Configurational analysis of cubebins and bicubebin from Aristolochia lagesiana and Aristolochia pubescens

Article abstract of DOI:10.1016/j.phytochem.2006.01.019

(8S,8′R,9S)-, (8R,8′R,9R)-, and (8R,8′R,9S)-cubebins, together with (8R,8′R,8″R,8?R,9R,9″S)-bicubebin, were isolated from Aristolochia lagesiana and Aristolochia pubescens. Their structures were determined by spectroscopic methods, including ¹H and ¹³C NMR spectroscopy at low temperatures, and by chemical transformations.

Full text of DOI:10.1016/j.phytochem.2006.01.019

PHYTOCHEMISTRY
Phytochemistry 67 (2006) 735–742
www.elsevier.com/locate/phytochem
Configurational analysis of cubebins and bicubebin from
Aristolochia lagesiana and Aristolochia pubescens
a
b
a,
*
Inara C. de Pascoli , Isabele R. Nascimento , Lucia M.X. Lopes
a
Instituto de Quı´mica, Universidade Estadual Paulista, UNESP, C.P. 355, 14801-970, Araraquara, Sa˜o Paulo, Brazil
ˆ
Unidade Diferenciada de Sorocaba/Ipero´, Universidade Estadual Paulista, UNESP, Av. Tres de Marc¸o, 511, 18087-180, Sorocaba, Sa˜o Paulo, Brazil
b
Received 29 September 2005; received in revised form 12 December 2005
Available online 28 February 2006
Abstract
(8S,80R,9S)-, (8R,80R,9R)-, and (8R,80R,9S)-cubebins, together with (8R,80R,800R,8000R,9R,900S)-bicubebin, were isolated from Aristol-
1
ochia lagesiana and Aristolochia pubescens. Their structures were determined by spectroscopic methods, including H and 13C NMR
spectroscopy at low temperatures, and by chemical transformations.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Keywords: Aristolochia pubescens; Aristolochia lagesiana; Aristolochiaceae; Lignan; Cubebin; Butyrolactol; Bicubebin
1. Introduction
Generally, cubebin, whether obtained from a natural
source or synthesized, has been identified as a mixture of
two 8,80-trans diastereomers at C-9 (Koul et al., 1983; Blu-
menthal et al., 1997; Wei-Ming et al., 1987; Badheka et al.,
1987; Chatterjee et al., 1968; Rehnberg and Magnusson,
1988, 1990; Ward, 1990; Koul et al., 1988). However, the
configuration of this anomeric center is still unclear, as
with other butyrolactol lignans (Brown and Daugan,
1989; Li et al., 2003; Go¨zler et al., 1996; Bhandari et al.,
Previous studies on Aristolochia pubescens led to the iso-
lation of 7 lignans, 5 neolignans, 2 sesquiterpenes, 2 diter-
penes,
8 aristolochic acids and derivatives, and 3
aristolactams (Nascimento and Lopes, 1999, 2000, 2003;
Nascimento et al., 2000). As part of our continuing studies
on the Aristolochiaceae family, in this paper we report the
isolation and structural elucidation of 8,80-cis-cubebin (1),
8,80-trans-cubebins (2a and 2b) and 8,80-trans-bicubebin
A (3) from A. pubescens and Aristolochia lagesiana.
1998; Rucker and Langmann, 1978; Barrero et al., 1994;
¨
Cambie et al., 1985).
Cubebin is a dibenzylbutyrolactone lignan type, which is
known to reduce larval viability in Anticarsia gemmatalis to
give rise to malformed adult insects (Nascimento et al.,
2004), to exhibit antifeedant activity for insect pests
(Harmatha and Dinan, 2003), and to show moderate
anti-inflammatory and analgesic activities (Silva et al.,
2005). Cubebin has been isolated from several species in
various families, such as Aristolochiaceae, Myristicaceae,
Rutaceae, and Piperaceae (Lopes et al., 2001; Bastos
et al., 2001; Koul et al., 1983; Blumenthal et al., 1997).
2. Results and discussion
Complex mixtures of cubebins were obtained from an
ethanol extract of the tubercula of A. pubescens and an ace-
tone extract from the roots of A. lagesiana. These mixtures
were subjected to preparative TLC followed by semi-pre-
parative HPLC to give 1, 2 (2a + 2b), and 3.
The 1H and 13C NMR, UV, and IR data for compound
1 were very similar to those reported for cubebin (Koul
et al., 1983; Blumenthal et al., 1997; Wei-Ming et al.,
1987; Badheka et al., 1987). Compound 1 was suggested
*
Corresponding author. Tel.: +55 16 3301 6663; fax: +55 16 3322 7932.
E-mail address: lopesxl@iq.unesp.br (L.M.X. Lopes).
0031-9422/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phytochem.2006.01.019

736
I.C. de Pascoli et al. / Phytochemistry 67 (2006) 735–742
OH
OR
O
OR
O
H
H
H
H
H
H
2
5
7
9
O
O
O
O
O
O
O
6
7'
9'
2'
6'
5'
O
O
O
O
O
O
R
H
R
H
1
2a
2c
2b
2d
CH3
CH3
O
2
5
H
H
H
9''
2''
5''
7
7''
9
O
O
O
O
6
7'
7'''
O
O
9'
9'''
H
2'
2'''
6'
5'
6'''
5'''
O
O
O
O
3
to be cubebin since it displayed quasi-molecular ions at m/z
379 [M + Na]+ and m/z 357 [M + H]+, which were consis-
butyrolactol ring. Analysis of 1H–1H COSY, gHMQC,
and gHMBC data enabled the precise assignments of all
hydrogens and carbons in the basic structure.
tent with the molecular formula C20H20O6. The 1H and 13
C
NMR spectra showed signals for two methylenedioxyben-
zyl, two methynic (dC: 52.3, 45.7; d2H: 2.02), anomeric
(dC: 104.3, dH: 5.04), and carbinolic (dC: 72.1, dH: 3.83,
3.32) groups (Tables 1 and 2). These latter comprise the
Compound 2 was also suggested to be a cubebin based
on the similarity of its ESI-MS, UV, IR, 1H and 13C
NMR spectra to those of 1. The main differences between
1
the H and 13C NMR spectra of 2 and 1 were signals for
Table 1
Table 2
13C NMR spectroscopic data for compounds 1, 2a and 2b (30 ꢁC, CDCl3,
1H NMR spectroscopic data for compounds 1, 2a and 2b (30 ꢁC, CDCl3,
500 MHz, J in Hz)
126 MHz)
C
1
2a
2b
H
1
2a
2b
1
2
3
4
5
6
7
8
133.5
109.2
147.6
145.8
108.1a
121.8
38.4
133.3
108.9
147.6
145.7
108.0
121.4
38.4
133.8
108.9
147.7
145.7
108.2a
121.3
33.6
2
5
6
7
6.49 d (1.5)
6.61 d (7.5)
6.47 dd (7.5, 1.5)
2.33 dd (14.0, 8,0), 2.37 m, 2.60 dd
2.56 dd (14.0, 7.5)
2.02 m
5.04 d (1.5)
6.42 d (1.5)
6.61 d (7.5)
6.41 dd (7.5, 1.5)
2.44 m
6.45 d (1.5)
6.61 d (8.5)
6.44 dd (8.5, 1.5)
6.56 d (1.5)
6.66 d (8.0)
6.52 dd (8.0, 1.5)
2.37 m, 2.52 m
(13.5, 7.5)
2.07 m
8
9
20
50
60
70
1.93 m
5.15 d (1.5)
6.51 d (1.5)
6.63 d (7.5)
6.49 dd (7.5, 1.5)
2.50 m, 2.52 m
5.15 d (4.5)
6.67 d (2.0)
6.66 d (8.0)
6.62 dd (8.0, 2.0)
2.50 m, 2.70 dd
(14.0, 10.0)
2.37 m
52.3
53.0
51.9
98.8
9
104.3
134.3
108.9
147.6
145.8
108.0a
121.4
39.0
103.3
134.1
109.1
147.5
145.9
108.0
121.7
39.2
10
134.5
109.3
147.5
145.9
108.1a
121.6
38.8
20
30
40
80
90
2.02 m
3.32 t (8.5),
3.83 dd (8.5, 7.0)
2.07 m
3.73 dd (8.5, 8.0),
3.93 dd (8.5, 7.0)
4 · 5.85 d (W1/2 1.5) 4 · 5.85 d
(W1/2 1.5)
50
3.50 dd (8.5, 7.5),
4.03 t (8.5)
60
70
OCH2O 5.85 d (W1/2 1.5),
5.84 d (W1/2 1.5),
80
45.7
72.1
45.8
72.1
42.9
72.5
90
5.82 d (W1/2 1.5),
2 · OCH2O
100.8, 100.7
2 · 100.8
2 · 100.8
5.81 d (W1/2 1.5)
2.02 br s
OH
1.72 br s
2.86 br s
a
Assignments may be interchangeable within the same column.

I.C. de Pascoli et al. / Phytochemistry 67 (2006) 735–742
737
two sets of: methylenedioxybenzyl, two methynic, ano-
meric, and carbinolic groups (Tables 1 and 2). Based on
integration of the signals observed in the H NMR spec-
gens of each species of compound 2 (Tables 1–3). Further-
more, the simulated 1H NMR spectra by using the
FOMSC3 program (FOMSC3, 2004) for first order cou-
pling hydrogens of 2a and 2b compounds were in total
agreement with the experimental spectra.
1D-gNOESY experiments with compound 1 showed
interactions between the carbinolic hydrogen H-90b1 (at
lower frequency, d 3.32) and H-90b2 (at higher frequency,
d 3.83), as well as between H-90b1 and H-20 (d 6.42), H-
60 (d 6.41), 2H-70 (d 2.44), H-9 (d 5.04), and H-8 and/or
H-80 (d 2.02), whereas H-90b2 was spatially correlated with
H-90b1 and with the methynic(s) hydrogen(s) at d 2.02. H-9
did not show any correlation with the methynic hydrogens.
These data suggested a syn orientation of H-90b1 and H-9,
as well as a syn orientation of H-90b2, H-80, and H-8
(Fig. 1).
1D-gNOESY experiments with compound 2 showed
spatial interactions between the carbinolic hydrogens H-
90b1 (2a: d 3.73, 2b: d 3.50) and H-90b2 (2a: d 3.93, 2b: d
4.03), and between H-90b1 and H-20 (2a: d 6.51, 2b: d
6.67), H-60 (2a: d 6.49, 2b: d 6.62), and 2H-70 (2a: d 2.50,
2.52, 2b: d 2.50, 2.70). Furthermore, the anomeric hydro-
gen of 2b showed a strong correlation with H-8, and the
latter gave a strong correlation with H-90b1, which sug-
gested that they were in a syn orientation, with the hydro-
xyl group cis to the methylenedioxybenzyl group at C-8
(Fig. 1). This suggestion is consistent with the upfield
chemical shifts of C-7 (d 33.6) due to the c effect of the
hydroxyl group. Correlations between H-9 and H-8, 2H-
7, and H-90b2 were observed for 2a. These correlations sug-
gested that H-9 was in an ‘‘equatorial’’ position, and in a
syn orientation with H-90b2 (Fig. 1).
1
trum and on the intensity of the signals in the 13C NMR
spectrum (obtained at 30 ꢁC), as well as on the observed
correlations between hydrogens by 1H–1H COSY, and
between hydrogens and carbons by gHMQC and gHMBC,
it was determined that compound 2 consisted of two cube-
bins in a 3:2 proportion (2a:2b). Interestingly, the anomeric
hydrogens of 2a and 2b showed the same chemical shift at d
5.15, but different coupling constants (d, J = 1.5 and
4.5 Hz) at 30 ꢁC, whereas the anomeric carbons showed
signals at d 103.3 and 98.8 for 2a and 2b, respectively.
1H–1H COSY experiments showed strong correlations
0
between all four carbinolic hydrogens (dH-9 : 4.03, 3.93,
3.73, and 3.50). These data suggested two hypotheses: there
are two epimers (at C-9 and/or C-8) in equilibrium, or the
cubebin species exist in conformational equilibrium.
Lactols can undergo ring-opening and -closing in solu-
tion. This phenomenon has been observed, for example,
in carbohydrates (Pihlaja and Kleinpeter, 1994), trityl-
oxymethyl butyrolactols (Li et al., 2004), and butyrolactol
lignans (Wei-Ming et al., 1987). Therefore, dynamic molec-
ular processes were investigated using NMR techniques.
Under appropriate NMR conditions (temperatures above
26 ꢁC, and long periods of storage in CDCl3 solution),
the resulting interconversion, 2a ꢀ 2b, was manifested
through the presence of additional signals at dH: 8.74 and
dC: 203.0, which are characteristic of an aldehyde group,
among others signals. The effects of temperature on the
chemical shifts and the correlations between hydrogens of
1
compounds 1 and 2 were also investigated. The H NMR
experiments, including 1D-gNOESY and 1H–1H COSY,
were performed from 30, 26, 20 to 10, 0, ꢀ10 ꢀ20, ꢀ30,
ꢀ40, and ꢀ50 ꢁC, while keeping other conditions (solvent,
concentration, pulse sequence, and recording conditions)
constant.
While the change in temperature did not significantly
affect the hydrogen or carbon chemical shifts of both com-
pounds, this did aid in the unequivocal assignment of the
structures, since some regions of the spectra such as the
aromatic and anomeric regions were less crowded at lower
temperature. Moreover, these Dd allowed to determine the
multiplicities and chemical shifts for the anomeric hydro-
The substituents on the lactol rings could favor the pos-
sibility that these molecules exist as a single conformation
in solution (Lightner and Gurst, 2000; Lambert et al.,
1974). To determine the most stable conformation, the rel-
ative configuration, and the absolute configuration for
these compounds, they were transformed into hinokinins,
the absolute configurations of which have been well estab-
lished (Lopes et al., 1983; Batterbee et al., 1969; Burden
et al., 1969). While no oxidation product was obtained
from compound 1, only one oxidation product was
obtained from compound 2 under the same experimental
conditions. The CD curve and aD (ꢀ28.9ꢁ) for the oxidized
Table 3
Selected 1H and 13C NMR spectroscopic data for compounds 1–3 at ꢀ10 ꢁC
H and C
1
2a
2b
3
a
b
a
b
a
b
a
b
dH
dC
dH
dC
dH
dC
dH
dC
9
5.06 d (1.0)
3.28 t (8.5),
3.85 dd (8.5, 7.0)
103.9
72.1
5.18 d (1.0)
3.74 dd (8.5, 8.0),
3.95 dd (8.5, 7.0)
103.1
72.0
5.17 d (4.0)
3.52 dd (8.5, 7.5),
4.01 t (8.5)
98.6
72.6
4.76 d (1.5)
3.63 t (8.0),
3.92 dd (8.0, 7.0)
4.80 d (4.5)
3.56 dd (8.5, 7.0),
4.03 dd (8.5, 8.0)
108.0
72.5
90
900
9000
104.4
72.5
a
Recorded in CDCl3, 500 MHz, J in Hz.
Recorded in CDCl3, 126 MHz.
b

738
I.C. de Pascoli et al. / Phytochemistry 67 (2006) 735–742
OH
OH
H
H
H
H
H
H
H
H
H
OH
O
O
O
O
O
O
O
H
O
O
H
H
H
O
H
H
O
O
O
O
O
1
2a
2b
Fig. 1. Selected nOe interactions observed for compounds 1, 2a and 2b.
product allowed it to be identified as (ꢀ)-hinokinin, which
has an 8R,80R configuration. Therefore, the anomers 2a
and 2b should have the same configuration at the corre-
sponding stereogenic centers (8R,80R).
suggested that the Cotton effects observed at 290 and
230 nm depended on the stereogenic center C-80 (Fig. 3).
Based on the above spectroscopic findings, the cubebins
were characterized as (8S,80R,9S)-cubebin (1), (8R,80R,9R)-
cubebin (2a), and (8R,80R,9S)-cubebin (2b), which is consis-
tent with most of the configurations previously established
for cubebin 2a (a-cubebin or epi-cubebin) and 2b (b-cubebin
or cubebin) (Wei-Ming et al., 1987; Badheka et al., 1987;
Even though the problems posed to establish the relative
and absolute configurations of saturated hetero five-mem-
bered rings are more complex than those for six-membered
rings (Pihlaja and Kleinpeter, 1994; Lightner and Gurst,
2000; Lambert et al., 1974), the coupling constants between
the carbinolic hydrogens (H-90) and between these and H-
80 were the same magnitude (J @ 8 Hz) for 1, 2a, and 2b,
which suggested that they have the same spatial arrange-
ments, and the torsion angles that involve C-80 and C-90
are very similar in all three molecules. Considering the
two possible conformations for the lactol ring in the three
cubebins, half-chair and envelope, and all data previously
discussed, the conformations that better represented these
molecules should be 4 for 1, 5 and 6 for 2a, and 7 and 8
for 2b, in which the carbinolic hydrogens are bisected
(Fig. 2). The CD curves of 1 and 2 were very similar, which
Rucker et al., 1981). Furthermore, two acetals were
¨
obtained from the anomers 2a and 2b, and they were,
respectively, identified as a-methylcubebin (2c) and b-meth-
ylcubebin (2d) (Blumenthal et al., 1997; Rucker et al., 1981).
¨
The 1H and 13C NMR spectra and gHMQC experiments
showed a total of 40 carbons and 38 hydrogens for lignan
3. The ESI-MS spectra displayed quasi-molecular ions at
m/z 717 [M + Na]+ and 695 [M + H]+, which were consis-
tent with the molecular formula C40H38O11 and the results
of an elemental analysis. This formula corresponded to
dehydration of two cubebin units. The NMR spectra were
also very similar to those of cubebins discussed previously
a
H
a
a
Pi
Pi
H
Pi
b2
b2
H
e
e H
e
H
H
H
H
e
e,
O
H
, H
H
e
O
O
OH
H
b
b
H
e
Pi
a,
Pi
a
a,
H
OH
e
e
b1
b1
4
OH
a
H
a
a
Pi
6
5
a
OH
a
OH
a
H
a,
H
Pi
b2
H
e
e
H
e
H
e
O
Pi
Pi
e,
H
a
H
e
b2
H
b1
O
H
H
Pi
b1
a
7
8
Fig. 2. Conformations for 1 (4), 2a (5 and 6) and 2b (7 and 8), showing axial (a), pseudoaxial (a0), equatorial (e), pseudoequatorial (e0), and bisected (b)
exocyclic positions.

I.C. de Pascoli et al. / Phytochemistry 67 (2006) 735–742
739
Table 4
2
1
13C and 1H NMR spectroscopic data for compound 3
C and
H
d
13Ca,b,c
d
1Hc,d,e
1D-gNOESY
0
-1
-2
-3
-4
-5
-6
1
2
3
4
5
6
7
133.4
109.0
147.7
145.8
108.1
121.7
38.5
6.46 d (1.5)
6.60 d (8.0)
1
6.40 dd (8.0, 1.5)
2.32 dd (13.5, 8.0),
2.47 dd (13.5, 7.5)
2.11 dddd (8.0, 7.5, 6.0, 2.0)
4.76 d (2.0)
2a+2b
3
8
9
52.7
108.6
134.6
109.0
147.6
146.0
108.2
121.4b
39.4
210
230
250
270
290
310
330
H-7, H-8, H-900
λ (nm)
10
20
30
40
50
60
70
80
90
6.43 d (1.5)
Fig. 3. CD curves for lignans 1–3.
6.58 d (8.0)
6.42 dd (8.0, 1.5)
2.51 d (8.0, 2H)
2.04 dddt (8.5, 7.5, 6.0, 8.0)
3.92 t (8.5), 3.62 dd (8.5, 7.5) H-80, H-90, H-20,
(Table 4). Indeed, they showed signals for two sets of: two
methylenedioxybenzyl groups, one anomeric group, one
carbinolic group, and two methynic groups, as previously
observed for 2. Furthermore, a detailed analysis of
gHMBC, 1D-gNOESY, and 1H–1H COSY experiments
allowed us to establish two cubebin units (Table 4 and
Fig. 3). 1H selective irradiation NMR experiments also
aided in determining the J values and chemical shifts
ascribed for first order coupling hydrogens, which were fur-
ther confirmed by spectral simulations using the FOMSC3
program (FOMSC3, 2004). The main differences between
these spectra from those of 2 were that the proportion of
the cubebins units was 1:1 instead of 3:2, and the absence
of signals for hydroxyl hydrogens. Furthermore, 1H–1H
COSY experiments did not show any correlations between
the carbinolic hydrogens of one unit with those of the other
unit (2H-90 with 2H-9000), even at low temperatures. How-
ever, nOes were observed between the anomeric hydrogen
from each unit. Moreover, gHMBC experiments showed
correlations between the anomeric carbon from one unit
with anomeric hydrogen from the other unit (dC: 108.6 with
dH: 4.81, and dC: 104.2 with dH: 4.76), which suggested that
the units were linked through C-9 ! O ! C-900 (Fig. 4).
This suggestion was further supported by analysis of the
chemical shifts of C-9 and C-900. Considering that the
replacement of a hydroxyl group in lactols by an ether
group caused Dd @ +6 ppm for the anomeric carbon (Blu-
menthal et al., 1997), as observed for 2a and 2c, as well
as for 2b and 2d, it could be deduced that the unit contain-
ing C-9 should be similar to 2a and that containing C-900
should be similar to 2b. This deduction was corroborated
by a comparison of all NMR spectroscopic data obtained
from 3 with those obtained from 2, including those
obtained by 1D-gNOESY experiments (Figs. 1 and 4).
Based on an analysis of all these data, including the cou-
pling constants observed for anomeric hydrogens, as well
as the spatial interactions between the hydrogens and the
substituents on the lactol ring, as previously described for
1, 2a, and 2b, the most stable conformations and the rela-
tive configurations could be established as being 5 and/or 6
for unity I and 7 and/or 8 for unity II for 3. Taking into
45.4
72.4
H-70, H-8, H-90
100
200
300
400
500
600
700
134.1
109.1
147.7
145.7
108.1
121.3b
33.9
6.51 d (2.0)
6.60 d (8.5)
6.44 dd (8.5, 2.0)
2.49 dd (13.5, 8.5),
2.43 dd (13.5, 6.5)
1.95 tdd (8.5, 6.5, 5.0)
4.81 d (5.0)
6.54 d (2.0)
800
900
51.7
104.2
134.2
109.0
147.6
145.8
108.2
121.5
39.4
H-800, H-9
1000
2000
3000
4000
5000
6000
7000
6.64 d (8.0)
6.50 dd (8.0, 2.0)
2.36 dd (13.5, 9.5),
2.60 dd (13.5, 5.0)
2.27 dddt (9.5, 7.0, 5.0, 8.5)
8000
9000
43.6
72.4
3.51 dd (8.5, 7.0), 4.00 t (8.5) H-2000, H-6000, H-7000,
H-800, H-9000, H-8000,
H-9000
OCH2O 4 · 100.8 5.77 d (W1/2 1.5),
5.82 d (W1/2 1.5),
6 · 5.85 br s
a
Recorded in CDCl3, 126 MHz (30 ꢁC).
Assignments may be interchangeable within the same column.
The 1H and 13C NMR data were assigned with the assistance of
b
c
gHMQC and gHMBC experiments.
d
Recorded in CDCl3, 500 MHz, J in Hz (30 ꢁC).
Multiplicities were determined with the assistance of 1H–1H COSY
e
and TOCSY, 500 MHz.
account that compound 3 was isolated together with 2
and that they showed very similar CD curves, the structure
of 3 was determined to be (8R,80R,800R,8000R,9R,900S)-cube-
bin (named bicubebin A). Moreover, bicubebin A did not
give acetals or hinokinins when subjected, respectively, to
etherification and oxidation under the same experimental
conditions as compound 2, which is consistent with the
proposed structure.

740
I.C. de Pascoli et al. / Phytochemistry 67 (2006) 735–742
on a JASCO J-720 spectrometer. HPLC analyses were car-
H
H
H
O
H
H
H
H
ried out using a Shimadzu liquid chromatograph 10Avp
equipped with a UV–Vis detector. Columns were RP-18
(Shimadzu, C18, 250 · 4.6 mm for analytical analysis and
250 · 20 mm for semi-preparative analysis), and chromato-
O
O
O
O
O
O
H
H
H
grams were acquired at 254 nm. TLC: Silica gel 60 PF254
.
O
O
4.2. Plant material
O
O
I
II
Plant materials were collected in Ituiutaba, SP, Brazil,
and identified as A. pubescens Will. ex Duch., and A. lage-
siana Ule. var. intermedia Hoehne by Dr. Condorcet Ara-
nha and Dr. Lindolpho Cappellari Ju´nior. Vouchers
specimens of A. lagesiana (ESA 88885 02/13/2003) and
A. pubescens (ESA 88882 01/20/1998) were deposited at
the herbarium of the Escola Superior de Agricultura, Luiz
de Queiroz (ESALQ), Piracicaba, SP, Brazil. The materials
were separated by plant parts, dried (ꢁ45 ꢁC) and ground.
3
O
H
H
H
H
O
O
O
O
O
O
O
O
4.3. Extraction and isolation
O
O
3
Ground roots (245.6 g) of A. lagesiana and tubercula
(728.1 g) of A. pubescens were extracted exhaustively at
room temperature with hexane, Me2CO and EtOH, succes-
sively, and the extracts were then individually concen-
trated. Some of the crude acetone extract of the roots of
A. lagesiana (2.3 g) was washed with hexane, Me2CO and
MeOH, successively. The Me2CO fraction (0.5 g) was sub-
jected to preparative TLC (hexane/EtOAc, 7:3) to give a
mixture of cubebins (1 + 2a + 2b + 3, 38.2 mg). This mix-
ture was subjected to preparative TLC (hexane–EtOAc
7:3) followed by semi-preparative HPLC (C-18, MeOH/
H2O 4:1) to give 1 (18.4 mg), 2a + 2b (4.8 mg), and 3
(7.7 mg). The crude ethanol extract of the tubercula of A.
pubescens (13.4 g) was washed with hot EtOH as described
previously (Nascimento and Lopes, 2003). The solution
was concentrated and subjected to partitioning between
MeOH/H2O/CHCl3 (1:1:1). After concentration, the
organic phase was subjected to CC (silica gel, CHCl3/
MeOH gradient) to give 33 fractions (Nascimento and
Lopes, 2003). Fraction 6 (23 mg) after preparative TLC
(hexane–EtOAc 7:3) gave three sub-fractions that by
semi-preparative HPLC (C-18, MeOH/H2O 4:1) gave 1
(8.4 mg), 2a + 2b (2.8 mg), and 3 (3.9 mg).
Fig. 4. (a) Selected nOe interactions (M), (b) selected gHMBC correla-
tions (!) for compound 3.
3. Conclusion
The ring-opening of cubebins 2 in CDCl3 solution, lead-
ing to inversion at the stereogenic center where the hydro-
xyl group is located, was confirmed by detection of the
intermediary aldehyde by NMR. The equilibrium between
the anomers 2a and 2b from 30 to ꢀ50 ꢁC was on the same
time-scale as the observation of signals by NMR spectros-
copy. Even though it was not possible to establish whether
both anomers (2a or 2b) are natural compounds, this paper
reports their absolute configurations. Furthermore, an
8,80-cis-cubebin and a bicubebin A were described for the
first time.
4. Experimental
4.1. General
The 1D (1H, 13C, DEPT, and gNOESY) and 2D (1H–1H
gCOSY, gHMQC, gHMBC, and gTOCSY) NMR experi-
ments were recorded on a Varian INOVA 500 spectrometer
(11.7 T) at 500 MHz (1H) and 126 MHz (13C), using the
solvents as an internal standard. Mass spectra (ESI-MS)
were obtained on a Fisons Platform II, and flow injection
into the electrospray source was used. IR spectra were
obtained on a Perkin–Elmer 1600 FT-IR spectrometer
using KBr discs. UV absorptions were measured on a Per-
kin–Elmer UV–Vis Lambda 14P spectrophotometer. Opti-
cal rotations were measured on a Polamat A Carl Zeiss
Jena polarimeter. Circular dichroism spectra were recorded
4.4. (8S,80R,9S)-cubebin (1)
26
White crystal; ½aꢂ ꢀ102.3 (c 0.64, CHCl3); CD (c 0.1,
D
CH3OH): [h]223 ꢀ9569, [h]230 ꢀ12,769, [h]254 ꢀ455, [h]270
ꢀ4124, [h]292 ꢀ15,145, [h]306 ꢀ908. UV and IR data agree
with those reported in the literature (Koul et al., 1983; Blu-
menthal et al., 1997; Wei-Ming et al., 1987; Badheka et al.,
1
1987). For H and 13C NMR spectra, see Tables 1 and 2;
positive ESI-MS (probe) 70 eV, m/z (rel. int.): 379
[M + Na]+ (100), 357 [M + H]+ (72). Found: C, 67.3; H,
5.9. C20H20O6 requires: C, 67.4; H, 5.7%.

I.C. de Pascoli et al. / Phytochemistry 67 (2006) 735–742
741
4.5. (8R,80R,9R)- and (8R,80R,9S)-cubebin (2a + 2b)
Acknowledgements
26
D
`
White crystal; ½aꢂ ꢀ53.8 (c 2.58, CHCl3); CD (c 0.1,
The authors thank the Fundac¸ao de Amparo a Pesquisa
˜
CH3OH): [h]219 ꢀ3476, [h]229 ꢀ7267, [h]254 +178, [h]269
ꢀ2939, [h]272 ꢀ2498, [h]290 ꢀ8157, [h]304 ꢀ526. UV and
IR data agree with those reported in the literature (Koul
et al., 1983; Blumenthal et al., 1997; Wei-Ming et al.,
do Estado de Sao Paulo (FAPESP) for financial support
˜
and Coordenac¸ao de Aperfeic¸oamento de Pessoal de Nıvel
˜
Superior (CAPES) for a fellowship to I.C. de Pascoli. We
thank Dr. Leila Beltramini (USP, Sao Carlos, Brazil)
˜
for measurements of CD spectra, Dr. Condorcet Aranha
and Dr. Lindolpho Cappellari Ju´nior for botanical
identification.
´
1987; Badheka et al., 1987). For H and 13C NMR spec-
1
tra, see Tables 1 and 2; positive ESI-MS (probe) 70 eV,
m/z (rel. int.): 379 [M + Na]+ (100), 357 [M + H]+ (70).
Found: C, 67.4; H, 5.8. C20H20O6 requires: C, 67.4; H,
5.7%.
References
4.6. (8R,80R,9R)- and (8R,80R,9S)-methylcubebin
(2c + 2d)
Badheka, L.P., Prabhu, B.R., Mulchandani, N.B., 1987. Lignans of Piper
cubeba. Phytochemistry 26, 2033–2036.
´
To dry DMSO (400 lL) was added KOH (11.3 mg).
After stirring for 20 min, a solution of 2a + 2b (4.3 mg,
0.012 mmol) in DMSO (600 lL) was added, followed
immediately by 10 lL of CH3I (0.032 mmol) at room
temperature. Stirring was continued for 1 h, and the mix-
ture was then poured into H2O (10 mL) and extracted
with CH2Cl2 (3 · 10 mL). The combined organic phases
were washed with H2O (10 mL), dried (Na2SO4) and
concentrated (Johnstone and Rose, 1979). Purification
of the product by preparative TLC (hexane/EtOAc,
7:3) yielded 2c + 2d (1.8 mg, 0.005 mmol). 1H and 13C
NMR, UV and IR data agree with those reported in
Barrero, A.F., Ha¨ıdour, A., Dorado, M.M., Gravalos, D.G., Quesada,
T.G., 1994. Lignans from the wood of Abies pinsapo. J. Nat. Prod. 57,
713–719.
Bastos, J.K., Carvalho, J.C.T., Souza, G.H.B. de, Pedrazzi, A.H.P., Sarti,
S.J., 2001. Anti-inflammatory activity of cubebin, a lignan from the
leaves of Zanthoxyllum naranjillo Griseb. J. Ethnopharm. 75, 279–282.
Batterbee, J.E., Burden, R.S., Crombie, L., Whiting, D.A., 1969. Chemistry
and synthesis of the lignan (ꢀ)-cubebin. J. Chem. Soc. (C), 2470–2477.
Bhandari, S.P.S., Babu, U.V., Garg, H.S., 1998. A lignan from Piper
chaba stems. Phytochemistry 47, 1435–1436.
Blumenthal, E.E. de A., Silva, M.S. da, Yoshida, M., 1997. Lignoids,
flavonoids and polyketides of Virola surinamensis. Phytochemistry 46,
745–749.
Brown, E., Daugan, A., 1989. Lignanes. 10. Preparationdes (R) et (S)-(ꢀ)-b-
piperonyl et b-veratryl-c-butyrolactones et leur utilization dans la
synthese totale de lignanes optiquement actifs. Tetrahedron 45, 141–154.
Burden, R.S., Crombie, L., Whiting, D.A., 1969. The extractives of
Heliopsis scabra: constitution of two new lignans. J. Chem. Soc. (C),
693–701.
the literature (Blumenthal et al., 1997; Rucker et al.,
¨
1981).
4.7. (8R,80R,800R,8000R,9R,900S)-bicubebin A (3)
Cambie, R.C., Clark, G.R., Craw, P.A., Jones, T.C., Rutledge, P.S.,
Woodgate, P.D., 1985. Chemistry of the Podocarpaceae. LXIX.
Further lignans from the wood of Dacrydium intermedium. Aust. J.
Chem. 38, 1631–1645.
Chatterjee, A., Basa, S.C., Ray, A.B., 1968. Spectral properties of cubebin.
J. Indian Chem. Soc. 45, 723–725.
26
D
White amorphous solid; ½aꢂ ꢀ54.8 (c 0.36, CHCl3); CD
(c 0.1, CH3OH): [h]220 ꢀ9508, [h]231 ꢀ16,260, [h]253 + 91,
[h]270 ꢀ6742, [h]273 ꢀ6104, [h]290 ꢀ20,587, [h]306 ꢀ451.
For H and 13C NMR spectra, see Table 4; positive ESI-
1
MS (probe) 70 eV, m/z (rel. int.): 717 [M + Na]+ (78),
695 [M + H]+ (100). Found: C, 68.9; H, 5.8. C40H38O11
requires: C, 69.2; H, 5.5%.
FOMSC3 (First Order Multiplet Simulator/Checker), 2004. Available
from: <http://artemis.ffclrp.usp.br/Software_P>.
¨
Go¨zler, B., Rentsch, D., Go¨zler, T., Unver, N., Hesse, M., 1996. Lignans,
alkaloids and coumarins from Haplophyllum vulcanicum. Phytochem-
istry 42, 695–699.
4.8. (ꢀ)-Hinokinin
Harmatha, J., Dinan, L., 2003. Biological activities of lignans and
stilbenoids associated with plant-insect chemical interactions. Phyto-
chem. Rev. 2, 321–330.
Chromium trioxide (40.0 mg, 4.0 mmol) was added to
a magnetically stirred solution of dry pyridine (54.2 lL)
in CH2Cl2 (800 lL). The mixture was stirred for 15 min
at room temperature. At the end of this period, a solu-
tion of 2a + 2b (20.0 mg, 0.056 mmol) in 900 lL of
CH2Cl2 was added in one portion. After stirring for
4 h, the solution was decanted from the residue, which
was washed with Et2O (Ratcliffe and Rodehorst, 1970).
The combined organic solutions were dried and the crude
product was purified by preparative TLC (hexane/
EtOAc, 7:3) to give hinokinin (13.6 mg, 0.038 mmol):
Johnstone, R.A.W., Rose, M.E., 1979. A rapid, simple, and mild
procedure for alkylation of phenols, alcohols, amides and acids.
Tetrahedron 35, 2169–2173.
Koul, S.K., Taneja, S.C., Dhar, K.L., Atal, C.K., 1983. Lignans of Piper
clusii. Phytochemistry 22, 999–1000.
Koul, S.K., Taneja, S.C., Pushpangadan, P., Dhar, K.L., 1988. Lignans of
Piper trichostachyon. Phytochemistry 27, 1479–1482.
Lambert, J.B., Papay, J.J., Khan, S.A., Kappauf, K.A., Magyar, E.S.,
1974. Conformational analysis of five-membered rings. J. Am. Chem.
Soc. 96, 6112–6118.
Li, L., Thompson, R., Sowa Jr., J.R., Clausen, A., Dowling, T., 2004.
Kinetic study on the epimerization of trityloxymethyl butyrolactol by
liquid chromatography. J. Chromatogr. A 1043, 171–175.
Li, N., Wu, J.-L., Sakai, J.-I., Ando, M., 2003. Dibenzylbutyrolactone and
dibenzylbutanediol lignans from Peperomia duclouxii. J. Nat. Prod. 66,
1421–1426.
26
½aꢂ ꢀ28.9ꢁ (c 0.31, CHCl3); 1H and 13C NMR, UV,
D
and IR data agree with those reported in the literature
(Lopes et al., 1983).

742
I.C. de Pascoli et al. / Phytochemistry 67 (2006) 735–742
Lightner, D.A., Gurst, J.E., 2000. Organic Conformational Analysis and
Stereochemistry from Circular Dichroism Spectroscopy. Wiley-VCH,
New York.
Lopes, L.M.X., Nascimento, I.R., Silva, T. da, 2001. Phytochemistry of
the Aristolochiaceae family. In: Mohan, R.M.M. (Ed.), Research
Advances in Phytochemistry, vol. 2. Global Research Network,
Kerala, pp. 19–108.
Rehnberg, N., Magnusson, G., 1988. Total synthesis of the lignans (ꢀ)-
and (+)-burseran, (ꢀ)-cubebin, and (ꢀ)-hinokinin by diastereoselec-
tive conjugate addition of benzyl anions to 2-(R)- and (S)-benzyloxy-
2,5-dihydro-4-(3,4-methylenedioxybenzoyl)furan. Tetrahedron Lett.
29, 3599–3602.
Rehnberg, N., Magnusson, G., 1990. General conjugate-addition method
for the synthesis of enantiomerically pure lignans. Total synthesis of
(ꢀ)- and (+)-burseran, (ꢀ)-dehydroxycubebin, (ꢀ)-trichostin, (ꢀ)-
cubebin, (ꢀ)-500-methoxyhinokinin, and (ꢀ)-hinokinin. J. Org. Chem.
55, 4340–4349.
Lopes, L.M.X., Yoshida, M., Gottlieb, O.R., 1983. Dibenzylbutyrolac-
tone lignans from Virola sebifera. Phytochemistry 22, 1516–1518.
Nascimento, I.R., Lopes, L.M.X., 1999. 2,3-Dihydrobenzofuran neolign-
ans from Aristolochia pubescens. Phytochemistry 52, 345–350.
Nascimento, I.R., Lopes, L.M.X., 2000. 2,3-Dihydrobenzofuran neolign-
ans from Aristolochia pubescens. Phytochemistry 53, 621.
Nascimento, I.R., Lopes, L.M.X., 2003. Diterpene esters of aristolochic
acids from Aristolochia pubescens. Phytochemistry 63, 953–957.
Nascimento, I.R., Lopes, L.M.X., Davin, L.B., Lewis, N.G., 2000.
Stereoselective synthesis of 8,9-lidarinediols. Tetrahedron 56, 9181–
9193.
Nascimento, I.R., Murata, A.T., Bortoli, S.A., Lopes, L.M.X., 2004.
Insecticidal activity of chemical constituents from Aristolochia pubescens
against Anticarsia gemmatalis larvae. Pest Manag. Sci. 60, 413–416.
Pihlaja, K., Kleinpeter, E., 1994. Carbon-13 NMR Chemical Shifts in
Structural and Stereochemical Analysis. Wiley-VCH, New York.
Ratcliffe, R., Rodehorst, R., 1970. Improved procedure for oxidations with
the chromium trioxide-pyridine complex. J. Org. Chem. 35, 4000–4002.
Rucker, G., Langmann, B., 1978. 3,4-Dimethoxy-3,4-desmethylendioxy-
¨
cubebin, ein neues lignan aus Aristolochia triangularis. Tetrahedron
Lett. 5, 457–458.
Rucker, G., Langmann, B., Siqueira, N.S. de, 1981. Inhaltsstoffe von
¨
Aristolochia triangularis. Planta Med. 41, 143–149.
Silva, R. da, Souza, G.H.B. de, Silva, A.A. da, Souza, V.A. de, Pereira,
A.C., Royo, V. de A., Silva, M.L.A. de, Donate, P.M., Arau´jo, A.L.S.
de M., Carvalho, J.C.T., Bastos, J.K., 2005. Synthesis and biological
activity evaluation of lignan lactones derived from (ꢀ)-cubebin.
Bioorg. Med. Chem. Lett. 15, 1033–1037.
Ward, R.S., 1990. Asymmetric synthesis of lignans. Tetrahedron 46, 5029–
5041.
Wei-Ming, C., Mayer, R., Rucker, G., 1987. Orientierung der hydroxyl-
¨
gruppe des (ꢀ)-cubebins und seines epimeren. Arch. Pharm. 320, 374–
376.