Inhibition of COX-1 activity and COX-2 expression by 3-(4′- geranyloxy-3′-methoxyphenyl)-2-trans propenoic acid and its semi-synthetic derivatives

Article abstract of DOI:10.1016/j.bmcl.2011.07.040

Oxyprenylated naturally occurring cinnamic acids displayed efficient and promising biological activities. Aim of this study was to characterize the effects of 3-(4′-geranyl-3′-methoxy)phenyl-2-trans propenoic acid and its selected semi-synthetic analogues

Full text of DOI:10.1016/j.bmcl.2011.07.040

Bioorganic & Medicinal Chemistry Letters 21 (2011) 5995–5998
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Inhibition of COX-1 activity and COX-2 expression
by 3-(40-geranyloxy-30-methoxyphenyl)-2-trans propenoic acid and its
semi-synthetic derivatives
⇑
,
Salvatore Genovese a, , Massimo Curini b, , Paolo Gresele c, , Teresa Corazzi c, , Francesco Epifano a,
a Dipartimento di Scienze del Farmaco, Università ‘G. D’Annunzio’ di Chieti-Pescara, Via dei Vestini 31, 66100 Chieti Scalo (CH), Italy
b Dipartimento di Medicina Interna, Sezione di Medicina Interna e Cardiovascolare, Università degli Studi di Perugia, Via E. Dal Pozzo, 06126 Perugia, Italy
c Dipartimento di Chimica e Tecnologia del Farmaco, Sezione di Chimica Organica, Università degli Studi di Perugia, Via del Liceo, 06123 Perugia, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Oxyprenylated naturally occurring cinnamic acids displayed efficient and promising biological activities.
Aim of this study was to characterize the effects of 3-(40-geranyl-30-methoxy)phenyl-2-trans propenoic
acid and its selected semi-synthetic analogues, on COX-2 expression and activity, and on COX-1 activity,
in purified systems or in whole cell systems. The anti-inflammatory activity of title compounds (1) was
Received 6 June 2011
Revised 9 July 2011
Accepted 11 July 2011
Available online 29 July 2011
tested as inhibition of COX-2 on isolated monocytes stimulated with LPS (10
l
g/ml). COX-2 expression
was completely suppressed when monocytes were incubated with 100
l
M of 3-(40-geranyl-30-methoxy)-
Keywords:
Anti-inflammatory activity
Cycloxygenases
Oxyprenylated natural cinnamic acids
Semi-synthetic oxyprenylated cinnamic
acids
phenyl-2-trans propenoic acid (1) or 3-(40-isopentenyloxy)phenyl-2-trans propenoic acid (4). Moreover
compounds (1) and (4) inhibit dose-dependently LPS-induced COX-2 expression.
Ó 2011 Elsevier Ltd. All rights reserved.
Different functional cycloxygenase (COX) isoforms exist: COX-1
which is assumed to be constitutively expressed in almost all tis-
sues, and COX-2, that is inducible in many tissues by inflammatory
stimuli and may be elicited in vitro in human monocytes by LPS,
some cytokines, and other stimuli.1–3 COX isoforms are encoded
by different genes with distinct pattern of expression,4 but share
the capacity to transform arachidonic acid into prostaglandins
(PGs) and thromboxanes (TXs) in a two-step process involving per-
oxidation and reduction.5 COX activity is inhibited by aspirin and
related non-steroidal anti-inflammatory drugs (NSAID) and has a
role in the production of gastro-protective prostaglandins, in the
regulation of platelet and in kidney function.6 COX-2 and its re-
lated products have renal and cardiovascular functions, are impli-
cated in the pathogenesis of several inflammatory disorders, and in
the pathogenesis of atherosclerosis and tumour formation. In par-
ticular, recent studies have suggested that the overexpression of
COX-2 and resultant overproduction of prostaglandins may be
involved in the development of colon cancer. Selective COX-2
inhibition has thus been actively pursued as a means to obtain
anti-inflammatory agents devoid of gastric and renal toxicity,7
and to obtain candidates agents against the development of cancer
of the colon and other organs.8
However, clinical use of selective COX-2 inhibitors has revealed
unexpected increased incidence of ischemic cardiovascular events
in patients at risk and treated for prolonged periods. One possible
explanation is the unbalance in prothrombotic and antithrombotic
prostanoids generated by COX-2 inhibitors, because these agents
inhibit the production from endothelium of COX-2-generated anti-
thrombotic PGI2, and do not suppress prothrombotic TXA2, derived
from platelet COX-1.9
Oxyprenylated secondary metabolites have been considered for
decades merely as biosynthetic intermediates of C-prenylated
compounds and only in the last 15 years were characterized as
phytochemicals exerting interesting and valuable biological
activities. Considering the length of the carbon chain, three types
of oxyprenyl skeletons can be identified: C5 (isopentenyl), C10
(geranyl) and C15 (farnesyl). Isopentenyloxy and geranyloxy chains
are quite abundant in nature, while farnesyloxy ones are less com-
mon. To date about 300 oxyprenylated derivatives were isolated
and/or synthesized and shown to possess a wide variety of valu-
able and promising pharmacological activities.10 Among these
3-(40-geranyloxy-30-methoxyphenyl)-2-trans propenoic acid (1),
also known as 40-geranyloxyferulic acid, is a prenyloxycinnamic
acid biosynthetically related to ferulic acid in which a geranyl
chain is attached to the phenolic group. It was isolated in 1966
from the bark of Acronychia baueri Schott, an Australian small tree
belonging to the family of Rutaceae (Curini, 2006).11 Although
known for more than four decades, only in the last years some of
⇑
Corresponding author.
E-mail address: s.genovese@unich.it (S. Genovese).
All authors contributed equally to the work.
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.07.040

5996
S. Genovese et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5995–5998
the pharmacological properties of this compound began to be char-
acterized. Since its discovery, this natural product showed valuable
properties, including an anti-inflammatory activity, that have been
recently reviewed.12
The first report about the anti-inflammatory properties of (1)
was reported in 2000 by Koshimizu and co-workers.12 These
authors tested the methyl and ethyl ester of compound (1) and
found that both attenuated the expression of iNOS and to a lesser
70 °C for 24 h to obtain the corresponding cinnamic acid methyl
esters (X = –Cl, 99% yield), (X = –Me, 89% yield). 19All these deriva-
tives were in turn alkylated in position 40 by reaction respectively
with 3,3-dimethylallyl and geranyl bromide promoted by K2CO3 as
the base in acetone at 80 °C for 2 h, followed by basic hydrolysis in
the same vessel and acid/base workup to afford pure isopentenyl
or geranyloxycinnamic acids (5 and 6) (X = –Cl, 99% and 97% yields)
and (7 and 8) (X = –Me, 87% and 92% yields), (Scheme 1).
extent of COX-2, in RAW 264.7 cells stimulated by LPS and IFN-
c
.
The anti-inflammatory activity of compound (1) and its deriva-
tives (2–8) was tested as inhibition of COX-2 on isolated mono-
cytes stimulated with LPS (10 lg/ml). We used rabbit polyclonal
antibodies directed against a unique amino acid sequence present
in human COX-2, but not in COX-1, to detect COX-2-like immuno-
reactivity in isolated monocytes incubated for 20 h in the absence
Soon after, both derivatives were found also to inhibit TPA-induced
Epstein-Barr virus activation in Raji cells without notable cytotox-
icity (81% and 78% inhibition, respectively), but were by far less
effective in attenuating superoxide radicals generation (35% and
45% inhibition, respectively) in HL-60 cells.13 In DLD-1 cells the
methyl ester of 40-geranyloxyferulic acid was shown to suppress
or in the presence of LPS (10 lg/ml). As shown in Figure 2, the anti-
the COX-2 promoter activity at a concentration of 100
l
M with a
COX-2 serum did not recognize any protein in extracts of mono-
cytes incubated for 20 h in the absence of LPS (Ctrl, lane b). After
preincubation with LPS, a band of approximately 72 kDa, exhibit-
ing the same electrophoresis migration of standard purified
COX-2, was evident in monocyte lysates (lane a). As expected,
LPS-induced COX-2 expression was completely suppressed by
50% decrease without marked cytotoxicity in a way similar to nat-
ural polyphenolic compounds such quercetin, genistein, and
resveratrol.13,14
More recently detailed studies on the inflammatory properties
were carried out by Epifano and co-workers using the Croton-oil-
induced ear oedema formation model in mice.15 These authors
showed that that 40-geranyloxyferulic acid (1) inhibited by 41%
5 lM dexamethasone (lane e) (an inhibitor of cycloxygenase-2
[COX-2] induction) used as reference drug.
oedema formation at the dose of 0.3
l
mol/cm2, while the reference
COX-2 expression was completely suppressed when monocytes
anti-inflammatory drug used, indomethacin, gave a 62% inhibition.
On the basis of the reported properties of (1), further evaluation
and deeper insights on the mechanism of the anti-inflammatory
action of 40-geranyloxyferulic acid, of its novel natural and semi-
synthetic derivatives as well as the search of new alternatives to
steroids and non-steroidal anti-inflammatory drugs (NSAIDs) are
fields of current and growing interest. The identification of new
pharmacological agents able to modulate COXs activity and/or
expression might represent a way to obtain an anti-inflammatory
effect without incurring in unwanted adverse events in the cardio-
vascular system. Aim of this study was to characterize the effects
on COX-1 and COX-2 activity and expression, in a purified system
or in human monocytes, of 3-(40-geranyl-30-methoxy)-phenyl-2-
trans propenoic acid (1) and some of its selected natural and
semi-synthetic analogues (2–8) (Fig. 1).
We chose to synthesize and test structural analogues deriving
from shortening the O-side chain from geranyl to isopentenyl
and from substitution of the methoxy group with groups with sim-
ilar steric hindrance but different electronic effects, like methyl
and chlorine. 3-(4’-Geranyloxy)-phenyl-2-trans propenoic acid
from Acronychia baueri (Rutaceae)16 (2) and boropinic acid from
Boronia pinnata Sm. (Fam. Rutaceae),17 and the semi-synthetic com-
pounds, 3-(40-isopentenyloxy)phenyl-2-trans propenoic acid (4)
were synthesized as already reported.18 The synthesis of the other
semi-synthetic 30-chloro and 30-methyl-40-prenyloxycinnamic
acids (5–8) was accomplished starting from commercially avail-
able 3-chloro-4-hydroxybenzaldehyde and 3-methyl-4-hydroxy-
benzaldehyde, respectively.
were incubated with 100 l
M of 3-(40-geranyl-30-methoxy)phenyl-
2-trans propenoic acid (1) or 3-(40-isopentenyloxy)phenyl-2-trans
propenoic acid (4) (lane c and f, respectively). COX-2 expression
was instead not affected by 100
(1.7 0.12 ng/ g protein vs 1.53 0.3 ng/
lated monocytes) and by 100 M (3) (1.78 0.1 ng/
1.53 0.3 ng/ g protein in LPS stimulated monocytes).
l
l
M
of compound (2)
g protein in LPS stimu-
g protein vs
l
l
l
l
Compounds (5–8) had no activity on COX-2 expression (data
not shown). The use of specific anti COX-1 antibody showed that
monocytes not exposed to LPS contained COX-1-like immunoreac-
tivity that was not increased after 24 h of incubation with LPS.
None of the drugs tested affected COX-1 expression (Fig. 3).
Encouraged by the results obtained in this preliminary screening,
we performed further assays using the same test system in order to
calculate IC50 values. To this aim the dose-dependent effect of com-
pounds (1) and (4) on COX-2 expression was tested. Compounds (1)
and (4) inhibit dose-dependently LPS-induced COX-2 expression:
the inhibition was significant at 10
bition, p <0.05 versus LPS stimulated monocytes) and at 25
compound (4) (13% inhibition, p = 0.05), respectively (Fig. 4).
l
M for compound (1) (20% inhi-
lM for
oCOX-1 and oCOX-2 activity assays were then carried out with
purified oCOX-1 and oCOX-2 to determine and compare the effects
of the compounds on oCOX-1 and oCOX-2 activity. After 15 min
CHO
COOMe
a
HO
HO
X
X
These were submitted to a Wittig–Horner olefination by reac-
tion with methyl trimethylphosphonacetate and Na in MeOH at
b, c
COOH
COOMe
R2
R1
O
O
R1
X
7 X = -CH3 R1 = -H
8 X = -CH3 R1 = isopentenyl
5 X = -Cl
6 X = -Cl
R1 = -H
R1 = isopentenyl
1 R1 = -OCH3 R2 = isopentenyl 5 R1 = -Cl
2 R1 = -H R2 = isopentenyl
3 R1 = -OCH3 R2 = -H
4 R1 = -H R2 = -H
R2 = -H
R2 = isopentenyl
7 R1 = -CH3 R2 = -H
8 R1 = -CH3 R2 = isopentenyl
6 R1 = -Cl
Scheme 1. Reagents and conditions: (a) (MeO)3POCH2COOMe (3 equiv), Na
(3 equiv) MeOH, 70 °C, 24 h; (b) isopentenyl bromide (1 equiv) or geranyl bromide
(1 equiv) K2CO3 (1 equiv), acetone , 80 °C, 2 h; (c) NaOH 2 N (aq), reflux, 30 min,
acid/base workup.
Figure 1. Structures of synthesized and tested oxyprenylated cinnamic acids.

S. Genovese et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5995–5998
5997
preincubation, compounds (1) and (4) inhibited oCOX-1 with IC50
values of 54.2 4.5 M and 35.1 0.5 M, respectively. The other
l
l
six compounds (2, 3 and 5–8) did not affect oCOX-1 activity (data
not shown).
The effects of these compounds were tested also on oCOX-2
activity. We found that 3-(40-geranyl-30-methoxy)phenyl trans
propenoic acid (1) and 3-(40-isopentenyloxy)phenyl-2-trans prope-
noic acid (4) (300 lM, respectively) inhibited oCOX-2 (97.5% and
74%, respectively), while compounds (2, 3 and 5–8) did not affect
oCOX-2 activity (data not shown).
From these data we can conclude that 40-geranyloxyferulic acid
(1) and 3-(40-isopentenyloxy)phenyl-2-trans propenoic acid (4) in-
hibit COX-2 expression and the activity of oCOX-1 and oCOX-2.
Trying to explain the mechanism of action underlying the observed
effects on these two pro-inflammatory enzymes, it could be no-
ticed that expression and activity of COXs can be modulated by
several factors,20 among which activation of peroxisome prolifera-
tor receptors (PPARs) play a pivotal role.21 To this aim we have re-
cently demonstrated that 40-geranyloxyferulic acid is a selective
inhibitor of PPARs b/d.22 In terms of structure–activity relationship,
the presence of a geranyloxy side chain coupled to a sterically hin-
dered group, like OCH3, in position 3 of the aromatic ring seems to
represent the best combination for COX inhibitory activities. When
such a group is not present in position 3 shortening the O-side
chain to isopentenyl provided a better activity compared to com-
pounds having a geranyloxy moiety in the para position. The pres-
ence of medium sterical hindrance group like CH3 or Cl abolished
the activity either in the presence of a C10 or C5 O-side chain.
In conclusion, our findings indicate that some natural and
semi-synthetic prenyloxycinnamic acids, structurally derived
from the natural product (1), can be regarded as potential novel
and effective anti-inflammatory agents, the potency of which is
not dissimilar from that of the well known anti-inflammatory
drug dexamethasone. All tested compounds were easily synthe-
sized starting from widely available starting materials, with good
yield and cheap synthetic routes. Two of these derivatives,
namely compounds (1) and (4), showed a very interesting and
promising anti-inflammatory activity in the inhibition of COX-2
expression. The inhibition of both COX-1 activity and of COX-2
expression pointed out by compounds (1) and (4) should under-
line the innovative mechanism of action of this class of com-
pounds. For these reasons the present findings could be
considered as a basis for further studies aiming at better defin-
ing the pharmacological profile of semi-synthetic prenyloxycin-
namic acids. Studies to get further insights into their
mechanism of action as well as to investigate the biological
activity of compounds substituted in positions other than 40
and 30 on the aromatic ring, as well as the effects of functional-
Figure 2. (A) Graphic representation of western blot analysis of COX-2 in human
monocytes. Isolated human monocytes (3 ꢀ 106 cells/ml) were treated for 20 h with
100 lM of compounds 1–4, vehicle of the compounds (Ctrl) and with 5 lM
dexamethasone (a selective COX-2 expression inhibitor) or LPS (10
lg/ml). Cells
were lysed and equal amounts of protein (20 g) were analyzed by SDS–
l
polyacrylamide gel and immunoblotting techniques using specific rabbit polyclonal
antibody directed against COX-2. This figure is representative of three experiments.
(B) Images of Western blot. The upper pannel reports COX-2 expression, while in
the lower pannel beta-actin was shown as equal amounts of protein loading.
Figure 3. Western blot analysis of COX-1 in human monocytes. Isolated human
monocytes (3 ꢀ 106 cells/ml) were treated for 20 h with LPS (10
lg/ml), vehicle of
the compounds (Ctrl), or 100 lM of compounds 1–4. Cells were lysed and equal
amounts of protein (20 lg) were analyzed by SDS–polyacrylamide gel and
immunoblotting techniques using specific rabbit polyclonal antibody directed
against COX-1.
ized O-side chains and
a,b-unsaturated double bonds and iso-
sters of the carboxylic groups, are now ongoing in our
laboratories.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.07.040.
References and notes
1. Porreca, E.; Reale, M.; Di Febbo, C.; Di Gioacchino, M.; Barbacane, R. C.;
Castellani, M. L.; Baccante, G.; Conti, P.; Cuccurullo, F. Immunology 1996, 89,
424.
2. Fu, J. Y.; Masferrer, L. J.; Siebert, K.; Raz, A.; Needleman, P. J. Biol. Chem. 1990,
265, 16737.
Figure 4. Dose-dependent inhibition of LPS-induced COX-2 expression. Graphic
representation of western blot analysis of COX-2 in human monocytes incubated
with increasing concentrations of the compounds 1 and 4. Cells were lysed and
3. Corazzi, T.; Leone, M.; Roberti, R.; Del Soldato, P.; Gresele, P. Biochem. Biophys.
Res. Comm. 2003, 311, 897.
20 lg of protein were analyzed by western blotting techniques using specific rabbit
polyclonal antibody directed against COX-2. This figure is representative of three
4. Crofford, L. J. J. Rheumatol. 1997, 24, 15.
5. Smith, W. L.; Marnett, L. J. Biochim. Biophys. Acta 1991, 1083, 1.
experiments (⁄p <0.05 vs LPS).

5998
S. Genovese et al. / Bioorg. Med. Chem. Lett. 21 (2011) 5995–5998
6. Vane, J. R. Nat. New Biol. 1971, 231, 232.
7. Seibert, K.; Zhang, Y.; Lehay, K.; Hauser, S.; Masferrer, J.; Perkins, W.; Lee, L.;
Isakson, P. Proc. Natl. Acad. Sci. U.S.A. 1991, 91, 12013.
8. Warner, T. D.; Mithchell, J. A. FASEB J. 2004, 18, 790.
9. Fitzgerald, G. A. N. Engl. J. Med. 2004, 351, 1709.
15. Epifano, F.; Genovese, S.; Sosa, S.; Tubaro, A.; Curini, M. Bioorg. Med. Chem. Lett.
2007, 17, 5709.
16. Prager, R. H.; Thregold, H. M. Aust. J. Chem. 1966, 19, 451.
17. Ito, C.; Itoigawa, M.; Otsuka, T.; Tokuda, H.; Nishino, H.; Furukawa, H. J. Nat.
Prod. 2000, 63, 1344.
10. Epifano, F.; Genovese, S.; Menghini, L.; Curini, M. Phytochemistry 2007, 68,
939.
18. Epifano, F.; Curini, M.; Genovese, S.; Blaskovich, M.; Hamilton, A.; Sebti, S. M.
Bioorg. Med. Chem. Lett. 2007, 17, 2639.
11. Curini, M.; Epifano, F.; Genovese, S.; Marcotullio, M. C.; Menghini, L. Anticancer
Agents. Med. Chem. 2006, 6, 571.
12. Hosoda, A.; Nomura, E.; Murakami, A.; Koshimizu, K.; Ohigashi, H.; Mizuno, K.;
Yaniguchi, K. Bioorg. Med. Chem. 2002, 10, 1855.
13. Hosoda, A.; Ozaki, Y.; Kashiwada, A.; Muoth, M.; Wakabayashi, K.; Mizuno, K.;
Nomura, E.; Taniguchi, H. Bioorg. Med. Chem. 2002, 10, 1189.
14. Gresele, P.; Cerletti, C.; Guglielmini, G.; Pignatelli, P.; de Gaetano, G.; Violi, F. J.
Nut. Biochem. 2011, 22, 201.
19. Yang, L. X.; Huang, K. X.; Li, H. B.; Gong, J. X.; Wang, F.; Feng, Y. B.; Tao, Q. F.;
Wu, Y. H.; Li, X. K.; Wu, X. M.; Zeng, S.; Spencer, S.; Zhao, Y.; Qu, J. J. Med. Chem.
2009, 52, 7732.
20. Scher, J. U.; Pillinger, M. H. J. Invest. Med. 2009, 57, 703.
21. Du, H.; Chen, X.; Zhang, J.; Chen, C. Br. J. Pharmacol. doi: 10.1111/j.1476-
5381.2011.01444.x.
22. Genovese, S.; Foreman, J. E.; Borland, M. G.; Epifano, F.; Gonzalez, F. J.; Curini,
M.; Peters, J. M. Life Sci. 2010, 86, 493.