In-vitro and in-vivo antimalarial activity of caffeic acid and some of its derivatives

Article abstract of DOI:10.1111/jphp.12982

Objectives: To explore the in-vitro and in-vivo antimalarial potential of caffeic acid and derivatives. Methods: Two common phenolic acids (caffeic acid and chlorogenic acid) were evaluated for in-vitro and in-vivo antiplasmodial activity in comparison with some semi-synthetic derivatives that were synthesized. An in-vitro assay based on plasmodial lactate dehydrogenase activity, and the classical in-vivo 5-day suppressive test from Peters on an artemisinin-resistant Plasmodium berghei strain was used. Parasitic stage sensitivity to ethyl caffeate was determined in this work. Key findings: Phenolic acid esters derivatives showed better antiplasmodial activity than corresponding phenolic acids. The derivative with the highest in-vitro activity being caffeic acid ethyl ester, exhibiting an IC₅₀?=?21.9?±?9.4?μm. Ethyl caffeate and methyl caffeate were then evaluated for antimalarial activity in?vivo and ethyl caffeate showed a growth inhibition of 55% at 100?mg/kg. Finally, it seems that ethyl caffeate blocks the growth of young parasitic forms. Conclusions: Our study provides evidence for an antimalarial potential of caffeic acid derivatives which are common in several medicinal plants traditionally used against malaria. It also demonstrates the possibility to use such derivatives in the treatment of malaria.

Full text of DOI:10.1111/jphp.12982

Research Paper
In-vitro and in-vivo antimalarial activity of caffeic acid and
some of its derivatives
Sylvain G. Alsona,b, Olivia Jansena, Ewa Cieckiewicza, Hajatiana Rakotoarimananab,
Herintsoa Rafatrob, Gilles Degottea,c, Pierre Francottec and Michel Fredericha
a
ꢀ
ꢀ
ꢁ
ꢁ
Laboratoire de pharmacognosie, Centre Interdisciplinaire de Recherches sur les Medicaments (CIRM), Universite de Liege, Liege, Belgium,
b
ꢀ
ꢀ
Laboratoire d’Evaluation Pharmaco Clinique (LEPC), Institut Malgache de Recherches Appliquees (IMRA), Fondation Albert et Suzanne Rakoto-
Ratsimamanga, Antananarivo, Madagascar and cLaboratoire de Chimie Pharmaceutique, Centre Interdisciplinaire de Recherches sur les
ꢀ
ꢀ
ꢁ
ꢁ
Medicaments (CIRM), Universite de Liege, Liege, Belgium
Keywords
Abstract
caffeic acid; ethyl caffeate; Malaria;
Plasmodium falciparum
Objectives To explore the in-vitro and in-vivo antimalarial potential of caffeic
acid and derivatives.
Correspondence
Methods Two common phenolic acids (caffeic acid and chlorogenic acid) were
evaluated for in-vitro and in-vivo antiplasmodial activity in comparison with
some semi-synthetic derivatives that were synthesized. An in-vitro assay based on
plasmodial lactate dehydrogenase activity, and the classical in-vivo 5-day sup-
pressive test from Peters on an artemisinin-resistant Plasmodium berghei strain
was used. Parasitic stage sensitivity to ethyl caffeate was determined in this work.
Key findings Phenolic acid esters derivatives showed better antiplasmodial activ-
ity than corresponding phenolic acids. The derivative with the highest in-vitro
activity being caffeic acid ethyl ester, exhibiting an IC50 = 21.9 ꢀ 9.4 lM. Ethyl
caffeate and methyl caffeate were then evaluated for antimalarial activity in vivo
and ethyl caffeate showed a growth inhibition of 55% at 100 mg/kg. Finally, it
seems that ethyl caffeate blocks the growth of young parasitic forms.
Conclusions Our study provides evidence for an antimalarial potential of caffeic
acid derivatives which are common in several medicinal plants traditionally used
against malaria. It also demonstrates the possibility to use such derivatives in the
treatment of malaria.
Michel Frederich, Laboratoire de
pharmacognosie, Centre Interdisciplinaire de
ꢀ
Recherches sur les Medicaments (CIRM),
ꢀ
ꢁ
Universite de Liege, Avenue Hippocrate 15,
ꢁ
B36, 4000, Liege, Belgium.
E-mail: m.frederich@uliege.be
Received April 5, 2018
Accepted June 27, 2018
doi: 10.1111/jphp.12982
particularly at risk of severe malaria, if infected with Plas-
modium falciparum. Malaria remains a major killer of
Introduction
Malaria is a major public health problem worldwide, espe-
cially in tropical and subtropical countries. The global
number of malaria cases and deaths was estimated to have
decreased since 2005 due to the expansion of access to
rapid diagnostic tests, quality assured artemisinin combina-
tion therapy, known as ACT (311 million courses were
produced in 2015) and increased funding for malaria con-
trol programs (US$ 2.9 billion in 2015).[1] However,
malaria is still the most deadly parasitic disease in the
world, with an estimated 216 million cases in 2016.[1] For
the first time since long ago, the number of registered cases
has not decreased between 2015 and 2016 (from 212 to
216 million cases). Children, pregnant women, HIV
patients and travellers to sub-Saharan African countries are
under-fives, claiming the life of one child every 2 min.[1,2]
The emergence and rapid spread of antimalarial drug-
resistant strains of P. falciparum, of insecticides resistant
mosquitoes and the absence of effective vaccine (Mosquirix
vaccine currently evaluated shows only a protection of about
30%)[3] made it more difficult to fight against this scourge.
This alarming situation highlights the urgent need to discover
new antimalarial molecules. Natural phenolic compounds,
commonly found in various plants, and some of their simple
esters are known for several biological activities, such as anti-
inflammatory,[4,5] antioxidant,[6] anti-diabetic,[7] anti-carci-
noma,[8] antibacterial[9] and immunomodulatory.[10] Among
phenolic metabolites, caffeic acid and derivatives (such as
chlorogenic) are the most frequently found and could be
© 2018 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2018), pp. **–**
1

Antimalarial activity of caffeic acid derivatives
Sylvain G. Alson et al.
present in high quantities in some plants used as antimalarial
traditional medicines, but also in foods (coffee beans, thyme,
mint, wine, several fruits and vegetables.. .).[11,12]
sulfuric acid (0.2 ml). After 2 h, the mixture was cooled
down and ethyl acetate was added. The resulting organic
phase was washed by a NaHCO3 aqueous solution (5% w/v),
water and then dried over anhydrous magnesium sulfate.
After filtration of the drying agent, removal of the solvent
under vacuo gave the pure expected compound.
Caffeic acid or simple derivatives have never been evalu-
ated for antimalarial activity, even if present in numerous
medicinal plants traditionally used as antimalarials. Having
found in our previous works on antimalarial traditional
plants a moderate antimalarial activity for caffeic acid that
could explain in part the activity of some traditional
medicines (data not published), we wanted, in this publica-
tion, to confirm this finding and to evaluate the in-vitro
and particularly in-vivo activity of some semi-synthesized
close derivatives to caffeic acid. For in-vivo activity assays,
an artemisinin-resistant Plasmodium berghei strain was
developed, to demonstrate the potential of evaluated com-
pounds against such resistant strains. This study shows for
the first time that caffeic acid ethyl ester possess in-vitro
antiplasmodial activity against P. falciparum and in-vivo
antimalarial activity against P. berghei in mice.
Methyl caffeate (2a)
Yield = 95%. mp = 155.6–158.9 °C (lit.: 158 °C); 1H
NMR (DMSO-d6) d 3.69 (CH3, s, 3H), 6.27 (Ar-CH = CH,
d, J = 15.9 Hz, 1H), 6.76 (5-H, d, J = 8.1 Hz, 1H), 7.00
(6-H, d, J = 8.2 Hz, 1H), 7.03–7.09 (2-H, m, 1H), 7.48
(Ar-CH = CH, d, J = 15.9 Hz, 1H), 9.37 (Ar-OH, bs, 2H).
Anal. (C10H10O4) theoretical: C, 61.85; H, 5.19. Found: C,
61.97; H: 5.32. MS m/z: 193 [M]+, 161, 134. UVmax: 218,
295 et 324 nm.
Ethyl caffeate (2b)
Yield = 84%. mp = 140.8–143.1 °C (lit. 143–146 °C); 1H
NMR (DMSO-d6) d 1.24 (CH2CH3, t, J = 7.1 Hz, 3H), 4.15
(CH2CH3, q, J = 7.1 Hz, 2H), 6.25 (Ar-CH = CH, d,
J = 15.9 Hz, 1H), 6.75 (5-H, d, J = 8.2 Hz, 1H), 7.00 (6-H, d,
J = 8.1 Hz, 1H), 7.04 (2-H, d, J = 2.1 Hz, 1H), 7.47 (Ar-CH
= CH, d, J = 15.9 Hz, 1H), 9.61 (Ar-OH, bs, 2H). Anal.
(C11H12O4) theoretical: C, 63.45; H, 5.81. Found: C, 63.79; H,
5.91. MS m/z: 208 [M]+, 179, 161; UVmax: 218, 295 et 324 nm.
Materials and Methods
Chemicals and reagents
All chemical solvent and reagent were purchased from
LiChrosolv, Merck Germany and HiPerSolv Chromanorm,
VWR Chemicals EC, abcr, Acros Organics, Alfa Aesar,
Fisher Chemical, Janssen Chimica, Merck, Sigma-Aldrich.
General experimental procedures
Isopropyl caffeate (2c)
1
Melting points were determined on a Stuart SMP3 capillary
1
apparatus and are uncorrected. The H NMR spectra were
Yield = 62%. mp = 129.8–135.0 °C (lit. 141–143 °C); H
NMR (DMSO-d6) d 1.24 (CH(CH3)2, d, J = 6.3 Hz, 6H),
4.99 (CH(CH3)2, q, J = 6.2 Hz, 1H), 6.22 (Ar-CH = CH,
d, J = 15.9 Hz, 1H), 6.75 (5-H, d, J = 8.1 Hz, 1H), 6.99
(6-H, dd, J = 2.1/8.2 Hz, 1H), 7.03 (2-H, d, J = 2.1 Hz,
1H), 7.40 (Ar-CH = CH, d, J = 15.9 Hz, 1H), 9.35 (Ar-OH,
bs, 2H). Anal. (C12H14O4) theoretical: C, 64.85; H, 6.35.
Found: C, 64.78; H, 6.36.
taken on a Bruker Advance 500 (500 MHz) instrument
equipped with cryoprobe in DMSO-d6 with TMS as an
internal standard; chemical shifts are reported in d values
(ppm) relative to internal TMS. The abbreviation s = sin-
glet, d = doublet, t = triplet, q = quadruplet, quint = quin-
tuplet, m = multiplet, and b = broad are used throughout.
Elemental analyses (C, H, N, S) were realized on a FlashEA
1112 series (Thermo-Interscience, Louvain-la-Neuve, Bel-
gium) and were within 0.4% of the theoretical values. All
reactions were routinely checked by TLC on silica gel
Merck 60F254. MS detection was carried out using Micro-
mass ESI-Q-TOF II instrument using ESI ionization in the
positive mode (Waters, Zellik, Belgium). The UV spectra
were acquired using a U-2910 spectrophotometer.
Ethyl cinnamate (2f)
1
Yield = 48%. Oil. H NMR (DMSO-d6) d 1.25 (CH2CH3,
t, J = 7.1 Hz, 3H), 4.18 (CH2CH3, q, J = 7.1 Hz, 2H), 6.50
(Ar-CH = CH, d, J = 16.1 Hz, 1H), 7.43 (3-H/4-H/5-H,
m, 3H), 7.67 (Ar-CH = CH, d, J = 16.0 Hz, 1H), 7.73 (2-
H/6-H, m, 2H). Anal. (C11H12O3) theoretical: C, 68.74; H,
6.29. Found: C, 68.35; H, 6.33.
Fischer esterification (for 2a-g)
Ethyl 3-hydroxycinnamate (2d)
A solution of the suitable carboxylic acid 1-2 (for instance
1, 200 mg, 1.11 mmol) or methyl ester 3 in the appropriate
alcohol (8 ml) was heated under reflux with concentrated
Yield = 68%. mp = 62.9–68.5 °C (lit. 69 °C); 1H NMR
(DMSO-d6) d 1.25 (CH2CH3, t, J = 7.1 Hz, 3H), 4.18
2
© 2018 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2018), pp. **–**

Sylvain G. Alson et al.
Antimalarial activity of caffeic acid derivatives
(CH2CH3, q, J = 7.1 Hz, 2H), 6.50 (Ar-CH = CH, d,
J = 16.0 Hz, 1H), 6.83 (6-H, ddd, J = 1.0/2.5/8.0 Hz, 1H),
7.03 (2-H, t, J = 2.0 Hz, 1H), 7.13 (4-H, dt, J = 1.2/7.8 Hz,
1H), 7.21 (5-H, t, J = 7.8 Hz, 1H), 7.55 (Ar-CH = CH, d,
J = 16.0 Hz, 1H), 9.64 (Ar-OH, s, 1H). Anal. (C11H12O3)
theoretical: C, 68.74; H, 6.29. Found: C, 68.35; H, 6.33.
with concentrated sulfuric acid (0.2 ml). After 1 h at
70 °C, the mixture was cooled down, and water (10 ml)
and ethyl acetate (10 ml) were added. The solution was
extracted three times with AcOEt, and the combined
organic phases were dried with MgSO4 and concentrated in
vacuo.
Ethyl 4-hydroxycinnamate (2e)
Methyl chlorogenate (5a)
Yield = 67%. mp = 70.0–71.0 °C (lit. 73–74 °C); 1H NMR
(DMSO-d6) d 1.24 (CH2CH3, t, J = 7.1 Hz, 3H), 4.16
(CH2CH3, q, J = 7.1 Hz, 2H), 6.39 (Ar-CH = CH, d,
J = 16.0 Hz, 1H), 6.78 (Ar-H, m, 2H), 7.55 (Ar-H/Ar-
CH = CH, m, 3H), 10.01 (Ar-OH, s, 1H). Anal. (C11H12O3)
theoretical: C, 68.74; H, 6.29. Found: C, 68.48; H, 6.41.
92 mg, yield = 59%. m.p.: 160 °C (dec). 1H NMR
(DMSO-d6) d 1.76 (60-H, dd, J = 13.1, 7.8 Hz, 1H), 1.91
(20-H, dd, J = 13.2, 4.0 Hz, 1H), 2.10 (20-H/60-H, m, 2H),
3.58 (4-H, m, 1H), 3.56 (CH3, s, 3H), 3.89 (50-H, m, 1H),
5.07 (30-H, td, J = 6.9, 4.0 Hz, 1H), 6.11 (Ar-CH = CH, d,
J = 15.6 Hz, 1H), 6.77 (5-H, d, J = 8.0 Hz, 1H), 6.96 (6-H,
dd, J = 2.1/8.2 Hz, 1H), 7.01 (2-H, d, J = 2.1 Hz, 1H),
7.39 (Ar-CH = CH, d, J = 15.8 Hz, 1H).
Ethyl 30,40-dimethoxycinnamate (2g)
Yield = 47%. m.p.: 56,1 °C (lit.: 56 °C). 1H NMR (DMSO-d6)
d 1.25 (CH2CH3, t, J = 7.1 Hz, 3H), 3.80 (OCH3, s, 6H), 4.17
(CH2CH3, q, J = 7.1 Hz, 2H), 6.54 (Ar-CH = CH, d,
J = 15.9 Hz, 1H), 6.99 (5-H, d, J = 8.3 Hz, 1H), 7.23 (6-H,
dd, J = 1.9/8.3 Hz, 1H), 7.36 (2-H, d, J = 1.9 Hz, 1H), 7.58
(Ar-CH = CH, d, J = 15.9 Hz, 1H). Anal. (C13H16O4) theoret-
ical: C, 66.09; H, 6.83. Found: C, 66.47; H, 6.96.
Ethyl chlorogenate (5b)
57 mg, yield = 35%. m.p. : 140 °C (dec). 1H NMR
(DMSO-d6) d 1.14 (CH2CH3, t, J = 7.1 Hz,3H), 1.76 (60-H,
dd, J = 13.1, 7.8 Hz, 1H), 1.91 (20-H, dd, J = 13.2, 4.0 Hz,
1H), 2.10 (20-H/60-H, m, 2H), 3.58 (4-H, m, 1H), 3.75
(CH2CH3, q, J = 7.1 Hz, 2H), 3.89 (50-H, m, 1H), 5.07
(30-H, td, J = 6.9, 4.0 Hz, 1H), 6.11 (Ar-CH = CH, d,
J = 15.6 Hz, 1H), 6.77 (5-H, d, J = 8.0 Hz, 1H), 6.96 (6-H,
dd, J = 2.1/8.2 Hz, 1H), 7.01 (2-H, d, J = 2.1 Hz, 1H),
7.39 (Ar-CH = CH,d, J = 15.8 Hz, 1H).
Methyl 30,40-dimethoxycinnamate (3)
A mixture of caffeic acid (1a, 500 mg, 2.78 mmol) and
anhydrous potassium carbonate (1.92 g) in acetone
(30 ml) was stirred during 15 min at 40 °C. Dimethylsul-
fate (2.6 ml, 27 mmol) was added to the medium, and the
mixture was heated under reflux at 70 °C during 5 h. The
solid was removed by filtration and washed with acetone.
The filtrate was evaporated under vacuo. To the residue
was added ethyl acetate; the resulting solution was washed
with water and a NaHCO3 aqueous solution (5% w/v),
dried over anhydrous magnesium sulfate and evaporated
under vacuo. The solid obtained by crystallization in a cold
ethanol-water mixture was collected by filtration and char-
acterized as methyl 30,40-dimethoxycinnamate (3, 0.34 g,
55%). mp = 70.1 °C (lit. 68–69 °C). 1H NMR (DMSO-d6)
d 3.71 (COOCH3, s, 3H), 3.81 ((OCH3)2, s, 6H), 6.57 (Ar-
CH = CH, d, J = 16.0 Hz, 1H), 6.98 (5-H, d, J = 8.3 Hz,
1H), 7.23 (6-H, dd, J = 1.9/8.3 Hz, 1H), 7.36 (2-H, d,
J = 1.9 Hz, 1H), 7.59 (Ar-CH = CH, d, J = 16.0 Hz, 1H).
Anal. (C12H14O4) theoretical: C, 64.85; H, 6.35. Found; C,
64.80;H, 6,25.
Antiplasmodial activity assay
Activity against P. falciparum chloroquine-sensitive 3D7
strains was assessed following the procedure already
described in Frꢀedꢀerich et al.[13] The parasites were obtained
from Prof. Grellier (Museum d’Histoire Naturelle, Paris,
France). Each compound, fraction and extract was applied
in a series of eight twofold dilutions (final concentrations
ranging from 0.8 to 100 lg/ml for an extract and from 0.08
to 10 lg/ml for a pure substance) on two rows of a 96-well
microplate and were tested in triplicate (n = 3). Parasite
growth was estimated by determination of lactate dehydro-
genase activity as described previously in Jonville et al.[14]
Artemisinin (98%; Sigma-Aldrich, Darmstadt, Germany)
was used as positive control.
Selection and cloning of artemisinin
resistant parasites
Plasmodium berghei sensitive to artemisinin, stored in liq-
uid nitrogen, was used. Two hundred microlitres of para-
site suspension was inoculated by caudal intravenous into
donor mice. 105 infected red blood cells from donor mice
Fischer esterification (for 5a-b)
A solution of the chlorogenic acid 4 (0.15 g, 0.42 mmol) in
the appropriate alcohol (20 ml) was heated under reflux
© 2018 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2018), pp. **–**
3

Antimalarial activity of caffeic acid derivatives
Sylvain G. Alson et al.
were diluted in 0.2 ml of 9& NaCl and inoculated by tail
intravenous into two groups of five mice. Immediately, first
group received orally an infra therapeutic dose of artemisi-
Doses of various compounds were administered to mice
for 4 days (day 0–3). 24 h after the last treatment, blood
smears are prepared from tail incision and per cent
reduction in parasitemia was calculated in various groups
and ED50 was determined from doses–effect curves.
nin freshly prepared in
a
solvent system (wa-
ter : DMSO : Tween-80; 85 : 14.7 : 0.3) consisting of
1.25 mg/kg. These mice received artemisinin treatment
once a day for 3 days. After 7 day, parasites which survived
this treatment were subinoculated into uninfected mice
and treatment was repeated with 2.5 mg/kg. Subsequently,
increasing doses have been scaled each week from 2.5 mg/
kg to achieve finally 30 mg/kg (Figure S1). To discard the
possibility that increases in drug tolerance were due to
increased virulence caused by multiples sub-inoculation, an
artemisinin sensitive parasite line was maintained in paral-
lel and passaged in untreated mice the same number of
times as the drug selected line. The evolution of parasitemia
of the two lines within 27 weeks of subinoculation is shown
in Figure S2.
Identification of in-vivo stage sensitivity to
ethyl caffeate
Identification of stage sensitivity for ethyl caffeate was per-
formed on Plasmodium vinckei petteri purchased from Prof.
Grellier (Musꢀeum National d’Histoire Naturelle, Paris,
France). The strain was selected for this study because of its
high synchronicity.[16] Another advantage when working
with this strain is the fact that schizogony occurs 24 h after
inoculation and the timing of the various parasitic stages
can be set up precisely.[17] For convenience, the intraery-
throcytic cycle which occurs within 24 h was subdivided
into five different stages: ring (R), young trophozoite (YT),
mid-term trophozoite (MT), old trophozoite (OT) and
schizont (S). The synchronicity was reinforced by rapid
freezing and thawing which eliminates all or most parasite
stages except the merozoites. At the time of inoculation,
merozoites of P. vinckei petteri rapidly penetrate the ery-
throcytes.[18] When mice were inoculated at T0, R appeared
at T0 + 3 h, YTs at T0 + 6 h, MTs at T0 + 12 h and OTs
at T0 + 18 h.
Parasitic stages were classified using microscope and the
following procedures[19]: R, less than 1/3 of the erythrocyte
(RBC) in size, displaying a large vacuole and a tiny ring of
cytoplasm; YT, about 1/3 of the RBC in size, having a large
vacuole and little or no pigment; MT, between 1/3 and 2/3
of the RBC in size, smaller vacuole than in the previous
stage, and containing tine pigment particles; OT, 2/3 of the
RBC in size, exhibiting small or no vacuole, dense cyto-
plasm, coarse and abundant pigment and S more than 2/3
of RBC in size, no vacuole, dense cytoplasm, abundant pig-
ment and exhibiting more merozoites nuclear.
Mice were infected intraperitoneally with 107 infected
red blood cells. Treatment with ethyl caffeate was started
3 days postinfection when parasitemia was comprised
between 1 and 8%. Animals were distributed into six
groups: five treated groups and one control group. Each
treated group received the same dose of ethyl caffeate at,
respectively, H0, H0 + 3, H0 + 6, H0 + 12 and H0 + 18
(when one of five stages defined above was highly predomi-
nant). They were treated with a single dose of 100 mg/kg
prepared as described for the in-vivo antimalarial test. Fol-
lowing treatment, parasitemia and parasitic pattern (per-
centage of each stages) were evaluated at H0, H0 + 3,
H0 + 6, H0 + 12, H0 + 18, H0 + 24 on thin blood smears
and compared with those of control mice. Analysis of these
parameters enabled the identification of the most sensitive
Evaluation of artemisinin resistance
Resistance quantification in drug-selected parasites clones
was established in the following way. Using a dose–re-
sponse curve, ED50 was assessed in drug-selected parasites
and untreated control lines. A resistance index (RI) was
determined using the following equation: RI = DE50 drug-
selected parasites/DE50 artemisinin-sensitive parasite. At
the end of the experiment, the ED50 of the resistant strain
and that of the susceptible strain were determined and the
results were as following: 36.11 ꢀ 1.39 and 6.36 ꢀ
0.54 mg/kg, respectively, for the isolated resistant strain
and the starting sensitive strain, so a RI of 5.67 compared
to initial strain.
In-vivo antimalarial assay
Animal experiments were approved by the ethical com-
mittee of the «comitꢀe scientifique de recherche et de
dꢀeveloppement de l’Institut Malgache de Recherches
Appliquꢀees» under the number LEPC-IMRA/2014/07/01,
the 18th of July 2014. In-vivo antimalarial activity was
evaluated in the previously developed artemisinin resis-
tant strain of P. berghei with the method described by
Peters.[15] Test compounds were dissolved in 20% Tween
and 80% distilled water, different concentration of com-
pound (25, 50 and 100 mg/kg) were prepared. Twenty-
five mice were infected intraperitoneally with 107 infected
red blood cells on day 0. Animals were divided into five
groups of five mice each (vehicle control group, artemisi-
nin group and 3 experimental groups receiving, respec-
tively: 25, 50 and 100 mg/kg). One hour postinfection
mice were treated with a single dose of test compound
by intraperitoneal administration (0.2 ml/mouse per day).
4
© 2018 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2018), pp. **–**

Sylvain G. Alson et al.
Antimalarial activity of caffeic acid derivatives
stage (or stages) and provided an estimation of the dura-
tion of drug action.
This procedure was also followed to carry out a transesteri-
fication starting from methyl 30,40-dimethoxycinnamate 3
to obtain the corresponding ethyl ester 2g. Minor modifica-
tions of this method were applied to obtain the chlorogenic
derivatives 5a-b (see Figure 1).
Statistical analysis
In-vivo results reflecting inhibition of parasite growth were
analysed using the Kruskal–Wallis test. Individual differ-
ences between treatments were evaluated using Dunn’s test.
Analysis was performed using Prism 5.0 (GraphPad Soft-
ware, La Jolla, CA, USA).
In-vitro antiplasmodial activity
Caffeic and chlorogenic acid and the seven synthetic deriva-
tives were evaluated for their in-vitro antiplasmodial activ-
ity against P. falciparum. Artemisinin was used as positive
control and showed an IC50 value of 3.9 ng/ml. All results
are presented in Table 1.
Determination of LogP
Lipophilicity was estimated (clogP) using Molinspiration
Cheminformatics http://www.molinspiration.com/.
In-vivo antimalarial activity
Methyl and ethyl esters of caffeic acid (2a and 2b), the most
active derivatives, were then evaluated in vivo against
P. berghei. The experiment was conducted on a P. berghei
strain resistant to artemisinin. This strain was developed in
the LEPC laboratory as explained in experimental part. The
methyl and ethyl caffeate derivatives showed a dose-depen-
dent antimalarial activity for the following doses (25, 50
Results
Synthesis
The compounds of interest were prepared from diversely
substituted cinnamic acids 1a-d using Fischer’s method in
the appropriate alcohol with concentrated sulfuric acid.
Figure 1 Schemes for obtaining compounds of interest 2a-g and 5a-b.
© 2018 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2018), pp. **–**
5

Antimalarial activity of caffeic acid derivatives
Sylvain G. Alson et al.
and 100 mg/kg), but only ethyl caffeate at 100 mg/kg
showed a statistically significant activity with a growth inhi-
bition of about 55%. The inhibitions of parasites growth
are detailed in Figure 2.
simple esters.[11,12] Chlorogenic acid and its esters were sig-
nificantly less active than caffeic acid and its esters, but
these compounds could be transformed in caffeic acid in
the digestive track.
Esterification of the carboxylic acid of 1 is correlated
with a four times increase of the in-vitro activity. Small
alkyl esters present similar IC50 values. The importance of
the presence of the 2 phenolic functions was confirmed
after the evaluation of compounds 2d-g: for these four
compounds, a huge reduction of the in-vitro activity was
observed, compared to the corresponding ethyl caffeate
(2b). The presence of two phenolic functions appeared then
mandatory, and this could suggest a link between the
antiplasmodial activity and some antioxidant potential, as
other phenolic compounds with antioxidant activity such
as ellagic acid or gallic acid have demonstrated antiplas-
modial activity.[20,21] In this later case, it is also an ethyl
ester of gallic acid which has demonstrated better activ-
ity.[21] This point has to be investigated in future.
Determination of stage specificity for ethyl
caffeate
Parasites were able to achieve a full cycle throughout the
experiment which allowed us to interpret the results.
Indeed, in the control group which received only the sol-
vent, overall parasitemia increased from 7 to 24% in one
cycle. At beginning (H0), predominant parasitic form is the
YT. Mid-term trophozoites, OTs and S were, respectively,
found at H0 + 3 (until H0 + 6), at H0 + 12 and at
H0 + 18. This kinetic is taken as reference for result inter-
pretation.
Concerning the group treated at H0 with ethyl caffeate,
the parasitemia observed at the end of the cycle (at
H0 + 24), is 8% inferior to that observed in control group.
This difference is not observed when treatment is given at
H0 + 3, H0 + 6 or H0 + 12. We could then conclude that
ethyl caffeate is particularly active on YTs as artemether,
which makes difference with chloroquine. Detailed kinetics
are given in Tables S3–S7.
Many biological activity are attributed to esters deriva-
tives of caffeic acids, such as anti-inflammatory activity,
anti-cancer by induction of apoptosis, antimicrobial,
immunomodulatory, as described in the introductory part
of this manuscript. Furthermore, it has been shown previ-
ously that caffeic is absorbed in rats and has an absolute
bioavailability in humans around 95%.[22,23] To the best of
our knowledge, there are no data available about caffeic
acid ethyl ester bioavailability, but a publication showed
that caffeic acid phenetyl ester was resorbed in rats.[24] Caf-
feic acid and derivatives could then explain, at least in part,
the oral antimalarial activity of traditional medicinal plants
where these compounds are abundant. One hypothesis for
better activity of ethyl caffeate compared to caffeic acid
could be the ability to reach intracellular targets which
Discussion
Ethyl and isopropyl, followed by methyl-caffeate presented
the best antiplasmodial activity with an IC50 around 20 lM
against the 3D7 chloroquine sensitive strain of P. falci-
parum. This range of activity is not extremely potent, but
caffeic acid is present in numerous plants used as food or
medicine and is considered as nontoxic, as probably its
Table 1 In-vitro antiplasmodial activity and Log P estimation of caf-
feic acid derivatives and analogues against Plasmodium falciparum
chloroquine-sensitive (3D7) strain
cLog
Pb
a
Compound
R
R3
R4
IC50
1a
2a
2b
2c
2d
2e
2f
H
OH
OH
OH
OH
H
OH
OH
OH
OH
OH
H
80.5 ꢀ 22.8
0.9
Methyl
Ethyl
Isopropyl
Ethyl
Ethyl
Ethyl
Ethyl
H
27.3 ꢀ 3.4
21.9 ꢀ 9.4
21.9 ꢀ 19.8
181.6 ꢀ 23.0
95.4 ꢀ 21.7
>280
1.6
1.9
2.3
2.4
OH
H
2.4
H
2.9
2g
4
OCH3
OH
OH
OH
OCH3
OH
OH
OH
196.6 (n = 1)
>280
2.6
ꢁ0.4
0.2
5a
5b
Methyl
Ethyl
173.8 ꢀ 87.7
168.9 ꢀ 61.3
0.5
Figure 2 In-vivo antimalarial activity of ethyl caffeate at day 4, fol-
lowing the standard test of Peters, using intra-peritoneal route. Para-
sitemia (%).
aResults expressed in lM ꢀ standard deviation (n = 3–5). bMolinspira-
tion Cheminformatics. http://www.molinspiration.com/.
6
© 2018 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2018), pp. **–**

Sylvain G. Alson et al.
Antimalarial activity of caffeic acid derivatives
could be related to the better lipophilicity of ester deriva-
tives, as illustrated by Log P value in Table 1. Previous
works found in the literature showed that the esterification
of caffeic acid or other phenolics increases their lipophilic-
ity thus facilitating their transmembrane passage.[25,26]
In-vivo antimalarial activity evaluation was conducted
only on the methyl and ethyl caffeic acid derivatives which
were among the three most active compounds and which
were the most easily synthesized compounds. In-vivo assay
confirmed the best activity and dose-dependent effect for
ethyl caffeate.
exhibit interesting activity in vivo, with a growth inhibi-
tion of P. berghei resistant to artemisinin of about 55%
at 100 mg/kg. This is not particularly high compared to
artemisinin or other standard drugs, but this is interest-
ing considering the non-toxicity of ethyl caffeate, and
the abundance of caffeic acid and derivatives in numer-
ous plants, including medicinal plants and food. The
presence of phenols seems to be essential for the activity,
even if methylated derivative maintain some activity.
Caffeic acid derivatives could participate in the anti-
malarial activity of several plants traditionally used
against malaria. It would be interesting, in future, to
evaluate the bioavailability and pharmacokinetics param-
eters of caffeic acid esters, and to evaluate their action
in association with standard antimalarial treatments, par-
ticularly artemisinins.
Only ethyl caffeate was studied for chronotherapy
because of its better in-vivo antimalarial activity. Results
showed that ethyl caffeate would act preferentially in the 6
first hours of the schizogonic cycle by blocking the growth
of young parasites (R and YTs), similarly to what is
described for artemisinin derivatives.
Declarations
Conclusions
Acknowledgements
Caffeic acid and close derivatives or analogues were
shown to exhibit some antiplasmodial activity against
P. falciparum. Particularly, the ethyl ester of caffeic acid
was shown to be the most active derivative of caffeic
acid in vitro and was also shown for the first time to
This work was supported by the University of Liꢁege
Research Grants, n° FSRC-16/39 and by the Belgian Fund
for Scientific Research (grant T.0190.13). G. Degotte
received a PhD fellowship from FRIA-FRS-FNRS.
6. Chalas J et al. Effect of ethyl esterifi-
cation of phenolic acids on low-den-
foods. Crit Rev Food Sci Nutr 1989;
28: 315–347.
sity lipoprotein oxidation. Biomed 12. De P et al. Cinnamic acid derivatives
References
1. World malaria report 2017. Geneva:
World Health Organization; 2017.
DOI: ISBN 978-92-4-156552-3.
Pharmacother 2001; 55: 54–60.
7. Gandhi GR et al. Antihyperglycemic
as anticancer agents – a review. Curr
Med Chem 2011; 18: 1672–1703.
activity and antidiabetic effect of 13. Frederich
M
et al. Antiplasmodial
2. Bellanger A-P et al. Malaria outbreak
in French troops returning from Co^te
d’Ivoire. Scand J Infect Dis 2011; 43:
230–233.
methyl caffeate isolated from Solanum
torvum Swartz. fruit in streptozotocin
induced diabetic rats. Eur J Pharmacol
2011; 670: 623–631.
activity of alkaloids from various
Strychnos species. J Nat Prod 2002; 65:
1381–1386.
NP020070E.
https://doi.org/10.1021/
3. Partnership SCT. Efficacy and safety
of RTS, S/AS01 malaria vaccine with
or without a booster dose in infants
and children in Africa: final results of
a phase 3, individually randomised,
controlled trial. Lancet 2015; 386: 31–
45.
8. Lim DY et al. (+)-2-(1-Hydroxyl-4- 14. Jonville MC et al. Dimeric bisindole
oxocyclohexyl) ethyl caffeate sup-
presses solar UV-induced skin car-
cinogenesis by targeting PI3K, ERK1/
alkaloids from the stem bark of
Strychnos nux-vomica L. Phytochem-
istry 2013; 87: 157–163.
2, and p38. Cancer Prev Res (Phila) 15. Peters W, Robinson BL. The
2014; 7: 856–865.
chemotherapy of rodent malaria. LVI.
Studies on the development of resis-
tance to natural and synthetic
endoperoxides. Ann Trop Med Para-
sitol 1999; 93: 325–329.
9. Balachandran C et al. Antimicrobial
and antimycobacterial activities of
methyl caffeate isolated from Solanum
torvum Swartz. Fruit. Indian J Micro-
biol 2012; 52: 676–681.
4. Chiang Y-M et al. Ethyl caffeate sup-
presses NF-kappaB activation and its
downstream inflammatory mediators,
iNOS, COX-2, and PGE2 in vitro or
in mouse skin. Br J Pharmacol 2005;
146: 352–363.
16. Cambie
G et al. Chronotherapy of
10. Gangan VD et al. Ethyl caffeate ether
derivatives as future potential drug. J
Pharm Res 2014; 88: 818–821.
malaria: identification of drug-sensitive
stage of parasite and timing of drug
delivery for improved therapy. Ann
Parasitol Hum Comp 1991; 66: 14–21.
5. Xu S et al. Ethyl caffeate ameliorates
collagen-induced
suppressing Th1 immune response.
Immunol Res 2017; 2017:
7416792.
arthritis
by
11. Herrmann K, Nagel CW. Occurrence
and content of hydroxycinnamic and 17. Montalvo-Alvarez AM et al. Experi-
J
hydroxybenzoic acid compounds in
mental modifications of the circadian
© 2018 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2018), pp. **–**
7

Antimalarial activity of caffeic acid derivatives
Sylvain G. Alson et al.
rythm of Plasmodium vinckei petteri
metabolic properties. Phytomedicine
2013; 20: 904–912.
resistance of the merozoıte to thaw- 25. Filipe HAL et al. Differential targeting
artemisinin (n = 3).
following cryopreservation; probable
Table S3. Growth kinetics, predom-
inant form and global parasitemia
during a full cycle of Plasmodium
vinckei petteri inoculated on mice
(control group).
€
ing. C R Acad Sci III 1988; 307: 5–10.
18. Cambie G et al. [Timing niches of 3
species of Plasmodium coexisting in a
of membrane lipid domains by caffeic
acid and its ester derivatives. Free
Radic Biol Med 2018; 115: 232–245.
rodent in Central Africa]. C R Acad 26. Wu X-W et al. Anti-inflammatory
Table S4. Group 2 (Treated at Ho),
growth kinetics, predominant para-
sitic form and global parasitemia dur-
ing a full cycle of Plasmodium vinckei
petteri inoculated on mice.
Sci III 1990; 310: 183–188.
phenolic acid esters from the roots
and rhizomes of Notopterygium inci-
sium and their permeability in the
human Caco-2 monolayer cell model.
Molecules 2017; 22: 935.
19. Caillard V et al. Plasmodium vinckei
petteri: identification of the stages sen-
sitive to arteether. Exp Parasitol 1992;
75: 449–456.
Table S5. Group 3 (Treated at
Ho + 3), growth kinetics, predomi-
nant parasitic form and global para-
20. Soh PN et al. In vitro and in vivo
properties of ellagic acid in malaria
treatment. Antimicrob Agents Che-
mother 2009; 53: 1100–1106.
Supporting Information
Additional supporting information
may be found online in the Supporting
Information section at the end of the
article:
sitemia during
a
full cycle of
Plasmodium vinckei petteri inoculated
on mice.
21. Jansen O et al. Antiplasmodial activity of
Mezoneuron benthamianum leaves and
identification of its active constituents. J
Ethnopharmacol 2017; 203: 20–26.
Table S6. Group 4 (Treated at
Ho + 6), growth kinetics, predomi-
nant parasitic form and global para-
Figure S1. Selection and cloning of
artemisinin resistant parasites.
Figure S2. In vivo evolution of
resistance to artemisinin.
22. Olthof MR et al. Chlorogenic acid
and caffeic acid are absorbed in
humans. J Nutr 2001; 131: 66–71.
sitemia during
a
full cycle of
Plasmodium vinckei petteri inoculated
on mice.
Table S1. Ethyl caffeate antimalarial
effect against Plasmodium berghei
strain presenting resistance to artemi-
sinin (n = 3).
23. Wang S-J et al. Bioavailability of caf-
feic acid in rats and its absorption
properties in the Caco-2 cell model.
Pharm Biol 2014; 52: 1150–1157.
Table S7. Group 5 (Treated at
Ho + 12), growth kinetics, predomi-
nant parasitic form and global para-
24. Guo X et al. Antitumor activity of
caffeic acid 3,4-dihydroxyphenethyl
ester and its pharmacokinetic and
Table S2. Methyl caffeate anti-
malarial effect against Plasmodium
berghei strain presenting resistance to
sitemia during
Plasmodium vinckei petteri inoculated
a
full cycle of
on mice.
8
© 2018 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ** (2018), pp. **–**