Electrochemical behaviour of new dimeric esters and amides derived from caffeic acid in dimethylsulfoxide

Article abstract of DOI:10.1039/c4ob00823e

The electrochemical oxidation in DMSO of four new derivatives of caffeic acid (CA), two dimeric amides and two dimeric esters, is reported in this article. Although all of them contain two caffeoyl electroactive moieties in their structures, small differences in the connectors result in interesting changes in the electrochemical behaviour of this type of compound. Voltammograms of both esters do not show appreciable differences between them; however, an electrografting process occurs during the electrochemical oxidation of one of them, which suggests that the identity of the connector has an influence on the ability of the diesters to interact with the electrode surface. On the other hand, voltammograms of dimeric amides were more complex than those corresponding to dimeric esters. Electronic effects of diamine connectors seem to be related to the fact that caffeoyl moieties suffer from separate oxidation processes in both compounds. In contrast to their ferulic acid (FA) analogues, which have been studied by our group before, CA dimeric amides do not interact in an appreciable way with the electrode surface. In addition, a relationship between the oxidation potential and the inhibition percentage of the DPPH (2,2′-diphenyl-1-picrylhydrazyl) radical was not observed for the symmetrical CA derivatives studied here. However, the molecular flexibility seems to play a very important role in the Free Radical Scavenging Activity (FRSA) of this type of compound. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4ob00823e

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Electrochemical behaviour of new dimeric
esters and amides derived from caffeic acid
in dimethylsulfoxide†
Cite this: Org. Biomol. Chem., 2014,
12, 5981
Analilia Sánchez, Omar Martínez-Mora, Evelin Martínez-Benavidez,
Javier Hernández, Zaira Domínguez* and Magali Salas-Reyes*
The electrochemical oxidation in DMSO of four new derivatives of caffeic acid (CA), two dimeric amides
and two dimeric esters, is reported in this article. Although all of them contain two caffeoyl electroactive
moieties in their structures, small differences in the connectors result in interesting changes in the
electrochemical behaviour of this type of compound. Voltammograms of both esters do not show
appreciable differences between them; however, an electrografting process occurs during the electro-
chemical oxidation of one of them, which suggests that the identity of the connector has an influence on
the ability of the diesters to interact with the electrode surface. On the other hand, voltammograms of
dimeric amides were more complex than those corresponding to dimeric esters. Electronic effects of
diamine connectors seem to be related to the fact that caffeoyl moieties suffer from separate oxidation
processes in both compounds. In contrast to their ferulic acid (FA) analogues, which have been studied by
our group before, CA dimeric amides do not interact in an appreciable way with the electrode surface.
In addition, a relationship between the oxidation potential and the inhibition percentage of the DPPH
(2,2’-diphenyl-1-picrylhydrazyl) radical was not observed for the symmetrical CA derivatives studied here.
However, the molecular flexibility seems to play a very important role in the Free Radical Scavenging
Activity (FRSA) of this type of compound.
Received 22nd April 2014,
Accepted 10th June 2014
DOI: 10.1039/c4ob00823e
www.rsc.org/obc
1. Introduction
Caffeic acid (CA in Fig. 1) is a phenolic compound that
belongs to the group of hydroxycinnamic acids. It is found
naturally in plants and vegetables such as carrots, tomatoes,
strawberries, blueberries, or potatoes.1–4 The antioxidant pro-
perties and biological importance of CA and some of its
natural esters, such as chloragenic5–10 and rosmarinic
acids,11–13 as well as the caffeic acid phenethyl ester
(CAPE),14–17 are well-documented in the literature. In addition,
caffeic acid amide derivatives have also been identified in
natural foods appreciated for their beneficial effect on the
human health, such as oats18 and amaranth.19
Fig. 1 Caffeic acid (CA) and natural dimeric derivatives containing two
caffeoyl units.
Unidad de Servicios de Apoyo en Resolución Analítica, Universidad Veracruzana,
Apdo. Postal 575, Xalapa, Ver., México. E-mail: magasalas@uv.mx,
zdominguez@uv.mx; Tel: +52 22 88 41 89 17
Dimeric esters containing two caffeoyl units are also
present in nature (Fig. 1). For example, 1,3-di-O-caffeoylquinic
acid (1) was identified from yerba mate (Ilex paraguariensis),20
whose benefits to health have been connected to its high
content of polyphenolic secondary metabolites. 1,2-di-O-
Caffeoylcyclopenta-3-ol (2) was isolated from Acanthopanax
†Electronic supplementary information (ESI) available: Additional ESI corres-
ponding to cyclic voltammetry of compounds 6 (Fig. S1) and 8 (Fig. S2) at
different scan rates are provided. The dependence of the anodic (Fig. S1) and
cathodic peak current (Fig. S2), respectively, on the square root of the potential
scan rate (v) is also shown. 13C NMR spectra of the new compounds 6–9 appear
as Fig. S3–S6. See DOI: 10.1039/c4ob00823e
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5981–5989 | 5981

View Article Online
Paper
Organic & Biomolecular Chemistry
koreanum,21 which is a plant used in traditional Korean medi-
cine in the treatment of rheumatism, hepatitis, and diabetes.
This compound has been found in Daphne feddei too,22 a kind
of shrub whose leaves and stems are used in some provinces
of China to treat injuries from falls and bruises. On the other
hand, chicoric acid (3), a natural diester derived from tartaric
and caffeic acids, was reported as the main component in the
leaves and roots of dandelion (Taraxacum officinale),23 which
has shown anti-inflammatory activity in animal studies. In
addition, L-chicoric acid is a potent and selective inhibitor of
type 1 HIV replication. This fact has motivated during the last
few years the synthesis of compounds that employ L-chicoric
acid as a template, and also several activity relationship (SAR)
studies, in order to identify the structural features involved in
selectivity and potency against integrase and HIV
replication.24–28 Less common are the reports about natural
dimeric amides derived from CA. An example of this kind of
compounds is N,N′-dicaffeoylspermidine (4), identified from
eggplant (Solanum melongena L),29 which is among the top 10
vegetables with oxygen radical trapping capacity.
Fig. 2 Caffeic acid ester and amide derivatives studied in the present
work.
We are interested in the antioxidant properties of dimeric
derivatives of hydroxycinnamic acids, since some of them have
been suggested as therapeutic agents against human health
disorders caused by oxidative stress.24–28,30,31 Cyclic voltamme-
try is a promising tool in the study of antioxidant compounds,
and numerous reports are found in the literature about the
application of this electrochemical technique to characterize
polyphenolic compounds.32–37 We have reported recently the
electrochemical behaviour of monomeric esters and amides
derived from CA and ferulic acid (FA),38 as well as mechanistic
aspects of electrochemical oxidation of dimeric FA amides.39
In those studies, we found that all compounds followed a
similar oxidation mechanism; however, FA amides additionally
undergo electrografting reactions. In addition, dimeric topo-
logy seems to favour this tendency. In contrast, caffeic acid
benzyl amide (CABzA) showed an oxidation potential even
lower than that of CAPE and CA. This amide was also the
compound, of the complete series, with a higher capacity to
scavenge the DPPH (2,2′-diphenyl-1-picrylhydrazyl) radical.
In this context, the aim of this study was to determine the
electrochemical behaviour of new compounds with dimeric
topology: two symmetrical esters (6 and 7 in Fig. 2) and two
symmetrical amides (8 and 9) derived from caffeic acid in an
aprotic medium. The influence of the connectors on the oxi-
dation mechanism is also discussed here. Since dimethylsulf-
oxide (DMSO) was the best solvent to dissolve all compounds
studied in this work, we decided to carry out the analysis of
the complete series in this medium. Cyclic voltammetry of
CABzA (10 in Fig. 2) in DMSO was also performed in order to
contrast the effect of monomeric versus dimeric topologies on
the electrochemical behaviour of compounds 8–10. In
addition, the radical scavenging capacity of the symmetrical
compounds was evaluated using the DPPH assay. The obtained
data were analyzed and compared with the oxidation potential
values in order to investigate whether there is a relationship
between them.
2. Results and discussion
2.1 Electrochemical study of diesters 6 and 7
Fig. 3 shows the voltammograms of dimeric esters 6 and 7 (3a
and 3b, respectively). It can be observed that there are no
appreciable differences between them. Typical chemically ir-
reversible oxidation waves (I) and the corresponding broadened
reduction waves (II) appear in both cases. The values registered
for these peaks are the following: 0.84 and 0.14 V vs. SCE in
the case of 6, and 0.89 and 0.08 V vs. SCE for compound 7. A
very similar electrochemical behaviour was observed before for
CA and CAPE employing the same experimental conditions,
which allows us to infer that these dimeric compounds follow
Fig. 3 Cyclic voltammetry of (a) compound 6 and (b) compound 7,
2 mM in DMSO + 0.1 M n-Bu4NPF6, on a glassy carbon electrode (3 mm ϕ)
at 0.1 V s−1
.
5982 | Org. Biomol. Chem., 2014, 12, 5981–5989
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
an ECE mechanism during their electrochemical oxidation. electrochemical response is recovered after the electrode
This mechanism involves the loss of one electron followed by a surface is polished. This behaviour suggests that parallel to
fast deprotonation and the subsequent loss of a second elec- the oxidation mechanism described before, an electrografting
tron from each electroactive catechol moiety. A further depro- process also takes place during the experiment. Interestingly,
tonation can be the final step in the oxidation of this kind of the electrochemical response of compound 6 remains almost
compounds, since o-quinone is the product suggested by spec- the same after several cycles, which suggests that the connec-
troscopic studies after the oxidation of CA, which has been tor could exert an influence on the ability of these dimeric
reported by other authors.40 However, dimeric topology pro- esters to interact with the surface of the glassy carbon elec-
duces a slight increment in the oxidation potential, since the trode (GCE). In order to know more details about the electro-
corresponding values reported for CA and CAPE were 0.78 and chemical processes occurring during the oxidation and the
0.73 V vs. SCE, respectively, under the same experimental electrografting process of compound 7, we carried out a series
conditions.38
of additional experiments, which are described below.
In addition, in the case of compound 6, the anodic peak
In the first place, after the complete blocking of the elec-
current is linearly dependent on the square root of the scan trode surface, the electrochemical reduction of the solution of
rate, which indicates that transport of this compound toward compound 7 was performed immediately. As is shown in
the electrode surface is diffusion controlled (see Fig. S1 in the Fig. 4c, a cathodic peak at −1.92 V vs. SCE was registered,
additional ESI†).41 On the other hand, when the scan rate is which corresponds to the reduction of the carbonyl of the
increased, the peak potential is anodically shifted, presenting ester group present in molecules of the sample in solution.43
a linear variation with respect to the logarithm of the scan In addition, a further oxidation cycle was also carried out in
rate. The transition from a mechanism, the rate controlled by the same cell and the electrochemical response is shown in
the follow-up chemical reaction, to a mechanism where the Fig. 4d. The recovery of peak I is clear, which implies that, at
electron transfer is the rate determining step has been well least in part, a desorption phenomenon occurred at cathodic
characterized.42 Thus, at 25 °C, this transition presents a vari- potentials as a consequence of an oxidation–reduction process
ation of δEp/δlog ν from 29.6 to 59.2 mV dec−1. Our experi- between the modified GCE surface and molecules of com-
mental results are in an intermediate situation (40 mV dec−1), pound 7 in solution.
which suggests that the initial electron and proton transfer
reactions are stepwise and they occur under a mixed kinetic (Fig. 5a and b), the electrode was rinsed several times with
control. acetone, followed by sonication in pure DMSO and then trans-
In a new experiment and after the electrografting process
Such as in the case of monomeric and dimeric FA amides ferred to a cell containing only the solution of the supporting
studied before by our group, the current intensity of the oxi- electrolyte. No peak was registered during the electrochemical
dation peak of compound 7 decreased until it disappeared reduction of the modified carbon electrode (Fig. 5c), which
completely after 25 cycles (see Fig. 4b). Nevertheless, the allows us to infer that the pending chain is not electroactive.
Fig. 5 Cyclic voltammetry in DMSO + 0.1 M n-Bu4NPF6, on a glassy
Fig. 4 Cyclic voltammetry in DMSO + 0.1 M n-Bu4NPF6, on a glassy
carbon electrode (3 mm ϕ) at 0.1 V s−1 of: (a) compound 7, 2 mM;
(b) cycle 25; (c) cyclic voltammetry to the cathodic direction of blocked
GCE into a solution of 7; and (d) further cyclic voltammetry to the
anodic direction after the procedure described in c.
carbon electrode (3 mm ϕ) at 0.1 V s−1 of: (a) compound 7, 2 mM;
(b) cycle 25; (c) cyclic voltammetry to the cathodic direction of blocked
GCE into a solution of the supporting electrolyte; and (d) cyclic voltam-
metry to the anodic direction of GCE after the procedure described in c,
employing a fresh solution of compound 7.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5981–5989 | 5983

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 6 Cyclic voltammetry of compounds 8 (a) and 9 (b), 2 mM in
DMSO + 0.1 M n-Bu4NPF6, on a glassy carbon electrode (3 mm ϕ)
Scheme 1 (a) Electrografting process taking place during oxidation of
compound 7. (b) Desorption phenomenon of compound 7 at cathodic
potentials.
at 0.1 V s−1
.
behaviour is clearly different, we suspected that two structural
features could be responsible for the oxidation peak II,
observed only in the case of 8 and 9: (1) the presence of an
additional electroactive functional group, such as an amide
group, since its oxidation is well documented in the litera-
ture,44 or (2) an effect of the diamine spacer, which provokes
that caffeoyl units give rise to separate oxidation processes,
similar to those described for bis-ferrocenyl compounds.45 In
order to rule out any of these hypotheses, we decided to carry
out cyclic voltammetry of CABzA (compound 10) in DMSO
(electrochemical behaviour of this monomeric amide in aceto-
nitrile was reported before by our group38). We rationalized
Immediately after that, the electrode was transferred to a cell
containing a fresh solution of compound 7, and a further oxi-
dation cycle was carried out. Interestingly, peak I was not
recovered, which means that the surface of the electrode was
still blocked (Fig. 5d), and that the desorption phenomenon
described above is only possible at cathodic potentials and in
the presence of molecules of compound 7 in the solution.
Scheme 1 summarizes the mechanism that could explain the
electrochemical behaviour of compound 7.
2.2 Electrochemical study of dimeric amides 8 and 9
The electrochemical response of CA dimeric amides differs that two peaks must be observed if, in addition to the caffeoyl
from the data registered for both esters described above. moieties, the amide group undergoes an oxidation process
Fig. 6a and b show the voltammograms of compounds 8 and during the electrochemical experiment on compound 10. On the
9, respectively. As can be observed, a prewave and two peaks, other hand, if only one peak is observed in the voltammogram
labeled as I′, I, and II respectively, appear in the anodic direc- of 10 that would mean that caffeoyl moieties undergo separate
tion in both compounds at 0.32, 0.84, and 1.14 V vs. SCE for oxidation processes, in the case of dimeric amides.
the dimeric amide 8, whereas 0.08, 0.51, and 1.13 V vs. SCE
Fig. 7 shows the electrochemical behaviour of amide 10,
were the corresponding values registered during the electro- which is the monomeric analogue of compound 8. Only a
chemical oxidation of compound 9. In addition, a broad wave sharp oxidation peak I at 0.70 V vs. SCE and a broad reduction
III was observed at cathodic potentials for compound 8 at wave II at 0.18 V vs. SCE are observed. Such as in the case of
0.08 V vs. SCE, whereas two waves at 0.17 and −0.08 V vs. SCE diesters 6 and 7, an ECE mechanism with a further deprotona-
appear in the voltammogram of 9, which were labelled as III tion must be involved in the oxidation of this compound,
and IV, respectively. From these data, it is clear that at least since its voltammogram is very similar to those reported
two main oxidation phenomena occur during the experiment before for CA and CAPE in the same solvent.38 More interest-
in both cases, and also that the identity of the connector exerts ing is the effect of topology on the electrochemical response of
a type of influence on the electrochemical behaviour of this monomeric versus dimeric amides. This result confirms that
kind of compounds, since prewave I′ and peak I in compound caffeoyl moieties in 8 and 9 undergo separate oxidation pro-
8 are shifted to more anodic potentials with respect to the cesses, as a consequence of the effect of diamine connectors.
corresponding values registered in the voltammogram of com- Looking at Fig. 6a and b, it is evident that insertion of two
pound 9.
Since compounds 6–9 share dimeric topology with the nylenediamine) makes the oxidation of the first caffeoyl
same electroactive caffeoyl moieties but their electrochemical moiety in compound
more difficult with respect to
methylene units in the spacer (m-xylylenediamine vs. 1,3-phe-
8
5984 | Org. Biomol. Chem., 2014, 12, 5981–5989
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 8 Cyclic voltammetry of compounds
employing Eλ = 0.90 and 0.65 V vs. SCE respectively, in DMSO +
0.1 M n-Bu4NPF6, on a glassy carbon electrode (3 mm ϕ) at 0.1 V s−1
8 (a) and 9 (b), 2 mM
Fig. 7 Cyclic voltammetry of compound 10,
2 mM in DMSO +
0.1 M n-Bu4NPF6, on a glassy carbon electrode (3 mm ϕ) at 0.1 V s−1
.
.
efficient. The difference in the behaviour of the two types of
hydroxycinnamic derivatives could be related to the fact that
after the first steps (EC) involved in the oxidation of both of
them, a further delocalization of the unpaired electron of the
radical toward the lateral chain occurs in a favourable way in
the case of FA amides, which reacts with the electrode surface,
as proposed by us in previous work,39 whereas formation of
o-quinone is preferred in the case of the analogous CA deriva-
tives, through the loss of a second electron and a subsequent
deprotonation.
compound 9, and that electronic effects could account for this
phenomenon. In addition, the difference between anodic
potentials of peaks I and II (ΔEp) decreases in a very important
way, also as a consequence of this structural change (620 mV
and 300 mV in the case of 9 and 8, respectively).
On the other hand, very modest changes in the intensities
of current peaks or in anodic potential values were observed
after several cycles for both dimeric amides. Nevertheless,
prewave I′ of both compounds decreases very slightly after
cycling the electrode potential between the initial potential
and the foot of peak I, which is consistent with the occurrence
of a modest adsorption phenomenon during the experiments
with solutions of compounds 8 and 9.46–48 In order to know
more about the relationship between the anodic peaks I and II
and the cathodic peak III observed in the voltammograms of
symmetrical amides, we carried out additional electrochemical
experiments, which are shown in Fig. 8. Inversion potentials
(Eλ) at 0.90 and 0.65 V vs. SCE were employed during the acqui-
sition of voltammograms of 8 (Fig. 8a) and 9 (Fig. 8b), respect-
ively. As can be observed, peak III remains in both cases,
which is evidence that it is coupled to the oxidation peak I. In
addition, in the case of compound 8, peak III becomes two
peaks, labelled as IIIa and IIIb in Fig. 8a. It should be men-
tioned that linear relationships were obtained between current
peaks of both IIIa and IIIb and the square root of the scan
rate. From this analysis, we can deduce that these peaks corres-
pond to diffusion waves (see Fig. S2 in additional ESI†).41
The behaviour of dimeric amides studied here contrasts
with their analogues derived from FA reported before by our
group, which after two or three cycles block, in a very efficient
way, the electrode surface when the experiment is performed
in acetonitrile. From these results it seems that electroactive
guaiacol moieties, present in FA amides, contribute in a deci-
sive manner to their ability to react with the electrode, whereas
catechol units, present in compounds 8 and 9, are less
2.3 Oxidation potentials versus free radical scavenging
activity (FRSA)
Electrochemical parameters such as the oxidation potential,
the number of electrons, and the electron transfer rate have
been used to evaluate the oxidant capabilities of organic com-
pounds.49,50 Particularly, oxidation potentials have been used
to study the electron-donating capacity of molecules and as an
indicator of their radical scavenging ability, since it has been
shown that compounds with a less positive oxidation potential
possess higher radical scavenging activities.32,38 The DPPH
assay is an easy test commonly employed to evaluate FRSA of
phenolic compounds, hydroxycinnamic acids and related
systems. In order to see whether there is a relationship
between the oxidation potential and the FRSA of symmetrical
CA derivatives reported here, we carried out a DPPH assay with
four new compounds.
The oxidation potentials of compounds 6–9 are shown in
Table 1. CAPE was also included in the last entry as a refer-
ence, due to the fact that this compound has been recognized
by its high antioxidant capacity and it is structurally related to
the new dimeric CA derivatives. Since two waves (I and II) were
observed for compounds 8 and 9 in the anodic direction, two
values appear in Table 1 for each symmetrical amide. However,
we take into account only the lower oxidation potentials of
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5981–5989 | 5985

View Article Online
Paper
Organic & Biomolecular Chemistry
shifts were recorded in δ (ppm) values from CD3OD or d6-DMSO
Table 1 Oxidation potential of CA derivatives and their corresponding
inhibition percentage of 2,2’-diphenyl-1-picrylhydrazyl (DPPH) radical
1
(δ 3.31 and 2.50 ppm for H, and δ 49.0 and 39.4 ppm for 13C
respectively, based on the middle peak of the solvents). Signal
patterns are indicated as s, singlet; d, doublet; dd, double
doublet; t, triplet; m, multiplet and br, broad signal. Coupling
constants (J) are given in hertz. HR-ESI-TOF MS spectra were
measured on a Gi969A Agilent spectrometer. Cyclic voltamme-
try experiments were performed on a potentiostat PGZ-301
(Radiometer Copenhagen) with positive feedback resistance
Inhibition (%)
Compound
Ep/V vs. SCEa
( SD) DPPHb
6
7
8
0.84
0.89
0.84
1.14
0.51
1.13
0.73c
70.14 ( 0.17)
94.07 ( 0.06)
65.80 ( 0.55)
9
40.57 ( 0.92)
55.72 ( 0.94)d
compensation.
A spectrophotometer Multiskan Spectrum
CAPE
(Thermo Electron Corporation) was employed to carry out the
DPPH assay.
a Oxidation potentials of caffeic acid derivatives were obtained at the
same concentration (2 mmol L−1) and at 0.1 V s−1 in DMSO. b The
values represent means S.D. from three experiments and indicate the
percentage of inhibition for solutions 35 μmol L−1 of CA derivatives.
c Oxidation potential of CAPE reported before by us.38 This value was
observed under the same experimental conditions employed to study
3.2 Chemicals
All chemical reagents and tetrabutylammonium hexafluoro-
phosphate 99% (n-Bu4NP6) were purchased from Sigma-
Aldrich and were used without purification. Dimethylform-
amide (DMF) and CH2Cl2 were of reagent grade and were dis-
tilled before use. DMSO (content of H2O < 0.1%) spectra grade
(Merck Uvasol) was used as the solvent during the electro-
chemical experiments, whereas ethanol chromasolv® was
employed to perform the DPPH assay. Products were purified
by flash column chromatography on silica gel 230–400 mesh
using mixtures of EtOAc–hexanes as eluents.
6–9. d Inhibition percentage evaluated at 35 μmol L−1
.
both compounds to make a comparison with the inhibition
percentage of the DPPH radical.
As can be observed, compound 9 has the lower oxidation
potential of the complete series (0.51 V vs. SCE), whereas the
symmetrical ester 7 showed the higher value (0.89 V vs. SCE).
On the other hand, the anodic peaks of compounds 6 and 8
were registered at the same potential (0.84 V vs. SCE). Never-
theless three of the four compounds exhibited higher oxi-
dation potentials with respect to CAPE (0.73 V vs. SCE);
oxidation of compound 9 occurred at less anodic values. In a
different way from the behaviour observed before by our group
in the case of monomeric FA and CA derivatives,38 there is no
clear relationship between the oxidation potential of the new
symmetrical compounds reported here and the corresponding
inhibition percentages of the DPPH radical.
Compound 7 showed the higher inhibition percentage
(94.07%), whereas the symmetrical amide 9 showed the lower
one (40.57%). In addition, compounds 6 and 8 exhibited
similar inhibition percentages (70.14 and 65.80%, respect-
ively). From the data shown in Table 1, it is clear that the pres-
ence of two caffeoyl moieties in the dimeric compounds does
not have a twofold effect on their capacity to react with the
DPPH radical. Nevertheless, it seems that molecular flexibility
could play the main role in the FRSA of this type of compound,
since the compound with the aliphatic longer connector
showed a higher inhibition percentage of the DPPH radical (7),
whereas the compound with the more rigid aromatic connec-
tor showed a lower value (9).
3.3 Synthesis of esters 6 and 7
The dimeric esters were prepared following the procedure
reported before by Son51 with some modifications. In a
250 mL flask fitted with a dropping funnel and a stir bar,
8.88 mmol (1.600 g) of caffeic acid, 15 mL of DMF, and 1.4 mL
of 25% NaOH solution (8.88 mmol) were placed. After that,
2.77 mmol of either trans-1,4-dibromo-2-butene or 1,8-di-
bromooctane previously dissolved in 6 mL DMF were added drop-
wise through the dropping funnel and the resulting solution
was stirred for seven days. The crude from the reaction was
washed with 30 mL of water at 0 °C and the product was
extracted several times with EtOAc, dried over sodium
sulfate and evaporated under reduced pressure. Purification
of the crude products was accomplished through flash
chromatography.
3.4 Synthesis of amides 8 and 9
In a 250 mL flask with a stir bar, 5.25 mmol (0.945 g) of caffeic
acid, 20 mL of DMF and 5.25 mmol (0.73 mL) of Et3N were
placed. The solution was allowed to stir for 10 minutes at 0 °C,
after that 2.5 mmol of either m-xylylenediamine or 1,3-pheny-
lenediamine was added directly to the solution. About
5.25 mmol (2.32 g) of (benzotriazol-1-yloxy)tris(dimethyl-
amino)phosphonium hexafluorophosphate (BOP) dissolved in
20 mL of CH2Cl2 were added via a syringe and the mixture was
stirred at 0 °C for another 30 minutes. The reactions proceeded
3. Experimental
3.1 Instrumentation
Melting points were determined with Melt-Temp apparatus at room temperature for 48 h, after which CH2Cl2 and DMF
and are not corrected. 1H and 13C spectra were recorded on were removed, and the crude from the reaction was diluted in
Varian Mercury Plus (300 MHz) and Agilent Technologies 400/ 50 mL of EtOAc. The product was filtered, evaporated to
54 Premium Shielded (400 MHz) spectrometers, using deuter- dryness, and then washed with 1 M HCl , water, and an
ated methanol (CD3OD) and d6-DMSO as solvents. Chemical aqueous solution of 1 M NaHCO3. Finally, it was diluted again
5986 | Org. Biomol. Chem., 2014, 12, 5981–5989
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
in EtOAc and dried over sodium sulfate. The solvent was through a salt bridge containing the same supporting electro-
removed in a rotary evaporator and the residue was chromato- lyte concentration as the working solution. All electrochemical
graphed on silica gel employing mixtures of hexanes–EtOAc as experiments were performed at 25 °C under a high purity
eluents with a gradient from 80 : 20 to 50 : 50. Yields are given nitrogen atmosphere.
for isolated products.
3.7 Free radical scavenging activity (DPPH assay)
3.5 Synthesis of monomeric amide 10
The free radical scavenging activity (FRSA) was measured
Compound 10 was prepared by us previously38 following the
according to the following procedure: 100 μL of a DPPH etha-
nolic solution (300 μmol L−1) was mixed with the same volume
of the test compound solution (35 μmol L−1). The reaction mix-
tures were vigorously vortexed for 10 s. The absorbance of the
resulting solutions was read at 520 nm, after 30 min of incu-
bation at room temperature in darkness. The scavenging
activity was determined by comparing the absorbance with
that of the blank containing only the 2,2′-diphenyl-1-picryl-
hydrazyl (DPPH) radical and ethanol. Results were expressed
as inhibition percentage with respect to control values.53 CAPE
was used as a standard antioxidant.
procedure reported by Rajan.52
Compound 6: 1,4-di-O-caffeoyl-trans-2-butenediol (0.23 g,
20.5%). m.p. 227–230 °C. δH(400 MHz; d6-DMSO) 4.64 (4H, m,
–OCH2), 5.91 (2H, m, –CHvCH–), 6.25 (2H, d, J = 15.9 Hz,
–CHCO), 6.73 (2H, d, J = 8.1 Hz, –CHAr), 6.97 (2H, dd, J =
8.2 Hz, J = 2.0 Hz, ArH), 7.02 (2H, d, J = 2.0 Hz, ArH), 7.47 (2H, d,
J = 15.9 Hz, ArH), 9.24 (4H, br, OH). δC(100.5 MHz; d6-DMSO)
63.6, 114.1, 115.3, 116.1, 121.8, 125.9, 128.4, 145.9, 146.0,
148.9, 166.6. HRMS (ESI-TOF) calculated for [C22H20O8Na]+
435.1056, found [M + Na]+ 435.1046.
Compound 7: 1,8-di-O-caffeoyloctanediol (0.06 g, 5.0%).
m.p. (decomposes before melting). δH(400 MHz; d6-DMSO) 1.29
(8H, m, –CH2), 1.59 (4H, m, –CH2), 4.07 (4H, t, J = 6.6 Hz,
–CH2O), 6.21 (1H, d, J = 15.9 Hz, –CHCO), 6.72 (1H, d, J =
8.1 Hz, –CHAr), 6.95 (1H, dd, J = 2.0 Hz, J = 8.1 Hz, ArH), 7.00
4. Conclusions
(1H, d, J = 2.0 Hz, ArH), 7.42 (1H, d, J = 15.9 Hz, ArH), 9.00 New dimeric derivatives of CA, two diesters and two diamides
(2H, s, –OH), 9.44 (2H, s, –OH). δC(100.5 MHz; d6-DMSO) 25.7, were prepared and their electrochemical behaviour was investi-
28.7, 28.9, 64.1, 114.4, 115.2, 116.1, 121.6, 126.0, 145.4, 146.0, gated in DMSO. Although all of them contain two caffeoyl elec-
148.7, 166.9. HRMS (ESI-TOF) calculated for [C26H31O8]+ troactive moieties in their structures, small differences
471.2019, found [M + H]+ 471.2018.
between the connectors provoke interesting changes in the
Compound 8: N,N’-dicaffeoylxylylenediamine (0.64 g, electrochemical responses of this type of compound. Although
25.5%). m.p. 119 °C. δH(300 MHz; CD3OD) 4.46 (4H, s, appreciable differences were not observed from voltammo-
–CH2Ar), 6.39 (2H, d, J = 15.6 Hz, –CHCO), 6.74 (2H, d, J = grams of both diesters, only one of them (compound 7) under-
8.1 Hz, ArH), 6.87 (2H, d, J = 8.1 Hz, ArH), 7.00 (2H, s, ArH), goes an electrografting process during the electrochemical
7.24 (2H, m, ArH), 7.41 (2H, d, J = 15.6 Hz, –CHAr), 7.46–7.55 experiment. Nevertheless, a desorption phenomenon occurs
(1H, m, ArH), 7.72–7.87 (1H, m, ArH). δC(75.5 MHz; CD3OD) through a redox process at cathodic potentials between the
44.3, 114.9, 116.3, 118.1, 122.1, 127.4, 127.7, 128.2, 129.8, molecules of compound 7 in solution and the modified
140.2, 142.4, 146.6, 148.5, 169.0. HRMS (ESI-TOF) calculated electrode.
for [C26H25N2O6]+ 461.1713, found [M + H]+ 461.1707.
Voltammograms of the dimeric amides derived from CA,
Compound 9: N,N′-dicaffeoylphenylenediamine (0.14 g, studied here, resulted in being more complex with respect to
5.8%). m.p. (decomposes before melting). δH(300 MHz; those obtained from dimeric esters. Electronic effects of the
CD3OD) 6.57 (2H, d, J = 15.5 Hz, –CHCO), 6.75 (2H, d, J = diamine connector seem to be related to the fact that caffeoyl
8.1 Hz, ArH), 6.96 (2H, d, J = 8.1 Hz, ArH), 7.08 (2H, s, ArH), moieties suffer from separated oxidation processes in this type
7.26 (2H, m, ArH), 7.37 (1H, m, ArH), 7.45 (2H, d, J = 15.5 Hz, of dimeric amide. In contrast to their FA analogues, which
–CHAr), 7.69 (1H, m, ArH). δC(75.5 MHz; CD3OD) 112.4, 115.2, have been studied by our group before, CA dimeric amides do
116.5, 118.6, 122.5, 125.2, 128.1, 130.1, 140.4, 143.6, 146.9, not interact in an appreciable way with the electrode surface.
149.1, 167.3. HRMS (ESI-TOF) calculated for [C24H21N2O6]+ The difference in the behaviour of FA and CA diamides could
433.1400, found [M + H]+ 433.1388.
be related to the fact that formation of o-quinone is mainly
favoured during electrochemical oxidation of CA derivatives,
whereas delocalization of the unpaired electron of the result-
3.6 Voltammetry experiments
A Pyrex glass three-electrode cell was used in the electrochemi- ing radical toward the lateral chain could occur in the case of
cal experiments. The working electrode was a 3 mm diameter FA amides, which reacts with the electrode surface. Taking all
glassy carbon disk (Sigradur G from HTW, Germany). This these into account, we can conclude that although the rep-
electrode was polished with 0.4 μm alumina powder and resentative electrochemical oxidation mechanism of hydroxycin-
rinsed in an ultrasound bath with distilled water and ethanol namic acids CA and FA and their derivatives involves ECE
before each run. A platinum mesh was used as a counter elec- processes, small changes in the structure of dimeric com-
trode and a saturated calomel electrode (SCE) as a reference pounds exert interesting effects on their electrochemical
electrode. The reference electrode was connected to the cell behaviour.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5981–5989 | 5987

View Article Online
Paper
Organic & Biomolecular Chemistry
On the other hand, a clear relationship between the oxi- 14 J. Hernández, F. Goycoolea, J. Quintero, A. Acosta,
dation potential and the inhibition percentage of DPPH was
not observed for the symmetrical CA derivatives studied here.
M. Castañeda and Z. Domínguez, Planta Med., 2007,
73(14), 1469.
However, the molecular flexibility seems to play an important 15 T. Usia, A. J. Banskota, Y. Tezuka, K. Midorikawa,
role in the FRSA of this type of compound, since the com- K. Matsushige and S. J. Kadota, Nat. Prod., 2002, 65, 673.
pound with the aliphatic longer connector showed the highest 16 T. Nagaoka, A. J. Banskota, Y. Tezuka, K. Midorikawa,
inhibition percentage of the DPPH radical (7), whereas the
compound with the more rigid aromatic connector showed the
K. Matsushige and S. Kadota, Biol. Pharm. Bull., 2003,
26(4), 487.
lowest value (9). Considering that in the last few years sym- 17 S. Sopheak and B. A. Lewis, J. Agric. Food Chem., 2002, 50,
metrical derivatives of hydroxycinnamic acids have attracted 468.
the attention of research groups as potential therapeutic 18 K. Bratt, K. Sunnerheim, S. Bryngelsson, A. Fagerlund,
agents, it should be kept in mind that the type of connector is
an important key to take into account in the molecular design.
L. Engman, R. E. Andersson and L. H. Dimberg, J. Agric.
Food Chem., 2003, 51, 594.
19 H. A. Pedersen, S. K. Steffensen, C. Christophersen,
A. G. Mortensen, L. N. Jørgensen, S. Niveyro, R. M. de
Troiani, J. Rodríguez-Enríquez, A. P. Barba-de la Rosa and
I. S. Fomsgaard, J. Agric. Food Chem., 2010, 58, 6306.
Acknowledgements
This work was supported by CONACyT-Mexico grants 134275 20 R. Jaiswal, T. Sovdat, F. Vivan and N. Kuhnert, J. Agric. Food
and COVECyT-Mexico FOMIX VER-2009-C03-127523. The Chem., 2010, 58, 5471.
authors thank Dr Felipe González (Departamento de Química, 21 N. X. Nhiem, K. C. Kim, A.-D. Kim, J. W. Hyun, H. K. Kang,
CINVESTAV) for access to the facilities used to acquire the
mass spectroscopy data of the compounds reported here. We
are very grateful to Dr Myrna H. Matus for her comments.
P. V. Kiem, C. V. Minh, V. K. Thu, B. H. Tai, J. A. Kim and
Y. H. Kim, J. Asian Nat. Prod. Res., 2011, 13(1), 56.
22 Y. Lu, X. Li, H. Mu, H. Huang, G.-P. Li and Q. Hu, J. Braz.
Chem. Soc., 2012, 23(4), 656.
23 C. A. Williams, F. Goldstone and J. Greenham, Phyto-
chemistry, 1996, 42(1), 121.
24 P. J. King, G. Ma, M. Miao, Q. Jia, B. R. Mc Dougall,
M. G. Reinecke, C. Cornell, J. Kuan, T. R. Kim and
W. R. Robinson, J. Med. Chem., 1999, 42, 497.
References
1 R. Santiago, R. Malvar, M. Baamonde, P. Revilla and
X. Souto, J. Econ. Entomol., 2005, 98(4), 1349.
2 H. F. Harrison, J. K. Peterson, M. E. Snook, J. R. Bohac and 25 Z. Lin, N. Neamati, H. Zhao, Y. Kiryu, J. A. Turpin,
D. M. Jackson, J. Agric. Food Chem., 2003, 51, 2943.
3 P. Mattila and J. Kumpulainen, J. Agric. Food Chem., 2002,
50, 3660.
C. Aberham, K. Strebel, K. Kohn, M. Witvrouw,
C. Pannecouque, Z. Bebyser, E. De Clercq, W. G. Rice,
Y. Pommier and T. R. Burke, J. Med. Chem., 1999, 42, 1401.
4 S. Sellapan, C. C. Akoh and G. Krewer, J. Agric. Food Chem., 26 R. A. Reinke, P. J. King, J. G. Victoria, B. R. McDougall,
2002, 50, 2432.
G. Ma, Y. Mao, M. G. Reinecke and W. E. Robinson, J. Med.
Chem., 2002, 45, 3669.
27 D. C. Crosby, X. Lei, C. G. Gibbs, B. R. Mc Dougall,
W. E. Robinson and M. G. Reinecke, J. Med. Chem., 2010,
53, 8161.
5 H. Shi, L. Dong, J. Jiang, J. Zhao, G. Zhao, X. Dang, X. Lu
and M. Jia, Toxicology, 2013, 303, 107.
6 S.-H. Kwon, H.-K. Lee, J.-A. Kim, S.-I. Hong, H.-C. Kim,
T.-H. Jo, Y.-I. Park, C.-K. Lee, Y.-B. Kim, S.-Y. Lee and
C.-G. Jang, Eur. J. Pharmacol., 2010, 649, 210.
28 T. T. Charvart, D. J. Lee, W. E. Robinson and
A. R. Chamberlin, Bioorg. Med. Chem., 2006, 14, 4552.
7 N. Cinkilic, S. K. Cetintas, T. Zorlu, O. Vatan, D. Yilmaz,
T. Cavas, S. Tunc, L. Ozkan and R. Bilaloglu, Food Chem. 29 B. D. Whitaker and J. R. Stommel, J. Agric. Food Chem.,
Toxicol., 2013, 53, 359. 2003, 51, 3448.
8 L. Ji, P. Jiang, B. Lu, Y. Sheng, X. Wang and Z. Wang, 30 M. Font, E. Ardaiz, L. Cordeu, E. Cubedo, J. García-Foncil-
J. Nutr. Biochem., 2013, 24, 1911.
las, C. Sanmartin and J.-A. Palop, Bioorg. Med. Chem., 2006,
14, 1942.
31 A. A. Reinke and J. E. Gestwicki, Chem. Biol. Drug Des.,
2007, 70, 206.
32 A. Masek, E. Chrzescijanska and M. Zaborski, Electrochim.
Acta, 2013, 107, 441.
9 Y. Li, W. Shi, Y. Li, Y. Zhou, X. Hu, C. Song, H. Ma, C. Wang
and Y. Li, Environ. Toxicol. Pharmacol., 2008, 26, 13.
10 J.-G. Xu, Q.-P. Hu and Y. Liu, J. Agric. Food Chem., 2012, 60,
11625.
11 R. Domitrović, M. Škoda, V. V. Marchesi, O. Cvijanocić,
E. P. Pugel and M. B. Štefan, Food Chem. Toxicol., 2013, 51, 33 M. A. N. Manaia, V. C. Diculescu, E. de Souza Gil and
370.
A. M. Oliveira-Brett, J. Electroanal. Chem., 2012, 682, 83.
34 E. de Souza Gil, T. A. Enache and A. M. Oliveira-Brett,
Comb. Chem. High Throughput Screening, 2013, 16, 92.
12 S.-S. Huang and R.L. Zheng, Cancer Lett., 2006, 239, 271.
13 F. Rossi, V. Jullian, R. Pawlowiez, S. Kumar-Roiné,
M. Hadd, H. T. Darius, N. Gaertner-Mazouni, M. Chinain 35 F. M. Roleira, C. Siquet, E. Orrù, E. M. Garrido, J. Garrido,
and D. Laurent, J. Ethnopharmacol., 2012, 143, 33.
N. Milhazes, G. Podda, F. Paiva-Martins, S. Reis,
5988 | Org. Biomol. Chem., 2014, 12, 5981–5989
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
R. A. Carvalho, E. J. Tavares da Silva and F. Borges, Bioorg. 44 N. L. Weinberg and H. R. Weinberg, Chem. Rev., 1968, 68, 449.
Med. Chem., 2010, 18, 5816.
45 G. Ferguson, C. Glidewell, G. Opromolla, C. M. Zakaria and
36 A. Gaspar, E. M. Garrido, M. Esteves, E. Quezada,
P. Zanello, J. Organomet. Chem., 1996, 517, 183.
N. Milhazes, J. Garrido and F. Borges, Eur. J. Med. Chem., 46 R. H. Wopschall and I. Shain, Anal. Chem., 1967, 39, 1514.
2009, 44, 2092.
37 A. Barzegar, Food Chem., 2012, 135, 1369.
47 M. Galicia and F. J. González, J. Electrochem. Soc., 2002,
149, D46.
38 M. Salas-Reyes, J. Hernández, Z. Domínguez, F. González, 48 L. S. Hernández-Muñoz, N. A. González-Fuentes, B. R. Díaz-
P. D. Astudillo, R. E. Navarro, E. Martínez-Benavidez,
C. Velázquez-Contreras and J. S. Cruz-Sánchez, J. Braz.
Chem. Soc., 2011, 22, 693.
Sánchez, R. Fragoso-Soriano, C. Vázquez-López and
F. J. González, Electrochim. Acta, 2012, 63, 287.
49 M. Born, P.-A. Carrupt, R. Zini, F. Brée, J.-P. Tillement,
K. Hostettmann and B. Testa, Helv. Chim. Acta, 1996, 79,
1147.
50 H. Hotta, S. Nagano, M. Ueda, Y. Tsujino, J. Koyama and
T. Osakai, Biochim. Biophys. Acta, 2002, 1572, 123.
51 S. Son, E. B. Lobkowsky and B. A. Lewis, Chem. Pharm.
Bull., 2001, 46(2), 236.
39 A. Sánchez, R. C. Guillén-Villar, R. Sánchez, R. Vargas,
J. Garza, M. H. Matus, M. Salas-Reyes and Z. Domínguez,
Electrochim. Acta, 2014, 133, 546.
40 R. Petrucci, P. Astolfi, L. Greci, O. Firuzi, L. Saso and
G. Marrosu, Electrochim. Acta, 2007, 52, 2461.
41 A. J. Bard and L. R. Faulkner, in Electrochemical Methods,
Fundamentals and Applications, Wiley & Sons, New York, 52 P. Rajan, I. Vedernikova, P. Cos, V. D. Berghe,
2001, ch. 4 and 6, pp. 151, 231 and 236.
42 L. Nadjo and L. J. M. Savéant, J. Electroanal. Chem., 1973,
48, 113.
43 J. Grimshaw, in Electrochemical Reactions and Mechanisms
in Organic Chemistry, Elsevier, Amsterdam, The Netherlands,
2000, ch. 10, pp. 354.
K. Ausgustyns and A. Haemers, Bioorg. Med. Chem. Lett.,
2001, 11, 215.
53 C. Velazquez, M. Navarro, A. Acosta, A. Angulo, Z. Domínguez,
R. Robles, R. Robles-Zepeda, E. Lugo, F. M. Goycoolea,
E. F. Velázquez, H. Astiazaran and J. Hernández, J. Appl.
Microbiol., 2007, 103, 1747.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5981–5989 | 5989