Potent antimalarial activity of 2-aminopyridinium salts, amidines, and guanidines

Article abstract of DOI:10.1021/jm0704752

We describe the design, synthesis, and antimalarial activity of 60 bis-tertiary amine, bis-2(1H)-imino-heterocycle, bis-amidine, and bis-guanidine series. Bis-tertiary amines with a linker from 12 to 16 methylene groups were active against the in vitro growth of Plasmodium falciparum within the 10 ^-6-10^-7 M concentration range. IC₅₀ decreased by 2 orders of magnitude for bis-2-aminopyridinium salts, bis-amidines, and bis-guanidines (27 compounds with IC₅₀ < 10 nM). Increasing the alkyl chain length from 6 to 12 methylene groups led to increased activity, while beyond this antimalarial activity decreased. Antimalarial activities appear to be strictly related to the basicity of the cationic head with an optimal pK_a over 12.5. Maximal activity occurs for bis-2-aminopyridinium, two C-duplicated bis-amidines, and three bis-guanidines, with IC₅₀ values lower than 1 nM. In comparison to similar quaternary ammonium salts, amidinium compounds have distinct structural requirements for antimalarial activity and likely additional binding opportunities on account of their hydrogen-bond-forming properties.

Full text of DOI:10.1021/jm0704752

J. Med. Chem. 2007, 50, 6307–6315
6307
Articles
Potent Antimalarial Activity of 2-Aminopyridinium Salts, Amidines, and Guanidines
Michèle Calas,*,†,‡ Mahama Ouattara,†,‡,§ Gilles Piquet,† Zyta Ziora,† Y. Bordat,| Marie L. Ancelin,| Roger Escale,† and
Henri Vial*,|
Institut des Biomolécules Max Mousseron (IBMM) CNRS UMR 5247, UniVersité Montpellier 1, Faculté de Pharmacie, 15, AVenue C. Flahault,
BP 14491, 34093 Montpellier Cedex 5 and UniVersité Montpellier 2, Place E. Bataillon, CP19, 34095 Montpellier Cedex 5, France,
Laboratoire de Chimie Thérapeutique et Synthèse de Médicaments, Faculté de Pharmacie, UniVersité d’Abidjan-Cocody, BP V, 34 Abidjan,
Côte d’IVoire, and Dynamique des Interactions Membranaires Normales et Pathologiques, CNRS UMR 5235, CP 107, UniVersité de
Montpellier II, Place E. Bataillon, 34095 Montpellier Cedex 5, France
ReceiVed April 23, 2007
We describe the design, synthesis, and antimalarial activity of 60 bis-tertiary amine, bis-2(1H)-imino-
heterocycle, bis-amidine, and bis-guanidine series. Bis-tertiary amines with a linker from 12 to 16 methylene
groups were active against the in Vitro growth of Plasmodium falciparum within the 10-6-10-7
M
concentration range. IC50 decreased by 2 orders of magnitude for bis-2-aminopyridinium salts, bis-amidines,
and bis-guanidines (27 compounds with IC50 < 10 nM). Increasing the alkyl chain length from 6 to 12
methylene groups led to increased activity, while beyond this antimalarial activity decreased. Antimalarial
activities appear to be strictly related to the basicity of the cationic head with an optimal pKa over 12.5.
Maximal activity occurs for bis-2-aminopyridinium, two C-duplicated bis-amidines, and three bis-guanidines,
with IC50 values lower than 1 nM. In comparison to similar quaternary ammonium salts, amidinium
compounds have distinct structural requirements for antimalarial activity and likely additional binding
opportunities on account of their hydrogen-bond-forming properties.
Introduction
antagonists consist of mono-9,10 or bis-11,12 quaternary
ammoniums, which are potent inhibitors of the PL-related
choline carrier in eukaryotic cells, such as erythrocytes,13
and the acetylcholine-related choline carrier in the central
nervous system.14,15 These compounds, which mimic the
choline structure, subsequently lead to selective inhibition
of de noVo malarial PC synthesis,9,16,17 while choline
antagonizes the in Vitro antimalarial activity of the com-
pounds (unpublished). The exact steps that mediate selective
inhibition of this metabolic pathway have yet to be clarified.
Recently, we have shown that compounds also interact with
hemin and/or hemozoin, the hemoglobin-degradation product
inside the parasite food vacuole, likely contributing to
antimalarial activity.18
Malaria is a public health problem today for 40% of the
world’s population. It causes from 1.5 to 2.7 million deaths each
year, and the disease is estimated to be responsible for 45 million
“disability-adjusted life years”.1–3 The latest analysis indicates
that the disease is an even greater problem than previously
thought because there were more than a half billion cases of
deadly Plasmodium falciparum malaria in 2002.4 The lack of
vaccine and the absence of a widely accessible vector control
strategy mean that drugs remain the key control measure for
malaria. The rapid extension of multidrug-resistant parasites,
particularly P. falciparum, responsible for the most lethal form
of malaria, calls for new pharmacological approaches, leading
to original molecules with novel mechanisms of action.
P. falciparum intraerythrocytic growth is associated with
a dramatic increase in the total membrane, essentially
composed of phospholipids (PLs) but not cholesterol. The
mature erythrocyte is devoid of any lipid biosynthetic
activity;5 therefore, the PL biosynthetic machinery is crucial
for parasite membrane biogenesis, which is a prerequisite
for its development and growth.6,7 We have identified a
family of antimalarial compounds that target the membrane
biogenesis of parasites in its erythrocytic stage through
the blockage of de noVo biosynthesis of phosphatidylcholine
(PC), the major malarial phospholipid.8 These choline
The structure of these antimalarial compounds has been
optimized using pharmacochemistry principles [structure-activity
relationship (SAR)]. The most active compounds (IC50
∼
nanomolars) are bisquaternary ammonium salts, whose linker
between the two nitrogen atoms is composed of at least 12
methylene groups and the cationic head volume is roughly 500
Å3. The bis-quaternary ammonium salt G25 (IC50 ) 0.65 nM)
and the bis-thiazolium are the lead compounds in this new
antimalarial family.12,19
* To whom correspondence should be addressed. Telephone: 33-
467143817. Fax: 33-467144866 E-mail: michele.calas@univ-montp2.fr
(M.C.); Telephone: 33-467143745. E-mail: vial@univ-montp2.fr (H.V.).
†
Université Montpellier 1 and Université Montpellier 2.
These authors contributed equally to this work.
Université d’Abidjan-Cocody.
Université de Montpellier II.
‡
The antimalarial activity appears to be related to the
capacity of these compounds to mimic the choline structure.
The compounds thus inhibit choline influx and induce a very
§
|
10.1021/jm0704752 CCC: $37.00
2007 American Chemical Society
Published on Web 11/16/2007

6308 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 25
Calas et al.
Table 1. Structures and Antimalarial Activities of Bis-tertiary Amines
and Their Structurally Related Bis-quaternary Ammonium Salts
Table 2. Structures and Antimalarial Activities of Mono and
Bis-2-aminopyridinium Salts
Table 3. Structures and Antimalarial Activities of Bis-2(1H)-imino
Heterocycles
a The activity of these compounds has already been reported.10,12
specific and early alteration of malarial de noVo biosynthesis
of PC, while the lethal effect is closely correlated with the
blockage of PC biosynthesis.16 Finally, the pharmacological
approach has also been fully validated in ViVo against rodent
malaria11 and also against P. falciparum and Plasmodium
cynomolgi malaria in monkeys.17,20 The potency of these
compounds seems to be related to their high accumulation
inside infected erythrocytes, which ensures both potency and
selectivity. Their dual mechanism of action should allow the
developed molecules to be active against polypharmacore-
sistant isolates of P. falciparum and limit the risk of
emergence of cross-resistance.
Despite this potent antimalarial activity, one major draw-
back of quaternary ammonium salts for their therapeutic use
is their very weak oral absorption because of the permanent
positive charge of the quaternary ammonium moiety. One
way to circumvent this poor oral absorption would be to
replace the quaternary ammonium heads by bioisosteric
groups that are highly basic, ionized at physiological pH,
and capable of creating bonds with the target(s) similar to
or stronger than those of bis-quaternary ammonium salts.
Tertiary amines, amidines, or guanidines are thus potential
bioisosteric substitutes for cationic heads and should more
easily cross the intestinal barrier.
In this study, we designed 60 new compounds, classified into
four groups: (a) 7 bis-tertiary amines and three bis-quaternary
ammonium salts (Table 1), (b) 22 bis-aromatic 2(1H)-imine-
heterocycles and one pyridin-2(1H)-imine (Tables 2 and 3), (c)
20 bis-amidines (Tables 4 and 5), and (d) 7 bis-guanidines
(Table 6), to which one intermediate imidate I1 was added
(Table 4), and we evaluated their in Vitro antimalarial activity
against the human P. falciparum parasite.
refluxing solvent, as shown in Scheme 1. C7 was obtained by
the reaction of 1,12-dibromododecane with diethylethanolamine
and KOH in dimethylsulfoxide (DMSO).
Alkylation of C6 and C7 with iodomethane and bromoethane
generated quaternary ammonium salts G6 and G7, respectively.
The reaction of N-methyl-2-hydroxymethylpyrrolidine and 1,16-
dibromohexadecane led to G5.
Results
Chemistry. 1. Bis-tertiary Amines and Bis-quaternary
Ammonium Salts (C and G Series, Table 1). Except for
commercial C1, tertiary amines C2, C3, C4, C5, and C6 were
prepared by the reaction of R,ω-dihalogenoalkane (bromide or
iodide) with an excess of an appropriate secondary amine in
2. Bis-2(1H)-imino-heterocycle Salts (H Series, Tables
2 and 3). Mono- and bis-2-aminopyridinium salts and bis-2-

Potent Antimalarial ActiVity
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 25 6309
Table 6. Structures and Antimalarial Activities of Bis-guanidines
Table 4. Structures and Antimalarial Activities of C-Duplicated
Bis-amidines and Imidate (I1)
solvent as described in Scheme 2. H6 was obtained by the
reaction of 1,4-bis(3-bromopropyl)benzene21 with an excess of
2-amino-4-methylpyridine.
These compounds can occur in protonated form (Scheme 2a),
with an aromatic character (where the positive charge is shared
throughout the aromatic heterocycle), or neutral form (Scheme
2b).
3. Bis-amidines (A Series, Tables 4 and 5). Bis-amidine
compounds (A series) with a linker attached to the functional
carbon atom (C-duplicated compounds, Table 4) were prepared
in three steps from alkane dibromide (Scheme 3): (a) substitution
of bromine atoms by cyanide groups, (b) addition of EtOH on
cyanide, leading to imidates (I1 or I2), and then (c) reaction
with an appropriate amine (leading to A1-A5) or diamine
(leading to A6-A11).
a These compounds are in the form of free bases.
1,12-Bis(imidazol-2-yl)dodecane A1222 was prepared by
cyclocondensation of tetradecanedial with a bisulfite combina-
tion of glyoxal (hydrate) and ammonium bicarbonate in water
and isopropanol. 1,12-Bis(benzimidazol-2-yl)dodecane A1323
was synthesized by acid condensation of o-phenylenediamine
with tetradecanedioic acid.
Table 5. Structures and Antimalarial Activities of N-Duplicated
Bis-amidines
Bis-amidines, with a linker on the functional nitrogen atom
(N-duplicated compounds, Table 5) and with R1 ) H
(A14-A17), were prepared by the reaction of 1,12-diamin-
ododecane with an appropriate imidate (Scheme 4).
A19 and A20 (Table 5) were prepared by N-alkylation of
imidazole and 2-methylimidazole, respectively, with 1,12-
dibromodecane in the presence of KOH (Scheme 4).
A18 (Table 5) was obtained by alkylation of pyrrolidin-2-
one and 1,12-dibromododecane in the presence of sodium and
the reaction of 1,12-di(pyrrolidinyl-2-one)dodecane obtained
with isocyanatidosulfuryl chloride (Scheme 5).24
4. Bis-guanidines (A Series, Table 6). Bis-guanidine com-
pounds A21, A23, A25, A26, and A27 (Table 6) were prepared
by the reaction of 1,12-diaminododecane with an appropriate
alkyl imidothiocarbamate CH3-S-C(dNR1)NHR2 or C2H5-
S-C(dNR1)NHR2 (Scheme 6).
A22 and A24 (Table 6) were prepared by the route shown in
Scheme 7. The addition of 1,12-diaminododecane on CS2
generated 1,12-diisothiocyanatododecane. The latter, treated with
dimethylamine, was converted to 1,12-bis(thioureido)dodecane,
which was alkylated by iodomethane to give the imidothiocar-
bamate I3. A22 and A24 were obtained by the reaction of I3
with ammonia in MeOH or NC-NH2, respectively.
Antimalarial Activity. 1. Bis-tertiary Amines, C Series
(Table 1). A total of 7 bis-tertiary amines, whose functions were
separated by an alkyl chain of 6-16 methylene groups and
whose N substituents were either low alkyl groups (methyl or
ethyl) or formed a pyrrolidine cycle with N, possibly substituted,
a These compounds are in the form of free bases.
amino-aryl salts represented our H series. A total of 10 bis-2-
aminopyridinium salts (Table 2) consist of two pyridinium
moieties, substituted by a methyl, phenyl, or chloride atom,
bound to one another by an alkyl chain of 6-16 methylene
groups, for H6 inclusion of a phenyl cycle. A total of 12
analogues, in which pyridinium moieties were replaced by other
differently substituted heterocycles [(benzo)thiazole, benzimi-
dazole, pyrimidine, isoquinoline, thiadiazole, and pyridazine]
(Table 3) were also synthesized. Lastly, one monopyridinium
salt, H1 (Table 2), was also prepared.
All symmetrical biscations were prepared by the reaction of
R,ω-dihalogenoalkane (bromide or iodide) with excess nucleo-
philic compound (2-aminopyridine eventually 3, 4, or 5
substituted or 2-amino nitrogenous heterocycle) in refluxing

6310 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 25
Calas et al.
Scheme 1a
a Reagents and conditions: (a) (C2H5)2NH, NaOH, HCl; (b) (C2H5)2N-CH2CH2OH, HNa/THF, HCl; (c) C2H5Br in EtOH; (d) (C2H5)2NH; (e) pyrrolidine,
NaOH, HCl; (f) 2-hydroxymethylpyrrolidine; (g) 2(p-trifluoromethylphenoxymethyl)pyrrolidine; (h) CH3I in EtOH.
Scheme 2a
group on each nitrogen atom), regardless of the nitrogen
substituents and inter-nitrogen chain (Table 1). This was
probably due to the less lipophilic environment of the cationic
heads of amines (three substituents) in comparison to those of
ammonium compounds (four substituents). Indeed, this lipo-
philic environment around nitrogen atoms is required for the
antimalarial activity.12
2. Bis-2-amino-N-heterocycle Salts, H Series (Tables 2
and 3). Because bis-tertiary amines had much lower antimalarial
activity than their corresponding quaternary ammonium, we
synthesized bis-2-aminopyridinium salts (Table 2) and bis-2-
amino-N-heterocycle salts (Table 3), which represented our H
series (Scheme 2). In these series, the antimalarial activity of
bis-cationic compounds was far more potent (by 3 orders of
a (a) Protonated form: salt whose cationic moiety possesses an aromatic
character. (b) Neutral form: bis-2(1H)-imino heterocycle.
magnitude) than monocationic compounds (compare H1 and
H4). Consequently, we have chosen to focus on duplicated
molecules in the following part of this work.
Bis-2-aminopyridinium salts, differently substituted (Table
2), were very potent against P. falciparum in Vitro growth, with
an IC50 ranging from 0.5 to 13 nM. A variation in the number
of methylene groups in the spacer indicated maximal activity
for 12 carbon atoms (compare H3, H4, and H5), with an IC50
of 0.5 nM for the H4 compound. Inclusion of an aromatic cycle
in the linker, corresponding to a linker length of 10 methylene
groups (H6, IC50 ) 2.9 nM), did not change the antimalarial
activity. Thus, we kept the linker to the 12 methylene groups
that appeared to be optimal for antimalarial activity.
We then substituted the pyridine cycle by other aromatic
heterocycles [(benzo)thiazole (H12 and H13), isoquinoline
(H16), benzimidazole (H14 and H15), pyrimidine (H17 and
H18), thiadiazole (H19, H20, H21, and H22), and benzo[h-
]cinnoline (H23), Table 3]. The highest activity, was obtained
for H18 (IC50 ) 12.3 nM), which contained a 5-phenyl
pyrimidinium as the cationic head. Other aromatic heterocycles
were synthesized. One compound possessed two oxygen atoms
in the interchain of 16 methylene groups. Their antimalarial
activities were evaluated in Vitro by adding the compounds to
a P. falciparum suspension for a complete 48 h parasite cycle
before assessing the parasite viability. Among these 7 bis-
amines, 6 were active in the micromolar range, with IC50 values
ranging from 5.4 to 0.11 µM. The activity appeared to be related
to the presence of a long alkyl chain (number of methylene
groups g 12) between the two nitrogen atoms. Progressive
elongation of this linker from 6 to 16 carbon atoms (C1/C2/
C3) increased the antimalarial potency. The presence of oxygen
atoms in the linker did not affect the activity against the malarial
parasite (C5/C7).
However, except for the poorly active C1 compound, the
antimalarial activity of tertiary bis-amines was 1-2 orders of
magnitude lower than that of their structurally related bis-
quaternary ammonium salts (with an additional methyl or ethyl

Potent Antimalarial ActiVity
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 25 6311
Scheme 3a
a Reagents and conditions: (a) NaCN; (b) EtOH, Et2O, HCl gas; (c) amines (ammonia, dimethylamine, pyrrolidine, or piperidine); (d) diamines
(ethylenediamine, N-methylethylenediamine, 1,3-diaminopropane, 1,2-diaminopropane, 1,2-diamino-2-methylpropane, or 1,2-diaminocyclohexane).
Scheme 4
bulky cationic head, had lower activity, while A24, whose
nitrogen is substituted by an electron-withdrawing CN group,
was poorly active.
In this study, we also measured antimalarial activity of the
synthesis intermediate imidate I1 (Table 4). This compound was
104-fold less active than amidines and guanidines.
4. Relation between Compound Basicity and Anti-
malarial Activity. We determined the pKa of most of the
compounds using potentiometric titration. Bis-2-amino-pyri-
dinium salts had a pKa in the 12.6–13.5 range (Table 2); bis-
amidines had a pKa ∼ 11.0–13.6 (Tables 4 and 5); and bis-
guanidines had a pKa ∼ 13–14 (Table 6). The pKa values of
other bis-2-amino-N-heterocycle salts in Table 3 were lower
(pKa ∼ 8–11). All of these biscations were fully protonated
at pH 7.4. Figure 1 shows the antimalarial activity versus
the pKa for compounds belonging to A (bis-amidines and
bis-guanidines) and H (bis-2-amino-N-heterocycle salts) series
that had the same inter-nitrogen spacer, (CH2)12, and that
differed only by the polar head. It was striking that the in
Vitro activity against the human P. falciparum parasite was
strictly related to the pKa values of the molecules. Biscations,
with a linker of 12 carbon atoms, possessing a pKa of more
than 12, had an IC50 of below 13 nM, regardless of the
nitrogen substitution.
resulted in a dramatic decrease of in Vitro antimalarial activity
(IC50 g 50 nM). The antimalarial activity thus did not seem to
be related to the cationic head volume but rather to the pKa
(see below).
3. Bis-amidines (Tables 4 and 5) and Bis-guanidines
(Table 6), A Series. We also synthesized A series involving
bis-amidines and bis-guanidines, possessing various substituents
on nitrogen atoms and 12 or 16 methylene groups as the linker.
Two bis-amidine types were designed, i.e., those with the
linker on the carbon atom (named C-duplicated compounds,
A1-A11, Table 4), and on the nitrogen atom (named N-
duplicated compounds, A14-A18, Table 5). In each of these
two series, two compounds had an amidine function included
in the heteroaromatic cycle (imidazoles A12, A19, and A20 and
benzimidazole A13, Tables 4 and 5). The last four compounds
were not very active, with an IC50 in the micromolar range. On
the other hand, bis-amidine and bis-guanidine compounds had
very potent in Vitro antimalarial activities against P. falciparum,
with an IC50 in the very low nanomolar range when the linker
contained 12 carbon atoms. Increasing the length of the linker
chain from 12 to 16 methylene groups significantly decreased
the antimalarial activity (compare A1, 0.3 nM with A2, 100
nM).
When the bis-amidines had the optimal linker (12 C), the
antimalarial activities all appeared to be more potent because
the substitution degree on the nitrogen atoms was low for both
the C- and N-duplicated amidine series [compare A1 (no
substitution, ∼0.3 nM) with A3-A6, A8-A11, and A14-A18
(1 or 2 substitutions, 0.8–6.3 nM) and with A7 (3 substitutions,
43 nM), Tables 4 and 5].
When equally substituted, bis-guanidines (Table 6) appeared
to be slightly more active, or as active as bis-amidines (compare
A21/A1, A22/A3, A25/A6, and A26/A8). A23, which has a very
Discussion
From a molecular standpoint, our pharmacological approach
is based on the capacity for biscations to mimic the choline
structure. The SARs assessed for more than 280 quaternary
ammonium salts allowed us to optimize their structures for
potent antimalarial activity, with IC50
against P. falciparum in
the low nanomolar range. The hydroxyethyl group of choline
is not a prerequisite for antiplasmodial activity. On the other
hand, the antimalarial activity appeared to be dependent upon
nitrogen substitution with one long lipophilic chain, while
duplication of the quaternary ammonium increased the inhibitor
potency by at least 2 orders of magnitude. Bis-quaternary
ammonium salts with two cationic heads separated by a
lipophilic spacer (at least 12 methylene groups) are among the
most potent antimalarials with this molecule type (G25 as a
leader).12
To remedy the low oral absorption of quaternary ammonium
salts, inherent to the permanent charge on nitrogen atoms, the
Scheme 5a
a Reagents and conditions: (a) Na; (b) OdCdN-SO2Cl and then NaOH.

6312 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 25
Calas et al.
Scheme 6
quaternary polar heads were replaced by bioisosteric groups,
which could, at physiological pH, induce the same drug-target
interactions as quaternary ammonium salts. These new analogues
are tertiary amines, 2-aminopyridinium salts, 2-amino-N-
substituted aryl salts, amidines, and guanidines. Their functions
are ionized at physiological pH, potentially mimicking the
cationic head of the choline. Because of the equilibrium between
protonated and unprotonated forms, these compounds should
diffuse more easily through the cellular membrane and tissue
because of the neutral form.
Bis-tertiary amines, whose inter-nitrogen linker contains at
least 12 carbon atoms, had significant antimalarial activity but
1-2 orders of magnitude lower than their corresponding bis-
quaternary ammonium salts (Table 1). The amine cationic head
was probably not lipophilic enough to completely match the
target.
Then, we replaced the quaternary ammonium heads by
strongly basic bioisosteric groups, ionized at physiological pH,
which could form the same bonds with the target as bis-
quaternary ammonium salts. The structure of amidine systems
is midway between that of tertiary amines and quaternary
compounds, at least regarding the ionization and electronic
properties, but differ from the latter in shape (planar for amidines
and guanidines and tetrahedral for ammoniums). Nevertheless,
we investigated amidinium systems for their additional distinc-
tive features. The driving force for the primary docking of
amidinium cations with an anionic counterpart of the target is
assumed to be an electrostatic interaction, but hydrogen-bond
formation could also strengthen the affinity.25–27
Among these compounds, 2-amino-pyridinium salts (H
series), amidines, and guanidines (A series) showed very potent
antimalarial activities, with an IC50 in the very low nanomolar
range. A total of 27 molecules had an IC50 of less than 10 nM
against the human parasite P. falciparum. Maximal activity was
obtained for two bis 2-aminopyridiniums (H4 and H8, with IC50
of 0.5 nM), two C-duplicated bis-amidines (A1 and A4, with
IC50 of 0.4 and 0.8 nM, respectively), and three bis-guanidines
(A21, A26, and A27, with IC50 of 0.35, 0.5, and 0.6 nM,
respectively). The compounds were as potent as bis-quaternary
ammonium salts.12
Dimerization of the cationic head produced effectors with
enhanced (by 2 orders of magnitude) antimalarial activity
relative to the monomeric congener. This feature, which was
also found in a previous study, suggests a bivalent interaction
between the compound and the target.12
The length of the linker between the two cationic heads
proved to be a critical parameter for antimalarial activity. The
optimal length, in the A and H series, was 12 methylene groups
(estimated to be around 16 Å in the extended state of the
molecule). This differed from 16, as observed with bis-amines
(Table 1) and bis-quaternary ammonium salts.12
Most compounds with a (CH2)12 linker, belonging to the bis-
2-aminopyridinium series, with a small substituent (methyl
group or chloride atom) on the pyridinium cycle (Table 2), and
to the A series (amidines and guanidines, Tables 4-6) are potent
in Vitro antimalarials (IC50 < 10 nM).
With the pyridinium cycle, amidine or guanidine function
replaced by other aromatic heterocycles, including (benzo)thia-
zole, isoquinoline, (benz)imidazole, pyrimidine, thiadiazole, and
benzo[h]cinnoline (all compounds in Table 3 and A12 and A13
in Table 4), led to much less active compounds. As illustrated
in Figure 1, the basicity (pKa values) of the compound polar
heads appeared to be strictly correlated with the antimalarial
activity and is likely a crucial parameter for antimalarial activity.
The antimalarial activity of the compounds occurred at lower
compound concentrations as their basicity increased. In Vitro
antimalarial activity in the low nanomolar range appears to be
restricted to compounds with a pKa between 12.5 and 14.5.
The positive charge of the H and A compound series was
strongly delocalized, and the ability of these groups to form
bonds (hydrogen or ionic bonds) was higher than that of tertiary
amines and quaternary ammoniums. The pKa values of these
compounds thus do not only indicate the percentage of the
protonated compound at physiologic pH but likely also its ability
to form strong (hydrogen and electrostatic) bonds with the
target.27–29
In contrast to compounds containing a quaternary ammonium,
the polar head of amidines, guanidines, and 2-aminopyridiniums
are planar and thus do not occupy the same bulk at the active
site. Hydrogen-bond formation may influence the positioning
of the interacting molecule, thus explaining the difference in
optimum linker length in the two biscation types. However, both
types of compounds exerted early effects on PC biosynthesis
of Plasmodium (unpublished results) and seemed to share the
same mechanism of action.
In general, symmetrical biscations, such as amidines, guanidines,
and heterocyclic amidines, are known to have various pharmaco-
logical effects (antimicrobial30 and antifungals31,32). They act at
different targets, including receptors (GPCR muscarinic recep-
tors33 and channel receptors34) and enzymes (protein kinase C).
Bis-aromatic benzamidines (pentamidine I and its analogues as
II and III, Figure 2 ), with different linkers connecting the
aromatic groups, were extensively studied for their physiological
effects on the central nervous system35 and their inhibition of
various microbes and parasites (Pneumocystis carinii,36–38
Leishmania,39 Trypanosoma brucei,40 and Babesia canis41),
including the malarial parasite P. falciparum.42 A and H series
compounds are structurally different from pentamidine ana-
logues, for which the cationic heads include an additional phenyl
cycle conjugated with the amidine function, and the linker is
often shorter and rigid.43,44 The structure-antiplasmodial activ-
ity relationships of the benzamidines and our compounds
differed substantially.45 This suggests that the therapeutic targets
also differed. Plausible mechanisms of action of these bioactive
benzamidines include initial binding to AT-rich sites of DNA
followed by inhibition of one or more of several DNA-dependent
enzymes or direct inhibition of the transcription process. This
concerns furamidine (II), a rigid, curved, and planar molecule;
therefore, it has the appropriate steric and electrostatic properties
for high-affinity binding to the minor groove of DNA. For
pentamidine congeners II and III, however, Mayence et al.44
recently reported that the DNA binding of benzamidine ana-
logues was not correlated with their antiplasmodial activities.
Finally, benzamidines, particularly pentamidine, which is con-
centrated 500-fold by erythrocytes infected with P. falciparum,42
were shown to form complexes with ferriprotoporphyrin IX and,
therefore, to be involved in the inhibition of hemozoin formation
in a cell-free system.46 Such an accumulation remains to be
clarified for our compounds series. Besides, there was no
indication of an interaction of pentamidine-like biscations with
plasmodial phospholipid biosynthesis.
In conclusion, we have shown that a series of novel
bis-amidines, bis-guanidines, and 2-aminopyridinium salts,

Potent Antimalarial ActiVity
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 25 6313
Scheme 7a
a Reagents and conditions: (a) CS2, DCC; (b) (CH3)2NH in MeOH; (c) CH3I; (d) NH3 in MeOH; (e) N′C-NH2 in MeOH.
All key target compounds possessed purity superior to 98%
(results of combustion analyses for C, H, and N were within (0.4%
of the theoretical value).
General Procedure for the Synthesis of Bis-tertiary
Amines (Table 1). A mixture of R,ω-dihalogenoalkane (0.15 mol)
and excess (4-10 equiv) of secondary amine in 75 mL of EtOH
was refluxed for 12 h. The solvent and amine excess were removed.
When the amine was isolated with the dihydroiodate, the residue
was directly recrystallized in acetone. When the amine was isolated
as its dihydrochloride, the residue was diluted with 30 mL of water
and 10% NaOH was added until basic pH. Diamine was extracted
with 100 mL of diethyl ether, in which HCl gas was added. The
resulting solid was recrystallized in acetone.
General Procedure for the Synthesis of Bis-2-amino-
pyridinium Salts (Table 2) and Bis-2(1H)-imino-nitrogenous
Heterocycle Salts (Table 3). Procedure A. A mixture of R,ω-
dihalogenoalkane (0.1 mol) and an excess (about 3 equiv) of a
nucleophilic compound (2-aminopyridine, possibly 3, 4, or 5
substituted, or 2(1H)-imino-nitrogenous heterocycle) in butanone
(100 mL) was refluxed in an argon atmosphere for 48 h. After the
solution was cooled, the sought-after compound was precipitated.
It was recrystallized in an iPrOH/MeOH mixture (9:1).
Procedure B. A mixture of R,ω-dihalogenoalkane (0.1 mol) and
an excess (about 3 equiv) of a nucleophilic compound [2(1H)-imino-
nitrogenous heterocycle] in sulfolane (100 mL) was refluxed in an
argon atmosphere for 1 week. A total of 60 mL of CH3CN was
added. After the solution was cooled, the sought-after compound
was precipitated. It was recrystallized in an iPrOH/MeOH mixture
(9:1).
Figure 1. Antimalarial activity (IC50) of bis-2-aminopyridinium salts
([), bis-2(1H)-imine heterocycles (9), C-duplicated bis-amidines (2),
N-duplicated bis-amidines (4), bis-guanidines (O), and one imidate
(b) (Tables 2-6) possessing the same inter-nitrogen spacer, (CH2)12,
as a function of basicity (pKa) (35 compounds).
When the compound was isolated as a dioxalate (H13-H15),
the salt was diluted in K2CO3 (10% aqueous). This solution was
gently heated (50 °C), and the free diamine was extracted with
AcOEt. This compound was diluted in a mixture of iPrOH/MeOH
(1:1), and 2 equiv of oxalic acid in solution in iPrOH was added.
The precipitate is filtered and recrystallized in an iPrOH/MeOH/
(iPr)2O mixture.
Figure 2. Pentamidine and analogues.
Determination of Ionization Constants (pKa). The compound
pKa values were obtained, at room temperature, using potentiometric
titration.47 A Jenway 3040 ion analyzer pH meter (calibrated
according to the instructions of the manufacturer) equipped with
an Ingold pH electrode was used. The method consisted of
measuring the pH values within a potentiometric titration.
bioisosteres of bis-quaternary ammonium salts, with a lipophilic
linker [(CH2)12] and whose pKa values were over 12, are potent
antimalarial compounds, which exert their activity in a very low
nanomolar range. Among them, 27 compounds had an IC50
<
An appropriate amount of basic compound (30–50 mg) was
dissolved in water or a water-EtOH medium (e20% EtOH) to
overcome the solubility problems associated with the compounds
and then titrated to pH 2.0 with 0.1 N HCl. During titration, the
titrant was added in increments of 0.1 mL after each stable reading
and the corresponding pH values were recorded. The resulting
values were plotted (pH versus added HCl), and the pKa values
were relatively accurately estimated from the graphs.
In Vitro Antimalarial Activity. In Vitro antimalarial activity
was measured using the [3H]-hypoxanthine incorporation assay with
the Nigerian strain of P. falciparum, as previously described.19
Results were expressed as the concentration resulting in 50%
inhibition (IC50).
10 nM and 15 molecules (H4, H7-H9, A1, A4, A6, A9, A10,
A14, A21, A22, A25-A27) had an IC50 of less than or equal
to 2 nM. In comparison to similar quaternary ammonium
salts, these compounds have binding opportunities on account
of their hydrogen-bond-forming capacity with the target(s).
These compounds, which can exist in the neutral form and in
equilibrium with the cationic form, could be more bio-
available.
Experimental Section
Chemistry. N,N,N′,N′-Tetramethylhexane-1,6-diamine, C1, was
commercially available (Sigma-Aldrich).

6314 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 25
Calas et al.
bis-thiazolium salts have potent antimalarial activity. J. Med. Chem.
2005, 48, 3639–3643.
Acknowledgment. This study was supported by the Euro-
pean Community (QLK2-CT-2000-01166 and PCRDT FP6,
Integrated Projects Antimal “Development of New Drugs for
the treatment of Malaria”), Ministère de l’Education Nationale
et de la Recherche Scientifique (PAL+), the UNDP/World
Bank/WHO special program for Research and Training in
Tropical Diseases, and Sanofi-Aventis Laboratories (Montpellier,
France).
(20) Wengelnik, K.; Vidal, V.; Ancelin, M.-L.; Cathiard, A. M.; Morgat,
J. L.; Kocken, C. H.; Calas, M.; Herrera, S.; Thomas, A. W.; Vial,
H. J. A class of potent antimalarials and their specific accumulation
in infected erythrocytes. Science 2002, 295, 1311–1314.
(21) Koizumi, T.; Tsutsui, K.; Tanaka, K. Selective formation of inter-
and intramolecular A-D-A-p-p stacking: Solid-state structures of
bis(pyridiniopropyl)benzenes. Eur. J. Org. Chem. 2003, 23, 4528–
4532.
(22) Christidis, Y.; Depernet, D. Imidazoles and 2-alkylimidazoles and
method for their manufacture. U.S. Patent 4,074,054, 1976.
(23) Okamoto, T.; Murai, T.; Kikukawa, Y.; Hirao, H.; Yoshioka, T.
Surface-treating agent containing alkylenedibenzimidazole compound
for copper (alloy) of printed circuit. Jpn. Kokai Tokkio Koho 1994, 5,
JP06173022.
(24) Rao, K. R.; Nageswar, Y. V. D.; Srinivasan, T. N.; Sattur, P. B. A
facile synthesis of 2-imino-1-methylazacarbocycles. Synth. Commun.
1988, 18, 877–880.
Supporting Information Available: General methods, experi-
mental procedures and details for the synthesis and analytical data
of all compounds; assessment of biological activity, in Vitro and
in ViVo antimalarial activity and acute toxicity in mouse. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(25) Oszczapowicz, J. Basicity, H-bonding, tautomerism and complex
formation of imidic acid derivatives. In The Chemistry of Amidines
and Imidates; Patai, S, Rappoport, Z., Eds.; John Wiley and Sons:
New York, 1991; Vol. 2, pp 623–688.
References
(1) World Health Organization (WHO). World Health Report: Changing
History; WHO: Geneva, Switzerland, 2004.
(26) Häfelinger, G. General and theorical aspects of amidines and related
compounds. In The Chemistry of Amidines and Imidates; Patai, S.,
Rappoport, Z., Eds.; John Wiley and Sons: New York, 1991; Vol. 2,
pp 1–100.
(27) Kneeland, D. M.; Ariga, K.; Lynch, V. M.; Huang, C. Y.; Anslyn,
E. V. Bis(alkylguanidinium) receptors for phosphodiesters: Effect of
counterions, solvent mixtures, and cavity flexibility on complexation.
J. Am. Chem. Soc. 1993, 115, 10042–10055.
(2) World Health Organization (WHO)/CDS/MAL/2003. Africa Malaria
Report; WHO: Geneva, Switzerland, 2003; p 1093.
(3) http://www.who.int/whr/2004/en.
(4) Snow, R. W.; Guerra, C. A.; Noor, A. M.; Myint, H. Y.; Hay, S. I.
The global distribution of clinical episodes of Plasmodium falciparum
malaria. Nature 2005, 434 (7030), 214.
(5) Van Deenen, L. L. M.; De Gier, J. Lipids of the red cell membrane.
In The Red Blood Cell; Surgenor, G., Ed.; Academic Press: New York,
1975; pp 147–211.
(28) Schiessl, P.; Schmidtchen, F. P. Abiotic molecular recognition of
dicarboxylic anions in methanol. Tetrahedron Lett. 1993, 34, 2449–
2452.
(6) Vial, H. J.; Mamoun, C. Plasmodium lipids: Metabolism and function.
In Molecular Approach to Malaria; Sherman, I., Ed.; ASM Press:
Washington, D.C., 2005; pp 327–352.
(29) Schiessl, P.; Schmidtchen, F. P. Binding of phosphates to abiotic hosts
(7) Vial, H. J.; Eldin, P.; Tielens, A. G.; van Hellemond, J. J. Phospholipids
in parasitic protozoa. Mol. Biochem. Parasitol. 2003, 126, 143–154.
(8) Vial, H. J.; Calas, M. Inhibitors of phospholipid metabolism. In
Antimalarial Chemotherapy, Mechanisms of Action, Modes of Resis-
tance, and New Directions in Drug DeVelopment; Rosenthal, P., Ed.;
Humana Press: Totowa, NJ, 2001; pp 347–365.
(9) Ancelin, M.-L.; Calas, M.; Bompart, J.; Cordina, G.; Martin, D.; Ben
Bari, M.; Jei, T.; Druilhe, P.; Vial, H. J. Antimalarial activity of 77
phospholipid polar head analogs: Close correlation between inhibition
of phospholipid metabolism and in vitro Plasmodium falciparum
growth. Blood 1998, 91, 1426–1437.
in water. J. Org. Chem. 1994, 59 (3), 509–511.
(30) Bailey, D. M.; DeGrazia, C. G.; Hoff, S. J.; Schulenberg, P. L.;
O’Connor, J. R.; Paris, D. A.; Slee, A. M. Bispyridinamines: A new
class of topical antimicrobial agents as inhibitors of dental plaque.
J. Med. Chem. 1984, 27 (11), 1457–1464.
(31) Del Poeta, M.; Schell, W. A.; Dykstra, C. C.; Jones, S.; Tidwell, R. R.;
Czarny, A.; Bajic, M.; Kumar, A.; Boykin, D.; Perfect, J. R.
Structure-in vitro activity relationships of pentamidine analogues and
dication-substituted bis-benzimidazoles as new antifungal agents.
Antimicrob. Agents Chemother. 1998, 42, 2495–2502.
(32) Del Poeta, M.; Schell, W. A.; Dykstra, C. C.; Jones, S. K.; Tidwell,
R. R.; Kumar, A.; Boykin, D. W.; Perfect, J. R. In vitro antifungal
activities of a series of dication-substituted carbazoles, furans, and
benzimidazoles. Antimicrob. Agents Chemother. 1998, 42, 2503–2510.
(33) Ilien, B.; Franchet, C.; Bernard, P.; Morisset, S.; Weill, C. O.;
Bourguignon, J. J.; Hibert, M.; Galzi, J. L. Fluorescence resonance
energy transfer to probe human M1 muscarinic receptor structure and
drug binding properties. J. Neurochem. 2003, 85, 768–778.
(34) Galanakis, D.; Ganellin, C. R.; Malik, S.; Dunn, P. M. Synthesis and
pharmacological testing of dequalinium analogues as blockers of the
apamin-sensitive Ca2+-activated K+ channel: Variation of the length
of the alkylene chain. J. Med. Chem. 1996, 39, 3592–3605.
(35) Tao, B.; Huang, T. L.; Sharma, T. A.; Reynolds, I. J.; Donkor, I. O.
Novel bisbenzamidines and bisbenzimidazolines as noncompetitive
NMDA receptor antagonists. Bioorg. Med. Chem. Lett. 1999, 9, 1299–
1304.
(36) Montgomery, A. B.; Debs, R. J.; Luce, J. M.; Corkery, K. J.; Turner,
J.; Brunette, E. N.; Lin, E. T.; Hopewell, P. C. Aerosolised pentamidine
as sole therapy for Pneumocystis carinii pneumonia in patients with
acquired immunodeficiency syndrome. Lancet 1987, 2, 480–483.
(37) Sepkowitz, K. A.; Armstrong, D. Treatment of opportunistic infections
in AIDS. Lancet 1995, 346, 588–589.
(38) Rahmathullah, S. M.; Hall, J. E.; Bender, B. C.; McCurdy, D. R.;
Tidwell, R. R.; Boykin, D. W. Prodrugs for amidines: Synthesis and
anti-Pneumocystis carinii activity of carbamates of 2,5-bis(4-amidi-
nophenyl)furan. J. Med. Chem. 1999, 42, 3994–4000.
(10) Calas, M.; Cordina, G.; Bompart, J.; Bari, M. B.; Jei, T.; Ancelin,
M.-L.; Vial, H. J. Antimalarial activity of molecules interfering with
Plasmodium falciparum phospholipid metabolism. Structure-activity
relationship analysis. J. Med. Chem. 1997, 40, 3557–3566.
(11) Ancelin, M.-L.; Calas, M.; Bonhoure, A.; Herbute, S.; Vial, H. J. In
vivo antimalarial activities of mono- and bis-quaternary ammonium
salts interfering with Plasmodium phospholipid metabolism. Antimi-
crob. Agents Chemother. 2003, 47, 2598–2605.
(12) Calas, M.; Ancelin, M.-L.; Cordina, G.; Portefaix, P.; Piquet, G.; Vidal-
Sailhan, V.; Vial, H. J. Antimalarial activity of compounds interfering
with Plasmodium falciparum phospholipid metabolism: Comparison
between mono- and bis-quaternary ammonium salts. J. Med. Chem.
2000, 43, 505–516.
(13) Deves, R.; Krupka, R. M. The binding and translocation steps in
transport as related to substrate structure. A study of the choline carrier
of erythrocytes. Biochim. Biophys. Acta 1979, 557, 469–485.
(14) Fisher, A.; Hanin, I. Choline analogs as potential tools in developing
selective animal models of central cholinergic hypofunction. Life Sci.
1980, 27 (18), 1615–1634.
(15) Holden, J. T.; Rossier, J.; Beaujouan, J. C.; Guyenet, P.; Glowinski,
J. Inhibition of high-affinity choline transport in rat striatal synapto-
somes by alkyl bisquaternary ammonium compounds. Mol. Pharmacol.
1975, 11 (1), 19–27.
(16) Ancelin, M.-L.; Calas, M.; Vidal-Sailhan, V.; Herbute, S.; Ringwald,
P.; Vial, H. J. Potent inhibitors of Plasmodium phospholipid metabo-
lism with a broad spectrum of in vitro antimalarial activities.
Antimicrob. Agents Chemother. 2003, 47, 2590–2597.
(17) Vial, H. J.; Wein, S.; Farenc, C.; Kocken, C.; Nicolas, O.; Ancelin,
M.-L.; Bressolle, F.; Thomas, A.; Calas, M. Prodrugs of bisthiazolium
salts are orally potent antimalarials. Proc. Natl. Acad. Sci. U.S.A. 2004,
101, 15458–15463.
(39) Bell, C. A.; Hall, J. E.; Kyle, D. E.; Grogl, M.; Ohemeng, K. A.;
Allen, M. A.; Tidwell, R. R. Structure-activity relationships of analogs
of pentamidine against Plasmodium falciparum and Leishmania
mexicana amazonensis. Antimicrob. Agents Chemother. 1990, 34,
1381–1386.
(40) Donkor, I. O.; Assefa, H.; Rattendi, D.; Lane, S.; Vargas, M.; Goldberg,
B.; Bacchi, C. Trypanocidal activity of dicationic compounds related
to pentamidine. Eur. J. Med. Chem. 2001, 36, 531–538.
(41) Lourie, E. M.; Yorke, W. Studies in chemotherapy. XXI. The
trypanocidal action of certain aromatic diamidines. Trop. Dis. Bull.
1940, 37, 404.
(18) Biagini, G. A.; Richier, E.; Bray, P. G.; Calas, M.; Vial, H.; Ward,
S. A. Heme binding contributes to antimalarial activity of bis-
quaternary ammoniums. Antimicrob. Agents Chemother. 2003, 47,
2584–2589.
(19) Hamze, A.; Rubi, E.; Arnal, P.; Boisbrun, M.; Carcel, C.; Salom-
Roig, X.; Maynadier, M.; Wein, S.; Vial, H.; Calas, M. Mono- and

Potent Antimalarial ActiVity
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 25 6315
(42) Stead, A. M.; Bray, P. G.; Edwards, I. G.; DeKoning, H. P.; Elford,
B. C.; Stocks, P. A.; Ward, S. A. Diamidine compounds: Selective
uptake and targeting in Plasmodium falciparum. Mol. Pharmacol.
2001, 59 (5), 1298–1306.
(43) Ismail, M. A.; Brun, R.; Easterbrook, J. D.; Tanious, F. A.; Wilson,
W. D.; Boykin, D. W. Synthesis and antiprotozoal activity of aza-
analogues of furamidine. J. Med. Chem. 2003, 46 (22), 4761–4769.
(44) Mayence, A.; Vanden Eynde, J. J.; LeCour, J. L.; Walker, L. A.;
Tekwani, B. L.; Huang, T. L. Piperazine-linked bisbenzamidines: A
novel class of antileishmanial agents. Eur. J. Med. Chem. 2004, 39,
547–553.
(45) Salom-Roig, X. J.; Hamzé, A.; Calas, M.; Vial, H. J. Dual molecules
as new antimalarials. Comb. Chem. High Throughput Screening 2005,
8, 49–62.
(46) Mayence, A.; Vanden Eynde, J. J.; Huang, T. L. Evidences for the
formation of bisbenzamidine-heme complexes in cell-free systems.
Bioorg. Med. Chem. Lett. 2004, 14, 1625–1628.
(47) Albert, A.; Seargent, E. P. The Determination of Ionisation Constants—A
Laboratory Manual, 3rd ed.; Chapman and Hall: New York, 1984.
JM0704752