SAR Studies of Dihydro-β-agarofuran Sesquiterpenes as Inhibitors of the Multidrug-Resistance Phenotype in a Leishmania tropica Line Overexpressing a P-Glycoprotein-Like Transporter

Article abstract of DOI:10.1021/jm0309699

A series of dihydro-β-agarofuran sesquiterpenes isolated from the leaves of Maytenus cuzcoina (1-10) or semisynthetic derivatives (11-30) have been tested on a multidrug-resistant Leishmania tropica line overexpressing a P-glycoprotein-like transporter to

Full text of DOI:10.1021/jm0309699

576
J . Med. Chem. 2004, 47, 576-587
SAR Stu d ies of Dih yd r o-â-a ga r ofu r a n Sesqu iter p en es a s In h ibitor s of th e
Mu ltid r u g-Resista n ce P h en otyp e in a Leish m a n ia tr opica Lin e Over exp r essin g
a P -Glycop r otein -Lik e Tr a n sp or ter
Fernando Corte´s-Selva,† Mercedes Campillo,‡ Carolina P. Reyes,§ Ignacio A. J ime´nez,§ Santiago Castanys,†
Isabel L. Bazzocchi,§ Leonardo Pardo,*,‡ Francisco Gamarro,*,† and Angel G. Ravelo*,§
Instituto de Parasitolog´ıa y Biomedicina “Lo´pez-Neyra”, Consejo Superior de Investigaciones Cient´ıficas, c/ Ventanilla 11,
18001 Granada, Spain, Laboratori de Medicina Computacional, Unitat de Bioestadistica, Institut de Neurocie`ncies,
Universitat Auto`noma de Barcelona, 08193 Cerdanyola del Valle`s, Barcelona, Spain, and Instituto Universitario de
Bio-Orga´nica “Antonio Gonza´lez”, Universidad de La Laguna, Avenida Astrof´ısico Francisco Sa´nchez, 2,
38206 La Laguna, Tenerife, Spain
Received J uly 23, 2003
A series of dihydro-â-agarofuran sesquiterpenes isolated from the leaves of Maytenus cuzcoina
(1-10) or semisynthetic derivatives (11-30) have been tested on a multidrug-resistant
Leishmania tropica line overexpressing a P-glycoprotein-like transporter to determine their
ability to revert the resistance phenotype and to modulate intracellular drug accumulation.
Almost all natural compounds showed potent reversal activity with different degrees of
selectivity. Compounds 2, 7, and 8 are the most effective reversal agents tested against the
multidrug resistance phenotype of Leishmania. Three-dimensional quantitative structure-
activity relationships using the comparative molecular similarity indices analysis (CoMSIA)
were employed to characterize the steric (contribution of 5.4%), electrostatic (58.9%), lipophilic
(10.0%), and hydrogen-bond-donor (13.3%) and -acceptor (7.5%) requirements of these sesqui-
terpenes as modulators at the P-glycoprotein-like transporter. The most salient features of
these requirements are the H-bond interaction between the substituents at the C-2 and C-6
positions with the receptor.
In tr od u ction
molecular mechanics procedures showed that the trans-
fused A and B rings formed a trans chair-chair decalin
system, slightly distorted by the presence of the 1,3-
diaxial bond responsible for the tetrahydrofuran C ring,
practically perpendicular to the plane formed by carbons
C-5, C-7, C-8, and C-10 (Figure 1).2
Natural products play important roles in both drug
discovery and chemical biology. In fact, many approved
therapeutics as well as drug candidates are derived from
natural sources. Additionally, natural products have
been extensively used to elucidate complex cellular
mechanisms, including signal transduction and cell
cycle regulation, leading to the identification of impor-
tant targets for therapeutic intervention. As a result of
recent advances in biology, there is now an increased
demand for new natural product-like small molecules.
Specifically, the fields of genomics and proteomics
promise the rapid identification of a large numbers of
gene products for which small molecule modulators will
be of both biological and medicinal interest.1
Over the last 30 years, a large number of secondary
metabolites exhibiting a wide range of bioactivity have
been isolated from Celastraceae, the sesquiterpenes
being the most widespread and characteristic metabo-
lites of this family. Generally, they occur as polyesters
of variously polyoxygenated tricyclic scaffolds, all based
on a core C15 skeleton known as dihydro-â-agarofuran
[5,11-epoxy-5â,10R-eusdesm-4(14)-ene], and they are
considered to be chemotaxonomic indicators of the
family. X-ray data and conformational studies using
The interest generated by sesquiterpenes from Celas-
traceae has increased in line with the complexity of the
substances isolated and, more importantly, with their
wide range of biological actions, suggesting that deriva-
tives of this sesquiterpene motif may be capable of
interacting with a variety of cellular targets. Recently,
sesquiterpenes have shown immunosuppressive,3 anti-
HIV,4 reversal of multidrug resistance (MDR) pheno-
type,5,6 and antitumor-promoting activities.7 In addition,
the fact that many of the compounds are active in cell-
based assays suggested that products with a dihydro-
â-agarofuran unit remain sufficiently lipophilic to cross
cell membranes, a key feature of any biologically
relevant compound. On the basis of the biological and
structural properties, we propose the selection of ses-
quiterpene polyesters with a dihydro-â-agarofuran skel-
eton as a privileged structure. This term describes
selected structural types, like polycyclic heteroatomic
systems, capable of orienting varied substituent pat-
terns in a well-defined three-dimensional space and bind
to multiple, unrelated classes of protein receptors as
high-affinity ligands.8
* Corresponding authors. L.P.: tel, 34-93-5812797; fax, 34-93-
5812344; e-mail, leonardo.pardo@uab.es. F.G.: tel, 34-958-805185; fax,
34-958-203323; e-mail, gamarro@ipb.csic.es. A.G.R.: tel, 34-922-
318576; fax, 34-922-318571; e-mail, agravelo@ull.es.
† Consejo Superior de Investigaciones Cient´ıficas.
‡ Universitat Auto`noma de Barcelona.
Drug resistance has emerged in the last years as one
of the major impediments for the treatment of diseases
produced by protozoan parasites. ABC (ATP-binding
cassette) transporters are involved, at least in vitro, in
§ Universidad de La Laguna.
10.1021/jm0309699 CCC: $27.50 © 2004 American Chemical Society
Published on Web 12/30/2003

SAR of Dihydro-â-agarofuran Sesquiterpenes
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 3 577
the C-2, C-6, C-9, and C-15 positions of the dihydro-â-
agarofuran skeleton, to characterize the steric, electro-
static, hydrophobic, and hydrogen-bond-donor and -ac-
ceptor requirements needed at the active sites of the
receptors for ligand recognition. One of the unique
features of CoMFA is its ability to represent the 3D-
QSAR model in terms of color contour maps that depict
locations on the ligands where structural modifications
might enhance their biological activity (e.g. binding
affinity). These maps can serve as a guide when design-
ing analogues within the same series of compounds that
possess desirable biological properties.
F igu r e 1. Dihydro-â-agarofuran skeleton.
Meth od s
3D-QSAR/CoMSIA Meth od . The equilibrium struc-
tures of the sesquiterpenes were obtained by full ab
initio geometry optimization with the 3-21G basis set.
A critical step in CoMSIA is to select a proper alignment
rule. The entire set of sesquiterpene analogues was
oriented in space by superimposing the common and
rigid dihydro-â-agarofuran skeleton. The percentage of
inhibition in both MDR and wild-type (WT) parasites
is a function of both the stabilization of the complexes
between the ligand molecules and the protein receptor
and the solvation energy of the ligands. Thus, the QSAR
Tables 3 and 4 consist the percentage of inhibition
values (dependent variable) and the electrostatic, steric,
hydrophobic, and hydrogen-donor and hydrogen-accep-
tor fields (independent variables), to mimic the stabi-
lization energy of the receptor-ligand complex, and the
solvation energy (independent variable). The atom-
centered atomic charges used in CoMSIA to evaluate
the electrostatic contributions were computed from the
molecular electrostatic potential22 using 6-31G* basis
set, a common procedure for the simulation of proteins,
nucleic acids, and organic molecules.23 Solvation free
energies (∆Gsolv) of sesquiterpenes were calculated with
the PM3-SR5.42R procedure within the AMSOL 6.7.2
program.24 The potential fields were calculated at each
lattice intersection of a regularly spaced grid of 2 Å. A
sp3 carbon atom with a van der Waals radius of 1.52 Å
carrying a charge of +1.0 served as a probe atom to
calculated the fields with an attenuation factor of 0.3.25
Partial least squares (PLS) analysis26-29 was used to
derive linear equations from the resulting matrixes.
Leave one out (LOO) cross-validation was employed
to select the number of principal components and
to calculate the cross-validated statistics. The final
CoMSIA model was generated using non-cross-valida-
tion and the number of components suggested by the
LOO validation run. The 3D-QSAR/CoMSIA study was
carried out with the QSAR module of the SYBYL 6.6
program,30 using default parameters. All the quantum
mechanical calculations were performed with the Gauss-
ian 98 system of programs.31
F igu r e 2. Sesquiterpenes assayed for the chemosensitization
of a MDR L. tropica line.
the resistance to many antiparasitic drugs.6,9-11 In
addition, new putative antiparasitic drugs such as
anthracyclines,12 taxol,13 or azoles,14,15 are known sub-
strates of ABC transporters, and thus could induce an
MDR phenotype. One ABC transporter homologous to
the human MDR-P-glycoprotein (Pgp) confers an MDR
phenotype in Leishmania6,11,16 similar to that character-
ized in cancer cells. Moreover, Leishmania Pgp also
confers resistance against alkyllysophospholipids, such
as miltefosine, the most promising leishmanicidal agent.6
Therefore, the development of inhibitors of these trans-
porters is of high clinical relevance. The main difference
between the MDR phenotypes conferred by Pgp in
mammalian cells and the Pgp-like transporter in Leish-
mania is the absence of significant reversal effects by
classical Pgp inhibitors such as verapamil, among
others.11 This observation led us to explore new inhibi-
tors against this protein. Agarofuran sesquiterpenes are
new, promising reversal agents; they have efficiently
overcome the MDR phenotype in Leishmania,5,6,17 in-
cluding resistance to alkyllysophospholipids, probably
due to their binding to the transmembrane domains of
the Leishmania Pgp-like transporter and blocking drug
efflux.
As a part of an intensive investigation into active
metabolites as reversal agents of MDR phenotype, we
describe in this paper the biological evaluation as
modulators of the MDR phenotype of a Leishmania
tropica line and a three-dimensional quantitative struc-
ture-affinity relationship analysis (3D-QSAR)18 of 30
(1-30) sesquiterpenes with a dihydro-â-agarofuran
skeleton (Figure 2), among them 10 (1-10) that were
isolated from Maytenus cuzcoina7 and 20 semisynthetic
derivatives (11-30) not previously described. The 3D-
QSAR was performed using an extension of the CoMFA
methodology,19,20 the comparative molecular similarity
indices analysis (CoMSIA).21 This methodology was
applied to these sesquiterpenes bearing substituents at
Resu lts a n d Discu ssion
The products considered for biological investigation
(Figure 2) were isolated from M. cuzcoina (compounds
1-107) or prepared by standard methods (compounds
11-30), as described in Scheme 1.
Compounds 11, 13, and 22 (Figure 2) were prepared
by partial basic hydrolysis of 1, 2, and 4, respectively.

578 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 3
Corte´s-Selva et al.
Ta ble 1. Effect of Sesqu iter p en es on th e Gr ow th of WT a n d MDR L. tr opica Lin es
growth inhibitiona (%)
15 µΜ
7 µΜ
3 µΜ
1 µΜ
compd
WT
MDR
WT
MDR
WT
MDR
WT
MDR
b
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
23.4 ( 3.1
93.5 ( 1.7
94.4 ( 1.8
28.7 ( 8.1
92.7 ( 5.5
95.8 ( 1.6
94.7 ( 2.5
94.3 ( 3.0
90.9 ( 5.0
89.9 ( 4.8
-
18.1 ( 6.4
86.2 ( 0.2
90.9 ( 2.1
9.1 ( 6.3
93.7 ( 1.5
94.8 ( 4.5
92.0 ( 1.6
89.5 ( 2.0
89.9 ( 5.2
89.0 ( 1.3
-
13.8 ( 4.9
52.9 ( 10.1
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
10.0 ( 9.8
15.2 ( 3.5
9.7 ( 6.5
-
83.5 ( 9.2
24.5 ( 0.7
-
-
-
-
-
35.3 ( 5.8
21.4 ( 8.6
16.5 ( 4.1
74.8 ( 5.9
11.7 ( 4.7
26.8 ( 2.9
20.6 ( 2.5
15.9 ( 5.7
89.1 ( 3.4
30.3 ( 2.5
31.5 ( 6.2
24.9 ( 8.0
21.8 ( 7.7
87.6 ( 3.8
25.7 ( 7.5
7.2 ( 6.0
-
-
79.6 ( 6.8
25.7 ( 7.6
15.2 ( 5.6
11.0 ( 6.6
6.9 ( 6.9
77.5 ( 5.0
18.0 ( 7.1
24.6 ( 7.1
16.4 ( 4.9
14.8 ( 2.8
66.9 ( 9.9
20.3 ( 5.1
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
44.5 ( 2.1
-
28.5 ( 13.4
-
-
-
-
-
-
-
10.9 ( 7.9
50.6 ( 9.3
-
-
-
-
-
17.1 ( 6.7
-
-
-
-
-
-
-
80.5 ( 4.5
95.2 ( 1.3
96.7 ( 0.0
7.1 ( 2.7
98.4 ( 0.7
59.5 ( 2.8
81.9 ( 3.3
45.0 ( 5.2
28.0 ( 9.2
9.2 ( 4.2
8.6 ( 3.5
77.6 ( 4.5
36.9 ( 4.6
11.8 ( 0.7
51.1 ( 4.1
29.2 ( 10.6
50.7 ( 8.5
82.5 ( 5.9
-
8.8 ( 1.3
-
10.6 ( 0.6
16.5 ( 8.8
-
-
-
44.5 ( 1.0
18.1 ( 7.4
-
-
17.2 ( 0.6
85.0 ( 3.2
15.5 ( 3.0
41.1 ( 6.7
14.3 ( 7.3
10.4 ( 6.3
-
35.4 ( 8.6
11.9 ( 6.7
-
-
-
8.0 ( 4.6
11.1 ( 9.1
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
25.9 ( 3.5
9.7 ( 2.3
-
11.2 ( 4.7
-
-
-
-
-
-
-
10.8 ( 2.5
26.4 ( 1.3
13.5 ( 4.4
6.4 ( 3.7
a
WT and MDR parasites were exposed to 15, 7, 3, and 1 µΜ of different sesquiterpenes, in the absence or presence of 150 µΜ DNM,
respectively. The results are expressed as percentage of growth inhibition relative to control growth in the absence of sesquiterpene. The
data shown are the average of three independent experiments (SD. Growth inhibition values between 0 and 6% are indicated by a
dash for simplicity.
b
Ta ble 2. Effect of Sesquiterpenes 2, 7, and 8 on IC50 Values for DNM in a MDR L. tropica Line
MDR
WT
c
7 µM
15 µM
2
7
8
2
7
8
c
IC50a (µM)
30.0 ( 3.6
33.3 ( 9.8
39.0 ( 6.0
9.0 ( 1.4
11.0 ( 2.8
20.0 ( 4.2
290 ( 24.8
3.8 ( 0.9
RIb
7.8
8.7
10.2
2.3
2.9
5.2
75.5
1
a
Parasites were exposed to increasing concentrations of DNM in the presence or in the absence of two different concentrations (7 and
15 µM) of sesquiterpenes. The results are expressed as the concentration of DNM necessary to inhibit the parasites growth by 50%. The
b
data shown are the average of three independent experiments (SD. Resistance index. Ratio among IC50 of the MDR line and IC50 of the
WT line. c No sesquiterpenes were used.
The furoyl group at C-6 was regioselectively cleaved
with 0.1 M NaHCO3 without affecting the other esters.
The corresponding esters 12, 14-21, and 23-30 were
prepared by treatment of 11, 13, and 22 with appropri-
ate acyl chloride or anhydride. Due to the dissimilar
reactivity of acylating reagents, different reaction condi-
tions were adopted. The structures of compounds 11-
from δ 4.56 in 11 to δ 5.61 in 12. The structures of
compounds 14-21 and 23-30 were determined in a
similar manner.
The reversal effects of dihydro-â-agarofuran sesqui-
terpenes 1-30 in an MDR L. tropica line grown in the
presence of daunomycin (DNM) were studied by using
an MTT-based assay. Their intrinsic cytotoxicity was
determined by using the same concentration of modula-
tors in both parental WT (Table 1) and MDR parasites
in the absence of DNM (data not shown). The results
obtained were similar in both parasite lines, suggesting
that MDR parasites do not show cross-resistance against
agarofuran sesquiterpenes. Table 1 shows that after 72
h incubation of MDR parasites in the presence of 150
µM DNM with increasing amounts of sesquiterpenes, a
concentration-dependent growth inhibition (GI) was
observed as compared with MDR control parasites,
grown with the same DNM concentration but in the
absence of a modulator.
1
30 were elucidated by spectroscopic data. Thus, the H
NMR spectrum of compound 11 indicated it is the
6-defuroyl derivative of 1, since the signal of the H-6
proton was shifted from δ 5.43 in 17 to δ 4.56 in 11 and
the signals for the aromatic protons of the furoyl group
at δ 6.83, 7.44, and 8.18 were not observed. In the same
way, the structures of compounds 13 and 22 were
determined as the 6-defuroyl derivative of 2 and 4,
respectively. Spectroscopic data showed that compound
12 is the 6-acetyl derivative of 11, because of the
presence of an additional acetate methyl signal at δ 2.14
and the fact that the signal of the H-6 proton was shifted

SAR of Dihydro-â-agarofuran Sesquiterpenes
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 3 579
Ta ble 3. Experimental and CoMSIA Predicted Percent Inhibition of Grown in MDR and WT L. tropica Lines for Sesquiterpenes
1-30 (3 µM) and Their Solvation Free Energy (∆Gsolv
)
MDR line
WT line
pred
compd
C-2
OFu
OAc
OH
OBz
OPr
OMeBut
OAc
OAc
H
OAc
OFu
OFu
OAc
OAc
OAc
OAc
OAc
OAc
OAc
OAc
OAc
OBz
OBz
OBz
OBz
OBz
OBz
OBz
OBz
OBz
C-6
C-9
C-15
CH3
CH3
CH3
CH3
CH3
CH3
CH2OAc
CH2OAc
CH3
CH2OAc
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
exp
pred
exp
∆Gsolv
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
OFu
OFu
OFu
OFu
OFu
OFu
OFu
OBz
OFu
OAc
OH
OAc
OH
OAc
OMeBut
OLau
ONap
OFu
OFu
OFu
OFu
OFu
OFu
OFu
OBz
OFu
OAc
OFu
OFu
OFu
OFu
OFu
OFu
OFu
OFu
OFu
OFu
OFu
OFu
OFu
OFu
OFu
OFu
OFu
OFu
OFu
OFu
52.9
83.5
4.9
74.8
89.1
87.6
79.6
77.5
66.9
0.0
1.5
0.0
5.5
6.0
1.4
8.8
16.5
44.5
1.7
35.4
4.9
11.1
4.1
4.1
3.0
2.4
11.2
5.9
0.8
51.6
85.1
4.9
71.8
90.8
87.8
79.7
77.6
67.2
-
6.2
-2.3
4.7
6.8
2.6
6.8
15.5
43.0
1.5
35.0
-
7.2
5.8
3.0
4.9
3.5
13.8
2.3
2.3
16.5
15.9
21.8
3.1
6.9
14.8
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
0.0
14.3
-
2.3
16.1
15.7
21.8
3.1
-15.5
-13.7
-15.5
-14.0
-12.8
-12.0
-17.1
-16.3
-11.4
-22.4
-15.7
-15.6
-13.8
-13.8
-11.5
-11.4
-13.4
-11.2
-10.5
-13.7
-15.7
-14.2
-14.1
-11.8
-11.7
-13.7
-11.5
-10.7
-14.1
-16.0
a
6.9
14.8
-
0.0
-0.4
-0.2
-0.2
0.0
0.2
-0.1
0.0
-0.1
0.4
0.1
0.2
0.6
0.1
-0.2
0.1
OPiv
OTFAc
O4-MeO-Bz
O4-NO2-Bz
OH
OAc
OMeBut
OLau
ONap
OPiv
OTFAc
O4-MeO-Bz
O4-NO2-Bz
12.6
6.2
1.3
0.0
0.1
-0.3
-0.1
13.5
13.6
a
A dash indicates that the compound was not included in the CoMSIA model.
Ta ble 4. Statistical Results of Percent Inhibition in MDR and
WT L. tropica Lines in CoMSIA Models
in the MDR line using different concentrations of the
most active and selective sesquiterpenes (Table 2). In
fact, 15 µM of 2, 7, and 8 compounds reduced the
resistance index (IC50 ratio among MDR and WT lines)
from 75.5 to 2.3, 2.9, and 5.2, respectively.
MDR line
WT line
q2 a
0.790
19
28
0.997
155.752
58.9
5.4
13.3
7.5
10.0
4.9
0.715
18
28
0.999
426.658
54.5
8.2
9.6
7.3
17.1
3.3
Nb
nc
DNM resistance in the MDR L. tropica line is related
to a decreased intracellular drug accumulation, mainly
due to the Pgp-like transporter overexpression.11 To
analyze if the reversal effect observed by some sesqui-
terpenes correlated with an increased drug accumula-
tion, as a consequence of the Pgp inhibition, we studied
by laser flow cytometry their effect on calcein (CAL)
accumulation as described.5 Flow cytometry analysis
(Figure 3) demonstrated that, as expected, the MDR line
accumulated a significantly lower amount of dye, ex-
pressed as mean fluorescence channel (m ) 384), than
the WT line (m ) 582). Coincubation of the MDR
parasites with 5 µM of one of the most active sesqui-
terpenes (compound 2) resulted in a significant shift of
the peak of fluorescence distribution to the right, almost
to the WT level (m ) 528); this reversal effect was a
consequence of an increased CAL accumulation, prob-
ably due to Pgp-like transporter inhibition. The same
concentration of sesquiterpenes that gave a moderate
reversal effect only produced a slight increase of the
intracellular dye (m ) 382-399). Indeed, Figure 3 shows
that different sesquiterpenes with distinct reversal
efficiencies restored the dye accumulation in the MDR
line with an order of efficiency similar to that obtained
with the DNM chemosensitization experiments, 7 > 2
. 14 > 3 > 13, with no significant effects in the WT
line (not shown). The effects of sesquiterpenes as
reversal agents of the MDR phenotype seem not to be
r2 d
F
electrostatice
sterice
H-bond donore,f
H-bond acceptore,f
hydrophobicitye
solvatione
a
b
Leave-one-out correlation coefficient. Optimal number of
principal components. c Number of compounds. Non-cross-vali-
dated correlation coefficient. e Percentage of contribution. f On the
receptor.
d
Sch em e 1
The chemosensitization to 150 µM DNM was very
efficient for compounds 2 and 4-8. In this form, 3 and
7 µM of these sesquiterpenes produced 75-89% GI and
90-95% GI, respectively. All compounds tested showed
a low intrinsic cytotoxicity at concentrations below 15
µM. All the chemical derivatives (compounds 11-30)
were inactive or less active than the natural ones. To
confirm the ability of sesquiterpenes to overcome the
drug resistance, we determined the IC50 values for DNM

580 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 3
Corte´s-Selva et al.
F igu r e 3. Differential modulation by sesquiterpenes of CAL
accumulation in a MDR L. tropica line. Fluorescence intensity
histograms were obtained by flow cytometry after incubation
for 1 h at 28 °C with 2 µM of CAL-acetoxymethyl ester (CAL-
AM) in the presence or absence of 5 µM sesquiterpenes. A total
of 10 000 cells were counted for each histogram. Experiments
were repeated three times and gave essentially the same
profiles as the ones shown here. WT (white bar) and MDR
(black bar) parasites were incubated with CAL-AM and used
as control of CAL accumulation. MDR parasites (grey bars)
incubated with CAL-AM in the presence of different sesqui-
terpenes (2, 3, 7, 13, and 14).
F igu r e 4. Plot of the predicted versus the experimental
percent inhibition for the MDR and WT L. tropica lines by
applying CoMSIA models.
and hydrophobic (iv) maps for the MDR (A) and WT (B)
models, using compounds 7 (Figure 5) and 29 (Figure
6) as the reference structures. The color code of the maps
is as follows: (i) areas where a high electron density
provided by the ligand increases or decreases GI are
shown in red or blue, respectively; (ii) green and yellow
areas depict zones of the space where occupancy by the
ligands increases or decreases GI, respectively; (iii)
areas where H-bond donors on the receptor are pre-
dicted to enhance or disfavor GI are shown in magenta
and orange, respectively, whereas cyan contours show
areas where H-bond-acceptor zones on the receptor are
predicted to increase GI; and (iv) yellow and white areas
define regions of space where hydrophobes and hydro-
philic groups, respectively, are predicted to enhance GI.
related with the binding to the cytosolic nucleotide
domains of Pgp-like transporter,17 suggesting the bind-
ing to the transmembrane domains and the block of
DNM efflux.
3D-QSAR/CoMSIA analysis was performed on the
sesquiterpenes 1-30 (Table 3). Compounds 10 and 21
for the MDR line and compounds 2 and 10 for the WT
line were not included in the CoMSIA model, because
their residual values were greater than two standard
deviations. Thus, the % GI produced by 3 µM of
sesquiterpenes 1-9 and 11-30 in the MDR line and
by sesquiterpenes 1, 3-9, and 11-30 in the WT line
were related to the independent variables by the PLS
methodology, to evaluate the potency and selectivity of
these compounds.
The CoMSIA analysis allowed us to rationalize the
observed data of GI. The H-bond donor (the second field
contribution in the MDR model, 13.3%; see Table 4) map
(Figures 5Aiii and 6Aiii) shows three magenta areas: at
the upper-left side (substitutions at C-2), on the right
side (substitutions at C-6 and C-15), and at the bottom-
center (substitution at C-6) of the figures. Structure-
activity relationship (SAR) studies of the A-ring of the
sesquiterpenes 1-6 and 9 suggest that a substituent
at the C-2 position seems to be essential for the reversal
activity in the MDR line. The introduction of the
carbonyl group of the ester moiety (OFu, 1; OAc, 2; OBz,
4; OPr, 5; and OMeBut, 6), capable of acting as a H-bond
acceptor in the H-bond with the receptor, produces a
10-fold higher chemosensitization with respect to the
presence at the same position of a hydroxyl group (OH,
3) [e.g. GI(3) ) 5 vs GI(1) ) 53, GI(2) ) 83, GI(4) ) 75,
GI(5) ) 89, and GI(6) ) 88] (Table 1). Clearly, there is
a contact of this carbonyl group with the magenta area,
which suggests a direct interaction with the receptor.
It seems, on the basis of both the experimental data and
the CoMSIA model, that the hydroxyl group of 3 or the
equivalently positioned ether group of 1, 2, 4, 5, or 6 is
Table 4 shows the statistical properties of the model.
From a statistical viewpoint, the high values of the
obtained cross-validated correlation coefficient q2 (0.790
and 0.715 for MDR and WT lines, respectively) reveal
that both models are useful tools for predicting the
biological activity. In addition, the models yielded
conventional r2 of 0.997 (19 principal components) and
0.999 (18 principal components) for the MDR and WT
lines, respectively. The theoretically predicted and
experimentally determined inhibition values, for the
whole set of compounds, are listed in Table 3 and plotted
in Figure 4. The relative contributions in the MDR (58.9:
5.4:13.3:7.5:10.0:4.9) and WT (54.5:8.2:9.6:7.3:17.1:3.3)
CoMSIA models for the electrostatic, steric, hydrogen-
bond-donor and -acceptor, hydrophobic, and solvation
terms, respectively, are also shown in Table 4. Notably,
the hydrophobic term is higher in the WT than in the
MDR model, whereas the steric and the hydrogen-bond-
donor contributions are higher in the MDR model. The
other contributions remain similar in both models.
Figures 5 and 6 illustrate the CoMSIA electrostatic
(i), steric (ii), hydrogen-bond-donor and -acceptor (iii),

SAR of Dihydro-â-agarofuran Sesquiterpenes
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 3 581
F igu r e 5. Electrostatic (i), steric (ii), hydrogen-bond-donor and -acceptor (iii), and hydrophobic (iv) maps for MDR (A) and WT
(B) L. tropica lines from CoMSIA models. Compound 7 is shown as reference structure. The color code is as follows: (i) areas
where a high electron density provided by the ligand increases (red) or decreases (blue) GI; (ii) green and yellow areas depict
zones of the space where occupancy by the ligands increases or decreases GI, respectively; (iii) areas where H-bond donors on the
receptor are predicted to enhance (magenta) or disfavor (orange) GI, whereas cyan contours show areas where H-bond acceptor
zones on the receptor are predicted to increase GI; and (iv) yellow and white areas defines regions of space where hydrophobes
and hydrophilic groups, respectively, are predicted to enhance GI.
not directly involved in the interaction with the receptor.
Thus, removal of the hydroxyl group at the C-2 position
(H, 9) also enhances the GI for the MDR line [e.g. GI(3)
) 5 vs GI(9) ) 67]. The lack of interaction of the
hydroxyl group with this domain of the receptor does
not compensate for the energy cost of desolvating the
HO-substituted ligand [∆Gsolv(3) ) -15.5 kcal/mol vs
∆Gsolv(9) ) -11.4 kcal/mol; see Table 3]. Moreover, the
hydrophobicity map (the third field contribution in the
MDR model, 10.0%) displays both a yellow and a small
white area at this C-2 position (Figures 5Aiv and 6Aiv).
Thus, the hydrophobic and bulky groups Bz, Pr, and
MeBut of the OBz (4), OPr (5), and OMeBut (6)
substituents, respectively, elicit some of the highest
values of GI in the MDR line (Table 3). It is important
to note that the Bz group has the lowest value of GI
among these substituents. This finding explains the
presence of the white area (regions where hydrophilic
groups enhance GI or hydrophobic groups diminishes
GI) at the upper-left part of the map, since the hydro-
gens of the Bz moiety of the OBz substituent reach this
white area (Figure 6Aiv). The steric map (the fifth field
contribution in the MDR model, 5.4%) reinforces this
findings. Thus, OPr (5) and OMeBut (6) substituents
at the C-2 position occupy the favorable green area
(Figure 5Aii) increasing GI in the MDR line, whereas
the OBz (4) reaches the unfavorable yellow (Figure 6Aii)
zone leading to lower value than expected (Table 3).
The OFu substituent at C-6 in compounds 1, 2, and
4 was replaced by a hydroxyl (11, 13 and 22) or an ester
(12, 14-21, 23-30) group. Comparison of their activi-
ties shows that the substituent at the C-6 position is
crucial for the modulation of the activity. The presence
of a hydroxyl (11, 13 and 22) or acetate (12, 14 and 23)
group at the 6-position has a detrimental effect on GI
with respect to the presence of a furoate group (1, 2,
and 4) [e.g. GI(1) ) 53 vs GI(11) ) 1, GI(12) ) 0; GI(2)
) 84 vs GI(13) ) 5, GI(14) ) 6; and GI(4) ) 75 vs GI-
(22) ) 11, GI(23) ) 4] (Table 1). Thus, the oxygen of
the furan ring of the OFu substituent seems to act as a
H-bond acceptor in the interaction with the receptor (see
magenta area at the bottom side of Figure 5Aiii).
Accordingly, the electrostatic map (the highest field
contribution in the MDR model, 58.9%) shows a red area
(Figure 5Ai) at this position of the furan ring, emphasiz-
ing the importance of this part of the molecule in the
interaction with the receptor. While the energy cost of
desolvating the hydroxyl or the acetate substituent is
similar to that of the corresponding furoate [∆Gsolv(1)
) -15.5 vs ∆Gsolv(11) ) -15.7, ∆Gsolv(12) ) -15.6;

582 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 3
Corte´s-Selva et al.
F igu r e 6. Electrostatic (i), steric (ii), hydrogen-bond-donor and -acceptor (iii), and hydrophobic (iv) maps for MDR (A) and WT
(B) L. tropica lines from CoMSIA models. Compound 29 is shown as reference structure. The color code is the same as in Figure
5.
∆Gsolv(2) ) -13.7 vs ∆Gsolv(13) ) -13.8, ∆Gsolv(14) )
-13.8; and ∆Gsolv(4) ) -14.0 vs ∆Gsolv(22) ) -14.2,
∆Gsolv(23) ) -14.1], their shortest side chain impedes
the interaction with this part of the receptor, leading
to lower values of GI. Additionally, the hydrophobicity
map (Figure 5Aiv) also displays a yellow area in the
vicinity of this furan ring. Thus, it seems reasonable to
propose that the electron-poor C-H hydrogens of the
furan ring also interact with the receptor.32 Conse-
quently, the steric map (Figure 5Aii) shows at the C-6
position a favorable green area near these furan hydro-
gens. Notably, the OPiv (18) substituent (derived from
compound 2 that contains the optimal OAc substituent
at the C-2 position) lacks the polar oxygen of the furan
ring but contains electron-poor C-H hydrogens, thus
possessing intermediate GI values [GI(2) ) 84 vs GI-
(18) ) 45]. Longer and hydrophobic side chains (15-
17) would occupy the yellow (disfavorable) area in the
steric map, whereas the longer and polar O4-NO2-Bz
side chain (21) requires higher solvation energy (see
Table 3), which is not compensated by the interaction
with the receptor, leading to low values of GI. The
intermediate values of GI of compound 20 (GI ) 35) led
us to suggest that the OCH3 moiety of the O4-MeO-
Bz substituent is able to H-bond the receptor as the
oxygen atom of the furan ring, but its longer side chain
occupies the yellow (disfavorable) area in the steric map
(Figure 6Aii).
On the other hand, the regiosubstitution seems to be
an important element for affinity, since the tetrasub-
stituted 4â-hydroxydihydro-â-agarofuran sesquiter-
penes 2 and 12 have the same molecular formula, the
only difference being the presence of the acetate or
furoate at the C-2 and C-6 positions. Compound 2 has
OAc at C-2 and OFu at C-6, contrarily to 12, because
12 is completely inactive and 2 shows strong activity
[e.g. GI(2) ) 83 vs GI(12) ) 0].
Moreover, compounds 2 (CH3) and 7 (CH2OAc) led us
to study the effect of the substituent at the C-15 position
in GI. The addition of the polar OAc moiety does not
affect GI [GI(2) ) 83 vs GI (7) ) 80]. However, the
energy cost of desolvating the CH2OAc-substituted
ligand [∆Gsolv(7) ) -17.1 kcal/mol] is higher than for
the CH3-substituted ligand [∆Gsolv (2) ) -13.7 kcal/mol].
Thus, this additional energy penalty of 7 relative to 2
due to solvation must be compensated by additional
interactions of 7 with the receptor. Accordingly, Figures
5 and 6 depict a magenta H-bond-donor area and a
favorable green area (steric map) on this part of the
receptor that enhances GI.
Finally, the OH substituent at the C-4 position of
these sesquiterpenes is located close to the orange
contours (Figure 5Aiii). This orange region predicts the
position where an H-bond donor zone/residue in the
receptor probably disfavors binding. On the other hand,
the OAc and the OFu/OBz substituents at the C-1 and
C-9 positions, respectively, are located close to the

SAR of Dihydro-â-agarofuran Sesquiterpenes
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 3 583
magenta area (Figure 5Aiii). Thus, these groups may
form hydrogen bonds with H-bond donor zones/residues
in the receptor. However, the absence of variability at
any of these positions in the studied series of compounds
prevents reaching any further conclusion.
Compounds 1-2, 4-6, and 9 produced limited effects
on the WT parasites, which could correspond to some
low binding to other cellular targets. The presence of
an acetate (2), protons (9), furoate (1), propianate (5),
benzoate (4), or methylbutyrate (6) at C-2 increases the
cytotoxic effects of these compounds in the rank order
of 2 < 1 < 9 < 5 < 4 < 6. These compounds produced in
the WT line a 2%, 14%, 15%, 16%, 17%, or 22%
inhibition of growth, respectively. These studies allowed
us to establish that the acetate group is the optimal
structural feature at C-2, in terms of both potency and
selectivity for the MDR line; thus, compound 2 repre-
sents the most active member of these sesquiterpenes
(1-6 and 9). The B panels in Figures 5 and 6 illustrate
the CoMSIA maps for the WT line. The substituent at
the C-2 position also seems important for GI in the WT
line as shown in the steric (Bii), H-bond-donor and
-acceptor (Biii), and hydrophobic (Biv) maps. The effect
on GI of the different substituents at this position of
the molecule followed the same trend as for the MDR
line (see above) with the exception of compound 2 (OAc).
The CH3 moiety of the OAc substituent in C-2 does not
contact the favorable green region in the steric map
(Figure 5Bii). Thus, the bulky substituents of compounds
1 (OFu), 4 (OBz), 5 (OPr), and 6 (OMeBut) and the
unsubstituted compound 9 (H) have larger values of GI
than compounds 2 (OAc) and 3 (OH). The steric interac-
tions of the bulky substituents (OFu, 1; OAc, 2; OBz, 4;
OPr, 5; and OMeBut, 6) with the receptor (Figures 5
and 6, panel Bii) or the facility of desolvating compound
9 are responsible for high values of GI.
F igu r e 7. Most salient structural elements of the ligands that
are key to high growth inhibition obtained with the 3D-QSAR/
CoMSIA methodology. Magenta spheres represent areas of the
receptor that interact with H-bond-acceptor moieties of the
ligand (i.e. the carbonyl group at the C-2 position and the
oxygen of the furan ring at the C-6 position). Yellow spheres
represent areas of the receptor that interact with hydrophobic
moieties of the ligand (i.e. the C-H groups of the furan ring
at the C-6 position).
Macrocyclic compound 10 showed no activity due to
its too bulky cycle that might not fit into the active site.
These results are in agreement with previous observa-
tions that the steric properties could modulate the
activity.5
Con clu sion s
3D-QSAR/CoMSIA methodology has been successfully
applied to explain the reversal activity of a series of
sesquiterpenes 1-30 against the MDR phenotype of
Leishmania. The derived computational model has
facilitated the identification of the structural elements
of the ligands, depicted in Figure 7, that are key to high
growth inhibition. The most salient features of the
electrostatic, steric, hydrogen-bond-donor and -acceptor,
and hydrophobic maps obtained with the 3D-QSAR/
CoMSIA methodology are the following. The carbonyl
group of the OFu (1), OAc (2), OBz (4), OPr (5), and
OMeBut (6) substituents at the C-2 position acts as a
H-bond acceptor in the H-bond with an area of the
receptor depicted as a magenta sphere (Figure 7). The
oxygen of the furan ring at the C-6 position seems to
form a hydrogen bond with the receptor (see magenta
sphere in Figure 7). The C-H moieties of the furan ring
at the C-6 position are also involved in the interaction
with the receptor (yellow sphere). Moreover, the CH2-
OAc group at the C-15 position in compound 7 is
engaged in the H-bond interaction with an area of the
receptor depicted as a magenta sphere (Figure 7). The
substituents at the C-1, C-4, and C-9 positions of
sesquiterpenes may also be engaged in H-bond interac-
tions with the receptor. However, the absence of vari-
ability at any of these positions in the studied series of
compounds impedes any further conclusion. Finally,
replacement of the furoate group by a benzoate group
at both C-6 and C-9 does not modify GI but slightly
increases the cytotoxic effect. These results support the
However, the most important position for GI in the
WT line is C-6. Substitution of the polar OFu group (1-
7, 9), involved in electrostatic interaction [the highest
field contribution in the WT model is 54.5% (see the red
contour at the bottom-center part of Figure 5Bi)], with
other groups in 11-30 [by OH (11 and 13), OAc (12 and
14), OMeBut (15 and 24), OLau (16 and 25), ONap (17
and 26), OPiv (18 and 27), OTFAc (19 and 28), O4-
MeO-Bz (20 and 29), or O4-NO2-Bz (21 and 30)] leads
to compounds with the desirable lack of activity in WT
(GI values of 0, Table 3). It seems that precisely the
absence of the negative charge generated by the oxygen
of the OFu substituent impedes compounds 11-30 from
interacting with the receptor at this position (panels B
in Figure 6).
This is corroborated by the orange region (area of the
receptor where H-bond donors disfavor GI, panels Biii
in Figures 5 and 6) at this position in the WT line that
contrast with the magenta region (area of the receptor
where H-bond donors enhance GI, panels Aiii in Figures
5 and 6) in the MDR line.
Another significant difference between the WT and
MDR maps resides in the H-bond-donor/acceptor map
at the C-15 position. While the map for the MDR line
also contains a magenta area at this position, the map
for the WT line contains an orange area (Figures 5 and
6, parts Aiii and Biii).

584 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 3
Corte´s-Selva et al.
8.00 (1H, d, J ) 0.7 Hz); MS (EI) m/z (%) 480 (M+, 3), 465
(25), 420 (10), 353 (35), 233 (13), 168 (12), 95 (100); HRMS
(EI) m/z calcd for C24H32O10 480.19955, found 480.20226.
importance of both hydrogen-bond donors and acceptors
in inhibitor binding to a Pgp-like transporter.
Sesquiterpenes 2, 7, and 8 showed a high reversal
activity on the DNM-resistant phenotype in Leishmania
with values of GI at 3 µM higher than 75% and very
low values of intrinsic toxicities, being, at present, the
most effective agarofuran sesquiterpenes tested. These
results together with other obtained in previous
works5,6,17 will be used to design and synthesize more
effective and specific new Pgp-like inhibitors. Moreover,
further studies are in progress to increase our knowl-
edge about the mode of action of these compounds.
1r,2r,6â-Tr ia cetoxy-9â-fu r oyloxy-4â-h yd r oxyd ih yd r o-
â-a ga r ofu r a n (14). Compound 13 (3.7 mg) was treated under
the conditions already described for the synthesis of 12,
affording 14 (3.7 mg): colorless lacquer; [R]25D ) +7.6° (c 0.34,
CHCl3); IR γmax (film) 3547, 2927, 2855, 1747, 1720, 1367, 1309,
1243, 1160, 762 cm-1 1H NMR (CDCl3) δ 1.49 (3H, s), 1.50
;
(6H, s), 1.54 (3H, s), 1.74 (3H, s), 2.05 (3H, s), 2.14 (3H, s),
1.98-2.18 (4H, m), 2.47 (1H, m), 2.94 (1H, s), 4.89 (1H, d, J )
6.7 Hz), 5.39 (1H, d, J ) 3.5 Hz), 5.50 (1H, m), 5.57 (1H, s),
6.73 (1H, t, J ) 0.7 Hz), 7.42 (1H, t, J ) 0.7 Hz), 8.01 (1H, t,
J ) 0.7 Hz); MS (EI) m/z (%) 522 (M+, 1), 507 (2), 480 (3), 462
(4), 420 (4), 402 (19), 233 (9), 192 (17), 95 (50), 57 (100); HRMS
(EI) m/z calcd for C26H34O11 522.21011, found 522.20712.
Exp er im en ta l Section
1r,2r-Dia cetoxy-9â-fu r oyloxy-6â-(S)-(+)-2-m eth ylbu ty-
r oyloxy-4â-h yd r oxyd ih yd r o-â-a ga r ofu r a n (15). A mixture
of (S)-(+)-2-methylbutyric anhydride (6 drops), triethylamine
(10 drops), compound 13 (4.0 mg), and 4-(dimethylamino)-
pyridine (2.0 mg) in dry dichloromethane (1 mL) was refluxed
for 16 h. The mixture was evaporated to dryness, and the
residue was purified by preparative TLC using n-hexane/ethyl
Gen er a l Exp er im en ta l P r oced u r es. Optical rotations
were measured on a Perkin-Elmer 241 automatic polarimeter,
and the [R]D are given in 10-1 deg cm2 g-1. IR (film) spectra
were recorded on a Bruker IFS 55 spectrophotometer. 1H NMR
spectra were recorded on a Bruker Avance 400 or a Bruker
Avance 300 spectrometer. EIMS and HREIMS were recorded
on
a Micromass Autospec spectrometer. Monitoring and
acetate (3/2) to give 15 (3.2 mg): colorless lacquer [R]20
)
D
purification of the reactions were performed using silica gel
(TLC, plastic sheets, silica gel 60-250 UV250, Panreac). All
reagents were purchased from Aldrich and used without
further purification.
+5.0° (c 0.24, CHCl3); IR γmax (film) 3547, 2962, 2928, 1745,
1731, 1366, 1243, 1147, 1070, 1027, 762 cm-1; 1H NMR (CDCl3)
δ 0.91 (3H, t, J ) 7.5 Hz), 1.19 (3H, d, J ) 7.0 Hz), 1.46 (1H,
m), 1.50 (3H, s), 1.51 (3H, s), 1.55 (3H, s), 1.56 (3H, s), 1.73
(3H, s), 1.74 (1H, m), 2.05 (3H, s), 1.94-2.20 (4H, m), 2.50 (2H,
m), 2.92 (1H, s), 4.90 (1H, d, J ) 6.8 Hz), 5.39 (1H, d, J ) 3.5
Hz), 5.50 (1H, m), 5.62 (1H, s), 6.73 (1H, d, J ) 1.6 Hz), 7.41
Ch em istr y. 1r-Acetoxy-2r,9â-d ifu r oyloxy-4â,6â-d ih y-
d r oxyd ih yd r o-â-a ga r ofu r a n (11). A mixture of 1 (10.0 mg),
methanol (3 mL), acetone (2 mL), and NaHCO3 (1 mL, 0.1 M)
was refluxed for 5 h. Then the mixture was concentrated under
reduced pressure, water was added to the residue, and the
aqueous residue was extracted three times with ethyl acetate.
The collected organic layers were dried over magnesium
sulfate and evaporated under reduced pressure to give a thick
oil, which was purified by preparative TLC using n-hexane/
(1H, d, J ) 1.6 Hz), 8.01 (1H, s); MS (EI) m/z (%) 549 (M+
-
15, 4), 504 (1), 480 (5), 462 (1), 437 (15), 402 (32), 290 (18),
233 (14), 192 (23), 95 (74), 57 (100); HRMS (EI) m/z calcd for
C
28H37O11 549.23359, found 549.23331.
1r,2r-Diacetoxy-9â-fu r oyloxy-6â-lau r oyloxy-4â-h ydr oxy-
ethyl acetate (1/1) to give 11 (8.2 mg): colorless lacquer; [R]25
d ih yd r o-â-a ga r ofu r a n (16). Lauroyl chloride (0.1 mL) was
added to a solution of 13 (4.0 mg), triethylamine (0.4 mL), and
4-(dimethylamino)pyridine (2.0 mg) in dry dichloromethane (1
mL) at 0 °C. The resulting mixture was stirred for 2 h at room
temperature. The reaction was quenched by the addition of
ethanol (0.5 mL) followed by stirring for 30 min. The mixture
was evaporated to dryness, the residue was purified by flash
column chromatography on silica gel (eluting 50% to 100%
ethyl ether in n-hexane), and finally the compound was
purified by preparative TLC using dichloromethane/ethyl ether
D
) +39.4° (c 0.34, CHCl3); IR γmax (film) 3433, 2925, 2853, 1746,
1
1720, 1312, 1244, 1162, 1140, 763 cm-1; H NMR (CDCl3) δ
1.52 (6H, s), 1.60 (3H, s), 1.71 (3H, s), 1.83 (3H, s), 2.16 (5H,
m), 3.31 (1H, s), 4.56 (1H, s), 4.87 (1H, d, J ) 6.9 Hz), 4.97
(1H, s), 5.49 (1H, d, J ) 3.7 Hz), 5.73 (1H, m), 6.70 (1H, s),
6.73 (1H, s), 7.42 (1H, d, J ) 1.5 Hz), 7.45 (1H, d, J ) 1.5 Hz),
7.97 (1H, d, J ) 0.7 Hz), 8.01 (1H, d, J ) 0.7 Hz); MS (EI) m/z
(%) 532 (M+, 1), 517 (4), 420 (3), 405 (11), 345 (1), 233 (4), 205
(4), 168 (6), 149 (16), 95 (100); HRMS (EI) m/z calcd for
(5/1) to give 16 (2.3 mg): colorless lacquer; [R]20 ) +8.3° (c
C
27H32O11 532.19446, found 532.19138.
D
0.18, CHCl3); IR γmax (film) 3554, 2924, 2853, 1747, 1366, 1310,
1r,6â-Diacetoxy-2r,9â-difu r oyloxy-4â-h ydr oxydih ydr o-
1
1243, 1158, 762 cm-1; H NMR (CDCl3) δ 0.88 (3H, t, J ) 5.5
â-a ga r ofu r a n (12). A mixture of acetic anhydride (4 drops),
compound 11 (4.0 mg), and 4-(dimethylamino)pyridine (2.0 mg)
in pyridine (2 drops) was stirred at room temperature for 16
h. The mixture was evaporated to dryness, and the residue
was purified by preparative TLC using n-hexane/ethyl acetate
Hz), 1.25 (14H, s), 1.50 (3H, s), 1.57 (3H, s), 1.63 (6H, s), 1.57-
1.65 (4H, m), 1.74 (3H, s), 2.04 (3H, s), 1.94-2.17 (4H, m), 2.37
(2H, dt, J ) 2.4, 7.3 Hz), 2.46 (1H, m), 2.92 (1H, s), 4.89 (1H,
d, J ) 6.6 Hz), 5.38 (1H, d, J ) 3.4 Hz), 5.50 (1H, m), 5.59
(1H, s), 6.72 (1H, d, J ) 1.1 Hz), 7.41 (1H, d, J ) 1.1 Hz), 8.01
(1H, s); MS (EI) m/z (%) 647 (M+ - 15, 9), 602 (2), 535 (11),
480 (14), 463 (3), 420 (9), 402 (67), 192 (34), 126(47), 95 (100);
HRMS (EI) m/z calcd for C35H51O11 647.34314, found 647.34390.
(3/2) to give 12 (3.5 mg): colorless lacquer; [R]25 ) +30.0° (c
D
0.10, CHCl3); IR γmax (film) 3446, 2925, 2853, 1731, 1368, 1309,
1
1260, 1160, 1024, 799 cm-1; H NMR (CDCl3) δ 1.50 (3H, s),
1.56 (3H, s), 1.59 (6H, s), 1.73 (3H, s), 2.14 (3H, s), 2.10-2.19
(4H, m), 2.47 (1H, m), 3.59 (1H, s), 4.91 (1H, d, J ) 6.7 Hz),
5.46 (1H, d, J ) 3.5 Hz), 5.61 (1H, s), 5.69 (1H, m), 6.69 (1H,
d, J ) 1.3 Hz), 6.73 (1H, d, J ) 1.3 Hz), 7.42 (1H, d, J ) 1.7
Hz), 7.44 (1H, d, J ) 1.7 Hz), 7.95 (1H, s), 8.03 (1H, s); MS
(EI) m/z (%) 574 (M+, 1), 559 (2), 514 (1), 447(5), 402 (14), 290
(8), 192 (10), 95 (100); HRMS (EI) m/z calcd for C29H34O12
574.20502, found 574.20781.
1r,2r-Dia cetoxy-9â-fu r oyloxy-6â-(1)-n a p h th oyloxy-4â-
h yd r oxyd ih yd r o-â-a ga r ofu r a n (17). A solution of 13 (3.0
mg), 1-naphthoyl chloride (0.1 mL), 4-(dimethylamino)pyridine
(1.0 mg), and triethylamine (0.4 mL) in dry dichloromethane
(1 mL) was refluxed for 2 h. The reaction was quenched by
the addition of ethanol (0.5 mL) followed by stirring for 30 min
at room temperature. The mixture was evaporated to dryness,
the residue was purified by flash column chromatography on
silica gel (eluting 50% to 100% dichloromethane in n-hexane),
and finally the compound was purified by preparative TLC
using n-hexane/ethyl acetate (3/2) to give 17 (3.0 mg): colorless
1r,2r-Diacetoxy-9â-fu r oyloxy-4â,6â-dih ydr oxydih ydr o-
â-a ga r ofu r a n (13). Compound 2 (120.0 mg) was treated under
the conditions already described for the synthesis of 11,
affording 13 (114.1 mg): colorless lacquer; [R]25 ) +12.8° (c
D
0.25, CHCl3); IR γmax (film) 3536, 2959, 1746, 1721, 1366, 1244,
lacquer; [R]20 ) +23.9° (c 0.28, CHCl3); IR γmax (film) 3546,
D
1
1140, 1026, 762 cm-1
;
1H NMR (CDCl3) δ 1.46 (3H, s), 1.51
2924, 2853, 1747, 1716, 1366, 1244, 1134, 787, 761 cm-1; H
(3H, s), 1.58 (3H, s), 1.70 (3H, s), 1.78 (3H, s), 2.06 (3H, s),
2.00-2.32 (5H, m), 3.27 (1H, s), 4.53 (1H, d, J ) 5.3 Hz), 4.85
(1H, d, J ) 6.7 Hz), 4.96 (1H, d, J ) 5.3 Hz), 5.41 (1H, d, J )
3.6 Hz), 5.54 (1H, m), 6.72 (1H, t, J ) 0.7 Hz), 7.41 (1H, s),
NMR (CDCl3) δ 1.53 (3H, s), 1.56 (6H, s), 1.60 (3H,s), 1.76 (3H,
s), 2.05 (1H, m), 2.07 (3H, s), 2.22 (2H, m), 2.40 (1H, s), 2.64
(1H, m), 3.19 (1H, s), 4.98 (1H, d, J ) 6.8 Hz), 5.46 (1H, d, J
) 3.4 Hz), 5.55 (1H, m), 5.93 (1H, s), 6.75 (1H, d, J ) 1.1 Hz),

SAR of Dihydro-â-agarofuran Sesquiterpenes
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 3 585
7.43 (1H, d, J ) 1.1 Hz), 7.56 (2H, m), 7.65 (1H,m), 7.91 (1H,
d, J ) 8.2 Hz), 8.03 (1H, s), 8.07 (1H, d, J ) 8.5 Hz), 8.71 (1H,
d, J ) 7.4 Hz), 9.20 (1H, d, J ) 8.6 Hz); MS (EI) m/z (%) 634
(M+, 1), 619 (1), 574 (1), 507 (2), 462 (1), 402 (9), 290 (2), 192
(2), 155 (100), 127 (8), 95 (9); HRMS (EI) m/z calcd for
(72), 120 (70), 95 (100); HRMS (EI) m/z calcd for C30H32N1O13
614.18737, found 614.18866.
1r-Acetoxy-2r-benzoyloxy-9â-furoyloxy-4â,6â-dihydroxy-
d ih yd r o-â-a ga r ofu r a n (22). Compound 4 (104.0 mg) was
treated under the conditions already described for the syn-
thesis of 11, affording 22 (88.0 mg): colorless lacquer; [R]25
C
35H38O11 634.24141, found 634.24117.
D
) +60° (c 0.36, CHCl3); IR γmax (film) 3433, 2928, 1720, 1366,
1277, 1233, 1105, 1025, 761, 711 cm-1; 1H NMR (CDCl3) δ 1.53
(3H, s), 1.58 (3H, s), 1.60 (3H, s), 1.70 (3H, s), 1.86 (3H, s),
2.08-2.35 (5H, m), 3.33 (1H, s), 4.58 (1H, d, J ) 5.4 Hz), 4.88
(1H, d, J ) 6.7 Hz), 4.99 (1H, d, J ) 5.4 Hz), 5.53 (1H, d, J )
4.0 Hz), 5.81 (1H, m), 6.73 (1H, d, J ) 1.6 Hz,), 7.43 (2H, m),
7.50 (1H, d, J ) 1.6 Hz), 7.59 (1H, m), 7.98 (1H, d, J ) 7.4
Hz), 8.02 (1H, s); MS (EI) m/z (%) 527 (M+ - 15, 6), 420 (5),
415 (11), 360 (1), 349 (3), 308 (3), 289 (3), 105 (100), 94 (63),
77 (21), 57 (26); HRMS (EI) m/z calcd for C28H31O10 527.19172,
found 527.18518.
1r,2r-Diacetoxy-9â-furoyloxy-6â-pivaloyloxy-4â-hydroxy-
d ih yd r o-â-a ga r ofu r a n (18). A solution of 13 (4.0 mg), piv-
aloyl chloride (0.1 mL), 4-(dimethylamino)pyridine (1.5 mg),
and pyridine (0.4 mL) in dry chloroform (1 mL) was refluxed
for 16 h. The reaction was quenched by the addition of ethanol
(0.5 mL) followed by stirring for 30 min at room temperature.
The mixture was evaporated to dryness, the residue was
purified by flash column chromatography on silica gel (eluting
50% to 100% dichloromethane in n-hexane), and finally the
compound was purified by preparative TLC using n-hexane/
ethyl acetate (3/2) to give 18 (3.5 mg): colorless lacquer; [R]20
D
) +10.6° (c 0.34, CHCl3); IR γmax (film) 3553, 2958, 2925, 2853,
1748, 1721, 1366, 1244, 1148, 1070, 1027, 762 cm-1; 1H NMR
(CDCl3) δ 1.24 (9H, s), 1.49 (6H, s), 1.50 (3H, s), 1.55 (3H, s),
1.73 (3H, s), 2.04 (3H, s), 1.94-2.19 (4H, m), 2.46 (1H, m), 2.90
(1H, s), 4.90 (1H, d, J ) 6.8 Hz), 5.40 (1H, d, J ) 3.5 Hz), 5.50
(1H, m), 5.62 (1H, s), 6.72 (1H, d, J ) 1.1 Hz), 7.41 (1H, t, J
) 1.1 Hz), 8.01 (1H, s); MS (EI) m/z (%) 549 (M+ - 15, 6), 437
(15), 402 (52), 290 (20), 233 (16), 192 (40), 173 (12), 149 (23),
57 (100); HRMS (EI) m/z calcd for C28H37O11 549.23359, found
549.23498.
1r,6â-Diacetoxy-2r-benzoyloxy-9â-furoyloxy-4â-hydroxy-
d ih yd r o-â-a ga r ofu r a n (23). Compound 22 (3.0 mg) was
treated under the conditions already described for the syn-
thesis of 12, affording 23 (2.1 mg): colorless lacquer; [R]20
)
D
+30.6° (c 0.16, CHCl3); IR γmax (film) 3547, 2925, 2853, 1723,
1368, 1275, 1236, 1099, 1070, 760, 712 cm-1; 1H NMR (CDCl3)
δ 1.51 (3H, s), 1.56 (3H, s), 1.60 (3H, s), 1.61 (3H, s), 1.73 (3H,
s), 2.14 (3H, s), 2.10-2.30 (4H, m), 2.50 (1H, m), 2.98 (1H, s),
4.91 (1H, d, J ) 6.8 Hz), 5.51 (1H, d, J ) 3.6 Hz), 5.64 (1H, s),
5.78 (1H, m), 6.74 (1H, d, J ) 1.1 Hz), 7.42 (1H, d, J ) 1.1
Hz), 6.74 (2H, m), 7.57 (1H, m), 7.97 (2H, dd, J ) 1.2, 7.0 Hz),
8.03 (1H, s); MS (EI) m/z (%) 569 (M+ - 15, 1), 542 (1), 524
(3), 457 (4), 420 (3), 402 (22), 192 (26), 149 (100), 105 (86), 94
(51), 57 (47); HRMS (EI) m/z calcd for C30H33O11 569.20229,
found 569.19952.
1r,2r-Dia cetoxy-9â-fu r oyloxy-6â-tr iflu or oa cetoxy-4â-
h yd r oxyd ih yd r o-â-a ga r ofu r a n (19). Trifluoroacetic anhy-
dride (3 drops) was added to a solution of 13 (3.0 mg) and
pyridine (4 drops) in dry dichloromethane (1 mL) at 0 °C. After
5 min of stirring the mixture was evaporated to dryness, and
the residue was purified by preparative TLC using n-hexane/
1r-Acet oxy-2r-b en zoyloxy-9â-fu r oyloxy-6â-(S)-(+)-2-
meth ylbu tyr oyloxy-4â-h ydr oxyd ih ydr o-â-a ga r ofu r a n (24).
Compound 22 (4.0 mg) was treated under the conditions
already described for the synthesis of 15, affording 24 (3.2
ethyl acetate (3/2) to give 19 (2.4 mg): colorless lacquer; [R]20
D
) +13.6° (c 0.22, CHCl3); IR γmax (film) 3449, 2925, 2854, 1775,
1750, 1369, 1221, 1174, 1142, 1038, 762 cm-1; 1H NMR (CDCl3)
δ 1.48 (3H, s), 1.51 (3H, s), 1.57 (3H, s), 1.73 (3H, s), 1.94 (3H,
s), 2.07 (3H, s), 2.22-2.54 (4H, m), 3.12 (1H, dd, J ) 3.4, 14.7
Hz), 4.94 (1H, d, J ) 6.7 Hz), 5.51 (1H, d, J ) 3.6 Hz), 5.63
(1H, m), 5.68 (1H, s), 6.72 (1H, d, J ) 1.7 Hz), 7.42 (1H, d, J
) 1.7 Hz), 8.02 (1H, s); MS (EI) m/z (%) 558 (M+ - 18, 1), 543
(1), 503 (1), 449 (8), 405 (0.9), 329 (1), 290 (1), 247 (2), 230 (2),
215 (2), 149 (9), 95 (100); HRMS (EI) m/z calcd for C26H29O10F3
558.17128, found 558.17567.
mg): colorless lacquer; [R]20 ) +6.3° (c 0.3, CHCl3); IR γmax
D
(film) 3553, 2924, 2854, 1727, 1462, 1273, 1144, 1098, 1071,
1
1026, 800, 712 cm-1; H NMR (CDCl3) δ 0.91 (3H, t, J ) 7.4
Hz), 1.20 (3H, d, J ) 7.0 Hz), 1.51 (3H, s), 1.58 (6H, s), 1.62
(3H, s), 1.72 (3H, s), 1.74 (1H, m), 1.98-2.27 (4H, m), 2.44-
2.50 (2H, m), 2.97 (1H, s), 4.92 (1H, d, J ) 7.0 Hz), 5.52 (1H,
d, J ) 3.6 Hz), 5.69 (1H, s), 5.78 (1H, m), 6.73 (1H, s), 7.47
(3H, m), 7.58 (1H, m), 7.97 (2H, dd, J ) 1.3, 7.0 Hz), 8.03 (1H,
s); MS (EI) m/z (%) 611 (M+ - 15, 4), 542 (2), 524 (1), 499 (7),
462 (1), 402 (43), 290 (8), 192 (25), 105 (100), 95 (66), 57 (83);
HRMS (EI) m/z calcd for C33H39O11 611.24924, found 611.24674.
1r,2r-Dia cetoxy-9â-fu r oyloxy-6â-(4)-m eth oxyben zoyl-
oxy-4â-h yd r oxyd ih yd r o-â-a ga r ofu r a n (20). Compound 13
(3.0 mg) was treated with 4-methoxybenzoyl chloride (0.1 mL)
under the conditions already described for the synthesis of 18,
affording 20 (2.7 mg): colorless lacquer; [R]20D ) +14.5° (c 0.22,
CHCl3); IR γmax (film) 3547, 2924, 2852, 1746, 1717, 1606, 1366,
1312, 1254, 1163, 1027, 761 cm-1; 1H NMR (CDCl3) δ 1.51 (3H,
s), 1.53 (3H, s), 1.54 (3H, s), 1.56 (3H, s), 1.75 (3H, s), 2.04
(3H, s), 1.99-2.24 (3H, m), 2.36 (1H, m), 2.56 (1H, m), 3.13
(1H, s), 3.87 (3H, s), 4.94 (1H, d, J ) 6.7 Hz), 5.43 (1H, d, J )
3.4 Hz), 5.53 (1H, m), 5.70 (1H, s), 6.74 (1H, t, J ) 1.1 Hz),
6.96 (2H, dd, J ) 2.7, 8.9 Hz), 7.42 (1H, t, J ) 1.7 Hz), 8.03
(1H, t, J ) 0.7 Hz), 8.15 (2H, dd, J ) 2.7, 8.9 Hz); MS (EI) m/z
(%) 614 (M+, 1), 599 (1), 533 (3), 502 (10), 402 (39), 290 (14),
192 (47), 149 (82), 120 (97), 95 (100); HRMS (EI) m/z calcd for
1r-Acetoxy-2r-ben zoyloxy-9â-fu r oyloxy-6â-la u r oyloxy-
4â-h yd r oxyd ih yd r o-â-a ga r ofu r a n (25). Compound 22 (6.0
mg) was treated under the conditions already described for
the synthesis of 16, affording 25 (4.3 mg): colorless lacquer;
[R]20D ) +27.5° (c 0.43, CHCl3); IR γmax (film) 3554, 2925, 2854,
1
1724, 1311, 1273, 1228, 1160, 1106, 1070, 762, 711 cm-1; H
NMR (CDCl3) δ 0.87 (3H, t, J ) 6.2 Hz), 1.26 (14H, s), 1.51
(3H, s), 1.51-1.66 (4H, m), 1.56 (3H, s), 1.59 (3H, s), 1.67 (3H,
s), 1.73 (3H, s), 2.15-2.50 (7H, m), 2.98 (1H, s), 4.92 (1H, d, J
) 6.7 Hz), 5.51 (1H, d, J ) 3.6 Hz), 5.66 (1H, s), 5.78 (1H, m),
6.74 (1H, t, J ) 1.1 Hz), 7.46 (3H, m), 7.58 (1H, m), 7.97 (2H,
dd, J ) 1.2, 7.1 Hz), 8.03 (1H, t, J ) 0.7 Hz); MS (EI) m/z (%)
709 (M+ - 15, 4), 597 (7), 560 (1), 542 (1), 524 (3), 420 (6), 402
(56), 192 (4), 105 (100), 95 (59), 57 (32); HRMS (EI) m/z calcd
for C40H53O11 709.35879, found 709.35315.
C
32H38O12 614.23633, found 614.23883.
1r,2r-Dia cetoxy-9â-fu r oyloxy-6â-(4)-n itr oben zoyloxy-
4â-h yd r oxyd ih yd r o-â-a ga r ofu r a n (21). Compound 13 (3.0
mg) was treated with 4-nitrobenzoyl chloride (0.1 mL) under
the conditions already described for the synthesis of 18,
affording 21 (2.8 mg): colorless lacquer; [R]20D ) +22.3° (c 0.13,
CHCl3); IR γmax (film) 3546, 2924, 2853, 1746, 1724, 1530, 1366,
1282, 1254, 1119, 1027, 760, 721 cm-1; 1H NMR (CDCl3) δ 1.52
(3H, s), 1.55 (3H, s), 1.56 (6H, s), 1.75 (3H, s), 2.04 (3H, s),
2.00-2.27 (3H, m), 2.39 (1H, m), 2.57 (1H, m), 3.09 (1H, s),
4.95 (1H, d, J ) 6.7 Hz), 5.43 (1H, d, J ) 3.5 Hz), 5.53 (1H,
m), 5.74 (1H, s), 6.74 (1H, d, J ) 1.6 Hz), 7.42 (1H, t, J ) 1.6
Hz), 8.03 (1H, s), 8.32 (2H, dd, J ) 1.9, 6.9 Hz), 8.41 (1H, dd,
J ) 1.9, 6.9 Hz); MS (EI) m/z (%) 614 (M+ - 15, 1), 590 (7),
569 (12), 502 (12), 402 (40), 290 (15), 233 (16), 192 (49), 150
1r-Acet oxy-2r-b en zoyloxy-9â-fu r oyloxy-6â-(1)-n a p h -
t h oyloxy-4â-h yd r oxyd ih yd r o-â-a ga r ofu r a n (26). Com-
pound 22 (3.5 mg) was treated under the conditions already
described for the synthesis of 17, affording 26 (3.1 mg):
colorless lacquer; [R]20D ) +31.8° (c 0.28, CHCl3); IR γmax (film)
3434, 2924, 2853, 1724, 1366, 1277, 1244, 1134, 759, 712 cm-1
;
1H NMR (CDCl3) δ 1.55 (3H, s), 1.59 (3H, s), 1.70 (3H, s), 1.71
(3H, s), 1.75 (3H, s), 2.04-2.33 (4H, m), 2.66 (1H, m), 3.24 (1H,
s), 5.01 (1H, d, J ) 6.6 Hz), 5.58 (1H, d, J ) 3.6 Hz), 5.84 (1H,
m), 5.99 (1H, s), 6.76 (1H, d, J ) 1.6 Hz), 7.46 (3H, m), 7.56
(2H, m), 7.65 (1H, m), 7.90 (2H, m), 7.97 (2H, m), 8.05 (3H,
m), 8.72 (1H, d, J ) 7.3 Hz), 9.20 (1H, d, J ) 8.7 Hz); MS (EI)

586 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 3
Corte´s-Selva et al.
m/z (%) 681 (M+ - 15, 1), 569 (1), 533 (1), 420 (1), 402 (7), 155
(100), 127 (14), 105 (24); HRMS (EI) m/z calcd for C39H37O11
681.23359, found 681.23106.
2. DNM Ch em osen sitiza tion Exp er im en ts. The viability
of parasites in the presence of the different sesquiterpenes was
analyzed by an MTT-based assay as previously described for
Leishmania.5 The screening was performed in flat-bottomed
96-well plastic plates maintained at 28 °C. Promastigote forms
from a logarithmic phase culture were suspended in fresh
medium to yield 6 × 106 cells/mL. Each well was filled with
50 µL of the parasite suspension (3 × 105 cells). Stock solutions
of sesquiterpenes dissolved in DMSO were diluted directly in
the culture medium at the suitable concentrations, and 50 µL
was added to each well. The final DMSO content did not exceed
0.3%, which had no effect on parasite growth. To assess the
chemosensitizing activity of sesquiterpenes, promastigotes of
the L. tropica MDR line were exposed to both DNM (150 µM)
and sesquiterpenes. To determine the intrinsic toxicity of the
sesquiterpenes, the WT and MDR L. tropica lines were exposed
to sesquiterpenes in the absence of DNM. After 72 h of
incubation at 28 °C, the viability of promastigotes was
determined by the colorimetric MTT assay. A 10-µL portion
of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
(MTT) (5 mg/mL in PBS) was added to each well, and plates
were incubated for an additional period of 4 h. Water-insoluble
formazan crystals were dissolved by adding 100 µL of SDS
20%, and absorbance was read at 540 nm using a microplate
reader (Beckman Biomek 2000). Cell survival was determined
by dividing the absorbance at a given sesquiterpene concentra-
tion by the absorbance of control cells. The results are
expressed as percent growth inhibition (GI). IC50 values, the
concentration of DNM that decreases the rate of parasite
growth by 50%, were determined for the WT and MDR lines.
Resistance indexes were calculated as the IC50 ratio between
the MDR and WT lines in the presence of different concentra-
tions of sesquiterpenes.
1r-Acetoxy-2r-ben zoyloxy-9â-fu r oyloxy-6â-p iva yloxy-
4â-h yd r oxyd ih yd r o-â-a ga r ofu r a n (27). Compound 22 (3.0
mg) was treated under the conditions already described for
the synthesis of 18, affording 27 (2.9 mg): colorless lacquer;
[R]20D ) +40.8° (c 0.25, CHCl3); IR γmax (film) 3553, 2958, 2925,
2853, 1722, 1366, 1313, 1275, 1231, 1118, 1106, 1068, 1026,
1
760, 712 cm-1; H NMR (CDCl3) δ 1.25 (9H, s), 1.51 (3H, s),
1.58 (6H, s), 1.62 (3H, s), 1.72 (3H, s), 2.10-2.27 (4H, m), 2.50
(1H, m), 2.94 (1H, s), 4.92 (1H, d, J ) 6.7 Hz), 5.52 (1H, d, J
) 3.6 Hz), 5.69 (1H, s), 5.79 (1H, m), 6.74 (1H, d, J ) 1.6 Hz),
7.44 (3H, m), 7.58 (1H, m), 7.97 (2H, dd, J ) 1.3, 7.4 Hz), 8.03
(1H, s); MS (EI) m/z (%) 611 (M+ - 15, 1), 499 (3), 402 (26),
290 (6), 192 (8), 105 (100), 95 (54), 57 (63); HRMS (EI) m/z
calcd for C33H39O11 611.24924, found 611.24870.
1r-Acet oxy-2r-b en zoyloxy-9â-fu r oyloxy-6â-t r iflu or o-
a cetoxy-4â-h yd r oxyd ih yd r o-â-a ga r ofu r a n (28). Compound
22 (5.0 mg) was treated under the conditions already described
for the synthesis of 19, affording 28 (4.2 mg): colorless lacquer;
[R]25D ) +46.6° (c 0.36, CHCl3); IR γmax (film) 3448, 2926, 2851,
1775, 1726, 1369, 1271, 1223, 1174, 1142, 1095, 1038, 760, 711
1
cm-1; H NMR (CDCl3) δ 1.51 (3H, s), 1.54 (3H, s), 1.67 (3H,
s), 1.73 (3H, s), 2.07 (3H, s), 2.25-2.61 (4H, m), 3.29 (1H, dd,
J ) 3.3, 14.7 Hz), 4.96 (1H, d, J ) 6.6 Hz), 5.63 (1H, d, J )
3.8 Hz), 5.73 (1H, s), 5.91 (1H, m), 6.74 (1H, d, J ) 1.6 Hz),
7.45 (3H, m), 7.60 (1H, m), 7.94 (2H, dd, J ) 1.3, 7.1 Hz), 8.04
(1H, s); MS (EI) m/z (%) 638 (M+, 1), 623 (1), 499 (1), 457 (1),
363 (1), 345 (1), 307 (1), 289 (1), 231 (1), 173 (2), 137 (24), 105
(100), 95 (71); HRMS (EI) m/z calcd for C30H30O11F3 623.17403,
found 623.16983.
1r-Acet oxy-2r-b en zoyloxy-9â-fu r oyloxy-6â-(4)-m et h -
oxyb en zoyloxy-4â-h yd r oxyd ih yd r o-â-a ga r ofu r a n (29).
Compound 22 (3.0 mg) was treated with (4)-methoxybenzoyl
chloride (0.1 mL) under the conditions already described for
the synthesis of 18, affording 29 (2.5 mg). Colorless lacquer;
[R]20D ) +18.3° (c 0.18, CHCl3); IR γmax (film) 3546, 2924, 2853,
3. R ever sion of Ca lcein Accu m u la t ion in a MDR L.
tr opica Lin e Over exp r essin g a P gp -Lik e Tr a n sp or ter .
The accumulation of CAL fluorescent dye in the WT and MDR
Leishmania lines was estimated by flow cytometry using a
Becton Dickinson FacScan, as described.5 Briefly, parasites
were incubated with 2 µM CAL-acetoxymethyl ester (CAL-AM)
(Molecular Probes Europe BV, The Netherlands), for 1 h at
28 °C in HPMI-glucose buffer (10 mM HEPES, 120 mM NaCl,
5 mM Na2HPO4, 0.4 mM MgCl2, 0.04 mM CaCl2, 10 mM
NaHCO3, 10 mM glucose, 5 mM KCl, pH 7.4), in the presence
or in absence of different concentrations of sesquiterpenes.
Parasites were then extensively washed, resuspended in cold
phosphate-buffered saline (PBS; 1.2 mM KH2PO4, 8.1 mM Na2-
HPO4, 130 mM NaCl, 2.6 mM KCl adjusted to pH 7.4), and
immediately analyzed. Cells were gated on the basis of size
and complexity to eliminate dead cells and debris from the
analysis. Quantification of intracellular fluorescence was
carried out by scanning the emission between 515 and 545 nm
(FL-1) using the Cell Quest Software application.
1
1721, 1606, 1366, 1257, 1168, 1106, 1025, 760, 712 cm-1; H
NMR (CDCl3) δ 1.53 (3H, s), 1.57 (3H, s), 1.63 (3H, s), 1.65
(3H, s), 1.74 (3H, s), 2.20-2.38 (4H, m), 2.58 (1H, m), 3.18 (1H,
s), 3.86 (3H, s), 4.96 (1H, d, J ) 6.3 Hz), 5.55 (1H, d, J ) 3.6
Hz), 5.76 (1H, s), 5.81 (1H, m), 6.75 (1H, d, J ) 1.4 Hz), 6.96
(2H, dd, J ) 8.8, 1.9 Hz), 7.43 (3H, m), 7.55 (1H, m), 7.96 (2H,
dd, J ) 1.3, 7.1 Hz), 8.04 (1H, s), 8.17 (2H, dd, J ) 8.8, 1.9
Hz); MS (EI) m/z (%) 661 (M+ - 15,1), 549 (2), 402 (11), 135
(100), 105 (18); HRMS (EI) m/z calcd for C36H37O12 661.22850,
found 661.22408.
1r-Acetoxy-2r-ben zoyloxy-9â-fu r oyloxy-6â-(4)-n itr oben -
zoyloxy-4â-h yd r oxyd ih yd r o-â-a ga r ofu r a n (30). Compound
22 (6.0 mg) was treated with 4-nitrobenzoyl chloride (0.1 mL)
under the conditions already described for the synthesis of 18,
affording 30 (4.7 mg): colorless lacquer; [R]20D ) +26.5° (c 0.47,
CHCl3); IR γmax (film) 3546, 2925, 2853, 1721, 1530, 1348, 1277,
1106, 1025, 760, 715 cm-1 1H NMR (CDCl3) δ 1.56 (6H, s),
;
Ack n ow led gm en t. This work has been supported
by the Spanish Grants PPQ2000-1655-C02-02, BQU2000-
0870-CO2-01 and CICYT (SAF2003-02730). Some of the
simulations were run at the Centre de Computacio´ i
Comunicacions de Catalunya. F.C.S. was a recipient of
a fellowship (B.E.F.I. Exp. 00/9068) from the Fondo de
Investigacio´n Sanitaria, Instituto de Salud Carlos III,
Ministerio de Sanidad y Consumo (Spain). We thank
Pilar Navarro for her help in parasite culture. We would
also like to acknowledge Pharmacia & Upjohn (Barce-
lona, Spain) for providing the DNM used in this study.
1.63 (3H, s), 1.67 (3H, s), 1.75 (3H, s), 2.18-2.60 (5H, m), 4.98
(1H, d, J ) 6.7 Hz), 5.56 (1H, d, J ) 3.6 Hz), 5.82 (2H, m),
6.76 (1H, d, J ) 1.7 Hz), 7.44 (3H, m), 7.57 (1H, m), 7.96 (2H,
d, J ) 7.3 Hz), 8.05 (1H, s), 8.31 (2H, m), 8.42 (2H, d, J ) 8.7
Hz); MS (EI) m/z (%) 676 (M+ - 15, 1), 564 (6), 402 (49), 290
(11), 192 (42), 150 (29), 120 (20), 105 (100), 95 (68), 77 (12);
HRMS (EI) m/z calcd for
C35H34N1O13 676.20302, found
676.20139.
Biologica l Assa ys. 1. P a r a site Cu ltu r e. The wild-type
(WT) L. tropica LRC-strain was a clone obtained by agar
plating.11 A L. tropica line highly resistant to DNM (MDR line)
was maintained in the presence of 150 µM DNM and used as
previously described.6 This resistant line had an MDR phe-
notype similar to tumor cells, with a cross-resistance profile
to several drugs and an overexpressed drug-efflux Pgp-like
transporter.6 Promastigote forms were grown at 28 °C in RPMI
1640-modified medium (Gibco) and supplemented with 20%
heat-inactivated fetal bovine serum (Gibco).
Su p p or tin g In for m a tion Ava ila ble: IR, MS, HRMS, and
1H NMR spectra for the compounds 11-30 described. This
material is available free of charge via the Internet at http://
pubs.acs.org.

SAR of Dihydro-â-agarofuran Sesquiterpenes
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 3 587
(18) Green, S. M.; Marshall, G. R. 3D-QSAR: A current perspective.
Trends Pharmacol. Sci. 1995, 16, 285-291.
(19) Cramer, R. D.; Patterson, D. E.; Bunce, J . D. Comparative
Molecular Field Analysis (CoMFA). 1. Effect of shape on binding
of steroids to carrier proteins. J . Am. Chem. Soc. 1988, 110,
5959-5967.
(20) Clark, M.; Cramer, R. D. I.; J ones, D. M.; Patterson, D. E.;
Simeroth, P. E. Comparative Molecular Field Analysis (CoMFA).
2. Toward its use with 3D-structural databases. Tetrahedron
Comput. Methodol. 1990, 3, 47-59.
Refer en ces
(1) Nicolaou, K. C.; Pfefferkorn, J . A.; Roecker, A. J .; Cao, G. Q.;
Barluenga, S.; Mitchell, H. J . Natural product-like combinatorial
libraries based on privileged structures. 1. General principles
and solid-phased synthesis of benzopyrans. J . Am. Chem. Soc.
2000, 122, 9939-9953
(2) Spivey, A. C.; Weston, M.; Woodhead, S. Celastraceae sesqui-
terpenoids: Biological activity and synthesis. Chem. Soc. Rev.
2002, 31, 43-59.
(3) Duan, H.; Takaishi, Y.; Momota, H.; Ohmoto, Y.; Taki, T.; J ia,
Y.; Li, D. Immunosuppressive sesquiterpene alkaloids from
Tripterygium wilfordii. J . Nat. Prod. 2001, 64, 582-587
(4) Duan, H.; Takaishi, Y.; Imakura, Y.; J ia, Y.; Duan, L.; Cosentino,
M.; Lee, K. Sesquiterpene alkaloids from Tripterygium hy-
poglaucum and Tripterygium wilfordii: A new class of potent
anti-HIV agents. J . Nat. Prod. 2000, 63, 357-361.
(5) Kennedy, M. L.; Corte´s-Selva, F.; Pe´rez-Victoria, J . M.; J ime´nez,
I. A.; Gonza´lez, A. G.; Mun˜oz, O. M.; Gamarro, F.; Castanys, S.;
Ravelo, A. G. Chemosensitization of a multidrug- resistant
Leishmania tropica line by new sesquiterpenes from Maytenus
magellanica and Maytenus chubutensis. J . Med. Chem. 2001,
44, 4668-4676.
(6) Pe´rez-Victoria, J . M.; Pe´rez-Victoria, F. J .; Parodi-Talice, A.;
J imenez, I. A.; Ravelo, A. G.; Castanys, S.; Gamarro, F. Alkyl-
lysophospholipid resistance in multidrug-resistant Leishmania
tropica and chemosensitization by a novel P-glycoprotein-like
transporter modulator. Antimicrob. Agents Chemother. 2001, 45,
2468-2474.
(7) Gonza´lez, A. G.; Tincusi, B. M.; Bazzocchi, I. L.; Tokuda, H.;
Nishino, H.; Konoshima, T.; J ime´nez, I. A.; Ravelo, A. G.
Antitumor promoting effects of sesquiterpenes from Maytenus
cuzcoina (Celastraceae). Bioorg. Med. Chem. 2000, 8, 1773-
1778.
(8) Evans, B. E.; Rittle, K. E.; Bock, M. G.; DiPardo, R. M.;
Freidinger, R. M.; Whitter, W. L.; Lundell, G. F.; Veber, D. F.;
Anderson, P. S.; Chang, R. S. L.; Lotti, V. J .; Cerino, D. J .; Chen,
T. B.; Kling, P. J .; Kunkel, K. A.; Springer, J . P.; Hirshfield, J .
Methods for drug discovery: Development of potent, selective,
orally effective cholecystokinin antagonists. J . Med. Chem. 1988,
31, 2235-2246.
(9) 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.
(10) Ouellette, M.; Legare, D.; Haimeur, A.; Grondin, K.; Roy, G.;
Brochu, C.; Papadopoulou, B. ABC transporters in Leishmania
and their role in drug resistance. Drug Res. Updates 1998, I,
43-48.
(11) Chiquero, M. J .; Pe´rez-Victoria, J . M.; O’Valle, F.; Gonza´lez-Ros,
J . M.; del Moral, R.; Ferragut, J . A.; Castanys, S.; Gamarro, F.
Altered drug membrane permeability in a multidrug-resistant
Leishmania tropica line. Biochem. Pharmacol. 1998, 55, 131-
139.
(21) Klebe, G.; Abraham, U.; Mietzner, T. Molecular Similarity
Indices in a Comparative Analysis (CoMSIA) of drug molecules
to correlate and predict their biological activity. J . Med. Chem.
1994, 37, 4130-4146.
(22) Singh, U. C.; Kollman, P. A. An approach to computing electro-
static charges for molecules. J . Comput. Chem. 1984, 5, 129-
145.
(23) Cornell, W. D.; Cieplak, P.; Bayly, C.; Gould, I. R.; Merz, K. M.;
Ferguson, D. M.; Spellmeyer, D. C.; Fox, T.; Caldwell, J . W.;
Kollman, P. A. A second generation force field for the simulation
of proteins, nucleic acids, and organic molecules. J . Am. Chem.
Soc. 1995, 117, 5179-5197.
(24) Hawkins, G. D.; Giesen, D. J .; Lynch, G. C.; Chambers, C. C.;
Rossi, I.; Storer, J . W.; Li, J .; Zhu, T.; Winget, P.; Rinaldi, D.;
Liotard, D. A.; Cramer, C. J .; Truhlar, D. G. AMSOL 6.7.2;
University of Minnesota, Minneapolis, MN 55455-0431.
(25) Bo¨hm, M.; Stu¨rzebecher, J .; Klebe, G. Three-dimensional quan-
titative structure-activity relationship analyses using compara-
tive molecular field analysis and comparative molecular simi-
larity indices analysis to elucidate selectivity differences of
inhibitors binding to trypsin, thrombin, and factor xa. J . Med.
Chem. 1999, 42, 458-477.
(26) Dunn, W. J . I.; Wold, S.; Edlund, U.; Hellberg, S.; Gasteiger, J .
Multivariate structure-activity relationships between data from
a
battery of biological tests and an ensemble of structure
descriptors: The PLS method. Quant. Struct.-Act. Relat. 1984,
3, 131-137.
(27) Wold, S.; Ruhe, A.; Wold, H.; Dunn, W. J . I. The covariance
problem in linear regression. The partial least squares (PLS)
approach to generalizad inverses. SIAM J . Sci. Stat. Comput.
1984, 5, 735-743.
(28) Cramer, R. D.; Bunce, J . D.; Patterson, D. E. Crossvalidation,
bootstrapping and partial least squares compared with multiple
regression in conventional QSAR studies. Quant. Struct.-Act.
Relat. 1988, 7, 18-25.
(29) Wold, S.; Albano, C.; Dunn, W. J . I.; Edlund, U.; Esbenson, W.;
Geladi, P.; Hellberg, S.; J ohansson, E.; Lindberg, W.; Sjo¨stro¨m,
M. Multivariate data analysis in chemistry. In Chemometrics:
Mathematics and Statistics in Chemistry; Kowalsky, B. R., Ed.;
Reidel: Dordrecht, The Netherlands, 1984; pp 17-95.
(30) SYBYL 6.6; Tripos Inc., 1699 South Hanley Rd., St. Louis, MO,
63144.
(31) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery,
J . A.; Keith, T. A.; Petersson, G. A.; Raghavachari, K.; Al-Laham,
A.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi,
J .; Barone, V.; Cossi, M.; Cammi, R.; Menucci, B.; Pomelli, C.;
Adamo, C.; Clifford, S.; Ochterski, J .; Petersson, G. A.; Ayala,
P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi,
I.; Gomperts, R.; Martin, R. L.; Fox, D. J .; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Gonza´lez, C.; Challacombe,
M.; Gill, P. M. W.; J ohnson, B. G.; Chen, W.; Wong, W.; Andres,
J . L.; Head-Gordon, M.; Replogle, E. S.; Pople, J . A. Gaussian
98; Gaussian, Inc., Pittsburgh, PA, 1998.
(12) Kole, L.; Das, L.; Das, P. K. Synergistic effect on interferon-
gamma and mannosylated liposome-incorporated doxorubicin in
the therapy of experimented visceral leishmaniasis. J . Infect.
Dis. 1999, 180, 811-820.
(13) Kapoor, P.; Sachdeva, M.; Madhubala, R. Effect of the microtu-
bule stabilizing agent taxol on leishmanial protozoan parasites
in vitro. FEMS Microbiol. Lett. 1999, 176, 429-435.
(14) Urbina, J . A. Lipid biosynthesis pathways as chemotherapeutic
targets in kinetoplastid parasites. Parasitology 1997, 114, 91-
99.
(15) Marsella, R.; Ruiz de Gopegui, R. Leishmaniasis: A re-emerging
zoonosis. Int. J . Dermatol. 1998, 37, 801-814.
(16) Pe´rez-Victoria, J . M.; Parodi-Talice, A.; Torres, C.; Gamarro, F.;
Castanys, S. ABC transporters in the protozoan parasite Leish-
mania. Intern. Microbiol. 2001, 4, 159-166.
(17) Pe´rez-Victoria, J . M.; Tincusi, B. M.; J imenez, I. A.; Bazzocchi,
I. L.; Gupta, M. P.; Castanys, S.; Gamarro, F.; Ravelo, A. G. New
natural sesquiterpenes as modulators of daunomycin resistance
in a multidrug-resistant Leishmania tropica line. J . Med. Chem.
1999, 42, 4388-4393.
(32) Steiner, T.; Koellner, G. Hydrogen bonds with π-acceptors in
proteins: Frequencies and role in stabilizing local 3D structures.
J . Mol. Biol. 2001, 305, 535-557.
J M0309699