Antimalarial versus cytotoxic properties of dual drugs derived from 4-aminoquinolines and mannich bases: Interaction with DNA

Article abstract of DOI:10.1021/jm9018383

The synthesis and biological evaluation of new organic and organometallic dual drugs designed as potential antimalarial agents are reported. A series of 4-aminoquinoline-based Mannich bases with variations in the aliphatic amino side chain were prepared via a three-steps synthesis. These compounds were also tested against chloroquine-susceptible and chloroquine-resistant strains of Plasmodium falciparum and assayed for their ability to inhibit the formation of β-hematin in vitro using a colorimetric β-hematin inhibition assay. Several compounds showed a marked antimalarial activity, with IC₅₀ and IC₉₀ values in the low nM range but also a high cytotoxicity against mammalian cells, in particular a highly drug-resistant glioblastoma cell line. The newly designed compounds revealed high DNA binding properties, especially for the GC-rich domains. Altogether, these dual drugs seem to be more appropriate to be developed as antiproliferative agents against mammalian cancer cells than Plasmodium parasites.

Full text of DOI:10.1021/jm9018383

3214 J. Med. Chem. 2010, 53, 3214–3226
DOI: 10.1021/jm9018383
Antimalarial versus Cytotoxic Properties of Dual Drugs Derived From 4-Aminoquinolines and Mannich
Bases: Interaction with DNA
Nicole I. Wenzel,†,þ Natascha Chavain,‡,þ Yulin Wang,§ Wolfgang Friebolin,† Louis Maes, Bruno Pradines,^
§
#
O
3
‡
Michael Lanzer, Vanessa Yardley, Reto Brun, Christel Herold-Mende, Christophe Biot, Katalin Toth,*,[ and
ꢀ
Elisabeth Davioud-Charvet*,†,z
†
‡
Biochemie-Zentrum der Universitat Heidelberg, Im Neuenheimer Feld 504, D-69120 Heidelberg, Germany, Universite de Lille 1, Unite de
€
Catalyse et de Chimie du Solide-UMR CNRS 8181, ENSCL, Batiment C7, USTL, B.P. 90108, F-59652 Villeneuve d’Ascq Cedex, France,
ꢀ
ꢀ
^
§University of Heidelberg, Department of Infectiology, D-69120 Heidelberg, Germany, University of Antwerp, Groenenborgerlaan 171, B-2020
Antwerpen, Belgium, ^Institut de Recherche Biomedicale des Armees antenne de Marseille, Unite de Recherche en Biologie et Epidemiologie
ꢀ
ꢀ
ꢀ
ꢀ
#
ꢀ
ꢀ
Parasitaires, UMR 6236, Allee du Medecin-colonel Jamot, Le Pharo, BP 60109, F-13262 Marseille Cedex, France, Department of Infectious and
Tropical Diseases, London School of Hygiene & Tropical Medicine, Keppel Street, London WC1E 7HT, U.K., 3Department of Neurosurgery,
University of Heidelberg, D-69120 Heidelberg, Germany, OSwiss Tropical and Public Health Institute, Medical Parasitology and Infection
Biology, CH-4002 Basel, Switzerland and University of Basel, Basel, Switzerland, [Division Biophysics of Macromolecules, German Cancer
Research Center, Im Neuenheimer Feld 580, D-69120 Heidelberg, Germany, and zEuropean School of Chemistry, Polymers and Materials
(ECPM), Universiteꢀ de Strasbourg, 25 rue Becquerel, F-67087 Strasbourg Cedex 2, France. þ These authors contributed equally to this work.
Received December 13, 2009
The synthesis and biological evaluation of new organic and organometallic dual drugs designed as
potential antimalarial agents are reported. A series of 4-aminoquinoline-based Mannich bases with
variations in the aliphatic amino side chain were prepared via a three-steps synthesis. These compounds
were also tested against chloroquine-susceptible and chloroquine-resistant strains of Plasmodium
falciparum and assayed for their ability to inhibit the formation of β-hematin in vitro using a
colorimetric β-hematin inhibition assay. Several compounds showed a marked antimalarial activity,
with IC50 and IC90 values in the low nM range but also a high cytotoxicity against mammalian cells, in
particular a highly drug-resistant glioblastoma cell line. The newly designed compounds revealed high
DNA binding properties, especially for the GC-rich domains. Altogether, these dual drugs seem to be
more appropriate to be developed as antiproliferative agents against mammalian cancer cells than
Plasmodium parasites.
Introduction
Plasmodium falciparum. However, the development and clin-
ical use of antimalarial 4-aminoquinoline agents such as AQ is
limited in long-term treatments due to the cause of adverse
side effects including hepatotoxicity and agranulocytosis
when the drug is used for prophylaxis.1 The mode of action
of the 4-aminoquinoline antimalarial agents, which involves
the inhibition of heme biomineralization,2 is partly dependent
on drug accumulation due to the lysosomotropism of these
weak bases and a pH gradient between the cytosol and the
acidic food vacuole of the parasite where host hemoglobin
digestion occurs.3
Inside the red blood cells of the host, the malaria parasite
degrades the hemoglobin for the acquisition of amino acids, i.e.,
to construct its own proteins, essential for energy metabolism
and for parasitic growth and division. Digestion is carried out
in an acidic compartment (pH ca. 5), the digestive vacuole, in
the parasite. During this process, the parasite produces the
toxic and soluble molecule heme and biocrystallizes it within
or at the surface of lipids4,5 to form hemozoin as the major
detoxification product. CQ binds to heme (or ferriprotopor-
phyrin IX(FeIII), FPIX) to form a FPIX-CQ complex which
ishighly toxic tothe parasiteanddisruptsmembranefunction.
Action of the toxic FPIX-CQ and FPIX ultimately results in
parasite cell autodigestion. In essence, the parasite drowns in
its own metabolic products.6 Unfortunately the design and
Malaria remains one of the most prevalent diseases in the
developing world. The most severe malaria episodes including
cerebral malaria are due to Plasmodium falciparum. The rapid
spread of multidrug resistance, especially toward chloroquine
(CQa), the main drug in use for many years, is a cause for
major concern. The antimalarial agent amodiaquine (AQ) is
effective against both CQ-susceptible and -resistant strains of
*To whom correspondence should be addressed For E.D-C.: phone,
þ49-6221-54-4188; fax, þ49-6221-54-5586; E-mail, elisabeth.davioud@
gmx.de. For K.T.: phone, þ49-6221-42-3390; fax, þ49-6221-42-3391;
E-mail, kt@dkfz.de.
a Abbreviations: AQ, amodiaquine; AT-rich DNA, DNA with high
adenine (A)- and thymine (T)- base pair content; BCNU, 1,3-bis(2-
chloroethyl)-1-nitroso urea; CQ, chloroquine; ΔTm, difference between
the melting temperature of the DNA treated with the drug and the
temperature of the free DNA; FPIX, ferriprotoporphyrin IX; FQ,
Ferroquine; GC-rich DNA, DNA with high guanine (G)- and cytosine
(C)- base pair content; GSH, glutathione; HPLC, high pressure liquid
chromatography; MDAQ, monodesethylamodiaquine; polydAdT, syn-
thetic poly(dA-dT)-poly(dA-dT) double strand oligonucleotide; QN,
quinine; tR, retention time; RSH, thiol; Tm, the thermal dissociation
temperature of DNA in the absence of any drug; Tm CQ, melting
temperatures of DNA in the presence of CQ; Tm drug, melting tempera-
tures of DNA in the presence of the drug; Tm1/2, the middle of the peak
at its half height; Tmint, the middle of the integrated area below the
derivative melting peak; Tmmax, the maxima of the derivative melting
curves; TE buffer, 10 mM Tris-HCl 0.1 mM EDTA, at pH 7.4.
r
pubs.acs.org/jmc
Published on Web 03/23/2010
2010 American Chemical Society

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8 3215
Scheme 1. Putative Bioactivation of Tertiary Amines by Oxidative N-Dealkylation Reactions (Pathway A) and Michael Addition
Following Deamination of the Released Mannich Base (Pathway B)
Chart 1. Structures of CQ, FQ, and Their Designed Related Analogues 1-5
synthesis of new antimalarials are hindered by the fact that
the mechanism of resistance is not fully understood and
involves more than one altered function, e.g. related to drug
transport,7-10 redox equilibrium,11 and metabolic pathways.12
As previously mentioned, the food vacuole of the malarial
parasites is the locus of hemoglobin digestion leading to the
release of an important amount of free heme. In the presence
of reactive oxygen species, the released heme is thought to
catalyze numerous oxidation reactions,13 in particular with
tertiary amine drugs (Scheme 1, pathway A). Recent advances
aimed at understanding the drug catabolism were applied by
using various iron porphyrins as models of cytP450 enzymes to
mimic the oxidative conditions found in the target cells.14,15
Oxidative N-dealkylation of drugs is known to contribute
both to biological activity and toxic effects of drugs.16,17
Hence, drug bioactivation should be considered as a pro-drug
strategy. On the basis of this hypothesis and on the knowledge
of cytP450 catalyzed oxidative N-dealkylation reactions,18 we
aimed at the design of new 4-aminoquinolines with an amino
side-chain substituted by the aliphatic Mannich base accord-
ing to a reductive amination. The dual drugs 2, 3, and 5
(Chart 1) are expected to enter the food vacuole of the parasite
because of the recognized 4-aminoquinoline motif. The com-
pounds 2, 3, and 5 were designed with the expectation that
oxidative cleavage of such compounds in the food vacuole
might generate two active compounds, a short CQ-analogue
(active per se), and a thiol-reactive saturated Mannich base
(Scheme 1, pathway B).
Chart 2. Structures of the Mannich Bases 6 and 7 Used in the
Study
by varying the pKa value of amino group of the saturated
Mannich base. The Michael acceptor properties of two satu-
rated Mannich bases 6 and 7 (Chart 2) possessing a dimethyl-
amine and a piperidine as amino leaving group were evaluated
by the formation of the monoglutathionylated Mannich base
adduct upon addition of one equivalent of glutathione. By
increasing the stability of the drug, a higher amount of the
nonoxidized compound (pro-drug) is expected to reach the
parasite before oxidation in the host occurs.
Before evidencing the inhibition of β-hematin formation
previous studies aimed at the understanding of the mode of
action of CQ and other 4-aminoquinoline-based compounds
demonstrated interactions of CQ with DNA in vitro.19,20 But,
as CQ is an inhibitor of lysosome maturation with high
tropism for acidic compartments, relevance of correlation
was debated.21 The DNA:drug interaction has been studied
by using different methods like spectrophotometry,22 equilib-
rium dialysis23 and viscosimetry,24 fluorometry,25 and DNA
melting.26,27 CQ, for example, strongly elevated the thermal
dissociation temperature, Tm, of DNA. It was concluded that
the drug forms a complex with DNA by ionic interaction
The modulation of the thiol attack, after the oxidative
N-dealkylation, could be reached in this series of compounds

3216 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8
Wenzel et al.
Scheme 2. Synthesis of the 4-Aminoquinoline Derivatives 2-3, and 5
a Conditions: (i) NaOH, NaBH4; (ii) SOCl2, CH3Cl, reflux; (iii) N-(7-Chloro-quinolin-4-yl)-ethane-1,2-diamine or N-[2-(aminomethyl)ferrocenyl]-7-
chloroquinolin-4-amine, NEt3, EtOH.
between the protonated ring system of CQ and the anionic
phosphate groups of DNA and stabilizes the helical configu-
ration against thermal denaturation26,27 and a more specific
interaction involving the aromatic cycles of CQ and the
nucleotide bases. Later, it has been well documented that
the buffering activity of CQ could improve transfection
efficiency by facilitating DNA release from the endocytic
pathway (as it may generate swelling and destabilization of
endosomes) and/or by inhibiting lysosomal enzyme degrada-
tion. More recently, plasmid DNA, when bound by CQ, was
shown to achieve improved gene transfer due to changes in its
intracellular processing or its rate of intracellular degrada-
tion.28 These data prompted us to evaluate the DNA binding
capabilities of our new 4-aminoquinoline derivatives. Pollack
et al. reported that the GC content of the DNA of P.
falciparum is extremely low (17-19%), and the discrepancy
between this result and previously reported values is due to the
difference in purity of the free parasite preparations used for
isolation of DNA.29 However, CQ binds predominately to
studies involving our drugs with DNA preparations of two
base ratios were further investigated to allow the comparison
of the drug binding affinities specifically to GC- vs AT-rich
regions and their cytotoxicity against mammalian cells.
Taking into account the cytotoxicity of these dual drugs
against mammalian cancer cells versus Plasmodium parasites
implications for drug design will be discussed.
Results and Discussion
Chemistry. The synthetic route of the dual drugs described
in Scheme 2 involves multiple steps starting from the satu-
rated Mannich bases 6-7 obtained after a Mannich reaction
with acetophenone, paraformaldehyde, and different amino
hydrochlorides. Reduction of the keto group to the alcohols
8,10 with sodium borohydrate under basic conditions and
the nucleophilic substitution of the hydroxyl group for a
chloride upon thionyl chloride treatment led to compounds
9,10. The second nucleophilic substitution of the chloride
of 9,10 by N-(7-chloro-quinolin-4-yl)-ethane-1,2-diamine
yielded to the favored 4-aminoquinoline-based compounds
2-3. Compound 5 was prepared by nucleophilic substitution
of 10 by N-[2-(aminomethyl)ferrocenyl)-7-chloroquinolin-
4-amine. The nucleophilic substitution of 9 by N-[2-(amino-
methyl)ferrocenyl)-7-chloroquinolin-4-amine failed whatever
the conditions. The most difficult issue in the preparation of
these compounds is the purification by flash chromato-
graphy. Only a small amount of product was obtained after
two purification steps. Noteworthy is that the direct forma-
tion of the reductive amination products with the com-
pounds 6-7 and N-(7-chloro-quinolin-4-yl)-ethane-1,2-dia-
mine or N-[2-(aminomethyl)ferrocenyl)-7-chloroquinolin-
4-amine under various conditions, e.g., TiCl4/NaBH4 or
NaBH(OAc)3,33 and also the direct alkylation of the imine
by β-aminoalkyllithium34 failed, probably because of the
deamination of the Mannich bases under these various
conditions.
poly(dG-dC) poly(dG-dC) than to other synthetic polynu-
3
cleotides, and a conclusive link between DNA intercalation
and the antimalarialactivity ofCQhasnot been madeso far.30
At physiological and therapeutic conditions, a small number
of binding sites could be occupied by CQ and the binding to
these few sites could be enough to kill the parasite.31 The
unusually high AT content of the malaria parasite genome
offersexceptional, yet unexploited, therapeutic opportunity.32
In this paper, we showed that the newly synthesized 4-
aminoquinoline 1-5 derivatives interact with hematin in a
fashion similar to CQ and FQ with respect to prevention of
β-hematin formation in vitro. Furthermore, they displayed
high antimalarial activity with IC50 and IC90 values in the low
nM range but also a high cytotoxic potential against mamma-
lian cells. On the basis of the knowledge that CQ and other 4-
aminoquinoline-based compounds interact with DNA in
vitro,19 these data prompted us to evaluate the DNA binding
capabilities of our new 4-aminoquinoline derivatives. Com-
plex formation between DNA and drugs was first examined
spectrophotometrically in order togain information aboutthe
correlation between DNA binding and their cytotoxic proper-
ties. The possibility of the formation of complexes between
certain drugs and DNA and thereby the ability to affect the
biological properties and functions of DNA has been rather
reliable demonstrated by other investigators.
Glutathionylation of Mannich Base Derivatives. The
Michael acceptor properties of the compounds 6 and 7
possessing one site for thiol alkylation were evaluated by
the rate of formation of the monoglutathionylated Mannich
base adducts upon addition of one equivalent of glutathione.
The results were expressed as the time-dependent Mannich
base:monoSG-adduct ratio given by HPLC retention times
(HPLC tR) and were shown in Figure S1 from the Supporting
Information. The proposed mode of action toward thiols
(RSH), like glutathione (GSH) by the saturated Mannich
Due to the fact that the human DNA exhibit a much higher
GC content (43%) compared to P. falciparum, complexation

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8 3217
Table 1. In Vitro Inhibition of β-Hematin Formation by 4-Aminoqui-
nolines AQ, CQ, FQ, and Compounds 1-5 in MeOH/1N HCl (1:1)
max inhibition reached (%)
(at drug:hematin ratio)
compd
IC50
1
78 (3:1)
68 (3:1)
68 (3:1)
2.2
2.2
2.0
0.2
0.4
0.6
1.5
0.3
2
3
4a
5
90 (0.75:1)
63 (0.5:1)
100 (1.5:1)
73 (2:1)
AQ
CQ
FQ
72 (0.75:1)
a Data from ref 13.
bases 6 and 7 is shown in Scheme 1, pathway B. For the
compounds 6 (tR = 12.15 min) and 7 (tR = 12.94 min), both
generating the same glutathionyl-conjugate (tR monoSG-adduct =
13.11 min) after deamination, the Mannich base:monoSG-
adduct ratio reached the value of 1 after 7.5 and 13 h,
respectively. The replacement of the dimethylamine by a
piperidine as leaving group of the saturated Mannich base
derivatives gave a significant loss of reactivity. While the
reactivity of Mannich bases to thiols was initially the basis to
design antimalarial dual drugs described in this paper, we did
not study further metabolism of the dual drugs under the
biomimetic oxidative conditions expressed in the malarial
parasites because of the very high toxicity of the compounds.
Inhibition Assay of β-Hematin Formation. Seven represen-
tative compounds, including CQ, the short CQ-analogue 1,
two aminoquinoline-based Mannich base derivatives 2 and
3, and three ferrocenyl compounds, FQ, the FQ analogue 4,
and the ferrocenic dual drug 5, were assayed for their ability
to inhibit the formation of β-hematin in vitro using the
colorimetric β-hematin inhibition screening assay with slight
modifications described in the Experimental Section.35 The
results of the experiment were shown in Table 1, taking into
account both the IC50 values and the drug:hematin ratio
giving the percentage of maximal inhibition reached by the
drug (reproducible data from two experiments). AQ was
used as reference inhibitor of β-hematin formation because,
with an IC50 value of 0.6 and 100% max inhibition at a drug:
hematin ratio of 1.5:1, it presented the highest inhibitory
capacity in the β-hematin formation assay, compared to CQ
(IC50 of 1.5 and 73% max inhibition at a drug:hematin ratio
of 2:1) and FQ (IC50 of 0.3 and 72% max inhibition at a drug:
hematin ratio of 0.75:1) (Figure 1A). Under the conditions
described above, the short CQ-analogue 1 and the Mannich
base-derived 4-aminoquinoline dual drug 2 showed a similar
inhibition behavior with an IC50 value of 2.2 but different
maximal inhibition of 78% and 68%, respectively, at a drug:
hematin ratio of 3:1 (Figure 1B). The nonferrocenyl dual
drug analogue 3 and the ferrocenyl dual drug 5 displayed
IC50 values of 2.0 and 0.4 with a maximal inhibition of 68%
and 63% with a drug:hematin ratio of 3:1 and 0.5:1, respec-
tively (Figure 1C). It is noteworthy that the inhibition
behavior at high concentration of 5 is different compared
to 3, in a similar way between FQ and CQ (or AQ) observed
in this study and previously discussed.13 As observed for FQ
itself (Figure 1A), an unusual bell-shaped curve giving the
absorption at 405 nm versus drug:hematin ratio was also
obtained for the FQ-analogue 5, suggesting that apparent
decreased inhibition of β-hematin formation at apparent
high drug:hematin ratio might be due to an effective de-
creased drug:hematin ratio in the assay. Previous reports
Figure 1. Inhibition of β-hematin formation. Absorbance measure-
ments at 405 nm, performed in the β-hematin formation assay, in the
presence of CQ, FQ, and compounds 1-5 with AQ as reference,
were plotted as a function of the inhibitor:hematin ratio. The results
are given in Table 1 as the IC50 values determined by the number
of drug equivalents required to inhibit the β-hematin formation
by 50%.

3218 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8
Wenzel et al.
Table 2. Antimalarial Activities of the Short Analogues 1 and 4, and the
Dual Drugs 2, 3, and 5, with Respect to CQ and FQ as References
ferrocenic drugs 4 and 5 than for the nonferrocenic drugs 1,
2, and 3. Using 12 strains of P. falciparum, the antimalarial
activity of the compounds was represented as a function of
the decreased susceptibility of the P. falciparum strain to CQ,
QN, and MDAQ, expressed as IC50 (Figure 2A) and IC90
values (Figure 2B). We observed that the ferrocenic com-
pounds like FQ maintained a constant antimalarial activity
while the pure organic compounds, although remaining
sharply more active than CQ, exhibited a slightly decreased
activity on CQ- and MDAQ-resistant strains. These data
suggested that a crossed resistance might be quickly at risk
and developed for the organic series with respect to the
established 4-aminoquinolines CQ, QN, and MDAQ. As
showed previously, the presence of a ferrocenyl group seems
to be a crucial character for a potent antimalarial activity on
CQ-resistant strains.
The new 4-aminoquinolines showed a high antimalarial
activity but also a high cytotoxicity on MRC-5 cell line
(Table 2). The most cytotoxic molecules were the most active
on CQ-resistant strains, i.e., ferrocenic compounds. In the
organic series, the most cytotoxic compounds are the dual
drugs and particularly 3 with a piperidine group. The
cytotoxicity increased when the dimethylamino group is
replaced by a piperidine group. Because the cytotoxicity of
dual drugs was very important, we did not further design new
optimized antimalarial dual drugs analogues, but instead we
focused on a more comprehensive study on the factors
responsible for the cytotoxicity.
IC50
NF54a
(nM)
K1a
(nM)
Dd2a
(nM)
MRC-5b
compd
type
(μM)
CQ
1
reference
11.2
8.9
316.3
47.4
61.9
32.6
20.0
16.5
15.9
527.0
22.6
25.6
23.9
1.6
64.0
64.0
27.0
8.0
CQ analogue
dual drug
dual drug
reference
2
7.1
8.7
3
FQ
4
11.9
10.5
12.7
6.0
3.0
FQ analogue
dual drug
4.9
5.2
5
1.0
a The standard drug chloroquine (CQ) and ferroquine (FQ) served as
positive controls for the chloroquine-susceptible Plasmodium falciparum
strain NF54 and two chloroquine-resistant Plasmodium falciparum
strains K1 and Dd2. b The cytotoxicity is evaluated against this human
lung cell line.
already discussed the importance of the ferrocenyl moiety in
FQ in catalyzing the Fenton reaction under the specific
oxidizing conditions found in the parasitic digestive va-
cuole.36 While it has been well established that FQ, at high
concentration, was able to self-aggregate in aqueous solu-
tion,37 it rather seems that the appropriate redox behavior of
FQ is responsible for destruction of the hematin:drug ratio
by the Fenton reaction in the β-hematin assay.13 The com-
pound 5 might behave in the same way as FQ (Figure 1A,C).
Antiparasitic and Cytotoxic Activities. The antimalarial
activities of the short CQ and FQ analogues, 1 and 4, and the
Mannich base-derived 4-aminoquinoline dual drugs 2, 3, and
5, were first evaluated in a primary and secondary screenings
against one CQ-susceptible and two CQ-resistant strains of
P. falciparum, (Table 2). The cytotoxicity of the compounds
was determined in assays using the human lung MRC-5 cell
line (Table 2). Then, in a tertiary screening, the antimalarial
potencies of the same compounds were confirmed against 12
strains of P. falciparum expressing a varying degree of
resistance to established drugs, including CQ, quinine
(QN), and monodesethylamodiaquine (MDAQ), the active
metabolite of AQ. The data are presented as histograms in
Figure 2A,B. They were derived from the IC50 and IC90
values given in the Supporting Information (Table S1). CQ,
FQ, and MDAQ have been taken as internal references. All
the short analogues and the 4-aminoquinoline-based Man-
nich base derivatives inhibited growth of P. falciparum
strains in vitro whatever the susceptibility of the strain to
CQ and MDAQ. The organic series is slightly more active
than CQ or MDAQ on CQ- and MDAQ-susceptible strains,
while the ferrocenic series showed a similar activity than CQ.
On the other hand, the ferrocenic series presented a better
activity on the CQ-resistant strains than the organic series,
although this series showed a better activity than CQ (about
5 times more active). The dual drugs 2, 3, and 5 showed
antimalarial activity in the same range as the corresponding
short analogues 1 and 4 in the assays using NF54, K1, and
Dd2 strains, but the effects were always more potent for the
In Vivo Antimalarial Activity against Plasmodium berghei
in Mice. Two of the most antimalarial compounds 2 and 3, in
the organic series, were tested in P. berghei-infected ANKA
mice by intraperitoneal administration. Conditions of in
vivo screens for both compounds were conducted against
CQ-susceptible P. berghei-infected mice according to the
Peters’s four-day test.38 For comparative purposes, data
acquired in the same screens for CQ and the two derivatives
are included: CQ at 10 mg/kg po led to 99.2% reduction in
parasitemia (n = 5 mice). Antimalarial activity of compound
2 against P. berghei ANKA in CD1 mice was evaluated at 30
mg/kg ip ꢀ 4. Four of the five animals died following the first
drug injection, attesting for a very high toxicity of the
compound in vivo. Furthermore, following ip administra-
tion of compound 3 in four animals (Swiss mice), all animals
died due to serious toxic effects. While the dual drugs 2 and 3
showed high antimalarial activity, they also displayed very
high toxicity. The observation made with CQ and other 4-
aminoquinoline-based compounds interacting with DNA in
vitro prompted us to study the antiproliferative activity and
the binding of the newly designed dual drugs to DNA in
order to understand the origin of their cytotoxic properties.
Proliferation Assays. Because of the cytotoxicity and the
interaction of our compounds with DNA, we tested our 4-
aminoquinolines for their anticancer activity. The short
analogues and the dual drugs 1-5 were tested on the
glioblastoma carmustine-resistant cell line NCH89 (carmus-
tine namely BCNU) (Table 3). The obtained IC50 values on
glioblastoma carmustine-resistant cells are in accordance
with the cytotoxycity of the compounds. The most cytotoxic
molecules against human MRC-5 cells displayed the most
potent antiproliferative effect to kill the glioblastoma car-
mustine-resistant cells. Compared to CQ, FQ, and their
short analogues, we observed a significant higher antiproli-
ferative effect with the dual drugs (Figure 3). Also, the
ferrocenic was always found more active than the organic

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8 3219
Figure 2. Antimalarial activities of 4-aminoquinolines 1-5 against various P. falciparum strains.
Table 3. Antiproliferative Effects of CQ, FQ, and Compounds 1-5
(IC50 Values) on the NCH89 Glioblastoma Cells
partition coefficients of the dual drugs inside the parasite,
in agreement with the calculated pKa values of some repre-
sentatives of the organic series (Table S2 in the Supporting
Information). It is well-known that CQ increased the gene
transfection and that the number, nature, and location of
charges dramatically change the transfection properties of
the cationic aminoside chain of CQ analogues.28,40 Varying
the charge per molecule might have a profound effect on the
buffering power to promote endosomal escape from endo-
cytic pathway and DNA binding and trafficking, processing,
or degradation of DNA within the cells.
DNA Binding Studies. The spectrophotometric method
used in the DNA interaction study allowed the determina-
tion of the complex formation between the drugs and the
DNA evidenced by a change in the absorption spectra of the
ligand upon addition and binding to DNA. The stepwise
addition of small increments of DNA to a solution contain-
ing a constant concentration of the drug (e.g., 10 μM of 1) at
physiological pH is shown in Figure 4. An increase of ab-
sorption at 260 nm (characteristic for DNA), indicating the
higher concentration of DNA is ascertained. A progressive
compd
NCH89a cell line CQ
1
2
3
FQ
4
5
IC50 (μM)
SD (μM)
41.2 30.4
7.3
5.6
4.0
3.4
1.5
(7.4 (1.3 (0.4 (0.01 (0.1 (0.1 (0.008
a IC50 of carmustine = 615 ( 38 μM
series, both in terms of cytotoxicity and antiproliferative
properties. By far, combination of the ferrocene moiety and
the drug duality by linking a 4-aminoquinoline to the
Mannich base through reductive amination increased both
the cytotoxicity and the antiproliferative capability of final
molecules (Figure 3). Recently, combination of RuII and
chloroquine in the same molecular structure, stabilized by
arene ligands, was shown to result in enhanced activities
against resistant malaria parasites and against certain types
of cancer cells.39 Compared to the short CQ analogues, the
presence of both an additional amino group in the side
chain and an aryl group might significantly modify the

3220 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8
Wenzel et al.
Figure 3. Cytotoxicity versus antiproliferative activity of FQ and
compounds 1-5, relatively to CQ. The ratios of IC50 values for
studied drugs to the IC50 value for CQ in the assays performed (i) for
the evaluation of the cytotoxicity against human MRC-5 (empty
bars), (ii) for the evaluation of the antiproliferative effects against
human glioblastoma carmustine-resistant cells (blue bars), were
plotted (see IC50 values listed from Tables 2 and 3). Within each
series, organic or ferrocenic, the dual drugs showed the most
important toxicity relative to their parent representative, CQ or FQ.
decrease at 320 and 340 nm (characteristic for the quinoline
ring) and a shift of around 10 nm toward longer wavelengths
of the absorption peaks of the drugs is observed because
more and more drug is bound to DNA. The decrease of the
absorption is due to the intercalation of the quinoline into
DNA and is higher than the decrease due to the dilution
effect. Cohen et al. have already observed the same phenom-
enon for CQ in 1965 and has shown that the shift toward
longer wavelengths is characteristic for the formation of a
quinoline/DNA complex.27 The appearance of an isosbestic
point around 342 nm indicated that the limiting systems (i.e.,
the spectra of free and completely bound drug) intersect and
allowed the selection of a single wavelength for the study of
complex formation. This phenomenon was observed for all
studied compounds. The initial absorbance spectra of the
compounds are shown in Figure S2 in the Supporting
Information. Molar extinction coefficients were determined
in aqueous solutions, methanol and toluene, and shown in
the Supporting Information (Tables S3 and S4).
A more sensitive evaluation of the interaction of the 4-
aminoquinolines with DNA was achieved by using the
spectrofluorometry. It enables noise-free quantitative mea-
surements with one to two order lower concentrations of the
drugs than through the absorption. The emission of the
drugs was measured with the optimal excitation wavelength
of 320 nm, separating sufficiently the fluorescence emission
from the Raman scattering of water (360 nm) (Figure 5A,B).
The organic compounds showed high emission due to the
presence of the quinoline ring characterized by a large
fluorescence band at around 380 nm, with high variability
in fluorescence intensity and quantum yield (Figure 5A and
Table S5 in the Supporting Information). Only the spectrum
of CQ is 5 nm red-shifted. The high emission observed for 1
was found quenched in the case of both dual drugs 2, 3
containing one phenyl group. Compared to compounds 1
and 4, both substituted by a short amino side chain, CQ and
FQ displayed a much lower emission signal. Addition of
Figure 4. Absorption spectrum of a 10 μM inhibitor 1 solution
upon addition of a DNA solution in increasing concentration. The
binding to a linear double stranded DNA (containing 50% AT) was
investigated by UV-Vis spectrometry (A). Increased absorption at
260 nm corresponds to the DNA concentration. DNA induces
significant changes in the spectra with a marked decrease in
absorption of the free 4-aminoquinoline ligand around 320 and
340 nm, while an increased absorption is observed for the DNA-
ligand complex shifted to a longer wavelength around 350 nm ((B)
zoom between 280 and 380 nm).
DNA induces a monotonous decrease of the fluorescence
indicating the drug binding (data not shown). In contrast to
the organic drugs, the ferrocenic compounds showed very
low emission, around 10-fold less, and the emission signal
was compromised by the Raman scattering (Figure 5B). The
observed weak emission is probably due to a quenching of
fluorescence because of the presence of the ferrocene group.
In this way, the use of the spectrofluorometric method for the
quantitative titration of the ferrocenic compounds was not
possible.
A precise titration by addition of a DNA solution to a
1-10 μM drug solution was followed by spectrofluorometry
for the organic compounds and by UV-spectrophotometry
for all compounds (10 μM). To evaluate quantitatively the
binding of the drugs with DNA, the dissociation constants
were determined. In our case, due to the multiple binding
modes, the Scatchard representation is not usable and the fit
of the titration curves was too complex. However, we
estimated the dissociation constants by taking the total

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8 3221
Figure 6. Derivative melting curves of two different DNAs in the
presence of CQ, FQ, and 4-aminoquinolines 1-5. The derivative
melting curves of 50% (A,B) and 100% AT rich (C,D) DNA treated
with organic (A,C) and ferrocenic (B,D) compounds were illus-
trated as in Figure 4, while DNA alone is shown with mauve full line.
The thermal denaturation was measured in TE buffer containing 5
mM NaCl, at base pair:drug molar ratio of 15:1.
Figure 5. Fluorescence spectra of CQ, FQ, and 4-aminoquinolines
1-5 measured in aqueous solution. The fluorescence emission
spectra for organic (A) and ferrocenic (B) compounds were illus-
trated as CQ, (black full); FQ, (black, dashed); 1, (red, full); 4, (red,
dashed); 2, (green, full); 3, (blue, full); 5, (blue dashed). Measure-
ments were performed in TE buffer containing 5 mM NaCl, with
excitation at 320 nm, spectra are corrected for the buffer signal and
instrument response.
thermal strand separation and thus to increase Tm.26 The
effect of the drugs on the derivative melting curves of two
DNA types is presented in Figure 5. To study also the role
of DNA base composition in the drug binding, we com-
pared a DNA with 50% AT content: a linearized pUC18
plasmid (Figure 6A,B) to a synthetic polydAdT double-
strand oligonucleotide (Figure 6C,D). This oligonucleo-
tide has been chosen because of the unique AT-richness of
Plasmodium genome (80% AT, relative to 60% in human
DNA), especially in the intergenic regions (more than
90%).42
Table 4. Kd Values of CQ, FQ, and Compounds 1-5 Estimated by
UV-Vis Spectrophotometry and Spectrofluorometry at Low Ionic
Force (μ = 12 mM)
Kd, μM
compd
by absorption
by emission
CQ
1
7.5
17.5
7.0
9.0
8.0
8.0
4.0
2
3
1.0
For all compounds, the thermal stability of both DNA
increased, evidencing the interaction of the drugs with DNA
in a slightly varying extent. The stabilization was propor-
tional to the drug concentration and depended on the base
pair/drug molar ratio (data not shown). The data presented
in Figure 5 showed the results obtained at 1 μM drug
concentration in a 15 base pair:drug ratio. All charac-
teristic melting temperatures are listed in Table S5 of the
Supporting Information. As an observed feature, the ferro-
cenic dual drug 5 has significantly smaller effect on both
DNAs than any other compound. This is in apparent con-
trast with the very low dissociation constant evaluated for
this compound (Table 4). One possible cause might be linked
to the Fenton catalyst properties of these ferrocenic 4-
aminoquinolines,36 in agreement with the destruction of
the drug:hematin ratio observed in the β-hematin inhibition
assay. The dual drug with a terminal N,N-dimethyl amine 2 is
the most stabilizing ligand of 50% AT-containing DNA,
while the short FQ analogue 4 is the most stabilizing agent of
polydAdT. The difference between the melting temperature
of the DNA treated with the drug and the temperature of the
free DNA (ΔTm) gives quantitative information about the
force of the interaction between the drug and the DNA. On
the other side, this difference is also dependent on the free
DNA alone. To avoid this effect, we compared the DNA
stabilization by one compound with the stabilization caused
FQ
4
5.0
8.0
0.2
5
DNA concentration added in the curve when half of the drug
is bound to DNA. The dissociation constants were estimated
at a low ionic force of 12 mM at pH 7.5. The dissociation
constants indicated that all our drugs showed a high affinity
for DNA (Table 4). For CQ, the determined Kd value was in
the same order, as reported by Viola et al. in a similar
experiment.41 The dual drugs 3 and 5 containing the piper-
idine group displayed a better affinity toward the DNA with
Kd values of 1.0 and 0.2 μM, respectively. These results are in
accordance with the highest cytotoxicity and the most potent
antiproliferative effect of these two compounds in each
series, i.e., in the organic and the ferrocenic series.
The interaction between the drugs and DNA could also be
sensitively recognized due to the stability changes in macro-
molecules. The cooperativity of the thermal denaturation
amplifies the effect of the local interactions: one from
thousand base pairs involved in an interaction may be
detectable. The effect of the drugs on thermal stability of
DNA indicating possible structural changes was measured
by UV absorption melting. Intercalation of molecules into
the double helix is known to stabilize the DNA against

3222 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8
Wenzel et al.
organic dual drugs through the buffering capabilities of their
polyamine side chain.
Experimental Section
€
Chemistry. Melting points were determined on a Buchi melt-
ing point apparatus and were not corrected. 1H (300 MHz) and
13C (75 MHz) NMR spectra were recorded on a Bruker DRX-
300 spectrometer and a Bruker AC 300 spectrometer; chemical
shifts were expressed in ppm relative to TMS; the protons are
indicated as QnH (H linked to 4-aminoquinoline), H-Pip (H
linked to the piperidine moiety), and PhH (H linked to phenyl);
in the 1H spectra multiplicity is indicated as s (singlet), d
(doublet), t (triplet), and m (multiplet); C indicates a quaternary
carbon in 13C spectra. Elemental analyses were carried out at the
Figure 7. Illustration of the different stabilizing effects of FQ and
4-aminoquinolines 1-5 relative to CQ. ΔTm rel is the change in
melting temperature Tmint of a drug relative to the change produced
by CQ, expressed in %. Empty bars were plotted for 50% AT-
containing DNA and red bars for 100% AT-rich DNA. Tmint values
were taken from Figure 5 and listed in Table S5.
€
Mikroanalytisches Laboratorium der Chemischen Fakultat der
Universitat Heidelberg. MALDI-TOF and ESI-MS were re-
€
€
corded at facilities of the Institut fur Organische Chemie der
Universitat Heidelberg and at the Centre Commun de Mesures
by CQ on the same DNA type. The relative ΔTm (ΔTm rel)
was defined in the following equation:
€
ꢀ
de l’Universite des Sciences et Technologies de Lille, respec-
tively. Analytical TLC was carried out on precoated Sil G-25
UV254 plates from Macherey-Nagel, detection by UV lamp.
Flash chromatography was performed using silica gel G60
(230-400 mesh) from Macherey-Nagel. The unsaturated Man-
nich base 1-(2-chlorophenyl)-5-(dimethylamino)-pent-1-en-3-one
hydrochloride, the CQ-analogue N-(7-Chloro-quinolin-4-yl)-
N0-isopropyl-ethane-1,2-diamine 1, and N-(7-Chloro-quinolin-
4-yl)-ethane-1,2-diamine were synthesized as previously de-
scribed.43,44 The saturated Mannich base 7 (mp 192-193 ꢀC)
was prepared via a nucleophilic substitution reaction of
p-chloropropiophenone, piperidine, and CsCO3. 3-Dimethyla-
mino-1-phenyl-1-propanol 8 was prepared by NaBH4 reduction
of the commercially available N,N-dimethylaminopropiophe-
none hydrochloride 6 according to reported procedures.45
1-Phenyl-3-piperidin-1-yl-propan-1-one 7 was reduced with
an excess of sodium borohydride in methanol at 0 ꢀC to the
1-phenyl-3-piperidin-1-yl-propan-1-ol 10. 3-Chloro-N,N-dimeth-
yl-3-phenylpropan-1-amine hydrochloride 9 and 1-(3-chloro-3-
phenylpropyl)piperidine hydrochloride 11 were prepared from
the corresponding alcohols, 3-(dimethylamino)-1-phenylpropan-
1-ol 8, and 1-phenyl-3-(piperidin-1-yl)propan-1-ol 10, by action of
thionyl chloride in chloroform under heating to reflux as pre-
viously reported.46 7-Chloro-quinolin-4-yl-[3-(isopropylamino-
methyl)-ferrocenyl]-amine 4 was synthesized as previously de-
scribed.13 The purity of compounds 2, 3, and 5 was controlled by
melting points, and elemental analyses that agreed with the
calculated values within 0.4%. Analytical high pressure liquid
chromatography (HPLC) was performed on a Spectra system
equipped with a UV detector set at 254 nm and using two types of
high pressure liquid chromatography (HPLC) columns, a Macher-
ey-Nagel C18 Nucleosil column (4 mm ꢀ 300 mm, 5 μm, 100 A) or
Macherey-Nagel EC 250/4.6 Nucleodur 100-5 CN-RP (4 mm ꢀ
300 mm, 5 μm, 100 A). Compounds were dissolved in acetonitrile
and injected through a 50 μL loop. The following solvent systems
were used: eluent (A): 0,05% trifluoroacetic acid (TFA) in H2O,
eluent (B) 100% CH3CN. HPLC retention times (HPLC tR) were
obtained, at flow rates of 1 mL/min, using the following condi-
tions: 100% eluent A for 5 min, then a gradient run to 100% eluent
B over the next 20 min.
ΔTm rel ð%Þ ¼ ðTm drug - TmÞ : ðTm CQ - TmÞ
where Tm drug, Tm CQ, and Tm are the melting temperatures of
DNA in the presence of the drug, or of CQ, or in the absence
of any drug, respectively. For this calculation, we took the
int
Tm values obtained as the middle of the integrated area
below the derivative melting peaks, which is the most in-
sensitive parameter to the type of DNA. The ΔTm rel values
relative to CQ (Table S5, Supporting Information) were
plotted in Figure 7. Only the dual drugs 2, 3, and 5 showed
a significantly higher affinity toward the GC- vs the AT-rich
regions. Interestingly, the same dual drugs showed the high-
est antiproliferative effects compared to the activity of short
analogues (Table 3).
Conclusion
Our original reason for embarking on an investigation of
the interaction of 4-aminoquinoline derivatives with two
different types of DNA was for the purpose of understanding
the high cytotoxicity of our designed antimalarial dual drugs.
Research of DNA binding is important for the design of novel
antimalarial agentstoprevent the interaction of drug into host
DNA and to decrease their human cytotoxicity. Noteworthy
is the fact that DNA binding studies have to be taken with
caution toward the relevance of the mechanism of drug action
in vivo because the pharmacokinetics of the drug in the cells
do not necessarily reflect the observations made in experi-
ments in vitro with isolated targets. For more than 60 years,
the DNA binding properties of CQ have not compromised the
success of the use of CQ as antimalarial drug in human
medicine. More work need to be continued to know if FQ,
which is in clinical trials, will be a successful antimalarial drug
as much as CQ was. However, the structural changes in the
side chain of CQ might dramatically affect the endosomal
escape, the DNA binding and trafficking, processing, or
degradation of DNA within the cells, as observed in gene
expression and transgene expression by CQ analogues.28
Altogether our dual drugs seem to be more appropriate to
be developed as antiproliferative derivatives against mamma-
lian cancer cells than Plasmodium parasites. Furthermore, our
findings based on the observed interaction of newly designed
4-aminoquinolines with GC-rich DNA revealed the possibi-
lity to specifically target the GC-rich domains of DNA.
Future directions are aimed at the evaluation of gene transfec-
tion and delivery through endosomal escape properties of
N1-[2-(7-Chloro-quinolin-4-ylamino)-ethyl]-N3,N3-dimethyl-
1-phenyl-propane-1,3-diamine (2 Base and 2 Chlorhydrate Salt).
7-Chloro-(quinolin-4-yl)ethane-1,2-diamine(400 mg, 1.80 mmol)
was dissolved in anhydrous ethanol (35 mL) under heating.
After cooling to room temperature, triethylamine (1.26 mL, 9.02
mmol) was added. Then the 3-chloro-N,N-dimethyl-3-phenyl-
propylamine hydrochloride was added as solid in three portions:
first, 444 mg (1.89 mmol), then further 92 mg (0.39 mmol) after
18 h and then 51 mg (0.22 mmol) after further 7 h. After addi-
tional 15 h stirring at room temperature, the solvent was evapo-
rated and the residue was purified by flash chromatography

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8 3223
(first on neutral Al2O3 with CHCl3/CH3OH 40:1 and then, on
SiO2 with acetone/methanol/water/triethylamine 10:3:2:0.2) to
yield the base 2 as a colorless oil/solid (0.43 g, 1.12 mmol, 62%);
mp (ꢀC): 210-212. 1H NMR (300 MHz, CDCl3) δ (ppm): 8.37
(d, 1H, 3J = 6 Hz, QnH-2), 7.86 (d, 1H, J = 2 Hz, QnH-8), 7.74
(d, 1H, J = 9 Hz, QnH-5), 7.25 (mc, 5H, PhH), 7.18 (dd, 1H,
J = 2 Hz, J = 9 Hz, QnH-6), 6.26 (sbr, 1H, NH), 6.18 (d, 1H, 3
J = 6 Hz, QnH-3), 4.37 (sbr, 1H, NH), 3.66 (t, 1H, 3J = 6 Hz,
CH), 3.19 (m, 2H, NCH2), 2.84 (m, 1H, NCH2), 2.73 (m, 1H,
NCH2), 2.33 (m, 1H, NCH2), 2.21 (m, 1H, NCH2), 2.12 (s, 6H,
CH3), 1.92 (m, 1H, CH2NMe2), 1.77 (m, 1H, CH2NMe2). 13C
NMR (75 MHz, CDCl3) δ (ppm): 151.5 (CH), 149.9 (C), 148.6
(C), 142.9 (C), 134.6 (C), 128.5 (CH), 128.0 (CH), 127.3 (CH),
126.9 (CH), 124.9 (CH), 121.7 (CH), 117.2, 98.7 (CH), 61.9
(CH), 57.0 (CH2), 45.2 (CH3), 44.8 (CH2), 41.9 (CH2), 35.0
(CH2). The chlorhydrate salt of 2 was prepared by solubilizing
the compound in a minimum of methanol (2 mL) and 4 equiv of
TMSCl was added. The precipitated hydrochloride was sepa-
rated by filtration, crystallized in ethanol, washed with cold
ether, and dried. MS (FABþ-MS) m/z: 383.4 (MHþ). Anal.
C22H30N4 3HCl 0.35H2O.
CH2), 1.39 (m, 4H, H-Pip), 1.28 (m, 2H, H-Pip). 13C NMR (75
MHz, CDCl3) δ (ppm): 152.1 (CH), 149.6 (C), 149.2 (C), 146.7
(C), 134.8 (C), 128.6 (CH), 128.5 (CH), 127.3 (CH), 127.0 (CH),
125.1 (CH), 123.5 (CH), 118.5 (C), 99.7 (CH), 85.9 (C), 84.3 (C),
70.0 (CH), 69.1 (CH), 66.5 (CH), 64.2 (CH), 61.7 (CH), 56.2
(CH2), 54.5 (CH2), 45.8 (CH2), 42.0 (CH2), 34.3 (CH2), 25.9
(CH2), 24.3 (CH2). MS (ESI) m/z: 563 MHþ 37Cl, 609 Mþ• 37Cl,
608 MHþ 35Cl, 607 Mþ• 35Cl, 391 (C9H5N37ClNHCH2C10-
H8FeCH2)þ, 389 (C9H5N35ClNHCH2C10H8FeCH2)þ. Anal.
C35H39ClN4Fe 1.5H2O.
3
Time-Dependent Formation of Glutathione-Mannich Base
Monoadducts. To a solution of 650 μL of H2O and 250 μL of
aqueous NH4HCO3 solution (25 μM), 50 μL of GSH (20 mM in
H2O) and 50 μL of the inhibitor (20 mM in DMSO) were added.
The reaction mixture (pH 6.5-7) was incubated at room
temperature. After different times, the solution was injected in
HPLC to determine the Mannich base:monoSG-adduct versus
time (min). The HPLC analysis was performed on a Hitachi
Merck L-4000 equipped with a UV detector set at 254 nm.
HPLC retention times were obtained using the following con-
ditions: 100% eluent A (0.05% trifluoroacetic acid (TFA) in
H2O) for 5 min, a gradient up to 100% B (0.05% TFA in H2O/
CH3CN (1:4) within 10 min, 100% B for 5 min, then again a
gradient up to 100% A within 5 min at a flow rate of 1 mL/min.
Drugs. Ferroquine base (FQ), licensed as SR97193 today, was
obtained from Prof. J. Brocard (France). Chloroquine diphos-
phate (CQ) was purchased from Sigma Aldrich, and quinine
hydrochloride (QN) and monodesethylamodiaquine (MDAQ)
were obtained from the World Health Organization (Geneva,
Switzerland). FQ and synthetic compounds were resuspended
and then diluted in RPMI-DMSO (99v/1v) to obtain final
concentration ranging from 0.125 to 500 nM. CQ was resus-
pended in water in concentrations ranging between 5 and 3200
nM for CQ. QN and MDAQ were first dissolved in methanol
and then diluted in water to obtain final concentrations ranging
from 5 to 3200 nM and from 1.56 to 1000 nM, respectively.
Assay of β-Hematin Inhibition in Eppendorf Tubes. We deter-
mined the IC50 values for inhibition of β-hematin formation
using Egan’s test35 with very slight modifications as described
below. Drug solutions (44.6, 26.8, 13.4, 8.9, 6.7, 4.5, 2.2, 0.9
mM) were prepared by dissolving the drug in methanol/HCl 1:1.
Hematin stock solution (1.68 mM) was prepared by dissolving
bovine hemin (1.08 mg) in 0.1 M NaOH (986 μL). The solution
was incubated at room temperature for 60 min. In a series
of Eppendorf tubes, 4 μL of drug solution were dispensed and
12.9 M sodium acetate solution, pH 5.0, (11.7 μL) preincubated
at 60 ꢀC was added. Then the Eppendorf tubes were placed in an
incubator at 60 ꢀC. The β-hematin formation process was
initiated by adding the hematin stock solution (20.2 μL) pre-
pared above. The final hematin concentration was 1 mM, the
final drug concentrations were 5, 3, 1.5, 1, 0.75, 0.5, 0.25, 0.1,
and 0 mM, and the final solution pH was 4.5. The reaction
mixtures were incubated at 60 ꢀC for 60 min. After incubation,
the reaction mixture was quenched at room temperature by
adding 900 μL of 200 mM HEPES 5% (v/v) pyridine solution,
pH 8.2, to adjust the final pH of the mixtures to a value between
7.2 and 7.5. Then, 1100 μL of 20 mM HEPES 5% (v/v) pyridine
solution, pH 7.5, was added. The Eppendorf tubes were shaken,
and the precipitate of β-hematin was scrapped from the walls of
the Eppendorf tubes to ensure complete dissolution of hematin.
The β-hematin was allowed to settle at room temperature for at
least 15 min. The supernatant was carefully transferred to a
cuvette without disturbing the precipitate, and absorption was
measured at 405 nm.
3
3
N-[(7-Chloro-quinolin-4-yl)-ethyl]-N0-(1-hexazin-1-yl-3-piper-
idin-1-yl-propyl)-ethane-1,2-diamine hydrochloride 3. N1-(7-Chloro-
quinolin-4-yl)-ethane-1,2-diamine (400 mg, 1.80 mmol) was
dissolved in anhydrous ethanol under heating. After cooling
to room temperature, an excess of triethylamine (1.26 mL, 9.02
mmol) was added. Then 1-(3-chloro-3-phenylpropyl)piperidine
hydrochloride (693 mg, 2.53 mmol) was added as solid in three
portions during 2 days. After additional stirring at room tempe-
rature, the solvent was evaporated in vacuo and the residue was
purified by flash-chromatography on neutral Al2O3 (CHCl3/
CH3OH 40:1) to obtain the base 3 as a colorless solid (140 mg;
0.33 mmol, 18%). The chlorhydrate salt of 3 was prepared as
1
previously described for 2; mp (ꢀC): 121-125. H NMR (300
MHz, CDCl3) δ (ppm): 8.26 (d, 1H, J = 5.31 Hz, QnH-2), 7.73
(s, 1H, QnH-8), 7.59 (d, 1H, J = 8.94 Hz, QnH-5), 7.16-7.04
(m, 6H, PhH þ QnH-6), 6.07 (d, 1H, J = Hz, QnH-3), 5.91 (sbr,
1H, NH), 3.52 (t, 1H, J = 5.35 Hz, CH), 3.13-2.85 (m, 2H,
NCH2), 2.79-2.50 (m, 2H, NCH2), 2.31-1.90 (m, 5H, CH2 þ
H-Pip), 1.90-1.45 (m, 2H, CH2), 1.45-1.05 (m, 7H, H-Pip). 13
C
NMR (75 MHz, CDCl3) δ (ppm): 151.67 (CH), 149.82 (C),
148.81 (C), 143.42 (C), 134.47 (C), 128.34 (CH), 128.20 (CH),
127.07 (CH), 126.81 (CH), 124.80 (CH), 121.61 (CH), 117.22
(C), 98.81 (CH), 62.32 (CH), 56.65 (CH2), 54.37 (CH2), 44.96
(CH2), 42.17 (CH2), 34.36 (CH2), 25.57 (CH2), 23.96 (CH2). MS
(MALDI-TOF) m/z: 423.5 (Mþ). Anal. C25H31ClN4 2.8HCl
3
3
1.9H2O.
7-Chloro-N-(2-{[1-phenyl-3-(piperidin-1-yl)propylamino]methyl}-
ferrocenyl)quinolin-4-amine 5. N-[2-(Aminomethyl)ferrocenyl)-
7-chloroquinolin-4-amine (234 mg, 0.58 mmol) was dissolved in
anhydrous ethanol. After complete dissolution, an excess of
triethylamine (5 equiv) and 1-(3-chloro-3-phenylpropyl)piperi-
dine hydrochloride (134 mg, 0.52 mmol) were added. The
mixture was stirred for 5 h at room temperature. After the addi-
tion of 1-(3-chloro-3-phenylpropyl)piperidine hydrochloride
(28 mg, 0.11 mmol), the reaction mixture was allowed to stir
for further 3.30 h. Then the solvent was evaporated in vacuo and
the residue was purified by flash-chromatography on silica gel
(EtOAc/NEt3 97.5:2.5) to obtain the base 5 as a brown solid (28
1
mg; 0.04 mmol, 6%); mp (ꢀC): 59-62. H NMR (300 MHz,
CDCl3) δ (ppm): 8.49 (d, 1H, J = 3.9 Hz, QnH-2), 7.87 (d, 1H,
J = 4.2 Hz, QnH-8), 7.79 (d, 0.5H, J = 9.0 Hz, QnH-5), 7.64 (d,
0.5H, J = 9.0 Hz, QnH-5), 7.18-7.04 (m, 6H, PhH þ QnH-6),
6.59 (sbr, 0.5H, NH), 6.43 (d, 1H, J = 5.9 Hz, QnH-3), 6.25 (sbr,
0.5H, NH), 4.24 (d, 2H, J = 12.9 Hz, NCH2), 4.18 (m, 1H,
H-Cp), 4.12 (m, 1H, H-Cp), 4.01 (s, 5H, H-Cp0), 4.00 (m, 1H,
H-Cp), 3.63 (t, 0.5H, J = 6.4 Hz, CH), 3.57 (t, 0.5H, J = 6.4 Hz,
CH), 3.49 (d, 0.5H, J = 12.3 Hz, NCH2), 3.37 (d, 0.5H, J = 12.3
Hz, NCH2), 3.30 (d, 0.5H, J = 12.3 Hz, NCH2), 3.22 (d, 0.5H,
J = 12.3 Hz, NCH2), 2.22 (m, 6H, CH2 þ H-Pip), 2.02 (m, 2H,
In Vitro Antimalarial Activities. Parasite Cultures and
Primary Screening Against NF54 and K1 P. falciparum Strains.
CQ-Susceptible NF54 was cultivated in a variation of the
medium previously described,47 consisting of RPMI 1640 sup-
plemented with 0.5% ALBUMAX II, 25 mM Hepes, 25 mM
NaHCO3 (pH 7.3), 0.36 mM hypoxanthine, and 100 μg/mL

3224 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8
Wenzel et al.
neomycin. The chloroquine-, pyrimethamine-, and cycloguanil-
resistant K1 strain (Thailand) of P. falciparum were acquired
from MR4 (Malaria Research and Reference Reagent Resource
Center, Manassas, VA). P. falciparum in vitro culture was
carried out using standard protocols47 with modifications.
Briefly, parasites were maintained in tissue culture flasks in
human A Rhþ erythrocytes at 5% hematocrit in RPMI 1640
supplemented with 25 mM HEPES, 24 mM NaHCO3, 0.2%
(w/v) glucose, 0.03% L-glutamine, 150 μM hypoxanthine, and
0.5% Albumax II (Invitrogen) in a 5% CO2/air mixture at 37 ꢀC,
and the medium was changed daily. Asynchronous ring stage
cultures were prepared at 1% parasitemia and 50 μL added per
well, the top test drug final concentration being 30 μM. After 24
h incubation, 37 ꢀC, 5% CO2 5 μL (3H)-hypoxanthine was
added (0.2 μCi/well)48 and plates were shaken for 1 min and
then incubated for 48 h. The plates were freeze/thawed rapidly,
harvested, and dried. (3H)-hypoxanthine uptake was measured
using a microbeta counter (Wallac 1450). IC50 values were
calculated (Prism).
infected intravenously with 2 ꢀ 106 parasitized red cells on day
þ0. For administration, compounds were freshly prepared in
10% DMSO in sterile phosphate-buffered saline the day of use.
Two hours postinfection, mice received the first treatment by the
intraperitoneal (ip) route. Mice were further treated on days
þ1-3. Blood films from tail blood were prepared on day þ4,
and parasitemia was determined by microscopic examination of
Giemsa-stained blood films. Compounds 2 and 3 were tested at
a daily dose of 30 mg/kg (2) or 50 mg/kg (3) by the ip route.
Chloroquine treatment po at 10 mg/kg/day was included as a
positive control and resulted in complete inhibition (data not
shown). Intraperitoneal dosing of CQ have shown similar
activity (98.9% inhibition at 10 mg/kg ip) in a number of tests
but was not done specifically with this series of compounds.
Mice were treated and levels of parasitemia determined as
described.
Evaluation of the Cytotoxicity. Human MRC-5SV2 cells are
ꢁ
cultured in Earls MEM þ 5% FCSi. Assays are performed in 96-
well microtiter plates, each well containing about 104 cell/well.
After 3 days incubation, cell viability was assessed fluorimetri-
cally after addition of resazurin and fluorescence was measured
(λex 550 nm, λem 590 nm). The results are expressed as % reduc-
tion in cell growth/viability compared to untreated control wells
and CC50 is determined. Compounds are tested within the dose
range of 0.25 μM. When the CC50 is lower than 4 μM, the
compound is classified as highly toxic.
Parasite Cultures, Secondary Screening Against P. falciparum
Strain Dd2. The IC50 was tested by standard in vitro antipro-
liferation assays.48 Infected erythrocytes in ring stage (0.5%
parasitemia, 2.5% hematocrit) in 96-well plates were exposed to
the compounds for 48 h and then to radioactive hypoxanthine
for 24 h. The amount of radioactivity in precipitable material
served as an index of cell proliferation.
Proliferation Assays. Cells from the NCH89 glioblastoma cell
line were cultured as described.51 Proliferation assays were
performed, as described,52 using the BrdU Labeling and Detec-
tion Kit III (Roche Diagnostics, Germany). Cells (7 ꢀ 103 cells/
well) were seeded in replicas in 96-well plates for 24 h. Then CQ,
FQ, and drugs 1-5 were added in concentrations ranging from
0.1 to 100 μM to the wells, and 48 h later, BrdU was applied at a
final concentration of 100 μM and measured according to the
manufacturer’s instructions.
DNA Binding Studies. Preparation of the Drug Solutions.
Drug stock solutions (10 mM) were prepared by dissolving
the drug in methanol or water according to its lipophilicity.
TE buffer (10 mM Tris-HCl 0.1 mM EDTA, at pH 7.4) contain-
ing 5 mM NaCl was applied as buffer in the optical measure-
ments.
Spectroscopic Characterization of the Compounds. UV absorp-
tion spectra between 220 and 380 nm were measured on a Cary
4E spectrophotometer and fluorescence spectra on an SLM-
Aminco 8100 fluorimeter with excitation at 320 nm, detection
between 340 and 500 nm. Fluorescence spectra were corrected
for solvent signal and instrument response. Quartz cells of 1 cm
optical path were applied filled with 2 mL of sample of 3 mM
drug concentration. Spectra were registered in TE buffer con-
taining 5 mM NaCl, in methanol and in toluene. The specific
molar absorption coefficients were evaluated from the absorp-
tion spectra. The quantum yields in aqueous solution were
determined relative to that of CQ by dividing the integrated
emission spectra by the absorption at the common excitation
wavelength and expressed as % of the quantum yield of CQ. In
the case of the poorly emitting ferrocenic compounds, the Ra-
man scattering was taken into correction.
Evaluation of Drug Binding with DNA. The spectral changes
upon addition of DNA to the drugs were measured by absorp-
tion and fluorescence. Quartz cells of optical path length 3 mm
were used with 100-150 μL samples. The binding experiments
were accomplished by titrating a 1 or 10 μM drug solution with a
linear pUC18 DNA solution (50% AT) measuring the absorp-
tion or fluorescence spectra at room temperature. The final
concentrations of DNA in the cuvette were varying between 0.1
and 50 μM base pairs.
Parasite Cultures and Tertiary Screening Against Twelve
Strains Expressing Varying Susceptibilities to Chloroquine and
Quinine. Twelve parasite strains (well characterized laboratory
strains or strains obtained from isolates after growth in culture
for an extended period of time) from a wide panel of countries:
Africa (3D7), Brazil (Bres), Cambodia (K2 and K14), Cameroon
(FCM29), Djibouti (Voll), the Gambia (FCR3), Indochina
(W2), Niger (L1), Senegal (8425), Sierra Leone (D6), and
Uganda (PA)49,50 were maintained in culture in RPMI 1640
(Invitrogen, Paisley, U.K.), supplemented with 10% human
serum (Abcys SA, Paris, France) and buffered with 25 mM
HEPES and 25 mM NaHCO3. Parasites were grown in
A-positive human blood under controlled atmospheric condi-
tions that consists of 10% O2, 5% CO2, and 85% N2 at 37 ꢀC
with a humidity of 95%. For in vitro isotopic microtests, 25 μL/
well of antimalarial drug and 200 μL/well of the parasitized red
blood cell suspension (final parasitemia, 0.5%; final hematocrit,
1.5%) were distributed into 96-well plates. Parasite growth was
assessed by adding 1 μCi of tritiated hypoxanthine with a
specific activity of 14.1 Ci/mmol (Perkin-Elmer, Courtaboeuf,
France) to each well at time zero. The plates were then incubated
for 48 h in a controlled atmospheric condition. Immediately
after incubation, the plates were frozen and thawed to lyse
erythrocytes. The contents of each well were collected on
standard filter microplates (Unifilter GF/B, Perkin-Elmer)
and washed using a cell harvester (Filter-Mate Cell Harvester,
Perkin-Elmer). Filter microplates were dried, and 25 μL of
scintillation cocktail (Microscint O, Perkin-Elmer) was placed
in each well. Radioactivity incorporated by the parasites was
measured with a scintillation counter (Top Count, Perkin-
Elmer). The IC50, defined as the drug concentration able to
inhibit 50% of parasite growth, was assessed by identifying the
drug concentration corresponding to 50% of the uptake of
tritiated hypoxanthine by the parasite in the drug-free control
wells. The IC50 value was determined by nonlinear regression
analysis of log-based dose-response curves (Riasmart, Pack-
ard, Meriden, USA). The cutoff values, defined statistically (>2
SD above the mean with or without correlation with clinical
failures) for in vitro resistance or reduced susceptibility to CQ,
QN, and MDAQ were 100, 800, and 60 nM, respectively.
In Vivo Antimalarial Activities. Compounds 2 and 3 were
tested in the P. berghei ANKA mouse model by using the 4-day
suppressive test, as indicated by Peters,38 and using chloroquine
as a positive control. Briefly, naive 18-20 g Swiss mice were
Optical Melting Measurements. Thermal denaturation curves
of the DNA were recorded by absorbance at 260 nm on a
Cary 4E spectrophotometer (Varian, Mulgrave, Australia)
equipped with a Peltier thermoregulator and an automatic

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8 3225
cuvette changer. The samples contained 1 μM drug solution and
DNA of 15 μM base pair concentration, with initial absorbance
of 0.2 at 260 nm in quartz cells of 1 cm path length in TE buffer
with 5 mM NaCl. The heating rate was 0.5 ꢀC/min in the
temperature range 30-97 ꢀC, and the absorbance data were
collected at every 0.5 ꢀC. Data were treated using the program
KaleidaGraph. The curves were normalized to the absorbance
at room temperature and smoothed in five-point intervals.
Derivative melting curves were calculated from the smoothed
data and once more smoothed in five-point intervals. The
maxima of the derivative melting curves were read as corre-
sponding melting temperatures (Tmmax). To correct for differ-
ences in peak asymmetry, two further temperatures were
calculated to characterize the melting: the middle of the peak
at its half height (Tm1/2) and the middle of the integrated surface
below the derivative melting peak (Tmint). Two types of DNA
were used for this study: linear pUC18 DNA (containing 50%
AT) and synthetic polydAdT (100% AT)
(3) Zarchin, S; Krugliak, M; Ginsburg, H. Digestion of the host
erythrocyte by malaria parasites is the primary target for quino-
line-containing antimalarials. Biochem. Pharmacol. 1986, 35, 2435–
2442.
(4) Egan, T. J.; Chen, J. Y.-J.; de Villiers, K. A.; Mabotha, T. E.;
Naidoo, K. J.; Ncokazia, K. K.; Langford, S. J.; McNaughton, D.;
Pandiancherri, S.; Wood, B. R. Haemozoin (β-haematin) bio-
mineralization occurs by self-assembly near the lipid/water inter-
face. FEBS Lett. 2006, 580, 5105–5110.
(5) Egan, T. J. Recent advances in understanding the mechanism of
hemozoin (malaria pigment) formation. J. Inorg. Biochem. 2008,
102, 1288–1299.
(6) Slater, A. F. G. Malaria Pigment. Exp. Parasitol. 1992, 74, 362–
365.
(7) Sanchez, C. P.; Stein, W.; Lanzer, M. Trans stimulation provides
evidence for a drug efflux carrier as the mechanism of chloroquine
resistance in Plasmodium falciparum. Biochemistry 2003, 42, 9383–
9394.
(8) Sanchez, C. P.; McLean, J. E.; Stein, W.; Lanzer, M. Evidence for a
substrate specific and inhibitable drug efflux system in chloroquine
resistant Plasmodium falciparum strains. Biochemistry 2004, 43,
16365–16373.
(9) Sanchez, C. P.; McLean, J. E.; Rohrbach, P.; Fidock, D. A.; Stein,
W. D.; Lanzer, M. Evidence for a pfcrt-associated chloroquine
efflux system in the human malarial parasite Plasmodium falcipar-
um. Biochemistry 2005, 44, 9862–9870.
(10) Sanchez, C. P.; Rohrbach, P.; McLean, J. E.; Fidock, D. A.; Stein,
W. D.; Lanzer, M. Differences in trans-stimulated chloroquine
efflux kinetics are linked to PfCRT in Plasmodium falciparum. Mol.
Microbiol. 2007, 64, 407–420.
Acknowledgment. We are grateful to B. Jannack and M.
€
Bruckner, who contributed by excellent technical assistance to
the chemical preparation of starting materials in bulk, to
HPLC measurements and in performing the hematin poly-
merization assays, respectively. Dr. G. Furrer is acknowl-
edged for recording 1H, 13C NMR spectra (Organisch-
Chemisches Institut, Heidelberg University). Dr. A. Page
is acknowledged for recording the mass spectra (Centre
€
(11) Gilberger, T.-W.; Schirmer, R. H.; Walter, R. D.; Muller, S.
Deletion of the parasite-specific insertions and mutation of the
catalytic triad in glutathione reductase from chloroquine-sensitive
Plasmodium falciparum 3D7. Mol. Biochem. Parasitol. 2000, 107,
169–179.
ꢀ
Commun de Mesures de l’Universite des Sciences et Techno-
logies de Lille). The authors thank R. Amalvict, E. Baret, and
J. Mosnier for technical support (Institut de Recherche
€
(12) Meierjohann, S.; Walter, R. D.; Muller, S. Regulation of intracel-
lular glutathione levels in erythrocytes infected with chloroquine-
sensitive and chloroquine-resistant. Plasmodium falciparum. Bio-
chem J. 2002, 368, 761–768.
ꢀ
ꢀ
Biomedicale des Armees;Antenne de Marseille). Our work
is supported by the CNRS-DFG program of the Centre
National de la Recherche Scientifique (E.D.C.) and by the
Deutsche Forschungsgemeinschaft (SFB 544 “Control of
Tropical Infectious Diseases”, B14 project (E.D.C.)). N.W.
isgrateful for a fellowship from the SFB544-DFG program. A
BDI fellowship from CNRS and Region Nord-Pas-de-Calais
to N.C. is gratefully acknowledged. We are grateful to the
PROCOPE program that supported this French-German
collaboration. This investigation received financial support
from the UNICEF/UNDP/World Bank/WHO Special Pro-
gramme for Research and Training in Tropical Diseases
(TDR).
(13) Chavain, N.; Davioud-Charvet, E.; Trivelli, X.; Mbeki, L.;
Rottmann, M.; Brun, R.; Biot, C. Antimalarial antimalarial activi-
ties of ferroquine conjugates with either glutathione reductase
inhibitors or glutathione depletors via a hydrolyzable amide linker.
Bioorg. Med. Chem. 2009, 17, 8048–8059.
(14) Iley, J.; Constantino, L. Oxidation of tertiary benzamides by
5,10,15,20-tetraphenyl porphyrinatoironIII chloride-tert-butyl-
hydroperoxide. Org. Biomol. Chem. 2004, 2, 1894–1900.
(15) Bazin, M. J.; Shi, H.; Delaney, J.; Kline, B.; Zhu, Z.; Kuhn, C.;
Berlioz, F.; Farley, K. A.; Fate, G.; Lam, W.; Walker, G. S.; Yu, L.;
Pollastri, M. P. Efficient use of the iron ortho-nitrophenylporphy-
rin chloride to mimic biological oxidations of dimethylaminoanti-
pyrine. Chem. Biol. Drug Des. 2007, 70, 354–359.
(16) Bernadou, J.; Meunier, B. Biomimetic chemical catalysts in the
oxidative activation of drugs. Adv. Synth. Catal. 2004, 346, 171–
184.
Supporting Information Available: Glutathionylation rates of
saturated Mannich bases 6 and 7 as Michael acceptors, absorp-
tion spectra of CQ, FQ, and their related analogues 1-5 measu-
red in aqueous solution, IC50 and IC90 values of 4-aminoquino-
lines 1-5 against various P. falciparum strains, calculated and
measured pKa and log D values of some representatives in the
organic series, specific molar absorption coefficients and rela-
tive quantum yields of CQ, FQ, and 4-aminoquinolines 1-5
measured in aqueous solution, specific molar absorption coeffi-
cients of CQ, FQ, and 4-aminoquinolines 1-5 measured in
organic solvents, melting temperatures values of 50% and
100% AT-rich DNA evidencing their interaction with CQ,
FQ, and 4-aminoquinolines 1-5, and elemental analyses of
new compounds 2, 3, and 5. This material is available free of
charge via the Internet at http://pubs.acs.org.
(17) Meunier, B. Hybrid molecules with a dual mode of action: dream or
reality? Acc. Chem. Res. 2008, 41, 69–77.
(18) Meunier, B.; de Visser, S. P.; Shaik, S. Mechanism of oxidation
reactions catalyzed by cytochrome P450 enzymes. Chem. Rev. 2004,
104, 3947–3980.
(19) Irvin, J. L.; Irvin, E. M.; Parker, F. S. The interaction of anti-
malarials with nucleic acids. Science 1949, 110, 426–428.
(20) Meshnick, S. R. Chloroquine as intercalator: a hypothesis revived.
Parasitol. Today 1990, 6, 77–79.
(21) Ginsburg, H.; Nissani, E.; Krugliak, M.; Williamson, D. H.
Selective toxicity to malaria parasites by non-intercalating DNA-
binding ligands. Mol. Biochem. Parasitol. 1993, 58, 7–15.
(22) Parker, F. S.; Irvin, J. L. The interaction of chloroquine with
nucleic acids and nucleoproteins. J. Biol. Chem. 1952, 199, 897–909.
(23) Stollar, D.; Levine, L. Antibodies to denaturated deoxyribonucleic
acid in lupus erythematosus serum. V. Mechanism of DNA-anti-
DNA inhibition by chloroquine. Arch. Biochem. Biophys. 1963,
101, 335–341.
(24) Kurnick, N. B.; Radcliffe, I. E. Reaction between DNA and
quinacrine and other antimalarials. J. Lab. Clin. Med. . 1962, 60,
669–688.
(25) Bolton, P. H.; Mirau, P. A.; Shafer, R. H.; James, T. L. Interaction
of the antimalarial drug fluoroquine with DNA, tRNA, and poly-
(A): 19F-NMR chemical-shift and relaxation, optical absorption,
and fluorescence studies. Biopolymers 1981, 20, 435–449.
(26) Allison, J.-L.; O’Brien, R. L.; Hahn, F. E. DNA: reaction with
chloroquine. Science 1965, 149, 1111–1113.
References
(1) Neftel, K. A.; Woodtly, W.; Schmid, M.; Frick, P. G.; Fehr, J.
Amodiaquine induced agranulocytosis and liver damage. Br. Med.
J. 1986, 292, 721–723.
(2) Egan, T. J.; Mavuso, W. W.; Ncokazi, K. K. The mechanism of
beta-hematin formation in acetate solution. Parallels between
hemozoin formation and biomineralization processes. Biochemis-
try 2001, 40, 204–213.

3226 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 8
Wenzel et al.
(27) Cohen, S. N.; Yielding, K. L. Spectrophotometric studies of the
interaction of chloroquine with deoxyribonucleic acid. J. Biol.
Chem. 1965, 240, 3123–3131.
(28) Cheng, J.; Zeidan, R.; Mishra, S.; Liu, A.; Pun, S. H.; Kulkarni,
R. P.; Jensen, G. S.; Bellocq, N. C.; Davis, M. E. Structure-function
correlation of chloroquine and analogues as transgene expression
enhancers in nonviral gene delivery. J. Med. Chem. 2006, 49, 6522–
6531.
L; Elisei, F. Photophysical properties and photobiological beha-
vior of amodiaquine, primaquine and chloroquine. Photochem.
Photobiol. 2007, 83, 1415–1427.
(42) Aravind, L.; Iyer, L. M.; Wellems, T. E.; Miller, L. H. Plasmodium
biology: genomic gleanings. Cell 2003, 115, 771–785.
(43) Davioud-Charvet, E.; McLeish, M. J.; Veine, D. M.; Giegel, D.;
Arscott, L. D.; Andricopulo, A. D.; Becker, K.; Muller, S.;
Schirmer, R. H.; Williams, C. H., Jr.; Kenyon, G. L. Mechanism-
based inactivation of thioredoxin reductase from Plasmodium
falciparum by Mannich bases. Implication for cytotoxicity. Bio-
chemistry 2003, 42, 13319–13330.
(29) Pollack, Y.; Katzen, A. L.; Spira, D. T.; Golenser, J. The genome of
Plasmodium falciparum I: DNA base composition. Nucleic Acids
Res. 1982, 10 (2), 539–546.
(30) Kwakye-Berko, F.; Meshnick, S. R. Sequence preference of chloro-
quine binding to DNA and prevention of Z-DNA formation. Mol.
Biochem. Parasitol. 1990, 39, 275–278.
(44) Friebolin, W.; Jannack, B.; Wenzel, N.; Furrer, J.; Oeser,
T.; Sanchez, C. P.; Lanzer, M.; Yardley, V.; Becker, K.;
Davioud-Charvet, E. Antimalarial Dual Drugs Based on Potent
Inhibitors of Glutathione Reductase from Plasmodium falciparum.
J. Med. Chem. 2008, 51, 1260–1277.
(31) Kwakye-Berko, F.; Meshnick, S. R. Binding of chloroquine to
DNA. Mol. Biochem. Parasitol. 1989, 35, 51–56.
(32) Woynarowski, J. M.; Krugliak, M.; Ginsburg, H. Pharmacogeno-
mic analyses of targeting the AT-rich malaria parasite genome with
AT-specific alkylating drugs. Mol. Biochem. Parasitol. 2007, 154,
70–81.
(33) Baxter, E. W.; Reitz, A. B. Reductive aminations of carbonyl
compounds with borohydride and borane reducing agents. Organic
Reactions 2004, 59, 1–57.
(45) Perrine, D. M.; Sabanayagam, N. R.; Reynolds, K. J. Synthesis of
NMP, a fluoxetine (Prozac) precursor, in the introductory organic
laboratory. J. Chem. Educ. 1998, 75, 1266.
(46) Tsuji, T.; Mizuma, T.; Toyoshima, S. Chemotherapeutic drugs
against viruses. XXVII. Synthesis and antiviral activity of 1-(p-
alkylphenyl)-3-dimethylamino-1-propanol. Chem. Pharm. Bull.
1960, 8, 763–767.
(34) Glorian, G.; Maciejewski, L.; Brocard, J.; Agbossou, F. Enantio-
selective synthesis of (R)- and (S)-1-ferrocenylalkylamines. Reduc-
tion of enantiopure ferrocenylimines obtained from valinol and
phenylglycinol. Tetrahedron: Asymmetry 1997, 8, 355–358.
(35) Ncokazi, K. K.; Egan, T. E. A colorimetric high-throughput
β-hematin inhibition screening assay for use in the search for
antimalarial compounds. Anal. Biochem. 2005, 338, 306–319.
(36) Chavain, N.; Vezin, H.; Dive, D.; Touati, N.; Paul, J. F.; Buisine,
E.; Biot, C. Investigation of the redox behavior of ferroquine, a new
antimalarial. Mol. Pharm. 2008, 5, 710–716.
(37) Buisine, E.; de Villiers, K.; Egan, T. J.; Biot, C. Solvent-Induced
Effects: Self-Association of Positively Charged π Systems. J. Am.
Chem. Soc. 2006, 128, 12122–12128.
(38) Peters, W.; Robinson, B. L. Handbook of Animal Models of
Infection; Academic Press, London: 1999; pp 756-771.
´
(39) Rajapakse, C. S.; Martınez, A.; Naoulou, B.; Jarzecki, A. A.;
(47) Trager, W.; Jensen, J. B. Human malaria parasites in continuous
culture. Science 1976, 193, 673–675.
(48) Desjardins, R. E.; Canfield, C. J.; Haynes, J. D.; Chulay, J. D.
Quantitative assessment of antimalarial activity in vitro by a
semiautomated microdilution technique. Antimicrob. Agents
Chemother. 1979, 16, 710–718.
(49) Henry, M.; Briolant, S.; Zettor, A.; Pelleau, S.; Baragatti, M.;
Baret, E.; Mosnier, J.; Amalvict, R.;þFuþsai, T.; Rogier, C.;
Pradines, B. Plasmodium falciparum Na /H exchanger 1 trans-
porter is involved in reduced susceptibility to quinine. Antimicrob.
Agents Chemother. 2009, 53, 1926–1930.
(50) Parquet, V.; Briolant, S.; Torrentino-Madamet, M.; Henry, M.;
Almeras, L.; Amalvict, R.; Baret, E.; Fusai, T.; Rogier, C.;
Pradines, B. Atorvastatin is a promising partner for antimalarial
drugs in treatment of Plasmodium falciparum malaria. Antimicrob.
Agents Chemother. 2009, 53, 2248–2252.
(51) Herold-Mende, C.; Steiner, H. H.; Andl, T.; Riede, D.; Buttler,
A.; Reisser, C.; Fusenig, N. E.; Mueller, M. M. Expression
and functional significance of vascular endothelial growth factor
receptors in human tumor cells. Lab Invest. 1999, 79, 1573–
1582.
ꢀ
ꢀ
Suarez, L.; Deregnaucourt, C.; Sinou, V.; Schrevel, J.; Musi, E.;
Ambrosini, G.; Schwartz, G. K.; Sanchez-Delgado, R. A. Synthe-
ꢀ
sis, characterization, and in vitro antimalarial and antitumor
activity of new ruthenium(II) complexes of chloroquine. Inorg.
Chem. 2009, 48, 1122–1131.
(40) Remy, J. S.; Sirlin, C.; Vierling, P.; Behr, J. P. Gene transfer with a
series of lipophilic DNA-binding molecules. Bioconjugate Chem.
1994, 5, 647–654.
(52) Biot, C.; Taramelli, D.; Forfar-Bares, I.; Maciejewski, L. A.;
Boyce, M.; Nowogrocki, G.; Brocard, J. S.; Basilico, N.; Olliaro,
P.; Egan, T. J. Insights into the mechanism of action of ferroquine.
Relationship between physicochemical properties and antiplasmo-
dial activity. Mol. Pharmaceutics 2005, 2, 185–193.
(41) Viola, G; Salvador, A; Cecconet, L; Basso, G; Vedaldi, D;
Dall’Acqua, F; Aloisi, G. G.; Amelia, M; Barbafina, A; Latterini,