS
increases with the reaction period in the blank experiment [17,19],
and the addition of half-curcumin derivatives inhibits the genera-
tion of TBARS for a period. In this work the inhibition period (tinh) is
measured when different concentrations of half-curcumin deriva-
tives are added. The relationships between tinh and the concen-
trations of half-curcumin derivatives are illustrated in panel A of
Fig. 4. All the half-curcumin derivatives are able to protect DNA
against AAPH-induced oxidation since tinh increases with the
concentration of half-curcumin derivatives. Moreover, the linear
relationships in the panel A of Fig. 4 are quantitatively expressed by
the equations of tinh w [half-curcumin derivatives], and listed in
Table 3.
The linear regressive analysis results in the equations, in which
the constants balance the equation, and the coefficients reveal the
sensitivity of tinh to the variety of the concentration of half-curcu-
min derivatives. Higher coefficient in the equation implies that tinh
increases more remarkably with the concentration of the half-
curcumin, and the half-curcumin possesses higher antioxidant
capacity. It can be found in Table 3 that DDFT has the highest
activity to protect DNA, while BMFT cannot behave as an antioxi-
dant in this case. Hence, the antioxidant capacity of half-curcumin
derivatives is in the sequence of DDFT > DHFT > FT > THFT > EOFT.
illustrated in panel B of Fig. 4. The addition of 18 and 36
generates ꢂ80 and ꢂ56 min of tinh, respectively, and till the
concentration of BMFTexceeds 54 M, BMFTgenerates a positive tinh
This fact implies that low concentration of BMFT plays a prooxidant
role to improve the hemolysis. Furthermore, tinh increase with the
concentrations of other half-curcumin derivatives, and the equations
of tinh w [half-curcumin derivatives] are listed in Table 3. The
sequenceofhalf-curcuminderivativestoprotecterythrocytesagainst
AAPH-induced hemolysis is similar to that in protecting DNA, viz.,
DDFT > DHFT > BMFT > FT > THFT > EOFT.
Hemin can intercalate into the membrane lipid of erythrocyte,
accelerate the potassium leakage, and dissociate skeletal proteins,
leading to hemolysis eventually [29]. Thus, if an antioxidant can
protect erythrocyte against hemin-induced hemolysis, it can be
regarded as a membrane-stabilizer. The absorbance of hemoglobin
(535 nm) is measured when different concentrations of half-
curcumin derivatives are applied in hemin-induced hemolysis, and
outlined in Fig. 5. It can be found that only DDFT protects eryth-
rocytes against hemin-induced hemolysis with the concentration
increasing. FT and THFT cannot protect erythrocytes even high
concentration applied. EOFT and DHFT improve the hemolysis with
their concentrations increasing, and BMFT promotes the hemolysis
significantly when high concentration is employed.
mM BMFT
m
.
4.4. Effects of half-curcumin derivatives on AAPH- and
hemin-induced hemolysis of erythrocytes
AAPH-induced hemolysis is a convenient in vitro experimental
system to mimic erythrocytes undergoing oxidative stress [27].
Hemolysis does not occur immediately when erythrocytes encounter
AAPH because of the endogenous antioxidants in the erythrocyte
membrane, and takes place after the complete exhaustion of
Table 2
The inhibition period (tinh) generated by half-curcumin derivatives in protecting
methyl linoleate against AAPH-induced oxidation.
Fig. 3. The percentage of TBARS formed in the mixture of 2.0 mg/mL DNA, 5.0 mM
Cu2þ, 3.0 mM GSH and 0.2 mM half-curcumin derivatives after incubation for 2 h (left
column), and in the mixture of 2.0 mg/mL DNA, 4.0 mM TCHQ, 8.0 mM, H2O2, and
0.2 mM half-curcumin derivatives after incubation for 30 min (right column).
Half-curcumin derivatives
DDFT
224
FT
DHFT
160
THFT
117
EOFT
BMFT
tinh (min)
221
e
e

1202
J.-Y. Feng, Z.-Q. Liu / European Journal of Medicinal Chemistry 46 (2011) 1198e1206
A
B
Fig. 4. The relationships between tinh and the concentrations of half-curcumin
derivatives in AAPH-induced oxidation of DNA (panel A) and hemolysis of erythrocytes
(panel B). The concentration of DNA and AAPH are 2.0 mg/mL and 40 mM, respectively,
the concentrations of erythrocytes and AAPH are 3.0% (v/v) and 20 mM, respectively.
Fig. 5. The correlations of the absorbance at 535 nm and the concentrations of
5. Discussion
half-curcumin derivatives employed in 20.0
erythrocytes after incubation for 30 min.
mM hemin-induced hemolysis of 1.0%
The multiply methods applied to evaluate the ability of an
antioxidant avoid shortcomings from one-dimensional character-
ization [30,31]. Comparing the antioxidant effectiveness of struc-
tural analogs of an antioxidant gives more information on the
structureeactivity relationship. The treatment of the experimental
data obtained from biological samples by chemical kinetics results
in a quantitative expression of the antioxidant capacities in bio-
logical experimental systems [32]. FT usually acts as the structural
feature in the synthesis of curcuminoid molecules [33], and pre-
sented here is a comparison of the antioxidant property of FT with
its analogs, aiming to clarify the contribution of structural feature in
half-curcumin to the antioxidant effectiveness.
enolic-OH in DHFT, thus, the radical-scavenging activity of DHFT
can be regarded as the individual contribution from the phenolic
and enolic-OH. The decrease rates of DPPH and ABTSþ in the
ꢀ
presence of DHFT are lower than that in the presence of FT. So, the
conjugated system between phenolic and enolic-OH is of impor-
tance to the radical-scavenging property. Together with the fast
rates of DDFT to decrease the concentrations of DPPH and ABTSþꢀ, it
can be concluded that phenolic-OH linked with enolic-OH by the
conjugated system plays the main role in scavenging or reducing
radicals. The lowest rates of EOFT to react with these radicals reveal
that the enolic-OH from ester group attenuates the radical-scav-
enging ability of half-curcumin.
The interactions with DPPH and galvinoxyl radicals reveal the
ability of an antioxidant to donate hydrogen atom to N- and O-
centered radicals [34,35], and the interaction with ABTSþ reveals
ꢀ
The above rules are also valid in protecting methyl linoleate
against AAPH-induced oxidation. BMFT almost cannot protect
methyl linoleate, thus, enolic-OH is not active in this case. More-
over, no activity of EOFT demonstrates that enolic-OH from ester
group cannot prohibit AAPH-induced oxidation of methyl linoleate.
Other half-curcumin derivatives protect methyl linoleate against
AAPH-induced oxidation and generate tinh, in which tinh of DDFT
and FT are much larger than THFT and DHFT. The lowest tinh of
THFT (117 min) indicates that phenolic-OH itself cannot enhance
the antioxidant ability remarkably. Higher tinh of DHFT (160 min)
implies that individual phenolic and enolic-OH devotes to the
antioxidant capacity of half-curcumin significantly. Higher tinh of FT
reconfirms the importance of the conjugated linkage between
phenolic and enolic-OH to reinforce the antioxidant efficacy of half-
curcumin. More phenolic-OHs do not enhance the antioxidant
efficacy extraordinarily since tinh of DDFT (224 min) is almost the
same as that of FT (221 min).
the ability of the antioxidant to reduce radicals [36]. As shown in
Table 1, half-curcumin derivatives are active to decrease the
concentrations of DPPH more than galvinoxyl radical, demon-
strating that the hydrogen atom in half-curcumin derivatives are
more readily to be abstracted by N-centered radical than by O-
centered radical. However, BMFT cannot react with these radicals,
indicating that radicals cannot abstract hydrogen atom from enolic-
OH, and cannot be reduced by enolic-OH. Two C]O bonds are
reduced to form alcoholic-OH in THFT, the radical-scavenging
activity is just resulted from phenolic-OH. The decrease rate of
ABTSþꢀ in the presence of THFT is similar to that in the presence of
FT, whereas the rate of THFT to trap DPPH is much lower than that
of FT. Hence, phenolic-OH interacts with radical mainly by reducing
radicals other than by donating hydrogen atom to radicals. A CeC
bond cuts off the conjugated linkage between phenolic and
Half-curcumin derivatives exhibit similar activity in
bleaching test. BMFT, EOFT and DHFT exhibit low activities in
-carotene-bleaching test, revealing that enolic-OH either itself or
tautomerized from an -keto ester, or unconjugated with phenolic-
b-carotene-
Table 3
The equations of tinh w [half-curcumin derivatives], and n of half-curcumin deriv-
atives in protecting DNA and erythrocytes.
b
b
Half-curcumin Protect DNA
derivatives
Protect erythrocytes
OH does not make half-curcumin an efficient antioxidant to protect
LH against the autoxidation. After C]O is reduced to form alco-
holic-OH, the activity of THFT is similar to that of FT, demonstrating
that enolic-OH protects LH if only it connects with phenolic-OH by
a conjugated system. Furthermore, because DDFT exhibits the most
active to protect LH, the antioxidant effectiveness of half-curcumin
increases with more phenolic-OHs involved.
tinh (min) ¼ (n/Ri)
n
tinh (min) ¼ (n/Ri)
n
[Half-curcumin derivative
M)] þ constant
[Half-curcumin derivative
(
m
(
m
M)] þ constant
DDFT
DHFT
FT
THFT
EOFT
BMFT
curcumin
tinh ¼ 2.57 [DDFT] þ 30.4 8.6 tinh ¼ 3.92 [DDFT] þ 16.2 6.6
tinh ¼ 1.91 [DHFT] þ 45.2 6.4 tinh ¼ 3.48 [DHFT] þ 51.0 5.9
tinh ¼ 1.42 [FT] þ 186.3
4.8 tinh ¼ 2.38 [FT] þ 45.6
4.0
tinh ¼ 1.11 [THFT] ꢂ 40.0 3.7 tinh ¼ 2.05 [THFT] þ 19.2 3.4
tinh ¼ 0.82 [EOFT] þ 94.1 2.8 tinh ¼ 1.83 [EOFT] ꢂ 6.0
3.1
The interaction with 1O2 reveals the reductive abilities of half-
curcumin derivatives. Low activity of THFT to reduce 1O2 implies
that the ability of phenolic-OH itself to reduce 1O2 is very weak. FT
and DHFT show similar activity to reduce 1O2, and BMFT has high
e
e
tinh ¼ 3.3 [BMFT] ꢂ 155.8 5.5
a
tinh ¼ 2.43
8.2 tinh ¼ 3.45
5.8
[curcumin] þ 37.9
[curcumin] þ 18.7
a
The data of curcumin were cited from Ref. [17].

J.-Y. Feng, Z.-Q. Liu / European Journal of Medicinal Chemistry 46 (2011) 1198e1206
1203
activity to reduce 1O2, demonstrating that enolic-OH plays a main
role in reducing 1O2. Finally, DDFT and EOFT exhibit concentration-
dependent abilities to reduce 1O2, indicating that two phenolic-OHs
and ester-related enolic-OH are much active to reduce 1O2.
Fig. 3 reveals that the additions of half-curcumin derivatives
result in much more TBARS generated in Cu2þ/GSH-induced
oxidation of DNA, indicating that half-curcumin derivatives play
prooxidant role in this case. DDFT possesses the highest prooxidant
efficacy whereas FT has the lowest prooxidant efficacy, so, more
phenolic-OHs promote the prooxidant effectiveness. Meanwhile,
the prooxidant behavior of BMFT reveals that enolic-OH improves
Cu2þ/GSH-induced oxidation of DNA. Hydroxyl groups reduce Cu
(II) to form Cu(I), and Cu(I) can form a complex with DNA, DNAeCu
(I)OOH. Meanwhile, Cu(I) catalyzes O2 to form Oꢂ2 and H2O2 [37].
H2O2 improves the cleavage of DNAeCu(I)OOH, and form TBARS
consequently [38]. It was reported that phenolic-OH in hydrox-
ycinnamic acids accelerated the cleavage of DNA by reducing Cu(II)
to form Cu(I) [39]. DDFT possesses the highest prooxidant activity
because it contains two phenolic-OHs. Hence, the prooxidant
effectiveness of half-curcumin derivatives may be due to Cu(I)
resulting from reducing Cu(II) by phenolic-OH. The prooxidant
behavior of BMFT reveals that enolic-OH can also accelerate the
degradation of DNA.
haemolyze erythrocytes [43]. The n values of half-curcumin
derivatives are the products of the coefficients in the equation of
tinh w [half-curcumin derivatives] and Ri ¼ Rg ¼ 1.68
m
M minꢂ1, and
are listed in Table 3 as well. The n of half-curcumin derivatives in
protectingerythrocytes isDDFT> DHFT> BMFT> FT> THFT>EOFT,
which is similar to that in protecting DNA, and is largely close to the
orderof thesecompoundstoreact withDPPH. Asshownin Scheme1,
the tautomeric structure leads to the formation of a large conjuga-
tion system from benzene ring to C]O via two C]C bonds. The
common characteristics of these half-curcumins are planar mole-
cules. Thus, their antioxidant effects are mainly derived from the
different substituents rather than from the molecular configuration.
Moreover, it is necessary to interpret the antioxidant activities of
half-curcumin derivatives by integrating the results from trapping
radical, protecting DNA and erythrocytes, and even from curcumin
derivatives in our previous report [17].
BMFT and curcumin with phenolic-OH protected by benzyl
group are not active to trap radical and to protect DNA and eryth-
rocytes, indicating that enolic-OH itself does not have the antioxi-
dant activity. As shown in Scheme 1, the enolic-OH is the
tautomeric structure of CH2 in diketone, thus, H atom abstracted by
radicals from enolic-OH is equivalent to the H atom from CH2. The H
atom in CH2 is very difficult to be abstracted by radicals because the
bond dissociation energy of CeH is much higher than that of OeH
[13]. Hence, the enolic-OH cannot trap radicals directly. The
phenolic-OH is the group in THFT to contribute the antioxidant
capacity, but low n of THFT reveals that phenolic-OH itself does not
contribute largely to the antioxidant effectiveness. The low rate of
THFT to trap DPPH also demonstrates that a single phenolic-OH
cannot enhance the ability of half-curcumin to trap radicals. The
conjugated linkage of phenolic and enolic-OH in DHFT is cut off by
CeC. The n values of DHFTare much higher than that of FT, implying
that phenolic-OH connected with enolic-OH by a conjugated
system as in FT does not benefit for the antioxidant capacity. On the
other hand, more phenolic-OHs in DDFT increase the antioxidant
capacity significantly, indicating that two adjacent phenolic-OHs
enhance the radical-scavenging ability remarkably. This is in
agreement with the findings in the research on the antioxidant
effectiveness of curcumin derivatives [17]. Furthermore, the low
activity of EOFT demonstrates that ester-related enolic-OH
decreases the antioxidant property of half-curcumin. The acidity of
Fig. 3 outlines that half-curcumin derivatives are able to protect
DNA against ꢀOH-induced oxidation. In particular, BMFT has the
highesteffectivein this case, demonstrating thatenolic-OH playsthe
main role in inhibiting ꢀOH-induced oxidation of DNA. But the ability
of EOFT is lower than that of FT, indicating that enolic-OH from an
ester group does not protect DNAwith high ability. Furthermore, the
lowest activity of DDFT implies that more phenolic-OHs do not
enhance the antioxidant effectiveness of half-curcumin to prohibit
ꢀOH-induced oxidation of DNA.
From a chemical kinetic viewpoint, tinh generated by an antiox-
idant (AH) correlateslinearly with the concentration as expressed by
equation (1) [40].
tinh ¼ ðn=RiÞ½AHꢄ
(1)
In equation (1), n is designated as stoichiometric factor to express
the ability of the antioxidant to terminate the radical-chain prop-
agation, and is not related to the concentration of the antioxidant
employed. Ri is the initiation rate of the radical-induced reaction.
This equation has been applied to calculate n of curcumin deriva-
tives in protecting linoleic acid [41], DNA and erythrocytes against
AAPH-induced oxidation [17]. So, n of half-curcumin derivatives is
calculated by equation (1), in which the value of n is the product of
Ri and the coefficient in equation (1) on the basis of a known Ri.
However, Ri should be measured by using trolox (6-hydroxyl-
2,5,7,8-tetramethylchroman-2-carboxylic acid) as the reference
antioxidant, whose n is assigned as 2 [42]. Therefore, tinh are
measured with different concentrations of trolox employed, and
then, Ri is obtained from the equation of tinh w [trolox]. The addi-
tion of trolox to AAPH-induced oxidation of DNA cannot generate
tinh (data not shown), thus, it is impossible to obtain Ri by using
trolox as reference antioxidant. Because both AAPH and DNA are
dissolved in water phase, and radicals generated from the decom-
position of AAPH can attack DNA at the same phase, it is safely to
methylene in
tone (pKa ¼ 9) [44], meaning that the amount of enolic-OH deriving
from the tautomerization of methylene in -keto ester is lower than
that in -diketone. From the comparison of n values in Table 3, it
b-keto ester (pKa ¼ 11) is lower than that in
b-dike-
b
b
can be concluded that the antioxidant effects of half-curcumin are
generally lower than that of curcumin except two phenolic-OH
contained as in DDFT.
Hemin-induced hemolysis is due to the collapse in the micro-
structure of erythrocyte membrane. The improvement of BMFT to
hemolysis indicates that enolic-OH results in the collapse of eryth-
rocyte membrane concentration-dependently. THFT and FT exhibit
weak protective effects on erythrocytes with high concentration
employed. Contrarily, DHFT and EOFT exhibit weak promotive
effects on erythrocytes with high concentration employed. It is
worthy to note that more phenolic-OHs enhance the protective
effect of DDFT. Thus, half-curcumin with more phenolic-OHs is
potential erythrocyte membrane-stabilizer.
assume that Ri is equal to the radical generation rate (Rg
¼
(1.4 ꢅ 0.2) ꢃ 10ꢂ6 [AAPH] sꢂ1 [42]) in AAPH-induced oxidation of
DNA, viz., Ri ¼ Rg ¼ 1.4 ꢃ 10ꢂ6 ꢃ 40 mM sꢂ1 ¼ 3.36
m
M minꢂ1 with
6. Conclusion
40 mM AAPH employed [17,19]. Table 3 lists the n values of half-
curcumin derivatives accordingly. The n of half-curcumin deriva-
tives in protecting DNA is DDFT > DHFT > FT > THFT > EOFT.
We have demonstrated that Ri ¼ Rg in AAPH-induced hemolysis,
Taking the results from curcumin derivative into consideration,
some conclusions on the antioxidant capacity of curcumin-related
compounds can be made. The antioxidant ability is composed of the
reductive and radical-scavenging properties resulting from phenolic
Ri ¼ Rg ¼ 1.68
m
M minꢂ1 when 20 mM AAPH is employed to

1204
J.-Y. Feng, Z.-Q. Liu / European Journal of Medicinal Chemistry 46 (2011) 1198e1206
and enolic-OH. The enolic-OH itself does not possess any reductive
activity and cannot trap radicals, while the phenolic-OH itself does
not have marked antioxidant activity either. The phenolic and
enolic-OH contained in the same molecule enhances the antioxidant
efficacy significantly when they are not connected by a conjugated
system, so, the conjugated system does not benefit for the antioxi-
dant effectiveness of half-curcumin. Meanwhile, enolic-OH tauto-
7.3. Synthesis of DHFT
Feruloylacetone (1 mmol, 0.234 g) dissolved in 30 mL of ethyl
acetate was hydrogenated over 20 mg of 10% Pd/C for 2 h. After the
catalyst was filtered, and the solvent was evaporated under vacuum
pressure, the residue was purified via column chromatography
(silica gel, petroleum ether/ethanol ¼ 50:1), 0.210 g of DHFT (pale
yellow power) was obtained, yield 89%, m.p. 40e41 ꢁC 1H NMR
merized from b-keto ester cannot increase the antioxidant ability of
half-curcumin. On the other hand, more phenolic-OHs increase the
antioxidant capacity remarkably. The aforementioned information
will be useful for the designation of antioxidants involving half-
curcumin as structural feature.
(300 MHz, CDCl3)
d
: 2.03 (s, 3H, CH3COe), 2.55 (t, J ¼ 7.8 Hz, 2H,
CH2CH2e), 2.86 (t, J ¼ 7.8 Hz, 2H, CH2CH2e), 3.86 (s, CH3Oe), 5.46
(s, 1H, ]CHeC), 5.49 (s, 1H, HOe in phenyl), 6.64 (d, J ¼ 8.1 Hz, 1H,
CH]CHe in phenyl), 6.68 (s, 1H, eCH]C in phenyl), 6.80
(d, J ¼ 8.1 Hz, 1H, eCH]CH in phenyl); 13C NMR (75 MHz, CDCl3)
d:
7. Experimental section
193.2, 191.1, 146.3, 143.9, 132.6, 120.7, 114.3, 110.9, 100.1, 55.8, 40.4,
31.2, 24.8.
7.1. Materials and instrumentation
7.4. Synthesis of THFT
AAPH, diammonium of 2,20-azinobis(3-ethylbenzothiazoline-
6-sulfonate) (ABTS), DPPH and galvinoxyl radical, 4-nitroso-N,N-
dimethylaniline (NDMA), LH, methyl linoleate, GSH, hemin and
naked DNA sodium salt were purchased from ACROS ORGANICS,
Belgium. Other reagents were of analytical grade, and purchased
from Beijing Chemical Reagent Co., China. Human erythrocytes
were provided by the First Hospital of Jilin University, Changchun,
China. The structures of half-curcumin derivatives were identified
by 1H and 13C NMR (Varian Mercury 300 NMR spectrometer, USA).
Feruloylacetone (1 mmol, 0.234 g) was dissolved in 20 mL of
methanol, and cooled to 0 ꢁC. NaBH4 (4 mmol, 0.151 g) was added
slowly. The mixture was stirred at 0 ꢁC for 0.5 h, and acidified by
0.1 M HCl to pH ¼ 6. After methanol was evaporated under vacuum
pressure, the residue was extracted by ethyl acetate (3 ꢃ 20 mL).
The ethyl acetate layer was washed by saturated saline, and dried
over Na2SO4. After the solvent was evaporated under vacuum
pressure, the residue was recrystallized by petroleum ether/
ethanol (10:1), and 0.22 g of THFT (white power) was obtained,
7.2. Synthesis of FT, BMFT and DDFT
yield 92%. 1H NMR (300 MHz, DMSO-d6);
d
: 1.08 (d, J ¼ 6 Hz, 3H,
CH3e), 1.41e1.66 (m, 2H, eCH2CH3), 3.79 (s, 3H, CH3OC6H3e), 4.24
(m, 1H, eCH(OH)eCH]), 4.45 (m, 1H, eCH(OH)eCH3), 4.77 (s, 1H,
HOeCHeCH]), 4.86 (s, 1H, HOeCHeCH3), 6.10 (m, 1H, CH]CHe),
6.36 (d, J ¼ 15.6 Hz, 1H, eCH]CH), 6.70 (d, J ¼ 8.1 Hz, 1H, eCH]CH
in phenyl), 6.79 (d, J ¼ 8.1 Hz, 1H, CH]CHe in phenyl), 7.00 (s, 1H,
eCH]C in phenyl), 9.02 (s, 1H, HOeC6H3e); 13C NMR (75 MHz,
Benzaldehyde (0.01 mol vanillin, 3-methoxy-4-benzylox-
ybenzaldehyde, or protocatechualdehyde), 1.0 g of boric anhydride
(0.014 mol) and 2.0 mL of tributyl borate were dissolved in 2 mL of
acetylacetone (0.02 mol). The mixture was stirred at 90 ꢁC for
30 min, and cooled to 70 ꢁC. Then, 0.7 mL of butylamine was added
dropwisely during 30 min, and stirred at 100 ꢁC for 1.5 h. The
mixture was acidified by 20 mL of 0.4 M HCl at 50 ꢁC, stirred
at room temperature for 1 h, and extracted by ethyl acetate
(3 ꢃ 20 mL). The layer of ethyl acetate was washed by distilled
water, and dried over Na2SO4. After the solvent was evaporated
under vacuum pressure, the residue was purified via column
chromatography (silica gel, petroleum ether/ethyl acetate ¼ 4:1),
and feruloylacetone was a pale yellow power, yield 41%, m.p.
DMSO-d6) d: 147.7, 146.1, 131.7, 128.4, 127.8, 119.4, 115.4, 109.7, 68.1,
62.9, 55.5, 47.0, 23.9.
7.5. Synthesis of EOFT
Vanillin (1.52 g, 0.01 mol) was dissolved in 2 mL of ethyl ace-
toacetate (0.0157 mol). Then, 0.7 mL of butylamine was added
dropwisely at 60 ꢁC during 30 min. The mixture was stirred at
100 ꢁC for 1.5 h, and 15 mL of 1 M HCl was added at 30 ꢁC and stirred
for 1 h at room temperature. The mixture was extracted by ethyl
acetate (3 ꢃ 20 mL). The layer of ethyl acetate was washed by
distilled water, and dried over Na2SO4. After the solvent was
evaporated under vacuum pressure, the residue was purified via
column chromatography (silica gel, petroleum ether/ethyl
acetate ¼ 4:1), and 1.06 g of EOFT (a pale yellow power) was
142e143 ꢁC 1H NMR (300 MHz, DMSO-d6)
d: 2.12 (s, 3H, CH3COe),
3.82 (s, 3H, CH3OC6H3e), 5.85 (s, 1H, ]CHeC), 6.62 (d, J ¼ 15.9 Hz,
1H, CH]CHe), 6.79 (d, J ¼ 8.1 Hz, 1H, CH]CHe in phenyl), 7.10
(d, J ¼ 8.1 Hz, 1H, eCH]CH in phenyl), 7.29 (s, 1H, eCH]C in
phenyl), 7.46 (d, J ¼ 15.9 Hz,1H, eCH]CH), 9.64 (s,1H, HOeC6H3e);
13C NMR (75 MHz, DMSO-d6)
d: 178.3, 149.2, 148.0, 140.2, 126.3,
122.9, 120.0, 115.6, 111.2, 100.4, 55.6, 30.7.
BMFT was a yellow crystal, yield 46%, m.p. 98e99 ꢁC 1H NMR
obtained, yield 40%, m.p. 102e103 ꢁC. 1H NMR (300 MHz, CDCl3)
d:
(300 MHz, DMSO-d6)
d
: 2.13 (s, 3H, CH3COe), 3.83 (s, 3H,
1.32 (t, J ¼ 6.9 Hz, 3H, CH3CH2e), 2.41 (s, 3H, CH3COe), 3.90 (s, 3H,
CH3Oe), 4.34 (m, 2H, eCH2CH3), 5.94 (s, 1H, HOeC6H3e), 6.91 (d,
J ¼ 8.4 Hz, 1H, CH]CHe in phenyl), 7.03 (s, 1H, eCH]C in phenyl),
7.05 (d, J ¼ 8.4 Hz, 1H, eCH]CH in phenyl), 7.48 (s, 1H, eCH]C);
CH3OC6H3e), 5.14 (s, 2H, eCH2C6H5), 5.87 (s, 1H, ]CHeC), 6.70 (d,
J ¼ 15.9 Hz, 1H, CH]CHe), 7.07 (d, J ¼ 8.1 Hz, 1H, CH]CHe in
phenyl), 7.20 (d, J ¼ 8.1 Hz, 1H, eCH]CH in phenyl), 7.35e7.54 (m,
7H, C6H5e, eCH]C in phenyl and eCH]CH); 13C NMR (75 MHz,
13C NMR (75 MHz, CDCl3)
d: 194.7, 164.7, 148.5, 148.1, 143.8, 132.1,
DMSO-d6)
d
: 197.1, 177.7, 149.7, 149.3, 139.7, 136.7, 128.4, 127.9,
125.0, 123.5, 114.8, 111.3, 61.7, 55.8, 26.4, 14.0.
127.87, 122.8, 122.5, 120.8, 113.2, 110.6, 100.7, 69.8, 55.6, 26.5.
DDFT was a red power, yield 26%, m.p. 216e217 ꢁC 1H NMR
7.6. Scavenging ABTSþꢀ, DPPH and galvinoxyl radical
(300 MHz, DMSO-d6) d: 2.07 (s, 3H, CH3COe), 6.06 (s, 1H, ]CHeC),
6.55 (d, J ¼ 15.6 Hz, 1H, CH]CHe), 6.76 (d, J ¼ 8.1 Hz, 1H, CH]CHe
in phenyl), 6.99 (d, J ¼ 8.1 Hz, 1H, eCH]CH in phenyl), 7.06 (s, 1H,
eCH]C in phenyl), 7.47 (d, J ¼ 15.6 Hz, 1H, eCH]CH), 9.19 (s, 1H,
HOeC6H3), 9.65 (s, 1H, HOeC6H3); 13C NMR (75 MHz, DMSO-d6)
The experiments of scavenging ABTSþꢀ, DPPH and galvinoxyl
radical were performed following our previous report [17].
DPPH radical (w0.1 mM) and galvinoxyl radical (w2
mM) were
dissolved in ethanol to make the absorbance (Absref) w 1.00
d: 183.1, 148.4, 145.7, 140.7, 126.3, 121.6, 120.6, 115.8, 114.7, 101.0,
at 517 nm (3DPPH ¼ 4.09 ꢃ 103 Mꢂ1 cmꢂ1 [45]) and 428 nm
30.7.
(
3galvinoxyl ¼ 1.4 ꢃ 105 Mꢂ1 cmꢂ1 [46]), respectively. Two milliliter of

J.-Y. Feng, Z.-Q. Liu / European Journal of Medicinal Chemistry 46 (2011) 1198e1206
1205
4.0 mM ABTS aqueous solution was oxidized by 1.41 mM K2S2O8 for
16 h, and then,100 mL of ethanol was added to make the absorbance
the oxidation of DNA. Three tubes were taken out at every 30 min
and cooled immediately, to which 1.0 mL of EDTA solution
(30.0 mM EDTA in PBS1 as the stock solution), 1.0 mL of thio-
barbituric acid (TBA) solution (dissolving 1.00 g TBA and 0.40 g
NaOH in 100 mL PBS1) and 1.0 mL of 3.0% trichloroacetic acid
aqueous solution were added. The tubes were heated in boiling
water for 30 min. After the test tubes were cooled to room
temperature, 1.5 mL of n-butanol was added and shaken vigorously
to extract TBA reactive substance (TBARS). The absorbance of
n-butanol layer was measured at 535 nm.
þꢀ
(Abs0) w 0.70 at 734 nm (3ABTS ¼ 1.6 ꢃ 104 Mꢂ1 cmꢂ1 [36]). The
ethanol solutions of half-curcumin derivatives were added to the
aforementioned radical solutions at room temperature. The final
concentrations of half-curcumin derivatives were 50.0
mM in the
test of trapping ABTSþꢀ and DPPH radicals, and 25.0
mM in the test of
trapping galvinoxyl radical. The absorbance of the radical solutions
was recorded, and the decay rates for DPPH, ABTSþꢀ and galvinoxyl
radical were calculated by multiplying the corresponding 3.
7.7. Quenching 1O2
7.10. Effects of half-curcumin derivatives on ꢀOH-induced oxidation
of DNA
1O2 was prepared following a literature [47]. Briefly, 10 mM
histidine, 20 mM sodium hypochlorite, 20 mM H2O2 and 50
m
M
ꢀOH was generated by mixing tetrachlorohydroquinone (TCHQ)
with H2O2 [23]. DNA, H2O2 (dissolved in PBS2: 8.1 mM Na2HPO4,
1.9 mM NaH2PO4, 10.0 mM EDTA) and TCHQ (dissolved in DMSO)
NDMA were dissolved in 45 mM sodium phosphate buffer
(pH ¼ 7.1) to generate 1O2. Various concentrations of half-curcumin
derivatives were added to above mixture to a total volume of
2.0 mL, and the mixture was incubated at 30 ꢁC for 40 min to
measure the absorbance at 440 nm. The percentage of 1O2
quenched by half-curcumin and its derivatives was calculated by
(Adetect ꢂ Aref)/(A0 ꢂ Aref) ꢃ 100, where A0 and Aref were the
absorbance before and after the incubation in the control experi-
ment, respectively, and Adetect was the absorbance after the incu-
bation in the presence of half-curcumin derivatives.
were mixed, and the final concentrations of DNA, H2O2 and TCHQ
were 2.0 mg/mL, 8.0 mM, and 4.0 mM, respectively. Then, DMSO
solutions of half-curcumin derivatives (0.2 mM as the final
concentration) were added, and the mixture was delivered into test
tubes with each one containing 2.0 mL. The test tubes were incu-
bated at 37 ꢁC for 30 min and cooled immediately, to which 1.0 mL
of TBA solution and 1.0 mL of 3.0% trichloroacetic acid aqueous
solution were added. The tubes were heated in boiling water for
30 min and cooled to room temperature. Then, 1.5 mL of n-butanol
was added to extract TBARS. The absorbances of TBARS in the blank
experiment and in the presence of half-curcumin derivatives were
assigned as A0 and Adetect, respectively. The effects of half-curcumin
7.8.
b-Carotene-bleaching test and protecting methyl linoleate
against AAPH-induced oxidation
ꢀ
An emulsion was formed by dissolving 5.0 mg of
b
-carotene,
derivatives on OH-induced oxidation of DNA were expressed by
40 mg of LH and 400 mg of Triton X-100 in 5.0 mL of CHCl3. After
CHCl3 was evaporated under vacuum pressure, 100 mL of oxygen-
saturated water was added and shaken under ultrasonic vibration
Adetect/A0 ꢃ 100.
7.11. Effects of half-curcumin derivatives on AAPH-induced
oxidation of DNA
to form homogeneous
[48]. The ethanol solutions of half-curcumin derivatives (0.1 mL)
were added to 1.9 mL of -carotene-LH emulsion with the final
concentration of half-curcumin derivatives at 50.0 M. The absor-
b
-carotene-LH emulsion (lmax ¼ 460 nm)
b
AAPH-induced oxidation of DNA was carried out following our
previous report [50]. DNA and AAPH were dissolved in PBS2 with
the final concentration at 2.0 mg/mL and 40 mM, respectively. After
various concentrations of half-curcumin derivatives (dissolved in
DMSO) were added, the mixture was delivered into test tubes with
each one containing 2.0 mL. All the tubes were incubated in a water
bath at 37 ꢁC to initiate the reaction. Three tubes were taken out at
every 2 h and cooled immediately. The following operation was the
m
bance of the mixture was determined and plotted vs time.
The abilities of half-curcumin derivatives to protect methyl
linoleate against AAPH-induced oxidation were tested by
measuring the decay of the concentration for methyl linoleate [19].
Methyl linoleate, methyl palmitate, AAPH, and half-curcumin
derivatives were dissolved in tert-butanol-H2O (1:1, v/v) in a test
tube with a final concentration at 15.0 mM, 10.0 mM, 80.0 mM, and
0.2 mM, respectively. The test tube was incubated at 37 ꢁC to
initiate the oxidation. Aliquots were taken out at every 120 min,
and the concentration of methyl linoleate was analyzed by GC
(HewlettePackard 1890 equipped with an SE-54 30 m ꢃ 0.25 mm
ꢀ
same as OH-induced oxidation of DNA except the heating period
was 15 min after TBA and trichloroacetic acid were added.
7.12. Effects of half-curcumin derivatives on AAPH- and
hemin-induced hemolysis of erythrocytes
capillary column, 0.25 mm film thickness, N2) with methyl palmitate
as the internal standard. The temperatures of chromatograph
chamber, injector, and hydrogen flame ionization detector were at
260 ꢁC, 280 ꢁC, and 300 ꢁC, respectively.
AAPH- and hemin-induced hemolysis of erythrocytes was per-
formed according to our previous report [17]. Human erythrocytes
were washed by phosphate-buffered saline (PBS3: 150 mM NaCl,
8.1 mM Na2HPO4, 1.9 mM NaH2PO4, 10 mM EDTA) to remove plasma
7.9. Effects of half-curcumin derivatives on Cu2þ/GSH-induced
oxidation of DNA
and centrifuged at 1700 ꢃ g for exact 10 min to obtain a compacted
volume of erythrocytes. Erythrocytes were suspended in PBS3, to
which AAPH and DMSO solutions of half-curcumin derivatives
were added. The final concentrations of erythrocytes and AAPH
were 3.0% (v/v) and 20 mM, respectively. The above mixture was
incubated at 37 ꢁC to initiate the hemolysis. Aliquots (1.5 mL) were
taken out at every 1 h and centrifuged at 1700ꢃg to obtain the
supernatant, whose absorbance was measured at 535 nm and
plotted vs incubation time to express the hemolysis process.
Hemin was dissolved in 5 mM NaOH to reach 2.0 mM before
usage. Erythrocytes were suspended in PBS4 (150 mM NaCl, 8.1 mM
Na2HPO4, 1.9 mM NaH2PO4, pH 7.4), to which DMSO solutions of
The oxidation of DNA mediated by Cu2þ and GSH was carried
out following a previous report [49] with a little modification.
Briefly, DNA, CuSO4, and GSH were dissolved in phosphate buffered
solution (PBS1: 6.1 mM Na2HPO4, 3.9 mM NaH2PO4) with the final
concentration for DNA, Cu2þ and GSH at 2.0 mg/mL, 5.0 mM, and
3.0 mM, respectively. Dimethyl sulfoxide (DMSO) solution of half-
curcumin derivatives was added with the final concentration at
0.2 mM. The mixture was delivered into test tubes with each one
containing 2.0 mL. The test tubes were incubated at 37 ꢁC to initiate

1206
J.-Y. Feng, Z.-Q. Liu / European Journal of Medicinal Chemistry 46 (2011) 1198e1206
[17] J.-Y. Feng, Z.-Q. Liu, J. Agric. Food Chem. 57 (2009) 11041e11046.
[18] B. Zhou, Q. Miao, L. Yang, Z.-L. Liu, Chem. Eur. J. 11 (2005) 680e691.
[19] G.-X. Li, Z.-Q. Liu, J. Agric. Food Chem. 57 (2009) 3943e3948.
[20] D. Spiteller, G. Spiteller, Angew. Chem. Int. Ed. 39 (2000) 583e585.
[21] W.A. Prütz, Biochem. J. 302 (1994) 373e382.
[22] Y. Li, M.A. Trush, Carcinogenesis 14 (1993) 1303e1311.
[23] B.-Z. Zhu, N. Kitrossky, M. Chevion, Biochem. Biophys. Res. Commun. 270
(2000) 942e946.
[24] M.F. Janicek, W.A. Haseltine, W.D. Henner, Nucleic Acids Res. 13 (1985)
9011e9029.
[25] P. Zhang, S.T. Omaye, Food Chem. Toxicol. 39 (2001) 239e246.
[26] R. Aeschbach, J. Loliger, B.C. Scott, A. Murcia, J. Butler, B. Halliwell, O.I. Aruoma,
Food Chem. Toxicol. 32 (1994) 31e36.
[27] Y. Sato, S. Kamo, T. Takahashi, Y. Suzuki, Biochemistry 34 (1995) 8940e8949.
[28] A. Banerjee, A. Kunwar, B. Mishra, K.I. Priyadarsini, Chem. Biol. Interact. 174
(2008) 134e139.
[29] F. Wang, T. Wang, J. Lai, M. Li, C. Zou, Biochem. Pharmacol. 71 (2006)
799e805.
[30] E.N. Frankel, A.S. Meyer, J. Sci. Food Agric. 80 (2000) 1925e1941.
[31] M. Laguerre, J. Lecomte, P. Villeneuve, Prog. Lipid Res. 46 (2007) 244e282.
[32] G.-X. Li, Z.-Q. Liu, Eur. J. Med. Chem. 44 (2009) 4841e4847.
[33] A. Sundaryono, A. Nourmamode, C. Gardrat, A. Fritsch, A. Castellan, J. Mol.
Struct. 649 (2003) 177e190.
half-curcumin derivatives were added. The mixture was incubated at
37 ꢁCfor30min, andthen, heminwasadded. The finalconcentrations
of erythrocytes and hemin were 1.0% (v/v) and 20.0 mM, respectively.
Afterincubationfor30min, themixturewascentrifugedat1700ꢃgto
obtain the supernatant. The absorbance of the supernatant was
measured at 535 nm and plotted vs the concentration of half-
curcumin derivatives. The same volume of DMSO (<1.0% to the total
volume) was contained in all the control experiment.
7.13. Statistical analysis
The data were the average value from at least three independent
measurements with the experimental error within 10%. The linear
relationships between inhibition period and concentration were
analyzed statistically by one-way ANOVA on Origin 7.5 professional
Software, and p < 0.001 indicated a significance difference.
References
[34] G. Litwinienko, K.U. Ingold, J. Org. Chem. 69 (2004) 5888e5896.
[35] H. Shi, E. Niki, Lipids 33 (1998) 365e370.
[36] R. Re, N. Pellegrini, A. Proteggente, A. Pannala, M. Yang, C. Rice-Evans, Free
Radical Biol. Med. 26 (1999) 1231e1237.
[37] A. Jain, N.K. Alvi, J.H. Parish, S.M. Hadi, Mutat. Res. 357 (1996) 83e88.
[38] S. Oikawa, K. Murakami, S. Kawanishi, Oncogene 22 (2003) 3530e3538.
[39] L.-F. Zheng, F. Dai, B. Zhou, L. Yang, Z.-L. Liu, Food Chem. Toxicol. 46 (2008)
149e156.
[40] L. Zennaro, M. Rossetto, P. Vanzani, V.D. Marco, M. Scarpa, L. Battistin, A. Rigo,
Arch. Biochem. Biophys. 462 (2007) 38e46.
[41] P. Somparn, C. Phisalaphong, S. Nakornchai, S. Unchern, N.P. Morales, Biol.
Pharm. Bull. 30 (2007) 74e78.
[42] V.W. Bowry, R. Stocker, J. Am. Chem. Soc. 115 (1993) 6029e6044.
[43] Y.-Z. Tang, Z.-Q. Liu, J. Agric. Food Chem. 56 (2008) 11025e11029.
[44] A. Streitwieser Jr., C.H. Heathcock, Introduction to Organic Chemistry, third ed.
Macmillan Inc., New York, 1985, 872.
[45] S.-K. Yim, S.-J. Yun, C.-H. Yun, J. Biochem. Mol. Biol. 37 (2004) 629e633.
[46] M.Y. Bobrov, A.A. Lizhin, E.L. Andrianova, N.M. Gretskaya, L.E. Frumkina,
L.G. Khaspekov, V.V. Bezuglov, Neurosci. Lett. 431 (2008) 6e11.
[47] P.D. Maldonado, I. Rivero-Cruz, R. Mata, J. Pedraza-Chaverri, J. Agric. Food
Chem. 53 (2005) 1996e2001.
[48] J. Yu, L. Wang, R.L. Walzem, E.G. Miller, L.M. Pike, B.S. Patil, J. Agric. Food Chem.
53 (2005) 2009e2014.
[49] C.J. Reed, K.T. Douglas, Biochem. J. 275 (1991) 601e608.
[50] F. Zhao, Z.-Q. Liu, J. Phys. Org. Chem. 22 (2009) 791e798.
[1] B.B. Aggarwal, K.B. Harikumar, Int. J. Biochem. Cell Biol. 41 (2009) 40e59.
[2] V. Basile, E. Ferrari, S. Lazzari, S. Belluti, F. Pignedoli, C. Imbriano, Biochem.
Pharmacol. 78 (2009) 1305e1315.
[3] B.B. Aggarwal, B. Sung, Trends Pharmacol. Sci. 30 (2009) 85e94.
[4] B. Ray, D.K. Lahiri, Curr. Opin. Pharmacol. 9 (2009) 434e444.
[5] L. Cen, B. Hutzen, S. Ball, S. DeAngelis, C.-L. Chen, J.R. Fuchs, C. Li, P.-K. Li, J. Lin,
BMC Cancer 9 (2009) 99. doi:10.1186/1471-2407-9-99.
[6] D. Sharma, P. Sethi, E. Hussain, R. Singh, Biogerontology 10 (2009) 489e502.
[7] K.I. Priyadarsini, J. Photochem. Photobiol. C: Photochem. Rev. 10 (2009) 81e95.
[8] R. Gautam, S.M. Jachak, Med. Res. Rev. 29 (2009) 767e820.
ꢀ
[9] V. Galasso, B. Kovac, A. Modelli, M.F. Ottaviani, F. Pichierri, J. Phys. Chem. A 112
(2008) 2331e2338.
[10] M. Musialik, G.J. Litwinienko, J. Therm. Anal. Calorim. 88 (2007) 781e785.
[11] J. Tatsuzaki, M. Taniguchi, K.F. Bastow, K. Nakagawa-Goto, S.L. Morris-Natschke,
H. Itokawa, K. Baba, K.-H. Lee, Bioorg. Med. Chem. 15 (2007) 6193e6199.
[12] V.K. Parihar, J. Dhawan, S. Kumar, S.N. Manjula, G. Subramanian,
M.K. Unnikrishnan, C.M. Rao, Chem. Biol. Interact 170 (2007) 49e58.
[13] Y.-M. Sun, H.-Y. Zhang, D.-Z. Chen, C.-B. Liu, Org. Lett. 4 (2002) 2909e2911.
[14] K.I. Priyadarsini, D.K. Maity, G.H. Naik, M.S. Kumar, M.K. Unnikrishnan,
J.G. Satav, H. Mohan, Free Radical Biol. Med. 35 (2003) 475e484.
[15] R.A. Sharma, A.J. Gescher, W.P. Steward, Eur. J. Cancer 41 (2005) 1955e1968.
[16] L.R.C. Barclay, M.R. Vinqvist, K. Mukai, H. Goto, Y. Hashimoto, A. Tokunaga,
H. Uno, Org. Lett. 2 (2000) 2841e2843.