Antinociceptive properties of caffeic acid derivatives in mice

Article abstract of DOI:10.1016/j.ejmech.2009.06.029

Ten ester derivatives from caffeic acid were synthesized, and their antinociceptive properties are evaluated in mice. The most active compound, dodecyl ester derivative, exhibited potent and dose-related activity against the writhing test, with a calculated ID₅₀ value of 15.1 (11.9-19.1) μmol/kg and MI of 78.8% being several times more active than reference drugs. It was also effective in other experimental models, such as formalin, capsaicin and glutamate-induced pain tests, but was inactive in the hot-plate test. Although the mechanism of action has still not been elucidated, these results appear to support its therapeutic potential against painful diseases.

Full text of DOI:10.1016/j.ejmech.2009.06.029

European Journal of Medicinal Chemistry 44 (2009) 4596–4602
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Antinociceptive properties of caffeic acid derivatives in mice
a
a
a
a
´
Fatima de Campos Buzzi , Caroline Liandra Franzoi , Graziele Antonini , Mauricio Fracasso ,
Valdir Cechinel Filho a, Rosendo Augusto Yunes b, Rivaldo Niero a
,
*
a Programa de Mestrado em Cieˆncias Farmaceˆuticas e Nu´cleo de Investigaç˜oes Qu´ımico – Farmaceˆuticas (NIQFAR)/CCS, Universidade do Vale do Itaja´ı (UNIVALI),
88.302-202, Itaja´ı, SC, Brazil
b Curso de Po´s-Graduaç˜ao em Qu´ımica, Universidade Federal de Santa Catarina (UFSC), 88.040-900, Floriano´polis, SC, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
Ten ester derivatives from caffeic acid were synthesized, and their antinociceptive properties are eval-
uated in mice. The most active compound, dodecyl ester derivative, exhibited potent and dose-related
activity against the writhing test, with a calculated ID50 value of 15.1 (11.9–19.1) mmol/kg and MI of 78.8%
being several times more active than reference drugs. It was also effective in other experimental models,
such as formalin, capsaicin and glutamate-induced pain tests, but was inactive in the hot-plate test.
Although the mechanism of action has still not been elucidated, these results appear to support its
therapeutic potential against painful diseases.
Received 12 February 2009
Received in revised form
8 June 2009
Accepted 29 June 2009
Available online 3 July 2009
Keywords:
Ó 2009 Elsevier Masson SAS. All rights reserved.
Antinociceptive action
Caffeic acid derivatives
Structure–activity relationship
1. Introduction
reference drugs, acetyl salicylic acid (ASA), acetaminophen (ACE)
and morphine (MOR) were included for the purpose of comparison.
Caffeic acid (1) and its derivatives are widely distributed in
medicinal plant (fruits, vegetables, wine, and olive oil, among
others) and are therefore present in human plasma in a diet
dependent concentration [1,2]. Indeed, their esters and amide
2. Chemistry
A scheme of the synthesis of derivatives (2–11) is illustrated in
Fig. 1. All the compounds were synthesized using the esterification
procedure proposed by Fischer with some modifications [12]. In the
present work, acetyl chloride was used as the source of HCl catalyst
in situ (Fig. 1A). Subsequently occurs the nucleophilic attack of
alcohol on the carbonyl group of caffeic acid, protonated by HCl,
giving the respective ester and water (Fig. 1B). In all the syntheses
was used alcohol in excess looking for better yields, accelerate the
reaction time, and solubilize the caffeic acid. When caffeic acid was
not soluble in alcohol, acetone was used as dissolvent. Initially the
alcohol in excess was added to acetyl chloride drop by drop, under
stirring in an ice bath. Caffeic acid was added to this mixture, which
was refluxed until the end of the reaction, monitored by TLC.
derivatives exhibit
a broad spectrum of biological activities,
including anti-oxidative properties, and they have been shown to
scavenge a number of reactive species, including DPPH radicals, and
peroxyl and hydroxyl radicals [3–6]. Furthermore, caffeic acid has
been shown to possess anti-inflammatory and protective effects
against Ni induced oxidative liver damage [7–9]. Studies carried out
with COX enzyme inhibitory assays have demonstrated that some
caffeate derivatives inhibit the cyclo-oxygenases COX-1 and COX-2
enzymes [10]. We have previously demonstrated that some of its
derivative esters present promising anti-inflammatory effects [11].
This evidence prompted a search of antinociceptive agents from
natural sources, and related synthetic derivatives, such as caffeoyl
esters. The present report deals with the evaluation of different
caffeic acid ester derivatives against some models of pain in mice.
Some structural aspects are also discussed. In addition, some
3. Pharmacology
All the compounds (1–11, Table 1) were initially administered
intraperitoneally (i.p) at 10 mg/kg, 30 min before of writhing test,
in an attempt to select the most active compound. Thus, the
dodecyl caffeate (6) was then tested in more specific models using
different doses to evaluate the dose-response effect, calculate the
* Corresponding author. Tel.: þ55 47 3341 7664; fax: þ55 47 3341 7601.
E-mail address: niero@univali.br (R. Niero).
maximum inhibition and ID50
.
0223-5234/$ – see front matter Ó 2009 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2009.06.029

F. de Campos Buzzi et al. / European Journal of Medicinal Chemistry 44 (2009) 4596–4602
4597
O
Cl
Cl
H
R
A
H
H
O
+
O+
O-
O
O
O
H3C
Cl
R
R
R
CH3
CH3
O+
H
-
O
Cl
HCl
+
O
R
CH3
CH3
O+
H
O
B
+
HO
OH
HO
OH
+
HCl
1
HO
HO
HO
H
H
O
O
O
H
O+
O
O+
HO
R
R
H
H
H
HO
HO
O+
O
H
R
O
O
HCl
HO
R
+
H2O
+
HO
HO
HO
2 -CH3
7 -CH(CH3)2
3 -CH2CH3
8 -CH2CH(CH3)2
9 -(CH2)2CH(CH3)2
10 -CH2C6H5
4 -(CH2)3CH3
5 -(CH2)7CH3
6 - (CH2)11CH3
11 -(CH2)2C6H5
Fig. 1. Scheme of synthesis of caffeic acid derivatives.
4. Results and discussion
ten caffeic acid derivatives, first evaluating them against the
writhing test in mice.
Recently, natural products have attracted much attention as
sources of new antinociceptive agents [13–17]. Although caffeic
acid and some derivatives have been pharmacologically investi-
gated against several experimental models, few papers have
reported their antinociceptive activities. We therefore synthesized
Table 1 shows that all the derivatives with aliphatic chain
caused a similar or higher antinociceptive effect than acetyl sali-
cylic acid and acetaminophen, two well-known drugs used as
reference, while the two compounds with aromatic chain demon-
strated lower activity when analyzed against the writhing test in
mice, at 10 mg/kg. Considering that the compound 6 showed
Table 1
Antinociceptive activity of caffeic acid derivatives and reference drugs against acetic
acid-induced abdominal constriction in mice at 10 mg/kg, given intraperitoneally.
ID50 = 5.25 (4.16 - 6.64) mg/kg
70
[15.1 (11.9 - 19.1) µmol/kg]
60
Compounds
Inhibition (%)
1
2
3
4
5
6
7
8
34.6 (ꢀ3.9)*
54.5 (ꢀ2.6)**
58.1 (ꢀ4.4)**
45.0 (ꢀ2.2)**
46.2 (ꢀ2.1)**
78.8 (ꢀ1.5)**
31.0 (ꢀ5.5)*
52.0 (ꢀ2.5)**
75.8 (ꢀ4.5)**
8.8 (ꢀ4.8)
50
**
40
**
30
**
20
10
0
9
10
11
ACE
ASA
Control
3 mg/kg
Compound 6 (i.p.)
6 mg/kg
10 mg/kg
27.5 (ꢀ2.8)**
38.0 (ꢀ1.0)**
35.0 (ꢀ2.0)*
Fig. 2. Effect of 6 (3–10 mg/kg, i.p.) against acetic acid-induced abdominal constric-
tions in mice. Each column represents mean ꢀ S.E.M. of six experimental values.
*Significance levels, when compared with the control group *p < 0.05; **p < 0.01.
Each group represents the mean ꢀ SEM of 6 experiments. *p < 0.05 and **p < 0.01
compared with the respective control values.

4598
F. de Campos Buzzi et al. / European Journal of Medicinal Chemistry 44 (2009) 4596–4602
estimated ID50 value of 1434.8 mmol/kg (Fig. 3), being about two
times less potent than acetyl salicylic acid and equipotent to acet-
aminophen, which presented ID50 values of 605 (516–705) and
45
40
35
30
25
20
15
10
5
ID50 > 500 mg/kg
[~1,434.8 µmol/kg]
22.2%
**
32.4%
**
1145 (708–1846)
mmol/kg [17] respectively, in the same experi-
45.7%
**
mental model. These results suggest that it is absorbed by the
gastrointestinal tract. However, the effect was not as representative
as when administered by the intraperitoneal route.
In an attempt to verify this difference, some molecular proper-
ties and structural factors were analyzed for caffeic acid derivatives,
according to the method proposed by Lipinski [19,20], who evalu-
ates the oral bioavailability in silico (Table 2).
0
control
100 mg/kg
250 mg/kg
500 mg/kg
Recently, Leeson and Springthorpe [21] demonstrated the
influence of drug-like concepts on decision-making in medicinal
chemistry, comparing the physicochemical profiles of recently
discovered oral drugs from four large multinational organizations:
AstraZeneca, GlaxoSmithKline, Merck and Co., and Pflizer. They
demonstrated that the drug lipophilicity is changing less over time
than other physical properties, which suggests that this is an
especially important drug-like property, and its control is impor-
tant for ultimate success in drug development. This is not
surprising, since the role of log P in influencing drug potency,
pharmacokinetics, and toxicity has been established for many
years. This property essentially reflects the key event of molecular
desolvation in the transfer from the aqueous phases to the cell
membranes and protein binding sites, which are mostly hydro-
phobic in nature.
However, if the lipophilicity is too high, there is an increased
likelihood of binding to multiple targets and resultant pharmaco-
logically based toxicology, as well as poor solubility and metabolic
clearance. Therefore, the difference in activity between the two
routes evaluated for compound 6 can be explained in part by its
high lipophilicity (log P ¼ 7.04), since this value is higher than those
established by Lipinski, whose value for good oral bioavailability
must not exceed 5. An extension of the ‘‘rule of five’’ is a rotatable
bond count that is now a widely used filter following the finding
that rotatable bond greater than 10 are correlated with decreased
oral bioavailability. The mechanistic basis for the rotatable bond
Compound 6 (p.o.)
Fig. 3. Effect of
6 (100–500 mg/kg, i.p.) against acetic acid-induced abdominal
constrictions in mice by oral route. Each column represents mean ꢀ S.E.M. of six
experimental values. *Significance levels, when compared with the control group.
*p < 0.05; **p < 0.01.
a more pronounced effect and a better yield, it was evaluated in
other specific models, and compared with the standard drugs used
as reference, such as acetyl salicylic acid (ASA), acetaminophen
(ACE) and morphine (MOR). As can be observed in Fig. 2, the
compound 6 caused a dose-dependent antinociceptive effect with
an ID50 calculated value of 15.1 (11.9–19.1) mmol/kg and maximal
inhibition of 78.8%, being about 9-fold more active than acetyl
salicylic acid and acetaminophen whose values of ID50 were 133
(73–243) mmol/kg and 125 (104–250) mmol/kg, respectively [17].
Local peritoneal receptors are believed to be partly involved in
the abdominal constriction response. The method has been asso-
ciated with prostanoids in general, i.e., increased levels of PGE2 and
PGF2 in peritoneal fluids, as well as lipoxygenase products, by
a
some researchers. Therefore, the results of the acetic acid-induced
writhing strongly suggest that its mechanism of action may be
partly linked to lipoxygenases and/or cyclo-oxygenases [18]. We
have also evaluated this compound by the oral route at 500 mg/kg.
In this model, 6 was effective in dose-dependent form, with an
Table 2
Theoretical studies of solubility and permeability of caffeic acid derivatives by Lipinski’s rule of five.
O
R
HO
HO
O
No
R
No. atom
log Pa
MW
180.1
194.2
208,2
236.3
292.4
348.5
222.2
236.3
250.3
270.3
284.3
No. ONb
No. OHNHc
No. rotb.d
Volume
TPSAe
77.75
66.76
66,76
66.76
66.76
66.76
66.76
66.76
66.76
66.76
66.76
No. viol.f
1
2
3
4
5
6
7
8
–H
–CH3
–CH2CH3
13
14
15
17
21
25
16
17
18
20
21
0.94
1.56
1.93
2.99
5.02
7.04
2.30
2.68
3.21
3.15
3.36
4
4
4
4
4
4
4
4
4
4
4
3
2
2
2
2
2
2
2
2
2
2
2
3
4
154.5
172.0
188.8
222.4
289.6
356.8
205.4
222.2
239.0
243.7
260.5
0
0
0
0
1
1
0
0
0
0
0
–(CH2)3CH3
–(CH2)7CH3
–(CH2)11CH3
–CH(CH3)2
–CH2CH(CH3)2
–(CH2)2CH(CH3)2
–CH2C6H5
–(CH2)2C6H5
6
10
14
4
5
6
9
10
11
5
6
ASA
ACE
MOR
–
–
13
11
21
1.43
0.68
1.16
180.2
151.2
285.3
4
3
4
1
2
2
3
1
0
155.6
140.0
256.7
63.60
49.33
52.9
0
0
0
a
Method for log P prediction developed at Molinspiration (mi log P2.2 – November 2005) is based on group contributions.
b
c
d
e
f
Sum of N and O H-bond acceptors.
Sum of NH and OH H-bond donors.
Number of rotatable bonds.
Topological polar surface area.
Number of violations.

F. de Campos Buzzi et al. / European Journal of Medicinal Chemistry 44 (2009) 4596–4602
4599
ID50 = 23.8 (20.4 - 27.6) mg/kg
[68.3 (58.7 - 79.2) µmol/kg]
ID50 = 26.2 (23.3 -29.5) mg/kg
[75.2 (66.8 - 84.6) µmol/kg]
A
B
250
200
150
100
50
140
120
100
80
60
40
34,4%
**
20.7%
**
56,8%
**
56.2%
**
20
0
0
control
6 mg/kg 10 mg/kg 30 mg/kg
Compound 6 (i.p.)
control
6 mg/kg 10 mg/kg 30 mg/kg
Compound 6 (i.p.)
Fig. 4. Effect of 6 (6–30 mg/kg, i.p.) against formalin-induced pain in mice. A ¼ the first phase (0–5 min) and B ¼ second phase (15–30 min). Each column represents mean ꢀ S.E.M.
of six experimental values. *Significance levels, when compared with the control group. *p < 0.05; **p < 0.01.
filter is unclear, because the rotatable bond count does not correlate
with the in vivo clearance rate, but the filter is reasonable from an in
vitro screening viewpoint because ligand affinity decreases, on
average, by 0.5 kcal for each two rotatable bonds [21]. Thus, this is
another important property in evaluating the results found by the
and capsaicin-induced algesic response), seems to be relevant. It
has been well-documented that the majority of the nonsteroidal
anti-inflammatory drugs analyzed so far are usually ineffective in
preventing formalin or capsaicin-induced neurogenic pain [27].
When analyzed by the glutamate test (Fig. 6), 6 showed
maximum inhibition of 65.2% with calculated ID50 of 110.2 (98.7–
oral route, since compound
6 once again goes beyond the
threshold, showing 14 rotatable bonds.
123.1) mmol/kg, suggesting that this compound could be inhibiting
With respect to the formalin test, 6 caused marked, dose-
dependant inhibition, particularly in the second phase by the
systemic route. The calculated ID50 values were 75.2 (66.8–84.6)
the liberation of neuropeptides from sensory fibers, namely NKs
and kinins involved in the peripheral role of controlling the noci-
ceptive process [27].
and 68.3 (58.7–79.2)
m
mol/kg, with maximum inhibition of 56.2
In the hot-plate test, 6 did not induce any antinociceptive effect.
Since supraspinal and spinal opioid receptors play an important
role in this assay, it is possible that 6 do not act on central opioid
receptors or produce a release of endogenous opioid peptides [28]
(Fig. 7). Since the activity of the central nervous system (CNS)
involves blood–brain barrier (BBB) permeation, various kinds of in
silico prediction methods have been developed [29].
Norinder and Haeberlein [29] proposed two very simple rules to
a compound that could have a high chance of entering in the brain.
The sum of nitrogen and oxygen (N þ O) should be five or less and
log BBB (C log P ꢁ (N þ O)) > 0. Analyzing these parameters, the
compound 6 demonstrated a sum (N þ O) of four and the log BBB of
3.03, suggesting that could be active on the CNS. Although lip-
ophilicity was the first of the descriptors to be identified as impor-
tant for CNS permeation, and C log P correlates nicely with log BBB
and increasing lipophilicity, increasing brain penetration. Mean-
while, the mean value for C log P for marketed CNS drugs is 2.5 [29].
Other important descriptors are polar surface area (PSA) and
rotatable bonds, which has been used as a predictor for blood–brain
barrier (BBB) penetration by many researchers [30,31]. In general,
and 56.8%, respectively (Fig. 4). It is important to note that the
reference drugs practically prevented only the inflammatory effects
(second phase) with ID50 values of 123.0 (77.0–209.0) and 120.0
(90–161) mmol/kg [22], respectively to ASA and ACE and that the
compound 6 was 3-fold more active in this pain model. Drugs that
act primarily on the central nervous system inhibit both phases
equally, while peripherally acting drugs inhibit the late phase. The
early phase is probably a direct result of the stimulation of noci-
ceptors in the paw, and reflects centrally mediated pain, while the
late phase is due to inflammation with a release of serotonin,
histamine, bradylkinin and prostaglandins, and to a certain extent,
the sensitization of the central nociceptive neurons [23–25].
When evaluated against the capsaicin test, which provided
more direct evidence of the antinociceptive effect of this compound
on neurogenic pain, compound
response to intraplantar capsaicin in a dose-dependent manner,
presenting a maximum inhibition of 58.6% at 60 mg/kg, with ID50
6 reduced the licking/biting
calculated of 133.7 (122.7–145.5) mmol/kg (Fig. 5). This suggests its
involvement with the antagonism of the vanilloid receptor (VR1)
[26]. The facts that compound 6 exhibits significant antinociception
when assessed against the neurogenic (first phase of formalin test
300
ID50 = 38.4 (34.4 - 42.9) mg/kg
[110.2 (98.7 - 123.1) µmol/kg]
250
200
150
100
50
120
25.6%
**
ID50 = 46.6 (42.8 - 50.7) mg/kg
[133.7 (122.7 - 145.5) µmol/kg]
100
80
60
40
20
0
46.3%
**
36.9%
**
65.2%
**
58.6%
**
0
control
10 mg/kg
30 mg/kg
60 mg/kg
control
10 mg/kg
30 mg/kg
60 mg/kg
Compound 6
Compound 6
Fig. 6. Effect of 6 (10–60 mg/kg, i.p.) against glutamate-induced pain in mice. Each
column represents mean ꢀ S.E.M. of six experimental values. *Significance levels, when
compared with the control group. *p < 0.05; **p < 0.01.
Fig. 5. Effect of 6 (10–60 mg/kg, i.p.) against capsaicin-induced pain in mice. Each
column represents mean ꢀ S.E.M. of six experimental values. *Significance levels, when
compared with the control group. *p < 0.05; **p < 0.01.

4600
F. de Campos Buzzi et al. / European Journal of Medicinal Chemistry 44 (2009) 4596–4602
chromatography. Melting points were determined on a Micro-
35
30
25
20
15
10
5
**
´
Quımica PF-300 apparatus with a digital thermometer, and are
uncorrected. 1H NMR and 13C spectra were obtained using a Bruker
400 spectrometer. NMR samples were prepared using deuterated
chloroform (CDCl3) purchased from Aldrich Chemical Co. FT-IR
spectra were obtained using a BOMEM spectrophotometer, and the
samples were prepared as pressed KBr plates.
6.1.1. General synthesis of caffeic acid (CA) derivatives
The derivatives were obtained by reaction of caffeic acid with
the appropriate alcohol and acetyl chloride as catalyst under reflux
(2–4 h; 60–70 ꢂC), according to the methodology previously
described [12,32]. The respective products were purified by
recrystallization or chromatographic column over silica gel eluted
with a mixture of hexane:acetone with increasing polarity. The
reaction was considered complete, and the purity of all synthesized
compounds was examined by thin layer chromatography (TLC)
0
Control
10 mg/kg
Compound 6
Morphine
Fig. 7. Effect of 6 (10 mg/kg, i.p.) against hot-plate induced pain in mice. Each column
represents mean ꢀ S.E.M. of six experimental values. *Significance levels, when
compared with the control group. *p < 0.05; **p < 0.01.
using Merck silica gel pre-coated aluminum plates with 200 mm
layer thickness, with several solvent systems of different polarities.
The structural characterization was performed by conventional
spectroscopic data (IR, 1H and 13C NMR) and compared directly
with the literature data [33]. Considering that all the compounds
studied are well-known, we describe below only the synthesis of
the most active compound (6).
drugs aimed at the central nervous system (CNS) tend to have lower
polar surface areas than those of other classes. For molecules to
penetrate the BBB and thus act on receptors in the CNS, the PSA
should preferably be less than 60 Å2. Morphine is the agonist with
maximal intrinsic activity for the opioid system, and has 52.90 Å2 of
PSA, compared with compound 6, for which PSA is 66.76. This is
a negative point against the activity in the hot-plate test. In relation
to the rotatable bonds, CNS drugs have significantly fewer than
other drug classes. Most centrally acting compounds have a rotat-
able count of five or less [29]. Compared with morphine, a standard
drug, it has zero rotatable bonds, while compound 6 has fourteen,
making it unfavorable for action on the CNS.
6.1.2. Synthesis of dodectyl-3,4-dihydroxycinnamate
(dodecyl caffeate) (6)
Caffeic acid (1.8 g, 9.9 mmol), acetyl chloride (0.10 g, 0.5 mmol),
dodecyl alcohol (50 ml), and benzene (50 ml), were placed in
a 250 ml 3-neck round bottomed flask equipped with a magnetic
stirring bar. The mixture was stirred, heated to reflux, and
monitored by thin layer chromatography. TLC considered the
reaction complete after about 4 h of refluxing. The reaction
mixture was concentrated on a rotatory evaporator, and purified
by chromatographic column over silica gel eluted with a mixture
of hexane:acetone with increasing polarity. The purity of the
compound was examined by thin layer chromatography (TLC)
5. Conclusions
We have shown that all the caffeic acid derivatives with
aliphatic chain caused a similar or higher antinociceptive effect
than the drugs used as reference, and the two compounds with
aromatic chain demonstrate lower activity when analyzed against
the writhing test at 10 mg/kg in mice. Compound 6 showed a more
pronounced effect and was then selected for more detailed analysis,
exhibiting an antinociceptive profile in other experimental models
as well. However, the results indicate that it is only slightly absor-
bed by the oral route. This difference may be attributed to certain
molecular properties and structural factors, according to the
method proposed by Lipinski, which evaluates oral bioavailability
in silico and can be explained in part by its high lipophilicity.
Although the mechanisms underlying these effects have not
been elucidated, the findings are of interest because they support,
at least partly, the notion that caffeic acid, a natural compound
widely distributed in plants, may be useful for the development of
new and effective peripheral analgesics against painful diseases.
Studies are currently in progress to determine the possible action
mechanisms responsible for the antinociceptive properties of the
most active compound (6).
using Merck silica gel pre-coated aluminum plates with 200 mm
layer thickness. The final product was recovered as a white crys-
talline solid. Yield: 64%; mp: 161–163 ꢂC. IR (KBr plate) nmax 3495,
3312, 1684, 1635, 1602. 1H NMR (CDCl3)
d
7.52 (d, J ¼ 16 Hz, 1H),
7.15 (d, J ¼ 2 Hz, 1H), 7.04, 7.01 (dd, J ¼ 2.0, 2.1 and 8.2 Hz, 1H), 6.85
(d, J ¼ 8.2 Hz, 1H), 6.27 (d, J ¼ 15.9 Hz, 1H), 3.70 (s, 3H); 13C NMR
(CDCl3)
d 167.9, 148.6, 146.1, 145.7, 127.5, 122.5, 116.3, 115.2, 115.1,
115.1, 51.5.
6.1.3. Solubility and permeability estimate: Lipinski’s rule of five
Computational approaches were used to estimate the solubility
and permeability of the synthesized compounds, using the ‘‘rule of
5’’ proposed by Lipinski and its extensions [19,20]. This rule
predicts that poor absorption or permeation is more likely when
there are more than 5 H-bond donors and, more than 10 H-bond
acceptors; when the molecular weight (MW) is greater than 500;
the calculated C log P is greater than 5 (or M log P > 4.15); and its
extension parameters polar surface area (PSA) more than 140 Å2 or
the sum of the H-bond donors and acceptors is more than 12 and
the rotatable bond more than 10. These physicochemical parame-
ters are associated with acceptable aqueous solubility and intes-
tinal permeability, and comprise the first steps in oral
bioavailability. The values for MW, log P, number of H-bond
acceptors and donors, PSA and rotatable bond were obtained from
the on-line program free molinspiration, by JME Editor, courtesy of
Peter Ertl of Novartis, available on the website: http://www.
molinspiration.com/cgi-bin/properties.
6. Experimental protocols
6.1. Chemistry
Caffeic acid (3,4-dihydroxy-cinnamic acid, 97%) and all alcohols
were purchased from Aldrich Chemical Co. (Milwaukee, WI). The
chromatography column was carried out using 70–230 mesh silica
gel (Aldrich). Silica gel Merck pre-coated aluminum plates with
200
mm
layer thickness were used for the thin layer

F. de Campos Buzzi et al. / European Journal of Medicinal Chemistry 44 (2009) 4596–4602
4601
6.2. Pharmacological evaluation
6.2.6. Glutamate-induced nociception
The animals were treated with compound 6 via i.p. (10, 30 and
60 mg/kg) 30 min before the glutamate injection. A volume of 20
of glutamate solution (30 mol\paw) was injected intraplantarly
under the surface of the right hindpaw, as described by Beirith and
co-workers [27]. After injection, the animals were observed from
0 to 15 min. The time spent licking or biting the injected paw was
timed with a chronometer and considered as indicative of pain.
6.2.1. Animals
ml
Swiss mice (25–35 g), housed at 22 ꢀ 2 ꢂC under a 12-h light/
12 h dark cycle and with access to food and water ad libitum, were
acclimatized to the laboratory for at least 1 h before testing. The
experiments reported here were carried out in accordance with the
current ethical and care guidelines for the care of laboratory
animals and the investigation of experimental pain in conscious
animals [34]. The experiments were approved by the local Ethics
Committee (113/2005-03 UNIVALI). The number of animals (6–8 for
group of treatment) and intensities of noxious stimuli used were
the minimum necessary to demonstrate consistent effects of the
treatments.
m
6.2.7. Hot-plate test
The hot-plate test was used to measure response latencies,
according to the method described by Eddy and Leimback [28]. The
mice were treated with saline solution, morphine (10 mg/kg, sc) or
compound 6 (10 mg/kg, i.p.), and placed individually on a hot plate
maintained at 56 ꢀ 1 ꢂC. The time between placing the animal on
the hot-plate and the occurrence of either the licking of the hind-
paws, shaking the paw or jumping off the surface was recorded as
response latency. Mice with baseline latencies of more than 20 s
were eliminated from the study and the cut-off time for the hot-
plate latencies was set at 30 s. The animals were treated 30 min
before the assay.
6.2.2. Drugs
The following drugs were used: Acetyl salicylic acid, acetamin-
ophen, capsaicin and glutamate hydrochloride (Sigma–Aldrich),
acetic acid and formaldehyde (Merck) and morphine hydrochloride
´
(Cristalia). All the compounds were dissolved in Tween 80
(E. Merck), plus 0.9% of NaCl solution, with exception of capsaicin,
which was dissolved in absolute ethanol. The final concentration of
Tween 80 and ethanol did not exceed 5% and did not cause any
effect per se.
6.3. Statistical analysis
The results are represented as a mean ꢀ SEM, except for the ID50
values (i.e., the dose that reduced responses by 50% relative to the
control values), which are presented as geometric means accom-
panied by their respective 95% confidence limits. The ID50 values
were determined by linear regression GraphPad. Statistical signif-
icance between the groups was calculated by means of analysis of
variance followed by the Newman–Keuls multiple comparison
tests. P-values less than 0.05 (P < 0.05) were considered as indica-
tive of significance.
6.2.3. Acetic acid-induced writhing
Abdominal constriction induced by intraperitoneal injection of
acetic acid (0.6%) was carried out according to the procedures
described previously by Collier and co-workers [35] with minor
modifications. Male Swiss mice (25–30 g) were pretreated with
derivatives (10 mg/kg, i.p.) or compound 6 (3, 6 and 10 mg/kg, i.p. or
100, 250 and 500 mg/kg, p.o.) 30 min before acetic acid injection.
The control animals received a similar volume of vehicle (0.9% NaCl
and Tween 80, 10 ml/kg, i.p.). All the experiments were carried out
at 23 ꢀ 2 ꢂC. After the challenge, pairs of mice were placed in
separate boxes and the number of constrictions of the abdominal
muscles, together with stretching, were counted cumulatively over
a period of 20 min. Antinociceptive activity was expressed as the
reduction in the number of abdominal contractions between the
control animals and the pretreated animals.
Acknowledgments
The authors are grateful to CNPq, ProPPEC/UNIVALI, FAPESC-SC-
Brazil for financial support.
References
[1] M. Nardini, P. Pisu, V. Gentili, F. Natella, M. Di Felice, E. Piccolella, C. Scaccini,
Free Radic. Biol. Med. 25 (1998) 1098–1105.
6.2.4. Formalin-induced nociception
The procedure used was essentially similar to that described
previously [24]. Animals from the same strain were used and 20 ml
[2] E.A. Miles, P. Zoubouli, P.C. Calder, D. Phil, Nutrition 21 (2005) 389–394.
[3] P. Rajan, I. Vedernikova, P. Cos, D.V. Berghe, K. Augustynsa, A. Haemersa,
Bioorg. Med. Chem. Lett. 1 (2001) 215–217.
of 2.5% formalin solution (0.92% formaldehyde), which was injected
intraplantarly into the right hindpaw. After injection, the time
spent licking the injected paw was timed and considered as
indicative of pain. The initial nociceptive scores normally peaked
5 min after formalin injection (first phase) and 15–30 min after
formalin injection (second phase), representing neurogenic and
inflammatory pain, respectively. In order to investigate the possible
antinociceptive action 6, 10 and 30 mg/kg, i.p. of compound 6 was
used.
[4] S. Son, B.A. Lewis, J. Agric. Food Chem. 50 (2002) 468–472.
[5] H. Kikuzaki, M. Hisamoto, K. Hirose, K. Akiyama, H. Taniguchi, J. Agric. Food
Chem. 50 (2002) 2161–2168.
[6] C. Hung, W. Tsai, L.Y. Kuo, Y. Kuo, Bioorg. Med. Chem. 13 (2005) 1791–1797.
[7] P. Michaluart, J.L. Masferrer, A.M. Carothers, K. Subbaramaiah, B.S. Zweifel,
C. Koboldt, J.R. Mestre, D. Grunberger, P.G. Sacks, T. Tanabe, A.J. Dannenberg,
Cancer Res. 59 (1999) 2347–2352.
[8] M. Nardini, C. Scaccini, L. Packer, F. Virgili, Biochim. Biophys. Acta 1474 (2000)
219–225.
[9] L. Pari, A.P. Rasath, Chem. Biol. Interact. 173 (2008) 77–83.
[10] B. Jayaprakasam, M. Vanisree, A. Zhang, D.L. Dewitt, M.G. Nair, J. Agric. Food
Chem. 54 (2006) 5375–5381.
[11] F.M. Da Cunha, D. Duma, J. Assreuy, F.C. Buzzi, R. Niero, M.M. Campos,
J.B. Calixto, Free Radic. Res. 38 (2004) 1241–1253.
[12] D.W. Mayo, R.M. Piked, P.K. Trumper (Eds.), Microscale Organic Laboratory
with Multistep and Multiscale Syntheses, John Wiley & Sons, New York, 1994.
[13] C.R. McCurdy, S.S. Scully, Life Sci. 78 (2005) 476–484.
[14] J.A. Ojewole, J. Ethnopharmacol. 99 (2005) 13–19.
6.2.5. Capsaicin-induced nociception
The procedure used was similar to that described previously [36].
After the adaptation period, 20 ml of capsaicin (1.6 mg/paw) was
injected intraplantarly into the right hindpaw. The animals were
observed individually for 5 min following capsaicin injection. The
amount of time spent licking the injected paw was timed with
a chronometer and was considered as indicative of nociception. The
animals were treated with compound 6 via i.p. (10, 30 and 60 mg/kg)
30 min before capsaicin injection, respectively. The control animals
received a similar volume of saline, intraperitoneally.
[15] J.A. Ojewole, Phytother. Res. 20 (2006) 764–772.
[16] M. Kanegusuku, D. Sbors, S.E. Bastos, M.M. Souza, V. Cechinel-Filho, R.A. Yunes,
F. Delle Monache, R. Niero, Biol. Pharm. Bull. 30 (2007) 999–1002.
[17] L. Santos, L.A. Lima, V. Cechinel-Filho, R. Correˆa, F.C. Buzzi, R.J. Nunes, Bioorg.
Med. Chem. 16 (2008) 8526–8534.
[18] H.O. Vongatau, J. Abbah, I.E. Ngazal, J. Ethnopharmacol. 90 (2004) 115–121.
[19] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Adv. Drug Deliv. Rev. 43
(2001) 3–26.

4602
F. de Campos Buzzi et al. / European Journal of Medicinal Chemistry 44 (2009) 4596–4602
[20] C.A. Lipinski, Drug. Discov. Today Tech. 1 (2004) 337–341.
[21] P.D. Leeson, B. Springthorpe, Nat. Rev. Drug Discov. 6 (2007) 881–890.
[22] M.T. de Souza, F. de Campos-Buzzi, V. Cechinel-Filho, S. Hess, F.D. Monache,
R. Niero, Z. Naturforsch. 62 (2007) 550–554.
[23] A. Tjoelsen, O.G. Berge, S. Hunskaar, J.H. Rolanf, K. Hole, Pain 51 (1992) 5–17.
[24] S. Hunskaar, K. Hole, Pain 30 (1987) 103–114.
[25] M. Shibata, I. Ohkubo, H. Takahashi, R. Inoki, Pain 38 (1989) 347–352.
[26] M.L.J. Caterina, M.A.M. Schumacher, T.A. Tominaga, J.D. Levine, D. Julius,
Nature 389 (1997) 816–824.
[27] A. Beirith, A.R. Santos, A.L. Rodrigues, T.B. Creczynski-Pasa, J.B. Calixto, Eur. J.
Pharmacol. 345 (1998) 233–245.
[29] U. Norinder, M. Haeberlein, Adv. Drug Deliv. Rev. 54 (2002) 291–313.
[30] H. Pajouhesh, G.R. Lenz, NeuroRxÒ 2 (2005) 541–553.
[31] P. Ertl, B. Rohde, P. Selzer, J. Med. Chem. 43 (2000) 3714–3717.
[32] Y. Nudelman, E. Bechor, B. Falb, B.A. Fischer, Synth. Commun. 28 (1998)
471–475.
[33] B. Etzenhouser, C. Hansch, S.C. Kapur, D. Selassie, Bioorg. Med. Chem. 9 (2001)
199–201.
[34] M. Zimmermann, Pain 16 (1983) 109–110.
[35] H.D.J. Collier, L.C. Dinnin, C.A. Johnson, C. Schneider, Br. J. Pharmacol. Che-
mother. 32 (1968) 295–310.
[36] T. Sakurada, K. Katsumata, K. Tan-No, S. Sakurada, K. Kisara, Neuropharma-
cology 31 (1992) 1279–1285.
[28] N.B. Eddy, D. Leimback, J. Pharmacol. Exp. Ther. 107 (1953) 385–393.