Synthesis and pharmacochemical study of novel polyfunctional molecules combining anti-inflammatory, antioxidant, and hypocholesterolemic properties

Article abstract of DOI:10.1016/j.bmcl.2005.11.027

We have designed and synthesized a series of novel molecules having a residue of a classical NSAID and an antioxidant moiety, both attached through amide bonds to a known nootropic structure, an l-proline, trans-4-hydroxy-l- proline or dl-pipecolinic acid residue. The compounds were found to retain anti-inflammatory and antioxidant activities, to acquire hypocholesterolemic action, and to possess a greatly reduced gastrointestinal toxicity. The novel molecules could find useful applications, among others, in slowing the progression or delaying the onset of neurodegenerative diseases.

Full text of DOI:10.1016/j.bmcl.2005.11.027

Bioorganic & Medicinal Chemistry Letters 16 (2006) 825–829
Synthesis and pharmacochemical study of novel
polyfunctional molecules combining anti-inflammatory,
antioxidant, and hypocholesterolemic properties
Christos M. Doulgkeris,a Dimitrios Galanakis,a,b Angeliki P. Kourounakis,c
Karyofyllis C. Tsiakitzis,a Antonios M. Gavalas,a Phaedra T. Eleftheriou,a
Panagiotis Victoratos,d Eleni A. Rekkaa and Panos N. Kourounakisa,*
aDepartment of Pharmaceutical Chemistry, School of Pharmacy, Aristotelian University of Thessaloniki, 541 24 Thessaloniki, Greece
bPharmacy Department, ‘‘Papageorgiou’’ General Hospital, Ring Road, Nea Efkarpia, 564 03 Thessaloniki, Greece
cDepartment of Pharmaceutical Chemistry, School of Pharmacy, University of Athens,
Panepistimiopolis Zografou, 157 71 Athens, Greece
dDepartment of Genetics, Development and Molecular Biology, School of Biology, Faculty of Sciences,
Aristotelian University of Thessaloniki, 541 24 Thessaloniki, Greece
Received 24 September 2005; revised 5 November 2005; accepted 7 November 2005
Available online 23 November 2005
Abstract—We have designed and synthesized a series of novel molecules having a residue of a classical NSAID and an antioxidant
moiety, both attached through amide bonds to a known nootropic structure, an L-proline, trans-4-hydroxy-L-proline or DL-pipeco-
linic acid residue. The compounds were found to retain anti-inflammatory and antioxidant activities, to acquire hypocholesterolemic
action, and to possess a greatly reduced gastrointestinal toxicity. The novel molecules could find useful applications, among others,
in slowing the progression or delaying the onset of neurodegenerative diseases.
ꢀ 2005 Elsevier Ltd. All rights reserved.
AlzheimerÕs disease and related dementias (SDAT) are
the greatest unmet medical challenges in neurology, with
over 12 million sufferers in the world. This condition ac-
counts for the majority of dementias diagnosed over the
age of 60. It is characterized by a global, progressive de-
cline of cognitive functions and leaves end-stage patients
bedridden, dependent on custodial care, with death
occurring in about 10 years after diagnosis.1 Pharmaco-
therapy of SDAT has been based on the cholinergic
hypothesis, that is, the dysfunction of the acetylcholine
system contributes to cognitive derangement in SDAT.
Therefore, standard care includes treatment with acetyl-
cholinesterase inhibitors.2 These agents have been prov-
en of limited benefit, while this symptomatic treatment
fails to inhibit the progress of the disease itself.
It is well documented that inflammation3 as well as oxi-
dative stress4,5 are profoundly implicated in a number of
pathobiochemical processes related to neurodegenera-
tive diseases. Furthermore, increased plasma cholesterol
is related to neurodegeneration.6 Thus, the prevention of
these biochemical aberrations occurring in the demented
brain may be a more rational approach toward agents
against this pathological condition. In this investigation,
we have designed a series of novel polyfunctional mole-
cules combining anti-inflammatory, antioxidant, and
hypocholesterolemic properties, which may have the po-
tential to slow or interrupt the progress of the disease
and thus provide a more complete and effective treat-
ment approach.
We have recently shown that the chemical derivatization
of the carboxylic acid group of well-established non-
steroidal anti-inflammatory drugs (NSAIDs) may offer
a viable route to anti-inflammatory agents possessing
an increased safety profile.7 Hence, amidation of the
NSAID molecules with cysteamine8 or cysteine ethyl
ester7 resulted in compounds with increased anti-inflam-
Keywords: Neurodegeneration; L-Cysteine; NSAIDs; Proline amide
conjugates; Antioxidants; Anti-inflammatories; Lipoxygenase; Hypo-
cholesterolemics; Gastrointestinal toxicity.
*
Corresponding author. Tel.: +30 2310 997621; fax: +30 2310
997622; e-mail: panoskur@pharm.auth.gr
0960-894X/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.11.027

826
C. M. Doulgkeris et al. / Bioorg. Med. Chem. Lett. 16 (2006) 825–829
matory and antioxidant properties but with a signifi-
cantly lower GI toxicity. Furthermore, it has been
reported that a proline moiety is part of the pharmaco-
phore of molecules showing nootropic or antiamnesic
activity.9,10 Thus, we have designed molecules of the
general structure I (Table 1) in which a NSAID and
cysteamine or cysteine ethyl ester have been chemically
attached to a proline, 4-hydroxy-proline or pipecolinic
acid moiety.
in vitro for their antioxidant activity by evaluating
their ability to inhibit peroxidation of rat hepatic
microsomal membrane lipids.8,11 Four compounds
inhibited lipid peroxidation by 100% at 1 mM, while
cysteamine and cysteine ethyl ester inhibited lipid per-
oxidation by 37% and 49%, respectively, at the same
concentration. IC50 values of the new compounds
(incubation for 45 min) were (compound, IC50 lM):
2, 390; 4, 320; 6, 500; 7, 122. This action could be
attributed to a combination of proper lipophilicity
with the free HS-group which can readily donate
the sulfydryl H atom, acting as a chain breaking
antioxidant.12 It is noteworthy that the less lipophilic
cysteamine analogues 1, 3, and 5 were much less ac-
tive in this assay (IC50 > 1 mM) compared with the
cysteine ethyl ester analogues 2, 4, and 6. A similar
trend was observed in a series of simpler conjugates
of NSAIDs with cysteamine8 and cysteine ethyl es-
ter,7 the latter being significantly more potent than
the former. The parent NSAIDs and the proline
intermediates II presented negligible antioxidant activ-
ity at 1 mM.
The structures and the synthesis of the novel compounds
1–7 are presented in Table 1. They were prepared by two
successive amidations, initially of the amine group of the
sodium salts of L-proline, trans-4-hydroxy-L-proline or
DL-pipecolinic acid by the acid chloride of the NSAIDs,
and then of the free carboxylic group of the intermediate
II, using 1,1-carbonyldiimidazol (CDI) under the condi-
tions indicated in Table 1. This group of compounds
includes two representative NSAIDs (indomethacin
and naproxen) and two known radical scavengers
(cysteamine and cysteine ethyl ester). The structures
were confirmed spectroscopically (IR, 1H NMR) and
by elemental analysis.
The radical scavenging ability of the molecules was
determined from the extent of their interaction with
the stable free radical DPPH.13 At equimolar concen-
trations, the interaction of the compounds with
DPPH ranged from 56% to 86% (Table 2). The inter-
action of the parent NSAIDs with DPPH at the
same concentration was negligible. This interaction
indicates the reducing potential of the new
compounds.
Compounds 1–7, along with cysteamine and cysteine
ethyl ester as reference compounds, were assayed
Table 1. Synthesis and structures of the prepared compounds
(COCl)2
(a)
O
O
R
OH
R
Cl
Q
HO
(b) (c)
C
N
The anti-inflammatory activity of derivatives 1–7,
along with indomethacin and naproxen as reference
compounds, was assessed from their ability to inhibit
paw edema induced by carrageenan in female Fischer
rats7 (Table 3). The compounds were administered ip
at a dose of 300 lmol/kg and demonstrated a signifi-
cant inhibition of the edema ranging from 14% to
68%. Furthermore, experimental arthritis was produced
in rats by an id injection of complete FreundÕs adju-
vant.14 Compounds 1–4 were administered at a dose
of 300–600 lmol/kg for 14 days, and the arthritic score
was assessed on the 15th day post adjuvant injection.
The tested compounds inhibited arthritis by 50–100%
(Table 4). It can be seen from the results that the
new compounds acquire a good antioxidant and
reducing potential, while they retain considerable
anti-inflammatory activity.
O
H
SH
NH
X
O
OH
O
H
SH
O
O
C
C
H2N X
R
N
R
N
Q
Q
(d) (e) (f)
(I)
(II)
Reagents and conditions: (a) in dichloromethane at 0 ꢁC, 30 min,
3 h rt; (b) in 2 N aq NaOH, 0 ꢁC, 1 h; (c) acidification with 1 N
hydrochloric acid, extraction with CHCl3; (d) in DMF, CDI added
at 0 ꢁC, 1 h rt; (e) cysteamine HCl or cysteine ethyl ester HCl
added at rt and stirred for 24 h; (f) water added, extraction with
CHCl3
Compound
R
Q
X
CH3
O
CH2
CH3
1
2
CH2
CH2
H
COOC2H5
N
O
Table 2. Percent interaction of compounds 1–7 at various concentra-
tions with DPPH (0.2 mM) after 30 min of incubation
Cl
(Indomethacin residue)
Compound
0.2 mM
0.1 mM
0.05 mM
NAa
69.7
78.9
86.3
56.2
75.7
76.7
NAa
44.2
56.3
61.9
36.5
54.4
46.0
NAa
31.1
28.8
32.1
21.3
26.8
23.5
CH3
3
4
5
6
7
CH2
CH2
H
1
2
3
4
5
6
7
H3C
O
(Naproxen residue)
COOC2H5
H
CHOH
CHOH
(CH2)2
COOC2H5
COOH2H5
Compound, mp (ꢁC) and yield (% final reaction): 1, 138–140, 29; 2,
109–111, 30; 3, 151–152, 42; 4, oil, 60; 5, 140–141, 29; 6, semi-solid, 32;
7, oil, 39.
a Not active.

C. M. Doulgkeris et al. / Bioorg. Med. Chem. Lett. 16 (2006) 825–829
827
Table 3. Anti-inflammatory action of compounds 1–7 administered at
300 lmol/kg using the carrageenan paw edema model
dosage of the parent NSAIDs, producing ca. 50% mor-
tality in 4 days, and the compounds under investigation
were administered sc once daily for four days to female
rats. Percent mortality, incidence of perforating gastroin-
testinal (GI) ulcers, melena defecation, and body weight
loss were recorded 24 h after the last treatment (Table 5).
Evidently, the new compounds examined are practically
devoid of GI toxicity at the doses shown in Table 5, in
marked contrast to the parent NSAIDs, that caused
50% mortality and 50–100% GI perforating ulcers. It is
known that the formation of gastric lesions by NSAIDs
is mainly due to inhibition of prostaglandin biosynthesis.
However, the involvement of active oxygen species and
lipid peroxidation in the development of NSAID-in-
duced mucosal damage has also been proposed as an
important event. Thus, indomethacin administration to
rats results in a marked increase of thiobarbituric acid
reactive substance formation in the liver and in the in-
duced gastric lesions,18 while liver glutathione levels are
significantly lowered.19 These findings indicate that
NSAIDs increase the susceptibility of tissues to lipid
peroxidation20 and further suggest the potentially impor-
tant contribution of lipid peroxidation to NSAID-in-
duced ulceration. It is likely that the significantly
reduced GI toxicity of the molecules reported here is
due to a combination of the antioxidant properties of
the compounds and the masking of the free carboxylic
group of the parent NSAID. It is noteworthy that HS-
containing compounds have been shown to possess heal-
ing properties21, and cysteamine and other HS-deriva-
tives have been reported to exert a protective effect
against ethanol- and prostaglandin-inhibition-induced
mucosal damage.22
Compound
% paw weight increasea
% edema inhibition
Control
Indomethacin
58.5 4.3
27.0 4.0
48.5 4.5
50.1 7.0
28.7 4.2
42.9 11.2
44.6 8.9
44.4 9.4
18.4 2.9
34.9 2.9b
—
53.9**
17.0**
14.3*
51.0**
26.7**
23.7**
24.1**
68.5**
50.7**
1
2
Naproxen
3
4
5
6
7
a Each value is the mean from 4 to 9 rats in two independent experi-
ments. Statistical significance compared with controls.
b Compound 7 was tested at a later stage of the project using control
group of animals in which the % paw weight increase was 70.8 10.6.
* P < 0.05.
** P < 0.001 (StudentÕs t test).
Table 4. Effect of the test compounds on FreundÕs complete adjuvant-
induced experimental arthritis in rats
Compound
Arthritic scorea
% arthritis inhibition
1
Control
7.0 0.8
14.3 1.5
1.8 0.9
7.8 2.3
3.8 1.7
7.8 2.3
0.0 0.0
6.8 2.0
50.8**
—
2
77.4*
—
Control
3
51.6*
—
100**
—
Control
4
Control
a Each value is the mean from 4 to 6 rats. Statistical significance
compared with controls.
* P < 0.05.
** P < 0.001 (StudentÕs t test).
The involvement of inflammation3 and oxidative
stress4,5 in the pathobiochemistry of neurodegenerative
diseases is well established. Thus, a number of epidemi-
ological studies show a lower incidence of AD when
NSAIDs are taken on a regular basis.23 In a 15-year lon-
gitudinal analysis, the use of NSAIDs was associated
with a lower incidence of AD.24 However, chronic use
of NSAIDs in such conditions is seriously limited by
their GI-toxicity. Hence, in a double-blind AD treat-
ment trial in which indomethacin was studied with posi-
tive results, approximately one-third of the patients had
to discontinue because of gastrointestinal side effects.25
The new molecules 1–7, being efficient in vivo anti-
inflammatory, potent in vitro lipoxygenase inhibitory
and antioxidant agents, with significantly reduced
GI-toxicity are good candidates for potential use in neu-
rodegenerative diseases.
It has been reported that lipoxygenases and their meta-
bolic products are increased in the pathologically affect-
ed brain regions of AlzheimerÕs disease (AD) patients.15
Inflammatory cascade may take part in aging-associated
neurodegenerative conditions.16 Thus, we examined the
in vitro effect of three of the synthesized compounds
on soybean lipoxygenase [220 U/ml, using sodium lino-
leate, (100 lM), as substrate, at pH 9 and at 28 ꢁC for
5 min].17 They exhibited considerable lipoxygenase inhi-
bition, with IC50 values (compound, IC50 lM): 2, 200; 3,
35; 4, 48.
The gastrointestinal toxicity profiles of compounds 1–4
and of the parent NSAIDs were evaluated in vivo follow-
ing a four-day dosage scheme in rats.7 Thus, equimolar
Table 5. Gastrointestinal toxicity of indomethacin, naproxen, and the novel derivatives in rats
Compound
Dose (lmol/kg)
Mortality (%)a
Perforating ulcers (%)b
Body weight changec
Incidence of melena
Indomethacin
1
84
84
50
0
80
0
À14.9
+1.3
+3.9
+
—
—
+
2
Naproxen
84
0
0
1350
1350
1350
50
0
100
0
À15.8
+3.5
+1.0
3
4
—
—
0
0
a Dead per total · 100.
b Percent of animals developing perforating intestinal ulcers.
c In g/100 g body weight. Standard deviation of the weight change was always within 10% of the average value.

828
C. M. Doulgkeris et al. / Bioorg. Med. Chem. Lett. 16 (2006) 825–829
Increasing evidence indicates that several pathogenic
mechanisms promoting atherosclerosis are also involved
in neurodegenerative diseases. An insight into the
factors determining the susceptibility to atherosclerosis
as well as its long-term progression is of interest for
the understanding of the evolution of such diseases as
AlzheimerÕs.26 It is known that hypercholesterolemia is
a risk factor of SDAT.27 Also cholesterol has been dem-
onstrated to modulate the processing of amyloid precur-
sor protein (APP) to amyloid b peptide Ab42.28 In
addition, apolipoprotein E4 (ApoE4) is a pathological
finding in SDAT,5,29 while the role of atherosclerosis
in vascular dementia is well known.30 It is reported that
increased plasma cholesterol is related to neurodegener-
ation.6 A number of known NSAIDs present a good
anti-dyslipidemic action.31 Furthermore, new potent
and safer derivatives of classical NSAIDs, structurally
related to the compounds reported herein, reduce
considerably plasma LDL-cholesterol.7 We have also
demonstrated that antioxidant properties of novel
anti-dyslipidemic compounds are beneficial for their
action.32 Thus, the novel agents of the present study
were evaluated for their hypocholesterolemic activity
and preliminary results are presented in Table 6. Com-
pounds 2, 4, 5, 6, and 7 were administered ip to
hypercholesterolemic rats and 24 h later the plasma total
cholesterol (TC), LDL-cholesterol, and triglyceride
(TG) concentrations were determined in blood taken
from the aorta.33 The results of this experiment demon-
strate that the test compounds exhibit significant
hypolipidemic activity.
multicausal disorders, such as AlzheimerÕs disease,
can be treated effectively with agents designed to act
at different causes and stages of their pathogenesis.
Thus, we expect that our approach will provide lead
compounds that may be useful in slowing the progres-
sion or delaying the onset of this devastating disease.
These molecules may also find useful applications in
other pathologic conditions such as multiple sclerosis
and atheromatosis, where inflammation and oxidative
stress are also involved.4
References and notes
1. Davis, K. L.; Sammels, S. C. Dementia and Delirium. In
Pharmacological Management of Neurological and Psychi-
atric Disorders, Enna, S. J., Coyle, J.T., Eds.; McGraw-
Hill: New York, 1998; pp 267–316.
2. Palmer, A. M. Trends. Pharmacol. Sci. 2002, 23, 426.
3. Akiyamma, H.; Barger, S.; Barnum, S.; Bradt, B., et al.
Neurobiol. Aging 2000, 21, 383, 30 authors.
4. Halliwell, B., Gutteridge, J. M. C., Eds.; In Free Radicals
in Biology and Medicine, 3rd ed.; Oxford University Press:
Oxford, 1999; a: pp 661–662; b: pp 744–751; c: pp 755–
756; d: pp 625–638.
5. Bassett, C. N.; Montine, T. J. J. Nutr. Health Aging 2003,
7, 24.
6. Yanagisawa, K. J. Neurosci. Res. 2002, 70, 361.
7. Galanakis, D.; Kourounakis, A. P.; Tsiakitzis, K. C.;
Doulgkeris, C. M.; Rekka, E. A.; Gavalas, A.; Kravari-
tou, K.; Charitos, C.; Kourounakis, P. N. Bioorg. Med.
Chem. Lett. 2004, 14, 3639.
8. Kourounakis, A. P.; Galanakis, D.; Tsiakitzis, K.;
Rekka, E. A.; Kourounakis, P. N. Drug Dev. Res. 1999,
41, 9.
9. Gudasheva, T. A.; Voronina, T. A.; Ostrovstaya, R. U.;
Rozantsev, G. G.; Vasilevich, N. I.; Trofimov, S. S.;
Kravshenko, E. V.; Skoldinov, A. P.; Seredenin, S. B. Eur.
J. Med. Chem. 1996, 31, 151.
10. Rault, S.; Foloppe, M.; Robba, M.; Boulouard, M.;
Renard, P.; Devissaquet, M.; Adam, G.; Eur. Pat.
Appl. EP 486386, 1992; Chem. Abstr. 1992, 117,
111468n .
In conclusion, the novel compounds, as designed,
acquired all the desired properties, that is, in vivo
and in vitro anti-inflammatory, antioxidant, low GI-
toxicity, and considerable anti-dyslipidemic action. It
seems that the in vivo actions are due to the whole
molecular entity and not to any liberation of the parent
drug in the organism. This is supported by their much
lower GI toxicity compared with the parent NSAID, as
well as by existing evidence that amide derivatives of
indomethacin and naproxen are not hydrolyzed signif-
icantly in the body.22,34 Furthermore, structurally relat-
ed proline derivatives have been found to be powerful
nootropic and antiamnesic agents. We believe that
11. Vajragupta, O.; Toasaksiri, S.; Boonyarat, C.; Wongkraj-
ana, Y.; Peungvicha, P.; Watanabe, H.; Boonchoong, P.
Free Radical Res. 2000, 32, 145.
12. Halliwel, B. Free Radical Res. Commun. 1990, 9, 1.
13. Aruoma, O. I. Mutat. Res. 2003, 523, 9.
14. Hadjipetrou-Kourounakis, L.; Rekka, E.; Kourounakis,
A. Ann. N.Y. Acad. Sci. 1992, 650, 19.
15. Yao, Y.; Clark, Ch. M.; Trojanowski, J. Q.; Lee, V. M.-
`
Y.; Pratico, D. Annals Neurobiol. 2005, published on-line
21 July 2005.
Table 6. Effect of the test compounds on plasma total cholesterol,
triglyceride, and LDL-cholesterol levels of tyloxapol-induced (200 mg/
kg) hyperlipidemic rats
16. Sugaya, K.; Uz, T.; Kumar, V.; Manev, H. Jpn. J.
Pharmacol. 2000, 82, 85.
17. Rekka, E.; Chrysselis, M.; Siskou, I.; Kourounakis, A.
Chem. Pharm. Bull. 2002, 50, 904.
Compound
Dose (lmol/kg)
% decreasea
TC
LDL
TG
18. Tanaka, J.; Yuda, Y. Biol. Pharm. Bull. 1996, 19, 716.
19. Naik, S. R.; Kalyampur, R.; Sheth, U. K. Biochem.
Pharmacol. 1972, 21, 511.
20. Ursini, F.; Barsacchi, R.; Pelosi, G.; Benassi, A. J.
Biolumin. Chemilumin. 1989, 4, 241.
21. Szabo, S.; Trier, J. S.; Frankel, P. W. Science 1981, 214, 200.
22. Gyires, K. Acta Physiol. Hung. 1992, 80, 247.
23. McGeer, P. L.; Shculzer, M.; McGeer, E. G. Neurology
1996, 47, 425.
24. Stewart, W. F.; Kawas, C.; Corrada, M. Neurology 1997,
48, 626.
2
4
5
6
7
200
56
56
56
56
78.0**
58.0**
45.0*
35.5
48.4*
55.0*
53.5*
56.5*
52.0*
55.7*
68.7**
51.0
25.0
21.3
40.0
a Each value is the mean from 4 to 9 rats in two independent experi-
ments. Statistical significance compared with hyperlipidemic
controls.
* P < 0.05.
** P < 0.001 (StudentÕs t test).

C. M. Doulgkeris et al. / Bioorg. Med. Chem. Lett. 16 (2006) 825–829
829
25. Rogers, J.; Kirby, L. C.; Hempelman, S. R., et al.
Neurology 1993, 43, 1609.
26. Napoli, C.; Palinski, W. Neurobiol. Aging 2005, 26, 293.
27. Ong, W. Y.; Halliwell, B. Ann. NY Acad. Sci. 2004, 1012, 51.
28. Reiss, A. B. Am. J. Alzheimers Dis. Other Demen. 2005, 20,
91.
29. Puglielli, L.; Tanzi, R. E.; Kovacs, D. M. Nat. Neurosci.
2003, 6, 345.
30. Casserly, I.; Topol, E. Lancet 2004, 363, 1139.
31. Kourounakis, A. P.; Victoratos, P.; Peroulis, N.; Stefaniu,
N.; Yangou, M.; Hadjipetrou, L.; Kourounakis, P. N.
Exp. Mol. Pathol. 2002, 73, 135.
32. Chrysselis, M. C.; Rekka, E. A.; Kourounakis, P. N.
Expert Opin. Ther. Patents 2001, 11, 33.
33. Chrysselis, M. C.; Rekka, E. A.; Kourounakis, P. N.
J. Med. Chem. 2000, 43, 609.
34. Shanbhag, V. R.; Crider, A. M.; Gokhale, R.; Harpalani,
A.; Dick, R. M. J. Pharm. Sci. 1992, 81, 149.