A comparative study on the antioxidant properties of tetrahydrocurcuminoids and curcuminoids

Article abstract of DOI:10.1016/j.tet.2007.06.085

Several curcuminoids and tetrahydrocurcuminoids (THCs), bearing various hydroxyl and/or methoxy groups on their benzene rings, have been synthesized to study their antioxidant and hydrogen donating capacities using the DPPH method at 25 °C in methanol. The results show that the tetrahydrocurcuminoids are in general much more efficient than their curcuminoid analogs if they include a phenol group in meta- or para-position of the linking chain and a neighboring phenol or methoxy group. This efficiency gain of THCs by comparison to curcuminoids was attributed to the presence of benzylic hydrogens involved in the oxidation process of these compounds and not to the beta-diketone moiety in the chain.

Full text of DOI:10.1016/j.tet.2007.06.085

Tetrahedron 63 (2007) 9092–9099
A comparative study on the antioxidant properties of
tetrahydrocurcuminoids and curcuminoids
*
Elise Portes, Christian Gardrat and Alain Castellan
´
Universite Bordeaux 1, US2B, UMR 5103 CNRS-INRA-UBx1, Talence F-33405, France
Received 2 February 2007; revised 21 June 2007; accepted 26 June 2007
Available online 4 July 2007
Abstract—Several curcuminoids and tetrahydrocurcuminoids (THCs), bearing various hydroxyl and/or methoxy groups on their benzene
rings, have been synthesized to study their antioxidant and hydrogen donating capacities using the DPPH method at 25 ꢀC in methanol.
The results show that the tetrahydrocurcuminoids are in general much more efficient than their curcuminoid analogs if they include a phenol
group in meta- or para-position of the linking chain and a neighboring phenol or methoxy group. This efficiency gain of THCs by comparison
to curcuminoids was attributed to the presence of benzylic hydrogens involved in the oxidation process of these compounds and not to the
beta-diketone moiety in the chain.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
The free scavenging activity of curcumin and tetrahydrocur-
cuminoids can be due either to the phenolic hydroxyl group
or to the methylene group of the beta-diketone moiety. If the
phenolic group appears to play the major role in the antiox-
idant activity of curcuminoids,17 the question still remains
open for tetrahydrocurcuminoids. Many methods to evaluate
the antioxidative activity of specific compounds have been
described but the most widely documented one deals with
2,2-diphenyl-1-picrylhydrazyl (DPPH) radical.18–21 The an-
tioxidant is allowed to react with DPPH radical in methanol
solution at room temperature. The reduction of this radical is
followed by monitoring the decrease of its absorbance. In its
radical form, it absorbs at 515 nm, but after reaction the
absorption disappears. In the DPPH test, antioxidants are
typically characterized by their EC50 value, which is the
amount of antioxidant necessary to decrease the initial
DPPH concentration by 50%, after a plateau has been
reached for various antioxidant/DPPH molar ratios.18 The
antiradical power (ARP) equal to 1/EC50, is also used; the
larger the ARP, the more efficient the antioxidant. It is also
of interest to determine the time necessary to reach the
plateau (see above) at a molar ratio corresponding to EC50
(Time EC50); it depends on the reaction rate between the
antioxidant and the DPPH radical.19
Natural curcumin isolated from Curcuma longa L. rhizomes
(turmeric) contains curcumin, 1,7-bis(4-hydroxy-3-meth-
oxyphenyl)-1,6-heptadiene-3,5-dione, as the major component
but also demethoxycurcumin and bis-demethoxycurcumin
in smaller quantities.1 Structurally, curcumin belongs to
the diarylheptanoid series of naturally occurring 1,3-
diketones in which the carbonyl groups are directly linked
to olefinic carbons.2 Turmeric is one of the major spices
and food coloring in Asian cooking, notably Indian.3 Curcu-
min also shows remarkable pharmacological activity: it is
a very strong but safe anti-inflammatory agent;4 it displays
some inhibition of the HIV proteases5 and it seems to have
anti-cancer activity.6 Curcumin acts as a lipoxygenase sub-
strate7 and also as an inhibitor of cyclooxygenase enzymes.8
The main action of curcumin is due to its ability to inhibit the
formation of reactive oxygen species such as hydroxyl
radicals and superoxide anion.9–11
Tetrahydrocurcuminoids are obtained from curcuminoids by
hydrogenation and they are usually colorless. THCs are
useful in non-colored food and cosmetic applications that
currently employ synthetic antioxidants.12 Tetrahydrocurcu-
minoids appear to be the major active metabolites formed
when curcuminoids are intraperitoneally administered to
mice.13 Several independent studies reported the significant
antioxidant effects of the tetrahydrocurcuminoids.14–16
We have previously synthesized and studied curcuminoid
compounds as possible photoprotective molecules for ligno-
cellulosic materials.22 For this purpose, we have examined
the photochemical behavior of curcumin, dimethylcurcu-
min, and curcumin without substituents.23 The similar be-
havior of the three studied curcuminoids was indicative
of only a moderate role of phenol groups in the photode-
gradation process. Structural analysis of photodegradation
Keywords: Curcuminoids; Tetrahydrocurcuminoids; Antioxidant proper-
ties; DPPH.
* Corresponding author. Tel.: +33 5 4000 6280; fax: +33 5 4000 6439;
e-mail: a.castellan@us2b.u-bordeaux1.fr
0040–4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.06.085

E. Portes et al. / Tetrahedron 63 (2007) 9092–9099
9093
products showed among other products the formation of
a flavanone molecule. It represented a unique example of
photochemical conversion of a diarylheptanoid molecule
into a flavonoid, another very important class of natural
products.
was synthesized by the general procedure described in
Scheme 2 using feruloylacetone26 as 1,3-diketone, and p-
hydroxybenzaldehyde. The structures of the synthesized
curcuminoids were established mainly by 1H and 13C
NMR, high resolution mass spectrometry (HRLSIMS) and
UV–vis absorption spectroscopy.
In this paper, we report on the synthesis of 12 tetrahydrocur-
cuminoids, some of them have never been characterized;
their antioxidant properties were evaluated using the
DPPH method by comparison with their curcuminoid ana-
logs and with some monomers mimicking the different parts
of tetrahydrocurcuminoids. Their photochemical properties
have already been published.24 The information gained in
these studies would be very helpful to use tetrahydrocurcu-
minoids for the protection of cellulosic materials to improve
their durability for long term usages.
The tetrahydrocurcuminoids 1–12 were obtained by hydro-
genation of the curcuminoids on Pd/C. Their characteriza-
tion was achieved as for the curcuminoids. Compounds 1,
2, 8, and 10 were described in the literature.28–31 As for cur-
1
cumin,32 the H NMR spectra of the studied curcuminoids
and tetrahydrocurcuminoids show that in methanolic solu-
tions, they do exist in the enol form. Nevertheless, in deuter-
ated chloroform, acetone, and dimethyl sulfoxide two
signals can be observed, one corresponding to the enol
form (near 5.5 ppm) and the other one corresponding to
the diketone form (near 3.5 ppm), their relative proportions
depend on the experimental conditions.
2. Results and discussion
The name and formulae of the studied compounds are given
in Scheme 1.
2.2. Antioxidant properties according to DPPH
measurements
2.1. Syntheses
2.2.1. Results. The H-transfer reactions from the curcumi-
noids and tetrahydrocurcuminoids to DPPH were measured
by visible absorption spectroscopy at 515 nm in methanol at
25 ꢀC with antioxidant-DPPH ratios ranging from 0.05 to 2
according to Brand-Williams et al.18 This allows the deter-
mination of the different parameters (EC50, ARP, and Time
EC50). Some of the antioxidants studied have a very low
Curcuminoids were synthesized according to described pro-
cedures
25–27
by reacting, in presence of tributylborate and
n-butylamine, the acetylacetone boron complex (obtained
by action boric anhydride) with vanillin and other
substituted benzaldehydes, respectively. Compound 100
O
OH
O
OH
R3
R1
R6
R4
R3
R1
R6
R4
R2
R5
R5
R2
Curcuminoids
Tetrahydrocurcuminoids
Compounds
R1
R2
R3
R4
R5
R6
Compounds
1
2
3
4
5
6
7
8
9
10
H
H
H
H
H
H
H
H
H
H
1’
2’
OH
OMe
OH
H
OH
OMe
OH
H
OMe
OMe
OMe
OMe
OH
OMe
OMe
OMe
OMe
OH
H
3’
H
H
4’
OMe
OMe
OMe
H
OMe
OMe
OMe
H
H
5’
OMe
OMe
H
OMe
OMe
H
6’
7’
OH
OH
8’
OH
OH
H
H
OH
OH
H
9’ *
10’
OH
H
OH
OMe
H
11
12
H
OH
OH
OH
H
OH
H
OH
OH
OH
H
OH
11’
12’ *
* Not prepared
(H3C)3C
HO
O
OH
HO
HO
OCH3
C(CH3)3
OCH3
4-n-Propylguaiacol
Acetylacetone
BHT
Eugenol
Scheme 1. Formulae and abbreviated name of the studied compounds.

9094
E. Portes et al. / Tetrahedron 63 (2007) 9092–9099
+
+
Ar
'
Ar
O
O
O
O
O
O
O
O
O
O
i
ii
–
–
B
B
H2BO3
H2BO3
Ar
iii
O
OH
O
O
O
Curcuminoids
Ar'
Ar
Ar
Ar'
Ar'
Ar
iv
O
1'
OH
e
O
c
g
a
Ar
Tetrahydrocurcuminoids
Ar'
d
f
b
R3
R6
2'
6'
2
3
3'
4'
1
4
Ar' =
= Ar
R4
R1
5'
5
6
R5
R2
Scheme 2. Syntheses of the studied symmetrical curcuminoids and tetrahydrocurcuminoids given in Table 1. (i) B2O3; (ii) n-BuNH2, (n-BuO)3B; ArCHO;
(iii) HCl; and (iv) H2, Pd/C.
efficiency with respect to DPPH: so values EC10 (corre-
sponding to a molar ratio antioxidant/DPPH where 90% of
DPPH remains) have been determined. The stoichiometry
value (SV, twice the EC50 value), gives the theoretical con-
centration to reduce 100% of DPPH radical. The inverse of
SV gives the number of DPPH moles reduced by one mole
of antioxidant (NRD), which is indicative of the mechanism
involved.18–20 The data found for the curcuminoids, tetra-
hydrocurcuminoids, and some reference molecules (euge-
nol, BHT, 4-n-propylguaiacol) in the same experimental
conditions are given in Tables 1 and 2.
parameters (NRD and Time EC50) are relatively close to
the literature data.19
The reaction kinetics of BHT with DPPH also gives close
values for the first hydrogen atom abstraction rate constant
k1: 4.2 L molꢁ1 sꢁ1 versus 5 L molꢁ1 sꢁ1 20 Bondet et al.19
.
suggested that the antioxidant properties of BHT are due to
different pathways with a main contribution of benzylic rad-
icals. The rate constant k1 was also determined for the most
active compounds: 2, 20, 3, 30, 7, 70, 9, 10, 100, and 12, using
a procedure adapted from catechin.20 The relative rate con-
stants to BHT are given in Table 3. The curves shown in
Figure 1 for 2 are very similar to those given by Goupy et al.
for catechin.20
The ARP value found for eugenol, close to 2, is in accor-
dance with the literature.18 According to Bondet et al.,19
the mechanism involves first a reversible phenolic hydrogen
abstraction by a DPPH radical to give a non-sufficiently sta-
bilized antioxidant radical. The second reaction step con-
cerns radicals obtained after delocalization of the latter on
the aromatic ring (ortho- or para-position) to lead to
monoquinonoid species by dimerization. With BHT, the
2.2.2. Curcuminoids. The antiradical power efficiency
(ARP) and the number of reduced DPPH (NRD) of the phe-
nolic curcuminoids 20, 70, 100 are similar and close to the one
of 4-n-propylguaiacol (ARP¼4.8; NRD¼2.4). The Time
EC50 is related to the rate of reaction between the antioxidant
Table 1. Antiradical activity (EC50), antiradical power (ARP), plateau time, stoichiometric value and DPPH/antioxidant molar ratio of the most efficient studied
compounds
Compounds
EC50 ꢂ8%
0.22
0.11
0.60
ARP
Time EC50ꢂ5% (min)
300
150
480
320
240
90
480
15
300
360
15
Stoichiometric value (SV)
Number of reduced DPPH (NRD)
20
2
30
3
70
7
80
4.55
9.1
1.67
5.26
0.44
0.22
1.20
0.38
0.41
0.18
0.98
0.126
0.42
0.4
0.094
0.53, lit.18 0.54
0.44
2.3
4.5
0.83
2.6
2.4
5.5
1.0
7.9
0.19
0.205
0.091
0.49
0.063
0.21
0.20
0.047
0.265
0.22
4.88
10.99
2.04
15.9
4.76
5.0
9
100
10
2.4
2.5
10.6
12
Eugenol
BHT
4-n-Propylguaiacol
21.2
3.8, lit.18 3.7
4.5
4.8
300
1.9, lit.18 1.85
2.3, lit.19 20 ꢀC, 2.8
2.4
300, lit.19 20 ꢀC, 300
360
0.21
0.42

E. Portes et al. / Tetrahedron 63 (2007) 9092–9099
9095
Table 2. Antiradical activity, EC10 (amount of antioxidant to decrease
0.4
0.3
0.2
0.1
DPPH concentration by 10%), antiradical power ARP10 (¼1/EC10), and
Time10 (time to decrease the DPPH concentration by 10%) for the less
efficient studied compounds
Compounds
EC10ꢂ5%
ARP10¼1/EC10
Time10
2
0.02
0.45
0.48
0.60
0.62
0.40
0.30
0.30
0.40
0.40
0.35
50.0
2.2
2.1
1.7
1.6
2.5
3.3
3.3
2.5
2.5
2.9
15 min
4 days
3 days
4 days
2 days
4 days
2 days
4 days
2 days
2 days
1 day
10
1
40
4
50
5
60
6
8
11
0
0
0
0
50
50
100
100
150
150
and DPPH (Table 3). Among the three curcuminoids, 70,
which includes a syringyl phenol, is the most rapid. This is
in accordance with the well-known ability of syringyl phe-
nols to form very stable phenoxy radicals. The behavior of
isocurcumin 30 is peculiar when compared to the three other
curcuminoids: ARP and k1 are lower, Time EC50 longer and
the NRD less than one. The meta-position of the phenol
group related to the conjugated double bond of the heptadi-
enone moiety, which does not allow extended conjugation of
the radical formed by mesomeric effect is probably at the
origin of this low antioxidant activity. The absence of
methoxy or hydroxyl substituents in ortho-position of the
phenolic group explains the low ARP and longer Time EC50
of compound 80.
Time (s)
-1
-2
-3
-4
-5
Bond dissociation energies (BDE) of the O–H bond in phe-
nolic parts of the molecules will surely be an important fac-
tor in determining the efficacy of the antioxidant activity
since the weaker the OH bond, the faster will be the reaction
with free radicals.33 Recently, Wright et al. have proposed
a comprehensive set of optimized DBDE values derived
from calculations,33 which are in good agreement with the
experimental activity of phenolic antioxidants. Using these
values with symmetrical curcuminoids assimilated to vi-
nyl-substituted phenols, the following results were obtained:
70>20>80>30; this order is in good agreement with the
ARP’s values and k1 rate constants.
Time (s)
Figure 1. Decay of the visible absorbance (515 nm) of a DPPH (1.5 mL,
5.6ꢃ10ꢁ5 mol Lꢁ1) solution in MeOH (25 ꢀC) following addition of tetra-
hydrocurcumin (2, 1.5 mL, 2.4ꢃ10ꢁ4 mol Lꢁ1) (top). Determination of
the rate constant for first hydrogen abstraction k1 (bottom).
Priyadarsini et al.17a conclusions despite energetics to re-
move hydrogen from both phenolic OH and the CH2 group
of the b-diketo structure (or enol OH) are very close. It
can be concluded that phenolic hydroxyl groups play a major
role in the antioxidant activity of curcuminoids. The same
conclusion dealing with curcumin has been reported in a
recent review.17b
The antioxidant activity of non-phenolic curcuminoids 10, 40,
50, and 60 is very low (Table 2). This is in accordance with
Table 3. Relative rate constant for first hydrogen abstraction k1 by compar-
ison with BHTa (see Sections 2 and 4)
b
2.2.3. Tetrahydrocurcuminoids. The antioxidant activity
of the phenolic tetrahydrocurcuminoids follows the same
classification as the corresponding curcuminoids except
compounds 3 and 10. The antiradical power ratios 2/20 and
7/70 are similar, 2 and 2.25, respectively, whereas for 3/30
it is equal to 3.15 and for 10/100 it is 0.95. For 2 and 7 the
number of reduced DPPH (NRD) are almost twice than those
of 20 and 70, in accordance with the presence of an easy re-
movable benzylic hydrogen, as it was already discussed by
Brand-Williams et al.17 The lack of mesomeric effect due
to the heptadienone chain in tetrahydroisocurcumin, 3, con-
duces this compound to have very similar properties to tetra-
hydrocurcumin, 2. For both compounds the substituted alkyl
Compounds
k1 rel
BHT
2
20
3
30
7
70
10
100
9
1
7.6
3.8
0.8
0.02
235
177
5.2
1.4
80
250
12
This work: k1: 4.2 L molꢁ1 sꢁ1; lit.20: 5 L molꢁ1 sꢁ1
ꢂ10%.
.
a
b

9096
E. Portes et al. / Tetrahedron 63 (2007) 9092–9099
chain has minor electronic effect. By contrast, the antioxi-
dant activity measured for tetrahydrodemethoxycurcumin,
10, is similar to that of its curcumin precursor. The NRD cal-
culated for hydrogenated and non-hydrogenated compounds
are close, 2.5 and 2.4, respectively. This indicates that the
phenolic group of the p-hydroxyphenyl part is not playing
any role in the primary process but might participate in the
secondary processes sketched by Brand-Williams et al.18
This phenolic hydrogen might compete with the benzylic
one in 10. This hypothesis is confirmed by the poor antiox-
idant data found for the monophenolic compounds 8 and 11
(Table 2). Among the active tetrahydrocurcumins, 12 with
two trihydroxyphenyl rings (gallic unit) presents the best an-
tioxidant activity followed by 9. This is in accordance with
the known antioxidant activities of gallic and catechol units
in polyphenols.18–21
The data in Tables 1, 2, and 3 indicate that phenolic tetra-
hydrocurcuminoids present ARP values and first order
hydrogen abstraction rate constant k1 higher than their cur-
cuminoid analogs, if they have a hydroxyl or methoxy sub-
stituents neighboring the phenol group. This is likely due
to the presence of benzylic hydrogens, which are involved
in the oxidation process of these compounds, and not in cur-
cuminoids. Also, it has been claimed that electron-donating
substituents weaken the benzyl C–H bond.34 This is in accor-
dance with the claimed antioxidant properties of compound
2 described by Sabinsa Corporation in their web site.35
Moreover, the photochemical behavior of THCs is also
closely related to their antioxidant properties, their photo-
chemical reactive quantum yield being parallel with their
ARP values.24
According to Wright’s data33 and assimilating THCs to methyl-
substituted phenols, the following order was obtained
for the calculated antioxidant activity: 12>9>7>2>3. This
is in very good agreement with the times EC50, k1 rate
constants and ARP of THCs.
3. Conclusion
A series of curcuminoids and tetrahydrocurcuminoids, bear-
ing various hydroxyl and methoxy groups on their benzene
subunits, have been synthesized to systematically study their
antioxidant and hydrogen donating capacities using the
DPPH method at 25 ꢀC in methanol. The results obtained
show that the tetrahydrocurcuminoids are in general much
more efficient than their curcuminoid analogs if they include
a phenol group in meta- or para-position of the linking chain
and a phenol or methoxy group as neighbor. This gain in
efficiency of THCs by comparison to curcuminoids is not
attributed to the presence of the b-diketone moiety in the
chain, as it was already proposed, but more likely, to the
presence of benzylic hydrogens, which are involved in
the oxidation process of these compounds, and not in
curcuminoids.
The antiradical properties of the studied tetrahydrocurcumi-
noids have shown the importance of the presence of both
phenolic group and hydroxyl or methoxy in ortho-position
to stabilize the phenoxy radical after hydrogen transfer.
Formally the latter might be a consequence of an electron
transfer followed by deprotonation of the formed radical
cation. The first pKa of a series of reference compounds
and THCs measured in water for comparison by UV-
absorption spectrometry are indicated in Table 4. The pKa
of the phenol and of the enol in water are very similar, so
the deprotonation might be easy from both part of the
tetrahydrocurcuminoid molecules. We might expect similar
behavior in methanol because they are both protic and polar
solvents.
4. Experimental
The electron transfer depends on the ionization potential of
the molecule e.g. on the position of its highest occupied
molecular orbital (HOMO). Considering that the electron
donation in the electron transfer process originates from the
benzene subunits, electron-donating groups such as hydroxyl
and methoxy substituents should favor the process. The
inefficiency of the various methoxylated tetrahydrocurcumi-
noids such as 6 indicates that this process is not operating.
Thus a mechanism involving an electron transfer from one
benzene nucleus followed by deprotonation of the enolic
part of the tetrahydrocurcuminoids does not appear to occur
in the antioxidant action of these compounds; the involve-
ment of the b-diketone part to their antioxidant properties,
as it is reported by Sugiyama et al.,15 should be minor.
4.1. Materials and methods
The starting materials and solvents of appropriate grade (for
synthesis or for spectroscopy) were obtained from Aldrich
and used without further purification. The synthesized com-
pounds were purified on Merck silica gel 60. TLC analysis of
the synthesized compounds was carried out on Fluka silica
gel 60 F254 plates (thickness 0.20 mm). Melting points
were measured on a heating microscope Electrothermal
9100 Reichert. Studies by 1H and 13C NMR were made using
Bruker Avance 300 Fourier transform spectrometer. Infrared
spectra were obtained with a Paragon 1000 PC Perkin–
Elmer FTIR spectrometer. UV–vis spectra were recorded
on a Lambda 18 Perkin–Elmer spectrometer using just pre-
pared solutions to avoid product degradation.15 GC–MS
analyses of tetrahydrocurcuminoids 1–8 and 10 were per-
formed with a Finnigan Trace mass spectrometer interfaced
with a Finnigan Trace GC Ultra gas apparatus (line transfer
temperature: 250 ꢀC) equipped with a PTVinjector (splitless
mode) using helium as carrier gas. A fused silica capillary
RTX-5MS column, 15 m, 0.25 mm i.d., film thickness
0.25 mm was selected. The oven temperature was pro-
grammed from 40 ꢀC (initial hold time of 1 min) to 320 ꢀC
at a rate of 15 ꢀC minꢁ1; this final temperature was main-
tained for 15 min. The electron energy was fixed at 70 eV.
Table 4. First pKa of some reference compounds and tetrahydrocurcumi-
noids
Compounds
pKa
p-Cresol (4-methylphenol)
Acetylacetone
4-Propylguaiacol
1
2
8
10.0, lit.36 10.3
9.2, lit.37 8.9–9.4
9.8
9.0
8.6
9

E. Portes et al. / Tetrahedron 63 (2007) 9092–9099
9097
Only the most significant peaks are given. HPLC analyses of
curcuminoids and of tetrahydrocurcuminoids 9 and 12 used
a Thermo Separation Product line including a pump type
SP1000, an automatic injector AS 3000 and a UV detector
AS 2000 and a column type Lichrospher (250ꢃ4.6 mm;
4.1.2.4. 1,7-Bis(3,4-dimethoxyphenyl)-heptane-3,5-di-
one (5). Yield: 70%, white powder; mp 72–73 ꢀC; Rf¼0.70
(ethyl acetate/petroleum ether (1/1)); 1H NMR (CDCl3,
300 MHz): d 2.53–2.89 (m, 8H, –CH2 (Ha, Hb, Hf, Hg));
3.85 (s, 12H, –OCH3); 5.44 (s, 1H, Hd (enol)); 6.68–6.80
(m, 6H, J¼9.0 Hz, 3.0 Hz, HAr); 15.48 (br s, 1H, –OH
enol); 13C NMR (CDCl3, 75.5 MHz): d 193.06 (Cc, Ce);
148.83 (C30); 147.44 (C40); 133.29 (C10); 120.08 (C60);
111.61 (C50); 111.25 (C20); 99.80 (Cd); 55.90 (–OMe);
55.80 (–OMe); 40.30 (Cb, Cf); 31.16 (Ca, Cg); EIMS m/z
˚
100 A; 5 mm) using a mixture of methanol/water (20/80 v/
v) as eluent. High-resolution mass spectrum analyses
(HRLSIMS) were performed using a VG Micromass
AutoSpec Q operating with a positive LSIMS ionization
mode (Cs+, ion bombardment energy: 35 keV; matrix:
3-nitrobenzyl alcohol).
ꢄ
(%): 400 (M+ , 29); 165(5); 164(10); 151(100); HRLSIMS:
calcd for C23H28O6: 400.1886; found: 400.1883; UV (meth-
anol): lmax nm (3 L molꢁ1 cmꢁ1): 280 (14,932).
4.1.1. Preparation of curcuminoids. The different curcu-
minoids were prepared by a Knoevenagel condensation of
the corresponding benzaldehydes and boron protected 2,4-
pentanedione according to a well reported literature proce-
dure.25 All the studied curcuminoids have already been
described and the physical properties obtained are in good
accordance.27–29
4.1.2.5. 1,7-Bis(3,4,5-trimethoxyphenyl)-heptane-3,5-
dione (6). Yield: 60%, lightly yellowish powder; mp 75–
1
76 ꢀC; Rf¼0.63 (ethyl acetate/petroleum ether (2/1)); H
NMR (CDCl3, 300 MHz): d 2.55–2.89 (m, 8H, –CH2 (Ha,
Hb, Hf, Hg)); 3.81 (s, 6H, –OCH3); 3.82 (s, 12H, –OCH3);
5.45 (s, 1H, Hd (enol)); 6.40 (s, 4H, HAr); 15.46 (br s, 1H,
–OH enol); 13C NMR (CDCl3, 75.5 MHz): d 192.71 (Cc,
Ce); 153.11 (C30, C50); 140.51 (C10); 136.51 (C40); 105.44
(C20, C60); 99.99 (Cd); 60.95 (–OMe); 56.18 (–OMe);
4.1.2. Preparation of tetrahydrocurcuminoids. Curcumi-
noids were hydrogenated over palladium to give the cor-
responding tetrahydrocurcuminoids in good yields. As
a typical example, a curcuminoid (2.1 mmol) in a mixture
of ethyl acetate/methanol (15 mL/20 mL) and 10% palla-
dium on charcoal (0.08 g) was stirred under hydrogen for
2 h at room temperature. The catalyst was removed by filtra-
tion and the solvent was evaporated. The residue was puri-
fied on silica gel using ethyl acetate as eluent.
ꢄ
40.32 (Cb, Cf); 32.07 (Ca, Cg); EIMS m/z (%): 460 (M+ ,
11); 195(32); 194(5); 181(100); HRLSIMS: calcd for
C25H32O8: 460.2097; found: 460.2075; UV (methanol):
lmax nm (3 L molꢁ1 cmꢁ1): 276 (9737).
4.1.2.6. 1,7-Bis(4-hydroxy-3,5-dimethoxyphenyl)-hep-
tane-3,5-dione (7). Yield: 65%, white powder; mp 86–
87 ꢀC; Rf¼0.70 (ethyl acetate); 1H NMR (CDCl3,
300 MHz): d 2.08 (s, 2H, –OHAr); 2.51–2.86 (m, 8H, –CH2
4.1.2.1.
1,7-Bis(4-hydroxy-3-methoxyphenyl)-hep-
tane-3,5-dione (1) and 1,7-bisphenyl-heptane-3,5-dione
(2). Their preparations and characterizations have already
been described in the literature.31a,b
0
(Ha, Hb, Hf, Hg)); 3.55 (s, 0.4H, Hd (diketone)); 3.84 (s, 12H,
–OCH3); 5.42 (s, 0.8H, Hd (enol)); 6.37 (s, 4H, HAr); 15.49
(br s, 0.8H, –OH enol); 13C NMR (CDCl3, 75.5 MHz):
d 193.15 (Cc, Ce); 147.26 (C30, C50); 133.26 (C10); 131.85
(C40); 105.08 (C60, C20); 99.99 (Cd); 56.38 (–OMe); 40.58
(Cb, Cf); 31.92 (Ca, Cg); EIMS m/z (abundance %):
4.1.2.2.
1,7-Bis(3-hydroxy-4-methoxyphenyl)-hep-
tane-3,5-dione (3). Yield: 88%, white powder; mp 121 ꢀC;
Rf¼0.46 (ethyl acetate/petroleum ether (1/1)); 1H NMR
(CDCl3, 300 MHz): d 2.51–2.84 (m, 8H, –CH2 (Ha, Hb, Hf,
ꢄ
432(M+ , 11); 181(12); 180(6); 167(100); HRLSIMS: calcd
0
Hg)); 3.50 (s, 0.5H, Hd (diketone)); 3.85 (s, 3H, –OCH3);
for C23H28O8: 432.1784; found: 432.1790; UV (methanol):
lmax nm (3 L molꢁ1 cmꢁ1): 277 (13,800).
3.86 (s, 3H, –OCH3); 5.43 (s, 0.75H, Hd (enol)); 5.59 (s, 2H,
–OHAr); 6.61–6.78 (m, 6H, J¼9.0 Hz, 3.0 Hz, HAr); 15.41
(br s, 0.75H, –OH enol); 13C NMR (CDCl3, 75.5 MHz):
d 193.29 (Cc, Ce); 145.75 (C40); 145.24 (C30); 134.01 (C10);
120.02 (C60); 115.04 (C20); 111.25 (C50); 99.86 (Cd); 55.99
(–OMe); 40.34 (Cb, Cf); 31.33 (Ca, Cg); EIMS m/z (%):
4.1.2.7. 1,7-Bis(4-hydroxy-phenyl)-heptane-3,5-dione
(8). Yield: 66%, white powder; mp 101–102 ꢀC; Rf¼0.67
1
(ethyl acetate/petroleum ether (3/1)); H NMR (DMSO-d6,
300 MHz): d 2.48–2.73 (m, 8H, –CH2 (Ha, Hb, Hf, Hg));
ꢄ
372(M+ , 20); 151(6); 150(12);137(100); HRLSIMS: calcd
3.65 (s, 1H, Hd (diketone)); 5.69 (s, 0.5H, Hd (enol)); 6.61–
0
for C21H24O6: 372.1573; found: 372.1574; UV (methanol):
lmax nm (3 L molꢁ1 cmꢁ1): 281 (14,837).
6.64 (m, 4H, J¼9.0 Hz, 3.0 Hz, HAr); 6.88–6.91 (m, 4H,
J¼9.0 Hz, 3.0 Hz, HAr); 15.50 (br s, 0.5H, –OH enol); 13C
NMR (DMSO-d6, 75.5 MHz): d 194.35 (Cc, Ce); 155.37
(C40); 130.64 (C10); 129.07 (C60, C20); 115.06 (C30, C50);
99.11 (Cd); 44.73 (Cb, Cf); 29.98 (Ca, Cg). EIMS m/z (abun-
4.1.2.3. 1,7-Bis(4-methoxyphenyl)-heptane-3,5-dione
(4). Yield: 90%, white powder; mp 68–69 ꢀC; Rf¼0.80 (di-
ꢄ
1
chloromethane); H NMR (CDCl3, 300 MHz): d 2.52–2.89
dance %): 312 (M+ , 16); 121 (7); 120 (28); 107 (100);
(m, 8H, –CH2 (Ha, Hb, Hf, Hg)); 3.80 (s, 6H, –OCH3); 5.43
(s, 1H, Hd (enol)); 6.82–6.85 (m, 4H, J¼9.0 Hz, 3.0 Hz,
HAr); 7.08–7.11 (m, 4H, J¼9.0 Hz, 3.0 Hz, HAr); 15.45
(br s, 1H, –OH enol); 13C NMR (CDCl3, 75.5 MHz):
d 193.41 (Cc, Ce); 158.18 (C40); 132.93 (C10); 129.30
(C20); 114.01 (C30, C50); 99.74 (Cd); 55.37 (–OMe); 40.42
HRLSIMS: calcd for C19H20O4: 312.1362; found: 312.1354;
UV (methanol): lmax nm (3 L molꢁ1 cmꢁ1): 278 (13,810).
Different physical data are given in the literature for this
product.31
ꢄ
(Cb, Cf); 30.77 (Ca, Cg); EIMS m/z (%): 340(M+ , 14);
4.1.2.8. 1,7-Bis(3,4-dihydroxy-phenyl)-heptane-3,5-di-
one (9). This compound was prepared by hydrogenation
of 1,7-bis(3,4-dibenzyloxyphenyl)-1,6-heptadiene-3,5-di-
one already described in the literature.27
135(4); 134(16); 121(100); HRLSIMS: calcd for
C21H24O6: 340.1675; found: 340.1671; UV (methanol):
lmax nm (3 L molꢁ1 cmꢁ1): 277 (12,406).

9098
E. Portes et al. / Tetrahedron 63 (2007) 9092–9099
Yield: 85%, yellowish oil; Rf¼0.54 (ethyl acetate/petroleum
absorbance (lmax¼515 nm, 3¼11,240 L molꢁ1 cmꢁ1),
which follows the antioxidant addition to the DPPH solution.
The temperature in the cell was maintained at 25 ꢀC by cir-
culating in the cell holder a water/ethanol mixture (1/1 v/v)
maintained at 25 ꢀC by a thermostat. In a typical procedure,
5 mL of a freshly prepared methanol solution of DPPH
(6.10ꢁ5 mol Lꢁ1) is mixed with an appropriate volume of
a freshly prepared solution of the antioxidant in methanol
(10ꢁ4 mol Lꢁ1) to reach the desired antioxidant/DPPH con-
centration ratio. The absorption spectra were recorded every
2 min first and then every 5 min. For each antioxidant
concentration tested, the reaction kinetics were plotted, the
percentage of DPPH remaining at the steady state was deter-
mined and the values plotted on an other graph showing the
percentage of residual DPPH at the steady state as a function
of the molar ratio of antioxidant to DPPH. The values EC50
and Time EC50 were determined from this plot. For the less
reactive antioxidants, the ratio antioxidant/DPPH was mea-
sured after 90% remaining DPPH e.g., 10% of conversion.
Values EC10 and Time EC10 were also determined. The
DDPH/antioxidant molar ratios were set between 0.05 and
2.00. It was checked that in the experimental conditions
used, curcuminoids do not contribute to the absorbance at
515 nm. The experiments were run in triplicate for each
antioxidant evaluation.
1
ether (3/1)); H NMR (DMSO-d6, 300 MHz): d 2.32–2.73
0
(m, 8H, –CH2 (Ha, Hb, Hf, Hg)); 3.49 (s, 0.8H, Hd (di-
ketone)); 5.44 (s, 0.6H, Hd (enol)); 6.39–6.69 (m, 6H,
J¼9.0 Hz, 3.0 Hz, HAr); 8.06 (br s, 4H, –OHAr); 15.46 (br
s, 0.6H, –OH enol); 13C (DMSO-d6, 75.5 MHz): d 196.29
(Cc, Ce); 144.45 (C30); 141.74 (C40); 134.21 (C10); 121.92
(C60); 117.04 (C50); 100.15 (Cd); 42.54 (Cb, Cf); 30.63
(Ca, Cg); EIMS (direct introduction) m/z (%): 344(4);
123(100); HRLSIMS: calcd for C21H24O6: 344.1260; found:
344.1265; UV (methanol): lmax nm (3 L molꢁ1 cmꢁ1): 283
(11,750).
4.1.2.9. 1-(4-Hydroxy-phenyl)-7-(4-hydroxy-3-meth-
oxyphenyl)-heptane-3,5-dione (10). Yield: 75%, lightly
yellowish oil; Rf¼0.33 (dichloromethane); 1H NMR
(CDCl3, 300 MHz): d 1.73 (br s, 1H, –OHAr); 2.51–2.85
0
(m, 8H, –CH2 (Ha, Hb, Hf, Hg)); 3.49 (s, 0.4H, Hd (diketone));
3.86 (s, 3H, –OCH3); 5.41 (s, 0.8H, Hd (enol)); 5.57 (br s, 1H,
–OHAr); 6.63–6.83 (m, 5H, J¼9.0 Hz, 3.0 Hz, HAr); 6.99–
7.02 (m, 2H, J¼9.0 Hz, 3.0 Hz, –HAr); 15.50 (br s, 0.8H,
–OH enol); 13C NMR (acetone-d6, 75.5 MHz): d 197.71
(Cc, Ce); 157.01 (C40); 148.52 (C3); 146.11 (C4); 133.53
(C1); 133.24 (C10); 129.28 (C20, C60)); 121.61 (C6); 116.3
(C5); 116.02 (C30, C50); 113.03 (C2); 100.62 (Cd); 55.98
(–OMe); 45.74 (Cb); 40.61 (Cf); 32.22 (Ca, Cg); HRLSIMS:
calcd for C20H22O5: 342.14672; found: 342.14759; UV
(methanol): lmax nm (3 L molꢁ1 cmꢁ1): 280 (13,580).
The determination of the rate constant for first hydrogen
abstraction k1 from BHT and curcuminoids and tetrahydro-
curcuminoids to DPPH was made by measuring the decay
of the visible absorbance (515 nm) of a DPPH solution in
MeOH (1.5 mL, 5.6ꢃ10ꢁ5 mol Lꢁ1, 25 ꢀC) following addi-
tion of the antioxidant (1.5 mL, 2.4ꢃ10ꢁ4 mol Lꢁ1) accord-
ing to a procedure described by Goupy et al.20
4.1.2.10. 1,7-Bis(3-hydroxy-phenyl)-heptane-3,5-di-
one (11). Yield: 65%, orange oil; Rf¼0.77 (ethyl acetate/
1
petroleum ether (3/1)); H NMR (acetone-d6, 300 MHz):
d 2.39–2.75 (m, 8H, –CH2 (Ha, Hb, Hf, Hg)); 3.52 (s, 0.6H,
0
Hd (diketone)); 5.51 (s, 0.7H, Hd (enol)); 6.39–6.65 (m,
6H, J¼9.0 Hz, 3.0 Hz, HAr); 6.91–7.02 (m, 2H, J¼9.0 Hz,
3.0 Hz, HAr); 8.15 (2H, –OHAr); 15.55 (br s, 0.7H, –OH
enol); 13C NMR (acetone-d6, 75.5 MHz): d 193.26 (Cc,
Ce); 158.64 (C30); 143.58 (C10); 130.51 (C50); 120.47 (C60);
114.22 (C40); 116.45 (C20);100.46 (Cd); 40.64 (Cb, Cf); 32.23
(Ca, Cg); HRLSIMS: calcd for C19H20O4: 312.1362; found:
312.1365; UV (methanol): lmax nm (3 L molꢁ1 cmꢁ1): 279
(11,480).
The first pKa of studied compounds (see Table 4) was
obtained from the titration of a dilute water solution
(5.10ꢁ5 mol Lꢁ1) by adding small amounts of hydrochloric
acid or sodium hydroxide to change the pH in the range
3.0–12.0 according to the well known procedure.38 The titra-
tion vessel was thermostated at 25 ꢀC. At each pH, UV–vis
spectra of the solutions were recorded in quartz cuvette
(length 1 cm).
4.1.2.11. 1,7-Bis(3,4,5-trihydroxy-phenyl)-heptane-
3,5-dione (12). Yield: 40%, yellowish oil; Rf¼0.63 (ethyl
acetate); 1H NMR (acetone-d6, 300 MHz): d 2.26–2.73 (m,
8H, –CH2 (Ha, Hb, Hf, Hg)); 3.60 (s, 0.4H, Hd (diketone));
5.60 (s, 0.8H, Hd (enol)); 6.19–6.23 (m, 4H, J¼9.0 Hz,
3.0 Hz, HAr); 7.44 (br s, 6H, –OHAr); 15.48 (br s, 0.8H,
–OH enol); 13C NMR (acetone-d6, 75.5 MHz): d 204.26
(Cc, Ce); 144.19 (C30, C50); 132.41 (C10); 128.86 (C40);
105.84 (C20, C60); 97.86 (Cd); 43.46 (Cb, Cf); 30.03 (Ca,
Cg); HRLSIMS: calcd for C19H20O8Na: 399.1056; found:
399.1049; UV (methanol): lmax nm (3 L molꢁ1 cmꢁ1): 272
(4425).
References and notes
1. Tonnesen, H. H.; Karlsen, J. J. Chromatogr. 1983, 259, 367.
2. Tonnesen, H. H.; Karlsen, J.; Mostad, A. Acta Chim. Scand. B
1982, 36, 475.
3. Govindarajan, V. S. Crit. Rev. Food Sci. 1980, 12, 199.
4. Ali, M.; Bagati, A.; Gupta, J. Indian J. Chem. 1995, 34B, 884.
5. Sui, Z.; Salto, R.; Li, J.; Craik, C.; Ortiz de Montellano, P. R.
Bioorg. Med. Chem. 1993, 1, 415.
6. For a recent review see for example: Aggarwal, B. B.; Kumar,
A.; Aggarwal, M. S.; Shishodia, S. Phytopharmaceuticals in
Cancer Chemoprotection; Bagchi, D., Preuss, H. G., Eds.;
CRC: Boca Raton, FL, 2005; pp 349–387.
7. Jankun, E. S.; McCabe, N. P.; Selman, S. H.; Jankun, J. Int. J.
Mol. Med. 2000, 6, 521.
4.1.3. Antioxidant and pKa measurements. The H-transfer
reaction from curcuminoids and tetrahydrocurcuminoids to
DPPH in methanol solution was monitored by UV–vis ab-
sorption spectrometry using a Perkin Elmer Lambda 18
spectrometer (quartz cuvette, length 1 cm, solvent methanol)
following a procedure described by Brand-Williams et al.18
It consists in recording the decay of the DPPH visible
8. Huang, M. T.; Lysz, T.; Ferraro, T.; Abidi, T. F.; Laskin, J. D.;
Conney, A. H. Cancer Res. 1991, 51, 813.
9. Sharma, S. C.; Mukhtar, H.; Sharma, S. K.; Krishna Murti,
C. R. Biochem. Pharmacol. 1972, 21, 1210.

E. Portes et al. / Tetrahedron 63 (2007) 9092–9099
9099
10. Sharma, O. P. Biochem. Pharmacol. 1976, 25, 1811.
11. Scartezzini, P.; Speroni, E. J. Ethnopharmacol. 2000,
71, 23.
24. Castellan, A.; Ruggiero, R.; Da Silva, L. G.; Portes, E.; Grelier,
S.; Gardrat, C. J. Photochem. Photobiol., A: Chem. 2007, 190,
110.
12. Majeed, M.; Badmaev, V.; Shimakumar, U.; Rajendran, R.
Curcuminoids: Antioxidant Phytonutrients; Nutriscience:
Piscataway, NJ, 1990.
13. Pan, M. H.; Huang, T. M.; Lin, J.-K. Drug Metab. Dispos. 1999,
27, 486.
25. Krackov, M. H.; Bellis, H. E. WO Patent 9,716,403, 1997;
Chem. Abstr. 1997, 127, 17480.
26. Roughley, P. J.; Whiting, D. A. J. Chem. Soc., Perkin Trans. 1
1973, 2379.
27. Nurfina, A. N.; Reksohadiprodjo, M. S.; Timmerman, H.;
Jenie, U. A.; Sugiyanto, D.; Van der Goot, H. Eur. J. Med.
Chem. 1997, 32, 321.
14. Osawa, T.; Sugiyama, Y.; Inayoshi, M.; Kawakishi, S. Biosci.
Biotechnol. Biochem. 1995, 59, 1609.
15. Sugiyama, Y.; Kawakishi, S.; Osawa, T. Biochem. Pharmacol.
1996, 52, 519.
28. Itokawa, H.; Morita, H.; Midorikawa, I.; Aiyama, R.; Morita,
M. Chem. Pharm. Bull. 1985, 33, 4889.
16. Nakamura, Y.; Ohto, Y.; Murakami, A.; Osawa, T.; Ohigashi,
H. Jpn. J. Cancer Res. 1998, 89, 361.
29. Hoehle, S. I.; Pfeiffer, E.; Solyom, A. M.; Metzler, M. J. Agric.
Food Chem. 2006, 54, 756.
17. (a) Priyadarsini, K. I.; Maity, D. K.; Naik, G. H.; Kumar,
M. S.; Unnikrishnan, M. K.; Satav, J. G.; Mohan, H. Free
Radical Bio. Med. 2003, 35, 475; (b) Barclay, L. R. C.;
Vinquist, M. R. The Chemistry of Phenols; Rappoport, Z.,
Ed.; John Wiley and Sons: Chichester, UK, 2003; pp 868–
869, Part 2, Chapter 12.
30. Asakawa, Y. Bull. Chem. Soc. Jpn. 1972, 45, 1794.
31. (a) Venkateswarlu, S.; Rambabu, M.; Subbaraju, G. V.;
Satyanarayana, S. Asian J. Chem. 2000, 12, 141; (b)
Kamnaing, P.; Tsopmo, A.; Tanifun, E. A.; Tchuendem,
M. H. K.; Tane, P.; Ayafor, J. F.; Sterner, O.; Rattendi, D.;
Iwu, M. M.; Schuster, B.; Bacchi, C. J. Nat. Prod. 2003, 66, 364.
32. Pedersen, U.; Rasmussen, P. B.; Lawesson, S. O. Liebigs Ann.
Chem. 1985, 1557.
18. Brand-Williams, W.; Cuvelier, M. E.; Berset, C. Food Sci.
Technol. (London) 1995, 28, 25.
19. Bondet, V.; Brand-Williams, W.; Berset, C. Food Sci. Technol.
(London) 1997, 30, 609.
33. Wright, J. S.; Johnson, E. R.; DiLabio, G. A. J. Am. Chem. Soc.
2001, 123, 1173.
20. Goupy, P.; Dufour, C.; Loonis, M.; Dangles, O. J. Agric. Food
Chem. 2003, 51, 615.
21. Roche, M.; Dufour, C.; Mora, N.; Dangles, O. Org. Biomol.
Chem. 2005, 3, 423.
22. Sundaryono, A.; Nourmamode, A.; Gardrat, C.; Grelier, S.;
Despeyroux, O.; Castellan, A. Holz Roh- und Werkst 2003,
61, 377.
34. Arafat, A. M.; Mathew, S. K.; Akintobi, S. O.; Zavitsas, A. A.
Helv. Chim. Acta 2006, 89, 2226 and references cited therein.
35. www.Tetrahydrocurcuminoids.com.
36. Kim, J.-B.; Britz-McKibbin, P.; Hirokawa, T.; Terabe, S. Anal.
Chem. 2003, 75, 3986.
37. Mofaddel, N.; Bar, N.; Villemin, D.; Desbene, P. L. Anal.
Bioanal. Chem. 2004, 380, 664.
23. Sundaryono, A.; Nourmamode, A.; Gardrat, C.; Grelier, S.;
Bravic, G.; Chasseau, D.; Castellan, A. Photochem.
Photobiol. Sci. 2003, 2, 914.
38. Albert, A.; Serjeant, E. P. The Determination of Ionisation
Constants: A Laboratory Manual, 3rd ed.; Chapman and
Hall: London, 1984.