Antiplasmodial activity and cytotoxicity of bis-, tris-, and tetraquinolines with linear or cyclic amino linkers

Article abstract of DOI:10.1021/jm001096a

Bisquinoline heteroalkanediamines were structurally modified in order to study the effects of enhanced bulkiness and rigidity on both their activity on strains of Plasmodium falciparum expressing different degrees of chloroquine (CQ) resistance and their

Full text of DOI:10.1021/jm001096a

1658
J . Med. Chem. 2001, 44, 1658-1665
Expedited Articles
An tip la sm od ia l Activity a n d Cytotoxicity of Bis-, Tr is-, a n d Tetr a qu in olin es
w ith Lin ea r or Cyclic Am in o Lin k er s
Sophie Girault,† Philippe Grellier,‡ Amaya Berecibar,†,§ Louis Maes,| Pascal Lemie`re,† Elisabeth Mouray,‡
Elisabeth Davioud-Charvet,† and Christian Sergheraert*,†
UMR CNRS 8525, Universite´ de Lille II, Institut de Biologie et Institut Pasteur de Lille, 1 rue du Professeur Calmette,
B.P. 447, 59021 Lille Cedex, France, Laboratoire de Biologie Parasitaire, IFR CNRS 63, Muse´um National d’Histoire
Naturelle, 61 rue Buffon, 75005 Paris, France, and Tibotec, B-32800 Mechelen, Belgium
Received October 10, 2000
Bisquinoline heteroalkanediamines were structurally modified in order to study the effects of
enhanced bulkiness and rigidity on both their activity on strains of Plasmodium falciparum
expressing different degrees of chloroquine (CQ) resistance and their cytotoxicity toward
mammalian cells. While cyclization yielded molecules of greater rigidity that were not more
active than their linear counterparts, they were characterized by an absence of cytotoxicity.
Alternatively, dimerization of these compounds led to tetraquinolines that are very potent for
CQ-resistant strains and noncytotoxic.
In tr od u ction
promising molecule, 4 (Ro 47-7737),13 as well as that of
other bisquinolines such as piperaquine, hydroxypiper-
aquine, and dichloroquinazine,14 has been suspended for
reasons of toxicity. Compared with the numerous bis-
quinolines above, Ro 47-7737 differed by the absence of
a proton-accepting side chain and by a greater rigidity
likely to reduce its efflux from the parasite while
favoring its affinity for ferriprotoporphyrin IX by a
decrease in the cost of entropy. With the aim of
maintaining both steric hindrance and a reduction of
the degrees of freedom while introducing proton-accept-
ing and/or substitution sites, we have designed new bis-,
tris-, and tetraquinolines whose 4-amino group belongs
to tri- and tetraazamacrocycles (cyclams) (A, C, and D
series of Charts 2 and 3). Series B was synthesized as
a linear counterpart (Chart 2).
Fifty years after its discovery, chloroquine (CQ, 1,
Chart 1) is still a mainstream drug in the fight against
malaria, but its efficacy is being eroded by the emer-
gence of resistant parasites.1,2 CQ is believed to exert
its activity by inhibiting haemozoin formation in the
digestive vacuole of the malaria carrying the parasite
Plasmodium.3,4 Discussion of the exclusivity of this
mechanism has recently been renewed by Ginsburg and
co-workers who have proposed that inhibition of fer-
riprotoporphyrin IX degradation by glutathione-depend-
ent redox processes could be an additional mode of
action of CQ.5 Biochemical studies have indicated that
isolates of the CQ-resistant parasites accumulate less
drug content than their more sensitive counterparts;
however, a mechanistic explanation for this observation
remains the subject of continuing debate.6-9 Although
CQ resistance may involve several mechanisms, its
reversal by molecules such as verapamil, desipramine,
and chlorpromazine suggests that an enhanced CQ
efflux by a multidrug-resistant mechanism may be
implicated.6,10 A strategy having the potential to over-
come this mechanism is the design of quinoline-based
drugs that will not be recognized by the proteins
involved in drug efflux. In this regard, bulky bisquino-
lines 2-4 (Chart 1) likely to be extruded with difficulty
by a proteinaceous transporter11 have been synthesized
and were discovered to inhibit the growth of both CQ-
sensitive and CQ-resistant parasites with similar
efficacy.11-13 However, further development of the most
We report here the synthesis and the antiparasitic
activities on Plasmodium falciparum of these new bis-,
tris-, and tetraquinolines. Cytotoxic effects upon human
MRC-5 cells (diploid embryonic lung cell line) and mouse
peritoneal macrophages are also discussed.
Ch em istr y
New bis- and trisquinolines were obtained by reacting
polyamines with 5 equiv of 4,7-dichloroquinoline in
DMF at reflux in the presence of an inorganic base
potassium carbonate (Scheme 1). Thick-layer chroma-
tography was used for purification. 1,4,8,11-Tetraaza-
cyclotetradecane was reacted as described above, and
after purification, it afforded trisubstituted and disub-
stituted compounds 5 and 6 (A series of Chart 2).
Forcing conditions and a greater excess of 4,7-dichlo-
roquinoline did not yield the tetrasubstituted variant.
1,4,7-Triazacyclononane afforded compound 7, the di-
substituted analogue of 6, and the monosubstituted
compound 8 (A series of Chart 2) but not the trisubsti-
tuted derivative. This may have been the result of steric
* To whom correspondence should be addressed. Phone: (33) 3 20
87 12 11. Fax: (33) 3 20 87 12 33. E-mail: christian.sergheraert@
pasteur-lille.fr.
† Institut de Biologie et Institut Pasteur de Lille.
‡ Muse´um National d’Histoire Naturelle.
§ Present address: Pfizer Global Research and Development, 3-9
rue de la Loge, 94265 Fresnes Cedex, France.
| Tibotec.
10.1021/jm001096a CCC: $20.00 © 2001 American Chemical Society
Published on Web 04/27/2001

Tetraquinolines with Amino Linkers
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 11 1659
Ch a r t 1. Chloroquine (1, CQ) and Biologically Active Bisquinolines 2-411-13
Ch a r t 2. Compounds 5-14 (A and B Series)a
a
ClQ: 7-chloroquinol-4-yl. Q: quinol-4-yl.
hindrance or of the greater rigidity provided by the ring.
Under the same conditions, linear amines diethylene-
triamine, N-(3-aminopropyl)-1,3-propanediamine, and
spermidine were only substituted on primary amino
groups, yielding, respectively, amines 9, 10, and 11 (B
series of Chart 2). Compounds 12 and 13 (Chart 2) were
prepared by di- or trisubstitution of tris(2-aminoethyl)-
amine, while addition of a third quinoline moiety to
compound 9 by reductive amination with 4-quinoline-
carboxaldehyde (Scheme 1) afforded trisubstituted com-
pound 14 (Chart 2). Use of N-methylpyrrolidinone
(NMP) at reflux as solvent had been previously recom-
mended for the obtention of bisquinolines via a dis-
placement reaction with 4,7-dichloroquinoline, alkane-
or heteroalkanediamines, and triethylamine.11,15 Bis-
quinolines were then isolated by the addition of water
and diethyl ether or ethyl acetate to the cooled reaction
mixtures, which initiated product precipitation. Under
these conditions, yields ranged from 49 to 87% for
alkanediamines and only from 12 to 86% for hetero-
alkanediamines. This decrease explained the low yields
that we obtained with linear polyamines and a fortiori
with cyclic polyamines, especially since product precipi-
tation in NMP, as with our series of compounds, was
not observed. In general, whatever the choice of the
solvent (DMF or NMP), it was very difficult to know
the fate of polyamines.
Acids 15 and 16 were synthesized, as described in
Scheme 2, by reaction of compound 7 with succinic and
glutaric anhydride, respectively, while the amide de-
rivative 17 of acid 15 was prepared by treatment with
ammonium hydrogen carbonate. In the case of com-
pounds 18, 20, 21, and 25, the side chain was introduced
by coupling various carboxylic acids with the free,
secondary amino group of compound 7, using bromo-
tris-pyrrolidinophosphonium hexafluorophosphate (Py-
BroP) as coupling reagent (Scheme 2).16,17 For com-
pounds 26-28, transformation to the amino side chain
was achieved in two steps: reaction of secondary amino
compound 7 with the appropriate bromo acid followed
by substitution of the remaining bromo group by pip-
eridine (Scheme 2). The benzylhydroxyl protecting group

1660 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 11
Ch a r t 3. Compounds 15-37 (C and D Series), 38, and 39
Girault et al.
Sch em e 1. Synthesis of Compounds 5-14a
a
Reagents: (a) K2CO3, DMF; (b) 4-quinolinecarboxaldehyde, molecular sieves, absolute EtOH, then NaBH4, absolute EtOH.
was removed from compound 18 by treatment with
ammonium formate and Pd/C to give alcohol 19 (Scheme
2). N-Boc-amino protecting groups of compounds 20 and
21 were removed by treatment with a 1:1 mixture of
TFA/CH2Cl2 to give deprotected analogues 22 and 23,
respectively (Scheme 2). The guanidinium derivative 24
was synthesized by treating primary amino compound
23 with 3,5-dimethylpyrazole-1-carboxamide, employing
sodium hydrogen carbonate as base (Scheme 2).18
Compound 29 was synthesized in two steps: (i) forma-
tion of the acyl chloride corresponding to acid 16 and
(ii) reaction of this latter with compound 9 (Scheme 2).
Bis derivatives of 1,4,7-triazacyclononane and com-
pounds 30 and 32-35, were synthesized as described
in Scheme 3 by coupling various diacids with the free,
secondary amino group of compound 7, using the
method displayed in Scheme 2. Compound 31 was
prepared by coupling acid 15 and amine 7 (Scheme 3).
In the case of compound 36, the synthesis required four
steps: (i) reaction of the anhydride, obtained by treat-
ment with DCC19 of N-(tert-butoxycarbonyl)iminodiace-
tic acid, with compound 7,19 (ii) formation of the
corresponding acyl fluoride by treating acid with cya-
nuric fluoride,17 (iii) reaction of the latter with a second

Tetraquinolines with Amino Linkers
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 11 1661
Sch em e 2. Synthesis of Compounds 15-29a
a
Reagents: (a) anhydride, pyridine; (b) Boc2O, NH4HCO3, pyridine, DMF; (c) acid, PyBroP, DIEA, DMF for compounds 18, 20, 21, and
25; (d) (i) bromo acid, PyBroP, DIEA, DMF, (ii) piperidine, 1-pentanol for compounds 26-28; (e) ammonium formate, Pd/C, MeOH; (f)
TFA/CH2Cl2 (1:1); (g) 3,5-dimethylpyrazole-1-carboxamide, NaHCO3, EtOH; (h) (1) oxalyl chloride, Et3N, dry CH2Cl2, (2) compound 9,
dry CH2Cl2.
molecule of 7, and (iv) deprotection of the N-Boc-amino
protecting group by treatment with a 1:1 mixture of
TFA/CH2Cl2 to give deprotected derivative 36 (Scheme
3). The phenyl linker of compound 37 was introduced
by reaction of terephthaloyl chloride with compound 7,
using diisopropylethylamine as a base (Scheme 3).
Compounds 38 and 39 were synthesized by coupling
acid 16 with 1,4,7-triazacyclononane and its mono-
quinoline derivative 8, respectively, using PyBroP as
coupling reagent (Scheme 3).
mouse peritoneal macrophages while no cytotoxicity was
observed for compound 7 (Table 3). Compound 11
inhibited haem polymerization in the same range as CQ
(IC50 ) 83 µM). Compounds 12-14, corresponding to
compound 9 substituted at the central nitrogen atom
of the linker, were less active than the parent molecule
itself (IC50 was 605.7 nM, >1 µM, and >1 µM, respec-
tively). Acylation of the remaining cyclic nitrogen of
compound 7 led to a comparative weakening in the
antimalarial activity (IC50 values mostly about 1 µM)
irrespective of the nature of the terminal group: the
free carboxylic group (compounds 15 and 16) or car-
boxamide group (compound 17), benzyl-protected (com-
pound 18) or free alcohol group (compound 19), N-Boc-
protected (compounds 20 and 21) or free amino group
(compounds 22 and 23), and guanidinium group (com-
pound 24). Alternatively, the presence of a terminal
piperidine proved favorable to activity independent of
the chain length (compounds 25-28, IC50 between 38
and 132 nM), yet cytotoxic effects toward human MRC-5
cells and mouse peritoneal macrophages were observed
from a concentration of 8 µM.
Biologica l Resu lts
All of the compounds were first tested for their
antimalarial activity upon the CQ-resistant strain
FcB1R (Tables 1 and 2, IC50 ) 110 nM for CQ) and for
their cytotoxicity toward human MRC-5 cells and mouse
peritoneal macrophages (Tables 3 and 4). Bisquinoline
derived from 1,4,7-triazacyclononane, compound 7, dis-
played an activity similar to CQ (IC50 ) 112.8 nM),
while its monosubstituted derivative (compound 8) and
the di- and trisubstituted 1,4,8,11-tetrazacyclotetrade-
cane derivatives (compounds 6 and 5) yielded IC50
values greater than 1 µM. While its activity was similar
to that of CQ, compound 7 was found to inhibit haem
polymerization to a lesser extent (IC50 ) 170 µM) than
CQ (IC50 ) 65 µM). Among the bisquinolines derived
from linear amines (compounds 9-11), compounds 10
and 11 were more active than compound 7 but displayed
a high toxicity toward both human MRC-5 cells and
Whatever the nature of the linker, the presence of an
additional chloroquinoline disubstituted 1,4,7-triazacy-
clononane (compounds 30-37, D series) removed en-
tirely any cytotoxic properties, except in the case of
compound 34, from a concentration of 32 µM (Table 4).
Antimalarial activity clearly depended on the length and
the nature of the linker; most of the compounds were

1662 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 11
Sch em e 3. Synthesis of Compounds 30-39a
Girault et al.
Ta ble 1. In Vitro Sensitivity of P. falciparum FcB1R Strain to
Compounds 5-29 (A-C Series)
b
compd series
n
n′
Xa
IC50 (nM)
5
6
7
8
9
A
A
A
A
B
B
B
B
B
B
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
>1000c
>1000c
112.8 ( 24.9d
>1000c
2
3
3
2
2
2
2
3
2
3
3
0
2
0
2
2
2
2
3
4
2
2
2
NH
NH
NH
142.1 ( 10.2c
75.4 ( 22.6c
57 ( 6c
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
N-CH2-CH2-NH2
N-CH2-CH2-NH-ClQ
N-CH2-Q
COOH
COOH
CONH2
O-CH2-C6H5
OH
CH(CH3)-NHBoc
NHBoc
CH(CH3)-NH2
NH2
NH-C(NH2)dNH
piperidine
piperidine
605.7 ( 22.7c
>1000c
>1000c
>1000c
940e
980e
1000e
930e
462.6 ( 41.7c
> 1000e
228.3 ( 33.1c
> 1000c
> 1000c
38.6 ( 17.7f
132.2 ( 68.0c
81.1 ( 24.1c
108.8 ( 69.8c
4
7
11
3
piperidine
piperidine
C(O)-N((CH2)2-NH-ClQ)2 152.2 ( 9.2g
a
b
ClQ: 7-chloroquinol-4-yl. Q: quinol-4-yl. IC50 ) 110 nM for
CQ. c Number of expts n ) 3. Number of experiements n ) 5.
d
e Number of experiements n ) 2. f Number of experiements n )
g
6. Number of experiements n ) 4.
Ta ble 2. In Vitro Sensitivity of P. falciparum FcB1R Strain to
Compounds 30-37 (D Series), 38, and 39
a
compd series
n
n′
X
IC50 (nM)
a
30
D
CH2-CH2-piperazine- 76.9 ( 9.4b
CH2-CH2
Reagents: (a) diacid, PyBroP, DIEA, DMF; (b) (1) tert-butyl-
2,6-dioxo-4-morpholinecarboxylate, dry THF, (2) cyanuric fluoride,
pyridine, dry CH2Cl2, (3) compound 7, pyridine, dry CH2Cl2, (4)
TFA/CH2Cl2 (1:1); (c) terephthaloyl chloride, DIEA, dry CH2Cl2;
(d) compound 7, PyBroP, DIEA, DMF; (e) 1,4,7-triazacyclononane,
PyBroP, DIEA, DMF; (f) 1-(7-chloroquinol-4-yl)-1,4,7-triazacy-
clononane, PyBroP, DIEA, DMF.
31
32
33
34
35
36
37
38
39
D
D
D
D
D
D
D
(CH2)2
38.1 ( 14.3c
18.3 ( 9.3c
32.7 ( 13.5b
22.6 ( 7.2b
79.1 ( 27.6b
102.6 ( 20.1b
266.2 ( 30.1b
>1000d
(CH2)3
(CH2)5
(CH2)7
(CH2)10
CH2-NH-CH2
phenyl (para)
more active than both CQ and the starting molecule 7
(IC50 between 18 and 103 nM), while a phenyl ring
(compound 37, IC50 ) 266.2 nM) was found to be less
favorable. The presence of one (compound 36) or two
proton-acceptor sites (compound 30) in the linker, likely
to increase vacuolar pH, was also found to be unfavor-
able to antimalarial activity when compared with a
simple polymethylene chain (n ) 2 to n ) 10). The
partial or total absence of chloroquinoline moieties on
one of the cyclononanes led to a substantial decrease in
activity (compare compound 32, IC50 ) 18.3 nM, with
3
3
274.7 ( 32.7b
a
b
IC50 ) 110 nM for CQ. Number of experiments n ) 3.
d
c Number of experiments n ) 6. Number of experiments n ) 2.
the inhibition of parasite growth and that of haem
polymerization was found for many bisquinolines,15
inhibition of haem polymerization was evaluated for CQ
and the two most active compounds 25 (C series) and
32 (D series). When tested at 65 µM, the IC50 value of
CQ in the haem polymerization assay of compounds 25
and 32 displayed 70% and 66% inhibition, respectively.
its analogues 38 and 39, IC50 > 1000 nM and IC50
)
274.7 nM, respectively). In addition, the presence of the
second triazacyclononane seems to be crucial for potent
antimalarial activity because compound 29, a linear
analogue of compound 32, displayed a much lower
activity (IC50 ) 152.2 nM) compared with compounds 7
and 9 (see IC50 values).
Discu ssion
Although the relationship between the P-glycoprotein
homologue 1 protein (Pgh1) of P. falciparum and
resistance to CQ and mefloquine (MF) remains debated,
there is now evidence that mutation occurring on Pgh1
can confer resistance to MF and can influence the
parasite resistance to CQ along with the structurally
unrelated compound artemisinin.10 Bisquinolines likely
to be extruded with difficulty by a proteinaceous trans-
porter11 were synthesized with the aim of avoiding CQ
resistance. New products were discovered that inhibit
the growth of CQ-sensitive and CQ-resistant parasites.
However, further development was suspended for rea-
The most active and least cytotoxic compounds of each
series (bis- and monoderivatives of 1,4,7-triazacy-
clononane 7, 25, and 30-35) were subsequently evalu-
ated for their capacity to inhibit the growth of other
strains expressing different degrees of CQ resistance
(Table 5). With the exception of compound 7 they all
yielded similar IC50 values whatever the CQ resistance
of a particular strain. Since a clear correlation between

Tetraquinolines with Amino Linkers
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 11 1663
Ta ble 3. In Vitro Cytotoxicity of Compounds 5-29 on MRC-5
Ta ble 5. Efficiency of Compounds 7, 25, and 30-35 To Inhibit
Growth of Parasites Expressing Different Degrees of Resistance
to CQ
Cells and Mouse Peritoneal Macrophages
cytotoxicity on
a
cytotoxicity on
mouse peritoneal
IC50 (nM)
MRC-5 cells (%)
macrophagesa
P. falciparum strain
concn ) concn ) concn ) concn ) concn ) concn ) concn )
compd
W2
FcB1R
D6
F32
compd 32 µM
8 µM
1 µM 0.5 µM 32 µM
8 µM
2 µM
CQ
7
175 ( 31
110 ( 26
54 ( 12
74.5 ( 7.3
18.9 ( 4.9
59.7 ( 4.4
21.9 ( 5.8
18.5 ( 3.9
29.3 ( 8.1
41.7 ( 8.2
19 ( 4
5
6
7
8
9
18
98
0
13
0
0
0
0
98
100
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
98
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
T
T
-
-
T
-
-
-
-
-
T
-
-
-
-
-
T
170.1 ( 19.7 112.8 ( 24.9
64.6 ( 2.2
16.1 ( 2.5
68.3 ( 3.1
23.7 ( 2.2
17.9 ( 2.4
34.5 ( 1.7
46.3 ( 4.6
25
30
31
32
33
34
35
50.2 ( 7.0
111.9 ( 13.4
29.5 ( 5.7
29.2 ( 2.6
47.8 ( 6.4
67.2 ( 4.9
179.9 ( 5.5
38.6 ( 17.7
76.9 ( 9.4
38.1 ( 14.3
18.3 ( 9.3
32.7 ( 13.5
22.6 ( 7.2
0
81
100
100
51
0
0
0
0
0
14
0
0
100
88
0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
T
T
T
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
T
-
-
-
-
-
-
T
-
-
T
-
-
-
-
-
-
-
-
-
-
T
-
-
T
T
T
79.1 ( 27.6 123.2 ( 38.0 132.6 ( 4.1
a
Parasites were considered resistant to CQ for IC50 > 100 nM.
IC50 values are the mean ( standard deviation of three indepen-
dent experiments except for the FcB1R strain (see Tables 1 and
2).
less potency than CQ, it revealed nonetheless a CQ-like
behavior because a correlation was observed between
the growth inhibition of 12 different strains of P.
falciparum and the degree of resistance to CQ (data not
shown). In the absence of a possible rational design,
acylation of the remaining amino group of compound 7
by a number of different substituents with linkers of
varying length (C series) and the dimerization of the
resultant molecules (D series) were carried out, and the
impacts of these modifications were measured against
both CQ-resistant strains and haem polymerization. In
the C series, the presence of a terminal amino group
(compound 23) or a guanidino group (compound 24),
which likely allows interaction with haem propionate
groups, greatly reduces antimalarial activity. The same
effect is observed with the other terminal groups,
whatever their nature, and in the presence of a protec-
tive group, except for that with a hydrogen bond
acceptor such as piperidine (compounds 25-28). In the
latter case, the length of the linker proves to be
insignificant for antimalarial activity. However, these
derivatives are toxic on MRC-5 cells and mouse perito-
neal macrophages at concentrations of 8 and 2 µM,
respectively. Molecular variations based on the dimer-
ization of compound 7 (D series) led to an increase of
the antimalarial activity (6-fold for compound 32) except
for the case of a rigid spacer (compound 37). Further-
more, the absence of in vitro cytotoxic effects was noted.
The presence of one (compound 36) or two proton-
acceptor sites (compound 30) was found to be unfavor-
able for antimalarial activity when compared with a
simple polymethylene chain (n ) 2 to n ) 10). These
results confirm those previously reported, proving that
an increase in the alkalinity of the food vacuole by
quinoline drugs and alkylamines does not correlate well
with their antimalarial activity.20 The most active
compounds (30-35) inhibited, in the same range, the
growth of parasites expressing different degrees of
resistance to CQ and to MF, since the F32 and the W2
strains are respectively 10 and 5 times more resistant
to MF than to the FcB1R strain (data not shown). The
four isoquinoline moieties are vital for antimalarial
activity, as proved by the superior IC50 values of
compounds 39 and 38 when compared with that of
compound 32. The fact that the most active compounds
in the C and D series (compounds 25 and 30-35)
displayed similar IC50 values irrespective of the CQ
T
0
0
0
0
0
0
0
11
0
-
-
T
T
T
T
-
0
91
92
91
89
0
87
92
91
92
0
T
-
T
-
a
The letter T means that the compound is toxic at this
concentration; - denotes no toxicity.
Ta ble 4. In Vitro Cytotoxicity of Compounds 30-39 on MRC-5
Cells and Mouse Peritoneal Macrophages
cytotoxicity on mouse
cytotoxicity on
MRC-5 cells (%)
peritoneal
macrophagesa
concn ) concn ) conc ) concn ) concn ) concn ) concn )
compd 32 µM
8 µM
1 µM 0.5 µM 32 µM
8 µM
2 µM
30
31
32
33
34
35
36
37
38
39
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
-
-
-
-
T
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
a
The letter T means that the compound is toxic at this
concentration; - denotes no toxicity.
sons of toxicity, including that of the most potent
candidate, Ro 47-7737, unique for the steric hindrance
attributed to its cyclohexyl ring. Decreasing the degrees
of freedom of a drug is a well-known method of enhanc-
ing acutely its affinity for a target by reducing the loss
in entropy. Simultaneously, interactions with other
undesired targets responsible for certain levels in toxic-
ity can be limited. We have therefore sought to augment
the rigidity of bisquinolines linked via short alkanedi-
amines already reported as revealing a promising level
of activity.11 The new structures were obtained by
introducing chloroquinoline moieties to two azamacro-
cycles, retaining one or two amino groups for substitu-
tion. Bisquinoline obtained from the smallest cycle
(compound 7), noncytotoxic at 32 µM, was found to be
more active than its tetraazamacrocycle homologue
(compound 6) toward the CQ-resistant strain FcB1R and
was therefore selected for optimization. While compound
7 inhibited haem polymerization (IC50 ) 170 µM) with

1664 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 11
Girault et al.
effects toward host cells. The peritoneal macrophages were
collected from the peritoneal cavity 48 h after stimulation with
potato starch and seeded in 96-well microplates at 30 000 cells
per well. MRC-5 cells were seeded at 5000 cells per well. After
24 h, the cells were washed and 2-fold dilutions of the drug
were added in 200 µL standard culture medium (RPMI + 5%
FCS). The final DMSO concentration in the culture remained
below 0.5%. The cultures were incubated with four concentra-
tions of compounds (32, 8, 1, and 0.5 µM) at 37 °C in 5% CO2-
95% air for 7 days. Untreated cultures were included as
controls. For MRC-5 cells, the cytotoxicity was determined
using the colorimetric MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide (thiazolyl blue)) assay26 and
scored as a percent reduction of absorption at 540 nm of
treated cultures versus untreated control cultures. For mac-
rophages, scoring was performed microscopically.
resistance of the strain, while inhibiting haem polym-
erization in the same range as CQ (compounds 25 and
32), might suggest that their greater bulkiness results
in a weaker efflux by a parasite transporter, although
a mechanism of action differing from that of CQ should
not be excluded.
In conclusion, an increase in rigidity by cyclization
yielded molecules that were not more active than their
linear counterparts but yet differed by an absence of
cytotoxic effects. Dimerization led to tetraquinolines
that are both very potent for CQ-resistant strains and
noncytotoxic.
Ma ter ia ls a n d Meth od s
Biologica l Eva lu a tion . 1. In Vitr o P . fa lcipa r u m Cu l-
tu r e a n d Dr u g Assa ys. P. falciparum strains were main-
tained continuously in culture on human erythrocytes as
described by Trager and J ensen.21 In vitro antiplasmodial
activity was determined using a modification of the semi-
automated microdilution technique of Desjardins et al.22 P.
falciparum CQ-sensitive (F32/Tanzania and D6/Sierra-Leone)
and CQ-resistant (FcB1R/Colombia and W2/Indochina) strains
were used in sensitivity testing. FcB1R and F32 were strains
obtained by limit dilution. Stock solutions of chloroquine
diphosphate and test compounds were prepared in sterile,
distilled water and DMSO, respectively. Drug solutions were
serially diluted with culture medium and added to asynchro-
nous parasite cultures (0.5% parasitemia and 1% final hema-
tocrite) in 96-well plates for 24 h at 37 °C prior to the addition
of 0.5 µCi of [3H]hypoxanthine (1-5 Ci/mmol; Amersham, Les
Ulis, France) per well. The growth inhibition for each drug
concentration was determined by comparison of the radioactiv-
ity incorporated into the treated culture with that in the
control culture (without drug) maintained on the same plate.
The concentration causing 50% inhibition (IC50) was obtained
from the drug concentration-response curve, and the results
were expressed as the mean ( the standard deviations
determined from several independent experiments. The DMSO
concentration never exceeded 0.1% and did not inhibit the
parasite growth.
2. Ha em P olym er iza tion Assa y. The drug effects on haem
polymerization were assessed according to Raynes et al.23 The
haemozoin content was determined using the procedure of
Chou and Fitch.24 Briefly, a 50 µL aliquot of insoluble tropho-
zoite material of the FcB1R strain (approximately equivalent
to 4 × 107 parasites) was added to 900 µL of a haem/acetate
mixture (0.3 mM bovine haematin, 60 mM sodium acetate, pH
5). A total of 50 µL of drug solution at different concentrations
was mixed with the other components. Samples without drug
constituted controls. All of the samples contained the same
amount of DMSO (1%). After incubation for 4 h at 37 °C, the
samples were centrifuged at 27 000 g for 15 min at 4 °C. The
pellet was resuspended in 1 mL of buffer A (68 mM NaCl, 4.8
mM KCl, 1.2 mM MgSO4, 5 mM glucose, 50 mM sodium
phosphate, pH 7.4) and repelleted. This second pellet was
resuspended with 2.5% SDS in buffer A and sonicated for 10
min. The polymerized haem was collected by centrifugation
at 27 000 g for 30 min at 20 °C. The pellet was then washed
four times before being resuspended in 900 µL of 2.5% SDS in
buffer A, and 100 µL of 1 M NaOH was added to dissolve the
polymerized haem. After incubation for 1 h, the concentration
of haemozoin was determined by measuring the absorbance
at 404 nm.25 The amount of haemozoin formed during the
incubation period was corrected for the endogenous haemozoin
of the trophozoite preparation, and the concentration of drug
required to produce 50% inhibition of haem polymerization
(IC50) was determined. Data presented are the mean of two
independent experiments each performed in duplicate.
3. Cytotoxicity Test on MRC-5 Cells a n d Mou se P er i-
ton ea l Ma cr op h a ges. A human diploid embryonic lung cell
line (MRC-5, Bio-Whittaker 72211D) and mouse primary
peritoneal macrophages were used to assess the cytotoxic
Ack n ow led gm en t. We express our thanks to Ge´r-
ard Montagne for NMR experiments and Dr Steve
Brooks for proof reading. This work was supported by
CNRS (GDR 1077, IFR CNRS 63, UMR CNRS 8525)
and Universite´ de Lille II.
Su p p or tin g In for m a tion Ava ila ble: Details of chemical
procedures and analytical data. This material is available free
of charge via the Internet at http://pubs.acs.org.
Refer en ces
(1) Van Est, H. G.; Skamene, G. E.; Schurr, E. Chemotherapy of
malaria: a battle against the odds? Clin. Invest. Med. 1993, 16,
285-293.
(2) White, N. J . Antimalarial drug resistance: the pace quickens.
J . Antimicrob. Chemother. 1992, 30, 571-585.
(3) Dorn, A.; Stoffel, R.; Matile, H.; Bubendorf, A.; Ridley, R. G.
Malarial haemozoin/b hematin supports haem polymerization
in the absence of protein. Nature 1995, 374, 269-271.
(4) Dorn, A.; Vippagunta, S. R.; Matile, H.; J aquet, C.; Vennerstrom,
J . L.; Ridley, R. G. An assessment of drug-hematin binding as
a mechanism for inhibition of hematin polymerization by quino-
line antimalarials. Biochem. Pharmacol. 1998, 55, 727-736.
(5) Ginsburg, H.; Famin, O.; Zhang, J .; Krugliak, M. Inhibition of
Glutathione-Dependent Degradation of Heme by Chloroquine
and Amodiaquine as a Possible Basis for Their Antimalarial
Mode of Action. Biochem. Pharmacol. 1998, 56, 1305-1313.
(6) Krogstad, D. J .; Gluzman, I. Y.; Kyle, D. E.; Oduola, A. M.;
Martin, S. K.; Milhous, W. K.; Schlesinger, P. H. Efflux of
chloroquine from Plasmodium falciparum: mechanism of chlo-
roquine resistance. Science 1987, 238, 1283-1285.
(7) Ginsburg, H.; Stein, W. D. Kinetic modeling of chloroquine
uptake by malaria-infected erythrocytes. Biochem. Pharmacol.
1991, 41, 1463-1470.
(8) Wu¨nsch, S.; Sanchez, C. P.; Gekle, M.; GroBe-Wortmann, L.;
Wiesner, J .; Lanzer, M. Differential Stimulation of the Na+/H+
exchanger determines chloroquine uptake in Plasmodium fal-
ciparum. J . Cell Biol. 1998, 140, 335-345.
(9) Bray, P. G.; J anneh, O.; Raynes, K. J .; Mungthin, M.; Ginsburg,
H.; Ward, S. A. Cellular uptake of chloroquine is dependent on
binding to ferriprotoporphyrin IX and is dependent on NHE
activity in Plasmodium falciparum. J . Cell. Biol. 1999, 145, 363-
376.
(10) Reed, M. B.; Saliba, K. J .; Caruana, S. R.; Kirk, K.; Cowman, A.
F. Pgh1 modulates sensitivity and resistance to multiple anti-
malarials in Plasmodium falciparum. Nature 2000, 403, 906-
909.
(11) Vennerstrom, J . L.; Ellis, W. Y.; Ager, A. L.; Andersen, S. L.;
Gerena, L.; Milhous, W. K. Bisquinolines. 1. N,N-Bis(7-chloro-
quinolyl-4-yl)alkanediamines with potential against chloroquine-
resistant malaria. J . Med. Chem. 1992, 35, 2129-2134.
(12) Raynes, K.; Galatis, D.; Cowman, A. F.; Tilley, L.; Deady, L. W.
Synthesis and activity of some antimalarial Bisquinolines. J .
Med. Chem. 1995, 38, 204-206.
(13) Ridley, R. G.; Matile, H.; J aquet, C.; Dorn, A.; Hofheinz, W.;
Leupin, W.; Masciadri, R.; Theil, F. P.; Richter, W. F.; Girometta,
M. A.; Guenzi, A.; Urwyler, H.; Gocke, E.; Potthast, J . M.; Csato,
M.; Thomas, A.; Peters, W. Antimalarial activity of the bisquino-
line
trans-N1,N2-bis(7-chloroquinolin-4-yl)cyclohexane-1-di-
amine: comparison of two stereoisomers and detailed evaluation
of the S,S enantiomer, Ro 47-7737. Antimicrob. Agents Chemo-
ther. 1997, 41, 677-686.

Tetraquinolines with Amino Linkers
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 11 1665
(14) Basco, L. K.; Ruggeri, C.; Le Bras, J . Mole´cules antipaludiques;
Masson: Paris, 1994; pp 115-120.
(15) Vennerstrom, J . L.; Ager, A. L.; Dorn, A.; Andersen, S. L.;
Gerena, L.; Ridley, R. G.; Milhous, W. K. Bisquinolines. 2.
Antimalarial N,N-Bis(7-chloroquinolyl-4-yl)heteroalkanediamines.
J . Med. Chem. 1998, 41, 4360-4364.
(16) Coste, J .; Frerot, E.; J ouin, P. Coupling N-Methylated Amino
Acids Using PyBroP and PyCloP Halogenophosphonium Salts:
Mechanism and Fields of Application. J . Org. Chem. 1994, 59,
2437-2446.
(17) Girault, S.; Davioud-Charvet, E.; Salmon, L.; Berecibar, A.;
Debreu, M.-A.; Sergheraert, C. Structure-activity relationships
in 2-aminodiphenylsulfides against trypanothione reductase
from Trypanosoma cruzi. Bioorg. Med. Chem. Lett. 1998, 8,
1175-1180.
(18) Davis, P. D.; Elliott, L. H.; Harris, W.; Hill, C. H.; Hurst, S. A.;
Keech, E.; Hari Kumar, M. K.; Lawton, G.; Nixon, J . S.;
Wilkinson, S. E. Inhibitors of Protein Kinase C. 2. Substituted
Bisindolylmaleimides with Improved Potency and Selectivity. J .
Med. Chem. 1992, 35, 994-1001.
(19) Cathala, B.; Raouf-Benchekroun, K.; Galaup, C.; Picard, C.;
Cazaux, L.; Tisne`s, P. Synthesis and Structure of Macrocyclic
Dioxa-, Dithia-, Diazatetralactams and Derivatives. Tetrahedron
1996, 52, 9793-9804.
(20) Ginsburg, H.; Nissani, E.; Krugliak, M. Alkalinisation of the food
vacuole by quinoline drugs and alkylamines is not correlated
with their antimalarial activity. Biochem. Pharmacol. 1989, 38,
2645-2654.
(21) Trager, W.; J ensen, J . B. Human malarial parasites in continu-
ous culture. Science 1976, 193, 673-677.
(22) 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.
(23) Raynes, K.; Foley, M.; Tilley, L.; Deady, L. W. Novel Bisquinoline
Antimalarials. Biochem. Pharmacol. 1996, 52, 551-559.
(24) Chou, A. C.; Fitch, C. D. Control of heme polymerase by
chloroquine and other quinoline derivatives. Biochem. Biophys.
Res. Commun. 1993, 195, 422-428.
(25) Asakura, T.; Minakata, K.; Adachi, K.; Russel, M. O.; Schwartz,
E. Denatured hemoglobin in sickle erythrocytes. J . Clin. Invest.
1977, 59, 633-640.
(26) Mossman, T. Rapid Colorimetric Assay for Cellular Growth and
Survival: Application to Proliferation and Cytotoxicity Assays.
J . Immunol. Methods 1983, 65, 55-63.
J M001096A