Anti-Plasmodium activity of imidazolium and triazolium salts

Article abstract of DOI:10.1016/j.bmc.2010.05.020

We have previously reported that tetrazolium salts were both potent and specific inhibitors of Plasmodium replication, and that they appear to interact with a parasite component that is both essential and conserved. The use of tetrazolium salts in vivo is limited by the potential reduction of the tetrazolium ring to form an inactive, neutral acyclic formazan. To address this issue imidazolium and triazolium salts were synthesized and evaluated as Plasmodium inhibitors. Many of the imidazolium and triazolium salts were highly potent with active concentrations in the nanomolar range in Plasmodium falciparum cultures, and specific to Plasmodium with highly favorable therapeutic ratios. The results corroborate our hypothesis that an electron-deficient core is required so that the compound may thereby interact with a negatively charged moiety on the parasite merozoite; the side groups in the compound then form favorable interactions with adjacent parasite components and thereby determine both the potency and selectivity of the compound.

Full text of DOI:10.1016/j.bmc.2010.05.020

Bioorganic & Medicinal Chemistry 18 (2010) 6184–6196
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Anti-Plasmodium activity of imidazolium and triazolium salts
a,
Jason Z. Vlahakis a, Carmen Lazar a, Ian E. Crandall b, , Walter A. Szarek
*
*
a Department of Chemistry, Queen’s University, Kingston, Ontario, Canada K7L 3N6
b Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada M5S 1A8
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 18 January 2010
Revised 5 May 2010
Accepted 6 May 2010
Available online 12 May 2010
We have previously reported that tetrazolium salts were both potent and specific inhibitors of Plasmo-
dium replication, and that they appear to interact with a parasite component that is both essential and
conserved. The use of tetrazolium salts in vivo is limited by the potential reduction of the tetrazolium ring
to form an inactive, neutral acyclic formazan. To address this issue imidazolium and triazolium salts were
synthesized and evaluated as Plasmodium inhibitors. Many of the imidazolium and triazolium salts were
highly potent with active concentrations in the nanomolar range in Plasmodium falciparum cultures, and
specific to Plasmodium with highly favorable therapeutic ratios. The results corroborate our hypothesis
that an electron-deficient core is required so that the compound may thereby interact with a negatively
charged moiety on the parasite merozoite; the side groups in the compound then form favorable inter-
actions with adjacent parasite components and thereby determine both the potency and selectivity of the
compound.
Keywords:
Plasmodium falciparum
Imidazolium salts
Triazolium salts
Malaria
Selective inhibitors
Antimalarial drugs
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
ing a novel strategy, namely, the prevention of malaria parasites
from invading erythrocytes. The invasion of hepatocytes by Plas-
Malaria, which is caused by infection with protozoan parasites
of the genus Plasmodium, is a major global health problem and is
responsible annually for the deaths of over a million people, mainly
children in sub-Saharan Africa.1 At present, prevention and treat-
ment strategies are continuously hampered by the emergence of
resistance in parasites to newly introduced drugs. Further, there
are considerable costs involved in drug development, and the mar-
ket for antimalarials is hard to value and is prone to logistical prob-
lems.2 The development of effective malaria vaccines would
change this situation,3 however, in the absence of such vaccines,
and because of the widespread resistance to currently used anti-
malarial drugs such as chloroquine, new chemotherapeutic agents
are needed to help in the prevention and control of this disease.4
Two major approaches have been employed in the search for
new antimalarial drugs.5 The more widely used approach is the
development of chemical analogs of existing antimalarial agents.6
An alternative approach is the identification of novel drug targets
within the parasite, followed by the design of chemical entities act-
ing on these targets.7 Potential antimalarial chemotherapeutic tar-
gets have been classified into three main categories: inhibition of
processes occurring in the digestive vacuole, inhibition of the en-
zymes involved in macromolecular and metabolite synthesis, and
inhibition of membrane processing and signaling.8 We are pursu-
modium falciparum depends on the interaction of parasite surface
proteins with negatively charged carbohydrates, such as sialate
residues or sulfated glycosaminoglycans (GAGs), on the surface of
host cells. We observed9 that anionic agents similar to such nega-
tively charged carbohydrate structures also interfered with the en-
try of P. falciparum merozoites into human red blood cells, and,
hence our ‘first-generation’ inhibitors consisted of short-chain ali-
phatic polysulfonates that were effective at blocking merozoite en-
try into erythrocytes. Our ‘second-generation’ inhibitors were
based on sulfated cyclodextrins, and these provided a better defini-
tion of the structural features that inhibit the parasite/merozoite/
host red-cell interaction.10 During the course of an exploration of
the relationship between structure and activity of the sulfated
cyclodextrins, it was determined that the compounds interact with
a specific component of the erythrocyte membrane, namely, the
anion-transport protein AE1, causing a disruption to the invasion
process. The observation that nitrotetrazolium blue chloride, a
component of the parasite viability assays, formed stable com-
plexes with the inhibitory sulfated cyclodextrins led us to the
development of our ‘third-generation’ of anti-Plasmodium agents,
which were based on tetrazolium salts. In an extensive study,11
tetrazolium salts were found to be both potent and specific
inhibitors of Plasmodium invasion, and were found to interact with
a component of the parasite that is both essential and conserved.
This parasite component, which we believe to be a sulfated or phos-
phorylated moiety such as in the phosphoinositol component of the
Merozoite Surface Protein 1, is responsible for the recognition of
* Corresponding authors. Tel.: +1 416 581 7704 (I.E.C.); +1 613 533 2643 (W.A.S.).
E-mail addresses: ian.crandall@utoronto.ca (I.E. Crandall), walter.szarek@chem.
queensu.ca, szarekw@chem.queensu.ca (W.A. Szarek).
0968-0896/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2010.05.020

J. Z. Vlahakis et al. / Bioorg. Med. Chem. 18 (2010) 6184–6196
6185
erythrocytes by Plasmodium and allows the parasites to adhere to a
receptor domain in the AE1 found on red cells and thereby initiate
their invasion process.12,13 In the presence of tetrazolium inhibitors,
parasite entry into the erythrocyte is aborted and parasite cultures
become unviable. The structure of one such ‘third-generation’
inhibitor, namely, tetrazolium red (commercially available), is
shown in Figure 1. We hypothesize that the compound functions
by attracting the parasite ligand by way of its electron-deficient tet-
razolium ring and then forms favorable interactions involving its
side groups. While the tetrazolium salts can be highly potent and
selective for malaria parasites, a potential problem with this series
of compounds is the possible reduction of the tetrazolium ring to
form an inactive acyclic formazan (see Fig. 1); also, tetrazolium
compounds may be susceptible to photochemical degradation.
These issues of stability have been addressed in the present study
by the replacement of tetrazolium salts by imidazolium or triazoli-
um salts thereby allowing us to formulate the next generations of
potential Plasmodium invasion inhibitors. A general depiction of
the imidazolium and triazolium ring structures is shown in Figure
1. The anti-Plasmodium activities of a series of these cationic azo-
lium salts were evaluated; it was observed that many of the com-
pounds exhibited strong and specific inhibitory activity towards P.
falciparum cultures, similar to that observed for tetrazolium salts.
Indeed, these new imidazolium and triazolium containing com-
pounds have provided drug candidates possessing greater potency
and selectivity than the original tetrazolium compounds. Moreover,
the results corroborate our hypothesis that an electron-deficient
core is required for the compound’s interaction with a negatively
charged moiety on the parasite merozoite ligand and that the side
groups in the compound then form favorable interactions with adja-
cent parasite components and thereby determine potency.
methyl)naphthalene} in toluene following a modification (omission
of base) of a published procedure.15,16 Alternate procedures for the
formation of symmetrical 1,3-disubstituted imidazolium salts were
also employed. Compounds 16 and 17 were prepared (Scheme 1,
Protocol 2) by the dialkylation of imidazole with the appropriate
a-bromoketone in DMF. Compound 19 was prepared (Scheme 1,
Protocol 3) by the alkylation of 1-methylimidazole with iodometh-
ane in 1-propanol following a related published procedure.17 Also
shown in Scheme 1 (Protocol 3), the unsymmetrical 1,3-disubsti-
tuted imidazolium salts 2, 4, 5, and 18 were prepared by the alkyl-
ation of the appropriate 1-substituted imidazole with the
appropriate alkyl or benzyl halide in toluene following a reported
similar procedure.18
As shown in Scheme 3 and 1,4-disubstituted-[1,2,4]triazolium
salts were prepared by two main protocols. Compound 27 was pre-
pared (Protocol 5) by the alkylation of [1,2,4]triazole using benzyl
bromide in K2CO3–THF, following reported similar procedures.19,20
Similarly, compound 29 was prepared (Protocol 5) by the alkyl-
ation of [1,2,4]triazole using 4,20-dibromoacetophenone in DMF.
Using a different approach, compound 28 was prepared (Protocol
6) by the alkylation of 32 (free-base form) with benzyl bromide
in 1-propanol.
The synthesis of tri- and tetra-substituted[1,2,4]triazoles in-
volved the preparation of substituted triazole precursors. Thus,
3,5-diphenyl-[1,2,4]triazole was synthesized by the ring-forming
condensation reaction of benzoic hydrazide with benzonitrile.21
1-Benzyl-3,5-diphenyl-[1,2,4]triazole and 1-(4-bromobenzyl)-3,5-
diphenyl-[1,2,4]triazole were prepared by the alkylation of 3,5-di-
phenyl-[1,2,4]triazole with benzyl bromide and 4-bromobenzyl
bromide, respectively, in K2CO3–DMF.19 As shown in Scheme 2
(Protocol 4), the 1,3,4-trisubstituted-[1,2,4]triazolium salt 26 was
synthesized by the methylation of 4-phenyl-1-(3-phenyl-[1,2,4]-
triazol-1-yl)-butan-2-one in the 4-position using trimethyloxoni-
um tetrafluoroborate in 1,2-dichloroethane following a modifica-
tion of a reported similar procedure.22 Similarly, the 1,3,4,5-
tetrasubstituted-[1,2,4]triazolium salts (including 23–25) were
synthesized by the methylation of the appropriate 1,3,5-trisubsti-
tuted-[1,2,4]triazole in the 4-position using trimethyloxonium
tetrafluoroborate in 1,2-dichloroethane, as shown in Scheme 2
(Protocol 4). The wide diversity of available starting materials
(most notably substituted benzyl halides) allowed control over
the nature and position of substituents when designing these imi-
dazolium and triazolium compounds. This flexibility permitted a
systematic exploration of structure–activity relationships of sev-
eral imidazolium and triazolium salts (Table 1) with the goal of
finding effective antimalarial candidates.
2. Results and discussion
2.1. Synthesis
The imidazolium compounds (1–19) and triazolium compounds
(21–30) in Table 1 were synthesized by six main protocols depicted
in Schemes 1–3. As a precursor to some imidazolium compounds,
the 1-substituted imidazole, 1-(4-bromobenzyl)imidazole, was pre-
pared by thealkylation of imidazole with 4-bromobenzylbromide in
KOH–DMSO following a reported similar procedure.14 As shown in
Scheme 1 (Protocol 1), most of the symmetrical 1,3-disubstituted
imidazolium salts (including 1, 3, and 6–15) were synthesized di-
rectly by the treatment of imidazole with an excess amount of alkyl
halide {including substituted benzyl halides and 2-(bromo-
Figure 1. Tetrazolium red, generic structures of imidazolium and triazolium salts, and the reductive ring-opening reaction of tetrazolium salts to form acyclic formazans.

6186
J. Z. Vlahakis et al. / Bioorg. Med. Chem. 18 (2010) 6184–6196
Table 1
Activities of imidazolium and [1,2,4]triazolium compounds in P. falciparum and CHO cell culturesa
Compound
Structure
IC50
(lM)
IC50
(
lM)
IC50
P. falciparum
CHO cells
CHO/IC50
P. falciparum
Imidazolium compounds
N
N
N
N
Br
1
16
1
191
108
3
6
12
Br
2
3
0.9 0.2
120
N
N
Br
0.9 0.4
27
27
1
5
30
37
N
N
Br
4
0.73 0.03
Br
N
N
Br
5
2.9 0.9
484.5 0.9
167
Br
N
N
Br
6
17
3
2
56
4
1
3
F
F
N
N
Cl
7
9
105
12
Cl
Cl
N
N
Br
8
0.7 0.1
171.9 0.1
246
Br
Br
N
N
Br
9
0.35 0.06
33
1
94
I
I
N
N
Cl
Cl
Br
10
3.3 0.4
334.2 0.4
132 22
101
126
N
N
Br
Br
Br
11
1.05 0.06

J. Z. Vlahakis et al. / Bioorg. Med. Chem. 18 (2010) 6184–6196
6187
Table 1 (continued)
Compound
Structure
IC50
(l
M)
IC50
(
lM)
IC50
P. falciparum
CHO cells
CHO/IC50
P. falciparum
Cl
Cl
N
N
N
N
Br
12
13
2.0 0.2
110 23
55
Br
Br
Br
1.4 0.6
125 10
89
N
N
Cl
14
15
16
17
5
9
1
2
30.6 0.7
6
H3CO
OCH3
N
N
Br
227
7
2
25
20
9
O2N
Cl
NO2
Cl
H
Br
2.6 0.1
53
N
N
O
O
H
Br
N
N
0.5 0.1
4.3 0.3
O
O
N
N
N
N
Br
18
19
20
230 83
543 66
368
998 83
894 19
3675
4
2
I
HN
N
10
Triazolium compounds
CH3
N
N
21
3.1 0.8
307 37
99
N
I
S
CH3
(continued on next page)

6188
J. Z. Vlahakis et al. / Bioorg. Med. Chem. 18 (2010) 6184–6196
Table 1 (continued)
Compound
Structure
IC50
(lM)
IC50
(
lM)
IC50
P. falciparum
CHO cells
CHO/IC50
P. falciparum
H
N
N
22
69
8
224 21
3
N
Br
S
CH3
CH3
N
N
23
7.8 0.2
87
8
11
N
BF4
CH3
N
N
24
1.68 0.05
33
2
20
N
BF4
Br
CH3
N
N
25
0.10 0.08
143 14
1430
N
O
BF4
H
CH3
N
N
N
O
BF4
26
3.4 0.2
237 17
70
Br
N
N
27
100
8
74
4
0.7
N

J. Z. Vlahakis et al. / Bioorg. Med. Chem. 18 (2010) 6184–6196
6189
Table 1 (continued)
Compound
Structure
IC50
(l
M)
IC50
(
lM)
IC50
P. falciparum
CHO cells
CHO/IC50
P. falciparum
H
Br
N
N
28
73
3
261
3
4
N
O
Br
Br
H
Br
N
N
29
10.9 0.7
20.9 0.5
2
N
O
O
F
F
F
N
F
N
BF4
30
89
5
303 15
3
N
F
N
HN
N
31
2896
894
1273 35
0.4
0.4
O
HCl
N
N
32
334 11
N
a
Each IC50 value represents the mean of four determinations with standard error indicated. The ‘ratio of activities’ given in the fifth column represents the IC50 value
determined for CHO cells divided by the IC50 value determined for P. falciparum cultures. Values greater than unity indicate that the compound has greater potency in P.
falciparum cultures. The strain line used was ItG. Parallel experiments were performed using chloroquine as a positive control. IC50 values of 10–50 nM were obtained
indicating that the hematocrit, parasitemia, and incubation time used produced results that were comparable to those reported using other conditions.
2.2. Biological evaluation
whereas the general toxicity of these compounds was determined
using Chinese hamster ovary (CHO) cell cultures. It should be noted
that the Plasmodium strain employed is chloroquine-sensitive. The
IC50 values are listed in Table 1; a selectivity index was calculated
The imidazolium and triazolium compounds in Table 1 were
evaluated for anti-Plasmodium activity using P. falciparum cultures,
Scheme 1. Synthesis of symmetrical and unsymmetrical imidazolium salts.

6190
J. Z. Vlahakis et al. / Bioorg. Med. Chem. 18 (2010) 6184–6196
Scheme 2. Synthesis of 1,3,4-trisubstituted-[1,2,4]triazolium salts and 1,3,4,5-
tetrasubstituted-[1,2,4]triazolium salts.
by dividing the IC50 values obtained in CHO cell cultures by the IC50
values obtained in P. falciparum cultures. The relative toxicity of
these compounds was also compared graphically by plotting the
IC50 values obtained in P. falciparum cultures against the values ob-
tained in CHO cell cultures (Fig. 2). More than 90% of the com-
pounds tested in Table 1 were found to be selectively toxic to
Plasmodium as evidenced by the majority of data points in Figure 2
falling above the diagonal line that represents unity. This result is
not unexpected since our synthetic design was based on the pre-
mise that triazolium and imidazolium components would interact
with the parasite ligand, or some other parasite component, in a
manner similar to that seen for the tetrazolium structures. The lack
of a clear linear trend in this plot is also consistent with the com-
pounds having a mechanism of action, for example, invasion inhi-
bition, that is unique to Plasmodium, and, hence activity in the
parasite cultures was poorly correlated with CHO cell toxicity.
Microscopic examination of the inhibited assay wells containing
individual imidazolium or triazolium salts confirmed the presence
of a large number of extracellular merozoites and a lack of ring-
stage parasites (see Fig. 3), a feature consistent with the hypothesis
that inhibition of merozoite entry is also the possible primary
mechanism of action of the triazolium and imidazolium genera-
tions of compounds. Interestingly, 19 of the 29 imidazolium and
triazolium compounds exhibited modest selectivity (selectivity in-
dex >10) for Plasmodium. Furthermore, 11 of the 29 imidazolium
and triazolium compounds exhibited enhanced selectivity with
selectivity indexes >50, with one even approaching 1430 (com-
pound 25). Consistent with our hypothesis about the active motif,
the uncharged triazoles 31 and 32 were not selective for Plasmo-
dium. As regards anti-Plasmodium potency, 17 out of the 29 imi-
dazolium and triazolium compounds tested exhibit IC50 values of
Figure 2. IC50 values for P. falciparum cultures were plotted against values obtained
in CHO cell cultures. Points representing triazolium compounds (N) and imidazo-
lium compounds (.) are depicted.
progressive changes seen in the series of compounds 20, 19, 18,
and 1. Thus, there is a dramatic increase in both potency and selec-
tivity in going from the uncharged compound 20 to the charged
compound 1 having two benzyl moieties. In comparison to the par-
ent dibenzylated imidazolium compound
1
(IC50 = 16
1 lM,
selectivity index of 12), the replacement of one benzene ring by
a larger, more hydrophobic aromatic ring system such as that of
naphthalene was found to increase antimalarial potency and selec-
tivity. For example, compounds 2 and 3 both showed an 18-fold in-
crease in potency when compared to compound 1. Potency against
Plasmodium is also increased by the incorporation of a halogen
atom, apart from fluorine, in the phenyl substituents of these com-
pounds. The presence of halogen atoms would have an electron-
withdrawing effect on the imidazolium ring and thereby accentu-
ate the electron deficiency of this component. Thus, the addition of
halogens to the scaffold of 1 gave compounds 5 and 7–13, all of
which are significantly more potent against P. falciparum than 1.
For example, the addition of just one p-bromo substituent, gave
compound 5 which exhibits an IC50 value of 2.9 0.9 lM and a
selectivity index of 167, whereas the addition of two p-bromo sub-
stituents gave compound 8 which exhibits an IC50 value of
0.7 0.1 lM and a selectivity index of 246. The increased potency
afforded by halogen incorporation was also noticed when compar-
ing the results of imidazolium compounds 2 and 4, and also the
triazolium compounds 23 and 24. Symmetrical imidazolium com-
pounds having halogen atoms in the 4-position of the benzyl func-
tionality showed a clear trend in potency against P. falciparum that
was dependent on the size (and electronegativity) of the halogen.
For example, the IC50 values (P. falciparum) for the directly compa-
rable F, Cl, Br, I series of compounds (6, 7, 8, and 9) were 17 3,
less than 5 lM for P. falciparum, a result supportive of our hypoth-
esis that they interact with a specific parasite component.
Our analysis of the results obtained in Table 1 revealed many
specific structure–activity relationships. Our current hypothesis
is that potency and the selective interaction of the compounds re-
sult from a combination of an electron-deficient ring and hydro-
phobic side groups. This hypothesis is illustrated by the
Scheme 3. Synthesis of 1,4-disubstituted-[1,2,4]triazolium salts.

J. Z. Vlahakis et al. / Bioorg. Med. Chem. 18 (2010) 6184–6196
6191
Figure 3. Photomicrograph of parasite cultures treated with 7 lM of compound 8 (left) or untreated (right).
9
2, 0.7 0.1, and 0.35 0.06, respectively, the iodo compound 9
prepared. The imidazolium analog 1 was 6 times more potent
and 17 times more selective for Plasmodium. A comparison be-
tween imidazolium compound 16 and triazolium compound 29 re-
vealed similar results.
being the most potent; apart from the bromo and iodo compounds,
a parallel trend was observed for the selectivity of these com-
pounds for Plasmodium. As regards the effect of the positions of
the halogen substituents on the aromatic rings, in the chlorine ser-
ies (12, 10, and 7) the o-chloro compound 12 was the most potent
and the m-chloro compound 10 was the most selective, whereas,
for the bromine series (13, 11, and 8) the p-bromo compound 8
was the most potent and selective inhibitor. The presence of a p-
methoxy group (as in 14) or a p-nitro group (as in 15) resulted in
each case, in a lower IC50 value for P. falciparum when compared
to that of 1. Our hypothesis that an electron-deficient ring sur-
rounded by hydrophobic side groups could form an active motif
was also supported by the behavior of the triazolium compounds,
as it was observed that, of the compounds studied herein, those
having the more-substituted structures were the most-effective
drug candidates. Thus, the tetrasubstituted triazolium compounds
21, 23, 24, and 25 were more effective anti-Plasmodium agents than
the trisubstituted triazolium compounds 22, 26, and 30. Two direct
comparisons were evident. The addition of a methyl substituent to
the 5-position of 22 led to compound 21 which exhibited a 22-fold
increase in potency and a 30-fold increase in selectivity towards P.
falciparum. Similarly, the addition of a phenyl substituent to the 5-
position of 26 led to compound 25 which exhibited a 34-fold in-
crease in potency and a 21-fold increase in selectivity towards P.
falciparum. Compound 25, a 1,3,4,5-tetra-substituted triazolium
tetrafluoroborate, displayed remarkable antimalarial properties,
with an IC50 value for Plasmodium of 100 nM and a selectivity index
of 1430 over CHO cells. To exclude the possibility that the tetra-
fluoroborate counterion influenced the viability assays, sodium
tetrafluoroborate was also assayed as a control and IC50 values of
The precise nature of how these compounds affect the parasite
cultures is not clear. Non-synchronous parasite cultures were used
and the assay was allowed to incubate for 72 h. During this time
parasites remained at the ring stage of development for 24 h prior
to maturing to the point of merozoite production during the next
24 h. The parasitized cell then bursts releasing merozoites which
will then invade new erythrocytes and will form ring-stage para-
sites. There were several observations that suggested that the com-
pounds prevented merozoite entry instead of releasing rings from
previously infected cells. Thus, for example, (1) external parasite
forms were only seen after synchronized cultures had passed
through the mature stage of development, (2) a much larger num-
ber of parasites were observed after 48 h than were in the original
culture (suggesting replication had taken place), and (3) external
parasites can be immunostained with an antibody against merozo-
ite surface proteins (data not shown).
3. Conclusions
Several compounds that express excellent selectivity as inhibi-
tors of P. falciparum relative to CHO cells have been synthesized. In
particular, imidazolium salts 2, 5, and 8–11, and triazolium salts 21
and 24–26 are noteworthy for their high potency and selectivity
towards Plasmodium. The results obtained in P. falciparum and
CHO cell assays support the hypothesis that potency in parasite
cultures results from the presence of an electron-deficient ring at-
tached to hydrophobic side groups. Tetrazolium, triazolium, and
imidazolium salts all appear to provide candidates that show
acceptable levels of activity in P. falciparum cultures coupled with
selectivity for replicating parasites. These active compounds are
unrelated to currently used antimalarial agents and further they
have become very useful tools in elucidating the mechanistic as-
pects of how Plasmodium invades erythrocytes; their novel mech-
anism of action suggests that they could be developed for
prophylactic and therapeutic applications. The increased stability
of the current generation of compounds compared to the previ-
ously described tetrazolium series is a major step forward in the
development of antimalarial therapeutics, since it will permit the
evaluation of anti-invasion strategies in mouse models of malaria.
848 135 lM for P. falciparum and 1991 152 lM for CHO cells
were observed, results indicating it is a well-tolerated counterion
in both systems. The disubstituted triazolium compounds 27, 28,
and 29 were active, but were found to be neither highly potent
or selective agents against P. falciparum, suggesting that triazolium
and imidazolium structures have similar, but not interchangeable,
activities Thus, it was not surprising that the monosubstituted and
unsubstituted triazoles 32 and 31, respectively, were essentially
inactive, particularly as they do not conform to our putative motif.
Although the comparisons are not direct, it appears that the
presence of a b-keto functionality does increase potency, as can
be seen by comparison of results for compounds 27, 28, and 29,
and also for compounds 7 and 16. The electron-withdrawing prop-
erties of the keto functionality renders the central azolium ring
system more electron-deficient, and this feature may contribute
to the enhanced antimalarial activity. Indeed, some of our most-
effective tetrazolium inhibitors11 possessed strongly electron-
withdrawing substituents such as fluorine.
4. Experimental
4.1. General
Flash column chromatography was performed on Silicycle silica
gel (230–400 mesh, 60 Å). Analytical thin-layer chromatography
was performed on glass-backed pre-coated silica gel 60 F254 plates
(Silicycle), and the compounds were visualized either by UV illumi-
In an effort to determine the effect that the cationic core struc-
ture has on activity, the directly comparable dibenzylated imidazo-
lium and triazolium compounds 1 and 27, respectively, were

6192
J. Z. Vlahakis et al. / Bioorg. Med. Chem. 18 (2010) 6184–6196
nation (254 nm), or by heating after spraying with phosphomolyb-
dic acid in ethanol. Melting points were taken on a Mel-Temp II
apparatus and are uncorrected. 1H and 13C NMR spectra were re-
corded on a Bruker Avance 400 spectrometer in CDCl3, CD3OD, or
DMSO-d6. The chemical shifts are reported in d (ppm) relative to
tetramethylsilane.23 High-resolution ESI mass spectra were re-
corded on an Applied Biosystems/MDS Sciex QSTAR XL mass spec-
trometer with an Agilent HP1100 Cap-LC system. Samples were
oil; 1H NMR (400 MHz, DMSO-d6): d 5.43 (s, 4H), 7.39–7.43 (m,
10H), 7.83 (d, J = 1.5 Hz, 2H), 9.41 (s, 1H); 13C NMR (100 MHz,
DMSO-d6): d 52.1, 122.9, 128.3, 128.8, 129.0, 134.8, 136.3; HRMS
(ESI) [MꢁBr]+ Calcd for C17H17N2: 249.1391. Found: 249.1388.
4.5.2. 1,3-Bis-(naphthalen-2-ylmethyl)imidazolium bromide (3)
Compound 3 was prepared using imidazole and 2-(bromo-
methyl)naphthalene as starting materials to give the product
(754 mg, 48%) as a white solid; 1H NMR (400 MHz, DMSO-d6): d
5.61 (s, 4H), 7.55–7.58 (m, 7H), 7.90–7.97 (m, 7H), 8.00 (s, 2H),
9.50 (s, 1H); 13C NMR (100 MHz, DMSO-d6): d 52.3, 123.1, 125.7,
126.8, 127.6, 127.7, 127.9, 128.8, 132.2, 132.7, 132.8, 136.6; HRMS
(ESI) [MꢁBr]+ Calcd for C25H21N2: 349.1704. Found: 349.1700.
run in 50% aqueous MeOH at a flow rate of 6 lL/min. High-resolu-
tion EI mass spectra were recorded on a Waters/Micromass GC–
TOF instrument.
4.2. Materials
2-(Bromomethyl)naphthalene, 3-chlorobenzyl bromide, and 2-
chlorobenzyl bromide were obtained from Alfa AesarÒ (Ward Hill,
MA, USA). 5-Methyl-3-(methylthio)-1,4-diphenyl-1H-1,2,4-triazo-
lium iodide (21) and 3-(methylthio)-1,4-diphenyl-1H-1,2,4-triazo-
lium bromide (22) were obtained from Sigma–AldrichÒ Canada,
Ltd. (Oakville, ON, Canada). The rest of the chemical reagents were
obtained from Sigma–AldrichÒ Canada, Ltd. (Oakville, ON, Canada),
and were used without further purification. The syntheses of 40-
4.5.3. 1,3-Bis-(4-fluorobenzyl)imidazolium bromide (6)
Compound 6 was prepared using imidazole and 4-fluorobenzyl
bromide as starting materials to give the product (737 mg, 55%) as
an off-white oil; 1H NMR (400 MHz, DMSO-d6): d 5.40 (s, 4H), 7.24–
7.29 (m, 4H), 7.49–7.52 (m, 4H), 7.80 (d, J = 1.4 Hz, 2H), 9.36 (s,
1H); 13C NMR (100 MHz, DMSO-d6): d 51.4, 115.9, 116.1, 122.9,
131.0, 131.1, 136.2; HRMS (ESI) [MꢁBr]+ Calcd for C17H15F2N2:
285.1203. Found: 285.1196.
benzyl-2-bromoacetophenone,
1-(3,5-diphenyl-[1,2,4]triazol-1-
yl)-4-phenyl-butan-2-one, 4-phenyl-1-(3-phenyl-[1,2,4]triazol-1-
yl)-butan-2-one, and 32 will be reported elsewhere.
4.5.4. 1,3-Bis-(4-chlorobenzyl)imidazolium chloride (7)
Compound 7 was prepared using imidazole and 4-chlorobenzyl
chloride as starting materials to give the product (150 mg, 12%) as
a white sticky solid; 1H NMR (400 MHz, DMSO-d6): d 5.48 (s, 4H),
7.49–7.50 (m, 10H), 7.89 (d, J = 1.4 Hz, 2H), 9.68 (s, 1H); 13C NMR
(100 MHz, DMSO-d6): d 51.1, 122.9, 129.0, 130.5, 133.5, 133.8,
136.6; HRMS (ESI) [MꢁCl]+ Calcd for C17H15Cl2N2: 317.0612.
Found: 317.0628.
4.3. Representative procedure for the formation of 1-
substituted imidazole precursors
4.3.1. 1-(4-Bromobenzyl)imidazole
Imidazole (400 mg, 5.80 mmol, 1 equiv) was added to a mixture
of DMSO (10 mL) and potassium hydroxide (394 mg, 7.05 mmol,
1.2 equiv) and the mixture was stirred for 30 min at 80 °C under
an atmosphere of N2. 4-Bromobenzyl bromide (1.76 g, 7.05 mmol,
1.2 equiv) was added and the mixture stirred for 2.5 h at 80 °C. The
mixture was cooled to rt, water was added, and the mixture was
extracted with ether (2 ꢀ 25 mL). The combined organic phase
was washed with water, dried over Na2SO4, and concentrated.
The residue was purified by flash column chromatography on silica
gel (30:1 v/v CHCl3–CH3OH as eluent, then CH3OH) to give the
product (680 mg, 2.87 mmol, 49%) as an orange oil; 1H NMR
(400 MHz, DMSO-d6): d 5.41 (s, 2H), 7.40 (d, J = 8.5 Hz, 2H), 7.63
(d, J = 8.5 Hz, 2H), 7.84 (s, 1H), 9.39 (s, 1H); 13C NMR (400 MHz,
DMSO-d6): d 51.3, 119.8, 122.2, 122.9, 130.8, 132.0, 134.0, 136.4;
HRMS (EI) [M]+ Calcd for C10H9N2Br: 235.9949. Found: 235.9940.
4.5.5. 1,3-Bis-(4-bromobenzyl)imidazolium bromide (8)
Compound 8 was prepared using imidazole and 4-bromobenzyl
bromide as starting materials to give the product (263 mg, 47%) as
a white solid; 1H NMR (400 MHz, DMSO-d6): d 5.43 (s, 4H), 7.40 (d,
J = 8.5 Hz, 4H), 7.60 (d, J = 8.5 Hz, 4H), 7.85 (d, J = 1.6 Hz, 2H), 9.43
(s, 1H); 13C NMR (100 MHz, DMSO-d6): d 51.3, 122.2, 122.9, 130.7,
131.9, 134.0, 136.4; HRMS (ESI) [MꢁBr]+ Calcd for C17H15Br2N2:
404.9601. Found: 404.9571.
4.5.6. 1,3-Bis-(4-iodobenzyl)imidazolium bromide (9)
Compound 9 was prepared using imidazole and 4-iodobenzyl
bromide as starting materials to give the product (66 mg, 28%) as
a white solid; 1H NMR (400 MHz, DMSO-d6): d 5.39 (s, 4H), 7.22
(d, J = 8.3 Hz, 4H), 7.79 (d, J = 8.3 Hz, 4H), 7.82 (d, J = 1.5 Hz, 2H),
9.38 (s, 1H); 13C NMR (100 MHz, DMSO-d6): d 51.4, 95.4, 122.9,
130.6, 134.3, 136.4, 137.7; HRMS (ESI) [MꢁBr]+ Calcd for
4.4. General procedure for the formation of symmetrical di-
substituted imidazolium salts (Scheme 1)
C17H15I2N2: 500.9324. Found: 500.9327.
Under an atmosphere of N2, the benzyl or alkyl halide (11.01
mmol, 3 equiv) in toluene (5 mL) was added dropwise to a solution
of imidazole (3.67 mmol, 1 equiv) in toluene (10 mL). The mixture
was stirred for 30 min at rt and then for 72 h at 70 °C. After re-
moval of the toluene, the crude product was washed with diethyl
ether to remove the excess of benzyl or alkyl halide. Purification
by flash column chromatography on silica gel (9:1 v/v CH2Cl2–
CH3OH) followed by recrystallization from 2-propanol afforded
the corresponding symmetrical di-substituted imidazolium salt.
4.5.7. 1,3-Bis-(3-chlorobenzyl)imidazolium bromide (10)
Compound 10 was prepared using imidazole and 3-chloroben-
zyl bromide as starting materials to give the product (1.00 g,
68%) as a white solid; 1H NMR (400 MHz, DMSO-d6): d 5.45 (s,
4H), 7.38–7.43 (m, 2H), 7.45–7.48 (m, 4H), 7.55 (s, 2H), 7.85 (d,
J = 1.4 Hz, 2H), 9.41 (s, 1H); 13C NMR (100 MHz, DMSO-d6):
d 51.3, 122.9, 127.2, 128.4, 128.8, 130.9, 133.5, 136.6, 137.0;
HRMS (ESI) [MꢁBr]+ Calcd for C17H15Cl2N2: 317.0612. Found:
317.0601.
4.5. Characterization of the compounds synthesized following
the general procedure for the formation of symmetrical di-
substituted imidazolium salts
4.5.8. 1,3-Bis-(3-bromobenzyl)imidazolium bromide (11)
Compound 11 was prepared using imidazole and 3-bromoben-
zyl bromide as starting materials to give the product (581 mg, 32%)
as a white solid; 1H NMR (400 MHz, DMSO-d6): d 5.43 (s, 4H), 7.38–
7.45 (m, 4H), 7.60–7.62 (m, 2H), 7.69 (s, 2H), 7.85 (d, J = 1.4 Hz,
2H), 9.39 (s, 1H); 13C NMR (100 MHz, DMSO-d6): d 51.3, 122.1,
4.5.1. 1,3-Dibenzylimidazolium bromide (1)
Compound 1 was prepared using imidazole and benzyl bromide
as starting materials to give the product (724 mg, 87%) as a clear

J. Z. Vlahakis et al. / Bioorg. Med. Chem. 18 (2010) 6184–6196
6193
123.0, 127.6, 131.2, 131.2, 131.7, 136.7, 137.3; HRMS (ESI) [MꢁBr]+
a solution of 1-(4-benzylphenyl)-2-bromoethanone (1.21 g, 4.18
mmol, 1 equiv) in DMF (5 mL) was added imidazole (0.85 g,
12.49 mmol, 3 equiv) under an atmosphere of N2. The mixture
was stirred at rt for 3 h, then slowly poured into water (100 mL),
and stirring was continued for 0.5 h. The resulting white precipi-
tate was removed by filtration, and the solid was washed with boil-
ing toluene (50 mL) to remove the monoalkylated product. The
toluene-insoluble material was dried under high vacuum and then
recrystallized from EtOH (5 mL). The solid was removed by filtra-
tion and washed with EtOH (1 mL). High-vacuum drying gave 17
(131 mg, 0.23 mmol, 11%) as a white solid; mp 215–217 °C; 1H
NMR (400 MHz, DMSO-d6): d 4.07 (s, 4H), 6.11 (s, 4H), 7.18–7.34
(m, 10H), 7.50 (d, J = 8.4 Hz, 4H), 7.76 (d, J = 1.2 Hz, 2H), 7.99 (d,
J = 8.0 Hz, 4H), 9.08 (s, 1H); 13C NMR (100 MHz, DMSO-d6): d
41.0, 55.5, 123.7, 126.3, 128.5, 128.6, 128.8, 129.4, 131.8, 138.6,
140.4, 148.5, 190.8; HRMS (ESI) [MꢁBr]+ Calcd for C33H29N2O2:
485.2229. Found: 485.2219.
Calcd for C17H15Br2N2: 404.9601. Found: 404.961.
4.5.9. 1,3-Bis-(2-chlorobenzyl)imidazolium bromide (12)
Compound 12 was prepared using imidazole and 2-chloroben-
zyl bromide as starting materials to give the product (660 mg,
45%) as a white solid; 1H NMR (400 MHz, DMSO-d6): d 5.59 (s,
4H), 7.45–7.48 (m, 6H), 7.55–7.57 (m, 2H), 7.82 (d, J = 1.4 Hz,
2H), 9.44 (s, 1H); 13C NMR (100 MHz, DMSO-d6): d 50.2, 123.1,
128.0, 129.9, 130.9, 131.0, 131.9, 132.9, 137.4; HRMS (ESI) [MꢁBr]+
Calcd for C17H15Cl2N2: 317.0612. Found: 317.0613.
4.5.10. 1,3-Bis-(2-bromobenzyl)imidazolium bromide (13)
Compound 13 was prepared using imidazole and 2-bromoben-
zyl bromide as starting materials to give the product (846 mg, 47%)
as a beige solid; 1H NMR (400 MHz, DMSO-d6): d 5.56 (s, 4H), 7.36–
7.40 (m, 4H), 7.47–7.51 (m, 2H), 7.73 (d, J = 7.9 Hz, 2H), 7.81 (d,
J = 1.4 Hz, 2H), 9.37 (s, 1H); 13C NMR (100 MHz, DMSO-d6): d
52.4, 123.1, 123.2, 128.5, 130.7, 131.1, 133.2, 133.5, 137.5; HRMS
(ESI) [MꢁBr]+ Calcd for C17H15Br2N2: 404.9601. Found: 404.9616.
4.6.3. 1,3-Dimethylimidazolium iodide (19)
To
a solution of 1-methylimidazole (202 mg, 2.46 mmol,
1 equiv) in 1-propanol (2 mL) at rt was added iodomethane
(0.31 mL, 4.98 mmol, 2 equiv) under an atmosphere of N2. The mix-
ture was heated at 100 °C for 19 h, then cooled to 0 °C. Et2O was
added slowly with stirring resulting in a yellow precipitate that
was removed by filtration and washed sequentially with Et2O
(3ꢀ) and also with EtOAc (5ꢀ). High-vacuum drying gave 19
(470 mg, 2.10 mmol, 85%) as a yellow solid; mp 78–80 °C; 1H
NMR (400 MHz, CD3OD): d 3.93 (s, 6H), 7.56 (s, 2H), 8.87 (s, 1H);
13C NMR (100 MHz, CD3OD): d 36.6, 124.8, 138.6; HRMS (ESI)
[MꢁI]+ Calcd for C5H9N2: 97.0765. Found: 97.0765; HRMS (ESI)
[MꢁC5H9N2]ꢁ Calcd for I: 126.9045. Found: 126.9049.
4.5.11. 1,3-Bis-(4-methoxybenzyl)imidazolium chloride (14)
Compound 14 was prepared using imidazole and 4-methoxy-
benzyl chloride as starting materials to give the product (550 mg,
43%) as a white solid; 1H NMR (400 MHz, DMSO-d6): d 3.76 (s,
3H), 5.32 (s, 4H), 6.96 (d, J = 8.6 Hz, 4H), 7.37 (d, J = 8.6 Hz, 4H),
7.58 (d, J = 1.5 Hz, 2H), 9.32 (s, 1H); 13C NMR (100 MHz, DMSO-
d6): d 51.6, 55.2, 114.4, 122.6, 129.5, 130.2, 135.7, 159.6; HRMS
(ESI) [MꢁCl]+ Calcd for C19H21N2O2: 309.1603. Found: 309.1617.
4.5.12. 1,3-Bis-(4-nitrobenzyl)imidazolium bromide (15)
Compound 15 was prepared using imidazole and 4-nitrobenzyl
bromide as starting materials to give the product (731 mg, 47%) as
a white solid; 1H NMR (400 MHz, DMSO-d6): d 5.62 (s, 4H), 7.66 (d,
J = 8.5 Hz, 4H), 7.9 (d, J = 0.9 Hz, 2H), 8.29 (d, J = 8.5 Hz, 4H), 9.45 (s,
1H); 13C NMR (100 MHz, DMSO-d6): d 51.2, 123.3, 124.1, 129.6,
137.3, 141.9, 147.7; HRMS (ESI) [MꢁBr]+ Calcd for C17H15N4O4:
339.1093. Found: 339.110.
4.7. General procedure for the formation of unsymmetrical di-
substituted imidazolium salts (Scheme 1)
Under an atmosphere of N2, the benzyl or alkyl halide
(4.40 mmol, 1.2 equiv) in toluene (5 mL) was added dropwise to
a solution of 1-benzylimidazole, 1-methylimidazole, or another
1-substituted imidazole (3.67 mmol, 1 equiv) in toluene (10 mL).
The mixture was stirred for 30 min at rt and then for 72 h at
70 °C. After removal of the toluene, the crude product was washed
with Et2O to remove the excess of benzyl or alkyl bromide. Recrys-
tallization from 2-propanol gave the corresponding unsymmetrical
disubstituted imidazolium salt.
4.6. Alternate procedures for the formation of symmetrical
disubstituted imidazolium salts (Scheme 1)
4.6.1. 1,3-Bis-[2-oxo-2-(4-chlorophenyl)ethyl]imidazolium
bromide (16)
This compound was obtained as the minorproduct in a procedure
designedfor thesynthesisofthemonoalkylatedanalog. Toa solution
of 2-bromo-1-(4-chlorophenyl)ethanone (3.50 g, 15 mmol, 1 equiv)
in DMF (15 mL) was added imidazole (3.06 g, 45 mmol, 3 equiv) un-
der an atmosphere of N2. The mixture was stirred at rt for 3 h, then
slowly poured into water (250 mL), and stirring was continued for
0.5 h. The resulting white precipitate was removed by filtration,
and the solid was washed with boiling toluene (100 mL) to remove
the monoalkylated product. The toluene-insoluble material was
dried under high vacuum and then recrystallized from 1:1 v/v
MeOH–EtOH. The solid was removed by filtration and washed with
EtOH. High-vacuum drying gave 16 (438 mg, 0.96 mmol, 13%) as a
white solid; mp >260 °C; 1H NMR (400 MHz, DMSO-d6): d 6.17 (s,
4H), 7.74 (d, J = 8.4 Hz, 4H), 7.80 (s, 2H), 8.09 (d, J = 8.4 Hz, 4H),
9.09 (s, 1H); 13C NMR (100 MHz, DMSO-d6): d 55.6, 123.7, 129.2,
130.0, 132.5, 138.5, 139.3, 190.5; HRMS (ESI) [MꢁBr]+ Calcd for
4.8. Characterization of the compounds synthesized following
the general procedure for the formation of unsymmetrical
disubstituted imidazolium salts
4.8.1. 1-Benzyl-3-(naphthalen-2-yl-methyl)imidazoliumbromide
(2)
Compound 2 was prepared using 1-benzylimidazole and 2-
(bromomethyl)naphthalene as starting materials to give the prod-
uct (46 mg, 32%) as a white solid; 1H NMR (400 MHz, CD3OD): d
5.43 (s, 2H), 5.60 (s, 2H), 7.41–7.45 (m, 5H), 7.47 (dd, J = 2 Hz,
1H), 7.52–7.56 (m, 1H), 7.64–7.69 (m, 1H), 7.88–7.93 (m, 3H),
7.94 (s, 1H), 7.96 (s, 1H), 9.26 (s, 1H); 13C NMR (100 MHz, CD3OD):
d 54.3, 54.4, 124.1, 124.2, 126.5, 128.0, 128.1, 128.9, 129.1, 129.4,
129.7, 130.4, 130.5, 132.4, 134.8, 134.9, 135.2, 137.7; HRMS (ESI)
[MꢁBr]+ Calcd for C21H19N2: 299.1548. Found: 299.1553.
C19H15Cl2N2O2: 373.0510. Found: 373.0488.
4.6.2. 1,3-Bis-[2-oxo-2-(4-benzylphenyl)ethyl]imidazolium
bromide (17)
4.8.2. 1-(4-Bromobenzyl)-3-(naphthalen-2-ylmethyl)imidazoli-
um bromide (4)
This compound was obtained as the minor product in a proce-
dure designed for the synthesis of the monoalkylated analog. To
Compound 4 was prepared using 1-(4-bromobenzyl)imidazole
and 2-(bromomethyl)naphthalene as starting materials, recrystal-

6194
J. Z. Vlahakis et al. / Bioorg. Med. Chem. 18 (2010) 6184–6196
lized from 2:1 v/v 2-propanol–Et2O to give the product (312 mg,
64%) as a yellow-brown oil; 1H NMR (400 MHz, DMSO-d6): d 5.43
(s, 2H), 5.60 (s, 2H), 7.39 (d, J = 8.5 Hz, 2H), 7.52–7.58 (m, 3H),
7.63 (d, J = 8.5 Hz, 2H), 7.84 (s, 1H), 7.88 (s, 1H), 7.91–7.99 (m,
4H), 9.43 (s, 1H); 13C NMR (100 MHz, DMSO-d6): d 51.3, 52.3,
122.1, 122.8, 123.0, 125.7, 126.7, 126.7, 127.6, 127.8, 128.8,
130.6, 131.9, 131.9, 132.7, 132.7, 134.0, 136.5; HRMS (ESI) [MꢁBr]+
Calcd for C21H18BrN2: 377.0653. Found: 377.0641.
4.10.2. 1-(4-Bromobenzyl)-3,5-diphenyl-[1,2,4]triazole
Under an atmosphere of N2, mixture of 3,5-diphenyl-
a
[1,2,4]triazole (500 mg, 2.26 mmol) and potassium carbonate
(942 mg, 6.78 mmol) in DMF (10 mL) was stirred at 75 °C for 1 h.
A solution of 4-bromobenzyl bromide (1.41 g, 5.65 mmol) in DMF
(3 mL) was added, and the mixture was stirred at reflux tempera-
ture for 48 h; the progress of the reaction was monitored by TLC.
The mixture was cooled to rt, diluted with EtOAc (25 mL), and
the organic layer washed sequentially with water (2 ꢀ 25 mL)
and a saturated aqueous solution of NaHCO3 (2 ꢀ 25 mL), dried
(Na2SO4), and concentrated to an orange oil. Purification by flash
column chromatography on silica gel (CHCl3 as eluent to remove
impurities, followed by elution of the product with MeOH) gave
the product (563 mg, 64%) as an orange oil; 1H NMR (400 MHz,
DMSO-d6): d 5.55 (s, 2H), 7.11 (d, J = 8.3 Hz, 2H), 7.44–7.56 (m,
8H), 7.7–7.72 (m, 2H), 8.05 (s, 1H), 8.072–8.074 (d, J = 1 Hz, 1H);
13C NMR (100 MHz, DMSO-d6): d 51.79, 120.92, 125.85, 127.48,
128.52, 128.75, 128.98, 129.1, 129.31, 130.34, 130.64, 131.66,
135.7, 155.6, 160.42; HRMS (EI) [M]+ Calcd for C21H16N3Br:
389.0528. Found: 389.0536.
4.8.3. 1-Benzyl-3-(4-bromobenzyl)imidazolium bromide (5)
Compound 5 was prepared using 1-benzylimidazole and 4-bro-
mobenzyl bromide as starting materials to give the product
(330 mg, 51%) as a beige solid; 1H NMR (400 MHz, CD3OD): d
5.44 (s, 2H), 5.45 (s, 2H), 7.38–7.436 (m, 4H), 7.63–7.65 (m, 4H),
7.86 (d, J = 1.6 Hz, 2H), 9.47 (s, 1H); 13C NMR (100 MHz, CD3OD):
d 51.2, 52.0, 122.2, 122.8, 122.9, 128.4, 128.8, 129.0, 130.7, 131.9,
134.1, 134.7, 136.3; HRMS (ESI) [MꢁBr]+ Calcd for C17H16BrN2:
327.0496. Found: 327.0513.
4.8.4. 3-Methyl-1-propylimidazolium bromide (18)
Compound 18 was prepared using 1-methylimidazole and 1-
bromopropane as starting materials to give the product (519 mg,
83%) as an orange oil; 1H NMR (400 MHz, CD3OD): d 0.96 (t,
J = 7.4 Hz, 3H), 1.88 (m, 2H), 3.96 (s, 3H), 4.19 (t, J = 7.2 Hz, 2H),
7.60 (s, 1H), 7.67 (s, 1H), 9.02 (s, 1H); 13C NMR (100 MHz, CD3OD):
d 10.9, 24.5, 36.6, 52.3, 123.7, 125.0, 137.9; HRMS (ESI) [MꢁBr]+
Calcd for C9H13N2: 125.1078. Found: 125.1084.
4.11. Representative procedures for the formation of 1,3,4,5-
tetrasubstituted-[1,2,4]triazolium salts (Scheme 2)
4.11.1. 1-Benzyl-4-methyl-3,5-diphenyl-[1,2,4]triazolium
tetrafluoroborate (23)
To a solution of 1-benzyl-3,5-diphenyl-[1,2,4]triazole (200 mg,
0.64 mmol) in 1,2-dichlorethane (2 mL) was added a solution of
trimethyloxonium tetrafluoroborate (104 mg, 0.70 mmol) in 1,2-
dichlorethane (1 mL) under an atmosphere of N2. The mixture
was heated at 65 °C with stirring for 5 h; the progress of the reac-
tion was monitored by TLC (9:1 v/v CH2Cl2–CH3OH as eluent). The
mixture was concentrated to a clear oil that was purified by flash
column chromatography on silica gel (CHCl3 as eluent to remove
impurities, followed by elution of the product with MeOH). High-
vacuum drying gave 23 (175 mg, 66%) as a white solid; 1H NMR
(400 MHz, DMSO-d6): d 3.66 (s, 3H), 5.55 (s, 2H), 7.26–7.28 (m,
2H), 7.36–7.38 (m, 3H), 7.69–7.87 (m, 10H); 13C NMR (100 MHz,
DMSO-d6): d 34.6, 54.1, 119.3, 123.1, 128.1, 128.7, 128.8, 129.4,
129.5, 129.8, 130.2, 132.2, 133.3, 133.5, 152.3, 154.0; HRMS (ESI)
[MꢁBF4]+ Calcd for C22H20N3: 326.1657. Found: 326.1663.
4.9. Representative procedure for the formation of 3,5-disub-
stituted [1,2,4]triazoles
4.9.1. 3,5-Diphenyl-[1,2,4]triazole
Under an atmosphere of N2, a mixture of benzoic hydrazide
(4.00 g, 29.30 mmol, 1 equiv) and benzonitrile (39.60 g, 384.00
mmol, 13.1 equiv) was stirred at reflux temperature for 14 h in a
round-bottom flask equipped with a Dean–Stark apparatus to re-
move water. The mixture was cooled to rt, and the resulting pre-
cipitate collected by filtration and washed with 2-propanol.
Recrystallization from 2-propanol gave the product (3.77 g, 58%)
as a white solid; 1H NMR (400 MHz, DMSO-d6): d 7.41–7.59 (m,
6H), 8.07–8.10 (m, 2H), 8.12 (1s, 1H); 13C NMR (100 MHz, DMSO-
d6): d 125.9, 126.1, 128.7, 129.1, 130.2, 131.3; HRMS (EI) [M]+ Calcd
for C14H11N3: 221.0953. Found: 221.0948.
4.11.2. 1-(4-Bromobenzyl)-4-methyl-3,5-diphenyl-[1,2,4]triazo-
lium tetrafluoroborate (24)
To a solution of 1-(4-bromobenzyl)-3,5-diphenyl-[1,2,4]triazole
(300 mg, 0.77 mmol) in 1,2-dichlorethane (5 mL) was added a
solution of trimethyloxonium tetrafluoroborate (136 mg,
0.92 mmol) in 1,2-dichlorethane (2 mL) under an atmosphere of
N2. The mixture was heated at 65 °C with stirring for 10 h; the pro-
gress of the reaction was monitored by TLC (9:1 v/v CH2Cl2–CH3OH
as eluent). The mixture was concentrated to a clear oil. Purification
by flash column chromatography on silica gel (CHCl3 as eluent to
remove impurities, followed by elution of the product with MeOH)
gave 24 (184 mg, 49%) as a white solid; 1H NMR (400 MHz, DMSO-
d6): d 3.66 (s, 3H), 5.54 (s, 2H), 7.24 (d, J = 8.1 Hz, 2H), 7.57 (d,
J = 8.1 Hz, 3H), 7.71–7.84 (m, 8H), 7.86 (d, J = 0.7 Hz, 2H); 13C
4.10. Representative procedures for the formation of 1,3,5-
trisubstituted [1,2,4]triazoles
4.10.1. 1-Benzyl-3,5-diphenyl-[1,2,4]triazole
Under an atmosphere of N2,
a mixture of 3,5-diphenyl-
[1,2,4]triazole (250 mg, 1.13 mmol) and potassium carbonate
(471 mg, 3.39 mmol) in DMF (5 mL) was stirred at 75 °C for 1 h.
A solution of benzyl bromide (386 mg, 2.26 mmol) in DMF (3 mL)
was added, and the mixture heated at reflux temperature with stir-
ring for 48 h; the progress of the reaction was monitored by TLC.
The mixture was cooled to rt, diluted with EtOAc (25 mL), and
the organic layer washed sequentially with water (2 ꢀ 25 mL)
and a saturated aqueous solution of NaHCO3 (2 ꢀ 25 mL), dried
(Na2SO4), and concentrated to a beige solid. Recrystallization from
hexanes gave the product (282 mg, 3.2 mmol, 80%) as a white so-
lid; 1H NMR (400 MHz, CDCl3): d 7.41–7.59 (m, 6H), 8.07–8.10
(m, 2H), 8.12 (s, 1H); 13C NMR (100 MHz, DMSO-d6): d 52.4,
120.9, 125.8, 126.8, 127.6, 127.8, 128.5, 128.8, 129.0, 129.3,
130.3, 130.7, 136.3, 155.6, 160.3; HRMS (EI) [M]+ Calcd for
NMR (100 MHz, DMSO-d6):
d 34.58, 53.35, 119.21, 121.99,
123.03, 129.35, 129.46, 129.84, 130.19, 130.43, 131.64, 132.27,
132.65, 133.51, 152.42, 154.08; HRMS (ESI) [MꢁBF4]+ Calcd for
C22H19BrN3: 404.0762. Found: 404.0771.
4.11.3. 4-Methyl-1-(2-oxo-4-phenylbutyl)-3,5-diphenyl-[1,2,4]
triazolium tetrafluoroborate (25)
To a solution of 1-(3,5-diphenyl-[1,2,4]triazol-1-yl)-4-phenyl-
butan-2-one (107 mg, 0.29 mmol, 1 equiv) in 1,2-dichloroethane
(2 mL) was added trimethyloxonium tetrafluoroborate (46 mg,
C21H17N3: 311.1422. Found: 311.1426.

J. Z. Vlahakis et al. / Bioorg. Med. Chem. 18 (2010) 6184–6196
6195
0.31 mmol, 1.07 equiv) in 1,2-dichloroethane (1 mL) under an atmo-
sphere of N2. The mixture was heated at 65 °C with stirring for 3.5 h,
and then concentratedto a golden oil. High-vacuum drying left a foa-
my white solid which was purified by flash chromatography on silica
gel (CHCl3 as eluent to remove impurities, followed by elution of the
product with EtOAc) gave 25 (49 mg, 0.10 mmol, 34%) as a clear oil;
Rf = 0 (CHCl3); 1H NMR (400 MHz, CD3OD): d 2.82–2.93 (m, 4H), 3.76
(s, 3H), 5.42 (s, 2H), 7.08–7.25 (m, 5H), 7.60–7.95 (m, 10H); 13C NMR
(100 MHz, CD3OD): d 30.0, 35.3, 42.2, 60.4, 120.3, 124.4, 127.3, 129.3,
129.5, 130.5, 130.7, 131.0, 131.2, 133.5, 135.0, 141.5, 155.3, 156.4,
201.8; 19F NMR (376 MHz, CD3OD): dꢁ156; 1H–1H NOESY: no obser-
vable NOE between protons at 5.42 ppm and the protons at
3.76 ppm (confirming methylation at the 4-position); HRMS (ESI)
[MꢁBF4]+ Calcd for C25H24N3O: 382.1919. Found: 382.1903. HRMS
(ESI) [MꢁC25H24N3O]ꢁ Calcd for BF4: 87.0029. Found: 87.0031.
and the resulting precipitate was removed by filtration, washed
with Et2O (10ꢀ), and dried under high vacuum. Purification by
flash column chromatography on silica gel (9:1 v/v CHCl3–MeOH
as eluent) gave 28 (24 mg, 0.07 mmol, 12%) as a beige solid; mp
166–170 °C; 1H NMR (400 MHz, DMSO-d6): d 5.66 (s, 2H), 6.31 (s,
2H), 7.40–7.55 (m, 5H), 7.63 (app t, J = 7.8 Hz, 2H), 7.77 (app t,
J = 7.4 Hz, 1H), 8.00–8.10 (m, 2H), 9.45 (s, 1H), 10.22 (s, 1H); 13C
NMR (100 MHz, DMSO-d6): d 50.9, 58.4, 128.5, 128.8, 129.3,
129.4 (2C), 133.5, 133.7, 134.9, 144.4, 144.9, 190.6; HRMS (ESI)
[MꢁBr]+ Calcd for C17H16N3O: 278.1293. Found: 278.1283.
4.13.3. 1,4-Bis-[2-oxo-2-(4-bromophenyl)ethyl]-[1,2,4]triazoli-
um bromide (29)
This compoundwas obtained as the minor product in a procedure
designedforthesynthesisof themonoalkylatedanalog. Toa solution
of 2-bromo-1-(4-bromophenyl)ethanone (1.12 g, 4 mmol, 1 equiv)
in DMF (6 mL) was added [1,2,4]triazole (828 mg, 12 mmol, 3 equiv)
under an atmosphere of N2. The mixture was stirred at rt for 3 h, then
water was added. The resulting white precipitate was removed by
filtration, and the solid was washed with boiling benzene (50 mL)
to remove the monoalkylated product. The benzene insoluble mate-
rial was dried under high vacuum and then purified by flash column
chromatography on silica gel (9:1 v/v EtOAc–MeOH as eluent). The
yellow solid obtained was washed once with MeOH to remove the
yellow color. Recrystallization from MeOH gave 29 (100 mg,
0.18 mmol, 9%) as a white solid; mp 244–245 °C; 1H NMR (400
MHz, DMSO-d6): d 6.20 (s, 2H), 6.40 (s, 2H), 7.80–7.92 (m, 4H),
7.95–8.08 (m, 4H), 9.24 (s, 1H), 10.08 (s, 1H); 13C NMR (100 MHz,
DMSO-d6): d 54.3, 58.5, 129.0, 129.1, 130.3, 130.4, 132.3, 132.4
(2C), 132.6, 145.4, 145.9, 189.8, 189.9; HRMS (ESI) [MꢁBr]+ Calcd
for C18H14Br2N3O2: 461.9452. Found: 461.9440.
4.12. Representative procedure for the formation of 1,3,4-
trisubstituted [1,2,4]triazolium salts (Scheme 2)
4.12.1. 4-Methyl-1-(2-oxo-4-phenylbutyl)-3-phenyl-[1,2,4]tria-
zolium tetrafluoroborate (26)
Compound 26 was prepared using the procedure for the forma-
tion of 1,3,4,5-tetrasubstituted-[1,2,4]triazolium salts above to
give a white solid in 40% yield from 4-phenyl-1-(3-phenyl-
[1,2,4]triazol-1-yl)-butan-2-one and trimethyloxonium tetrafluo-
roborate; mp ꢂ50 °C; Rf = 0.05 (CHCl3); 1H NMR (400 MHz,
CD3OD): d 2.92–3.05 (m, 4H), 4.01 (s, 3H), 5.53 (s, 2H), 7.15–7.20
(m, 1H), 7.21–7.30 (m, 4H), 7.60–7.73 (m, 3H), 7.78–7.84 (m,
2H); 13C NMR (100 MHz, CD3OD): d 30.0, 35.3, 42.3, 60.8, 123.9,
127.3, 129.4, 129.6, 130.5 (3C), 133.5, 141.8, 156.1, 201.2; 1H–1H
NOESY: no observable NOE between protons at 4.01 ppm and the
protons at 5.53 ppm (confirming methylation at the 4-position);
HRMS (ESI) [MꢁBF4]+ Calcd for C19H20N3O: 306.1606. Found:
306.1603.
4.14. Determination of anti-Plasmodium activity
P. falciparum cultures were grown in O+ blood obtained by
venipuncture of volunteers. Cultures of the laboratory line 3D7
were maintained by the method of Trager and Jensen24 using
RPMI-1640 supplemented with 10% human serum (a kind gift
obtained under ethical consent from the Chemo Day Care Depart-
ment of the Princess Margaret Hospital, Toronto, Canada) and
4.13. Representative procedures for the formation of 1,4-disub-
stituted-[1,2,4]triazolium salts (Scheme 3)
4.13.1. 1,4-Dibenzyl-[1,2,4]triazolium bromide (27)
Under a atmosphere of N2, a mixture of [1,2,4]triazole (500 mg,
14.40 mmol) and potassium carbonate (426 mg, 3.09 mmol) in THF
(10 mL) was stirred at rt for 10 min. Benzyl bromide (5.00 g,
28.80 mmol) was added dropwise and the mixture stirred at reflux
temperature for 48 h. The mixture was cooled to rt, the white pre-
cipitate was removed by filtration, and the filtrate was concen-
trated to a yellow oil. The oil was treated with dichloromethane
and the resulting precipitate was removed by filtration. Recrystal-
lization from 2-propanol gave 27 (530 mg, 35%) as a white solid; 1H
NMR (400 MHz, DMSO-d6): d 5.52 (s, 2H), 5.61 (s, 2H), 7.40–7.51
(m, 10H), 9.33 (s, 1H), 10.32 (s, 1H); 13C NMR (100 MHz, DMSO-
50 lM hypoxanthine (referred to as RPMI-10Pf). The effects of
the test compounds on the viability of P. falciparum cultures were
determined using a Lactate Dehydrogenase (LDH) enzyme assay
specific for the enzyme found in P. falciparum (pLDH).25,26 Briefly,
compounds to be tested were dissolved in DMSO to afford a solu-
tion having a concentration of 10 mg/mL. Twofold serial dilutions
were then produced in 50
lL of RPMI-10Pf in a 96-well plate and
then 50 L of parasite culture (2% hematocrit, 2% parasitemia)
l
were added to each well and the plates were then incubated at
37 °C in 95% N2, 3% CO2, and 2% O2 for 72 h. The contents of
the wells were then re-suspended using a multi-channel pipettor
d6):
d 50.65, 54.89, 128.82, 128.87, 128.91, 128.96, 129.09,
and a 15-
lL sample was removed from each well and was added
133.13, 133.5, 142.74, 145.0; HRMS (ESI) [MꢁBr]+ Calcd for
to 100 L of pLDH enzyme assay mixture. After 1 h the absor-
l
C16H16N3: 250.1344. Found: 250.1341.
bance of the wells at 595 nm was determined using a Thermo-
Max microplate reader (Molecular Devices, Sunnyvale, CA, USA).
The IC50 values of individual compounds were determined using
a non-linear regression analysis of the data27 using the computer
program SigmaPlot (Jandel ScientificÒ, San Rafael, CA, USA). The
IC50 values represent the mean standard error calculated from
four independent determinations. To verify if the poor viability
of the cultures was related to inhibited merozoite invasion,
samples were taken from treated wells and the presence of
extracellular merozoites was confirmed by microscopy. We have
frequently used SYBR-Green 1 assays to determine parasite via-
bility, and the values obtained were virtually identical to those
obtained from the LDH assay. The LDH method was employed
4.13.2. 4-Benzyl-1-(2-oxo-2-phenylethyl)-[1,2,4]triazolium
bromide (28)
A solution of the hydrochloride salt 32 (132 mg, 0.59 mmol) in
water (ꢂ5 mL) was basified using an excess of K2CO3 (ꢂ132 mg,
0.96 mmol). The mixture was extracted with EtOAc (3ꢀ), and the
combined organic extracts were dried (Na2SO4), concentrated,
and dried under high vacuum to afford the free base (112 mg,
0.59 mmol, 100%). To
a solution of the free base (112 mg,
0.59 mmol, 1 equiv) in 1-propanol (2 mL) at rt was added benzyl
bromide (0.1 mL, 0.84 mmol, 1.4 equiv). The mixture was heated
at reflux temperature for 4 h, then cooled to 0 °C. Et2O was added,

6196
J. Z. Vlahakis et al. / Bioorg. Med. Chem. 18 (2010) 6184–6196
because some compounds caused interference with the SYBR-
Green 1 assay.
References and notes
1. Snow, R. W.; Guerra, C. A.; Noor, A. M.; Myint, H. Y.; Hay, S. I. Nature 2005, 434,
214.
2. Cohen, J. M.; Singh, I.; O’Brien, M. E. Malar. J. 2008, 7, 200.
3. Langhorne, J.; Ndungu, F. M.; Sponaas, A.-M.; Marsh, K. Nat. Immunol. 2008, 9,
725.
4. Ancelin, M. L.; Calas, M.; Bonhoure, A.; Herbute, S.; Vial, H. J. Antimicrob. Agents
Chemother. 2003, 47, 2598.
5. Wiesner, J.; Ortmann, R.; Jomaa, H.; Schlitzer, M. Angew. Chem., Int. Ed. 2003, 42,
5274.
4.15. Mammalian strains and culture
CHO cells (ATCCÒ, Manassas, VA, USA) were grown in RPMI-
1640 supplemented with 10% fetal calf serum (Sigma–AldrichÒ
Canada, Ltd., Oakville, ON, Canada), 25 mM HEPES, and gentimicin
(referred to as RPMI-10). Cells were seeded in 96-well plates and
grown to 50% confluency in 100 lL of RPMI-10 per well prior to
the addition of either DMSO alone, or a 10 mg/mL solution of a test
6. Gelb, M. H. Curr. Opin. Chem. Biol. 2007, 11, 440.
7. Woster, P. M. Ann. Rep. Med. Chem. 2003, 38, 203.
8. Olliaro, P. L.; Yuthavong, Y. Pharmacol. Ther. 1999, 81, 91.
9. Kisilevsky, R.; Crandall, I. E.; Szarek, W. A.; Bhat, S.; Tan, C.; Boudreau, L.; Kain,
K. C. Antimicrob. Agents Chemother. 2002, 46, 2619.
10. Crandall, I. E.; Szarek, W. A.; Vlahakis, J. Z.; Xu, Y.; Vohra, R.; Sui, J.; Kisilevsky,
R. Biochem. Pharmacol. 2007, 73, 632.
compound in DMSO. Compound gradients were prepared by add-
ing 90
l
L of RPMI-10 mixed with 10
l
L of compound solution to
L to the next
the first well in the series, mixing, transferring 100
l
well, and repeating until the next-to-last well was reached. After
48 h, the viability of the cells was determined by discarding the
media in the wells and adding 100 lL of 10 mg/mL of 3-[4,5-
dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT,
Sigma–AldrichÒ Canada, Ltd., Oakville, ON, Canada) in RPMI-10,
incubating the plates for a further 30 min, and then removing the
11. Cui, X.; Vlahakis, J. Z.; Crandall, I. E.; Szarek, W. A. Bioorg. Med. Chem. 2008, 16,
1927.
12. Crandall, I. E.; Szarek, W. A. Am. J. Trop. Med. Hyg. 2009, 81. abstract 514.
13. Szarek, W. A.; Crandall, I. E.; Vlahakis, J. Z.; Lazar, C.; Mitu, S. Am. J. Trop. Med.
Hyg. 2009, 81. abstract 516.
14. Wan, Y.; Wallinder, C.; Plouffe, B.; Beaudry, H.; Mahalingam, A. K.; Wu, X.;
Johansson, B.; Holm, M.; Botoros, M.; Karlén, A.; Pettersson, A.; Nyberg, F.;
Fändriks, L.; Gallo-Payet, N.; Hallberg, A.; Alterman, M. J. Med. Chem. 2004, 47,
5995.
15. Brenna, S.; Posset, T.; Furrer, J.; Bluemel, J. Chem. Eur. J. 2006, 12, 2880.
16. Starikova, O. V.; Dolgushin, G. V.; Larina, L. I.; Ushakov, P. E.; Komarova, T. N.;
Lopyrev, V. A. Russ. J. Org. Chem. 2003, 39, 1467.
17. Becker, H. G. O.; Hoffmann, G.; Gwan, K. M.; Knüpfer, L. J. Praktische Chem.
1988, 330, 325.
18. Chu, Y.; Deng, H.; Cheng, J.-P. J. Org. Chem. 2007, 72, 7790.
19. Owen, C. P.; Patel, C. H.; Dhanani, S.; Ahmed, S. Lett. Drug Des. Discov. 2006, 3,
761.
20. Owen, C. P.; Dhanani, S.; Patel, C. H.; Shahid, I.; Ahmed, S. Bioorg. Med. Chem.
Lett. 2006, 16, 4011.
21. Weidinger, H.; Kranz, J. DE Patent 1,076,136, 1958.
22. Lee, L. A.; Evans, R.; Wheeler, J. W. J. Org. Chem. 1972, 37, 343.
23. Gottlieb, H. E.; Kotlyar, V.; Nudelman, A. J. Org. Chem. 1997, 62, 7512.
24. Trager, W.; Jensen, J. B. Science 1976, 193, 673.
media and adding 100 lL of DMSO and reading the absorbance
at 650 nm (see Ref. 28). The IC50 values of individual compounds
were determined using a non-linear regression analysis of the sig-
moidal dose–response curves using the computer program Sigma-
Plot (Jandel ScientificÒ, San Rafael, CA, USA).
Acknowledgments
The authors gratefully acknowledge the generous support of a
CIHR Proof-of-Principle Grant (PPP-84167) and an MRI BioDiscov-
ery Grant (BDT-2012-S3a).
25. Prudhomme, J. G.; Sherman, I. W. J. Immunol. Methods 1999, 229, 169.
26. Makler, M. T.; Ries, J. M.; Williams, J. A.; Bancroft, J. E.; Piper, R. C.; Gibbins, B.
L.; Hinrichs, D. J. Am. J. Trop. Med. Hyg. 1993, 48, 739.
27. Smilkstein, M.; Sriwilaijaroen, N.; Kelly, J. X.; Wilairat, P.; Riscoe, M. Antimicrob.
Agents Chemother. 2004, 48, 1803.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2010.05.020.
28. Campling, B. G.; Pym, J.; Galbraith, P. R.; Cole, S. P. Leuk. Res. 1988, 12, 823.