Antioxidant properties in a non-polar environment of difluoromethyl bioisosteres of methyl hydroxycinnamates

Article abstract of DOI:10.1111/jphp.12507

Objectives Many natural antioxidants have poor pharmacokinetic properties that impair their therapeutic use. For hydroxycinnamic acids (HCAs) and other phenolic antioxidants, their major drawback is their low lipophilicity and a rapid metabolism. The difl

Full text of DOI:10.1111/jphp.12507

Research Paper
Antioxidant properties in a non-polar environment of
difluoromethyl bioisosteres of methyl hydroxycinnamates
a
b
c
b
a
Mario D. Mart ꢀı nez , Lorena Luna , Alvaro Y. Tesio , Gabriela E. Feresin , Fernando J. Dur aꢀ n and
Gerardo Burton
a
a
Departamento de Qu ꢀı mica Org ꢀa nica and UMYMFOR (CONICET-UBA), Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires,
b
c
Buenos Aires, Instituto de Biotecnolog ꢀı a, Facultad de Ingenier ꢀı a, Universidad Nacional de San Juan, San Juan and INQUIMAE (CONICET-UBA),
Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, Buenos Aires, Argentina
Keywords
Abstract
antioxidant activity; bioisosteres;
difluoromethyl group; hydroxycinnamic
acids; lipid peroxidation
Objectives Many natural antioxidants have poor pharmacokinetic properties
that impair their therapeutic use. For hydroxycinnamic acids (HCAs) and other
phenolic antioxidants, their major drawback is their low lipophilicity and a rapid
metabolism. The difluoromethyl group may be considered as a ‘lipophilic hydro-
xyl’ due to its hydrogen bond donor and acceptor properties; this prompted us
to assess it as a bioisosteric replacement of a phenolic hydroxyl for increasing the
lipophilicity of HCAs.
Correspondence
Gerardo Burton, Departamento de Qu ꢀı mica
Org ꢀa nica Facultad de Ciencias Exactas y
Naturales, Universidad de Buenos Aires,
Pabell oꢀ n 2, Ciudad Universitaria,
C1428EGA, Buenos Aires, Argentina.
E-mail: burton@qo.fcen.uba.ar
Methods Six difluoromethyl-substituted methyl cinnamates (4a-c, 5a-c)
related to caffeic acid were synthesized and their antioxidant activity evaluated
by chemical (FRAP, DPPH scavenging, inhibition of b-carotene bleaching, at
Received August 18, 2015
1
–200 lM), electrochemical (differential pulse voltammetry, cyclic voltamme-
try) and cell-based (inhibition of lipid peroxidation in erythrocytes, at 1 and
0 lM) assays.
Accepted November 19, 2015
5
doi: 10.1111/jphp.12507
Key findings Analogues 4a-c and 5a-c were inactive in FRAP and DPPH assays
and only those containing a free phenolic hydroxyl (4a and 5a) exhibited electro-
chemical activity although with high redox potentials. Compounds 4a,b and 5a,b
were active in the inhibition of b-carotene bleaching assay and all analogues
inhibited lipid peroxidation in the human erythrocytes assay.
Conclusions Lipophilic difluoromethyl-substituted cinnamic esters retain radical
scavenging capabilities that prove useful to confer antioxidant properties in a
non-polar environment.
exceeds their capacity, the accumulated highly reactive spe-
cies will damage most cell components and may also dis-
Introduction
Reactive oxygen species (ROS) are produced in living
organisms under normal physiological conditions and
comprise both free radicals as the superoxide anion and
hydroxyl radical and non-radical species as peroxides and
singlet oxygen. Several defence mechanisms are present in
biological systems to eliminate most of these species, as well
as to repair the damage they produce in cells, mostly to
their DNA, but also to proteins and lipids. Thus, endoge-
nous antioxidant compounds as glutathione, or enzyme
systems as catalase or superoxide dismutase, can convert
ROS into harmless compounds. When these defence mech-
anisms are impaired or an excessive production of ROS
rupt cellular signalling. This condition, termed oxidative
stress, is implicated in the pathogenesis of several human
diseases including cancer, certain cardiovascular diseases
and neurodegenerative diseases and may also contribute to
[
1–4]
the ageing process.
Besides the endogenous antioxi-
dants mentioned above, compounds with antioxidant
properties from exogenous sources are normally incorpo-
rated through the human diet, with polyphenols being the
most abundant. Although much controversy exists around
the benefits of incorporating this type of compounds as
dietary supplements in a regular basis, their ability to
counteract the harmful effects of ROS and regulate the
©
2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 68 (2016), pp. 233–244
233

Difluoromethyl bioisosteres of hydroxycinnamates
Mario D. Mart ꢀı nez et al.
physiological defence systems makes them useful candi-
dates for prevention or treatment of oxidative stress-related
boxylic acid moiety into different kinds of esters or
[
14–18]
amides.
However, it has been proposed that the
[
5,6]
diseases.
Phenolic acids are natural hydrophilic antioxidants
derived from benzoic or cinnamic acids that occur in
incorporation of long lipophilic chains into the HCA
molecules could lead to undesirable effects as self-aggre-
gation, reduced mobility or internalization into the lipid
core that may hamper the antioxidant efficacy in a living
[7]
fruits, vegetables, spices and herbs.
Hydroxycinnamic
[
19]
acids (HCAs) and their derivatives have drawn particular
attention, as a large number of beneficial health effects
have been correlated with their ingestion as part of the
human diet. This has led to extensive studies of their
organism.
The difluoromethyl group can be used as a lipophilic
isostere of the hydroxyl group taking advantage of its
[
20,21]
hydrogen bond donor and acceptor capacities.
It
[8,9]
antioxidant properties.
The most common substi-
can also generate radicals by homolytic cleavage of the
C-H bond, as the fluorine atoms stabilize the difluo-
romethyl radicals due to their p-donating ability and
tuted cinnamic acids found in fruits and vegetables are
shown in Figure 1. Most of these compounds also pre-
sent antibacterial, antiviral, antisclerotic and antitumor
participate in the formation of o-quinone methides in
[22–24]
[5]
activities, among others. HCAs can prevent or mini-
mize oxidative damage processes essentially by scaveng-
ing free radical species and/or boosting the endogenous
antioxidant system capacity, by stimulating the synthesis
of endogenous antioxidants. However, the major draw-
back of HCAs and other phenolic compounds lies in
their low lipophilicity and a rapid metabolism that result
in limited ADME properties (absorption, distribution,
basic conditions.
its use as a replacement for a phenolic hydroxyl
However, there are few reports of
[25–29]
and to the best of our knowledge, none pertaining its
radical scavenging properties. We envisaged that this
modification applied to HCAs would render analogues
with enhanced lipophilicity that should retain at least in
part the properties of the parent phenolic compound.
Specifically, we were interested in evaluating to what
extent the difluoromethyl group could mimic the antiox-
idant properties conferred by a phenolic hydroxyl group,
as well as its effect on the antioxidant properties of
neighbouring phenolic hydroxyls. In this work, we
report for the first time the synthesis of difluoromethyl
analogues of simple HCA esters and their antioxidant
properties evaluated in different systems with special
emphasis on the radical scavenging properties and their
performance in non-polar media.
[2,5,10]
metabolism and excretion).
The structure–activity relationships for a large number
of HCAs and synthetic derivatives have been the subject
[8,9,11]
of recent reviews.
The antioxidant activity has been
shown to be strongly dependent on certain structural
features as the unsaturated side chain and is usually
associated with the presence of free phenolic hydroxy
groups. However, a clear distinction can be made
between the direct antioxidant capacity that may be
evaluated in non-biological systems based on redox reac-
tions or radical scavenging processes and the effective
antioxidant activity in whole cells or complex matrices.
The latter is related not only to the redox or radical
scavenging properties but also to lipophilicity and to the
presence of enzymes that can expose masked phenolic
Materials and Methods
Melting points were taken on a Fisher-Johns apparatus
1
13
and are uncorrected. H and C NMR spectra were
measured in a Bruker Avance II 500 NMR spectrometer
(Bruker BioSpin, Karlsuhe, Germany) at 500.13 and
125.72 MHz, respectively. Exact mass spectra (HRMS)
were measured on a Bruker micrOTOF-Q II mass spec-
trometer with positive electrospray ionization. Analytical
thin layer chromatography (TLC) was performed on
precoated silica gel plates (Merck KGaA, Darmstadt,
Germany; F254, 0.2 mm thickness); compounds were
visualized under 254 nm UV light. Flash column chro-
matography (FCC) was performed on silica gel Merck
[12,13]
hydroxyls, as demethylases and esterases.
Increased
lipophilicity that allows the passive crossing of the
blood–brain barrier is particularly important for antioxi-
dants that may be potentially useful in the prevention or
minimization of oxidative neuronal damage in neurode-
[14]
generative diseases. In this respect, most approaches
to improve antioxidant activity of HCAs in vivo have
sought to increase lipophilicity by conversion of the car-
9385 (0.0040–0.0063 mm) eluting with mixtures of ethyl
acetate–hexane of increasing polarity. Homogeneity of
all compounds was confirmed by TLC and NMR. Com-
pounds 6b,c were synthesized from 2-hydroxy-5-iodo-
benzaldehyde, and compounds 9b,c were obtained from
methyl 4-iodosalicylate (see supporting information for
details).
Figure 1 Common hydroxycinnamic acids found in fruits and
vegetables.
2
34
© 2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 68 (2016), pp. 233–244

Mario D. Mart ꢀı nez et al.
Difluoromethyl bioisosteres of hydroxycinnamates
Methyl 3-[(4-(difluoromethyl)-3-hydroxyphenyl)]-
Chemistry
(
E)-propenoate (5a)
Compound 5a was prepared from 5c (0.023 g,
.016 mmol) following the procedure described for 4a.
Recrystallization from n-hexane/chloroform gave 5a
0.016 g, 82%).
Chemical data of synthesized compounds
0
Structures of all compounds were confirmed by NMR and
HRMS (see Appendix).
(
Methyl 3-[(3-(difluoromethyl)-4-methoxyphenyl)]-
Electrochemical measurements
(E)-propenoate (4b)
Electrochemical measurements were carried out on 0.1 mM
solutions of 1-3, 4a-c and 5a-c in phosphate buffer
â
Deoxofluor 50% in toluene (0.061 ml, 0.248 mmol) was
added dropwise to a solution of 6b (0.028 g, 0.124 mmol)
in dry dichloromethane (0.5 ml) under argon. The mixture
was stirred at room temperature for 1.5 h and then perco-
lated through a silica gel pad eluting with dichloromethane.
The percolate was evaporated to dryness and the residue
purified by FCC to give 4b (0.029 g, 96%).
(
0.067 M, pH 7.3) using a homemade two-compartment
Pyrex cell. The working electrode was a GC disk (Bioanalyt-
ical System, Inc., West Lafayette, Indiana, USA, 3 mm
diameter). The counter electrode was a large-area platinum
mesh; reference electrode was Ag/AgCl (3 M KCl). Scan rate
in cyclic voltammograms was 0.050 V/s. For differential
pulse voltammograms, a modulation amplitude of 0.025 V
and a step potential of 0.005 V were used. All experiments
were recorded at 25°C on an AutoLab PGSTAT 30 poten-
tiostat, controlled by GPES 4.9 electrochemical software
Methyl 3-[4-acetyloxy-(3-(difluoromethyl)phe-
nyl)]-(E)-propenoate (4c)
Compound 4c was prepared from 6c (0.072 mg,
â
(
EcoChemie, Utrecht, NL).
0
.289 mmol) and Deoxofluor 50% in toluene (0.142 ml,
0.578 mmol) following the procedure described for 4b
(
0.070 g, 90%).
Antioxidant activity
Free radical scavenger assay
Methyl 3-[(3-(difluoromethyl)-4-hydroxyphenyl)]-
E)-propenoate (4a)
(
Free radical scavenger activity of extracts was assessed by
the fading of a methanolic solution of 1,1-diphenyl-2-
picrylhydrazyl radical, following the procedure of Tapia
To a solution of 4c (0.045 g, 0.167 mmol) in methanol
3.0 ml), conc. H SO (50 ll) was added and the mixture
(
2
4
[30]
et al.
The tested compounds were assayed at concentra-
stirred 2 h at 60°C. The reaction mixture was diluted with
water (15 ml), concentrated under reduced pressure and
extracted with ethyl acetate (2 9 30 ml). The organic layer
was dried with anhydrous sodium sulphate and the solvent
evaporated. The residue was purified by FCC to give 4a
tions of 1, 50, 100 and 200 lM. Activities were evaluated in
6-well microplates in triplicate at 517 nm, using a Labsys-
9
tems microplate reader (Helsinki, Finland). Quercetin was
used as reference compound. The percentage of decol-
oration at each concentration was calculated as follows:
(
0.036 g, 94%).
ꢀ
ꢁ
ðAsample ꢀ AcontrolÞ
DPPH
decoloration (%) ¼ 1 ꢀ
ꢁ 100;
Methyl 3-[(4-(difluoromethyl)-3-methoxyphenyl)]-
E)-propenoate (5b)
A
(
where Asample: sample absorbance; Acontrol: control absor-
bance; ADPPH: DPPH absorbance; values are reported in
terms of IC50 (concentration of the compound that pro-
duces 50% decoloration) as the mean ꢂ standard devia-
tion of three independent assays.
Compound 5b was prepared from 9b (0.028 g,
â
0.127 mmol) and Deoxofluor 50% in toluene (0.094 ml,
0.381 mmol) following the procedure described for 4b
(
0.027 g, 88%).
Methyl 3-[3-acetyloxy-4-(difluoromethyl)phenyl)]-
E)-propenoate (5c)
Ferric reducing antioxidant power (FRAP) assay
(
Ferric reducing antioxidant power was determined by the
ꢀ
3
ꢀ4
Compound 5c was prepared from 9c (0.055 g, 0.22 mmol)
â
direct reduction of [Fe(CN)6] to [Fe(CN)6] measuring
the absorbance of Perl’s Prussian blue complex after addi-
and a solution of Deoxofluor 50% in toluene (0.164 ml,
0.667 mmol) following the procedure described for 4b
3+ [31,32]
tion of excess Fe . The iron reducing power
[
was tested according to Oyaizu.
33]
(
0.044 g, 73%).
Briefly, 100 ll of
©
2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 68 (2016), pp. 233–244
235

Difluoromethyl bioisosteres of hydroxycinnamates
Mario D. Mart ꢀı nez et al.
compounds solution or gallic acid solution (1, 50, 100,
2
controls was 1%. After incubation, cells were exposed to
tert-butyl hydroperoxide (1 mM) for 15 min at 37°C under
vigorous shaking. Lipid peroxidation was determined indi-
rectly by TBARS formation at 540 nm using a Labsystems
microplate reader (Helsinki, Finland). Quercetin (1 and
50 lM) was used as reference antioxidant. Percentages of
inhibition relative to controls were expressed as means ꢂ s-
tandard error of three independent experiments. Statistical
analyses were performed with STATISTICA 6.0 (StatSoft
Inc., Tulsa, Oklahoma, USA) and consisted in one-way
ANOVA followed by Tukey’s multiple comparisons tests.
Differences were regarded as significant at P < 0.05 (bars
with different superscript letters are significantly different
from each other). Before statistical analysis, Q-Q plot and
Shapiro–Wilk test were performed for normality.
Homoscedasticity was assessed with Levene’s test.
00 lM) was added to 250 ll of phosphate buffer (0.1 M,
pH 6.6) and 250 ll of 1% potassium ferricyanide. The mix-
ture was incubated at 50°C for 20 min, and 250 ll of 10%
trichloroacetic acid was added and vortexed for 20 s. The
resulting solution (50 ll) was diluted with distilled water
(
50 ll) in a microplate and 10 ll of 0.1% FeCl added.
3
After 30-min incubation, absorbance was read at 700 nm
using a Labsystems microplate reader (Helsinki, Finland).
The reducing power was determined by linear regression
from a calibration plot and expressed as gallic acid equiva-
lent antioxidant capacity (millimolar concentration of a
gallic acid solution with antioxidant capacity equivalent to
a 1.0 mM solution of compound).
b-Carotene bleaching assay
The b-carotene bleaching assay was conducted according to
the method described by Mogana with modifications
Results
[34,35]
and recommendations by Prieto.
b-carotene (1 mg) in chloroform (5 ml) was mixed with
0 ll of linoleic acid and 250 ll of Tween-20. Chloroform
Briefly, a solution of
Chemistry
Three representative HCA methyl esters, methyl caffeate
(
2
1), methyl ferulate (2) and methyl isoferulate (3), were
was removed under vacuum at 35°C and deionized
water (50 ml) added with vigorous stirring to give the
b-carotene–linoleic acid emulsion that was used immedi-
ately. Twenty microlitre of methanolic solutions of the
compounds (1, 5, 10, 20, 50, 100 lM) was plated out in
duplicate in a 96-well microtiter plate and 180 ll of b-
carotene–linoleic acid emulsion added to each of the test
samples. Absorbance was measured at 470 nm against a
blank (linoleic acid emulsion without b-carotene) and a
control (emulsion and 20 ll of methanol) every 5 min for
chosen as templates for the bioisosteric replacement of
hydroxyl by the difluoromethyl group, leading to the fluo-
rinated analogues 4a,b and 5a,b (Figure 2). The acetylated
derivatives 4c and 5c were also selected, considering that
they could act as precursors of the free phenols in biological
systems.
The 3’-difluoromethyl analogues 4a-c were obtained
from salicylaldehyde as depicted in Scheme 1. Iodination
with iodine monochloride in glacial acetic acid gave
[
36]
5-iodo-2-hydroxybenzaldehyde
that was coupled with
4
h at 45°C using a PHERAstar F5 multiplate spectropho-
tometer reader (BMG Labtech, Cary, North Carolina, USA)
with stirring. The antioxidant activity was evaluated in
terms of bleaching of b-carotene as:
ðA0 ꢀ AtÞ
%
b-carotene bleached ¼
ꢁ 100;
A0
where A : initial absorbance and A : absorbance at the dif-
t
0
ferent incubation times for the test samples and control.
Lipid peroxidation in human erythrocytes
Studies on erythrocytes lipid peroxidation were conducted
as described by Tapia et al.
[30]
Cells were washed three
times in cold phosphate-buffered saline (PBS) by centrifu-
gation at 1200g, suspended in PBS and the density adjusted
to 1 mM haemoglobin. The cell suspension was incubated
with different concentrations (1 and 50 lM) of the test
compounds dissolved in DMSO and PBS for 10 min at
Figure 2 Structures of methyl hydroxycinnamates 1-3 and
difluoromethyl analogues from bioisosteric replacement of the 3’ or
4’ phenolic hydroxyl (4a-c or 5a-c, respectively).
3
7°C. The final concentration of DMSO in samples and
2
36
© 2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 68 (2016), pp. 233–244

Mario D. Mart ꢀı nez et al.
Difluoromethyl bioisosteres of hydroxycinnamates
conditions as described above gave cinnamate 9a in 60%
yield. Conversion of the latter compound to the fluorinated
analogues 5a-c was achieved by the same reaction sequence
1
13
as used for 4a-c. The H and C NMR spectra of com-
pounds 5a-c showed the corresponding triplets of the diflu-
oromethyl group as observed in 4a-c.
Electrochemical measurements
Voltammetric methods have been extensively applied to
characterize a variety of natural phenolic antioxidants and
synthetic antioxidants, mainly to get an insight of the
[
39,40]
underlying mechanism.
voltammetry (DPV) and cyclic voltammetry (CV) to inves-
The use of differential pulse
2
Scheme 1 Reagents and conditions: (i) 1. ICl, AcOH; 2. Pd(AcO) , (o-
tigate the electrochemical behaviour of different kinds of
[41]
Tol)
Ac
MeOH.
3
P, methyl acrylate, TEA, MeCN; (ii) MeI, K
2
CO
3
, Me
2
CO; (iii)
SO
HCA-related antioxidants was recently reviewed.
The
â
2
O, K
2
CO
3
, CH
2 2
Cl ; (iv) Deoxofluor , PhCH
, CH Cl ; (v) H
3 2
2
2
4
,
redox potentials of the synthesized compounds 4a-c and
a-c were measured using differential pulse and cyclic
5
methyl acrylate under Heck reaction conditions to give the
Attempts
voltammetries at physiological pH (7.3) and their electro-
chemical behaviour was compared with that of the hydrox-
ylated analogues 1, 2 and 3 measured under the same
conditions (Figures 3 and 4). The difluoromethyl ana-
logues 4b,c and 5b,c did not show any electrochemical
activity in the potential range studied, due to the absence of
a free phenolic hydroxyl (data not shown).
[37]
methyl cinnamate derivative 6a in 65% yield.
to carry out the deoxofluorination reaction on the latter
compound failed; thus, the phenolic hydroxyl was acety-
lated and the aldehyde group converted to the difluo-
â
romethyl group with Deoxofluor to give the isoferulate
[
38]
1
13
analogue 4c in 84% yield from 6a.
NMR spectra of 4c showed triplets at dH 6.76 ppm
JHF = 55.2 Hz) and at dC 111.5 ppm (JCF = 239 Hz) for
The H and
C
Consecutive cyclic voltammograms of compounds 1-3,
4a and 5a registered in the potential range between ꢀ0.2
and 1.2 V showed a marked decrease in the current signals
in consecutive sweeps. After the fifth successive sweep,
voltammetric signals almost disappeared (data not shown).
It is well known that the oxidation of phenols occurs at
potentials close to 1 V depending on the reaction medium.
This may be represented in principle by an E1C1E2C2
mechanism generating, in some cases, polymeric products
(
the CF H moiety. Mild acid hydrolysis of the phenolic acet-
2
ate gave the difluoromethyl phenol 4a in 79% yield from
6a. Methylation of 6a with methyl iodide followed by reac-
â
tion with Deoxofluor gave the isoferulate analogue 4b
(
91% from 6a).
The 4’-difluoromethyl analogues 5a-c were obtained by a
similar sequence but starting from commercially available
methyl 4-iodosalicylate (Scheme 2). Reduction to the ben-
zyl alcohol 7 with lithium aluminium hydride followed by
oxidation with manganese dioxide gave 4-iodo-2-hydroxy-
benzaldehyde 8. Coupling with methyl acrylate under Heck
[42]
that poison or passivate the working electrode surface.
Thus, in this work, we only analysed the electrochemical
behaviour of the first voltammogram. Compounds 4a
and 5a had higher redox potentials compared to 1-3.
Scheme 2 Reagents and conditions: (i) LiAlH
Me CO; (v) Ac O, K CO , CH Cl
4
, THF; (ii) MnO
â
2
, CH
2
Cl
2
; (iii) Pd(AcO)
2 3 2 3
, (o-Tol) P, methyl acrylate, TEA, MeCN; (iv) MeI, K CO ,
2
2
2
3
2
2
; (vi) Deoxofluor , PhCH
3
-CH
2
Cl ; (vii) H
2
SO , MeOH.
2 4
©
2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 68 (2016), pp. 233–244
237

Difluoromethyl bioisosteres of hydroxycinnamates
Mario D. Mart ꢀı nez et al.
Figure 3 Cyclic voltammograms for 0.1 mM solutions of the tested compounds (scan rate: 50 mV/s). (a) 1 (—), 2 (. . ..), 3 (‑‑‑); (b) 4a (—),
a (‑‑‑). Phosphate buffer (pH 7.3) was used as the supporting electrolyte.
5
[
43]
Comparison of regioisomers 4a and 5a showed the same
trend found for regioisomers 2 and 3 albeit enhanced with
a lower redox potential for the isomer with the hydroxyl
group in the para position relative to the acrylate side
chain. Both compounds exhibited two redox potential val-
ues that may be ascribed to adsorbed and free forms.
as in the original ferricyanide method (FRAP).
Other
methods that use acidic solutions may not reflect the true
behaviour of physiologically important antioxidants in a
cellular assay or in vivo. Radical scavenging capacity was
first evaluated with the 1,1-diphenyl-2-picrylhydrazyl
(DPPH) radical assay. Being these predominantly single
electron transfer (SET)-based methods, they may be used
[
41]
to compare with the electrochemical data. Compounds
a-c and 5a-c were inactive in the FRAP and DPPH assays
Antioxidant activity
4
(see supporting information, Table S1).
The antioxidant capacity profiles for the synthesized com-
pounds were evaluated by four representative assays. For
the total antioxidant capacity (TAC) levels, we used an iron
To evaluate the effect of the isosteric replacement in a
biological environment where the capacity to scavenge free
(
III)-based TAC assay at a pH close to physiological values,
Figure 5 Inhibition of lipid peroxidation in human erythrocytes at 1
and 50 lM of compounds 1-3, 4a-c and 5a-c. Quercetin (Q) was used
as positive control. Means ꢂ SE from three independent experiments
are shown. Differences were determined by one-way ANOVA fol-
lowed by Tukey’s multiple comparisons tests. At each concentration,
bars with different superscript letters are significantly different from
each other (P < 0.05). If one superscript letter is shared between bars,
then no significant differences were found.
Figure 4 Differential pulse voltammograms for 0.1 mM solutions of
the tested compounds (scan rate: 5 mV/s): 1 (—) E = 191 mV, 2 (. . ..
= 370 mV, 3 (—) E = 478 mV, 4a (‑.‑.‑) E = 582 and 761 mV
and 5a (‑..‑..‑) E = 748 and 836 mV. Phosphate buffer (pH 7.3) was
used as the supporting electrolyte.
p
)
E
p
p
p
p
2
38
© 2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 68 (2016), pp. 233–244

Mario D. Mart ꢀı nez et al.
Difluoromethyl bioisosteres of hydroxycinnamates
Figure 6 Inhibition of b-carotene bleaching at 1 (○), 5 (■), 10 (M), 20 (●), 50 (□) and 100 (▲) lM of methanolic solutions of compounds 1 (a),
a (b) and 5a (c). Methanol (♦) was used as control. All assays were carried out in duplicate.
4
Figure 7 Inhibition of b-carotene bleaching at 1 (○), 5 (■), 10 (M), 20 (●), 50 (□) and 100 (▲) lM of methanolic solutions of compounds 2 (a),
(b), 5b (c) and 4b (d). Methanol (♦) was used as control. All assays were carried out in duplicate.
3
radicals by hydrogen atom transfer (HAT) plays an impor-
tant role, we used a cellular assay based on the inhibition of
relative to the acrylate chain. The same tendency was
observed between compounds 5b,c and 4b,c, suggesting a
stabilization from the side chain although differences were
not statistically significant. Thus, despite their poor perfor-
mance in the single electron transfer-based methods, all
difluoromethyl analogues equalled or bettered the activity
of compounds 1-3 in the cell-based assay at the lower con-
centration. These results prompted us to determine the
ability of the difluoromethyl analogues to scavenge free
radicals by hydrogen donation in a direct chemical assay.
The b-carotene bleaching assay specifically evaluates the
antioxidant capacity by the hydrogen atom transfer (HAT)
[30,44]
lipid peroxidation in human erythrocytes.
All com-
pounds were active in this assay displaying significant activ-
ity at 1 and 50 lM concentration (Figure 5). The activity of
the fluorinated compounds 4a-c and 5a-c did not differ sig-
nificantly from the reference compound quercetin (Q) at
both concentrations. In the case of compounds 4a and 5a
with a free phenolic hydroxyl, both compounds had a sig-
nificantly better performance at 1 lM compared with their
bioisostere methyl caffeate (1); however, there was no sig-
nificant difference between the regioisomers. Isomers 2 and
[
34,45]
3
with a phenolic hydroxyl and a methoxy substituent also
method, in a non-polar environment.
Figure 6 shows
had the same activity, although at 50 lM compound 2
showed a tendency to increase the lipid peroxidation inhi-
bition consistent with the para position of the hydroxyl
the dose–response effects of caffeic acid (1) and analogues
4a and 5a. Replacing either hydroxyl of 1 by a difluo-
romethyl reduced the antioxidant activity although 4a, with
©
2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 68 (2016), pp. 233–244
239

Difluoromethyl bioisosteres of hydroxycinnamates
Mario D. Mart ꢀı nez et al.
the hydroxyl in the para position relative to the acrylate
side chain, was more active than its regioisomer. In the case
of the methoxylated analogues 4b and 5b, both compounds
were effective in inhibiting the b-carotene bleaching reac-
tion although their activity was reduced compared to their
hydroxyl bearing counterparts 3 and 2, respectively (Fig-
ure 7). As expected, methyl ferulate (2) with the hydroxyl
in the para position relative to the acrylate side chain was
more active than methyl isoferulate (3). Compound 5b
with the difluoromethyl group para to the acrylate side
chain followed the same trend, being more active than 4b
and comparable to compound 3.
in the b-carotene bleaching assay supports the participa-
tion of the difluoromethyl group in a hydrogen atom
transfer mechanism as opposed to a single electron trans-
[
47]
fer mechanism.
The capacity of the difluoromethyl
group to generate a stable carbon radical in lipophilic
environments was evidenced here on the higher activity
of compound 5b, in which this radical may be further
stabilized by delocalization on the acrylate side
[
22,23]
chain.
The same trend (although not statistically sig-
nificant) was observed for the inhibition of lipid peroxi-
dation in erythrocytes, suggesting an active participation
of the difluoromethyl moiety in the antioxidant process.
Based on the behaviour of other resonance-stabilized
carbon-centred radicals, dioxygen addition to the difluo-
romethyl radicals is expected to be reversible. Thus, their
effectiveness as chain-breaking antioxidants should be
enhanced at low oxygen partial pressure (e.g. in mam-
Discussion
Current evidence supports the correlation of the antioxi-
dant activity of HCAs with the presence of one or more
phenolic hydroxyls and structural features capable of sta-
[
47]
malian tissues).
The higher lipophilicity of the difluo-
[41]
bilizing the resulting phenoxyl radicals.
However, the
romethyl-substituted analogues compared to the phenolic
parent compounds (see supporting information, Table S2)
should give an improved membrane permeability and also
play a crucial role due to the ‘polar paradox’, by which
correlation between redox potentials and antioxidant
activity is not always evident when comparing different
compounds. Furthermore, the influence of electron-with-
drawing substituents on the redox potentials of HCAs
lipophilic antioxidants tend to concentrate in the lipid
[48]
[9]
has not been established. As shown in Figure 4, the
presence of the CF H moiety ortho to the phenolic
environment.
These properties would explain at least in
part, why these analogues exhibited either equal or better
inhibition of lipid peroxidation in the cellular assay, com-
pared to their hydroxylated counterparts.
2
hydroxyl increased the redox potential as compared with
1
-3. In agreement with this increase in Ep values,
replacement of either hydroxyl group of methyl caffeate
by a difluoromethyl moiety (compounds 4a and 5a)
resulted in loss of activity in both SET-based assays
Conclusions
(
FRAP and DPPH scavenging). Compounds without a
The radical scavenging activity of the difluoromethyl
derivatives described here was manifested only in the b-car-
otene bleaching assay and the lipid peroxidation cellular
assay, strongly suggesting that the difluoromethyl group
actively participates in the radical scavenging process by a
hydrogen atom transfer mechanism. Our results show that
as a bioisostere of the phenolic hydroxyl, the lipophilic
difluoromethyl group retains not only the H-bond donor
properties, but also radical scavenging capacities that prove
useful to confer enhanced antioxidant properties in lipo-
philic conditions. This bioisosteric replacement may thus
provide a strategy for modifying polar phenolic antioxi-
dants rendering them more effective in lipophilic environ-
ments and biological systems.
free phenolic hydroxyl (4b,c and 5b,c) were also inactive
in these assays as the difluoromethyl group cannot par-
ticipate in electron transfer processes.
Interestingly, all the fluorinated compounds inhibited
lipid peroxidation in the human erythrocytes assay, where
even compounds 4b and 5b with non-hydrolysable
methyl ethers achieved a high degree of cellular mem-
brane protection in red blood cells. Cinnamate-like
antioxidants without free phenolic hydroxyls are rare,
and until now, their activity appeared to be restricted to
whole cell systems. Li et al. reported a small but signifi-
cant inhibition of AAPH-induced red blood cell haemoly-
sis by 3,4-dimethoxycinnamic acid and this was ascribed
[
13]
to demethylation occurring in the erythrocytes.
In a
similar assay of the schisandrins that contain several aro-
matic methoxy groups but no free phenolic hydroxyls,
Declarations
[
12]
activity was lost upon inhibition of demethylases.
Other mechanisms not involving the phenolic hydroxyl
have been proposed for a series of eugenol derivatives
Acknowledgements
This work was supported by grants from CONICET,
Argentina (PIP 11220110100896CO), and Universidad de
Buenos Aires (UBACYT 20020090200143).
[46]
with antioxidant properties. In our case, the behaviour
of compounds 4b and 5b as chain-breaking antioxidants
2
40
© 2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 68 (2016), pp. 233–244

Mario D. Mart ꢀı nez et al.
Difluoromethyl bioisosteres of hydroxycinnamates
conjugated chains. Food Chem 2014; 27. Hanson SR et al. Synthesis and evalu-
References
1
58: 41–47.
ation of general mechanism-based
inhibitors of sulfatases based on (di-
fluoro)methyl phenyl sulfate and cyc-
lic phenyl sulfamate motifs. Bioorg
Med Chem 2006; 14: 8386–8395.
28. Liu Y et al. Synthesis and protection
of aryl sulfates using the 2,2,2-tri-
chloroethyl moiety. Org Lett 2004; 6:
209–212.
1
. Carocho M, Ferreira ICFR. A review on 14. Roleira FMF et al. Lipophilic phenolic
antioxidants, prooxidants and related
controversy: natural and synthetic com-
pounds, screening and analysis method-
ologies and future perspectives. Food
Chem Toxicol 2013; 51: 15–25.
antioxidants: correlation between
antioxidant profile, partition coeffi-
cients and redox properties. Bioorg
Med Chem 2010; 18: 5816–5825.
15. Teixeira J et al. Exploring nature prof-
its: development of novel and potent
lipophilic antioxidants based on gal-
2
. Saeidnia S, Abdollahi M. Antioxi-
dants: friends or foe in prevention or
treatment of cancer: the debate of the
century. Toxicol Appl Pharmacol 2013;
loyl–cinnamic hybrids. Eur
Chem 2013; 62: 289–296.
16. Garrido J et al. Alkyl esters of hydrox-
J
Med 29. Betley JR et al. Direct screening for
phosphatase activity by turnover-based
271: 49–63.
capture of protein catalysts. Angew
Chem Int Ed 2002; 41: 775–777.
antioxidant activity and lipophilicity 30. Tapia A et al. Free radical scavengers
3
. Benfeito S et al. Antioxidant therapy:
still in search of the ‘magic bullet’.
Mitochondrion 2013; 13: 427–435.
ycinnamic acids with improved
protect PC12 cells against oxidative
and antioxidants from Baccharis grise-
bachii. Ethnopharmacol 2004; 95:
155–161.
amide antioxidants: synthesis, radical 31. Prior RL et al. Standardized methods
4. Pinchuk I et al. Evaluation of antioxi-
stress. Biochimie 2012; 94: 961–967.
J
dants: scope, limitations and relevance 17. Aladedunye F et al. Novel caffeic acid
of assays. Chem Phys Lipids 2012; 165:
38–647.
6
scavenging activity and performance
under storage and frying conditions.
Food Chem 2012; 130: 945–952.
for the determination of antioxidant
capacity and phenolics in foods and
5. El-Seedi HR et al. Biosynthesis, natural
sources, dietary intake, pharmacoki-
dietary supplements.
Chem 2005; 53: 4290–4302.
antioxidant activity of long chain alkyl 32. G u€ lßcin I et al. Polyphenol contents
J Agric Food
netic properties, and biological activi- 18. Menezes JCJMDS et al. Synthesis and
ties of hydroxycinnamic acids. J Agric
Food Chem 2012; 60: 10877–10895.
. Murphy MP. Antioxidants as thera-
hydroxycinnamates. Eur J Med Chem
and antioxidant activity of lyophilized
aqueous extract of propolis from
Erzurum, Turkey. Food Chem Toxicol
2010; 48: 2227–2238.
6
7
2011;46:773–777.
pies: can we improve on nature? Free 19. Laguerre
M et al. How to boost
Radical Biol Med 2014; 66: 20–23.
. Quideau S et al. Plant polyphenols:
antioxidants by lipophilization? Bio-
chimie 2013; 95: 20–26.
33. Oyaizu M. Studies on products of
chemical properties, biological activi- 20. Erickson JA, McLoughlin JI. Hydro-
browning
reaction.
Antioxidative
ties, and synthesis. Angew Chem Int
Ed 2011; 50: 586–621.
. Sova M. Antioxidant and antimicro-
gen bond donor properties of the
difluoromethyl group. J Org Chem
1995; 60: 1626–1631.
activities of products of browning
reaction prepared from glucosamine.
Jpn J Nutr Die 1986; 44: 307–315.
8
9
bial activities of cinnamic acid deriva- 21. Gu Y et al. Fundamental properties of 34. Prieto MA et al. A critical point: the
tives. Mini-Rev Med Chem 2012; 12:
49–767.
. Razzaghi-Asl
the CH ꢃꢃꢃ O interaction: is it a true
hydrogen bond? J Am Chem Soc 1999;
121: 9411–9422.
problems associated with the variety
of criteria to quantify the antioxidant
capacity. J Agric Food Chem 2014; 62:
5472–5484.
7
N
et al. Antioxidant
properties of hydroxycinnamic acids: 22. Dolbier W Jr. Structure, reactivity and
a review of structure-activity relation-
chemistry of fluoroalkyl radicals. 35. Mogana R et al. Anti-Inflammatory,
ships. Curr Med Chem 2013; 20:
Chem Rev 1996; 96: 1557–1584.
23. Brahms DLS, Dailey WP. Fluorinated
carbenes. Chem Rev 1996; 96: 1585–
1632.
anticholinesterase, and antioxidant
potential of scopoletin isolated from
Canarium patentinervium Miq. (Burs-
eraceae Kunth). Evid Based Comple-
ment Alternat Med 2013; 2013: Article
ID 734824.
4436–4450.
1
1
0. Lafay S, Gil-Izquierdo A. Bioavailabil-
ity of phenolic acids. Phytochem Rev
2008; 7: 301–311.
24. Toteva MM, Richard JP. The genera-
tion and reactions of quinone
1. Touaibia M et al. Caffeic acid, a ver-
satile pharmacophore: an overview.
Mini-Rev Med Chem 2011; 11: 695–
methides. Adv Phys Org Chem 2011; 36. Cho YJ et al. Synthesis of conjugated
45: 39–91.
spiropyran dyes via palladium-cata-
lyzed coupling reaction. Synth Com-
mun 1999; 29: 2061–2068.
713.
25. Lenger J et al. Evaluation of sulfatase-
directed quinone methide traps for
1
1
2. Zheng RL et al. Difference in antioxida-
tion for schisandrins and schisantherin
between bio- and chemo- systems. Phy-
tother Res 1997; 11: 600–602.
proteomics. Bioorg Med Chem 2012; 37. Lee SH et al. An efficient and conve-
20: 622–627.
nient synthesis of 5-formylsalicylalde-
hyde. Synth Commun 2000; 30: 1003–
1008.
26. Kwan DH et al. Self-immobilizing flu-
orogenic imaging agents of enzyme
3. Li Y et al. An effective strategy to
develop active cinnamic acid-directed
antioxidants based on elongating the
activity. Angew Chem Int Ed 2011; 50: 38. Lal GS et al. Bis(2-methoxyethyl)
300–303.
aminosulfur Trifluoride:
a
new
©
2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 68 (2016), pp. 233–244
241

Difluoromethyl bioisosteres of hydroxycinnamates
Mario D. Mart ꢀı nez et al.
broad-spectrum
deoxofluorinating 44. Lekse JM et al. Plant catechols prevent
liferative and proapoptotic activity. J
Med Chem 2005; 48: 4931–4946.
agent with enhanced thermal stability.
J Org Chem 1999; 64: 7048–7054.
9. Born M et al. Electrochemical beha-
lipid peroxidation in human plasma
and erythrocytes. Mol Cell Biochem 50. Hodgson
HH,
Jenkinson
TA.
3
4
4
4
2001; 226: 89–95.
CCCCIII.-The Reimer-Tiemann reac-
tion with m-bromo- and m-iodo-phe-
viour and antioxidant activity of some 45. Prieto MA et al. b-Carotene assay
natural polyphenols. Helv Chim Acta
996; 79: 1147–1158.
0. Sochor J et al. Electrochemistry as a
revisited. Application to characterize
and quantify antioxidant and prooxi-
dant activities in a microplate. J Agric
Food Chem 2012; 60: 8983–8993.
nol. J Chem Soc (Resumed) 1927;
1
3041–3044.
tool for studying antioxidant proper-
Supporting Information
ties. Int J Electrochem Sci 2013; 8: 46. d’Avila Farias M et al. Eugenol deriva-
Additional Supporting Information
may be found in the online version of
this article:
8464–8489.
tives as potential anti-oxidants: is phe-
nolic hydroxyl necessary to obtain an
effect? J Pharm Pharmacol 2014; 66:
733–746.
1. Teixeira J et al. Hydroxycinnamic acid
antioxidants: an electrochemical over-
view. BioMed Res Int 2013;2013:Article
ID 251754.
Table S1 Antioxidant efficacy
DPPH and FRAP) and redox poten-
(
47. Foti MC, Amorati R. Non-phenolic
radical-trapping antioxidants. J Pharm
Pharmacol 2009; 61: 1435–1448.
2. Hammerich O, Svensmark B. The
anodic oxidation of oxygen-containing
tials (Ep).
Table S2 Calculated partition coef-
compounds. In: Lund H, Baizer MM, 48. Frankel EN et al. Interfacial phenom- ficients for compounds 1-3, 4a-c, 5a-c.
eds. Organic Electrochemistry: An
Introduction and a Guide, 3rd edn.
New York: Marcel Dekker Inc., 1991:
ena in the evaluation of 630 antioxi-
Appendix S1 Synthesis of interme-
dants: bulk oils vs emulsions. J Agric diates 6b, 6c, 9b and 9c.
1
Appendix S2 H (500.13 MHz) and
Food Chem 1994; 42: 1054–1059.
49. Cincinelli R et al. Synthesis and struc-
tureꢀactivity relationships of a new
series of retinoid-related biphenyl-4-
ylacrylic acids endowed with antipro-
13
6
15–657.
3. G u€ lßcin I. Antioxidant activity of caf-
feic acid (3,4-dihydroxycinnamic
acid). Toxicology 2006; 217: 213–220.
C (125.72 MHz) NMR spectra of
compounds 4a-c and 5a-c.
4
Appendix
(CDCl ) d: 189.3 (ArCHO), 167.5 (1-C), 163.1 (4’-C),
3
143.3 (3-C), 135.5 (6’-C), 128.3 (2’-C), 127.4 (1’-C), 125.1
(
(
3’-C), 117.3 (2-C), 112.3 (5’-C), 56.1 (CH OAr), 51.9
3
+
+
CH O); HRMS: calcd for C H O [M+H] : 221.0808,
3
12 13
4
Physical and spectroscopic data
found: 221.0805.
Methyl 3-(3-formyl-4-hydroxyphenyl)-(E)-
propenoate (6a)
Methyl 3-(4-acetyloxy-3-formylphenyl)-(E)-
propenoate (6c)
[49]
1
White solid, mp 104°C (lit.
100–101°C); H NMR
1
(
CDCl ) d: 11.21 (s, 1H, ArOH), 9.93 (d, J = 0.5 Hz, 1H,
White needles, mp 141–142°C; H NMR (CDCl
) d: 10.13
3
3
ArCHO), 7.73 (dd, J = 2.3, 8.6 Hz, 1H, 6’-H), 7.71 (d,
J = 2.2 Hz, 1H, 2’-H), 7.66 (d, J = 16.0 Hz, 3-H), 7.03 (d,
J = 8.5 Hz, 1H, 5’-H), 6.37 (d, J = 16.0 Hz, 1H, 2-H), 3.82
(bs, 1H, ArCHO), 8.02 (d, J = 2.3 Hz, 1H, 2’-H), 7.77 (ddd,
J = 0.4, 2.3, 8.4 Hz, 1H, 6’-H), 7.70 (d, J = 16.0 Hz, 1H, 3-
H), 7.25 (d, J = 8.4 Hz, 1H, 5’-H), 6.49 (d, J = 16.0 Hz,
1
3
13
C
(
s, 3H, CH O); C NMR (CDCl ) d: 196.4 (ArCHO),
1H, 2-H), 3.83 (s, 3H, CH
NMR (CDCl ) d: 188.2 (ArCHO), 169.1 (CH C(O)), 167.0
3
O), 2.41 (s, 3H, CH C(O));
3 3
3
3
1
67.4 (1-C), 163.2 (4’-C), 142.9 (3-C), 135.8 (6’-C), 134.0
2’-C), 126.8 (1’-C), 120.7 (3’-C), 118.8 (5’-C), 117.1 (2-
3
(
(1-C), 152.7 (4’-C), 142.4 (3-C), 134.3 (6’-C), 133.1 (1’-C),
130.5 (2’-C), 128.4 (3’-C), 124.4 (5’-C), 119.9 (2-C), 51.9
+
+
C), 51.9 (CH O); HRMS: calcd for C H O [M+H] :
3
11 11 4
+
2
07.0652, found: 207.0658.
(CH
[
O), 21.0 (CH C(O)); HRMS: calcd for C13H12NaO
3 3 5
+
M+Na] : 271.0577, found: 271.0580.
Methyl 3-(3-formyl-4-methoxyphenyl)-(E)-
propenoate (6b)
Methyl 3-[(3-(difluoromethyl)-4-
methoxyphenyl)]-(E)-propenoate (4b)
1
White solid, mp 123–124°C; H NMR (CDCl ) d: 10.46 (s,
1
3
1
H, ArCHO), 8.01 (d, J = 2.4 Hz, 1H, 2’-H), 7.71 (dd,
J = 2.4, 8.7 Hz, 1H, 6’-H), 7.65 (d, J = 16.0 Hz, 1H, 3-H),
.03 (d, J = 8.7 Hz, 1H, 5’-H), 6.40 (d, J = 16.0 Hz, 1H, 2-
White solid, mp 77–78°C; H NMR (CDCl
3
) d: 7.76–7.75
(m, 1H, 2’-H), 7.66 (d, J = 16.0 Hz, 1H, 3-H), 7.60–7.58
(m, 1H, 6’-H), 6.95 – 6.91 (m, 1H, 5’-H), 6.93 (t,
J = 55.5 Hz, 1H, CF H), 6.37 (d, J = 16.0 Hz, 1H, 2-H),
2
7
13
H), 3.98 (s, 3H, CH OAr), 3.80 (s, 3H, CH O); C NMR
3
3
2
42
© 2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 68 (2016), pp. 233–244

Mario D. Mart ꢀı nez et al.
Difluoromethyl bioisosteres of hydroxycinnamates
1
.91 (s, 3H, CH OAr), 3.80 (s, 3H, CH O); C NMR
3
3
196.2 (ArCHO), 161.5 (2-C), 134.4 (6-C), 129.6 (5-C),
127.4 (3-C), 120.1 (1-C), 105.3 (4-C).
3
3
(
CDCl ) d: 167.6 (1-C), 158.9 (t, J = 6 Hz, 4’-C), 143.7 (3-
3
C), 132.2 (6’-C), 127.3 (1’-C), 126.1 (t, J = 6 Hz, 2’-C),
23.4 (t, J = 23 Hz, 3’-C), 116.8 (2-C), 111.4 (5’-C), 111.2
t, J = 237 Hz, CF H), 56.0 (CH OAr), 51.9 (CH O);
1
Methyl 3-(4-formyl-3-hydroxyphenyl)-(E)-
propenoate (9a)
(
2
3
3
+
+
HRMS: calcd for C H F NaO [M+Na] : 265.0647,
1
2
12
2
3
1
found: 265.0656.
White solid, mp 87–88°C; H NMR (CDCl ) d: 11.03 (bs,
3
1H, ArOH), 9.91 (d, J = 0.6 Hz, 1H, ArCHO), 7.63 (d,
J = 16.0 Hz, 3-H), 7.59 (d, J = 8.0 Hz, 1H, 5’-H), 7.17 (dd,
J = 1.5, 8.0 Hz, 1H, 6’-H), 7.12 (d, J = 1.2 Hz, 1H, 2’-H),
Methyl 3-[4-acetyloxy-(3-(difluoromethyl)
phenyl)]-(E)-propenoate (4c)
13
6
.53 (d, J = 16.0 Hz, 1H, 2-H), 3.83 (s, 3H, CH O);
C
3
1
White solid, mp 101–102°C; H NMR (CDCl ) d: 7.77–
NMR (CDCl ) d: 196.1 (ArCHO), 166.8 (1-C), 161.9 (3’-C),
3
3
7
.75 (m, 1H, 2’-H), 7.69 (d, J = 16.0 Hz, 1H, 3-H), 7.66–
.62 (m, 1H, 6’-H), 7.25–7.23 (m, 1H, 5’-H), 6.76 (t,
143.0 (3-C), 142.6 (1’-C), 134.2 (5’-C), 122.1 (2-C), 121.4
(4’-C), 119.4 (6’-C), 117.1 (2’-C), 51.2 (CH O); HRMS:
7
3
+
+
J = 55.2 Hz, 1H, CF H), 6.45 (d, J = 16.0 Hz, 1H, 2-H),
calcd for C H O [M+H] : 207.0652, found: 207.0659.
11 11 4
2
1
.82 (s, 3H, CH O), 2.35 (s, 3H, CH C(O)); C NMR
3
3
3 3
(
CDCl ) d: 168.6 (CH C(O)), 167.1 (1-C), 149.7 (t, J =
3 3
Hz, 4’-C), 142.8 (3-C), 132.7 (1’-C), 131.2 (t, J = 2 Hz,
’-C), 127.1 (t, J = 23 Hz, 3’-C), 126.3 (t, J = 6 Hz, 2’-C),
Methyl 3-(4-formyl-3-methoxyphenyl)-(E)-
propenoate (9b)
5
6
1
5
1
24.0 (5’-C), 119.4 (2-C), 111.5 (t, J = 239 Hz, CF H),
White solid, mp 145°C; H NMR (CDCl ) d: 10.5 (d,
2
3
2.0 (CH O), 21.0 (CH C(O)); HRMS: calcd for
3
J = 0.8 Hz, 1H, ArCHO), 7.84 (d, J = 7.9 Hz, 1H, 5’-H),
7.68 (d, J = 16.0 Hz, 1H, 3-H), 7.23–7.17 (m, 1H, 6’-H),
3
+
+
C H F NaO [M+Na] : 293.0596, found: 293.0592.
13
12
2
4
7
.09 (d, J = 1.4 Hz, 1H, 2’-H), 6.53 (d, J = 16.0 Hz, 1H, 2-
13
H), 3.97 (s, 3H, CH OAr), 3.83 (s, 3H, CH O); C NMR
3
3
Methyl 3-[(3-(difluoromethyl)-4-
hydroxyphenyl)]-(E)-propenoate (4a)
(
CDCl ) d: 189.2 (ArCHO), 166.9 (1-C), 162.0 (3’-C),
3
143.6 (3-C), 141.6 (1’-C), 129.2 (5’-C), 125.9 (4’-C), 121.0
(2-C), 120.4 (6’-C), 111.1 (2’-C), 55.9 (CH OAr), 55.1
1
Light yellow solid, mp 132–133°C; H NMR (CDCl ) d:
3
3
+
+
7
7
.64 (d, J = 16.0 Hz, 1H, 3-H), 7.65–7.63 (m, 1H, 2’-H),
.53–7.48 (m, 1H, 6’-H), 6.93–6.90 (m, 1H, 5’-H), 6.90 (t,
(CH O); HRMS: calcd for C H NaO
3
[M+Na] :
12 12
4
243.0628, found: 243.0624.
J = 55.3 Hz, 1H, CF H), 6.45 (bs, 1H, ArOH), 6.35 (d,
2
1
3
J = 16.0 Hz, 1H, 2-H), 3.82 (s, 3H, CH O); C NMR
3
Methyl 3-(3-acetyloxy-4-formylphenyl)-(E)-
propenoate (9c)
(
(
CDCl ) d: 167.8 (1-C), 155.8 (t, J = 4.6 Hz, 4’-C), 143.8
3-C), 132.0 (6’-C), 127.5 (1’-C), 127.2 (t, J = 6 Hz, 2’-C),
21.3 (t, J = 22 Hz, 3’-C), 117.3 (5’-C), 116.7 (2-C), 112.8
3
1
1
White solid, mp 106–107°C; H NMR (CDCl ) d: 10.10
3
(
t, J = 237 Hz, CF H), 52.0 (CH O); HRMS: calcd for
(bs, 1H, ArCHO), 7.90 (d, J = 8.0 Hz, 1H, 5’-H), 7.67 (d,
J = 16.0 Hz, 1H, 3-H), 7.53 (dd, J = 1.5, 8.0 Hz, 1H, 6’-
H), 7.33 (d, J = 1.5 Hz, 1H, 2’-H), 6.53 (d, J = 16.0 Hz,
2
3
+
+
C H F NaO [M+Na] : 251.0490, found: 251.0500.
11
10
2
3
13
1
H, 2-H), 3.83 (s, 3H, CH O), 2.42 (s, 3H, CH C(O));
3
C
3
2
-Hydroxymethyl-5-iodophenol (7)
NMR (CDCl ) d: 188.1 (ArCHO), 169.2 (CH C(O)), 166.7
3
3
1
White solid, mp 138°C; H NMR (CD OD) d: 7.16 (dd,
J = 1.7, 7.9 Hz, 1H, 5-H), 7.12 (d, J = 1.6 Hz, 1H, 2-H),
7
(1-C), 151.9 (3’-C), 142.3 (3-C), 141.4 (1’-C), 131.9 (5’-C),
128.8 (4’-C), 126.0 (6’-C), 122.8 (2’-C), 122.0 (2-C), 52.2
3
13
+
.03 (d, J = 7.9 Hz, 1H, 6-H), 4.58 (s, 2H, ArCH OH);
C
(CH O), 21.0 (CH C(O)); HRMS: calcd for C H NaO
5
2
3
3
13 12
+
NMR (CD OD) d: 157.1 (2-C), 130.7 (6-C), 129.5 (5-C),
[M+Na] : 271.0577, found: 271.0568.
3
1
28.9 (1-C), 124.7 (3-C), 93.0 (4-C), 60.4 (ArCH OH);
2
+
+
HRMS: calcd for C H INaO (M+Na) : 272.9383, found:
7
7
2
Methyl 3-[(4-(difluoromethyl)-3-
methoxyphenyl)]-(E)-propenoate (5b)
272.9394.
1
White solid, mp 108–109°C; H NMR (CDCl ) d: 7.68 (d,
3
2
-Hydroxy-4-iodo-benzaldehyde (8)
J = 16.0 Hz, 1H, 3-H), 7.58 (d, J = 7.9 Hz, 1H, 5’-H), 7.20
(d, J = 8.0 Hz, 1H, 6’-H), 7.06–7.03 (m, 1H, 2’-H), 6.93 (t,
[50]
1
White solid, mp 85–86°C (lit.
87°C); H NMR (CDCl )
3
d: 11.02 (s, 1H, ArOH), 9.85 (d, J = 0.6 Hz, 1H, ArCHO),
J = 55.6 Hz, 1H, CF H), 6.48 (d, J = 16.0 Hz, 1H, 2-H),
3.91 (s, 3H, CH OAr), 3.83 (s, 3H, CH O); C NMR
3 3
2
13
7
5
.45–7.43 (m, 1H, 3-H), 7.40 (dd, J = 1.5, 8.1 Hz, 1H,
1
3
-H), 7.24 (d, J = 8.1 Hz, 1H, 6-H); C NMR (CDCl ) d:
(CDCl ) d: 167.2 (1-C), 157.7 (t, J = 6 Hz, 3’-C), 144.0
3
3
©
2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 68 (2016), pp. 233–244
243

Difluoromethyl bioisosteres of hydroxycinnamates
Mario D. Mart ꢀı nez et al.
(
3-C), 138.2 (t, J = 2 Hz, 1’-C), 126.9 (t, J = 6 Hz, 5’-C),
24.5 (t, J = 22.3 Hz, 4’-C), 120.7 (6’-C), 119.7 (2-C),
J = 2383 Hz, CF H), 52.1 (CH O), 20.9 (CH C(O));
HRMS: calcd for C H F O (M+H) : 271.0776, found:
13 13 2 4
271.0776.
2
3
3
+
+
1
1
5
2
11.3 (t, J = 236 Hz, CF H), 110.0 (2-C), 55.8 (CH OAr),
2
3
+
+
2.0 (CH O); HRMS: calcd for C H F NaO [M+Na] :
3
12 12
2
3
65.0647, found: 265.0653.
Methyl-3-[(4-(difluoromethyl)-3-
hydroxyphenyl)]-(E)-propenoate (5a)
Methyl 3-[3-acetyloxy-4-(difluoromethyl)
phenyl)]-(E)-propenoate (5c)
1
White solid, mp 129–130°C; H NMR (CDCl ) d: 7.64 (d,
3
J = 16.0 Hz, 1H, 3-H), 7.48 (d, J = 8.0 Hz, 1H, 5’-H), 7.17
(d, J = 8.0 Hz, 1H, 6’-H), 7.10–7.06 (m, 1H, 2’-H), 6.90 (t,
1
White solid, mp 73–74°C; H NMR (CDCl ) d: 7.65 (d,
3
J = 16.0 Hz, 1H, 3-H), 7.63 (d, J = 8.2 Hz, 1H, 5’-H), 7.47
J = 55.3 Hz, 1H, CF H), 6.46 (d, J = 16.0 Hz, 1H, 2-H),
2
13
(
d, J = 8.2 Hz, 1H, 6’-H), 7.34 (bs, 1H, 2’-H), 6.74 (t,
6.21 (t, J = 1.6 Hz, 1H, ArOH), 3.84 (s, 3H, CH O);
C
3
J = 55.1 Hz, 1H, CF H), 6.47 (d, J = 16.0 Hz, 1H, 2-H),
NMR (CDCl ) d: 167.6 (1-C), 154.5 (t, J = 5.0 Hz, 3’-C),
2
3
13
3
.82 (s, 3H, CH O), 2.36 (s, 3H, CH C(O)); C NMR
3
143.9 (3-C), 138.2 (t, J = 1.7 Hz, 1’-C), 127.7 (t, J = 6 Hz,
5’-C), 122.5 (t, J = 22 Hz, 4’-C), 120.5 (6’-C), 119.7 (2-C),
116.0 (2’-C), 112.8 (t, J = 236 Hz, CF H), 52.2 (CH O);
3
(
CDCl ) d: 168.8 (CH C(O)), 166.9 (1-C), 148.9 (t,
3 3
J = 5.3 Hz, 3’-C), 142.7 (3-C), 138.3 (t, J = 2 Hz, 1’-C),
2
3
+
+
127.8 (t, J = 23 Hz, 4’-C), 127.3 (t, J = 6.4 Hz, 5’-C),
125.8 (6’-C), 122.5 (2’-C), 120.7 (2-C), 111.6 (t,
HRMS: calcd for C H F NaO [M+Na] : 251.0490,
11 10 2 3
found: 251.0493.
2
44
© 2016 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, 68 (2016), pp. 233–244