Synthesis and antimalarial evaluation of a series of piperazinyl flavones

Article abstract of DOI:10.1016/j.bmcl.2006.11.051

A series of 27 flavonoid derivatives containing a piperazinyl chain have been synthesized and tested for their antiplasmodial activity. Diverse substitution patterns on piperazinyl and flavone moieties were examined and found to affect the activity differ

Full text of DOI:10.1016/j.bmcl.2006.11.051

Bioorganic & Medicinal Chemistry Letters 17 (2007) 959–963
Synthesis and antimalarial evaluation of a series of
piperazinyl flavones
Gwenola Auffret,a Mehdi Labaied,b Franc¸ois Frappier,c Philippe Rasoanaivo,d
Philippe Grellierb and Guy Lewina,*
aLaboratoire de Pharmacognosie, (Univ. Paris-Sud, BIOCIS, UMR-8076 CNRS), Faculte´ de Pharmacie, av. J.B. Cle´ment,
ˆ
92296 Chatenay-Malabry Cedex, France
bUSM 0504 Biologie Fonctionnelle des Protozoaires, Muse´um National d’Histoire Naturelle, RDDM, CP 52, 61 rue Buffon,
75231 Paris Cedex 05, France
cUSM 0502-UMR 5154 CNRS Chimie et Biochimie des Substances Naturelles, RDDM, CP 54, 63 rue Buffon,
75231 Paris Cedex 05, France
dLaboratoire de Phytochimie et de Pharmacologie Cellulaire et Parasitaire, Institut Malgache de Recherches Applique´es,
BP 3833, 101-Antananarivo, Madagascar
Received 17 October 2006; revised 13 November 2006; accepted 14 November 2006
Available online 18 November 2006
Abstract—A series of 27 flavonoid derivatives containing a piperazinyl chain have been synthesized and tested for their antiplasmo-
dial activity. Diverse substitution patterns on piperazinyl and flavone moieties were examined and found to affect the activity dif-
ferently. The most active compounds, which have a 2,3,4-trimethoxybenzylpiperazinyl chain attached to the flavone at the 7-phenol
group, showed in vitro activity against chloroquine-sensitive (Thai) and -resistant (FcB1,K1) Plasmodium falciparum strains in the
micromolar to submicromolar range. One of them was active when given orally in a Plasmodium yoelii nigeriensis infected mouse
model.
Ó 2006 Elsevier Ltd. All rights reserved.
The development of multidrug-resistant Plasmodium
strains represents a major problem in the treatment of
malaria. In a previous publication, we synthesized a ser-
ies of flavonoid derivatives containing a N-benzylpiper-
azine chain, and reported on their ability to potentiate
doxorubicin cytotoxicity on resistant K562/doxorubicin
cells through multidrug resistance (MDR)-modulating
activity.1 More recently, some synthesized chromone
derivatives with a N-substituted piperazine fragment
proved also to be powerful reversal agents of MDR.2,3
As verapamil, a well-known MDR modulator, was
shown to partially reverse Plasmodium falciparum chlo-
roquine resistance,4 we decided to explore the chloro-
quine-potentiating effect of this series of synthetic
flavonoids on P. falciparum. No synergistic effect was
noted, but an intrinsic antiplasmodial activity5 was ob-
served with most of studied compounds, especially flav-
ones 1 and 8 (IC50 0.5 and 0.18 lM) (Table 1). Some
clear structure–activity relationships emerged from this
study: a polymethoxylation pattern of the benzylpiper-
azinyl chain and its fixation at the 7-phenol group,
and an increased lipophilic flavone moiety were favor-
able to the activity. It is noteworthy that both parts (fla-
vone and benzylpiperazinyl) of these hybrid structures
were only weakly active on P. falciparum by themselves:
IC50 of diosmetin (11) and its diether 12 were, respec-
tively, 36 and 70 lM, in the range of literature data
for flavones,6,7 while 2,3,4-trimethoxybenzylpiperazine
(13) and its N-acetyl derivative were almost inactive
(IC50 100 lM) (data not shown). Starting from these
promising results, compound 1, slightly less active than
8 but devoid of cytotoxicity upon the mammalian cells
MCR58 (Table 1), was selected as a model for pharma-
comodulation studies. Therefore, we report in this letter
on the synthesis and the antiplasmodial activity of two
series of analogs of 1, different from the model in the
piperazinyl chain (compounds 20–30), or in the flavone
moiety (compounds 31–36, Figs. 1 and 2).
Keywords: Plasmodium falciparum; Antimalarial drugs; Flavonoids;
Diosmin; Piperazinyl chain.
*
Corresponding author. Tel.: +33 146835593; fax: +33 146835399;
e-mail: guy.lewin@cep.u-psud.fr
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.11.051

960
G. Auffret et al. / Bioorg. Med. Chem. Lett. 17 (2007) 959–963
Table 1. Structure and in vitro antiplasmodial activities of previously synthesized compounds 1–10
OMe
O
5"
3'
OR2
7
O
O
O
4"
3"
N
R1
N
5
2"
OH
Compound
R1
R2
Activitya IC50 (lM)
1
200,300,400-(OMe)3
200,300-(OMe)2
300,400,500-(OMe)3
300,400-(OMe)2
400-Cl
CH2CO2Et
CH2CO2Et
CH2CO2Et
CH2CO2Et
CH2CO2Et
CH2CO2Et
H
0.5b
2
3
4
5
6
7
8
2.4
4.0
4.7
6.0
H
9.5
200,300,400-(OMe)3
200,300,400-(OMe)3
5.3
O
N
0.18d
c
R1
N
OMe
OR2
O
EtO2CH2CO
OR1
O
O
N
9
H
6.2
5.4
c
R1
N
O
10
CH2CO2Et
c
N
N
R1
Chloroquine was used as positive control (IC50 0.18 lM).
a Antiplasmodial activity was evaluated on the D6 strain of P. falciparum.
b Cytotoxicity on MRC5 cells: IC50 > 70 lM.
c
R1 = 200,300,400-(OMe)3.
d Cytotoxicity on MRC5 cells: IC50 = 1 lM. Determinations are means of three independent experiments.
Synthesis of analogs of 1 by modification of the piperazi-
nyl moiety. The synthetic strategy for the preparation of
amides 20–29 involved the preparation of substituted N-
(chloroacetyl)benzylpiperazines 19 and their coupling
through the 7-phenol alkylation of 15, a semi-synthetic
flavone readily available from the natural flavonoid
diosmin (14).9 N-(Chloroacetyl)benzylpiperazines 19a–
e were synthesized from the appropriate benzaldehyde
according to a previously described method.1 N-(Chlo-
roacetyl)benzylpiperazine 19f resulted particularly from
an initial Mannich reaction between 3-methoxycatechol
and ethyl 1-piperazinecarboxylate followed by dialkyla-
tion of the catechol system by CH2Br2.10 Lastly, conden-
sation of tert-butyl 1-piperazinecarboxylate with
chloroacetyl chloride led to the corresponding chloroac-
etamide. Coupling of all these chloroacetylpiperazines

G. Auffret et al. / Bioorg. Med. Chem. Lett. 17 (2007) 959–963
961
mide 19 a (corresponding to the piperazinyl chain of 1)
was condensed with various flavonic moieties. Com-
pounds 31–33 differed from 1 only in the ether group
at C-30, and they were obtained as 1 by replacement of
flavone 15 by its analogs 16–1812 that were synthesized
from diosmetin (11) in three steps (7-O-benzylation; 30-
O-alkylation; hydrogenolysis of the benzyl ether). Com-
pounds 34–36 arose from the coupling of 19a with easily
available or previously prepared 7-hydroxyflavones.13
OMe
OR2
R1O
O
OH O
R1
H
R2
H
11
12
14
Antimalarial evaluation. The in vitro antiplasmodial
activity of the new compounds was first assayed against
the chloroquine-resistant FcB1 strain of P. falciparum,5
and their toxicity was evaluated against the human
diploid embryonic lung cell line MRC5.8 Index of
selectivity (IS = IC50MRC5/IC50P. falciparum) indicates the
specificity of compounds against the parasite. The anti-
plasmodial evaluation of compounds 20–30 (Table 2),
analogs of 1 modified on the piperazinyl chain, confirms
the positive effect of a polymethoxylation pattern (21–24
vs 20, 25), and the importance of a lipophilic character
(quaternization of the piperazinyl chain in 28 and 29
greatly decreases the antiplasmodial activity). Unexpect-
edly, the antiplasmodial activity does not seem to be
correlated with the basicity of compounds: carbamate
26 is much more potent than the N-methylpiperazinyl
compound 27, while reduction of the amide function
weakens the activity (30 vs 1). The only compound with-
in this series with a better activity than 1 is the phenolic
analog 24, but with a less favorable index of selectivity
(IS > 140 and 10, respectively). All the other polyoxy-
genated patterns of the benzylpiperazinyl chain (com-
pounds 21–25) resulted in much lower specificities
toward the parasite, which accounts for the choice of
the 2,3,4-trimethoxybenzylpiperazinyl moiety in the
structure of the following synthesized compounds. The
examination of the second series of analogs (Table 3)
indicates that replacement of the ethoxycarbonylmethyl
ether of 1 by some other ether groups at 30 slightly
decreases antiplasmodial activity (31–33). In other re-
spects, an unsubstituted B ring on the flavone moiety
seems to be detrimental (34, 35 vs 1), while the lipophilic
3,5-di-tert-butyl-4-hydroxyphenol substitution pattern
in 36 maintains the antiplasmodial activity with a good
specificity (IC50 on P. falciparum 1.2 lM, IS > 125).
CH2CO2Et
CH2CO2Et
H
rutinosyl
15
16
17
18
H
H
H
H
CH2CO2Et
CH2CO2t-But
n-hexyl
cyclopentyl
Figure 1.
R4
N
N
O
R3
R1
R1
R2
CH Cl
2
R2
OMe
R3
R4
19 a OMe
19 b
OMe
OMe
OMe
OMe
OBn
OMe
H
H
H
H
19 c OMe
19 d OMe
19 e OMe
H
H
H
OMe
H
OMe
19 f
O C H2 O
H
Figure 2.
with the flavone 15 provided the compounds 20–23, 25,
and 26. 23 was subsequently transformed to 24 by cata-
lytic hydrogenolysis, while 26 led to 27 by hydrolysis of
the Boc function and then N-methylation; finally, the
quaternized derivatives 28 and 29 were prepared from
1 by reaction of the 4000-tertiary amine group, respective-
ly, with meta-chloroperbenzoic acid and methyl iodide.
As easy access to 30 by selective reduction of 1 seemed
uncertain, then we chose to synthesize 30 by the follow-
ing two-step procedure: reductive amination of 2,3,4-tri-
methoxybenzaldehyde by N-(2-hydroxyethyl)piperazine
(NaBH3CN in AcOH) giving the crude benzylpiperazine
which was reacted with 15 under Mitsunobu conditions
(DEAD, PPh3 in THF).11
Considering these results, the four most active com-
pounds (1, 24, 31, and 36) have been evaluated on the
chloroquine-sensitive strain Thai and the chloroquine-
resistant strain K1 of P. falciparum (Table 4). The anti-
plasmodial potencies of compounds 24, 31, and 36 were
homogeneous with a maximum difference of IC50 values
of a factor 3 between the different strains. In contrast,
compound 1 showed a marked weaker activity against
the Thai and K1 strains compared to the FCB1 strain
(IC50 13 lM vs 1 lM, respectively). No apparent corre-
lation was observed with the chloroquine resistance of
strains tested. The antimalarial activity of compounds
1 and 36 was evaluated in vivo against mice infected
by Plasmodium yoelii nigeriensis14 (Table 5). A similar
dose-dependent reduction of parasitemia was observed
for both compounds at day 4 (around 40% of parasite
growth inhibition at 50 mg/kg/day). However, no
Synthesis of analogs of 1 by modification of the flavone
moiety. In the second part of the study, the chloroaceta-

962
G. Auffret et al. / Bioorg. Med. Chem. Lett. 17 (2007) 959–963
Table 2. Synthesis, structure and in vitro antiplasmodial activities of compounds 20–30
OMe
5"
O
R
O
4"
6"
OCH2CO2Et
N
R1
3"
4'''
N
2"
OH O
1
1
R = CO, R = 2",3",4"-trimethoxy
OMe
1
20
21
22
23
24
25
R = CO, R = 4"-methoxy
R = CO, 1= 2",4"-dimethoxy
HO
O
R
OCH2CO2Et
a
e
1
R = CO, R = 2",4",6"-trimethoxy
1
R = CO, R = 4"-benzyloxy-2",3"-dimethoxy
OH O
15
b
1
R = CO, R = 2",3"-dimethoxy-4"-hydroxy
c
OMe
1
O
R = CO, R = 4"-hydroxy-2" and 3"-methylenedioxy
O
O
OCH2CO2Et
N
1
R
30
R = CH2,
= 2",3",4"-trimethoxy
N
26 R = CO2-tbutyl
27 R = CH3
R
d
OH O
Compound
Formula
P. falciparum (FcB1 strain) IC50 (lM)
Cytotoxicity MRC5 IC50 (lM)
ISa
1
C36H40N2O12
C34H36N2O10
C35H38N2O11
1
>140
ndb
5.6
9.5
10
>140
20
21
22
23
24
25
26
27
8.9
1.7
1.9
2.8
0.6
10
3.3
5
C36H40N2O12
C42H44N2O12
C35H38N2O12
C35H36N2O12
3.6
16.7
3.7
2.1
10
37
C31H36N2O11
C27H30N2O9
1.8
26
3.8
nd
28(1, N-oxide)
29(1, iodomethylate)
30
C36H40N2O13
C37H43 I N2O12
19
41
nd
nd
C36H42N2O11
4.3
18
4.2
Reagents and conditions: (a) KHCO3, 19a–f in DMF (120 °C, 2.5 h); (b) H2, Pd–C in MeOH (rt, 1 atm, 3 h); (c) KHCO3, 1-Boc-4-chloroacetyl-
piperazine in DMF (120 °C, 2.5 h); (d) CH2Cl2–TFA 1–1 (rt, 1 h); CH2O, NaBH3CN in AcOH (rt, 2 h); (e) 1-(2-hydroxyethyl)-4-(2,3,4-trimeth-
oxybenzyl)piperazine, P(Ph)3, DEAD in THF (rt, 15 h). Chloroquine was used as positive control (IC50 0.2 lM).
a Index of selectivity defined by the ratio IC50 on MRC5 cells/IC50 on P. falciparum.
b Not determined. Determinations were means of three independent experiments.
Table 3. Structure and in vitro antiplasmodial activity of compounds 31–36
R2
OMe
R3
R4
O
OMe
N
MeO
O
O
O
N
R1
Compound R1
R2
R3
R4
Formula
P. falciparum Cytotoxicity
(FcB1 strains) IC50 (lM) (MRC5 cells) IC50 (lM)
ISa
31
32
33
34
35
36
OH
H
H
H
H
H
OMe OCH2CO2tert-Butyl C38H44N2O12 1.7
OMe O-Hexyl
OMe O-Cyclopentyl
>140
ndb
>82
OH
OH
H
C
38H46N2O10 3.9
C37H42N2O10 2.8
>150
nd
nd
>53
H
H
H
C31H32N2O7
13
14
1.2
OH
H
H
tert-Butyl
C
C
31H32N2O8
39H48N2O8
tert-Butyl OH
>150
>125
Chloroquine was used as positive control ((IC50 0.2 lM).
a Index of selectivity defined by the ratio IC50 on MRC5 cells/IC50 on P. falciparum.
b Not determined. Determinations were means of three independent experiments.

G. Auffret et al. / Bioorg. Med. Chem. Lett. 17 (2007) 959–963
963
Table 4. In vitro antiplasmodial activities of products on chloroquine-
sensitive and chloroquine-resistant strains of P. falciparum
4. Carlton, J. M.; Fidock, D. A.; Djimde, A.; Plowe, C. V.;
Wellems, T. E. Curr. Opin. Microbiol. 2001, 4, 415.
5. Plasmodium falciparum strains were maintained in contin-
uous culture on human erythrocytes as described by
Trager, W.; Jensen, J. B. Science 1976, 193, 673–675. In
vitro antiplasmodial activity was determined by [3H]hy-
poxanthine incorporation according to Desjardins, R. E.;
Canfield, C. J.; Haynes J. D.; Chulay, J. D. Antimicrob.
Agents Chemother. 1979, 16, 710–718. Assays were
performed on the chloroquine-sensitive Thai/Thailand
strain and the chloroquine-resistant FcB1/Colombia and
K1/Thailand strains. The concentration causing 50% of
growth inhibition (IC50) was obtained from the drug
concentration–response curve, and the results were
expressed as mean from three independent experiments.
6. Yenjai, C.; Prasanphen, K.; Daodee, S.; Wongpanich, V.;
Kittakoop, P. Fitoterapia 2004, 89.
Compound
FcB1 strain
IC50 (lM)
Thai strain
IC50 (lM)
K1 strain
IC50 (lM)
1
24
1
13
1.2
13
0.6
1.7
1.2
0.2
0.4
1.4
0.4
0.27
31
36
1.5
0.9
CQ
0.014
Chloroquine (CQ) was used as positive control. Determinations were
means of three independent experiments.
Table 5. In vivo antimalarial activities of compounds 1 and 36a
Dosesb
(mg/kg)
1
36
7. Tasdemir, D.; Lack, G.; Brun, R.; Ruedi, P.; Scapozza, L.;
Perozzo, R. J. Med. Chem. 2006, 49, 3345.
8. Cytotoxicity evaluation on the human diploid embryonic
lung cell line MRC5 was performed according to Labaied,
% inhibitionc
DEADd
% inhibitionc
DEADd
10
50
27.1 9.0
43.1 14.1
2/5
2/5
24.1 5.3
38.5 12.1
5/5
4/5
a Assays were performed on Plasmodium yoelii nigeriensis N67 infected
mice according to the 4-day suppressive test described by Peters
et al.14
b Drugs were administered orally at 0.5 ml/mouse/day
c % inhibition was determined by comparison of the parasitemia
measured in treated mice to that of non-treated mice at day 4.
d Number of dead mice at day 14 (5 mice per group). All mice were
dead in the non-treated group at day 14. Chloroquine (10 mg/kg/ml)
was used as control and cured totally the mice from parasites.
´
M.; Dagan, A.; Dellinger, M.; Geze, M.; Egee, S.;
Thomas, S. L.; Wang, C.; Gatt, S.; Grellier, P. Malaria
J. 2004, 3, 49. Cells were maintained for 5 days in culture
in the presence of drug and the cytotoxicity was deter-
mined using the colorimetric MTT assay according to the
manufacturer’s recommendations (cell proliferation kit I,
Roche Applied Science, France).The concentration caus-
ing 50% of growth inhibition (IC50) was obtained from the
drug concentration–response curve, and the results were
expressed as mean from three independent experiments.
9. ADIR et Cie, French Patent FR 2 623 808.
difference of mice survival was observed between the
non-treated group and the group treated with com-
pound 36. In contrast, a significant increase of mice sur-
vival was constantly observed with compound 1.
10. Clark, J. H.; Holland, H. L.; Miller, J. M. Tetrahedron
Lett. 1976, 38, 3361.
11. Mitsunobu, O. Synthesis 1981, 1.
12. For compound 18, see : ADIR et Cie, European Patent EP
832 886.
13. For compound 36, see: Lewin, G.; Rolland, Y.; Privat, S.;
Breugnot, C.; Lenaers, A.; Vilaine, J.-P.; Baltaze, J.-P.;
Poisson, J. J. Nat. Prod. 1995, 58, 1840.
In conclusion, this pharmacomodulation study showed
that the marked in vitro antiplasmodial potency of some
of the compounds comes from the combination of two
moieties, which are individually of weak activity. A
same observation had already been noticed in this series
for the MDR modulation.1 One of the synthesized com-
pounds exhibits, though in a medium range, an antima-
larial effect in vivo. Until now, only few flavonoids, such
as licochalcone A,15,16 were reported to have in vivo
antimalarial activity. Interestingly, most of the studied
compounds were semi-synthesized from diosmin, an eas-
ily available Citrus flavonoid, which confirms the inter-
est of natural products as raw materials for medicinal
chemistry.
14. In vivo antimalarial assays were performed against Plas-
modium yoelii nigeriensis N67 according to the 4-day
suppressive test of Peters (Peters, W.; Portus, J.H.;
Robinson, B.L. Ann. Trop. Med. Parasitol. 1975, 69,
155–171). Compounds were initially dissolved in DMSO
and subsequent dilutions were made in 0.9% NaCl.
DMSO quantity did not exceed 5%. Day 0, swiss mice
aged between 6 and 8 weeks (25 2 g) were infected
intravenously with 107 parasitized erythrocytes. Mice were
then treated daily starting on day 0 with 10 and 50 mg/kg
of test compound by oral route (0.5 ml). A group of
untreated mice received only the vehicle and a group of
mice was treated orally with chloroquine (10 mg/kg). Each
group was composed of five animals. On day 4, parasi-
temia was determined on a blood smear stained with Diff
Quick reagent and mean of parasitemia standard devi-
ation was determined for each group. Percentage of
parasite growth.
References and notes
´
1. Ferte, G.; Kuhnel, J.-M.; Chapuis, G.; Rolland, Y.;
15. Chen, M.; Theander, T. G.; Christensen, S. B.; Hviid, L.;
Zhai, L.; Kharazmi, A. Antimicrob. Agents Chemother.
1994, 38, 1470.
16. Gutteridge, C. E.; Nichols, D. A.; Curtis, S. M.; Thota, D.
S.; Vo, J. V.; Gerena, L.; Montip, G.; Asher, C. O.; Diaz,
D. S.; DiTusa, C. A.; Smith, K. S.; Bhattacharjee, A. K.
Bioorg. Med. Chem. Lett. 2006, 16, 5682.
Lewin, G.; Schwaller, M. A. J. Med. Chem. 1999, 42, 478.
2. Hadjeri, M.; Barbier, M.; Ronot, X.; Mariotte, A.-M.;
Boumendjel, A.; Boutonnat, J. J. Med. Chem. 2003, 46,
2125.
3. Boumendjel, A.; Nicolle, E.; Moraux, T.; Gerby, B.;
Blanc, M.; Ronot, X.; Boutonnat, J. J. Med. Chem. 2005,
48, 7275.