Overcoming human P-glycoprotein-dependent multidrug resistance with novel dihydro-β-agarofuran sesquiterpenes

Article abstract of DOI:10.1016/j.ejmech.2011.07.048

Sixteen (1-16) dihydro-β-agarofuran sesquiterpenes were isolated from the fruits of Maytenus jelskii and evaluated against mammalian cells with a multidrug resistance phenotype mediated by the overexpression of the human P-glycoprotein. Their stereostruct

Full text of DOI:10.1016/j.ejmech.2011.07.048

European Journal of Medicinal Chemistry 46 (2011) 4915e4923
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Overcoming human P-glycoprotein-dependent multidrug resistance with novel
dihydro-
b
-agarofuran sesquiterpenes
,
,
Nayra R. Perestelo a 1, María P. Sánchez-Cañete b 1, Francisco Gamarro b, Ignacio A. Jiménez a,
**
Santiago Castanys b , Isabel L. Bazzocchi
,
a,
*
a Instituto Universitario de Bio-Orgánica “Antonio González”, Departamento de Química Orgánica, Universidad de La Laguna, Avenida Astrofísico Francisco Sánchez 2,
38206 La Laguna, Tenerife, Spain
b Instituto de Parasitología y Biomedicina López-Neyra, Consejo Superior de Investigaciones Científicas (IPBLN-CSIC), Parque Tecnológico de Ciencias de la Salud,
Avenida del Conocimiento s/n, 18100 Armilla, Granada, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Sixteen (1e16) dihydro-b-agarofuran sesquiterpenes were isolated from the fruits of Maytenus jelskii and
Received 24 May 2011
Received in revised form
25 July 2011
Accepted 26 July 2011
Available online 4 August 2011
evaluated against mammalian cells with a multidrug resistance phenotype mediated by the over-
expression of the human P-glycoprotein. Their stereostructures have been elucidated on the basis of
spectroscopic analysis, including 1D and 2D NMR techniques, CD studies, chemical correlations and
biogenetic means. Eight compounds from this series were discovered as potent chemosensitizers (1, 2, 4,
6, 8, 9, 11 and 14), showing similar effectiveness to or higher than the classical P-glycoprotein reversal
agent verapamil, a first-generation chemosensitizer, when reversing resistance to daunomycin and
vinblastine. Detailed structureeactivity relationships revealed that aromatic substituents at the 6 and
9-position of the sesquiterpene scaffold were able to modulate the intensity of inhibition.
Ó 2011 Elsevier Masson SAS. All rights reserved.
Keywords:
Maytenus jelskii
Celastraceae
Dihydro-b-agarofuran sesquiterpenes
Human MDR1/P-glycoprotein inhibitors
Structureeactivity relationship
1. Introduction
however, they lack specificity, require high doses to reverse
multidrug resistance, and are associated with unacceptable toxic-
Cancer is one of the most dangerous, fast propagating with quite
high mortality rate diseases of present century. Following heart
disease, cancer is the second leading cause of death in the devel-
oped world [1]. One of the major problems of cancer chemotherapy
is intrinsic or acquired drug resistance, and searching to overcome
it has been a major effort in clinical oncology [2]. A primary
mechanism of multiple-drug resistance (MDR) may due to the
overexpression of the ABC (ATP-binding cassette) transporter
P-glycoprotein (Pgp) in the plasma membrane of resistant cells,
which mediates the efflux of MDR drugs, reducing intracellular
accumulation of anticancer drugs [3]. Pgp-mediated drug transport
could be inhibited or altered by a wide range of agents (named as
MDR reversal agents, modulators or chemosensitizers). Verapamil,
a calcium channel blocker, was the first discovered Pgp inhibitor.
Since then, various reversal agents were brought to light, including
first-generation Pgp inhibitors, e.g. quinidine, and cyclosporine A,
ities. Second-generation agents as e.g. valspodar, elacridar, bir-
icodar, and dexverapamil, are more potent, specific and less toxic
compared to the former inhibitors, but they show interactions with
other transporter proteins. Third-generation Pgp inhibitors, e.g.
tariquidar, zosuquidar, laniquidar, and ONT-093, have high potency
and specificity for Pgp and in addition have shown no clinically
significant drug interactions with common chemotherapeutic
agents. However, these compounds were ineffective or toxic at the
doses required to attenuate Pgp function, or induced unfavorable
pharmacokinetic interactions, resulting in failure in their clinical
trials. Although progress in the development of ideal Pgp inhibitors
has been slow, it is believed that this approach is still realistic and
promising for overcoming MDR [4].
Natural products from plant and their synthetic derivatives play
an important role in the development of innovative drugs, especially
for the treatment of infections and cancer. Some natural products of
relatively low toxicity are reported to be inhibitors of one or more
ABC drug efflux pumps, such as terpenoids, coumarins and steroids,
and there is a wealth of literature demonstrating the modulator
effects of plant flavonoids and carotenoids. As yet, the potential use
of such natural products has not been satisfactorily explored [5].
* Corresponding author. Tel.: þ34 922 318594; fax: þ34 922 318571.
** Corresponding author. Tel.: þ34 958 181666; fax: þ34 958 181632.
E-mail addresses: castanys@ipb.csic.es (S. Castanys), ilopez@ull.es (I.L. Bazzocchi).
1
Both authors contributed equally to this work.
0223-5234/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2011.07.048

4916
N.R. Perestelo et al. / European Journal of Medicinal Chemistry 46 (2011) 4915e4923
In our previous research on reversal agents of the Pgp-dependent
MDR phenotype from plant, dihydro- -agarofuran sesquiterpenes
LH-20 to yield 16 sesquiterpene esters (1e16, Scheme 1). The
structures of the new compounds, 1e5 and 9e16, were deduced
as described.
b
(DAS) isolated from Celastraceae species have demonstrated potent
reversing effect on the cytotoxicity of the antitumoral drugs
daunomycin (DNM) and vinblastin (VBT) against MDR human
cancer cells [6,7] and Leishmania [8], which was comparable to
verapamil. Thus, to continue our research toward the discovery of
new DAS as potential Pgp inhibitors, we report herein on the
isolation, structure elucidation and bioactivity of thirteen new (1e5
and 9e16) and three known (6 [9], 7 [10] and 8 [9]) sesquiterpenes
Compound 1 was obtained as a colorless lacquer and gave the
molecular formula C38H40O13 by HRESIMS. The IR spectrum showed
absorption bands for hydroxy (3547 cmꢀ1), ester (1746, 1720 cmꢀ1
)
and aromatic (759, 713 cmꢀ1) groups. The mass spectrum exhibited
peaks attributable to the presence of methyl ([M ꢀ 15]þ, m/z 689),
benzoate ([M ꢀ 122]þ, m/z 582 and [M ꢀ 2 ꢁ122]þ, m/z 460),
furoate ([M ꢀ 460 ꢀ 112]þ, m/z 348) and acetate ([M ꢀ 122 ꢀ 60]þ,
m/z 522) groups. This was confirmed by the 1H and 13C NMR data,
which included signals of thirteen aromatic protons (dH 6.78e8.19),
sixteen aromatic carbons (dC 110.1e148.6) and three carboxyl
groups at dC 161.7, 165.7 and 166.0, assigned to one furoyl and two
benzoyl moieties, in addition to signals corresponding to two acetyl
groups [dH 1.73 (s), dC 20.1 (q), 169.5 (s) and dH 2.36 (s), dC 20.5 (q),
170.7 (s)]. Its 1H NMR spectrum (Table 1) displayed signals for two
with a dihydro-b-agarofuran skeleton from Maytenus jelskii Zahl-
bruchner. Their stereostructures were elucidated by application of
1D and 2D NMR techniques, including COSY, HSQC, HMBC and
ROESY experiments, and their absolute configurations were deter-
mined by CD studies, chemical correlations or biosynthetic grounds.
The isolated compounds have been tested on human MDR1-
transfected NIH-3T3 mammalian cells, in order to determine their
ability to reverse the MDR phenotype due to Pgp overexpression. All
compounds were able to block Pgp transport with low intrinsic
cytotoxicity, and eight of them (1, 2, 4, 6, 8, 9, 11 and 14) showed an
effectiveness that was similar to (or higher than) the classical Pgp
inhibitor verapamil for the reversion of resistance to DNM and VNB.
The structureeactivity relationships are discussed.
methylene systems at
d
2.27 (2H, m) and 2.32 (1H, m), 2.72 (1H,
4.52,
ddd, J ¼ 3.5, 7.0, 15.9 Hz) and one oxymethylene system at
d
5.31 (2H, dAM, J ¼ 12.8 Hz), assigned to protons H2-3, H2-8 and
H2-15, respectively. Resonances for four oxymethine protons at
d
5.47 (d, J ¼ 7.0 Hz, H-9),
J ¼ 3.5 Hz, H-2) and 6.33 (s, H-6) and one aliphatic methine proton
at
d
5.70 (d, J ¼ 3.5 Hz, H-1),
d 5.83 (br d,
d
d
2.41 (t, J ¼ 3.5 Hz, H-7) were also observed. When 1 was treated
with acetic anhydride in pyridine, compound 1 was unaltered,
this fact together with the presence of a singlet at dH 3.12, inter-
changeable with D2O, confirmed the presence of a tertiary hydroxy
2. Results and discussion
2.1. Chemistry
group. A methyl group at
d 1.60 (H3-14) attached to a carbon at
d
69.8, bearing a hydroxy group, and two methyls at d 1.63 (H3-12)
The hexaneeEt2O extract of the fruits of M. jelskii was sub-
jected to multiple chromatographic steps, involving vacuum
liquid chromatography (VLC), medium-pressure liquid chroma-
tography (MPLC) and preparative TLC on silica gel and Sephadex
and 1.60 (H3-13) were confirmed from the 13C NMR spectrum
(Table 1). All these data indicate that 1 is a polyester sesquiterpene
with a 1,2,4,6,9,15-hexasubstituted dihydro-b-agarofuran skeleton.
OAc
OR3
OAc
AcO
OFu
AcO
AcO15
OFu
BzO
R1O
RO
1
9
2
5
7
4
6
12
11
O
O
HO
O
HO
HO
14
OBz
OR2
OBz
13
9
4 R1 = H, R2 = R3 = Bz
5 R1 = R2 = Ac, R3 = Fu
6 R1 = Ac, R2 = R3 = Bz
7 R1 = R2 = Ac, R3 = Bz
8 R1 = Ac, R2 = R3 = Fu
1 R = Bz
2 R = Ac
3 R = H
a)
b)
FuO
OBz
R1
O
FuO
OBz
FuO
OR
R2
AcO
AcO
AcO
15 R1 = OAc, R2 = H
16 R1 = H, R2 = OH
O
O
R
HO
10 R = OH
11 R = H
12 R = Bz
13 R = Fu
14 R = Cin
a) BzCl, imidazole, DMAP
b) Ac2O, Et3N, DMAP
O
O
O
β
O
1
3
3
3
α
1
Bz =
Ac =
Cin =
Fu =
2
5
O
5
5
Scheme 1. Structure of compounds 1e16 and chemical correlations.

N.R. Perestelo et al. / European Journal of Medicinal Chemistry 46 (2011) 4915e4923
4917
The regiosubstitution of 1 was determined by an HMBC exper-
iment, showing three-bond correlations between the carboxyl
signals of the acetate groups at dC 169.5 and dC 170.7 and the
resonances at dH 5.70 (H-1) and dH 4.52, 5.31 (H2-15), respectively.
The carboxyl signal of the furoate group at dC 161.7 was correlated
with the signal at dH 5.47 (H-9), whereas the carboxyl signal of the
benzoate groups at dC 165.7 and dC 166.0 were linked to the reso-
nance at dH 6.33 (H-6) and dH 5.83 (H-2), respectively. The relative
stereochemistry of 1 was established on the basis of the coupling
constants and confirmed by a ROESY experiment. Therefore, the
J1,2 (3.5 Hz) indicated a cis relationship between H-1/H-2, and ROE
effects of H2-15 to H-6, H-9 and H3-14, and those of H-1 to H-2 and
H-3b were observed in the ROESY experiment (Fig. 1). The absolute
configuration of 1 was resolved by the dibenzoate chirality method,
an extension of the circular dichroism exciton chirality procedure
[11]. The angle (57.3ꢂ) between the two chromophores (benzoate
groups) was calculated by molecular mechanic calculations using
the PC model [12]. Therefore, compound 1 was considered suitable
for a CD study, showing a Davidoff-type split curve with a first
negative Cotton effect at 235.7 nm (
effect at 221.3 nm (
þ5.7), due to the couplings of the two
benzoate groups at C-2 and C-6 (Fig. 1). Thus, the absolute
D
3 ꢀ8.3) and a second positive
D
3
a
b
configuration of 1 was established as (1R,2S,4S,5S,6R,7R,9S,10R)-
1,15-diacetoxy-2,6-dibenzoyloxy-9-(3-furoiloxy)-4-hydroxy-dihydro-
b-agarofuran.
Compound 2 showed a molecular formula of C33H38O13 by
HRESIMS. The 1H and 13C NMR data (Table 1) showed that it had
similar features to those of compound 1, with the only difference
being the replacement of the benzoate group at C-2 in 1 by an
acetate group in 2. 2D NMR (COSY, ROESY, HSQC and HMBC)
experiments allowed the complete and unambiguous assignment
of the chemical shifts, regiosubstitution and relative configuration
of compound 2. The HRESIMS of compound 3 gave a molecular
formula of C31H36O12. Its 1H and 13C NMR data (Table 1) were
closely related to those of 2, the main difference was that the
resonances assigned to the acetate group at C-2 in 2 were replaced
by those for a hydroxy group in 3, together with the shift of the
signal corresponding to the H-2 from dH 5.50 in 2 to dH 4.33 in 3.
These data suggested that 3 was the 2-deacetyl derivative of 2, this
was supported by 2D NMR analysis, which verified the 1D NMR
data, relative configuration and regiosubstitution assignments. The
absolute configuration of 2 and 3 was established by chemical
correlation. Thus, benzoylation and acetylation of
3 yielded
compounds whose spectroscopic data were identical to those of 1
and 2, respectively (Scheme 1).
Compounds 4 and 5 showed spectroscopic data similar to 1e3
and to the other known isolated compounds (6e8) [9,10]. Conse-
quently, the stereostructure elucidation of these molecules was
greatly aided by comparison of their spectroscopic data. Even so, it
should be noted that a complete set of 2D NMR spectra (COSY, HSQC
and HMBC) was acquired for each metabolite in order to gain the
13
13
H
H
H
H
O
O
H
H
OFu
OAc
OFu
OAc
12
12
1
1
H
H
2
2
HO
BzO
H
HO
BzO
H
9
9
4
4
H
H
BzO
AcO
BzO
6
6
15
14
14
15
H
H
H
H
CD [-]
CD [-]
Fig. 1. ROE effects (solid line) and CD exciton coupling (dashed line) for compounds 1
(left) and 9 (right).

4918
N.R. Perestelo et al. / European Journal of Medicinal Chemistry 46 (2011) 4915e4923
complete and unambiguous assignment of the 1H and 13C NMR
resonances as listed in Table 1. The same relative stereochemistry
previously assigned to 1 was also confirmed for 4 and 5 by 2D NMR
ROESY experiments, and their absolute configurations were
assumed based on their having the same polyhydroxy sesquiter-
pene core as compound 1.
Compound 9, with a molecular formula of C36H38O11 (HRESIMS),
was revealed to be the 15-deacetoxy derivative of 1 by analysis of its
1D and 2D NMR spectra (Table 1 and Experimental section, Fig. 1).
Thus, comparison of their NMR data showed the replacement of the
resonances for an acetate group at C-15 (dH 4.52, 5.31, dC 66.0) in 1
by resonances for a methyl group (dH 1.69, dC 21.8) in 9, along with
the 1He13C long-range correlations for Me-15/C-1, C-5, C-9 and
C-10 observed in the HMBC experiment. The absolute configuration
of 9 was resolved by a circular dichroism study [11], showing
a Davidoff-type split curve with a first negative Cotton effect at
235.5 nm (
þ6.3), due to the couplings of the two benzoate groups at
and C-6 (Fig. 1). Therefore, the absolute configuration of
was established as (1R,2S,4S,5S,6R,7R,9S,10R)-1-acetoxy-2,6-
D
3
ꢀ7.6) and a second positive effect at 221.6 nm
(D 3
C-2
a
b
9
dibenzoyloxy-9-(3-furoyloxy)-4-hydroxy-dihydro-b-agarofuran.
Compound 10 was obtained as a colorless lacquer. The 1H and
13C NMR spectra (Table 2) combined with the mass data were used
to established its structure as that of a 1,2,3,4,9-pentasubstituted
dihydro-b-agarofuran sesquiterpene polyester containing two
acetate, one benzoate, one furoate and one hydroxy groups. The
structure and NMR data assignments of 10 were proved by the 2D
NMR data. Particularly, the regiosubstitution was determined by
the long-range 1He13C HMBC couplings observed between the H-1
(dH 6.18), H-2 (dH 5.19), H-3 (dH 5.24) and H-9 (dH 5.02) proton
resonances and the carboxyl signals of the furoate, (dC 160.8)
acetate (dC 170.1 and 170.2) and benzoate (dC 165.2) groups,
respectively. The relative configuration was established on the basis
of the coupling constants and confirmed by a ROESY experiment
(Fig. 2). Therefore, the J1,2 (10.8 Hz) and J2,3 (2.8 Hz) values indicated
a
trans and cis relationship between H-1/H-2 and H-2/H-3,
respectively. The ROESY experiment showed ROE effects of H-2 to
H3-14 and H3-15, those of H-3 to H3-14 and H-9 to H3-15. The
absolute configuration of 10 was assumed based on it having the
same polyhydroxy core as another sesquiterpene previously iso-
lated from the leaves of M. jelskii and whose absolute configuration
has been established by a CD study [13].
The molecular formula C31H36O10 of 11 indicated one fewer
oxygen atoms than 10. Comparison of their NMR data (Table 2 and
Experimental section, Fig. 2) showed that the resonances of the
characteristic tertiary hydroxy group at C-4 (dH 3.43 interchange-
able with D2O, dC 69.7) in 10 were replaced by those of an aliphatic
methine group (dH 2.41, dC 43.4) in 11. These data suggested that 11
was the 4-deoxy-derivative of 10. The HREIMS of compound 12
gave a molecular ion at m/z 526.2185, corresponding to a molecular
formula of C29H34O9. Its 1H and 13C NMR data (Table 2) were closely
related to those of 10, the predominant differences being the
replacement of the resonances corresponding to the acetate group
on C-3 by those of an aliphatic methylene (dH 2.09, dC 42.7) in 12.
These data suggested that 12 was the 3-deacetoxy derivative of 10.
Compounds 13 and 14 gave molecular formulas of C27H32O10 and
C29H34O9, respectively, by HREIMS. Their 1H and 13C NMR data
(Table 2) showed the same 1,2,4,9-tetrasubstituted dihydro-b-
agarofuran skeleton as 12. Consequently, their stereostructural
elucidation was greatly aided by the comparison of their spectro-
scopic data with those for 12, the only difference being the ester
group located at C-9. In fact, the NMR data for compounds 13 and 14
showed signals corresponding to a furoate and a cinnamate group,
respectively, at this position. Even, so, it should be noted that
a complete set of 2D NMR spectra (COSY, ROESY, HSQC and HMBC)

N.R. Perestelo et al. / European Journal of Medicinal Chemistry 46 (2011) 4915e4923
4919
13
configuration was previously established. Compounds 11e14 are
also likely to be biosynthetic intermediates of compound 10. The
H
AcO
O
OAc
OBz
OFu
new compounds have the basic polyhydroxy-dihydro-
b-agarofuran
12
1
sesquiterpenoid core of 3-deoxymaytol [14] (1e5),
2a,4b-
H
H
2
dihydroxycelorbicol [15] (9), 6-deoxymagellanol [16] (10), 4,6-
dideoxymagellanol [16] (11), 3,6-dideoxymagellanol [16] (12e14),
R
H
9
4
H
2-epi-isocelorbicol [15] (15) and 4
[15] (16).
b-hydroxy-6-deoxycelorbicol
H
6
15
14
H
H
2.2. Biological evaluation
Fig. 2. ROE effects for compounds 10 and 11.
The biological activity of the isolated sesquiterpenes (1e16) was
monitored through their ability to inhibit Pgp-mediated DNM
efflux and to reverse cellular MDR in NIH-3T3 MDR1G-185
mammalian cells (MDR1 cells). Flow cytometry analysis shows that
was acquired for each metabolite in order to gain the complete and
unambiguous assignment of the 1H and 13C NMR resonances as
listed in Table 2.
all of them demonstrated a Pgp inhibitory activity at 10 mM, and the
eight most active compounds were able to block up to 70% of DNM
transport activity of the protein (Fig. 3). Afterward, the Ki
(concentration that inhibits DNM transport by 50%) was deter-
mined with values of 0.34 ꢃ 0.09, 0.59 ꢃ 0.16, 0.47 ꢃ 0.09,
The molecular formula C29H34O8 of compound 15 (HREIMS)
had one fewer oxygen atom than 12. Comparison of their NMR
data (Table 2 and Experimental section) showed that the reso-
nances of the characteristic tertiary hydroxyl group at C-4 (dH
2.91, interchangeable with D2O; dC 70.6) in 12 were replaced by
those of an aliphatic methine group (dH 2.10, dC 40.7) in 15. These
data suggested that 15 was the 4-deoxy-derivative of 12,
proposal supported by combined analysis of 1D and 2D spectra.
The HREIMS of compounds 16 gave a molecular ion at m/z
468.2129, corresponding to a molecular formula C27H32O7. Its 1H
and 13C NMR data (Table 2) were closely related to those of 12,
except for the absence of the signals assigned to the acetate
group at C-2 and the shift of the signals corresponding to the C-2
position from dH 5.11 and dC 69.1 in 12 to dH 1.56, 2.09 and dC 23.8
in 16. These data indicated that 16 was the 3-deacetoxy deriv-
ative of 12. Detailed assignments of the 1H and 13C NMR data and
its relative configuration were determined based on 2D spectra
analysis.
0.39 ꢃ 0.05, 0.51 ꢃ0.04, 0.38 ꢃ 0.10, 0.64 ꢃ 0.16, and 0.57 ꢃ 0.18
mM
for compounds 1, 2, 4, 6, 8, 9, 11 and 14, respectively. These eight
sesquiterpenes possess Imax values (maximal percentage inhibition
of DNM transport) higher than 90% (data not shown). This effi-
ciency in blocking DNM transport was correlated with their ability
to sensitize MDR1 cells to the cytotoxic drugs DNM and VNB. In this
regard, the eight most efficient sesquiterpenes in blocking DNM
efflux were able to reverse DNM and VNB resistance in MDR1 cells
with reversal index values comparable to or higher than the clas-
sical Pgp inhibitor verapamil (Table 3).
These sesquiterpenes showed similar toxicity to verapamil for
both wild-type and MDR1 cells (Table 4). One conclusion can be
drawn from these results: Pgp does not confer cross-resistance to
sesquiterpenes or, in other words, the sesquiterpenes are either
poor Pgp substrates or not transported at all by the Pgp. Similar
results were found for other sesquiterpenes in our previous
work [7].
The absolute configuration of compounds 11e16 could be
assumed based on biosynthetic considerations. We propose that
the biosynthetic pathway involves stereospecific introduction of
hydroxyl groups while retaining the configuration in the poly-
hydroxy sesquiterpene core. Thus, compounds 15 and 16 could
be considered early precursors to the more functionalized basic
polyhydroxy sesquiterpene core of compound 10, whose absolute
2.3. SAR analysis
The influence of the substitution pattern of the dihydro-
b-
agarofuran skeleton on the MDR reversal activity of the
Fig. 3. Inhibition of Pgp-mediated DNM transport across the cell membrane in intact mammalian MDR1 cells. The cells were co-incubated with 2
mM DNM and 10 mM of
sesquiterpenes, the DNM retention was determined by flow cytometry analysis as described in Biological assays. DNM fluorescence intensity is expressed as relative fluorescence
compared to a control (5 mM ortho-vanadate), representing 100% inhibition of Pgp. C ¼ MDR1 cells incubated with DNM and without compounds. V ¼ 10
Pgp reversal agent used for comparison. Data are the mean ꢃ S.D. of three independent experiments performed in triplicate.
mM verapamil, a classical

4920
N.R. Perestelo et al. / European Journal of Medicinal Chemistry 46 (2011) 4915e4923
Table 3
Drug resistance reversal ability of compounds in MDR1 cells.a
Compound
Reversal index with DNMb
Reversal index with VNBc
1
mM
3
m
M
10
m
M
1
mM
3
m
M
10 mM
1
2
4
6
8
9
11
14
7.82 ꢃ 2.79
4.50 ꢃ 0.80
2.21 ꢃ 1.02
14.07 ꢃ 2.82
11.97 ꢃ 3.67
6.25 ꢃ 1.54
19.80 ꢃ 3.22
4.37 ꢃ 1.13
10.35 ꢃ 1.68
3.68 ꢃ 0.72
7.77 ꢃ 1.78
8.79 ꢃ 2.35
17.11 ꢃ 3.68
13.27 ꢃ 2.33
8.82 ꢃ 2.57
10.34 ꢃ 4.41
7.51 ꢃ 1.54
2.55 ꢃ 1.15
28.60 ꢃ 3.44
3.58 ꢃ 1.09
11.14 ꢃ 1.81
3.50 ꢃ 0.71
4.89 ꢃ 1.21
6 .23 ꢃ 0.56
14.26 ꢃ 4.43
13.21 ꢃ 2.77
13.26 ꢃ 0.80
52.56 ꢃ 2.06
18.85 ꢃ 2.54
30.32 ꢃ 5.77
22.36 ꢃ 2.46
19.90 ꢃ 1.16
20.85 ꢃ 2.84
51.67 ꢃ 2.69
58.13 ꢃ 5.41
36.60 ꢃ 4.36
165.10 ꢃ 5.82
65.64 ꢃ 2.07
125.31 ꢃ 6.25
73.92 ꢃ 5.07
83.66 ꢃ 3.88
76.81 ꢃ 7.01
12.73 ꢃ 1.39
2.72 ꢃ 0.62
4.36 ꢃ 1.04
1.93 ꢃ 0.38
1.59 ꢃ 0.44
3.04 ꢃ 0.57
21.24 ꢃ 3.09
14.63 ꢃ 0.17
33.97 ꢃ 4.21
10.44 ꢃ 1.85
16.57 ꢃ 2.60
10.64 ꢃ 1.97
Verapamil
a
The reversal index is the fined as the ratio between the IC50 of cells without compound and the IC50 with compound. IC50 values were determined as described in the
Biological assays section. Results are expressed as the mean ꢃ S.D. of three independent experiments performed in triplicate.
b
The maximum reversal index with DNM (ratio between IC50 for MDR1 and parental drug-sensitive cells) was 55.
The maximum reversal index with VNB (ratio between IC50 for MDR1 and parental drug-sensitive cells) was 250.
c
sesquiterpenes studied was examined, revealing the following
trends on the preliminary structureeactivity relationship of this
series of natural sesquiterpenes. In general, the more active
compounds against DNM present an ester group at C-6
(compounds 1e9). Compounds with an aromatic ester group
(benzoate or furoate) at C-6 are more active than those with an
acetate group in such position (6 and 8 versus 5 and 7, respectively).
by the human Pgp overexpression. Compounds 6 and 9 were the
most active agents, reversing both DNM and VBN resistance and
showed higher activity than the reference reversal agent verapamil.
Structureeactivity relationships among sesquiterpenes showed
that the overall esterification level of the compound and the
presence of aromatic ester groups at C-6 and C-9 seems to be an
important element to enhance the MDR reversal activity, findings
that are consistent with our previous studies. The elucidation of the
pharmacophore will be helpful for the rational design of more
potent and specific Pgp inhibitors with potential use in the treat-
ment of MDR cancers.
Results showed
a reversal activity sequence of cinnama-
te > benzoate > furoate groups at C-9 (14 versus 12 and 13; 6 versus
2). Furthermore, an acetate group at C-3 (10 versus 12) or C-15 (2
versus 9) reduces the reversal activity. In addition, the presence of
a hydroxyl group at C-2 produces a decrease in activity (3 versus 1
and 2; 4 versus 6). No straightforward conclusion can be drawn
4. Experimental
from the presence/absence of the hydroxy group at C-4b, given that
its presence (11 versus 10; 12 versus 15) decreases or increases,
respectively, the DNM accumulation in MDR1 cells. Similar corre-
lation was found for the VNB reversal activity, especially when the
analysis is performed with lower concentrations of sesquiterpenes.
These trends, based upon substitution patterns, are consistent with
our previous SAR findings [6e8], and reinforce our statement that
the overall esterification level of the molecule and the presence of
aromatic ester groups seems to be responsible for the potent MDR
reversal activity of this type of metabolites.
4.1. General
Optical rotations were measured on a Perkin Elmer 241 auto-
matic polarimeter in CHCl3 at 25 ꢂC and the [
a]D values are given in
10ꢀ1 deg cm2 gꢀ1. UV spectra were obtained in absolute EtOH on
a JASCO V-560 instrument. IR (film) spectra were measured in
CHCl3 on a Bruker IFS 55 spectrophotometer. 1H (300 and 400 MHz)
and 13C (75 and 100 MHz) NMR spectra were recorded on a Bruker
Avance 300 and 400 spectrometers; chemical shifts were referred
to the residual solvent signal (CDCl3: dH 7.26, dC 77.0); DEPT, COSY,
ROESY (spin lock field 2500 Hz), HSQC and HMBC (optimized for
J ¼ 7.7 Hz) experiments were carried out with the pulse sequences
given by Bruker. EIMS and HREIMS were recorded on a Micromass
Autospec spectrometer, and HRESIMS (positive mode) were
measured on an LCT Premier XE Micromass Electrospray spec-
3. Conclusion
A new series of dihydro-b-agarofuran sesquiterpenes isolated
from M. jelskii have been evaluated as reversal agents of the MDR
phenotype in mammalian cells. Eight of the isolated sesquiterpenes
showed a potent reversal activity of the MDR phenotype mediated
trometer. Silica gel 60 (15e40 mm) for column chromatography,
silica gel 60 F254 for TLC, and nanosilica gel 60 F254 for preparative
HPTLC were purchased from Macherey-Nagel, and Sephadex LH-20
was obtained from Pharmacia Biotech. The spots were visualized
by UV light and heating silica gel plates sprayed with H2Oe
H2SO4eAcH (1:4:20). All solvents used were analytical grade from
Panreac. Reagents were purchased from SigmaeAldrich and
used without further purification. Compounds used for CD were
purified by high-performance TLC and eluted with a mixture of
n-hexaneeEtOAc.
Table 4
Intrinsic toxicity of sesquiterpenes in parental and MDR1 cells.a
Compound
Parenteral cells
MDR1 cells
1
2
4
6
8
9
11
14
82.14 ꢃ 1.62
92.36 ꢃ 5.76
84.38 ꢃ 4.43
76.70 ꢃ 1.61
91.98 ꢃ 5.83
108.24 ꢃ 3.56
91.78 ꢃ 8.12
85.95 ꢃ 4.13
102.91 ꢃ 3.32
84.33 ꢃ 2.65
87.84 ꢃ 0.18
104.77 ꢃ 13.21
77.21 ꢃ 4.23
82.86 ꢃ 2.45
98.02 ꢃ 1.84
92.37 ꢃ 5.82
100.10 ꢃ 0.12
79.18 ꢃ 8.93
4.2. Plant material
Verapamil
M. jelskii was collected in the Urubamba Privince, Cuzco, Perú, in
December 2006. A voucher specimen (CUZ 29845) was identified
by Professor Alfredo Tupayachi Herrera and deposited in the
Herbarium of Missouri Botanical Garden, St. Louis, MO.
a
Toxicity was determined using 10
m
M of compounds. Results are expressed as
percentage mean of cell growth ꢃ S.D. with respect to the control without
compounds. Data are the mean of three independent experiments performed in
triplicate.

N.R. Perestelo et al. / European Journal of Medicinal Chemistry 46 (2011) 4915e4923
4921
4.3. Extraction and isolation
HRESIMS m/z 665.2202 [M þ Na]þ (calcd for C33H38O13Na,
665.2210).
The fruits (250 g) of M. jelskii were extracted three times with n-
hexaneeEt2O (1:1) for 24 h, each time in a Soxhlet apparatus.
Removal of the solvent under reduced pressure provided 42 g of
crude extract, which was chromatographed by dry flash chroma-
tography on a silica gel column, using mixtures of n-hexaneeEtOAc
of increasing polarity to afford 30 fractions which were combined
in six fractions (AeF) on the basis of their TLC profiles. Preliminary
NMR studies revealed that four fractions (CeF) were found to be
4.3.3. (1R,2S,4S,5S,6R,7R,9S,10R)-1,15-Diacetoxy-6-benzoyloxy-9-
(3-furoyloxy)-2,4-dihydroxy-dihydro-
b
-agarofuran (3)
25
Colorless laquer; [
(log ) 230 (3.3), 218 (3.1) nm; IR nmax (film) 3516, 2926, 2853, 1717,
1278, 1234, 1141, 762, 713 cmꢀ1; 1H NMR (CDCl3)
1.80 (3H, s, OAc-
a]
þ31.3 (c 0.4, CHCl3); UV lmax (EtOH)
D
3
d
1), 2.30 (3H, s, OAc-15), OFu [6.75 (1H, d, J ¼ 1.7 Hz, H-4), 7.45 (1H,
m, H-5), 8.03 (1H, d, J ¼ 1.7 Hz, H-2)], OBz-6 [7.46 (2H, m, H-3, H-5),
7.58 (1H, m, H-4), 8.19 (2H, d, J ¼ 7.0 Hz, H-2, H-6)], for other signals,
rich in dihydro-b-agarofuran sesquiterpenoids and were further
investigated. These fractions were purified after several chroma-
tographies on Sephadex LH-20 using n-hexaneeCHCl3eMeOH
(2:1:1) as eluant. The obtained fractions were chromatographed by
wet flash chromatography on a silica gel column, using mixtures of
CH2Cl2eMe2CO of increasing polarity, and further purified by
preparative TLC with n-hexaneedioxane (6:4), n-hexaneeEt2O
(2:8) or CH2Cl2eMe2CO (8:2) to give the compounds. Following this
procedure, fraction C led to the isolation of compounds 7 (2.7 mg), 9
(32.5 mg), 15 (5.0 mg) and 16 (8.0 mg). In the same way, fraction D
afforded compounds 1 (40.8 mg), 2 (225.7 mg), 6 (25.8 mg) and 11
(14.6 mg). Fraction E gave compounds 8 (3.4 mg), 10 (40.5 mg), 12
(18.6 mg), 13 (3.1 mg) and 14 (78.6 mg). Finally, fraction F yielded
compounds 3 (23.2 mg), 4 (1.3 mg) and 5 (6.5 mg). Compounds 1
and 9, used for CD studies, were further purified by high-
performance TLC, using as eluant n-hexaneeEtOAc 1:1 and 6:4,
respectively.
see Table 1; 13C NMR (CDCl3)
d OFu-9 [109.8 (d, C-4), 118.6 (s, C-3),
143.8 (d, C-5), 148.7 (d, C-2), 161.8 (s, COOe)], OBz-6 [128.6 (2 ꢁ d,
C-3, C-5), 129.9 (s, C-1), 130.1 (2 ꢁ d, C-2, C-6), 133.3 (d, C-4), 166.6
(s, COOe)], OAc [(20.8 q, 169.9 s, OAc-1), (21.3 q, 170.6 s, OAc-15)],
for other signals, see Table 1; ESIMS m/z 623 [M þ Na]þ (100);
HRESIMS m/z 623.2098 [M þ Na]þ (calcd for C31H36O12Na,
623.2104).
4.3.4. Benzoylation of 3
A mixture of benzoyl chloride (4 drops), compound 3 (3.2 mg),
4-(dimethylamino)-pyridine (2.0 mg), and imidazole (2.0 mL) in
dry pyridine (2.0 mL) was heated under reflux for 16 h. The mixture
was evaporated to dryness and the residue was dissolved in Et2O
and washed with H2O, 10% NaHCO3, and brine. The organic layer
was stirred over KF (64.0 mg) for 1 h and filtered through Celite.
The filtrate was dried over anhydrous MgSO4, filtered, and
concentrated under reduced pressure. The residue was purified by
preparative TLC using n-hexaneeEtOAc (1:1) to give 1 (2.8 mg).
4.3.1. (1R,2S,4S,5S,6R,7R,9S,10R)-1,15-Diacetoxy-2,6-dibenzoyloxy-
9-(3-furoyloxy)-4-hydroxy-dihydro-b-agarofuran (1)
Colorless laquer, [
a
3
]
25 þ55.3 (c 1.0, CHCl3); CD lext (MeCN) 235.7
4.3.5. Acetylation of 3
(D
3
¼ ꢀ8.3), 221.3 (
D
¼Dþ5.7) nm; UV lmax (EtOH) (log
3 ) 230 (4.0),
A mixture of acetic anhydride (2 drops), triethyl-amine (4 drops),
compound 3 (2.9 mg), and 4-(dimethylamino)pyridine in dichloro-
methane (2 mL) was stirred at room temperature 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 evap-
orated to dryness, and the residue was purified by preparative TLC,
using n-hexaneeEtOAc (1:1) to give compound 2 (2.5 mg).
213 (3.8) nm; IR nmax (film) 3547, 2928, 1746, 1720, 1273, 1223, 1108,
1023, 759, 713 cmꢀ1; 1H NMR (CDCl3)
d 1.73 (3H, s, OAc-1), 2.36 (3H,
s, OAc-15), OFu-9 [6.78 (1H, s, H-4), 7.49 (1H, m, H-5), 8.07 (1H, s, H-
2)], OBz-6 [7.50 (2H, m, H-3, H-5), 7.60 (1H, m, H-4), 8.14 (2H, d,
J ¼ 7.9 Hz, H-2, H-6)], OBz-2 [7.50 (2H, m, H-3, H-5), 7.60 (1H, m, H-
4), 8.19 (2H, d, J ¼ 7.9 Hz, H-2, H-6)], for other signals, see Table 1;
13C NMR (CDCl3)
d OFu-9 [110.1 (d, C-4), 118.5 (s, C-3), 143.8 (d, C-5),
148.6 (d, C-2), 161.7 (s, COOe)], OBz-6 [128.6 (2 ꢁ d, C-3, C-5), 129.3
(s, C-1),129.8 (d, C-2), 129.9 (d, C-6), 133.4 (d, C-4), 165.7 (s, COOe)],
OBz-2 [128.7 (2 ꢁ d, C-3, C-5), 129.7 (s, C-1), 130.0 (d, C-2), 130.2 (d,
C-6), 133.5 (d, C-4), 166.0 (s, COOe)], OAc [(20.1 q, 169.5 s, OAc-1),
(20.5 q, 170.7 s, OAc-15)], for other signals, see Table 1; EIMS m/z
689 [M ꢀ CH3]þ (1), 582 (1), 522 (1), 460 (22), 348 (1), 275 (7), 192
(7), 105 (100), HREIMS m/z 689.2234 [M ꢀ CH3]þ (calcd for
C37H37O13, 689.2241); HRESIMS m/z: 727.2352 [M þ Na]þ (calcd for
C38H40O13Na, 727.2367).
4.3.6. (1R,2S,4S,5S,6R,7R,9S,10R)-1,15-Diacetoxy-6,9-dibenzoyloxy-
2,4-hydroxy-dihydro-
Colorless laquer; [
(log ) 231 (4.1) nm; IR nmax (film) 3521, 2924, 2854, 1746, 1717,
1276, 1107, 1024, 772, 714 cmꢀ1; 1H NMR (CDCl3)
1.71 (3H, s, OAc-
b-agarofuran (4)
25
a
]
þ32.7 (c 0.1, CHCl3); UV lmax (EtOH)
D
3
d
1), 2.35 (3H, s, OAc-15), OBz-9 [7.50 (2H, m, H-3, H-5), 7.62 (1H, m,
H-4), 8.10 (2H, d, J ¼ 7.4 Hz, H-2, H-6)], OBz-6 [7.50 (2H, m, H-3, H-
5), 7.62 (2H, m, H-4), 8.23 (2H, d, J ¼ 7.6 Hz, H-2, H-6)], for other
signals, see Table 1; 13C NMR (CDCl3)
d
OBz-9 [128.4 (2 ꢁ d, C-3, C-
5), 129.1 (s, C-1), 130.1 (2 ꢁ d, C-2, C-6), 133.4 (d, C-4), 165.3 (s,
COOe)], OBz-6 [128.7 (2 ꢁ d, C-3, C-5),129.9 (s, C-1),130.2 (2 ꢁ d, C-
2, C-6), 133.6 (d, C-4), 166.0 (s, COOe)], OAc [(20.7 q, 169.7 s, OAc-1),
(21.4 q, 170.6 s, OAc-15)], for other signals, see Table 1; ESIMS m/z
633 [M þ Na]þ (100); HRESIMS m/z 633.2322 [M þ Na]þ (calcd for
C33H38O11Na, 633.2312).
4.3.2. (1R,2S,4S,5S,6R,7R,9S,10R)-1,2,15-Triacetoxy-6-benzoyloxy-
9-(3-furoyloxy)-4-hydroxy-dihydro-
b
-agarofuran (2)
25
Colorless laquer; [
(log ) 231 (3.9), 213 (3.7) nm; IR nmax (film) 3548, 2956, 2932, 1747,
1723, 1575, 1278, 1242, 1155, 758, 715 cmꢀ1; 1H NMR (CDCl3)
1.71
a]
þ40.5 (c 1.1, CHCl3); UV lmax (EtOH)
D
3
d
(3H, s, OAc-1), 2.10 (3H, s, OAc-2), 2.31 (3H, s, OAc-15), OFu-9 [6.74
(1H, d, J ¼ 0.7 Hz, H-4), 7.45 (1H, m, H-5), 8.02 (1H, d, J ¼ 0.7 Hz, H-
2)], OBz-6 [7.46 (2H, m, H-3, H-5), 7.57 (1H, m, H-4), 8.18 (2H, d,
J ¼ 8.1 Hz, H-2, H-6)], for other signals, see Table 1; 13C NMR (CDCl3)
4.3.7. (1R,2S,4S,5S,6R,7R,9S,10R)-1,2,6,15-Tetraacetoxy-9-(3-
furoyloxy)-4-hydroxy-dihydro-
b
-agarofuran (5)
25
Colorless laquer; [
(log ) 226 (2.9), 219 (3.8) nm; IR nmax (film) 3541, 2927, 2854, 1746,
1368, 1232, 1154, 1029, 760 cmꢀ1 1H NMR (CDCl3)
1.73 (3H, s,
a]
þ18.6 (c 0.6, CHCl3); UV lmax (EtOH)
D
d
OFu-9 [109.9 (d, C-4), 118.6 (s, C-3), 144.0 (d, C-5), 148.9 (d, C-2),
3
161.9 (s, COOe)], OBz-6 [128.8 (2 ꢁ d, C-3, C-5), 129.9 (s, C-1), 130.2
(2 ꢁ d, C-2, C-6), 133.5 (d, C-4), 166.1 (s, COOe)], OAc [(20.6 q, 169.6
s, OAc-1), (21.3 q, 169.8 s, OAc-2), (21.5 q, 170.7 s, OAc-15)], for other
signals, see Table 1; EIMS m/z 627 [M ꢀ CH3]þ (3), 582 (1), 515 (1),
460 (22), 418 (1), 348 (1), 275 (7), 192 (7), 105 (100), 95 (31);
HREIMS m/z 627.2087 [M ꢀ CH3]þ (calcd for C32H35O13, 627.2078);
;
d
OAc-1), 2.13 (3H, s, OAc-2), 2.14 (3H, s, OAc-6), 2.29 (3H, s, OAc-15),
OFu-9 [6.76 (1H, d, J ¼ 1.6 Hz, H-4), 7.45 (1H, d, J ¼ 1.6 Hz, H-5), 8.04
(1H, s, H-2)], for other signals, see Table 1; 13C NMR (CDCl3)
d OFu-9
[109.7 (d, C-4), 118.5 (s, C-3), 143.9 (d, C-5), 148.8 (d, C-2), 161.7 (s,
COOe)], OAc [(20.5 c,169.5 s, OAc-1), (21.1 q, 169.7 s, OAc-2), (21.3 q,

4922
N.R. Perestelo et al. / European Journal of Medicinal Chemistry 46 (2011) 4915e4923
170.6 s, OAc-15), (21.6 q, 170.3 s, OAc-6)], for other signals, see
Table 1; EIMS m/z 565 [M ꢀ CH3]þ (2), 520 (8), 460 (12), 453 (2), 351
(7), 291 (7), 228 (18), 192 (27), 95 (6); HREIMS m/z 565.1971
[M ꢀ CH3]þ, (calcd for C27H33O13 565.1921); HRESIMS m/z 603.2060
[M þ Na]þ (calcd for C28H36O13Na, 603.2054).
J ¼ 7.5 Hz, H-2, H-6)], for other signals, see Table 2; 13C NMR (CDCl3)
d
OFu-1 [109.4 (d, C-4), 118.8 (s, C-3), 143.4 (d, C-5), 147.3 (d, C-2),
161.1 (s, COOe)], OBz-9 [127.9 (2 ꢁ d, C-3, C-5), 129.8 (s, C-1), 130.2
(2 ꢁ d, C-2, C-6),132.9 (d, C-4),165.6 (s, COOe)], OAc (20.9 q,170.5 s,
OAc-2), for other signals, see Table 2; EIMS m/z 526 [M]þ (6), 510
(10), 465 (1), 388 (7), 362 (6), 329 (3), 232 (8), 105 (99), 95 (100);
HREIMS m/z 526.2185 [M]þ (calcd for C29H34O9, 526.2203).
4.3.8. (1R,2S,4S,5S,6R,7R,9S,10R)-1-Acetoxy-2,6-dibenzoyloxy-9-
(3-furoyloxy)-4-hydroxy-dihydro-b-agarofuran (9)
Colorless laquer; [
¼ ꢀ7.6), 221.6 (
a
]
25 þ49.5 (c 0.4, CHCl3); CD lext (MeCN) 235.5
4.3.12. (1S,2R,4S,5R,7R,9S,10R)-2-Acetoxy-1,9-di-(3-furoyloxy)-4-
D
(
D
3
D
3 ¼ þ6.3); UV lmax (EtOH) (log
3
) 229 (3.4) nm;
hydroxy-dihydro-
Colorless laquer; [
(log ) 274 (4.4), 223 (4.2) nm; IR nmax (film) 3517, 2925, 2853, 1738,
1311, 1241, 1160, 1008, 874, 758 cmꢀ1; 1H NMR (CDCl3)
1.86 (3H, s,
b-agarofuran (13)
25
IR nmax (film) 3546, 2955, 2927, 1719, 1275, 1110, 1024, 759,
a
]
þ45.5 (c 0.2, CHCl3); UV lmax (EtOH)
D
713 cmꢀ1; 1H NMR (CDCl3)
d
1.76 (3H, s, OAc-1), OFu-9 [6.76 (1H, d,
3
J ¼ 0.7 Hz, H-4), 7.46 (1H, m, H-5), 8.10 (1H, s, H-2)], OBz-2 [7.47
(2H, m, H-3, H-5), 7.60 (1H, m, H-4), 7.99 (2H, d, J ¼ 8.0 Hz, H-2, H-
6)], OBz-6 [7.50 (2H, m, H-3, H-5), 7.60 (1H, m, H-4), 8.25 (2H, d,
J ¼ 8.0 Hz, H-2, H-6)], for other signals, see Table 1; 13C NMR (CDCl3)
d
OAc-2), OFu-1 [6.45 (1H, s, H-4), 7.31 (1H, s, H-5), 7.76 (1H, s, H-2)],
OFu-9 [6.70 (1H, s, H-4), 7.39 (1H, s, H-5), 8.02 (1H, s, H-2)], for
other signals, see Table 2; 13C NMR (CDCl3)
d OFu-1 [109.3 (d, C-4),
d
OFu-9 [109.8 (d, C-4), 118.7 (s, C-3), 143.8 (d, C-5), 148.6 (d, C-2),
118.7 (s, C-3), 143.0 (d, C-5), 147.1 (d, C-2), 161.7 (s, COOe)], OFu-9
[109.8 (d, C-4), 118.9 (s, C-3), 143.3 (d, C-5), 148.5 (d, C-2), 162.4
(s, COOe)], OAc (20.7 q, 170.5 s, OAc-2), for other signals, see
Table 2; EIMS m/z 516 [M]þ (14), 501 (10), 474 (1), 456 (2), 389 (3),
362 (5), 250 (4), 95 (100); HREIMS m/z 516.1978 [M]þ (calcd for
C27H32O10, 516.1995).
161.9 (s, COOe)], OBz-2 [128.6 (2 ꢁ d, C-3, C-5), 129.5 (2 ꢁ d, C-2, C-
6), 129.7 (s, C-1), 133.4 (d, C-4), 165.8 (s, COOe)], OBz-6 [128.7
(2 ꢁ d, C-3, C-5), 129.9 (s, C-1), 130.2 (2 ꢁ d, C-2, C-6), 133.2 (d, C-4),
166.1 (s, COOe)], OAc (20.5 q, 169.8 s, OAc-1), for other signals, see
Table 1; EIMS m/z 631 [M ꢀ CH3]þ (1), 613 (1), 519 (2), 402 (12), 290
(6), 230 (5), 105 (100), 95 (23); HREIMS m/z 631.2168 [M ꢀ CH3]þ,
(calcd for C35H35O11 631.2179); HRESIMS m/z 669.2283 [M þ Na]þ
(calcd for C36H38O11Na, 669.2312).
4.3.13. (1S,2R,4S,5R,7R,9S,10R)-2-Acetoxy-9-trans-cynamoiloxy-1-
(3-furoyloxy)-4-hydroxy-dihydro-
b
-agarofuran (14)
25
Colorless laquer; [
(log ) 276 (4.1), 222 (4.0), 217 (4.0) nm; IR nmax (film) 3536, 2931,
1740, 1310, 1241, 1166, 1009, 757, 713 cmꢀ1; 1H NMR (CDCl3)
1.89
a]
D
þ124.7 (c 0.6, CHCl3); UV lmax (EtOH)
4.3.9. (1S,2S,3S,4S,5R,7R,9S,10R)-2,3-Diacetoxy-9-benzoyloxy-1-(3-
furoyloxy)-4-hydroxy-dihydro-
3
b
-agarofuran (10)
d
25
Colorless laquer; [
(log ) 231 (3.9) nm; IR nmax (film) 3524, 2927, 1745, 1714, 1305,
1245, 1137, 1011, 757, 714 cmꢀ1; 1H NMR (CDCl3)
1.77 (3H, s, OAc-
a
]
þ71.6 (c 0.4, CHCl3); UV lmax (EtOH)
(3H, s, OAc-2), OFu-1 [6.49 (1H, d, J ¼ 1.6 Hz, H-4), 7.30 (1H, t,
D
3
J ¼ 1.6 Hz, H-5), 7.80 (1H, d, J ¼ 1.6 Hz, H-2)], OCin-9 [6.44 (1H, d,
d
J ¼ 16.0 Hz, H-
H-
(CDCl3)
C-2), 161.6 (s, COOe)], OCin-9 [118.7 (d, C-
a
), 7.40 (3H, m, H-3, H-4, H-5), 7.52 (1H, d, J ¼ 16.0 Hz,
2), 2.26 (3H, s, OAc-3), OFu-1 [6.33 (1H, d, J ¼ 1.9 Hz, H-4), 7.66 (1H,
t, J ¼ 1.9 Hz, H-5), 7.31 (1H, d, J ¼ 1.9 Hz, H-2)], OBz-9 [7.45 (2H, m,
H-3, H-5)], 7.58 (1H, m, H-4), 8.01 (2H, d, J ¼ 7.7 Hz, H-2, H-6)], for
b
), 7.56 (2H, m, H-2, H-6)], for other signals, see Table 2; 13C NMR
OFu-1 [109.6 (d, C-4), 119.1 (s, C-3), 143.6 (d, C-5), 147.4 (d,
), 128.2 (2 ꢁ d, C-2, C-6),
),
d
a
other signals, see Table 2; 13C NMR (CDCl3)
d
OFu-1 [109.2 (d, C-4),
128.8 (2 ꢁ d, C-3, C-5), 130.1 (d, C-4), 134.6 (s, C-1), 144.6 (d, C-
b
118.7 (s, C-3), 143.3 (d, C-5), 147.1 (d, C-2), 160.8 (s, COOe)], OBz-9
[127.7 (2 ꢁ d, C-3, C-5), 129.7 (s, C-1), 129.9 (2 ꢁ d, C-2, C-6), 132.7
(d, C-4), 165.2 (s, COOe)], OAc [(20.4 q, 170.1 s, OAc-2), (20.8 q, 170.2
s, OAc-3)], for other signals, see Table 2; EIMS m/z 584 [M]þ (4), 569
(28), 524 (1), 447 (18), 430 (4), 370 (10), 248 (18), 105 (100), 95 (94);
HREIMS m/z 584.2266 [M]þ (calcd for C31H36O11, 584.2258).
166.2 (s, COOe)], OAc (20.9 q, 170.4 s, OAc-2), for other signals, see
Table 2; EIMS m/z 552 [M]þ (19), 537 (5), 421 (2), 389 (23), 362 (6),
249 (7), 232 (6), 131 (100), 95 (67); HREIMS m/z 552.2380 [M]þ
(calcd for C31H36O9, 552.2359).
4.3.14. (1S,2R,4R,5S,7R,9S,10R)-2-Acetoxy-9-benzoyloxy-1-(3-
furoyloxy)-dihydro-
Colorless laquer; [
(log ) 230 (3.8) nm; IR nmax (film) 2926, 2854, 1737, 1711, 1304,
1278, 1244, 1143, 1014, 757, 714 cmꢀ1; 1H NMR (CDCl3)
1.87 (3H, s,
b-agarofuran (15)
25
4.3.10. (1S,2R,3R,4R,5S,7R,9S,10R)-2,3-Diacetoxy-9-benzoyloxy-1-
a]
þ81.5 (c 0.2, CHCl3); UV lmax (EtOH)
D
(3-furoyloxy)-dihydro-
Colorless laquer; [
(log ) 230 (4.1) nm; IR nmax (film) 2926, 2854, 1738, 1277, 1238,
1150, 1010, 757, 714 cmꢀ1 1H NMR (CDCl3)
1.85 (3H, s, OAc-2),
b
-agarofuran (11)
3
25
a
]
D
þ18.1 (c 0.3, CHCl3); UV lmax (EtOH)
d
3
OAc-2), OFu-1 [6.37 (1H, s, H-4), 7.44 (1H, m, H-5), 7.70 (1H, s, H-2)],
OBz-9 [7.45 (2H, m, H-3, H-5), 7.57 (1H, m, H-4), 8.06 (2H, d,
J ¼ 7.7 Hz, H-2, H-6)], for other signals, see Table 2; 13C NMR (CDCl3)
;
d
2.26 (3H, s, OAc-3), OFu-1 [6.39 (1H, d, J ¼ 1.7 Hz, H-4), 7.31 (1H,
t, J ¼ 1.7 Hz, H-5), 7.71 (1H, d, J ¼ 1.7 Hz, H-2)], OBz-9 [7.45 (2H, m,
H-3, H-5), 7.57 (1H, m, H-4), 8.07 (2H, d, J ¼ 7.7 Hz, H-2, H-6)], for
d
OFu-1 [109.5 (d, C-4), 119.1 (s, C-3), 143.4 (d, C-5), 147.2 (d, C-2),
161.2 (s, COOe)], OBz-9 [127.9 (2 ꢁ d, C-3, C-5), 130.0 (s, C-1), 130.3
(2 ꢁ d, C-2, C-6), 132.8 (d, C-4), 165.8 (s, COOe)], OAc (21.1 q, 170.9 s,
OAc-2), for other signals, see Table 2; EIMS m/z 510 [M]þ (55), 495
(1), 450 (1), 388 (9), 346 (3), 277 (5), 234 (10), 105 (100), 95 (85);
HREIMS m/z 510.2248 [M]þ (calcd for C29H34O8, 510.2254).
other signals, see Table 2; 13C NMR (CDCl3)
d OFu-1 [109.3 (d, C-4),
118.6 (s, C-3), 143.3 (d, C-5), 147.1 (d, C-2), 160.8 (s, COOe)], OBz-9
[127.7 (2 ꢁ d, C-3, C-5), 130.0 (2 ꢁ d, s, C-2, C-6, C-1), 132.5 (d, C-
4), 165.4 (s, COOe)], OAc [(20.6 q, 170.4 s, OAc-2), (21.2 q, 171.0 s,
OAc-3)], for other signals, see Table 2; EIMS m/z 568 [M]þ (44), 553
(2), 446 (6), 354 (3), 311 (5), 232 (9), 157 (8), 105 (100), 95 (69);
HREIMS m/z 568.2306 [M]þ (calcd for C31H36O10, 568.2308).
4.3.15. (1S,4S,5R,7R,9S,10S)-9-Benzoyloxy-1-(3-furoyloxy)-4-
hydroxy-dihydro-
Colorless laquer; [
(log ) 230 (3.9) nm; IR nmax (film) 3543, 2930, 1720, 1307, 1278,
1160, 1092, 970, 760, 713 cmꢀ1; 1H NMR (CDCl3)
OFu-1 [6.34 (1H,
b-agarofuran (16)
25
a
]
þ69.7 (c 0.6, CHCl3); UV lmax (EtOH)
D
4.3.11. (1S,2R,4S,5R,7R,9S,10R)-2-Acetoxy-9-benzoyloxy-1-(3-
3
furoyloxy)-4-hydroxy-dihydro-
b
-agarofuran (12)
d
25
Colorless laquer; [
(log ) 231 (3.8) nm; IR nmax (film) 3539, 2929, 2855, 1740, 1714,
1277, 1241, 1156, 1012, 757, 714 cmꢀ1; 1H NMR (CDCl3)
1.86 (3H, s,
a]
þ65.1 (c 0.6, CHCl3); UV lmax (EtOH)
s, H-4), 7.25 (1H, s, H-5), 7.63 (1H, s, H-2)], OBz-9 [7.38 (2H, m, H-3,
D
3
H-5), 7.51 (1H, m, H-4), 7.91 (2H, d, J ¼ 7.7 Hz, H-2, H-6)], for other
d
signals, see Table 2; 13C NMR (CDCl3)
d OFu-1 [109.4 (d, C-4), 119.2
OAc-2), OFu-1 [6.35 (1H, s, H-4), 7.30 (1H, s, H-5), 7.67 (1H, s, H-2)],
OBz-9 [7.46 (2H, m, H-3, H-5), 7.57 (1H, m, H-4), 8.04 (2H, d,
(s, C-3), 143.2 (d, C-5), 147.1 (d, C-2), 162.1 (s, COOe)], OBz-9 [128.0
(2 ꢁ d, C-3, C-5), 129.7 (s, C-1), 129.8 (2 ꢁ d, C-2, C-6), 132.8 (d, C-4),

N.R. Perestelo et al. / European Journal of Medicinal Chemistry 46 (2011) 4915e4923
4923
165.6 (s, COOe)], for other signals, see Table 2; EIMS m/z 468 [M]þ
(5), 453 (27), 356 (10), 346 (9), 234 (14), 105 (100), 95 (58); HREIMS
m/z 468.2129 [M]þ (calcd for C27H32O7, 468.2148).
drug concentration that inhibit cell growth by 50%) for DNM and
VNB at each sesquiterpene concentration. The reversal index for
a given sesquiterpene at a given concentration is the ratio between
the IC50 for MDR cells without sesquiterpene and the IC50 with
sesquiterpene.
4.4. Biological assays
4.4.1. Chemicals
Acknowledgments
Vinblastine (VNB), verapamil, 3-(4,5-dimethyl thiazol-2-yl)-2,5-
diphenyltetrazolium bromide (MTT), sodium ortho-vanadate, and
colchicine were purchased from SigmaeAldrich (Madrid, Spain).
Daunomycin (DNM) was from Pfizer (Madrid, Spain).
This study was supported by the Agencia Canaria de Inves-
tigación,
Innovación
y
Sociedad
de
la
Información
(C200801000049), the Plan Andaluz de Investigación, Proyecto de
Excelencia P08-CTS-03625 and by FEDER Funds from the EU (SC),
and the Plan Andaluz de Investigación cod. BIO130 (FG). N.R.P.
thanks the Gobierno Autónomo de Canarias for the fellowship. We
also thank Dr. Ira Pastan (National Cancer Institute, NIH, Bethesda,
MD) for providing the NIH-3T3 and NIH-3T3 MDR-G185 cell lines.
We thank Dr J.T. Vázquez for carrying out the CD experiments.
Thanks are also due to Alfredo Tupayachi for collecting the plant
material.
4.4.2. Cell cultures
The parental drug-sensitive NIH-3T3 cell line and the one
transfected with human MDR1-G185 (MDR1 cells) were cultured in
DMEM (Invitrogen, Barcelona, Spain) with 10% heat-inactivated fetal
bovine serum (iFBS), with 2 mM
L-glutamine, 250 U/mL penicillin
G and 250 g/mL streptomycin, as previously described [17].
m
4.4.3. Modulation of DNM efflux by sesquiterpenes
The sesquiterpenes studied in the present work were previously
screened for their ability to block the Pgp-dependent DNM efflux
by flow cytometry analysis as previously described [17]. Briefly,
24 h before the experiment, MDR1 cells in logarithmic phase of
growth were seeded in 24-well plates at a density of 105 cells per
well. Afterward, cells were incubated for 30 min at 37 ꢂC in
Appendix. Supplementary data
Supplementary data associated with this article can be found in
the online version, at doi:10.1016/j.ejmech.2011.07.048.
References
DMEM þ 10% iFBS in the presence or absence of 10
mM sesquiter-
penes with 2 M DNM. Finally, cells were washed twice with ice-
m
[1] M. Ladanyi, C.W.P. Hogendoorn, J. Pathol. 223 (2011) 99e101.
[2] H.R. Mellor, R. Callaghan, Pharmacology 81 (2008) 275e300.
[3] Y. Li, H. Yuan, K. Yang, W. Xu, W. Tang, X. Li, Curr. Med. Chem. 17 (2010)
786e800.
[4] P.D.W. Eckford, F.J. Sharom, Chem. Rev. 109 (2009) 2989e3011.
[5] J. Molnár, H. Engi, J. Hohmann, P. Molnár, J. Deli, O. Wesolowska, K. Michalak,
Q. Wang, Curr. Top. Med. Chem. 10 (2010) 1757e1768.
[6] C.P. Reyes, F. Muñoz-Martínez, I.R. Torrecillas, C.R. Mendoza, F. Gamarro,
I.L. Bazzocchi, M.J. Núñez, L. Pardo, S. Castanys, M. Campillo, I.A. Jiménez,
J. Med. Chem. 50 (2007) 4808e4817.
[7] D. Torres-Romero, F. Muñoz-Martínez, I.A. Jiménez, S. Castanys, F. Gamarro,
I.L. Bazzocchi, Org. Biomol. Chem. 7 (2009) 5166e5172.
cold PBS, trypsinized and resuspended in 0.2 mL of ice-cold PBS
for immediate analysis. Fluorescence measurements of individual
cells were made with a Beckton Dickinson FacScan (BD European
HQ, Belgium). To calculate Ki for inhibition of DNM efflux, cells were
incubated with DNM as described above in the presence or absence
of different concentrations of sesquiterpenes. All intracellular
fluorescence values were converted to percentage inhibition of Pgp
normalized for MDR1 cells treated with 5 mM ortho-vanadate
(100% inhibition of Pgp), plotted and fitted to the Hill equation
for allosteric interactions, using SigmaPlot 2000 software:
i ¼ (Imax ꢁ Sn)/(Ki þ Sn); where i is the percentage inhibition of DNM
efflux at a given sesquiterpene concentration, Imax is the maximal
percentage inhibition, S is the concentration of sesquiterpene, and
n is the Hill coefficient.
[8] F. Cortés-Selva, I.A. Jiménez, F. Muñoz-Martínez, M. Campillo, I.L. Bazzocchi,
L. Pardo, A.G. Ravelo, S. Castanys, F. Gamarro, Curr. Pharm. Des. 11 (2005)
3125e3139.
[9] A.G. González, B.M. Tincusi, I.L. Bazzocchi, H. Tokuda, H. Nishino,
T. Konoshima, I.A. Jiménez, A.G. Ravelo, Bioorg. Med. Chem.
1773e1778.
[10] F. Muñoz-Martínez, C.R. Mendoza, I.L. Bazzocchi, S. Castanys, I.A. Jiménez,
F. Gamarro, J. Med. Chem. 48 (2005) 4266e4275.
8 (2000)
[11] N. Harada, K. Nakanishi, Circular Dichroism Spectroscopy Exciton Coupling in
Organic Stereochemistry. University Science Books, Mill Valley, CA, 1983.
[12] PC Model from Serena Software, P.O. 3076, Bloomington, IN 47402-3076,
version 9.2 with MMX force field.
[13] N.R. Perestelo, I.A. Jiménez, H. Tokuda, H. Hayashi, I.L. Bazzocchi, J. Nat. Prod.
73 (2010) 127e132.
[14] S.M. Kupchan, R.M. Smith, J. Org. Chem. 42 (1977) 115e118.
[15] C.R. Smith, R.W. Miller, D. Weisleder, W.K. Rohwedder, N. Eickman, J. Clardy,
J. Org. Chem. 41 (1976) 3264e3269.
[16] A.G. González, M.P. Núñez, I.A. Jiménez, A.G. Ravelo, J.A. Sazatornil,
J.T. Vázquez, I.L. Bazzocchi, E.Q. Morales, O.M. Muñoz, J. Chem. Soc., Perkin
Trans. 1 (1992) 1437e1441.
4.4.4. Reversion of Pgp-dependent DNM and VNB resistance
The doseeresponse curves of drug-sensitive and MDR1 cells
were determined by the MTT colorimetric assay [18] as described
previously [14]. To assess the chemosensitizing effect of the best
sesquiterpenes blocking the DNM efflux activity of Pgp
(compounds 1, 2, 4, 6, 8, 9, 11 and 14), both drug-sensitive and
MDR1 cells were exposed to increasing concentrations of DNM or
VNB (up to 500 ng/mL) in the presence/absence of fixed concen-
trations (1, 3 and 10 mM) of each sesquiterpene. Doseeresponse
[17] F. Muñoz-Martínez, P. Lu, F. Cortés-Selva, J.M. Pérez-Victoria, I.A. Jiménez,
A.G. Ravelo, F.J. Sharom, F. Gamarro, S. Castanys, Cancer Res. 64 (2004)
7130e7138.
curves were generated by non-linear regression (using SigmaPlot
2000 for Windows, SPSS Inc.) of the data points to a four-param-
eter logistic curve, in order to determine the IC50 (defined as the
[18] T. Mosmann, J. Immunol. Methods 65 (1983) 55e63.