Synthesis and biological evaluation of substituted quinolines: Potential treatment of protozoal and retroviral co-infections

Article abstract of DOI:10.1016/j.bmc.2003.09.007

We report the synthesis of substituted quinolines and their in vitro biological evaluation against the causal agents of cutaneous leishmaniasis, visceral leishmaniasis, African trypanosomiasis and Chagas' disease. Furthermore, several quinolines have also been tested for their anti-retroviral activity in HIV-1 infected cells. The structure-activity relationships of these new synthetic compounds are discussed and emphasis was placed on the treatment of leishmania/HIV co-infections.

Full text of DOI:10.1016/j.bmc.2003.09.007

Bioorganic & Medicinal Chemistry 11 (2003) 5013–5023
Synthesis and Biological Evaluation of Substituted Quinolines:
Potential Treatment of Protozoal and Retroviral Co-infections
Mohammed A. Fakhfakh,a Alain Fournet,a,b Eric Prina,c Jean-Francois Mouscadet,d
Xavier Franck,a Reynald Hocquemillera and Bruno Figaderea,*
a
´ ´ ´ ´
Laboratoire de Pharmacognosie (associe au CNRS-BioCIS), Faculte de Pharmacie, Universite Paris-Sud, rue J.B. Clement,
92296 Chaˆtenay-Malabry, France
´
b
Institut de Recherche pour le Developpement (IRD), 213 rue Lafayette, 75480 Paris, France
´
c
Institut Pasteur, Unite d’Immunophysiologie et Parasitisme Intracellulaire, 25 rue du Dr. Roux, 75724 Paris cedex 15, France
d
´
Laboratoire de Physicochimie et de Pharmacologie des Macromolecules Biologiques, URA 147 CNRS, PRII,
Institut Gustave Roussy, 39 rue Camille Desmoulins, 94805 Villejuif, France
Received 8 April 2003; accepted 1 September 2003
Abstract—We report the synthesis of substituted quinolines and their in vitro biological evaluation against the causal agents of
cutaneous leishmaniasis, visceral leishmaniasis, African trypanosomiasis and Chagas’ disease. Furthermore, several quinolines have
also been tested for their anti-retroviral activity in HIV-1 infected cells. The structure–activity relationships of these new synthetic
compounds are discussed and emphasis was placed on the treatment of leishmania/HIV co-infections.
# 2003 Elsevier Ltd. All rights reserved.
Introduction
able, chemotherapies are nowadays exclusively based on
nitroimidazoles treatments (e.g., benznidazole, Roche).
Although benznidazole has an efficiency in the acute
and short-term (up to a few years) chronic phases,4 no
benefits could be definitively evidenced for chronically
infected patients.
Parasitic diseases such as leishmaniasis and trypanoso-
miasis have significant impacts in developing countries,
with infections spread over several hundred millions of
people and are the cause of a mortality rate of several
millions per year. Conventional chemotherapies are
often inadequate, toxic or are becoming less effective
due to emergence of numerous resistances.1 There is an
urgent need for new drugs for the chemotherapy of
human African trypanosomiasis (caused by Trypano-
soma brucei gambiense and Trypanosoma brucei rho-
diense), Chagas disease2,3 [caused by Trypanosoma
(Schizotrypanum) cruzi] and leishmaniases (Leishmania
spp).
The leishmaniases5 constitute a diverse group of dis-
eases that varies from simple cutaneous leishmaniasis to
visceral leishmaniasis (VL), a very fatal disease if left
untreated. VL, also known as ‘kala-azar’ is a severe
disease and epidemics are generally devastating. More
recently, VL has also gained attention as an opportu-
nistic infection in AIDS. Owing to the fact that over-
lapping geographical distribution of leishmaniases and
HIV/AIDS is increasing, Leishmania–HIV co-infection
has been regarded as an emerging infectious disease.5,6
For 50 years, the pentavalent antimony compounds
sodium stilbogluconate (Pentostam, Glaxo Wellcome,
UK) and meglumine antimonate (Glucantime, Aventis,
France) have been the first-line treatments of leishma-
niases. However, relapse is frequent and resistance to
antimony is growing. Amphotericin B and pentamidine,
the parenteral alternatives to antimony are considered
to cause serious and irreversible toxic effects. New
chemotherapeutic methods have been proposed based
Treatments of Human African trypanosomiasis, com-
monly known as sleeping sickness, (pentamidine, sur-
amin and melarsoprol) often induce serious side effects.
Chemotherapy of Chagas’ disease is still unsatisfactory
and, since Nifurtimox (a nitrofuran) is no more avail-
*Corresponding author. Tel.:+33-1-4683-5592; fax+33-1-4683-5399;
e-mail: bruno.figadere@cep.u-psud.fr
0968-0896/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2003.09.007

5014
M. A. Fakhfakh et al. / Bioorg. Med. Chem. 11 (2003) 5013–5023
on the use of other non-parenteral agents such as the
aminoglycoside, aminosidine (topical application) or
oral agents like hexadecylphosphocholine (miltefosine).
The latter molecule, an antineoplastic agent, has been
identified as the first effective oral treatment in visceral
infection, and is currently in phase IV clinical studies.7
Results and Discussion
Chemistry
Compounds 2–4, 9, 10, 14, 15, 19, 50–52 were prepared
from N-oxyquinoline, by addition of isobutyl chloro-
formate followed by the desired Grignard reagent, to
afford the corresponding quinoline in good yields (32–
82%), as reported.13 Compounds 11 and 12 were
obtained through the same procedure but starting from
N-oxy-8-hydroxy- and N-oxy-6-methylquinoline, in 23
and 30% yield, respectively. Compound 1 was obtained
in 88% yield from 2 by TBAF treatment in THF
(Scheme 1).
To develop other alternative drugs, new compounds
have been looked for based on empirical screening or
ethnopharmacological studies, and 2-substituted quino-
lines isolated from a Bolivian medicinal plant, Galipea
longiflora Kr (Rutaceae),8 have shown efficacy in the
experimental treatment of cutaneous leishmaniasis of
the New World.9,10 The activity of these compounds has
also been described in the experimental treatment of VL
by oral and parenteral routes.11
Compounds 5–8, 22, and 23 were obtained in good
yields (54–84%) by treatment of 2-quinaldehyde 28 by
the desired alkyltriphenylphosphonium bromide, in the
presence of t-BuOK in toluene at 0 ꢀC. Compounds 21,
25 were obtained in moderate non-optimized yields (21,
15%, respectively) from 6-quinaldehyde 29, under the
same reaction conditions, whereas 24 was obtained
from 22 in 63% yield. Compound 26 was obtained in
63% yield from 27 (which was prepared by selenium
oxidation of commercially available 2,6-dimethylquino-
line) under the same reaction conditions (Scheme 2).
Recently, the effects of oral treatment of 2-n-propylquin-
oline, a major alkaloid isolated from G. longiflora Kr
(Rutaceae) were evaluated in Balb/c mice chronically
infected with T. cruzi, and preliminary data indicate
that the serological response, evidenced by ELISA
(enzyme linked immuno assay), is significantly different
from that of the controls and of the benznidazole-trea-
ted mice.12
In order to get information on the structure–activity
relationship in this family, Fakhfakh et al.,13 have syn-
thesised a library of 2-substituted quinolines in solution
from Grignard reagents and a chloride salt of N-oxy-
quinoline. Then 2-alkenylquinolines were prepared
either from-1,1-dibromo-2-(2-quinolyl)-ethylene14 or
from 2-quinaldehyde.15 Finally, 3-substituted,16 and
disubstituted quinolines were also synthesised.15
Carboxylic acid 30 was obtained in 60% yield from
ester 23, by KOHtreatment, whereas alcohols 31 and 32
were obtained in 49 and 60% yield by DIBAL reduction
of 23, depending on the reaction conditions (stoıchio-
metric and excess of DIBAL, respectively) (Scheme 2).
Alcohols 35 and 37 were obtained in 53 and 42% yield,
from 2-quinaldehyde, by addition of the required
Scheme 1. (a) i-BuCOCl, then RMgBr, ꢁ78 ꢀC, THF; (b) TBAF, THF.
Scheme 2. (a) Ph3P¼CHR, toluene, 0 ꢀC; (b) DIBAL, ꢁ78 ꢀC; (c) DIBAL, excess; (d) KOH, MeOH, H2O.

M. A. Fakhfakh et al. / Bioorg. Med. Chem. 11 (2003) 5013–5023
5015
Grignard reagent. Compounds 33 and 43 were obtained
in 57% yield, from 2-quinaldine, by treatment with
pyridine followed by addition of the desired carboxylic
anhydride, whereas when 2-quinaldine was treated by
butyllithium and then by acetaldehyde, compound 34
was obtained in 60% yield. Under the same reaction
conditions (but with 2 equiv of butyllithium), 8-
hydroxy-2-methylquinoline gave 36 in 53% yield
(Scheme 3).
from 38 by butyllithium treatment). When 1 in Et2O
was mixed with E-dichloroethylene in the presence of
piperidine and catalytic amounts of Pd(PPh3)4 and CuI,
compound 44 was obtained in 52% yield (Scheme 5).
Compound 13 was obtained in 34% yield when 3-bro-
moquinoline was treated by propenylmagnesium bro-
mide in a 1/1 mixture of NMP/THF, in the presence of
a catalytic amount of Co(acac)2. Whereas 20 was pre-
pared in 46% yield when 3-bromoquinoline was treated
at 30 ꢀC by phenylmagnesium bromide in THF in the
presence of a catalytic amount of Fe(acac)3. Com-
pounds 18 and 48 were obtained in 98 and 70% yield,
respectively, from 3-bromoquinoline, in pyrrolidine,
and in the presence of the desired alkyne and catalytic
amounts of Pd(PPh3)4 and CuI (Scheme 6).
Compounds 38 and 39 were obtained in 89 and 22%
yield, respectively, by treatment of 28 and 29 by PPh3
and CBr4. Then when 38 was treated by 1 equiv of i-
PrMgCl in the presence of a catalytic amount of Fe-
(acac)3 in a 1/1 mixture of THF/NMP, 42 was obtained
in 84% yield, however if a second equiv of a Grignard
reagent was added, thus compounds 16 and 17 were
obtained in 63 and 47% yield, respectively. Compounds
45 and 46 were obtained in 60 and 72% yield, respec-
tively, when 38 was mixed with the desired alkyne in
toluene in the presence of N,N-diisopropylethylamine
and catalytic amounts of Pd(PPh3)4 and CuI. However,
compound 47 was obtained in 20% yield (non-opti-
mized) when 42 was treated by 1 in THF in the presence
of pyrrolidine and catalytic amounts of Pd(PPh3)4 and
CuI (Scheme 4).
Finally, compound 49 was obtained in 42% yield by
heating 2-quinaldine and 2-quinaldehyde 28 in the pre-
sence of acetic anhydride (Scheme 7).
Biological results
Anti-protozoal, anti-retroviral and cytotoxic activities
of various synthetic compounds were presented in four
tables. Table 1 shows the activities of 21 unfunctional-
ized alkenyl and alkynylquinolines (1–21) against intra-
cellular amastigotes of L. amazonensis, L. infantum, and
Compound 41 was obtained from 38 in 87% yield by
t-BuOK treatment (whereas 1 was alternatively obtained
Scheme 3. (a) RMgBr; (b) BuLi, then RCHO; (c) pyridine, then RCO2COR.
Scheme 4. (a) PPh3, CBr4; (b) (i) i-PrMgCl, Fe(acec)3; (c) alkyne, Pd(PPh3)4, CuI.
Scheme 5. BuLi (2 equiv); (b) t-BuOK (2 equiv); (c) (E)-ClCH¼CHCl, piperidine, Et2O, Pd(PPh3)4, CuI.
Scheme 6. (a) RMgBr, Co(acac)2 or Fe(acac)3, THF; (b) alkyne, Pd(PPh3)4, CuI.

5016
M. A. Fakhfakh et al. / Bioorg. Med. Chem. 11 (2003) 5013–5023
Table 1. In vitro cytotoxicity of unfunctionalized alkenyl and alkynylquinolines upon bone marrow-derived macrophages and amastigote or
trypomastigote forms of L. amazonensis, L. infantum, T. brucei and T. cruzi parasites
Compd
IC50 (mM)
Cytotoxicity
macrophages
L. amazonensis
amastigotesa
L. infantum
amastigotesb
T. brucei
trypomastigotesc
T. cruzi
amastigotesd
1
>41
<5
32
>32
8
2
3
4
5
6
7
8
7
40
56
21
8
32
>32
13
16
17
16
4
8
12
74
37
12
>32
>32
>32
11
137
34
ND
>32
16
>200
>200
111
63
>200
56
28
4
11
9
47
>200
>200
47
>200
>200
ND
>32
14
>32
>32
>32
>32
>32
19
10
11
12
13
14
15
16
137
148
68
148
ND
ND
16
ND
ND
13
ND
ND
19
>200
>200
>200
>200
>200
>200
30
ND
ND
ND
ND
>32
17
>200
<100
11
4
14
(continued)

M. A. Fakhfakh et al. / Bioorg. Med. Chem. 11 (2003) 5013–5023
5017
Table 1 (continued)
Compd
IC50 (mM)
Cytotoxicity
macrophages
L. amazonensis
amastigotesa
L. infantum
amastigotesb
T. brucei
trypomastigotesc
T. cruzi
amastigotesd
18
19
>200
>200
>200
<100
ND
ND
ND
ND
ND
ND
20
21
>200
>200
<100
>200
ND
ND
ND
ND
ND
ND
Amphotericin B
Glucantime1
Suramin
25
>200
ND
0.13
20
ND
ND
ND
7
ND
ND
ND
ND
0.09
ND
ND
ND
ND
0.4
Nifurtimox
ND
aThirty-hour contact with parasites.
bFive-day contact with parasites.
cFour-day contact with parasites.
dTwenty four-hour contact with parasites.
T. cruzi and extracellular trypomastigotes of T. brucei.
Cytotoxicity against host cell macrophages is also
represented for all compounds tested. In the case of
leishmanicidal activity, compound 1 (2-ethynylquino-
line) displayed against L. amazonensis amastigotes a
lower IC50 (around 5 mM) than the reference drug, glu-
cantime1 (20 mM) which shows identical activity than
compound 3 (2-vinylquinoline). All compounds have a
lower activity against L. infantum amastigotes than glu-
cantime1 (7 mM) excepted for compound 2 (2-tri-
methylsilylethynylquinoline with IC50=8 mM), but this
compound showed a high cytotoxicity against macro-
phages. Among these compounds, the alkenylquinolines
7, 8 and 17 inhibited the growth of trypomastigote
forms of T. brucei with an IC50 around 4 mM. However,
these quinolines are 20 times less active than the refer-
ence drug, suramin (0.09 mM). Against the amastigote
forms of T. cruzi, compounds 1 and 2 were the more
active ones (IC50=8 mM), but still much less active than
nifurtimox (IC50=0.4 mM). Increasing the carbon chain
length did not result in an increase activity against the
parasites.
activity on the inhibition of L. amazonensis and L.
infantum amastigote proliferation in macrophages (both
compounds have an IC50 of 4 mM and 2 mM, respec-
tively). These results show the importance of the sub-
stitution, at the 2 position, by an a-b-unsaturated
aldehyde or an allylic alcohol, and the presence of a
double bond on the lateral chain, which induces an
increase activity against the intracellular amastigotes of
L. amazonensis and L. infantum. However, the unsatu-
rated ester (compound 23), less electron attractor than
compound 22, induced a lost of activity but was less
cytotoxic. Among the functionalized quinolines, com-
pounds 22 and 31 were also found the most efficient
ones for inhibiting the amastigote forms of T. cruzi,
with an IC50 around 5 mM. Compound 35 (1-quinol-2-
yl-propan-1-ol) was the most efficient one against try-
pomastigotes of T. brucei with IC50=3 mM. On the
other hand, compounds 31, 34, 36 and 37, which all
possess an hydroxyl group on the lateral chain in posi-
tion 2 or 3, showed weaker activity against T. brucei.
In Table 3, 12 halogeneous and miscellaneous function-
alized quinolines were tested for their anti-leishmanial
and anti-trypanosomal activities. Compounds 38 (1,1-
dibromo-2-(quinolyl)-1-ethene and 42 [1-bromo-2-(2-
quinolyl)ethylene] displayed notably high activity
against amastigotes of L. infantum (IC50=2 mM) and
From Table 2, presenting the data obtained from 16
functionalized quinolines (compounds 22–37), some
observations can be made on the structure–activity
relationship. Compounds 22 and 31 have a significant
Scheme 7.

5018
M. A. Fakhfakh et al. / Bioorg. Med. Chem. 11 (2003) 5013–5023
Table 2. In vitro cytotoxicity of functionalized quinolines upon bone marrow-derived macrophages and amastigote or trypomastigote forms of L.
amazonensis, L. infantum, T. brucei and T. cruzi parasites
Compd
IC50 (mM)a
Cytotoxicity
macrophages
L. amazonensis
amastigotes
L. infantum
amastigotes
T. brucei
trypomastigotes
T. cruzi
amastigotes
22
23
24
9
>200
105
4
59
47
2
26
19
16
5
15
26
>32
>32
25
26
27
28
29
117
>200
>200
>200
>200
117
>200
>200
>200
>200
ND
ND
>32
>32
ND
ND
ND
25
8
>32
ND
ND
>32
ND
ND
30
31
32
33
34
63
34
63
4
>32
2
>32
>32
ND
10
>32
4
>200
>200
>200
67
ND
>32
ND
ND
>32
>32
>200
134
16
35
36
37
>200
>200
>200
>200
>200
>200
16
ND
>32
3
16
14
>32
16
>32
aFor lengths of treatments and data for reference drugs, see Table 1.

M. A. Fakhfakh et al. / Bioorg. Med. Chem. 11 (2003) 5013–5023
5019
Table 3. In vitro cytotoxicity of halogenous and miscellaneous functionalized quinolines upon bone marrow-derived macrophages and amastigote
or trypomastigote forms of L. amazonensis, L. infantum, T. brucei and T. cruzi parasites
Compd
IC50 (mM)a
Cytotoxicity
macrophages
L. amazonensis
amastigotes
L. infantum
amastigotes
T. brucei
trypomastigotes
T. cruzi
amastigotes
38
80
159
2
4
>32
39
40
>200
>200
ND
ND
ND
ND
ND
ND
100
100
41
42
43
44
54
13
54
3
>32
2
>32
1
>32
0.15
ND
ND
>200
>200
>200
59
ND
ND
ND
ND
45
46
72
72
ND
ND
ND
ND
ND
ND
>200
>200
47
48
49
>200
>200
>200
>200
>200
>200
17
17
8
ND
13
ND
>32
>32
24
aFor lengths of treatments and data for reference drugs, see Table 1.
trypomastigotes of T. brucei (IC50=4 and 1 mM,
respectively). The 1,1-dibromoalkene (compound 38)
was significantly less cytotoxic towards macrophages
(IC50=80 mM) than 1-monobromo compound 42
(IC50=13 mM). Compound 42 showed also a good anti-
L. amazonensis activity with an IC50 value of 3 mM
against amastigotes. Looking at the regression of para-
sitophorous vacuoles and the overall decrease in para-
site number, compound 42 showed complete clearance
of amastigotes (at 3 mM) (Fig. 1).
Otherwise, this same compound 42 was more active
against the amastigote forms of T. cruzi (IC50=0.15
mM) than the reference drug, nifurtimox (IC50=0.4 mM,
Table 1). Indeed, compound 42 is the 2-substituted
quinoline which holds the most potent anti-protozoal
activity described in this study. All compounds within
the group of bisquinolines (47–49) or compounds 39–41
and 43–44 from the halogenated product family were
either inactive against L. amazonensis or possessed the
same or higher level of cytotoxicity against host cells.

5020
M. A. Fakhfakh et al. / Bioorg. Med. Chem. 11 (2003) 5013–5023
Table 4. Inhibition of HIV-1 replication in CEM4fx cells by mono or polysubstituted quinolines
Compd
Antiviral activity IC50 (mM)a
Cytotoxicity on CEM4fx Cells TC50 (mM)b
2
0.5
1.8ꢂ0.2
3
5
6
7.2
6.0
7.9
9.4ꢂ4
5.5ꢂ0.4
8.5ꢂ1.2
10
14
16
4.6
2.9
5.9ꢂ3
7.5ꢂ2.5
>100
>100
22
23
0.5
1.0ꢂ0.2
>100
>100
26
31
32.2
3.6
30.6ꢂ5
1.7ꢂ0.3
36
2.7
4.5ꢂ0.9
37
38
48
9.8
3.3
20.4ꢂ6
5.3ꢂ0.6
168ꢂ51
135
49
50
51
>300
78
>300
84ꢂ21
110
108
52
>300
>300
Ref.
L-731,988
1.2ꢂ0.3
>100 mM
aExperimental data of 3 days treatment of three independent experiments were averaged and fitted on a sigmoıdal dose–response model in order to
evaluate a TC50 for each compound. SE are standard deviations of best-fit values.
bCytotoxicity of most compounds masked the true antiviral activity of most compounds thus preventing fitting experimental data to a significant
pharmacological model. Therefore, the antiviral activity was merely estimated by graphical determination of the IC50 corresponding to the drug
concentration for which 50% of viral production was observed as compared to the non-treated control.

M. A. Fakhfakh et al. / Bioorg. Med. Chem. 11 (2003) 5013–5023
5021
Finally, among the 49 compounds tested for anti-pro-
tozoal activities, 19 were evaluated for their in vitro
antiviral activity, looking for the inhibition of HIV-1
replication in CEM4fx cells (Table 4). Compounds 2
and 22 exhibited submicromolar antiviral activity (IC50
around 0.5 mM) and compounds 14, 31, 36 and 38
showed IC50 in the micromolar range. We observed that
two of these compounds possess unsaturated side chain
with three carbon atoms (22 and 31), whereas com-
pound 14 is an alkylquinoline.
intervention of host-specific immunoregulatory mol-
ecules like TNF-a, a molecule which has been shown to
play an important role in the pathogenesis of L. infan-
tum and HIV-1 infections.17,18
Interestingly, several quinolines (2, 3, 36, 48) were found to
exhibit significant activity against HTLV-1 transformed
cells.19 Indeed, at 10 mM, all these compounds inhibited
more than 80% of the virus replication. Further studies are
necessary to evaluate the efficiency of these compounds
for the treatment of adult-T-cell leukemia/lymphoma
(ATLL), a severe HTLV-I-associated disease.20,21
The results presented herein confirm the crucial role of
the side chain length for antileishmanial or trypanoso-
mal activity. In general, the most active quinolines were
those with a three carbon alkenyl side chain, containing
reactive electrophilic functions such as a carbonyl
(compound 22), an hydroxyl (compound 31) or an
halogen (compound 42). Although compounds 22 and
31 showed a toxicity on HIV-1 host cells but not on
macrophages, they possessed significant in vitro activity
against HIV-1 replication and they were the most active
quinolines against Leishmania parasites. The mechanism
by which these quinolines are active against parasites or
HIV-1 is unknown. It could be due to either a direct
action of these molecules on molecular targets specific
of the pathogen or an indirect action involving the
The results obtained in the present study are interesting
and represent a screening of new quinolines with anti-
protozoal and/or anti HIV-1 activities. Our results
indicate that two compounds (22 and 31) show the best
activities against intracellular amastigotes forms of L.
amazonensis and L. infantum and also against the repli-
cation of HIV-1. The activities described herein repre-
sent a major advance in the search for the treatment of
co-infections protozoal/retrovirus,22 and particularly
Leishmania/HIV, but these findings must be validated in
experimental in vivo studies.
Experimental
Chemistry
The compounds were prepared through the methods
published.13ꢁ16 Full experimental procedures with spec-
troscopic data of the 52 compounds can be obtained as
supplementary material, which can be found in the
online version of this paper.
In vitro activity against intracellular L. amazonensis
amastigotes
Female BALB/c mice aged 2–4 months are obtained
from the breeding center of the Pasteur Institute. L.
amazonensis strain LV79 (MPRO/BR/1972/M1841) was
propagated in BALB/c mice. L. amazonensis amasti-
gotes were isolated from lesions and purified as descri-
bed earlier1 Bone marrow plugs from tibias and femurs
of BALB/c mice were suspended in RPMI 1640 medium
(Seromed) supplemented with 10% heat-inactivated
fetal calf serum (FCS, Dutscher, Brumath, France), 50
mg/mL of streptomycin, 50 IU/mL of penicillin (culture
medium) and with 15% L-929 fibroblast-conditioned
medium. Cells were then distributed in bacteriologic
Petri dishes (Greiner, Germany) and were incubated at
37 ꢀC in a 5% CO2 atmosphere. Five days later, adherent
macrophages were washed with Dulbecco’s phosphate
buffered solution (PBS) and taken off by treatment for
20 min at 37 ꢀC with Ca2+-and Mg2+-free Dulbecco’s
PBS containing 2 mg/mL of glucose. Recovered macro-
phages were suspended in culture medium and they
were then deposited in flat-bottom 96-well plates (Tan-
ner, Switzerland) at a density of 4ꢃ104 cells/well.
Twenty-four hours after replating, macrophages were
infected at a multiplicity of five amastigotes per host cell
and were incubated at 34 ꢀC, which is the permissive
Figure 1. The effect of compound 42 on L. amazonensis-infected mac-
rophages. Bone marrow-derived macrophages were infected with L.
amazonensis amastigotes. After 24 h of infection and for an additional
30-h period, infected cells were exposed either to DMSO (a) or to
compound 42 (b, 3 mM). (a) Amastigotes are lodged within large PV
and 100% of macrophages are infected. (b) PV and parasites are no
more detectable.

5022
M. A. Fakhfakh et al. / Bioorg. Med. Chem. 11 (2003) 5013–5023
temperature for the survival and multiplication of LV79
strain amastigotes. In most instances, more than 95% of
the macrophages were found to be infected.
In vitro activity against T. brucei trypomastigotes
Bloodstream forms of T. brucei (strain S427) were cul-
tivated in HMI-9 medium and set out in a 96-well
microplate at a density of 2ꢃ104 trypomastigotes per
100 mL of medium per well. Trypanosomes were incu-
bated at 37 ꢀC in the presence of various drug con-
centrations for 48 h, in triplicate. Alamar Blue1 (20 mL)
was then added to each well, and the plates incubated
for a further 24 h period. Drug effects on parasite
multiplication were estimated by monitoring the absor-
bance of AlamarBlueTM at two wavelengths 570 and
600 nm. The IC50, values were calculated from the dose–
response curves.
For all drugs, stock solutions were prepared in DMSO
at a concentration of 500 mg/mL. Two-fold serial dilu-
tions were made from 250 mg/mL in culture medium
supplemented with 0.5% DMSO final. Twenty-four
hours after infection, freshly prepared drugs are added
to the infected cultures in triplicate. The first final drug
concentration is 50 mg/mL and the final DMSO con-
centration is 0.1%. This DMSO concentration was
proven to have no effect on control cultures.
Thirty hours after drug addition, infected cultures were
examined using an inverted phase contrast Zeiss micro-
scope (magnification of 400). Toxic effects in the
macrophages were evidenced by the change in morpho-
logical features i.e., loss of refringency, vacuolation of
cytoplasm or loss of cytoplasmic material.
Antiviral assays
The lymphocytes cell line CEM4fx was maintained in
RPMI-1640 medium supplemented with 10% fetal calf
serum. Hela-CD4+-ßGal (P4) cells were grown in
DMEM with 10% fetal calf serum and 0.5 mg/mL
geneticin. Cell-free viral supernatants were obtained by
trans-fection of P4 cells with HIV-1 PLN4-3 genomic
clone.23 CEM4fx cells were plated in triplicate on a 96-
well plates (100 mL) and infected with cell-free virus.
Viral supernatants were removed 2 h after infection and
drugs dissolved in DMSO were added in fresh medium.
Infected cells were grown in the presence of drugs for 3
days. Supernatants were then collected at t=72 h and
used to infect P4 cells. P4 cultures were incubated for 24
h and subsequently lyzed in a phosphate buffer con-
taining 50 mM 2-mercaptoethanol, 10 mM MgSO4, 25
mM EDTA, 0.125% NP40. 20 mL of lyzate was incu-
bated with 100 mL of CPRG-containing buffer. The red
staining intensity was quantified on a multiscan photo-
meter at 570 nm. CEM4fx cell viability was estimated
by the MTT (Sigma) assay after 3 days treatment with
drugs (20 mL). A solution of (7.5 mg/mL) in phosphate
buffer was added to each well of microtiter trays. Plates
were further incubated at 37 ꢀC in a CO2 incubator for 4
h. Solubilization of formazan crystals was achieved by
adding 100 mL of 10% SDS, 10 mM HCI. Absorbance
was read in a multiscan photometer at 570 nm. Experi-
ments were performed in triplicate and averaged. The
anti-integrase compound L731,988 was used as a posi-
tive control for antiviral activity.
Leishmanicidal effects of drugs are detectable looking at
the regression of parasitophorous vacuoles and the
overall decrease in parasite number. For some drugs,
complete clearance of amastigotes was achieved.
In vitro activity against intracellular L. infantum
amastigotes
L. infantum (strain MHOM/MA/BE/67) was main-
tained in female Golden hamsters (Iffa Credo, France)
by passage every 6–8 weeks. Peritoneal macrophages
were harvested from female BALB/c mice by peritoneal
washing 24 h after injection of a 2% soluble starch
solution (Merck, France). After two washes in PBS the
exudate cells were dispensed into 16-well Lab-tek tissue
culture slides (Nunc Inc. IL) at 4ꢃ104/well in a volume
of 200 mL of RPMI-1640 medium (Sigma-Aldrich,
France) plus 10% FCS. After 24 h, amastigotes of L.
infantum (derived from the spleen of an infected ham-
ster) were added at an infection ratio of 10/1 together
with 2-fold serial dilutions of the drugs to be tested and
the cultures were incubated at 37 ꢀC in 5% CO2–95%
air. Five days later, cells were fixed by methanol and
stained with Giemsa. Leishmanicidal activity of the
drugs was determined by counting the percentage of
infected macrophages in treated and untreated cultures.
N-methylglucamine, antimonate (Aventis, France) was
used as a control drug.
Acknowledgements
The authors wish to thank all persons who have helped
in the field study, and especially the students who have
re-prepared some compounds (M. Dos Santos, B.
Vidovic, G. Epane, L. Courpotin, H. A. Tran), and J.
C. Jullian for NMR experiments. Particularly, we wish
to thank Dr. J. R. L. Pink of Special Programme for
Research and Training in Tropical Diseases (TDR) and
Programme Drug Discovery (DDR).
In vitro activity against intracellular T. cruzi amastigotes
Primary peritoneal macrophages were seeded in 96-well
microtiter plates at 3ꢃ104 cells/well/100 mL in RPMI
1640 medium with 10% FCS and 2 mM l-glutamine.
After 24 h, 105 trypomastigotes of T. cruzi (Tulahuen
strain) were added in 100 mL per well with 2-fold serial
dilutions of the drugs, in triplicate. The cultures were
incubated at 37 ꢀC in 5% CO2–95% air for 4 days. The
IC50 were determined spectrophotometrically at 570 nm
following addition of the substrate Chlorophenolred-b-
d-galactopyranoside (CPRG)/Nonidet during 4 h from
the dose response curve.
References and Notes
1. Antoine, J. C.; Jouanne, C.; Ryter, A. Parasitology 1989,
99, 1.

M. A. Fakhfakh et al. / Bioorg. Med. Chem. 11 (2003) 5013–5023
5023
2. Kirchoff, L. V. N. Engl. J. Med. 1993, 329, 639.
3. Urbina, J. Mem. Inst. Oswaldo Cruz 1999, 94 (Suppl. 1), 349.
4. Andrade, A. L. S.; Zicker, F.; Oliveira, R. M.; Silva, S. A.;
Luquetti, A.; Travassos, L. R.; Almeida, I. C.; Andrade, S. S.;
Andrade, J. G.; Martelli, C. M. T. Lancet 1996, 348, 1407.
5. World Health Organization. Leishmaniasis and Leishma-
nia/HIV co-Infection: http://www.who.int/emc-documents/
surveillance/docs/whocdsisr2001.html/Leishmniasis/Leishman
iasis.htm.
6. World Health Organization. Weekly Epidemiol. Rec. 1997,
72, 357.
7. Sundar, S.; Rosenkaimer, F.; Makharkia, M. K.; Goyal,
A. K.; Mandal, A. K.; Voss, A.; Hilgard, P.; Murray, H. W.
Lancet 1998, 352, 1821.
de Bilbao, N.; Torres, S.; Schinini, A.; Fournet, A. Phytother.
Res. 2001, 15, 630.
13. Fakhfakh, M. A.; Franck, X.; Fournet, A.; Hocquemiller,
R.; Figadere, B. Tetrahedron Lett. 2001, 42, 3847.
14. Fakhfakh, M. A.; Franck, X.; Hocquemiller, R.; Figadere,
B. J. Organomet. Chem. 2001, 624, 131.
15. Fakhfakh, M. A.; Franck, X.; Fournet, A.; Hocquemiller,
R.; Figadere, B. Synth. Commun. 2002, 32, 2863.
16. Quintin, J.; Franck, X.; Hocquemiller, R.; Figadere, B.
Tetrahedron Lett. 2002, 43, 3547.
17. Wolday, D.; Berhe, N.; Akuffo, H.; Britton, S. Parasitol.
Today 1999, 15, 182.
18. Bernier, R.; Turco, S. J.; Olivier, M.; Tremblay, M. J.
Virol. 1995, 69, 7282.
8. Fournet, A.; Hocquemiller, R.; Roblot, F.; Cave, A.;
Richomme, P.; Bruneton, J. J. Nat. Prod. 1993, 56, 1547.
9. Fournet, A., Angelo Barrios, A., Munoz, V., Hocquemiller,
R., Roblot, F., Bruneton, J., Richomme, P., Gantier, J.C.
PCT/FR92/00903, 29 September, 1992.
19. Fournet, A.; Mahieux, R.; Fakhfakh, M. A.; Franck, X.;
Hocquemiller, R.; Figadere, B. Bioorg. Chem. Lett. 2003, 13,
891.
20. Uchiyama, T.; Yodoi, J.; Sagawa, K.; Takatsuki, K. Blood
1977, 50, 481.
10. Fournet, A.; Ferreira, M. E.; Torres de Ortiz, S.; Fuentes,
S.; Nakayama, H.; Rojas de Arias, A.; Schinini, A.; Hocque-
miller, R. Antimicrob. Agents Chemother. 1996, 40, 2447.
11. Fournet, A.; Gantier, J. C.; Gautheret, A.; Leysalles, L.;
Munos, M. H.; Mayrargue, J.; Moskowitz, H.; Cave, A.;
Hocquemiller, R. J. Antimicrob. Chemother. 1994, 33, 537.
12. Nakayama, H.; Ferreira, M. E.; Rojas de Arias, A.; Vera
21. Gessain, A.; Barin, F.; Vernant, J. C.; Gout, O.; Maurs,
L.; Calender, A.; de The, G. Lancet 1985, Ii, 407.
22. Hallock, H. F.; Cardellina, J. H., II; Schaˆ ffer, M.; Bring-
mann, G.; Francois, G.; Boyd, M. R. Bioorg. Med. Chem.
Lett. 1998, 8, 1729.
23. Jeanson, L.; Subra, F.; Vaganay, S.; Hervy, M.; Marangoni,
E.; Bourhis, J.; Mouscadet, J. F. Virology 2002, 300, 100.