Synthesis, antimalarial evaluation and molecular modeling studies of hydroxyethylpiperazines, potential aspartyl protease inhibitors, Part 2

Article abstract of DOI:10.1016/j.ejmech.2009.03.041

The antimalarial acitivity of hydroxyethylamines, synthesized from the reaction of intermediated hydroxyethypiperazines with benzenesulfonyl chlorides or benzoyl chlorides, has been evaluated in vitro against a W2 Plasmodium falciparum clone. Some of the

Full text of DOI:10.1016/j.ejmech.2009.03.041

European Journal of Medicinal Chemistry 44 (2009) 3816–3820
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Short communication
Synthesis, antimalarial evaluation and molecular modeling studies of
hydroxyethylpiperazines, potential aspartyl protease inhibitors, Part 2
,
e
,
a
a
a
b
Wilson Cunico a , Claudia R.B. Gomes , Victor Facchinetti , Marcele Moreth , Carmen Penido ,
*
Maria G.M.O. Henriques b, Fernando P. Varotti c f, Luisa G. Krettli c, Antoniana U. Krettli c
,
,
,f
Franklin S. da Silva d, Ernesto R. Caffarena d, Camila S. de Magalha˜es d
a
´
Departamento de Sıntese, Farmanguinhos-Fiocruz, R. Sizenando Nabuco 100, 21041-250 Rio de Janeiro, RJ, Brazil
b Departamento de Farmacologia Aplicada, Farmanguinhos-Fiocruz, RJ, Brazil
c
´
´
´
Centro de Pesquisas Rene Rachou-Fiocruz, Laboratorio de Malaria, Belo Horizonte, MG, Brazil
d Programa de Computaç˜ao Cient´ıfica – PROCC-Fiocruz, RJ, Brazil
e
o
ˆ
´
Departamento de Qu´ımica Organica, UFPel, Campus Universitario s/n , 96010-900 Pelotas, RS, Brazil
f Programa de Farmacologia Bioqu´ımica e Molecular, UFMG, Belo Horizonte, MG, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
The antimalarial acitivity of hydroxyethylamines, synthesized from the reaction of intermediated
hydroxyethypiperazines with benzenesulfonyl chlorides or benzoyl chlorides, has been evaluated in vitro
against a W2 Plasmodium falciparum clone. Some of the nineteen tested derivatives showed a significant
activity in vitro, thus turning into a promising new class of antimalarials. In addition, a molecular
modeling study of the most active derivative (5l) was performed and its most probable binding modes
within plasmepsin II enzyme were identified.
Received 27 August 2008
Received in revised form
27 February 2009
Accepted 30 March 2009
Available online 8 April 2009
Ó 2009 Elsevier Masson SAS. All rights reserved.
Keywords:
Protease inhibitor
Malaria
Hydroxyethylamine
1. Introduction
proteases, known as the plasmepsins, appears to be involved in the
initial steps of the degradation pathway and therefore converts
Malaria accounts for more than a million deaths each year, of
which over 80% occur in tropical Africa, where this illness is the
leading cause of mortality in children under five years of age [1]. In
Brazil it is estimated that 500,000 cases occur annually, mainly in
the Amazon Rain Forest. Malaria is caused by protozoal parasites of
the genus Plasmodium. Four different Plasmodium species infect
humans: Plasmodium vivax, Plasmodium malarie, Plasmodium ovale
and the most dangerous specie Plasmodium falciparum. Despite the
importance of this disease, there has been little economic incen-
tive for the development of new drug-based antimalarial thera-
pies. There is a growing need for effective drugs with new
mechanism of action, due the high rate of mutation of the parasite,
which leads to the development of resistance. One of the critical
stages of the life cycle of the parasite during human infection is the
degradation of hemoglobin which provides nutrients for its
itself in an attractive antimalarial drug target [3]. There are four
aspartic proteases present in the food vacuole of P. falciparum,
plasmepsin I, II, IV and Histo-Aspartic-Protease (HAP). The most
attractive target from this group of enzymes is the plasmepsin II
[4]. Recent studies have shown that HIV protease inhibitors can
inhibit the plasmepsin in vitro [5] and in vivo [6] at pharmaco-
logically relevant concentrations.
A secondary alcohol is usually the structural element of choice to
inhibit aspartic protease. This element mimics the tetrahedral
intermediate during peptide bond cleavage by aspartic proteases
[3]. It has been successfully used to develop potent inhibitors of
these enzymes, for example norstatine [7,8] and hydroxyethyl-
amine-based [9,10] molecules (Fig. 1). In the course of our investi-
gations on hydroxyethylamine derivatives as antimalarial agents
[11,12], we describe in the present work the synthesis and in vitro
activity of novel hydroxyethylpiperazine-based compounds against
W2 clone of P. falciparum. The clone W2 was used for the test, as
described in our previous work [13], and activity compared to that
of chloroquine and lopinavir (the HIV protease inhibitor with the
best antimalarial activity) [5], used in parallel as control in each test.
growth and maturation [2].
A particular family of aspartic
* Corresponding author. Departamento de Sı´ntese, Farmanguinhos-Fiocruz,
R. Sizenando Nabuco 100, 21041-250, Rio de Janeiro, RJ, Brazil
E-mail address: wjcunico@yahoo.com.br (W. Cunico).
0223-5234/$ – see front matter Ó 2009 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2009.03.041

W. Cunico et al. / European Journal of Medicinal Chemistry 44 (2009) 3816–3820
3817
OH
R
HO
OH
OH
H
N
H
N
H
N
N
N
H
R
O
R
O
R
tetrahedral intermediate
of amide hydrolysis
norstatines
hydroxyethylamines
Fig. 1. Molecules developed for the synthesis of aspartyl protease inhibitors.
2. Pharmacological evaluation
presence of different concentrations of the compounds (0.1–
100 g/ml) for 20 h, when MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyl tetrazolium bromide) solution (5 mg/ml; 22.5 l/well)
was added to the culture for 4 h. Supernatant was discharged and
DMSO (150 l/well) was added for solubilization of formazan
m
Antimalarial assay. Parasites were cultured with human eryth-
rocytes (blood group Oþ) at 5% hematocrit in RPMI 1640 supple-
mented with 10% human plasma as previously described [14]. Test
compounds were solubilized in ethanol prior to in vitro tests. The
antiparasitic effects of the molecules was measured by the [3H]-
hypoxanthine incorporation assay [15]. Briefly, trophozoite stages
in sorbitol-synchronized blood, were cultured at 2% parasitaemia
and 2.5% hematocrit, in the presence of the test compounds (at
various concentrations), diluted with culture medium (RPMI 1640)
without hypoxanthine; a chloroquine control (as a reference anti-
malarial drug) was used in each experiment. Inhibition of parasite
growth was evaluated through the levels of [3H]-hypoxanthine
incorporation plotted to generate dose–response curves. The half-
maximal inhibitory response (IC50) compared with parasite growth
in the drug-free controls was estimated by curve fitting using
a software program [Microcal, Origin Software, Inc. (Northampton,
MA, USA)].
m
m
crystals. The absorbance was read at 540 nm.
3. Chemistry
The target compounds 5a–m and 6a–f were prepared as out-
lined in Scheme 1. The hydroxyethylamine transition-state
mimicking fragment was prepared by selective ring-opening of the
(2S,3S)boc-phenylalanine epoxide 1 with 4-(benzyl)piperazines
(2a–g) in refluxing isopropanol for 16 h to give intermediates 3a–g
with good yields (68–96%) [12]. Compounds 3a–g were depro-
tected with trifluoroacetic acid in dichloromethane at room
temperature for 4 h resulting the free amine 4a–g, which were
coupled with benzenesulfonyl chlorides by using potassium
carbonate in acetonitrile at room temperature to afford the
hydroxyethylpiperazines 5a–k in good yields after purification (60–
83%). Compounds 5l and 5m were prepared by a reduction reaction
from compounds 5j and 5e, respectively. The nitro group of
compounds 5e and 5j were reduced with H2 and Pd/C (10%) using
ethanol as the solvent, in good yields. The hydroxyethylpiperazines
6a–d were synthesized from reaction of amine 4g with
Cytotoxicity assay. Cytotoxicity was determined in the murine
monocyte/macrophage cell lineage J774, using the method previ-
ously described [16]. Cells were seeded in a flat bottom 96 well
plate (2 ꢀ 106 cells/well) cultured for 1 h (CO2 5% at 37 ꢁC) in RPMI
1640 enriched with 10% bovine fetal serum, 2 mM L-glutamine,
25 mg/ml gentamicin, pH 7.4. Adherent cells were cultured in the
R2
R2
R1
OH
Ph
N
N
i
ii
BocHN
N
BocHN
+
R1
HN
68-96%
3a-g
Ph
2a-g
1
R
R2
R1
iii
OH
Ph
N
H
N
60-83%
N
S
O
O
5a-k
R2
νi
νi
5e R1 = NO2
5m R1 = NH2
5j R = NO2
N
OH
Ph
65%
75%
5I R = NH2
H2N
N
R1
4a-g
R4
R5
R3
O
O
iν or ν
57-84%
R6
OH
Ph
N
H
N
N
R7
O
6a-f
Scheme 1. Reaction conditions: i: IPA, reflux, 16 h; ii: TFA/CH2Cl2 (1/3), r.t., 4 h; iii: ArSO2Cl, CH3CN, K2CO3, r.t., 4 h; iv: ArCOCl, CH3CN, K2CO3, r.t., 4 h; v: ArCOOH, THF, HOBT, EDC,
DIPEA, 0 ꢁC – r.t., 18 h; vi: H2, Pd/C 10%, EtOH, r.t., 16 h.

3818
W. Cunico et al. / European Journal of Medicinal Chemistry 44 (2009) 3816–3820
Table 1
Antimalarial activity, cytotoxicity activy and lipophilicity (Log P) of hydroxyethylpiperazines 5a–m and 6a–f.
Comp.
R
R1
R2
R3
R4
R5
R6
R7
IC50
(
m
M)a
Cytotoxicity (
m
g/ml)b
miLogPc
% of cell viability
W2
3D7
0.1
1
10
88
100
5a
5b
5c
5d
5e
5f
5g
5h
5i
5j
5k
5l
5m
6a
6b
6c
Me
Me
Me
Me
Me
Me
Me
Br
OMe
NO2
F
NH2
Me
–
H
F
Cl
OMe
NO2
OMe
H
H
H
H
H
OMe
–
–
–
–
–
–
–
–
–
–
–
–
–
H
F
Cl
H
H
Me
–
–
–
–
–
–
–
–
–
–
–
–
–
H
H
H
Cl
–
–
–
–
–
–
–
–
–
–
–
–
–
H
H
H
H
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
H
F
>90
43.0
35.1
>90
>90
>90
–
–
–
–
–
–
–
5.7
–
–
–
4.6
–
–
–
–
–
–
100
100
100
100
91
100
100
100
100
100
100
100
100
100
85
86
100
98
5
15
22
6
92
9
49
5
12
40
4
55
10
38
51
43
93
61
8
3.921
4.084
4.599
3.978
3.880
3.567
3.811
4.172
3.419
3.321
3.526
2.438
2.997
3.487
3.719
4.747
4.771
3.378
3.828
92
90
84
93
92
98
96
90
100
83
99
77
91
83
84
98
95
8
90
100
100
100
100
99
100
94
100
90
100
87
90
100
100
75
–OCH2O–
16.9
–OCH2O–
–OCH2O–
–OCH2O–
–OCH2O–
–OCH2O–
5.6
34.3
44.0
42.5
4.8
NH2
H
–
>90
–OCH2O–
–OCH2O–
–OCH2O–
–OCH2O–
–OCH2O–
–OCH2O–
H
H
H
Cl
H
H
69.4
47.4
55.5
47.2
25.4
5.9
–
–
–
–
Cl
H
H
H
100
98
100
100
6d
6e
6f
–OCH2O–
–
OH
H
–
Chloroquine
Lopinavir
0.34
2.7
0.03
–
a
IC50 represents concentration inhibitory dose of the parasite growth in relation to control cultures with no drugs.
Percentage of J774 cell viability 24 h after incubation with the compounds determined by the MTT method.
Calculated using online www.molinspiration.com/cgi-bin/properties site (Molinspiration Property Engine v2007.04).
b
c
corresponding benzoyl chlorides in good yields (73–84%). Piper-
indicated the best antimalarial activity of this work (IC50 ¼ 4.8
almost as active as standard lopinavir. The compound 5h (R ¼ Br)
also showed good activity (IC50 ¼ 5.6 M). The compounds 5h and
5l were also tested against P. falciparum 3D7 clone (chloroquine
M and 4.6 M, respectively. The third
mM),
onylic acid and 3-hydroxy-2-methylbenzoic acid were used as
precursors for the synthesis of compounds 6e and 6f, respectively.
These reactions were carried out with HOBT, EDC and DIPEA using
THF as the solvent in moderate yields. Both analytical and spectral
data (1H and 13C NMR) of all compounds are in full agreement with
the proposed structures (see Supplementary information). 2D-
NMR thecniques (HMBQ, HMQC and COSY) helped us to assign the
correct signals of compounds.
m
sensitive) with IC50 ¼ 5.7
m
m
series was obtained from reaction of compound 4g with benzoyl
chlorides. Only compound 6f showed good activity against P. fal-
ciparum (IC50 ¼ 5.9
macrophages cells at low concentration (cell reduction of 25% at
g/ml and cell reduction of 92% at 10 g/ml concentration). The
mM), however, this compound was cytotoxic at
1
m
m
lipophilicities of the synthesized compounds 5a–m and 6a–f,
which were expressed as log P values, were determined through
online www.molinspiration.com/cgi-bin/properties site.
4. Results and discussion
Table 1 shows the antimalarial activities of all compounds
expressed as 50% inhibitory concentration (IC50) of P. falciparum. In
the first series, the p-toluenesulfonyl group was fixed and different
benzylpiperazines were attached to compounds 5a–g and 5m. In
this preliminary study, we observed that compound 5g, with 4-
(piperonyl)piperazine moiety, presented the best antimalarial
activity (IC50 ¼ 16.9
mM). The second series was synthesized fixing
the 4-(piperonyl)piperazine moiety and switched the p-toluene-
sulfonyl group for different benzenesulfonyl derivatives to give
compounds 5h–l. The hydroxyethylpiperazine 5l (R ¼ NH2)
O
O
OMe
O
Ph
O
OH
Ph
H
N
N
N
MeO
O
O
O
Fig. 3. Hydrogen bonds of Plm II with compound 5l (conf. 5 in yellow) superposed
over the re-docked structure of EH58 ligand (in pink). (For interpretation of the
references to color in this figure legend, the reader is referred to the web version of this
article.)
EH58
Fig. 2. Structure of EH58 ligand.

W. Cunico et al. / European Journal of Medicinal Chemistry 44 (2009) 3816–3820
3819
Table 2
Results of the linear interaction energy (LIE) method.
c
c
c
c
Compounda
IC50
(
m
M)b
D
Gbind;calc
CV1vdꢂws DP
CV DP
1ꢂs
CV1vdꢂws DW
CV DW
1ꢂs
el
el
5l – conf. 4
5l – conf. 5
4.8
4.8
0.44 (2.62)
ꢂ1.61 (1.23)
ꢂ58.70 (1.67)
ꢂ80.00 (4.03)
ꢂ37.28 (0.88)
ꢂ91.66 (3.34)
ꢂ65.55 (1.15)
ꢂ80.78 (2.20)
ꢂ35.83 (0.86)
ꢂ91.04 (2.42)
a
Conf. 4 and conf. 5 indicate the two selected docking conformations of compound 5l.
b
c
IC50 represents concentration inhibitory dose of the parasite growth in relation to control cultures with no drugs.
The calculated average electrostatic and non-polar energies for ligand-surrounding interactions are in kcal/mol (standard deviation).
Fig. 4. Superposition of conf. 4(a) and conf. 5(b) (in blue and light blue, respectively) of compound 5l over the crystallographic structure of the EH58 ligand (in red). The generation
of the figure above was done with the software Pymol v0.99rc6, DeLano, W. L., 2002. (For interpretation of the references to color in this figure legend, the reader is referred to the
web version of this article.)
5. Molecular modeling
These conformations were used as starting structures for
molecular dynamics simulations and binding free energy evaluation
with the LIE method [19]. The calculated free energy of binding for
the two selected conformations of compound 5l is shown in Table 2.
The results revealed conf. 5 as the most favored binding mode.
When compared to conf. 5, conf. 4 presented similar energies for
ligand-surrounding interactions. However, the main difference was
in the non-polar protein–ligand interactions. This is probably due to
the absence of the ligand groups in subsites S2 and S20 in conf. 4.
The most favored conformation of compound 5l (conf. 5) fills the
Plm II subsites in a quite similar way to the experimental structure
of EH58 ligand (Fig. 4). This conformation also presents a better fit of
the non-polar groups of compound 5l in the enzyme hydrophobic
subsites contributing to a better overall affinity.
Molecular docking studies were also done for compound 5j, the
one with the lowest observed activity. The best automatically
docked conformation of 5j differ from the 5l one, mainly regarding
the position of the benzodioxolane and the benzylic rings at S20 and
S10 subsites, respectively. However, the benzylic ring with the
substituent group in 5j is located at S2 pocked in a similar position
to the one observed in 5l conformation. The most active compound
5l is hydrogen bonded to the two catalytic aspartates of Plm II
enzyme and to SER218, GLY216 and TYR192. The docked confor-
mation of compound 5j presents two hydrogen bonds with SER218
and a hydrogen bond to only one of the catalytic aspartates. None
specific interaction of the substituent groups that might explain
activity was observed.
The potential binding modes and the main interactions between
the most active synthesized compound 5l and Plm II enzyme were
investigated by computational analysis. Firstly, to validate the
computational docking protocol, redocking analysis of the EH58
ligand into the Plm II active site was carried out with the Autodock
program [17]. Since the EH58 ligand is a large and highly flexible
compound (19 rotatable bonds), conformations with RMSD values
up to 3.0 Å were considered correctly docked structures (Fig. 2).
Taken into account this criterion, the redocking of EH58 into plas-
mepsin II active site presented a success rate of 60% in finding
structures near the crystallographic one. The correctness of docked
conformations of EH58 was also confirmed by visual inspection.
Encouraged by the good agreement of docked structures with the
crystallographic structure of EH58, the same docking protocol was
used for automated docking of compound 5l.
The visual inspection along with the investigation of the main
interactions between the ligand and key residues of the enzyme is
a helpful strategy in selecting the most favored ligand binding
modes [18]. This fact prompted us to submit the 10 best confor-
mations of compound 5l to this kind of investigation. Visual
inspection of the docked conformations interacting with key
enzyme residues showed conformations 4 and 5 as the more
probable binding modes of compound 5l and their superposition
upon the EH58 structure reveals equivalent interactions together
with a good agreement in covering the enzyme subsites. The
hydrogen bonds of the selected binding modes of compound 5l are
shown in Fig. 3.
6. Conclusion
Superposition of conformations 4 and 5 of compound 5l sug-
gested distinct ways in the placement of ligand groups in the Plm II
subsites. In conf. 4 the piperazine ring was situated near the S20
pocket. The 4-aminosulfonamide group of this conformation was
set near the S2 pocket and the phenyl ring is located in the vicinity
of the S10 pocket. Positions of all chemical groups of compound 5l in
conf. 5 closely coincide with those of conf. 4, excepting the
piperonyl group that remained parallel to the phtalimide ring of
EH58, resulting in a much better free energy of binding within Plm
II active site (Fig. 4).
In summary, we easily synthesized nineteen hydrox-
yethylpiperazines derivatives in good yields. Despite the fact that
the most active compounds are at least 10- to 20 times less active in
vitro than chloroquine, their IC50 values are in the micromolar
concentration range comparable to recently reports results [13,20].
In addition, compounds 5h, 5l and 6f are as active as lopinavir, the
protease inhibitor of HIV with the highest activity against
P. falciparum. Compounds 5h and 5l also have low toxic activities at
high concentration (100 mg/ml). The conf. 5 was the most favored

3820
W. Cunico et al. / European Journal of Medicinal Chemistry 44 (2009) 3816–3820
[5] S. Parikh, J. Gut, E. Istvan, D.E. Goldberg, D.V. Havlir, P.J. Rosenthal, Antimicrob.
Agents Chemother. 49 (2005) 2983.
conformation of hydroxyethylapiperazine 5l, filling the Plm II
subsites and showing a better fit of the non-polar groups in the
enzyme hydrophobic subsites. The investigation of these
compounds as anti-HIV agents is ongoing.
[6] K.T. Andrews, D.P. Fairlie, P.K. Madala, J. Ray, D.M. Wyatt, P.M. Hilton,
L.A. Melville, L. Beattie, D.L. Gardioner, R.C. Reid, M.J. Stoermer, T. Skinner-
Adams, C. Berry, J.S. McCarthy, Antimicrob. Agents Chemother. 50 (2006) 639.
[7] K. Hidaka, T. Kimura, Y. Tsuchiya, M. Kamiya, A.J. Ruben, E. Freire, Y. Hayashi,
Y. Kiso, Bioorg. Med. Chem. Lett. 17 (2007) 3048.
Acknowledgements
[8] H.M. Abdel-Rahman, T. Kimura, K. Hidaka, A. Kiso, A. Nezami, E. Freire,
Y. Hayashi, Y. Kiso, Biol. Chem. 385 (2004) 1035.
[9] D. Noteberg, E. Hamelink, J. Hulte´n, M. Wahlgren, L. Vrang, B. Samuelsson,
A. Hallberg, J. Med. Chem. 46 (2003) 734.
The authors thank Farmanguinhos and PROCC for the financial
support of the research. We also thank FAPERJ (V.F.; F.S.C.; C.S.M.)
and CNPq (M.M).
[10] D. Muthas, D. Noteberg, Y.A. Sabnis, E. Hamelink, L. Vrang, B. Samuelsson,
A. Karle´n, A. Hallberg, Bioorg. Med. Chem. 13 (2005) 5371.
[11] W. Cunico, M.L.G. Ferreira, T.G. Ferreira, C. Penido, M.G.M.O. Henriques,
L.G. Krettli, F.P. Varotti, A.U. Krettli, Lett. Drug Des. Discov. 5 (2008) 178.
[12] W. Cunico, C.R.B. Gomes, M. Moreth, D.P. Manhanini, I.H. Figueiredo, C. Penido,
M.G.M.O. Henriques, F.P. Varotti, A.U. Krettli, Eur. J. Med. Chem. 44 (2009)
1363.
[13] W. Cunico, C.A. Cechinel, H.G. Bonacorso, M.A.P. Martins, N. Zanatta, M.V.N. de
Souza, I.O. Freitas, R.P.P. Soares, A.U. Krettli, Bioorg. Med. Chem. Lett. 16 (2006)
649.
Appendix. Supplementary information
There are supplementary informations with chemical general
procedure, yields, NMR 1H and 13C data and LC-MS data. We also
include information about molecular modeling study. Supple-
mentary data associated with this article can be found in the online
version, at doi:10.1016/j.ejmech.2009.03.041.
[14] V.F. de Andrade-Neto, M.O.F. Goulart, J.F. da Silva Filho, M.J. da Silva,
M.C.F.R. Pinto, A.V. Pinto, M.G. Zalis, L.H. Carvalho, A.U. Krettli, Bioorg. Med.
Chem. Lett. 14 (2004) 1145.
[15] R.E. Desjardins, C.J. Canfield, J.D. Haynes, J.D. Chulay, Antimicrob. Agents
Chemother. 16 (1979) 710.
References
[16] M. Ferrari, M.C. Fornasiero, A.M. Isetta, J. Immunol. Methods 131 (1990) 165.
[17] G.M. Morris, D.S. Goodsell, R.S. Halliday, R. Huey, W.E. Hart, R.K. Belew,
A.J. Olson, J. Comput. Chem. 19 (1998) 1639.
[1] W. Cunico, S.A. Carvalho, C.R.B. Gomes, G.H. Marques, Rev. Bras. Farm. 89
(2008) 49.
[18] V. Kasam, M. Zimmermann, A. Maa, H. Schwichtenberg, A. Wolf, N. Jacq,
V. Breton, M. Hofmann-Apitius, J. Chem. Inf. Model. 47 (2007) 1818.
[19] J. Åqvist, C. Medina, J.E. Samuelsson, Prot. Eng. 7 (1994) 385.
[20] V.R. Solomon, S.K. Puri, K. Srivastava, S.B. Katti, Bioorg. Med. Chem. 13 (2005)
2157.
[2] G.H. Coombs, D.E. Goldberg, M. Klemba, C. Berry, J. Kay, J.C. Mottram, Trends
Parasitol. 17 (2001) 532.
[3] K. Ersmark, B. Samuelsson, A. Hallberg, Med. Res. Rev. 26 (2006) 626.
[4] C. Boss, S. Richard-Bildstein, T. Weller, W. Fischli, S. Meyer, C. Binkert, Curr.
Med. Chem. 10 (2003) 883.