Design, synthesis and in vitro antimalarial activity of an acylhydrazone library

Article abstract of DOI:10.1016/j.bmcl.2005.09.058

A library of acylhydrazone iron chelators was synthesized and tested for its ability to inhibit the growth of a chloroquine-resistant strain of Plasmodium falciparum. Some of these new compounds are significantly more active than desferrioxamine DFO, the

Full text of DOI:10.1016/j.bmcl.2005.09.058

Bioorganic & Medicinal Chemistry Letters 16 (2006) 31–35
Design, synthesis and in vitro antimalarial activity of an
acylhydrazone library
Patricia Melnyk,a,* Virginie Leroux,a Christian Sergheraerta and Philippe Grellier,b
aInstitut de Biologie et Institut Pasteur de Lille, UMR CNRS 8525, Universite´ de Lille II,
1 rue du Professeur Calmette, BP 447, 59021 Lille, France
bLaboratoire de Biologie fonctionnelle des protozoaires, USM0504, Muse´um National dÕHistoire Naturelle,
61 rue Buffon, 75005 Paris, France
Received 18 February 2005; revised 21 September 2005; accepted 22 September 2005
Abstract—A library of acylhydrazone iron chelators was synthesized and tested for its ability to inhibit the growth of a chloroquine-
resistant strain of Plasmodium falciparum. Some of these new compounds are significantly more active than desferrioxamine DFO,
the iron chelator in widespread clinical use and also than the most effective chelators.
ꢀ 2005 Elsevier Ltd. All rights reserved.
Almost one-half of the worldÕs population is exposed to
the burden of malaria, a disease which is responsible for
the death of about 2 million people every year. The
spread of multidrug-resistant Plasmodium falciparum
has highlighted the urgent need to discover new antima-
larial drugs, preferably those affordable to developing
countries where malaria is prevalent.1,2
with these classes of compounds are side effects due to
toxicity.6
Nevertheless, DFO has been extensively used for the
treatment of Fe overload disease,7 has a detectable anti-
malarial activity in humans,8,9 but has not found a role
in the clinical treatment of malaria, probably due to
short plasma half-life, lack of oral activity and high
cost.7
Iron was identified as an essential nutrient for the devel-
opment of the parasite, many enzymes of the plasmodial
metabolic pathways (d-aminolevulinate synthase,
responsible for haemÕs de novo synthesis,3 or ribonucle-
otide reductase, involved in DNA synthesis4) depend on
the presence of this element. For this reason, iron chela-
tors gained a respectable, although still undeveloped,
place among compounds presenting antimalarial activi-
ty. Several possible sources for iron acquisition have
been postulated: host transferrin in the plasma, host
erythrocyte ferritin or, even, host haemoglobin.5
The antimalarial activity of DFO prompted several
authors10,11 to study three aroylhydrazone Fe chelators,
salicylaldehyde isonicotinoyl hydrazone (SIH, 1), 2-
hydroxy-1-naphthylaldehyde m-fluorobenzoyl hydra-
zone (HNFBH, 2) and pyridoxal isonicotinoyl hydrazone
(PIH, 3) (Fig. 1). As these compounds were orally effective
and inexpensive, the results were promising, the hydraz-
ones 1 and 2 providing the best results. An important pre-
requisite of an iron-chelating drug as an antimalarial is a
high affinity for iron. The affinity constant of acylhydraz-
ones12 for iron(III) is about 1028.
Several classes of chelators have been shown to suppress
the growth of P. falciparum in erythrocytes in vitro:
hydroxamate siderophores (DFO for instance, Fig. 1)
and derivatives, catecholamide and catecholate sidero-
phores, a-ketohydroxypyridinones, dihydroxycouma-
rins, acylhydrazones and aminophenols.5 The problems
Another study pointed out the interest of those com-
pounds together with their thiosemicarbazone
analogues.13
Some derivatives of these compounds are proteinase
inhibitors with antiparasitic activity against Trypanosoma
brucei14 and Nifurtimox, a furyl hydrazone, has been
commercialised for the treatment of ChagaÕs disease.
Keywords: Antimalarial; Hydrazone; Library; Iron chelator.
*
Corresponding author. Tel.: +33 3 20 87 12 19; fax: +33 3 20 87 12
33; e-mail: patricia.melnyk@ibl.fr
0960-894X/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.09.058

32
P. Melnyk et al. / Bioorg. Med. Chem. Lett. 16 (2006) 31–35
H
O
H
O
O
O
F
N
N
N
N
H
N
SIH, 1
HNFBH, 2
HO
O
H
O
H
N
N
N
O
O
N
N
H
O
N
N
HN
O
HO
O
HO
+
NH3
N
PIH, 3
Figure 1. Structures of DFO and acylhydrazones 1, 2 and 3.
DFO
OH
A library of 156 acylhydrazones was designed in order
to evaluate the potential of this class of compounds as
antimalarials and to study more in detail the struc-
ture–activity relationships. The libraryÕs diversity was
ensured by the large number of commercially available
aldehydes and hydrazides. This allowed the study of
SAR rules, influence of the compoundsÕ accumulation
in the parasite food vacuole and of the lipophilicity.
To allow a comparison with previous studies, the library
also contained compound 1. As one of the major prob-
lems with malaria is the development of resistant strains,
our efforts were focused towards FcB1, a chloroquine
(CQ)-resistant strain.
is included in the reference section, together with the
quality control (QC) results of the most active
compounds.15
Antimalarial activity and cytotoxicity. The crude com-
pounds were tested for their ability to inhibit parasite
growth (CQ-resistant strain FcB1, IC50 CQ = 126 nM)16
at 1 lM (Table 1). Compounds displaying inhibition of
parasite growth by more than 50% were directly submit-
ted to further pharmacological characterisation (IC50,
Table 2) as their purity was over 95%.
Most active compounds were submitted to a cytotoxicity
test (CC50) on a human diploid embryonic lung cell line
(MRC-5) using the colorimetric MTT assay17 (Table 2).
The acylhydrazones were synthesized in deepwell plates
at 10 lmol scale according to Scheme 1. The purity was
assessed by HPLC and all the mass spectra were consis-
tent with the anticipated product structure. Out of the
total 156 compounds of the library, 153 were obtained
with a purity over 85% and were submitted to biological
activity evaluation (Fig. 2). The experimental procedure
In vitro inhibition of b-hematin formation. Compounds
were tested for their ability to inhibit formation of b-he-
matin (the synthetic equivalent of hemozoin) induced by
1-monooleoyl glycerol (MOG)18,19 (Table 2).
Introduction of a variety of substituents either on the
aldehyde or on the hydrazone moiety provided com-
pounds with modest or low inhibition percentage of par-
asite growth at 1 lM (Table 1), except for 11
compounds (compounds 2–12). Reference compound 1
is representative of the results with 22% of growth
inhibition.
O
OH
OH
O
DMF
+
R'
NH
O
N
r.t. overnight
N
R'
NH2
R
R
H
H
Scheme 1. Synthesis of acylhydrazones.
Very low inhibition percentages were obtained in the
case of methylhydrazide (0–34%) or benzylhydrazide
(0–45%) whatever the structure of aldehyde partner.
The introduction of polar or ionizable groups led to
no improvement, either on hydrazide partner (pyridine,
phenol substituent), or on the aldehyde moiety (3, 4 or
5-hydroxy substituent). The highest inhibition was ob-
tained for aromatic partners, substituted by tert-butyl,
chloro, nitro, methyl or methoxy groups. The results
suggested the importance of rather hydrophobic or
bulky substituents.
160
140
120
100
80
100
90
80
70
60
50
40
30
20
10
0
60
40
20
0
purity
Compounds displaying inhibition of parasite growth by
more than 50% were submitted to an IC50 evaluation
Figure 2. Quality control of the library.

P. Melnyk et al. / Bioorg. Med. Chem. Lett. 16 (2006) 31–35
33
Table 1. Antimalarial activity on CQ-resistant strain FcB1 (inhibition of parasite growth at 1 lM)
N+ O
O
OH
Cl
O
R
R0:
CH3–
O
S
N
N
H
21%
42%
20%
41%
13%
45%
23%
21%
33%
37%
14%
35%
16%
27%
26%
19%
30%
0%
1: 22%
28%
11%
43%
nda
35%
36%
30%
10%
14%
32%
2: 60%
0%
26%
31%
16%
32%
23%
38%
2%
13%
14%
29%
16%
13%
24%
37%
19%
24%
23%
8: 52%
38%
21%
11%
5%
25%
24%
13%
36%
16%
8%
14%
30%
21%
37%
29%
9%
43%
35%
4%
0%
Naphthyl
3-OH
3-OMe
4-OH
4-OMe
4-NEt2
5-OH
5-OMe
5-Me
48%
24%
32%
18%
45%
4: 69%
28%
14%
46%
11: 81%
17%
31%
11%
29%
33%
nda
20%
16%
15%
14%
12%
4%
41%
39%
37%
43%
41%
nda
35%
18%
26%
34%
8%
4%
45%
25%
24%
5: 52%
9: 69%
36%
31%
39%
39%
18%
28%
10: 53%
29%
26%
3: 73%
38%
47%
34%
25%
3%
4%
7%
45%
19%
14%
33%
0%
38%
49%
7: 59%
47%
21%
24%
33%
32%
21%
19%
20%
24%
31%
10%
4%
6: 65%
48%
32%
32%
17%
22%
5-tBu
5-Br
5-NO2
40%
29%
15%
12: 52%
36%
a Not determined.
Table 2. Biological activity of the 11 most active compounds
a
R
R0
Antimalarial
CC50 (lM)
(MRC5 cells)
Inhibition of b-hematin
formation IC50 (lM)
VAR · E+3
a
a
activity IC50 (lM)
CQ
2
—
—
Phenyl
0.126 0.005
1.1 0.3
0.9 0.1
1.0 0.1
1.9 0.1
2.2 0.2
1.3 0.2
1.6 0.2
2.9 0.6
2.3 0.3
1.6 0.1
1.7 0.4
50
4
60 10
50 ꢀ 100
>100
54
68
120
86
59
60
71
60
59
60
57
46
4-NEt2
4-NEt2
4-NEt2
5-Me
5-Me
5-tBu
5-tBu
5-tBu
5-tBu
5-tBu
5-Br
1.1 0.1
1.1 0.1
2.5 0.7
3
4-tert-Butyl-phenyl
4-Nitro-phenyl
4-Chloro-phenyl
4-tert-Butyl-phenyl
Phenyl
4
>100
5
>100
16.3 2.6
2.6 0.8
50 ꢀ 100
50 ꢀ 100
>100
6
7
8
2-Thienyl
12
3.6 0.8
>100
1.1 0.2
>100
4
50 ꢀ 100
>100
>100
9
4-Chloro-phenyl
4-Methoxy-phenyl
4-Nitro-phenyl
4-tert-Butyl-phenyl
10
11
12
50 ꢀ 100
>100
a Mean IC50 SD values for two to three independent experiments are shown.
(Table 2). Highest inhibition was obtained for 2-hy-
droxy-4-diethylaminobenzaldehyde 4-tert-butyl-benzoyl
hydrazone 3 with an IC50 of 900 nM. But the range of
IC50 is not large enough to draw some rules.
better VAR values. Compounds presenting the highest
activity accumulate in the same range as CQ.
CQ inhibits the polymerization process, thus, leading to
poisoning of the parasite. DFO seems to have an oppo-
site behaviour.24 We have tested the ability of the com-
pounds to inhibit this process of detoxification of haem
(Table 2) to get some information on the mechanism of
action. Even if the compounds are known as inhibitors
via direct coordination of iron(III), the mechanism will
be different from CQ which inhibits the crystallisation
of the haem to form the hemozoin (and its equivalent
b-hematin) through p–p stacking. The most active com-
pounds are equivalent or less potent inhibitors than CQ
and the best inhibitors (2 compounds provided an IC50
of about 20 lM) are totally inactive on parasite growth
(see supplementary materials). The inhibition of this
process does not seem to be the base of the antimalarial
activity of this class of compounds. It can be notified
that no significant red blood cells (RBC) haemolysis
was recorded after 48 h of incubation with the more ac-
tive compound at a concentration totally inhibiting the
parasite growth (20 lM), as well as no alteration of
the RBC morphology was observed by phase contrast
microscopy suggesting that the inhibition of growth
The partition coefficient logD at pH 7.4 was evaluated
in silico20 to determine the influence of lipophilicity on
the antimalarial activity. As expected, the 11 most active
compounds provided the highest values, between 3.5
and 5.2. This direct correlation of a compound lipophil-
icity with its antimalarial activity has already been
experimentally demonstrated in the case of reversed
siderophores.11,21
Host haemoglobin is degraded in the food vacuole of the
parasite, and haem liberated in this process is polymer-
ized to hemozoin.
In order to evaluate the ability of the component to
accumulate in the parasite food vacuole, in silico calcu-
lations of vacuolar accumulation ratios (VAR) were car-
ried out based on a weak-base model.20,22 A wide range
of accumulation ratios were obtained (from 640 to
3.6 · 109, see supplementary materials). As expected,
compounds with phenol groups as substituents provided

34
P. Melnyk et al. / Bioorg. Med. Chem. Lett. 16 (2006) 31–35
10. Tsafack, A.; Loyevski, M.; Ponka, P.; Cabantchik, Z. I.
J. Lab. Clin. Med. 1996, 127, 575.
could not be attributed to indirect effects due to modifi-
cations of the host cell membrane.
11. Clarke, C. J.; Eaton, J. W. Clin. Res. 1990, 38, 300A.
12. Ponka, P.; Richardson, D. R.; Edward, J. T.; Chubb, F. L.
Can. J. Physiol. Pharmacol. 1994, 72, 659.
13. Walcourt, A.; Loyevsky, M.; Lovejoy, D. B.; Gordeuk, V.
R.; Richardson, D. R. Int. J. Biochem. Cell B 2004, 36,
401.
14. (a) Caffrey, C. R.; Schanz, M.; Nkemgu-Njinkeng, J.;
Brush, M.; Hansell, E.; Cohen, F. E.; Flaherty, T. M.;
McKerrow, J. H.; Steverding, D. Int. J. Antimicrob.
Agents 2002, 19, 227; (b) Greenbaum, D. C.; Mackey,
Z.; Hansell, E.; Doyle, P.; Gut, J.; Caffrey, C. R.; Lehman,
J.; Rosenthal, P. J.; McKerrow, J. H.; Chibale, K. J. Med.
Chem. 2004, 47, 3212.
The average cytotoxicities of acylhydrazones upon MRC-
5 cells extended from 1 lM to more than 100 lM (Table
2). Three compounds showed a reproducible lack of
toxicity at 100 lM: 2-hydroxy-5-methyl-benzaldehyde
4-chloro-benzoyl hydrazone 5, 2-hydroxy-5-tert-butyl
benzaldehyde 4-methoxy-benzoyl hydrazone 10 and
2-hydroxy-5-bromobenzaldehyde 4-tert-butyl-benzoyl
hydrazone 12. All the other compounds provided a selec-
tivity index (ratio CC50/IC50) between 1 and 16, too low
for a potential development as drug candidates.
15. In deepwell plates, 100 lL of aldehyde (0.1 M in DMF) was
added to 50 lL of hydrazide (0.2 M in DMF). The mixture
was stirred at room temperature overnight. Five microlitres
was removed for QC analysis. Reaction plates and QC
plates were evaporated in a Genevac EZ2 evaporator. QC
aliquots were solubilized with 5 lL DMF and 200 lL
CH3CN before HPLC and MALDI/TOF analysis.
A library of 153 acylhydrazones was synthesized with
high purity. Though the activity is in the micromolar
range against P. falciparum growth, these compounds
have the highest antimalarial activity of this class of iron
chelators. As three of the compounds present no detect-
able toxicity, further studies of their biological activity
are in progress. From a mechanistic point of view, as
ribonucleotide reductase is a target for HNFBH 2,25
an investigation of the potency of this library on this en-
zymeÕs activity is also intended.
Compound tR (min) Purity (%) m/z [M+H]+ found
(calcd)
2
3
5.29
7.18
5.90
7.60
8.56
8.14
8.05
8.74
8.20
8.41
9.05
97
96
312.2 (311.16)
368.2 (367.23)
357.1 (356.15)
289.1 (288.07)
311.2 (310.17)
297.1 (296.15)
303.1 (302.11)
331.1 (330.11)
327.1 (326.16)
342.1 (341.14)
375.1–377.4
4
>99
99
Acknowledgments
5
6
>99
>99
>99
>99
>99
>99
95
´
We express our thanks to Herve Drobecq for MALDI
spectra. These works are supported by Universite de
7
´
8
Lille II and PAL+ French Research Ministry
program.
9
10
11
12
Supplementary data
(374.06–376.06)
QC results, inhibition of haem polymerization, VAR in
silico calculations for the entire library. Supplementary
data associated with this article can be found, in the on-
line version, at doi:10.1016/j.bmcl.2005.09.058.
16. (a) Trager, W.; Jensen, J. B. Science 1976, 193, 673; (b)
Desjardins, R. E.; Canfield, C. J.; Haynes, J. D.; Chulay,
J. D. Antimicrob. Agents Chemother. 1979, 16, 710.
17. Mossman, T. J. Immunol. Methods 1983, 65, 55.
18. Fitch, C. D.; Cai, G.; Chen, Y.-F.; Shoemaker, J. D.
Biochim. Biophys. Acta 1999, 1454, 31.
19. Ayad, F.; Tilley, L.; Deady, L. W. Bioorg. Med. Chem.
Lett. 2001, 11, 2075.
20. Values of LogD and LogP at pH 5.0 and pH 7.4
were calculated using ACD/pKa DB software from
Advanced Chemistry Development Inc., Toronto,
Canada.
21. (a) Shanzer, A.; Libman, J.; Lytton, S. D.; Glickstein, H.;
Cabantchik, Z. I. Proc. Natl. Acad. Sci. U.S.A. 1991, 88,
6585; (b) Pradines, B.; Rolain, J. M.; Ramiandrasoa, F.;
Fusai, T.; Mosnier, J.; Rogier, C; Daries, W.; Baret, E.;
Kunesh, G.; Le Bras, J.; Parzy, D. J. Antimicrob.
Chemother. 2002, 50, 177.
References and notes
1. Geary, T. G.; Jensen, J. B.; Ginsburg, H. Biochem.
Pharmacol. 1986, 35, 3805.
2. Yayon, A.; Cabantchik, Z. I.; Ginsburg, H. EMBO J.
1984, 3, 2695.
3. Bonday, Z. Q.; Taketani, S.; Gupta, P. D.; Padmanaban,
G. J. Biol. Chem. 1997, 272, 839.
4. Wrigglesworth, J. M.; Baum, H. In The Biochemical
Function of Iron; Jacobs, A., Worwood, M., Eds.;
Academic Press: New York, 1980; p 29.
5. Loyevsky, M.; Gordeuk, V. R. In Antimalarial Chemo-
therapy; Rosenthal, P. J., Ed.; Humane Press Inc.:
Totowa, NJ, 2001, Chapter 17.
6. Bernhardt, P. V.; Caldwell, L. M.; Chaston, T. B.; Chin,
P.; Richardson, D. R. J. Biol. Inorg. Chem. 2003, 8, 866.
7. Richardson, D. R. Exp. Opin. Invest. Drugs 1999, 8, 2141.
8. Gordeuk, V. R.; Thuma, P. E.; Brittenham, G. M.; Zulu,
S.; Simwanza, G.; Mhangu, A.; Flesch, G.; Parry, D.
Blood 1992, 79, 308.
22. Vacuolar accumulation ratios (VARs) were calculated
from the equation below:
P
P
4
n¼1
n
ai
v
1 þ
i¼110pK -pH
VAR ¼
;
P
P
4
n
1 þ
10pKai-pH0
n¼1
i¼1
where pHv = pH inside the vacuole (assumed to be pH 5.0)
pH0 = pH externally (assumed to be pH 7.4). This equation
proceeds from a derivation of the Henderson–Hasselbach
equation, based on predicted values of drug pKa accord-
9. Gordeuk, V. R.; Thuma, P. E.; McLaren, C. E.; Biemba,
G.; Zulu, S.; Poltera, A. A.; Askin, J. E.; Brittenham, G.
M. Blood 1995, 85, 3297.

P. Melnyk et al. / Bioorg. Med. Chem. Lett. 16 (2006) 31–35
35
ing to previous works of Hawley et al.23 Values of pKa
were calculated using ACD/pKa DB software from Ad-
vanced Chemistry Development Inc., Toronto, Canada.
23. Hawley, S.; Bray, P. C.; OÕNeill, P. M.; Park, B. K.; Ward,
S. A. Biochem. Pharmacol. 1996, 52, 723.
24. Vippagunta, S. R.; Dorn, A.; Bubendorf, A.; Ridley, R.
G.; Vennerstrom, J. L. Biochem. Pharmacol. 1999, 58,
817.
25. Chaston, T.; Lovejoy, D.; Watts, R. N.; Richardson, D.
R. Clin. Cancer Res. 2003, 9, 402.