Effects of highly active novel artemisinin-chloroquinoline hybrid compounds on β-hematin formation, parasite morphology and endocytosis in Plasmodium falciparum

Article abstract of DOI:10.1016/j.bcp.2011.04.018

4-Aminoquinolines were hybridized with artemisinin and 1,4-naphthoquinone derivatives via the Ugi-four-component condensation reaction, and their biological activities investigated. The artemisinin-containing compounds 6a-c and its salt 6c-citrate were the most active target compounds in the antiplasmodial assays. However, despite the potent in vitro activities, they also displayed cytotoxicity against a mammalian cell-line, and had lower therapeutic indices than chloroquine. Morphological changes in parasites treated with these artemisinin-containing hybrid compounds were similar to those observed after addition of artemisinin. These hybrid compounds appeared to share mechanism(s) of action with both chloroquine and artemisinin: they exhibited potent β-hematin inhibitory activities; they caused an increase in accumulation of hemoglobin within the parasites that was intermediate between the increase observed with artesunate and chloroquine; and they also appeared to inhibit endocytosis as suggested by the decrease in the number of transport vesicles in the parasites. No cross-resistance with chloroquine was observed for these hybrid compounds, despite the fact that they contained the chloroquinoline moiety. The hybridization strategy therefore appeared to be borrowing the best from both classes of antimalarials.

Full text of DOI:10.1016/j.bcp.2011.04.018

Biochemical Pharmacology 82 (2011) 236–247
Contents lists available at ScienceDirect
Biochemical Pharmacology
journal homepage: www.elsevier.com/locate/biochempharm
Effects of highly active novel artemisinin–chloroquinoline hybrid compounds on
b
-hematin formation, parasite morphology and endocytosis in Plasmodium
falciparum
Tzu-Shean Feng a,b, Eric M. Guantai a,c, Margo Nell d, Constance E.J. van Rensburg d, Kanyile Ncokazi a,
a,e,
b,**
*
Timothy J. Egan a, Heinrich C. Hoppe
, Kelly Chibale
a Department of Chemistry, University of Cape Town, Rondebosch 7701, South Africa
b Division of Pharmacology, University of Cape Town, Observatory 7925, South Africa
c Division of Pharmacology, School of Pharmacy, University of Nairobi, Box 19676-00202, Kenya
d Department of Pharmacology, University of Pretoria, Hatfield 0028, South Africa
e Institute of Infectious Disease and Molecular Medicine, University of Cape Town, Rondebosch 7701, South Africa
A R T I C L E I N F O
A B S T R A C T
Article history:
4-Aminoquinolines were hybridized with artemisinin and 1,4-naphthoquinone derivatives via the Ugi-
four-component condensation reaction, and their biological activities investigated. The artemisinin-
containing compounds 6a–c and its salt 6c-citrate were the most active target compounds in the
antiplasmodial assays. However, despite the potent in vitro activities, they also displayed cytotoxicity
against a mammalian cell-line, and had lower therapeutic indices than chloroquine. Morphological
changes in parasites treated with these artemisinin-containing hybrid compounds were similar to those
observed after addition of artemisinin. These hybrid compounds appeared to share mechanism(s) of
Received 5 April 2011
Accepted 29 April 2011
Available online 8 May 2011
Keywords:
Artemisinin
b
-Hematin inhibition
action with both chloroquine and artemisinin: they exhibited potent
b-hematin inhibitory activities;
Chloroquinoline
they caused an increase in accumulation of hemoglobin within the parasites that was intermediate
between the increase observed with artesunate and chloroquine; and they also appeared to inhibit
endocytosis as suggested by the decrease in the number of transport vesicles in the parasites. No cross-
resistance with chloroquine was observed for these hybrid compounds, despite the fact that they
contained the chloroquinoline moiety. The hybridization strategy therefore appeared to be borrowing the
best from both classes of antimalarials.
Endocytosis
Parasite morphology
Plasmodium falciparum
ß 2011 Elsevier Inc. All rights reserved.
1. Introduction
century this species has developed significant degrees of resistance
to many classes of drugs [3,5,6]. Of particular concern is the recent
Malaria remains one of the most widespread infectious diseases
and poses a great challenge to world health. Over 200 million
episodes of malaria occur annually, occasioning approximately 1
million deaths every year, mainly among children and pregnant
women in sub-Saharan Africa [1–4]. Resistance of malaria
parasites to available antimalarial drugs remains the main
challenge to the effective control of the disease. Plasmodium
falciparum is the predominant Plasmodium species in Africa, the
most virulent species of the malaria parasite and the cause of the
bulk of the mortality associated with malaria. Over the last half a
evidence of the possible emergence of resistance to artemisinin-
based antimalarials in Southeast Asia, a situation that could
threaten the current clinical efficacy of the first-line artemisinin-
combination therapy (ACT) strategies [7,8].
This has led to extensive drug discovery efforts aimed at
developing novel antimalarial compounds or modifying existing
agents in the hope of identifying new, highly efficacious drugs to
supplement available drugs for the treatment of malaria [9]. As
part of these efforts, we undertook to synthesize a variety of hybrid
compounds combining features of artemisinin, 1,4-naphthoqui-
none and chloroquine for biological investigation. Artemisinins
and chloroquine are established antimalarial drugs, and their
antimalarial activities have been extensively studied and exploited
both experimentally and clinically [10–12]. In addition, the
precedent already exists for the fusion of artemisinin and
aminoquinoline components to yield highly potent hybrid
compounds [13,14].
* Corresponding author at: Department of Chemistry, University of Cape Town,
Rondebosch 7701, South Africa. Tel.: +27 21 650 2553; fax: +27 21 650 5195.
** Corresponding author. Tel.: +27 12 841 4363; fax: +27 12 841 4790.
E-mail addresses: HHoppe@csir.co.za (H.C. Hoppe), Kelly.Chibale@uct.ac.za
(K. Chibale).
0006-2952/$ – see front matter ß 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.bcp.2011.04.018

T.-S. Feng et al. / Biochemical Pharmacology 82 (2011) 236–247
237
Fig.
3 shows the chemical structures of the target hybrid
compounds 6a–e.
Following their synthesis, the target compounds were assessed
for their biological activities, including in vitro antiplasmodial
activity, inhibition of
b-hematin formation, effects on parasite
morphology, and effects on hemoglobin endocytosis by the parasite.
The latter experiments were motivated by the reported differential
activitiesofquinolines andartemisininonendocytosisandendocytic
trafficking in malaria parasites [24]. Chloroquine is thought to
interfere with the parasite endocytic pathway by blocking the fusion
of hemoglobin transport vesicles to the digestive food vacuole,
resulting in an accumulation of undigested hemoglobin in transport
vesicles. By contrast, artemisinin does not inhibit transport vesicle-
vacuole fusion, but rather the preceding endocytosis of hemoglobin
and formation of the transport vesicles, while there are additional
indications of an inhibition of the proteolytic digestion of hemoglo-
bin inside the food vacuole [24]. The biological effects of the test
compounds were used to obtain a preliminary indication of their
Fig. 1. Examples of 1,4-naphthaquinone inhibitors of glutathione reductase.
1,4-Naphthoquinones, on the other hand, are an important class
of inhibitors of glutathione (and thioredoxin) reductase, with
plumbagin, menadione and its hexanoic acid derivative M5 being
representative examples (Fig. 1) [15,16]. Glutathione reductase is
the enzyme responsible for regenerating the reduced form of
glutathione, a major (di)thiol responsible for the prevention of
oxidative damage in both human cells and the malaria parasite by
the detoxification of reactive oxygen radicals [17]. Glutathione has
also been suggested to be involved in the degradation of toxic
heme formed by the malaria parasite during hemoglobin digestion
[17], though the main route of detoxification has been shown to be
mainly via the conversion of heme to hemozoin (malaria pigment),
an inert, crystalline form of heme [18,19]. Double headed prodrugs
and other hybrid compounds based on 1,4-naphthoquinones and
4-aminoquinolines have previously been developed and proven to
be potent antimalarials in vitro and in vivo, exerting their effects by
inhibiting glutathione reductase [16,20,21].
1,4-Naphthoquinone and artemisinin were therefore chosen for
hybridization to 4-aminoquinolines to afford the target hybrid
compounds 6a–e (Fig. 3), using the Ugi four-component conden-
sation (Ugi-4CC) reaction [22] as the hybridization strategy. This
reaction is an example of a multi-component reaction (MCR), i.e.
chemical reactions that utilize three or more different starting
materials as chemical inputs in a one-step-one-pot synthesis to
yield one single product that contains features of all inputs. Such
reactions are highly efficient as they create molecular complexity
by generating more than two chemical bonds per operation, and
are widely used to study structure-activity relationships and also
to create libraries of organic molecules [23]. The Ugi-4CC utilizes
an aldehyde, an amine, an acid and an isocyanide, and converts
`
modes of action vis-a-vis chloroquine and artemisinin.
2. Materials and methods
Chemicals and reagents were purchased from either Sigma-
Aldrich or Merck, South Africa. Chromatography solvents were
purchased from Kimix Chemicals or Protea Chemicals, South
Africa, as Chemically Pure (CP grade) solvents and distilled before
use.
2.1. Chemistry
The Ugi-4CC utilizes an acid, an amine, an aldehyde and an
isocyanide, and converts them into an
a-acetamidoamide in one
step. The target Ugi adducts were synthesized using either artelinic
acid 2 or 1,4-naphthoquinone-acid 3 as the acid component; either
of the 4-aminoquinoline amines 4 or 5 as the amine component;
paraformaldehyde as the aldehyde input; and cyclohexyl- and tert-
butyl isocyanide as the isocyanide component (Scheme 1).
2.1.1. Methyl p-[(10-dihydroartemisininoxy)methyl]benzoate (1)
Dihydroartemisinin (1.00 g, 3.500 mmol, 1.0 eq) and 4-(hydro-
xylmethyl)-benzoate (2.036 g, 12.006 mmol, 3.43 eq) were stirred
in anhydrous Et2O under N2, where BF3ÁEt2O (0.50 ml, 3.955 mmol,
1.13 eq) was added drop-wise. The solution was stirred under N2 at
room temperature for 23 h, washed with 5% NaHCO3 (until pH 8),
H2O (2 Â 100 ml), dried (Na2SO4) and solvent evaporated. The crude
product was purified via column chromatography (EtOAc/Hex, 1:9)
them into an
yields.
a-acetamidoamide in one step and usually in good
Based on its versatility, the Ugi-4CC was selected as the
chemical approach for the synthesis of the target hybrid
compounds combining features of artemisinin, 1,4-naphthoqui-
none and chloroquine (Fig. 2). The synthesis and results from the
biological evaluation of the target compounds are reported herein.
to yield the pure product 1 as colourless oil (1.460 g, 97% (both
a-
and -isomers)). Rf (EtOAc/Hex 1:9) 0.21. -Isomer: H (400 MHz;
b
b
d
CDCl3) 8.01 (d, J 8.4, 2H, H19), 7.37 (d, J 8.4, 2H, H18), 5.44 (s, 1H,
H12), 4.95 (d, J 13.2, 1H, H17), 4.91 (d, J 3.2, 1H, H10), 4.57 (d, J 13.2,
1H, H17), 3.91 (s, 3H, OCH3), 2.69 (m, 1H, H9), 2.36 (ddd, J 14.0, 13.2
and 4.0, 1H, not known1), 2.20–1.50 (m, 4H, not known1), 1.45 (s, 3H,
H14), 0.96 (d, J 7.6, 3H, H15), 0.94 (d, J 6.4, 3H, H16). a-Isomer: dH
(400 MHz; CDCl3) 8.00 (d, J 8.4, 2H, H19), 7.43 (d, J 8.8, 2H, H18), 5.34
(s, 1H,H12),5.02(d,J14.2, 1H, H17), 4.68(d,J13.2, 1H, H17),4.52(d, J
9.2, 1H, H10), 3.91 (s, 3H, OCH3), 2.53 (m, 1H, H9), 2.39 (ddd, J 14.4,
13.2 and 4.0, 1H, not known1), 2.20–1.50 (m, 4H, not known1), 1.45
(s, 3H, H14), 0.93 (d, J 5.6, 3H, H16), 0.93 (d, J 7.2, 3H, H15).
2.1.2. 2.1.2 p-[(10-Dihydroartemisininoxy)methyl]benzoic acid
(Artelinic Acid) (2)
The methyl ether 1 (1.272 g, 2.932 mmol) was stirred in 2.5%
KOH/MeOH (25 ml) at room temperature for 3 days, and the
solvent removed under reduced pressure. The residue was
1
Fig. 2. General structure showing the design of the target compounds.
Protons could not be assigned due to complexity.

238
T.-S. Feng et al. / Biochemical Pharmacology 82 (2011) 236–247
Scheme 1. Reagents and conditions: (i) methyl 4-(hydroxymethyl)-benzoate (3.4 eq), BF3ÁEt3O (1.13 eq), Et3O, N2, r.t., 24 h; (ii) 2.5% KOH/MeOH, r.t., 3 days; (iii) succinic acid
(3.0 eq), silver nitrate (0.5 eq), ammonium persulfate (1.3eq), 30% aq. CH3N, 65–70 8C, 3 h.; (iv) 2 or 3 (1.0 eq), amine 4 or 5 (1.2 eq), aldehyde (1.0 eq), cyclohexyl- or tert-butyl
isocyanide (1.0 eq), MeOH, r.t., 3–5 days.
dissolved in H2O (25 ml), and the pH adjusted to 8 via addition of
acetic acid. The solution was extracted with Et2O (3 Â 50 ml), dried
(Na2SO4) and solvent evaporated. The resulting oil was seeded and
the pure compound crystallized out to yield white crystals
(1.091 g, 89%). m.p. 141–144 8C (lit. 142–145 8C) [25]. Rf (EtOAc/
precipitated on cooling were filtered and washed with brine and
H2O. For products that stayed in solution, NaOH was added, and the
organic product was immediately basified with 10% NaOH (w/v)
and extracted with EtOAc (5 times), washed with H2O (2 times)
and dried (Na2SO4). Recrystallization from EtOAc afforded the
amine in excellent yields.
Hex 3.3:6.7) 0.18. b-Isomer: dH (300 MHz; CDCl3) 8.01 (d, J 8.4, 2H,
H19), 7.40 (d, J 8.4, 2H, H18), 5.45 (s, 1H, H12), 4.98 (d, J 13.5, 1H,
H17), 4.93 (d, J 3.0, 1H, H10), 4.60 (d, J 13.5, 1H, H17), 2.70 (m, 1H,
H9), 2.38 (ddd, J 14.0, 13.2 and 4.0, 1H, not known1), 2.20–1.50 (m,
4H, not known1), 1.45 (s, 3H, H14), 0.98 (d, J 7.2, 3H, H15), 0.95 (d, J
2.1.5. N-(7-Chloroquinolin-4-yl)-ethane-1,2-diamine (4)
Pure product was recrystallized from EtOAc as white powder
(0.4861 g, 64%). m.p. 136–138 8C (lit. 137–139 8C) [27]. Rf (MeOH/
CH2Cl2, 0.5:9.5) 0.32. dH (400 MHz; DMSO-d6) 8.38 (d, J 5.6, 1H,
H2), 8.10 (d, J 8.4 Hz, 1H, H5), 7.76 (d, J 2.0, 1H, H8), 7.40 (dd, J 8.4
and 2.0, 1H, H6), 7.24 (br s, 1H, NH), 6.48 (d, J 5.6, 1H, H3), 3.25 (t, J
5.6, 2H, H9), 2.82 (t, J 6.0, 2H, H10).
6.0, 3H, H16).
a-Isomer: dH (300 MHz; CDCl3) 8.06 (d, J 8.4, 2H,
H19), 7.46 (d, J 8.4, 2H, H18), 5.35 (s, 1H, H12), 5.05 (d, J 13.5, 1H,
H17), 4.70 (d, J 13.2, 1H, H17), 4.53 (d, J 9.0, 1H, H10), 2.55 (m, 1H,
H9), 2.39 (ddd, J 14.4, 13.2 and 4.2, 1H, not known1), 2.20–1.50 (m,
4H, not known1), 1.46 (s, 3H, H14), 0.95 (d, J 6.0, 3H, H16), 0.94 (d, J
7.2, 3H, H15).
2.1.6. N-(7-Chloroquinolin-4-yl)-butane-1,4-diamine (5)
The
b
-isomer was the major product (diastereomeric ratio
a
:
b
Pure product was recrystallized from EtOAc as a light-yellow
solid (1.60 g, 87%). m.p. 43–46 8C (lit. 43–47 8C) [27]. Rf (NH3/
2:98) as confirmed by the integrations on 1H NMR spectroscopy.
MeOH, 1:49) 0.22. dH (300 MHz; CD3OD) 8.48 (d, J 5.7, 1H, H2), 7.96
2.1.3. 2.1.3 3-Carboxyethyl-5-hydroxy-2-methyl-1,4-
(d, J 9.0 Hz, 1H, H5), 7.69 (d, J 2.1, 1H, H8), 7.27 (dd, J 9.0 and 2.1, 1H,
H6), 6.34 (d, J 5.7, 1H, H3), 3.40 (t, J 7.2, 2H, H9), 3.04 (t, J 7.2, 2H,
H12), 1.88 (m, 2H, H10), 1.62 (m, 2H, H11).
naphthoquinone (3)
Plumbagin (1.00 g, 5.314 mmol, 1.0 eq) and succinic acid
(1.902 g, 15.942 mmol, 3.0 eq) were stirred in 30% aqueous
acetonitrile (40 ml) at 65 8C, followed by the addition of silver
nitrate (0.456 g, 2.657 mmol, 0.5 eq). A solution of ammonium
persulfate (1.609 g, 6.908 mmol, 1.3 eq) in 30% aqueous acetoni-
trile (20 ml) was added drop-wise over 2 h, and the resulting
solution stirred at 65–70 8C for 3 h. On cooling, the mixture was
extracted with EtOAc (4 Â 60 ml), and the organic layer washed
with H2O (3 Â 20 ml), dried (Na2SO4) and solvent evaporated. The
crude material was purified via column chromatography (DCM/
MeOH 9.5:0.5) to yield the pure product as an orange powder
(0.581 g, 58%). m.p. 155–159 8C [26]. Rf (CH2Cl2/MeOH, 9.5:0.5)
2.1.7. Ugi adduct (6a)
Purified via column chromatography, first with EtOAC and then
with EtOAC/MeOH 9.5:0.5 to yield the pure product as a white
powder (0.5115 g, 89%). (m.p. 121–124 8C. Rf (MeOH/EtOAC, 1:9)
0.18; nmax (CHCl3)/cmÀ1 3223 (N–H), 2920 (Ar–H), 1580 (C55O),
1216 (C–N), 781 (C–Cl); dH (400 MHz; CDCl3) 8.45 (br s, 1H, H24),
8.28 (br s, 1H, H26), 7.96 (br s, 2H, H19), 7.41 (d, J 8.8, 2H, H18),
7.39 (br s, 1H, H27), 7.03 (br s, 1H, H27), 6.35 (br s, 1H, H25), 5.42 (s,
1H, H12), 4.88 (br s, 2H, H17 + H10), 4.51 (br s, 1H, H17), 4.09 (m,
2H, H22), 3.61 (br s, 2H, H23), 2.68 (m, 1H, H9), 2.38 (ddd, J 14.4 and
13.2, 4.2 Hz, 1H, not known), 1.45–1.93 (m, 15H), 1.44 (s, 3H, H14),
0.95 (d, J 6.0, 6H, H16 + H15); dc (75.5 MHz; DMSO-d6) 171.6,
171.0, 170.2, 167.5, 151.9, 151.5, 150.1, 149.5, 149.1, 139.9, 139.2,
135.2, 134.9, 133.4, 127.5, 126.6, 126.3, 124.1, 123.7, 123.5, 117.5,
103.3, 100.6, 98.6, 87.0, 80.4, 52.0, 47.7, 36., 34.0, 32.1, 30.5, 25.6,
25.1, 24.4, 24.2, 20.0, 12.7; (Found: m/z, 761.3596 [M+],
0.49. dH (400 MHz; CDCl3) 7.62 (d, J 6.8, 1H, H8), 7.57 (t, J 7.6, 1H,
H7), 7.23 (d, J 7.6, H6), 2.98 (t, J 7.6, 2H, H11), 2.62 (t, J 7.6, 2H, H10),
2.23 (s, 3H, CH3).
2.1.4. General method A for the preparation of 4 and 5
Et3N (0.3 eq) was added drop-wise to a stirred solution of 4,7-
dichloroquinoline (1.0 eq) and K2CO3 (0.5 eq) in respective
diamines (20.0 eq) and refluxed at 110 8C for 4 h. Products that
C42H53ClN4O7 requires M, 761.3584) (Found: C, 66.00, H, 7.26, N,
7.67%; Requires C, 66.26, H, 7.02 N, 7.36%).

T.-S. Feng et al. / Biochemical Pharmacology 82 (2011) 236–247
239
2.1.8. Ugi adduct (6b)
1H, H14), 7.88 (d, J 1.8, 1H, H14), 7.88 (d, J 9.0, 1H, H12), 7.74 (d, J
9.0, 1H, H12), 7.55 (m, 4H, (H2 + H3) Â 2), 7.35 (dd, J 9.0 and 1.8,
1H, H13), 7.31 (dd, J 9.0 and 1.8, 1H, H13), 7.13 (d, J 7.5, 1H, H1),
7.13 (d, J 7.5, 1H, H1), 6.97 (br s, 1H, NH), 6.34 (d, J 5.4, 1H, H11),
6.33 (d, J 5.7, 1H, H11), 4.09 (s, 2H, H7),4.00 (s, 2H, H7), 3.88 (m, 2H,
H8), 3.70 (br s, 4H, H8 + H9), 3.50 (m, 2H, H9), 2.97 (t, J 7.8, 2H, H6),
2.72 (t, J 8.1, 2H, H5), 2.57 (t, J 7.8, 2H, H6), 2.28 (t, J 8.1, 2H, H5),
2.24 (s, 3H, H4), 1.92 (s, 3H, H4), 1.75 (m, 11H, cyclohexyl-H), 1.22
(m, 11H, cyclohexyl-H); dc (100 MHz; CDCl3) 174.9, 161.2, 151.6,
151.0, 137.3, 136.8, 136.2, 136.1, 132.0, 125.9, 125.8, 125.7, 123.9,
122.2, 122.0, 119.2, 119.1, 98.6, 98.0, 53.1, 43.0, 41.0, 32.9, 32.2,
31.5, 25.5, 25.2, 24.8, 24.7, 22.7, 22.4, 12.9; (Found: m/z 603.1182
[M+], C33H35ClN4O5 requires M, 603.1177) (Found: C, 66.67, H,
7.07, N, 7.29%; Requires C, 66.95, H, 7.28 N, 7.10%).
Purified via column chromatography, first with EtOAc and then
with EtOAc/MeOH 9.5:0.5 to yield the pure product as a white
powder (0.5009 g, 84%). (m.p. 121–123 8C. Rf (MeOH/EtOAc, 1:9)
0.48; nmax (CHCl3)/cmÀ1 3333 (N–H), 2926 (Ar–H), 2894 (alkyls),
1666 (C55O), 1580 (Ar–C), 1010 (C–N), 753 (C–Cl); dH (400 MHz;
CDCl3) 8.45 (d, J 5.2, 1H, H24), 8.23 (br s, 1H, H29), 7.99 (s, 2H, H31),
7.36 (br s, 5H, H30 + H18 + H19), 7.13 (br s, 1H, NH), 6.31 (br s, 1H,
H28), 6.10 (br s, 1H, NH), 5.44 (s, 1H, H12), 4.91 (d, J 12.4, 1H, H17),
4.90 (d, J 4.0, 1H, H10), 4.56 (d, J 12.4, 1H, H17), 4.11 (m, 2H, CH2,
H22), 3.81 (br s, 2H, CH2, H25), 3.52 (br s, 2H, CH2, H23), 3.48 (s, 2H,
H20), 3.26 (br s, 2H, CH2, H24), 2.52 (m, 1H, H9), 2.00–1.45 (m,
20H), 1.45 (s, 3H, H14), 0.96 (d, J 7.2, 3H, H15), 0.95 (d, J 5.6, 3H,
H16); dc (100 MHz; DMSO-d6) 170.9, 167.1, 151.8, 150.0, 149.0,
139.5, 135.7, 132.3, 127.4, 126.8, 126.6, 126.5, 124.4, 124.0, 123.9,
117.4, 103.3, 100.6, 98.6, 87.0, 80.4, 68.7, 52.0, 47.6, 45.8, 43.7,
42.3, 36.6, 36.0, 34., 32.4, 32.1, 30.5, 25.6, 24.7, 24.4, 20.0, 12.7;
(Found: m/z 789.4125 [M+], C44H57ClN4O7 requires M, 789.4122)
(Found: C, 66.80, H, 6.91, N, 6.92%; Requires C, 66.95, H, 7.28 N,
7.10%).
2.1.12. N-(2-(tert-Butylamino)-2-oxoethyl)-N-(2-(7-chloroquinolin-
4-ylamino)ethyl)-3-(3-methyl-1,4-dioxo-1,4-dihydronaphthalen-2-
yl)propanamide (6e)
Purified via column chromatography, first with EtOAc and then
with EtOAc/MeOH 9.5:0.5 to yield the pure product as an orange
powder (0.123 g, 28%). m.p. 205–207 8C; Rf (EtOAc) 0.11; nmax
(CHCl3)/cmÀ1 3320 (N–H), 3026 (Ar–H), 1736 (C55O), 1430 (Ar–C),
1216 (C–N); 753 (C–Cl); dH (400 MHz; CDCl3) 8.35 (d, J 5.2, 1H,
H10), 8.34 (d, J 5.2, 1H, H10), 7.90 (d, J 9.2, 1H, H12), 7.85 (d, J 2.0,
1H, H14), 7.84 (d, J 9.2, 1H, H12), 7.79 (d, J 2.0, 1H, H14), 7.51 (m,
4H, (H2 + H3) Â 2), 7.29 (dd, J 9.2 and 2.0, 1H, H13), 7.28 (dd, J 9.2
and 2.0, 1H, H13), 7.15 (d, J 7.2, 1H, H1), 7.14 (d, J 7.2, 1H, H1), 6.36
(d, J 5.6, 1H, H11), 6.35 (d, J 5.6, 1H, H11), 3.99 (s, 2H, H7), 3.79 (s,
2H, H7), 3.77 (d, J 5.6, 2H, H8), 3.66 (d, J 5.6, 2H, H9), 3.63 (d, J 5.6,
2H, H8), 3.49 (d, J 5.6, 2H, H9), 2.91 (t, J 8.0, 2H, H6), 2.72 (t, J 8.0, 2H,
H6), 2.51 (t, J 8.0, 2H, H5), 2.30 (t, J 8.0, 2H, H5, 2.19 (s, 3H, H4), 1.93
2.1.9. Ugi adduct (6c)
Purified via column chromatography, first with EtOAc and then
with EtOAc/MeOH 9.5:0.5 to yield the pure product as a white
powder (0.453 g, 82%). (m.p. 127–129 8C. Rf (MeOH/EtOAc, 1:9)
0.18; nmax (CHCl3)/cmÀ1 3385 (N–H), 3017 (Ar–H), 1520 (C55O),
1215 (C–N), 756 (C–Cl); dH (400 MHz; CDCl3) 8.47 (br s, 1H, H24),
8.24 (br s, 1H, H26), 8.12 (d, J 8.6, 2H, H19), 8.00 (s, 1H, H28), 7.40
(d, J 8.6, 2H, H18), 7.39 (br s, 1H, H27), 7.04 (br s, 1H, NH), 6.35 (br s,
1H, H25), 5.42 (s, 1H, H12), 4.87 (br s, 2H, H17 + H10), 4.51 (br s,
1H, H17), 3.93 (br s, 2H, H22), 3.64 (br s, 2H, H23), 2.68 (m, 1H, H9),
2.38 (ddd, J 14.4, 13.2 and 4.2, 1H, not known), 2.10–1.57 (m, 4H),
1.44 (s, 3H, H14), 1.35 (s, 9H, t-butyl), 0.95 (d, J 6.0, 6H, H16 + H15);
(s, 3H, H4), 1.37 (s, 9H, t-butyl), 1.27 (s, 9H, t-butyl); dc (100 MHz;
CDCl3) 151.6, 150.3, 145.6, 136.1, 127.2, 126.4, 125.6, 123.7, 123.6,
122.5, 122.4, 119.0, 98.9, 98.2, 52.7, 52.0, 49.4, 14.2, 49.0, 48.7,
48.5, 47.7, 42.1, 40.8, 32.0, 31.4, 28.4, 22.5, 22.3, 12.6; (Found: m/z
577.0805 [M+], C31H33ClN4O5 requires M, 577.0798) (Found: C,
66.80, H, 7.39, N, 7.48%; Requires C, 66.95, H, 7.28 N, 7.10%).
d
c (100 MHz; CDCl3) 129.9, 127.2, 126.8, 125.9, 104.2, 104.2, 101.6,
101.6, 98.2, 88.1, 81.1, 70.7, 69.5, 69.1, 52.7, 52.6, 52.1, 44.5, 44.4,
37.5, 36.4, 34.6, 31.0, 30.9, 29.1, 28.7, 26.2, 24.7, 24.6, 20.3, 13.1;
(Found: m/z 735.3211 [M+], C40H51ClN4O7 requires M, 735.3205)
(Found: C, 65.68, H, 7.28, N, 7.56%; Requires C, 65.34, H, 6.99 N,
7.62%).
2.2. Biological evaluation
2.1.10. Citrate salt of Ugi adduct 6c (6c-citrate)
2.2.1. Parasite cultivation
Citric acid (0.131 g, 0.680 mmol, 2.5 eq) in acetone (5 ml) was
added to a solution of 6c (0.200 g, 0.272 mmol, 1.0 eq) in acetone
(5 ml). The product precipitated out over 2 days at 4 8C to yield
D10 and K1 strains of P. falciparum were maintained in
continuous in vitro culture by standard methods [28]. The cultured
parasites were, whenever necessary, synchronized by treatment
white solid (0.247 g, 98%).
3317 (N–H), 3017 (Ar–H), 2962 (OH (acid)), 1707 (C55O), 1612 (Ar–
H), 1215 (C–N), 1007 (C–O), 752 (C–Cl); H (300 MHz; CD3OD) 8.51
n
max (CHCl3)/cmÀ1 3419 (OH (alcohol)),
with 5% D-sorbitol (Sigma) at the ring stage [29].
d
2.2.2. Evaluation of in vitro antiplasmodial activity
The in vitro antiplasmodial assays were carried out against D10
and K1 strains of P. falciparum using the parasite lactate
dehydrogenase assay, as previously reported by Guantai et al.
2010 [30].
(br s, 1H, H24), 8.43 (br s, 1H, H26), 7.91 (br s, 1H, H28), 7.61 (br s,
1H, H27), 7.39 (br s, 2H, H19), 6.99 (br s, 2H, H18), 6.42 (br s, 1H,
H25), 5.44 (s, 1H, H12), 4.04 (br s, 2H, H22), 3.95 (br s, 2H, H23),
2.58 (m, 1H, H9), 2.33 (td, J 14.4 and 13.0, 4.2 Hz, 1H, not known),
2.20–1.50 (m, 4H, not known), 1.44 (br s, 4H, H29 + H30), 1.37 (s,
3H, H14), 1.32 (s, 9H, t-butyl), 0.94 (d, J 6.0, 6H, H16 + H15);
(Found:, m/z 927.4471 [M+], C40H51N4ClO7 requires M, 927.4459)
(Found: C, 59.76, H, 6.63, N, 5.74%; Requires C, 59.57, H, 6.41, N,
6.04%).
2.2.3. Evaluation of
b-hematin inhibition
b-hematin inhibition assays were conducted as previously
described by Ncokazi et al. 2005 [31].
2.2.4. Cytotoxicity assays
2.1.11. N-(2-(7-Chloroquinolin-4-ylamino)ethyl)-N-(2-
Cytotoxicity assays were carried out on a human cervical
carcinoma cell line (HeLa) using the MTT [3-(4,5-dimethylthiazol-
2-yl)-2,5-diphenyltetrazolium bromide] assay. Briefly, cell cul-
tures were routinely maintained in Eagle’s Minimum Essential
Medium (EMEM) supplemented with 10% bovine foetal calf serum
(cyclohexylamino)-2-oxoethyl)-3-(3-methyl-1,4-dioxo-1,4-
dihydronaphthalen-2-yl)propanamide (6d)
Purified via column chromatography, first with EtOAc and then
with EtOAc/MeOH 9.5:0.5 to yield the pure product as an orange
powder (0.1138 g, 25%). (m.p. 135–138 8C. Rf (MeOH/EtOAc, 1:9)
0.11; nmax (CHCl3)/cmÀ1 3273 (N–H), 3020 (Ar–H), 2926 (alkyls),
1606 (C55O), 1430 (Ar–C), 1216 (C–N), 770 (C–Cl); dH (300 MHz;
CDCl3) 8.48 (d, J 5.7, 1H, H10), 8.44 (d, J 5.4, 1H, H10), 7.96 (d, J 1.8,
(FCS). For cytotoxicity evaluations, 5 Â 102 cells in 180
ml medium
were seeded in each well of a 96-well plate and incubated at 37 8C
under a 5% CO2 atmosphere for 1 h. Aliquots of 20
ml of serial
dilutions of the test compounds (stocks in DMSO) were added to

240
T.-S. Feng et al. / Biochemical Pharmacology 82 (2011) 236–247
Fig. 3. Target compounds synthesized via the Ugi 4-CC.
the wells. Untreated control wells received 20
m
l of culture
(SDS-PAGE) and Western blotting with anti-hemoglobin antise-
rum. To ensure that the different intensities of the hemoglobin
bands on the resultant blots were not the result of uneven loading
of parasite pellets, gels were also stained with Coomassie dye to
visualize protein bands in the gels. No variations in the intensities
of the major non-globin proteins were observed in the different
samples. Net hemoglobin band intensities on the blots were
determined with Kodak 1D image analysis software (version 3.5).
For the immunofluorescence assays, the drug-treated saponin-
released parasites were fixed onto poly-lysine-coated glass cover
slips with paraformaldehyde-glutaraldehyde, after which they
were treated with Triton to enhance their permeability. Parasite-
associated hemoglobin was labelled by incubation with anti-Hb
antiserum and tetramethylrhodamine-5-isothiocyanate (TRITC)-
conjugated secondary antibodies. The cover slips were viewed by
fluorescence microscopy and images captured with a CoolSNAP-
Pro CCD camera.
medium, while additional solvent controls were prepared with
medium containing DMSO to account for any possible effects of
DMSO on cell viability. The plates were then incubated at 37 8C
under an atmosphere of 5% CO2 for seven days.
After the incubation period, 20
ml MTT (5 mg/ml) was added to
each well. The plates were further incubated for an additional 4 h
at 37 8C under an atmosphere of 5% CO2, and then centrifuged for
10 minutes at 800 G. The supernatant was carefully aspirated from
each well without disturbing the pellet, and the cells were washed
by addition of 150
ml of phosphate-buffered saline (PBS) followed
by centrifugation for 10 minutes at 800 G. The supernatant was
again carefully aspirated, and the plates were left to dry off at 37 8C
for an hour. 100
ml ethanol was added to each well to solubilize the
resultant formazan crystals, aided by gentle mechanical shaking
for 1–2 h. Absorbances were measured on a Universal Microplate
Reader (ELx800 UV, Bio-tek Instruments) at a wavelength of
570 nm and used to calculate percentage cell growth in drug-
treated wells; these in turn were plotted versus log drug
concentration and used to determine the corresponding IC50
values by non-linear regression analysis.
3. Results and discussion
3.1. Antiplasmodial activity and cytotoxicity
2.2.5. Western blotting and immunofluorescence assays
Antiplasmodial activity against the chloroquine sensitive (CQS)
D10 strain and chloroquine resistant (CQR) K1 strain of P.
falciparum, as well as cytotoxicity against a human cervical cancer
cell line (HeLa), were determined for the target compounds and are
tabulated in Table 1. Therapeutic indices were calculated by
dividing the IC50 obtained for HeLa cells with those obtained for the
K1 parasite strain. Chloroquine was used as the control for both
antiplasmodial and cytotoxicity assays.
Western blotting and immunofluorescence assays were con-
ducted as previously reported by Hoppe et al. [24]. Briefly,
trophozoite-stage D10 P. falciparum parasites were respectively
incubated with chloroquine, artesunate and the hybrid compounds
for 8 h at five times the IC50 of the relevant compounds, released
from the red blood cells by saponin treatment and washed
extensively to remove extraneous hemoglobin.
Hemoglobin levels in the released parasites were determined
by sodium dodecyl sulfate-polyacrylamide gel electrophoresis
The three Ugi adducts containing the artemisinin core (6a–c)
displayed excellent in vitro antiplasmodial activities that were

T.-S. Feng et al. / Biochemical Pharmacology 82 (2011) 236–247
241
Table 1
Results obtained from the in vitro biological evaluation of the intermediates and target compounds for antiplasmodial, anticancer and
b-Hematin inhibition activity.
O
H
N
R1
N
R2
N
O
n
6a - e
HN
Cl
.
Cpd
R1
R2
n
D10 IC50
(
ÆSDa) mM
K1 IC50
0.006
(
ÆSDa) mM
RIb
HeLa IC50
(
mM)
TIc
b
-Hematin
Inhibition
IC50 (equiv.)
2
3
4
5
–
–
–
–
–
–
–
–
–
–
–
–
0.006
1
0.53
NDf
NDf
NDf
NDf
–
–
–
–
NDf
NDf
NDf
NDf
38.35 Æ 5.085
1.37
20.29 Æ 4.137
2.56
1.87
1.27
2.15
2.74
6a
ARTd
1
0.026 Æ 0.0014
0.035 Æ 0.0014
0.023 Æ 0.0014
0.021 Æ 0.0014
0.88
0.286
12
0.45 Æ 0.04
6b
ARTd
3
0.60
0.169
8
0.31 Æ 0.01
6c
ARTd
ARTd
NQe
1
1
1
0.027 Æ 0.0015
0.024 Æ 0.0012
0.638 Æ 0.0018
0.019 Æ 0.0015
0.018 Æ 0.0012
0.467 Æ 0.0018
0.70
0.75
0.73
0.496
0.356
5.211
26
20
11
0.44 Æ 0.04
0.40 Æ 0.01
Not activeg
6c-citrate
6d
6e
NQe
1
0.569 Æ 0.0019
0.454 Æ 0.0020
0.80
11
6.259
14
Not activeg
Chloroquine
Artemisinin
DHA
–
–
–
–
–
–
–
0.020 Æ 0.0033
0.023 Æ 0.0040
0.004
0.219 Æ 0.0023
0.014 Æ 0.0037
0.003
8.54
NDf
NDf
39
–
1.91 Æ 0.3
0.66 Æ 0.1
NDf
–
NDf
0.61
0.75
–
a
SD = standard deviation of the mean of three independent determinations.
RI, resistance index, calculated as [IC50 (K1)]/[IC50 (D10)].
TI, therapeutic index, calculated as [IC50 (HeLa)]/[IC50 (K1)].
ART, artelinic acid core.
b
c
d
e
f
NQ, naphthoquinone core.
ND, not determined.
g
Not active, IC50 > 10.0 eq.
comparable to chloroquine’s activity against the CQS D10 strain
(26, 35 and 27 nM for 6a, 6b and 6c respectively, compared with
20 nM for CQ). The salt derivative 6c-citrate did not exhibit a
significant enhancement of in vitro activity relative to 6c. There
was no evidence of these artemisinin–chloroquinoline hybrids
exhibiting cross-resistance with chloroquine, as shown by the low
resistance index (RI) values for this set of compounds relative to
chloroquine [RI was calculated by dividing the IC50 obtained for the
CQR strain (K1) by that of the CQS strain (D10)]. The 1,4-
naphthoquinone-chloroquinoline hybrid molecules 6d and 6e only
exhibited moderate antiplasmodial activities relative to the
artemisinin-based hybrids, with IC50 values >0.4
mM against both

242
T.-S. Feng et al. / Biochemical Pharmacology 82 (2011) 236–247
strains of the malaria parasite. Notably however, they also
displayed no evidence of cross-resistance with chloroquine and
had low RI values of <1.
assay, and further fuels the controversy that still surrounds the
effects of artemisinin compounds on
tion.
b-hematin/hemozoin forma-
The results from the cytotoxicity assay suggest that the hybrid
molecules possess moderate selectivity for the malaria parasite
relative to mammalian cells. None of the compounds matched the
selectivity of chloroquine, which had a calculated therapeutic
index (TI) of 39.
Furthermore, in previous work we have found no effect of
artesunate on -hematin formation, so we ourselves were initially
b
very surprised by this result with artemisinin. Nonetheless, repeat
experiments gave the same result and forced us to the conclusion
that artemisinin does not behave in the same way as artesunate, at
least in this assay. The failure to observe inhibition in previous
studies may be a result of the reliance on the intrinsic absorbance
of soluble hematin, whereas the current assay converts unreacted
hematin to a distinct low-spin heme-pyridine complex which can
be more reliably measured.
Notwithstanding the above mentioned controversies, all data
produced so far by various authors needs to be reviewed in light of
the more recent disclosure that beta-hematin formation occurs at
the interface of lipid droplet-like structures dubbed lipid nano-
spheres [37]. The physico-chemical properties of this interface
environment differ markedly from bulk aqueous medium and the
effects of artemisinins in this milieu remain to be explored. This
finding has serious implications on the aforementioned contro-
3.2.
b-Hematin Inhibition
The hybrid compounds were tested for their ability to inhibit
the crystallization of hematin to -hematin in vitro using a
colorimetric -hematin inhibition screening assay [31], and the
b
b
results are also tabulated in Table 1.
Chloroquine was used as the control, and had an IC50 value of
1.91 eq. All three hybrid molecules that contain the artemisinin
core displayed enhanced
b-hematin inhibitory activity when
compared to chloroquine and artemisinin, with IC50 values of 0.45,
0.31 and 0.44 eq for 6a, 6b and 6c, respectively. The salt derivative
6c-citrate displayed activity similar to the free base 6c.
Compounds 6d and 6e, which contain the 1,4-naphthoquinone
core, displayed no inhibitory activity at the highest concentration
tested (10 eq). This suggested that the presence of the 1,4-
versy surrounding inhibition of
b-hematin formation by the
artemisinin class of drugs. On the basis of this more recent finding,
the different physico-chemical properties of artemisinin, dihy-
droartemisin, arteether and artesunate should be factored into any
argument. A detailed investigation of the effect on beta-hematin
formation at the lipid bilayer interface needs to be conducted for
artemisinin, dihydroartemisin, arteether and artesunate.
naphthoquinone moiety may interfere with the
b-hematin
inhibitory mechanism of the 4-aminoquinoline pharmacophore,
and probably explains why these two compounds exhibited
relatively lower in vitro antiplasmodial activities. Interestingly, we
were surprised to observe that artemisinin inhibited
formation with a lower IC50 value of 0.66 eq in our assay. This
observed potency of artemisinin in inhibiting -hematin forma-
tion is quite controversial, especially in light of existing reports
b-hematin
3.3. Morphological changes in the parasites
b
In order to monitor for changes in parasite morphology, the
artemisinin–chloroquinoline hybrid compounds 6a–c were added
to ring- and trophozoite-stage cultures respectively at concentra-
tions five times their respective IC50 values. At appropriate time
intervals after addition of the compounds, parasite-infected
erythrocytes were smeared onto microscope slides, fixed with
methanol and stained with Giemsa before viewing under a light
microscope. Microscope images for untreated and drug-treated
rings are shown in Fig. 4A. Control (untreated) ring stage parasites
were still in the ring stage after 6 h, but had developed into
prominent trophozoites after 24 h and 30 h. At 48 h, parasites had
progressed through a round of asexual reproduction to produce
new rings. When ring-stage parasites were treated with chloro-
quine, no morphological changes were observable after 6 h. This
was expected, as chloroquine has been widely accepted to target
parasites at the actively metabolizing trophozoite stage [38,39].
However, 24 and 30 h after drug treatment, the majority of the
parasites in the cell culture still remained in the ring stage, as
opposed to developing into trophozoites. This suggested that
chloroquine action had retarded parasite life-cycle progression.
After 48 h, the majority of the parasites had developed into
trophozoites with apparently enlarged food vacuoles and had not
undergone nuclear division to form schizonts.
Artemisinin appeared to be much faster-acting than chloro-
quine on ring-stage parasites, as dying parasites could already be
observed in the artemisinin-treated culture at 24 h after addition,
evidenced by the decrease in size and the contracted, pyknotic
appearance of the rings. Smears of the culture at 30 and 48 h
indicated that the culture predominantly consisted of non-viable
ring remnants, appearing as tiny, dense dots. Non-viable rings in
the culture indicate that artemisinin displays marked parasiticidal
activity on the ring stage, completely preventing survival and
progression into the trophozoite stage. This is consistent with the
in vivo findings that artemisinins cause a rapid reduction of
parasitemia to below detectable levels [40,41].
suggesting that artemisinin antimalarials do not inhibit
hematin/hemozoin formation [32].
b-
The controversy surrounding the possible effects of artemisinin
antimalarials on hemozoin formation is longstanding, and a look at
the literature suggests that the presence or absence of observed
effects is highly dependent on the assay applied and the attendant
conditions. For example, both quinoline as well as endoperoxide
antimalarials showed notable inhibition of
b-hematin formation
catalyzed by the cell free homogenate of Plasmodium yoelii [33,34];
this inhibition was more pronounced in a more sensitive plasma-
based assay (pH 4.8), with IC50s of 17.6 mM (CQ), 4.5 mM
(artemisinin) and 1.9 M ( arteether) [34]. In related studies,
m
a/b
artemisinin and dihydroartemisinin showed a dose-dependent
inhibition in the heme polymerization inhibitory activity (HPIA)
assay (hematin in acetic acid at pH 3, 37 8C, 24 h), with a mole ratio
(dihydroartemisinin:hematin) of 1.61 inhibiting
merization by 50% [35]. In contrast, no inhibition was observed
with dihydroartemisinin in a -hematin inhibitory activity (BHIA)
b-hematin poly-
b
assay (hemin in dimethyl sulfoxide-acetate buffer at pH 5.0, 37 8C,
18 h) [32].
It has been argued that the inhibition of b-hematin formation in
the HPIA assay by artemisinin and dihydroartemisinin reflects a
reactivity (resulting from axial interactions between the artemi-
sinin and hematin) that is not related to their antimalarial action;
the absence of activity in the BHIA assay (which is believed to
screen for those molecules forming
p–p interactions with hematin
and thus inhibiting -hematin formation) is thought to suggest
b
that these endoperoxide compounds do not exert their antimalar-
ial activity by inhibiting hemozoin formation [32,36]. Interestingly,
our results were derived using
a standardized BHIA assay
(hematin, dimethyl sulfoxide-acetate buffer at pH 5.0, 60 8C, 1 h)
as reported by Ncokazi et al. [31], and showed significant inhibition
of
b-hematin formation by artemisinin; this is despite the
similarity in conditions with the previously reported [36] BHIA

T.-S. Feng et al. / Biochemical Pharmacology 82 (2011) 236–247
243
Fig. 4. (A) Light microscopy images of Giemsa-stained parasites at various time-points after the addition of drugs to ring stage parasites. (B) Light microscopy images of
Giemsa-stained parasites at various time-points after addition of drugs to trophozoite stage parasites.
Morphological changes in the ring-stage parasites treated with
hybrid molecules 6a–c and its salt all followed the same pattern as
that of artemisinin, with cultures mainly consisting of non-viable
rings by 24 h after drug addition. In the culture treated with
compound 6b, a small percentage of abnormal and pyknotic early
trophozoites were also observed at 30 and 48 h after drug addition.
In a second set of morphology experiments, compounds were
added to parasites in the trophozoite stage and incubated for 48 h.
Light microscope images of untreated and drug-treated tropho-
zoites are shown in Fig. 4B. After 6 h, the control (untreated)
trophozoites started undergoing nuclear division (the multiple
nuclei produce a marked punctate appearance in the parasite
cytoplasm) to become schizonts. These daughter nuclei in the
schizont develop into small, invasive merozoites, and at 24 h and
30 h these merozoites were observed to have invaded fresh red
blood cells and developed into rings which further developed into
trophozoites that predominated at 48 h.
observed to be smaller and denser, a clear indication of stress on
the parasites. After 24 h of drug treatment, deformed schizonts
with few nuclei were observed, and this unusual morphology was
maintained for a further 24 h. Morphological changes in artemi-
sinin-treated parasites were somewhat different to those incubat-
ed with chloroquine. After 6 h, parasites seemed to be initiating
nuclear division and entering the schizont stage, though they did
not appear to progress further than this as the parasites were still
predominantly in this stage up to 24 h after artemisinin treatment.
At 30 and 48 h after drug treatment, pyknotic, darkly stained
parasite remnants were mostly observed.
The artemisinin-based hybrid compounds 6a–c appeared to act
faster than chloroquine or artemisinin; a small percentage of the
parasites had already taken on the contracted appearance of dying
parasites after just 6 h of drug exposure, and after 24 h of drug
treatment most of the parasites appeared non-viable.
From the study of morphological changes of the drug-treated
parasites, it was evident that chloroquine’s parasiticidal action is
mainly directed against trophozoite stage parasites, while its effect
on ring stage parasites is to slow/disrupt development into
trophozoites. By marked contrast, artemisinin is parasiticidal
Effects of chloroquine treatment on trophozoites were much
more pronounced than on the ring stage. Morphological changes
after 6 h demonstrated that chloroquine had already exerted
considerable effects on these parasites as the trophozoites were

244
T.-S. Feng et al. / Biochemical Pharmacology 82 (2011) 236–247
Fig. 4. (Continued).
against all stages of parasites and has a more rapid action. These
findings corresponded to findings published in literature
[38,39,42,43]. Importantly, the blood smears of cultures treated
with the hybrid molecules indicate that these compounds shared
the rapid parasiticidal action and broad stage-specificity of
artemisinin, and produced similar morphological changes.
3.4. Effect on endocytosis: hemoglobin levels
It has previously been shown that chloroquine blocks
hemoglobin degradation and causes an increase in parasite-
associated hemoglobin [44,45]. Based on their pronounced
antiplasmodial activities (Table 1), compounds 6a–c and its salt
were investigated by Western blotting for their effects on
hemoglobin levels in the parasites and compared to chloroquine
and artesunate.
It was evident from the Western blots (Fig. 5) that both
chloroquine and artesunate treatment caused an increase in
hemoglobin level in the parasites relative to the untreated controls,
Fig. 5. Western blots of hemoglobin levels in parasites. Parasites were left untreated
(ctrl), or incubated with chloroquine (CQ), artesunate (Artes), or hybrid compounds
6a–c and its salt at 37 8C for 5 h. Net intensities of individual bands were
determined with Kodak 1D image analysis software, version 3.5. Normalized
values Æ SD obtained from five independent experiments are shown.

T.-S. Feng et al. / Biochemical Pharmacology 82 (2011) 236–247
245
Fig. 6. Immunofluorescence microscopy localization of hemoglobin in parasites following incubation with various drugs. The position of the food vacuole (FV) is indicated by
large arrows, whereas those of the hemoglobin-containing transport vesicles are indicated by small arrows. Control: untreated parasites; Phase: phase-contrast images, Hb:
localization of hemoglobin; DAPI: nucleic acids were stained using DAPI in order to determine the number of nuclei, to avoid including multi-nucleated schizonts in the
analyses.
although the increase associated with artesunate was considerably
less than that obtained with chloroquine. These results are in
agreement with published results [24,46]. The hybrid molecules
also produced an increase in parasite hemoglobin levels compared
to the controls, and these effects were intermediate between those
observed for chloroquine and artesunate.
Parasite-associated hemoglobin is predominantly located in the
food vacuole. Ultra-structural studies have suggested that the
increase in hemoglobin level in chloroquine-treated parasites may
be due to the presence of endocytic vesicles (or hemoglobin
transport vesicles) filled with undigested hemoglobin found either
in the food vacuole [47] or throughout the parasite cytoplasm [24].
To investigate the effects of the target compounds on endocytic
trafficking within the parasite, subcellular localization of hemo-
globin was visualized by immunofluorescence microscopy, with
particular emphasis on the amount of hemoglobin in the food
vacuole and the numbers and distribution of endocytic vesicles.
Representative images are shown in Fig. 6.
In the control parasites, hemoglobin was predominantly
located in the food vacuole (FV). The location of the FV is readily
determined by referring to the corresponding phase-contrast
images, where the shiny hemozoin crystals, formed from
hemoglobin degradation in the FV are prominently visible. The
food vacuoles are further indicated by large arrows in the
3.5. Effect on endocytosis: localization of transport vesicles by
immunofluorescence microscopy
To further investigate the effects of the artemisinin–chlor-
oquinoline hybrid molecules on the endocytic pathway, the
subcellular location of hemoglobin within parasites after drug
exposure was determined by immunofluorescence microscopy
using anti-hemoglobin antiserum. Compound 6b was selected for
the immunofluorescence assay from among the hybrid species as it
showed the most significant increase in hemoglobin as detected by
Western blotting.

246
T.-S. Feng et al. / Biochemical Pharmacology 82 (2011) 236–247
fluorescence images. Punctate endocytic vesicles, filled with
hemoglobin, were also visible. These are indicated by small arrows.
For chloroquine treated parasites, there was little fluorescence
(or in some cases, no fluorescence) in the FV, and hemoglobin was
mostly found enclosed in endocytic vesicles. Treatment with
artesunate and 6b resulted in fewer endocytic vesicles when
compared to control parasites. The average number of transport
vesicles per parasite was enumerated by counting the amount of
visible vesicles (appearing as fluorescent foci separate from the
food vacuole) in 100–400 parasites per treatment. It was found
that the average number of transport vesicles increased signifi-
cantly in chloroquine-treated parasites (4.5) relative to the
untreated parasites (1.54), but decreased in parasites treated with
artesunate and 6b (0.89 and 0.91, respectively).
In conclusion, 4-Aminoquinolines were hybridized with arte-
misinin and 1,4-naphthoquinone derivatives via the Ugi-4CC
reaction and their biological activities were investigated. No cross-
resistance with chloroquine was observed for these hybrid
compounds in chloroquine resistant malaria parasites, despite
the fact that they contained the chloroquinoline moiety. The
artemisinin-containing compounds 6a–c and its salt were the most
active target compounds in the antiplasmodial assays. They were
also potent inhibitors of
b-hematin formation. However, despite
the potent in vitro activities, they also displayed notable
cytotoxicity and had lower therapeutic indices than chloroquine.
Morphological changes in parasites treated with compounds 6a–c
and its salt have shown that they may act by mechanism(s) similar
to artemisinin, as both ring and trophozoite stage parasites were
The results obtained for chloroquine and artesunate agreed
with those reported in literature. As discussed above, Hoppe et al.
[24] suggested that the increase in hemoglobin accumulation and
the number of transport vesicles in chloroquine-treated parasites
may be the result of a block in transport vesicle-vacuole fusion,
which reduces the delivery of hemoglobin to the food vacuole for
digestion. In addition, the decrease in endocytic vesicle counts in
artesunate-treated parasites agrees with the reported inhibition of
endocytosis by this drug – decreased hemoglobin endocytosis
should decrease the rate of endocytic vesicle formation and hence
lower the vesicle content per parasite. Crespo et al. [48] also report
that ring-stage parasite-infected erythrocytes treated for 24 h with
artemisinin at two times its IC50 value advanced to the trophozoite
affected by the drug treatment. Additionally, the potent b-hematin
inhibitory activities of these hybrids may contribute to their
inhibitory mechanism(s) against trophozoite stage parasites. The
effects of the artemisinin-containing hybrids on endocytosis were
also investigated, and were found to result in an increase in
accumulation of hemoglobin within the parasites. Subsequent
immunofluorescence studies showed that, for the representative
compound 6b, there was a decrease in the number of transport
vesicles in the parasites. These hybrid molecules may therefore act
by inhibition of endocytosis, as suggested by the immunofluores-
cence results, as well as by inhibition of hemoglobin degradation,
as suggested by the observed increase in hemoglobin levels in the
parasites seen on the Western blots.
stage but showed
accompanied by the disintegration of many of the major
membrane-bound features.
a
marked loss of organellar structures
The morphological study reported here constituted a qualita-
tive assessment of the stage-specificity and rate of action of the
active hybrid compounds vis-a-vis chloroquine and artemisinin.
The results suggested a more artemisinin-like action for the hybrid
compounds. Attempts were not made to accurately quantify the
morphological changes, given the intrinsic difficulties in confi-
dently categorising the morphological state of each individual
parasite by light microscopy, caused by the heterogeneity and
graded nature of morphological responses of individual cells to the
drug interventions (unless the changes are drastic, uniform and
advanced). It is due to the inherent qualitative nature of the
morphological evaluations that additional independent and more
quantifiable read-outs of chloroquine and artemisinin action were
sought - in this case measuring hemoglobin levels and transport
vesicle numbers in treated parasites. The results from the latter
experiments agree with the morphological study in suggesting a
more artemisinin-like action for the hybrid compounds.
At concentrations much higher than those applied in this study,
artemisinin-treatment has been reported to result in loss of
digestive vacuole integrity in trophozoite-infected erythrocytes
treated for 4 h. After 8 h of treatment with the endoperoxide
antimalarials, more dramatic effects on parasite morphology were
observed, including the disruption of the digestive vacuole
membrane and the release of hemozoin crystals into the parasite
cytoplasm. [48].
The morphology and immunofluorescence assays suggest that
the mode of action of the hybrid molecules more closely resembles
that of the artemisinin parent drug rather than that of chloroquine.
The hybrid molecules and artemisinin rapidly kill parasites in the
ring stage, as opposed to the parasiticidal action of chloroquine
which is more prolonged and prevalent in the trophozoite stage. In
addition, the hybrid molecules and artemisinin reduced the
amount of hemoglobin-filled endocytic vesicles in the parasite,
likely due to an inhibition of endocytosis, in contrast to the vesicle-
vacuole fusion block and marked increase in vesicles caused by
chloroquine. The fact that the increase in overall hemoglobin levels
in the parasite obtained with the hybrids, as judged by Western
blotting, is intermediate between the increase observed with
artesunate and chloroquine respectively suggests that an element
of chloroquine’s mode of action may be maintained in the hybrid
molecules.
An obvious limitation to this work presented here, and which is
a
universal limitation when evaluating the cell biological
consequences of antimalarial drug treatment, is difficulty in
distinguishing between ‘‘effects that are the result of a primary
mechanism of action of the drug (e.g., a direct interaction of the
drugs or their adducts with the endocytosis or fusion machinery of
the cell), an important secondary effect of a more general
perturbation (e.g., pH changes or membrane damage), or a
nonspecific manifestation of a cell that is stressed and dying’’ [24].
The moderate increase in overall parasite hemoglobin levels
caused by artemisinin has been interpreted as a combination of
decreased hemoglobin uptake due to a block in endocytosis on the
one hand, balanced by increased hemoglobin content due to an
inhibition of hemoglobin proteolysis in the food vacuole on the
Acknowledgments
Financial support from the following sources is gratefully
acknowledged: the South African National Research Foundation
(T-SF, EMG, TJE); the South African Research Chairs Initiative
(SARChI) of the Department of Science and Technology (DST)
administered through the South African National Research
Foundation (KC); the South African Medical Research Council
(MRC) and the University of Cape Town (KC, TJE); the Cancer
Association of South Africa (CANSA) (KC); and the Welcome Trust
through a Senior International Research Fellowship (HH).
other [24]. The b-hematin inhibition assays suggest that the hybrid
compounds could be even more effective than chloroquine at
inhibiting the formation of hemozoin crystals in the food vacuole.
Vacuole damage caused by heme accumulation may result in more
extensive hemoglobin proteolysis inhibition by the hybrids than
with the artemisinin parent drug, thus explaining the greater
extent of hemoglobin accumulation observed by Western blotting.

T.-S. Feng et al. / Biochemical Pharmacology 82 (2011) 236–247
247
References
[26] Salmon-Chemin L, Bruisine E, Yardley V, Kohler S, Debreu M-A, Landry V, et al.
2- and 3-Substituted 1,4-naphthoquinone derivatives as subversive substrates
of trypanothione reductase and lipoamide dehydrogenase from Trypanosoma
cruzi: synthesis and correlation between redox cycling activities and in vitro
cytotoxicity. J Med Chem 2001;44:548–65.
[27] Peck RM, Preston RK, Creech HJ. Nitrogen mustard analogs of antimalarial
drugs. J Am Chem Soc 1959;81:3984–9.
[28] Trager W, Jensen JB. Human malaria parasites in continuous culture. Science
1976;193:673–5.
[29] Lambros C, Vanderberg JP. Synchronization of Plasmodium falciparum eryth-
rocyte stages in culture. J Parasitol 1979;65:418–20.
[30] Guantai EM, Smith PJ, Gut J, Rosenthal PJ, Chibale K. Design, synthesis and in
vitro antimalarial evaluation of triazole-linked chalcone and dienone hybrid
compounds. Bioorg Med Chem 2010;18:8243–56.
[1] Thwing J, Skarbinski J, Newman RD, Barber AM, Mali S, Roberts JM, et al.
Malaria Surveillance—United States, 2005. Surveil Summ 2007;56:23–8.
[2] WHO. World malaria report; 2008.
[3] WHO Expert Committee on Malaria. WHO technical report series twentieth
report; 2000.
[4] Kumar A, Katiyar SB, Agarwal A, Chauhan PMS. Perspective in antimalarial
chemotherapy. Curr Med Chem 2003;10:1137–50.
[5] Price RN, Nosten F. Drug resistant falciparum malaria: clinical consequences
and strategies for prevention. Drug Resist Update 2001;4:187–96.
[6] WHO. Guidelines for the treatment of malaria. WHO; 2006.
[7] Noedl H, Se Y, Schaener K, Smith BL, Socheat D, Fukuda MM. Evidence of
artemisinin-resistant malaria in Western Cambodia.
2008;359:2619–20.
[8] Dondorp AM, Norsten F, Yi P, Das D, Phyo AP, Tarning J, et al. Artemisinin
resistance in Plasmodium falciparum malaria. N Engl J Med 2009;361:455–67.
[9] Ridley RG, Medical need. scientific opportunity and the drive for antimalarial
drugs. Nature 2002;415:686–93.
[10] Coatney GR. Pitfalls in a discovery: the chronicle of chloroquine. Am J Trop
Med Hyg 1963;12:121–8.
[11] Haynes RK. From artemisinin to new artemisinin antimalarials: biosynthesis,
extraction, old and new derivatives, stereochemistry and medicinal chemistry
requirements. Curr Top Med Chem 2006;6:509–37.
[12] Woodrow CJ, Haynes RK, Krishna S. Artemisinins. Postgrad Med J 2005;81:71–
8.
[13] Jones M, Mercer AE, Stocks PA, la Pensee LJI, Cosstick R, Park BK, Kennedy ME,
Piantanida I, Ward SA, Davies J, Bray PG, Rawe SL, Baird J, Charidza T, Janneh O,
O’Neill PM. Antitumour and antimalarial activity of artemisinin-acridine
hybrids. Bioorg Med Chem Lett 2009;19:2033–7.
[14] Araujo NCP, Barton V, Jones M, Stocks PA, Ward SA, Davies J, Bray PG, Shone AE,
Cristiano MLS, O’Neill PM. Semi-synthetic and synthetic 1,2,4-trioxaquines
and 1,2,4-trioxolaquines: synthesis, preliminary SAR and comparison with
acridine endoperoxide conjugates. Bioorg Med Chem Lett 2009;19:2038–43.
[15] Biot C, Bauer H, Schirmer RH, Davioud-Charvet E. 5-Substituted tetrazoles as
bioisosteres of carboxylic acids. bioisosterism and mechanistic studies on
glutathione reductase inhibitors as antimalarials. J Med Chem 2004;47:5972–
83.
[16] Davioud-Charvet E, Delarue S, Biot C, Schwo¨bel B, Boehme CC, Mu¨ssigbrodt A,
et al. A prodrug form of a Plasmodium falciparum glutathione reductase
inhibitor conjugated with a 4-anilinoquinoline. J Med Chem 2001;44:4268–
76.
N Engl J Med
[31] Ncokazi K, Egan T. A colorimetric high-throughput
b-hematin inhibition
screening assay for use in the search for antimalarial compounds. Anal
Biochem 2004;338:306–19.
[32] Haynes RK, Monti D, Taramelli D, Basilico N, Parapini S, Olliaro P. Artemisinin
antimalarials do not inhibit hemozoin formation. Antimicrob Agents Che-
mother 2003;47:1175.
[33] Pandey AV, Tekwani BL, Singh RL, Chauhan VS. Artemisinin, an endoperoxide
antimalarial, disrupts the hemoglobin catabolism and heme detoxification
systems in malarial parasite. J Biol Chem 1999;274:19383–8.
[34] Tripathi AK, Gupta A, Garg SK, Tekwani B. In vitro b-hematin formation assays
with plasma of mice infected with Plasmodium yoelii and other parasite
preparations: comparative inhibition with quinoline and endoperoxide anti-
malarials. Life Sci 2001;69:2725–33.
[35] Basilico N, Pagani E, Monti D, Olliaro P, Taramelli D. A microtitre-based method
for measuring the haem polymerization inhibitory activity (HPIA) of antima-
larial drugs. J Antimicrob Chemother 1998;42:55–60.
[36] Parapini S, Basilico N, Pasini E, Egan TJ, Olliaro P, Taramelli D, et al. Standardi-
zation of the physicochemical parameters to assess in vitro the
b-hematin
inhibitory activity of antimalarial drugs. Exp Parasitol 2000;96:249–56.
[37] Pisciotta JM, Coppens I, Tripathi AK, Scholl PF, Shuman J, Bajad S, et al. The role
of neutral lipid nanospheres in Plasmodium falciparum haem crystallization.
Biochem J 2007;402:197–204.
[38] Yayon A, van de Waa JA, Yayon M, Geary TG, Jensen JB. Stage-dependent effects
of chloroquine on Plasmodium falciparum in vitro.
J Euk Microbiol
1983;30:642–7.
[39] Terkuile F, White NJ, Holloway P, Pasvol G, Krishna S. Plasmodium falciparum:
in vitro studies of the pharmacodynamic properties of drugs used for the
treatment of severe malaria. Exp Parasitol 1993;76:85–95.
[40] Bjo¨rkman A, Bhattarai A. Public health impact of drug resistant Plasmodium
falciparum malaria. Acta Trop 2005;94:163–9.
[17] Atamna H, Ginsburg H. Heme degradation in the presence of glutathione. J Biol
Chem 1995;270:24876–83.
[41] Krishna S, Uhlemann A-C, Haynes RK. Artemisinins: mechanisms of action and
potential for resistance. Drug Resist Update 2004;7:233–44.
[42] Meshnick SR, Artemisinin:. mechanisms of action, resistance and toxicity. Int J
Parasitol 2002;32:1655–60.
[43] Meshnick SR. Artemisinin antimalarials: mechanism of action and resistance.
Med Trop Mars 1998;58:13–7.
[44] Famin O, Ginsburg H. Differential effects of 4-aminoquinoline-containing
antimalarial drugs on hemoglobin digestion in Plasmodium falciparum-
infected erythrocytes. Biochem Pharmacol 2002;63:393–8.
[18] Weissbuch I, Leiserowitz L. Interplay between malaria, crystalline hemozoin
formation, and antimalarial drug action and design. Chem Rev
2008;108:4899–914.
[19] Egan TJ, Combrinck JM, Egan J, Hearne GR, Marques HM, Ntenteni S, et al. Fate
of haem iron in the malaria parasite Plasmodium falciparum. Biochem
J
2002;365:343–7.
[20] Friebolin W, Jannack B, Wenzel N, Furrer J, Oeser T, Sanchez CP, et al.
Antimalarial dual drugs based on potent inhibitors of glutathione reductase
from Plasmodium falciparum. J Med Chem 2008;51:1260–77.
[45] Zhang Y. Inhibition of hemoglobin degradation in Plasmodium falciparum by
chloroquine and ammonium chloride. Exp Parasitol 1987;64:322–7.
[46] Roberts L, Egan TJ, Joiner KA, Hoppe HC. Differential effects of quinoline anti-
malarials on endocytosis in Plasmodium falciparum. Antimicrob Agents Che-
mother 2008;52:1840–2.
[47] Yayon A, Timberg R, Friedman S, Ginsburg H. Effects of chloroquine on the
feeding mechanism of the intraerythrocytic human malarial parasite Plasmo-
dium falciparum. J Euk Microbiol 1984;31:367–72.
[21] Chavain N, Davioud-Charvet E, Trivelli X, Mbeki L, Rottmann M, Brun R, et al.
Antimalarial activities of ferroquine conjugates with either glutathione re-
ductase inhibitors or glutathione depletors via a hydrolyzable amide linker.
Bioorg Med Chem 2009;17:8048–59.
[22] Ugi I, Steinbru¨ckner C. On
1960;72:267–8.
a new condensation principle. Angew Chem
[23] Do¨mling A, Ugi I. Multicomponent Reactions with Isocyanides. Angew Chem
Int Ed 2000;39:3168–210.
[48] Crespo MP, Avery TD, Hanssen E, Fox E, Robinson TV, Valente P, et al.
Artemisinin and a series of novel endoperoxide antimalarials exert early
effects on digestive vacuole morphology. Antimicrob Agents Chemother
2008;52:98–109.
[24] Hoppe HC, van Schalkwyk DA, Wiehart UIM, Meredith SA, Egan J, Weber BW.
Antimalarial quinolines and artemisinin inhibit endocytosis in Plasmodium
falciparum. Antimicrob Agents Chemother 2004;48:2370–8.
[25] Lin AJ, Klayman DL, Milhous WK. Antimalarial activity of new water-soluble
dihydroartemisinin derivatives. J Med Chem 1987;30:2147–50.