This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
Author's personal copy
Phytochemistry 71 (2010) 1539–1544
Contents lists available at ScienceDirect
Phytochemistry
journal homepage: www.elsevier.com/locate/phytochem
Sesquiterpene lactones from Vernonia scorpioides and their in vitro cytotoxicity
Humberto Buskuhl a, Fabio L. de Oliveira a, Luise Z. Blind a, Rilton A. de Freitas a, Andersson Barison b,
Francinete R. Campos b, Yuri E. Corilo c, Marcos N. Eberlin c, Giovanni F. Caramori d, Maique W. Biavatti e,*
a
Curso de Farmácia, Centro de Ciências da Saúde, Universidade do Vale do Itajaí (UNIVALI), Itajaí, SC, Brazil
Departamento de Química, Centro Politécnico, Universidade Federal do Paraná (UFPR), Curitiba, PR, Brazil
c
Instituto de Química, Universidade Estadual de Campinas (UNICAMP), Campinas, SP, Brazil
d
Departamento de Química, CFM, Universidade Federal de Santa Catarina (UFSC), Florianópolis, SC, Brazil
e
Departamento de Ciências Farmacêuticas, CCS, Universidade Federal de Santa Catarina (UFSC), Florianópolis, SC, Brazil
b
a r t i c l e
i n f o
Article history:
Received 27 May 2009
Received in revised form 5 March 2010
Available online 3 July 2010
Keywords:
Vernonia scorpioides
Asteraceae
Hirsutinolide
Glaucolide
Piptocarphol esters
Luteolin
Apigenin
Ethyl caffeate
Cytotoxicity
a b s t r a c t
Fresh leaves of Vernonia scorpioides are widely used in Brazil to treat a variety of skin disorders. Previous
in vivo studies with extracts of this species had also demonstrated a high antitumor potential. This paper
reports isolation of four sesquiterpene lactones (hirsutinolides and glaucolides), together with diacetylpiptocarphol, 8-acetyl-13-etoxypiptocarphol, luteolin, apigenin, and ethyl caffeate from fresh leaves and
flowers of Vernonia scorpioides. The hypothesis that hirsutinolide 3 is formed during extraction was verified theoretically using Density Functional Theory. The effects of isolated compounds on in vitro tumor
cells were investigated, as well as their genotoxicity by means of an in vitro comet assay. The results indicate that glaucolide 2 and hirsutinolide 4 are toxic to HeLa cells. These compounds were genotoxic
in vitro, a property that appears to be related to the presence of their epoxy groups, which has been a
more reliable indication of toxicity than substitution on C-13 or the presence of a,b-unsaturated ketogroups. These results need to be replicated in vivo in order to ascertain their toxicity.
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
Vernonia scorpioides (Lam.) Pers., Asteraceae, is well known in
Brazil as Piracá, Enxuga or Erva-de-São-Simão, and usually grows
in poor, deforested, neotropical soils (Cabrera and Klein, 1980).
Topical use of the alcohol extract from fresh leaves is widespread
for treatment of various skin disorders, including chronic wounds
and ulcers. Previous studies using V. scorpioides crude extracts,
and its chloroform and hexane derived fractions, have shown fungicidal and moderate bactericidal activities (Freire et al., 1996), as
well as mild wound healing (Leite et al., 2002) and antitumor effects (Pagno et al., 2006).
Members of the genus Vernonia are good sources of sesquiterpene lactones (SLs). These are often members of the highly
oxygenated family of germacranolides, such as glaucolides and
hirsutinolides, but the cadinanolides have also been reported (Costa
et al., 2005). Hirsutinolides have been reported as having cytotoxic
(Chen et al., 2005), antibacterial and anti-inflammatory properties
* Corresponding author. Address: UFSC/CCS/CIF, Campus Universitário Trindade,
88040-900 Florianópolis, SC, Brazil. Tel./fax: +55 48 3721 5075/9542.
E-mail address: maique@ccs.ufsc.br (M.W. Biavatti).
0031-9422/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phytochem.2010.06.007
(Kos et al., 2006), as well as antiplasmodial (Chea et al., 2006; Pillay
et al., 2007) effects. Glaucolides have properties including those of
smooth muscle relaxants (Campos et al., 2003) and phytogrowth
inhibitors (Barbosa et al., 2004), as well as weak cytotoxic (Williams
et al., 2005) and molluscicidal effects (Alarcon et al., 1990). Some
SLs from V. scorpioides have also been reported (Drew et al., 1980;
Jakupovic et al., 1985; Warning et al., 1987), but no bioactivity for
any compound isolated from V. scorpioides has been reported to
date.
Hirsutinolides have also been named piptocarphol esters, and it
has been proposed that they are formed by rearrangement of
glaucolides when exposed to silica gel in alcohol solution during
chromatographic separations (Catalán et al., 1986). However, some
authors report that hirsutinolides are probably natural products, as
they were detected in fractions before silica gel column chromatography with methanol, and also without the use of silica gel in
the absence of either MeOH or EtOH (Kos et al., 2006).
The present paper reports isolation of four new sesquiterpene
lactones (hirsutinolides and glaucolides), together with diacetylpiptocarphol, 8-acetyl-13-etoxypiptocarphol, luteolin, apigenin and
ethyl caffeate from an ethanol extract of fresh leaves and flowers
of Vernonia scorpioides. The effect of the isolated compounds on a
tumor cell line and their genotoxic properties were investigated.
Author's personal copy
1540
H. Buskuhl et al. / Phytochemistry 71 (2010) 1539–1544
2. Results and discussion
The EtOH extract of leaves and flowers of V. scorpioides were
partitioned with solvents of increasing polarity. The resulting
CH2Cl2 extract exhibited cytotoxic activity (with an IC50 of
0.7 lg/mL; 55.1 lg/mL and >100 lg/mL on HeLa, L929 and
B16F10 cells, respectively). This extract was then subjected to
SiO2 flash chromatography and the sesquiterpene lactone-rich
fractions obtained were further purified to isolate sesquiterpene
lactones 1–6 (Fig. 1).
Glaucolide 1 was obtained, together with hirsutinolide 3, in a
similar concentration, as a white amorphous solid (according to
the 1H NMR spectrum). Both substances were identified based on
analysis of NMR and MS spectroscopic data. The molecular formula
of compounds 1 and 3 was C19H24O10, as deduced by HR-ESI(+)-MS,
which showed an accurate [M+H]+ ion at m/z 413.1350 (calcd.
glaucolides 413.14477). All isolated SLs were found to be isomers
with the same mass. The ESI( )-MS spectra of compounds 1–4
gave a [M H] species of m/z 411. The main fragments observed
in the MS spectrum of compounds 1 and 3 were m/z 369, m/z
351, m/z 309 and m/z 291, which were confirmed by ESI( )-MS/
MS experiments. These fragments indicated a loss of ketene from
the protonated molecule (m/z 411) to form the fragment m/z 369,
followed by loss of acetic acid to form the ion fragment m/z 309.
In contrast, the molecular ion may lose acetic acid to form the
ion fragment m/z 351, followed by the loss of ketene to form ion
fragment m/z 291. The ESI( )-MS/MS spectra of compounds 2
and 4 were similar, indicating that they dissociate mainly into
the ion m/z 369 from the loss of ketene, as well as the fragments
cited above. The presence of a hydroxyl at C-6 of compound 3
probably facilitates the loss of acetic acid to form the main peak
(m/z 351) in the MS spectra.
In the 1H NMR spectrum of compounds 1 and 3, four singlet signals were evident in the range of 2.06–2.15 ppm (3H each), interpreted as the presence of four acetate groups. Two doublet
signals in the range of 4.88–5.08 ppm (J = 13.0 Hz), characteristic
of geminal methylene hydrogens close to oxygen, were assigned
Fig. 1. Structures of compounds 1–6.
as H-13a and H-13b, and four singlet signals (3H) were inferred
to the methyl hydrogens H-14 (1.32 and 1.36 ppm) and H-15
(1.47 and 1.61 ppm). The 13C NMR spectrum presented 38 carbons,
showing characteristic signals of a,b-unsaturated c-lactone rings
at 167.2 and 167.5 ppm (C-12), 129.9 and 120.7 ppm (C-11), and
162.0 and 160.8 ppm (C-7) (Table 1). Glaucolide 1 is apparently
the precursor of hirsutinolide 3, which may have been formed during the extraction procedure in acidic silica. The formation of an
unusual C1–C5 epoxide bridge in compound 3 could be explained
by the strong correlation of H-5 at 3.54 ppm with C-1 at 95.0 ppm,
as observed in an HMBC experiment.
In order to evaluate the hypothesis concerning formation of 3
during the extraction, the proposed compound 3 was studied by
computational calculations using Density Functional Theory
(DFT), in which geometry optimization and stretching frequency
calculations were carried out using the B3LYP/6-31+G(d,p) level
(Becke, 1993; Rassolov et al., 2001). This study enabled the researcher to determine whether the amount of energy needed to
form the compound was prohibitive. The results suggest that the
proposed structure 3 corresponded to the minimum level of energy
on the potential energy surface and confirmed by harmonic vibration frequency analysis (Fig. 2A). This minimum energy structure
showed a short, strong intramolecular hydrogen bond (1.597 Å),
which is also in agreement with the O–H bond lengthening with
a concomitant red shift in the O–H stretching frequency (1.044 Å
and 3436 cm 1) compared with the other O–H (1.028 Å and
3689 cm 1) present in the structure, as depicted in Fig. 2B. This
structural feature would give the compound additional electronic
stability (Fig. 2A).
Glaucolide 2 was obtained as a colorless gum, with the same
molecular formula (C19H24O10) as compounds 1 and 3, as evidenced in its high-resolution MS spectra. The 13C NMR spectrum
presented nineteen signals, with a typical glaucolide carbonyl at
215.1 ppm (C-1), in addition to the acetate carbonyls at 169.1
and 170.2 ppm (8- and 13-acetate). Its 1H NMR spectrum showed
the presence of several singlet signals. The resonances at 4.24
(1H) and 4.97 (2H) ppm were attributed to H-5 and H-13 respectively, whereas the signals at 2.07 and 2.14 ppm were assigned
to the methyl groups of the 8- and 13-acetates, respectively. The
resonances at 1.31 and 1.36 ppm were related to the methyl hydrogens H-14 and H-15, respectively (Table 2). The correlation of H-15
at 1.36 ppm with C-5 at 67.0 ppm, and the correlations of H-8 at
6.11 and H-13 at 4.97 ppm with the C-6 at 87.6 ppm, as observed
in the 1H–13C long-range correlation experiment (HMBC), clearly
indicated the presence of an epoxide between C-5 and C-6. Typical
glaucolides isolated from Vernonia exhibited an epoxide between
C-4 and C-5 (Padolina et al., 1974) and are considered precursors
of the non-natural hirsutinolides (Jiménez et al., 1995). Glaucolide
2 may, therefore, be the precursor of hirsutinolide 4.
Hirsutinolide 4, also an isomer of the previously isolated compounds, gave characteristic signals of Vernonia hirsutinolides in
the 13C NMR spectrum, such as the hemiacetal carbon at
108.5 ppm (C-1) and the oxygen bearing carbon at 78.1 ppm (C4) (Table 2). The main differences were found in the 1H NMR spectrum, which showed a singlet at 3.48 ppm (1H) that correlated in
the HSQC experiment with the carbonyl carbon at 69.9 ppm (C5), and in the HMBC with the carbons at 78.1 (C-4) and 90.2 ppm
(C-6), suggesting an epoxide between C-5 and C-6 instead of the
usual double bond.
The relative configuration of 4 was established through 1D NOE
experiments. Selective irradiation of the resonance frequency of H5 at 3.48 ppm caused a NOE enhancement on the signal of the hydrogen H-15, indicating a b-orientation of the C5–C6 epoxide group
(this enhancement was not so clear for the C5–C6 epoxide group
of compounds 1 and 2). Moreover, selective irradiation of the resonance frequency of H-14 at 1.14 ppm showed NOE intensification
Author's personal copy
1541
H. Buskuhl et al. / Phytochemistry 71 (2010) 1539–1544
Table 1
NMR spectroscopic data for compounds 1 and 3.a
Carbon
1
3
1
H NMR mult. (J)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
CH3COO-8
CH3COO-8
CH3COO-13
CH3COO-13
2.16
2.23
2.18
2.51
m
m
m
m
3.78 s
5.47 dd (7.4, 2.7)
2.23 dd (14.5, 2.7)
2.61 dd (14.5, 7.4)
4.90
5.09
1.32
1.61
2.06
d (13.0)
d (13.0)
s
s
s
2.07 s
13
C NMR
107.9
36.7
33.6
80.5
71.4
91.5
160.8
63.9
47.1
75.4
129.9
167.2
55.6
24.8
29.7
20.8
169.5
20.6
170.4
b
1
HMBC
1,
1,
1,
1,
4,
4,
2,
4,
3
3
4
5
H NMR mult. (J)
and10
and10
and 15
and 15
1.77
2.32
1.82
2.65
m
m
m
ddd (14.2, 7.1, 3.3)
4, 6 and 15
3.54 s
6, 7, 9, 10, 11 and CH3COO-8
7, 8 and 10
7, 8, 10 and 15
6.73 dd (10.4, 7.8)
1.56 dd (13.3, 10.4)
2.91 dd (13.3, 7.8)
7, 11, 12 and CH3COO-13
7, 11, 12 and CH3COO-13
1, 9 and 10
3, 4 and 5
CH3COO-8
4.88
4.99
1.36
1.47
2.15
CH3COO-13
2.13 s
dd (12.3; 0.5)
dd (12.3; 1.2)
s
s
s
13
C NMR
95.0
33.2
35.0
70.2
78.7
102.2
162.0
67.9
39.9
82.6
120.7
167.5
54.8
23.0
26.7
20.9
169.5
20.8
170.4
HMBCb
1,
1,
2,
2,
3, 4 and 10
3 and 4
4, 5 and 15
4 and 15
1, 3, 4, 6, 7 and 15
7, 9, 11 and CH3COO-8
1, 8, 10 and 14
7, 8 and 10
7, 11, 12 and CH3COO-13
7, 11, 12 and CH3COO-13
1, 9 and 10
3, 4 and 5
CH3COO-8
CH3COO-13
a
The experiments were recorded in CDCl3 and all NMR chemical shifts are given in ppm related to the TMS signal at 0.00 ppm as internal reference and coupling constants
(J) are given in Hz. The unambiguous 1H and 13C NMR chemical shift assignments were established by a combination of 1D and 2D NMR including 1H–1H, one-bond and longrange 1H–13C correlation experiments.
b
Long-range 1H–13C HMBC correlations, optimized for 8 Hz, are from hydrogen(s) stated to the indicated carbon.
Fig. 2. Optimized structure of hirsutinolide 3 (A), O-H bond lengths (Å) and
stretching frequencies (cm 1) (in brackets) of 3 at B3LYP/6-31+G(d,p) level (B).
mainly on the signal of hydrogen H-9 at 2.07 ppm, whereas selective
irradiation of the resonance frequency of H-8 at 5.99 ppm caused a
strong NOE enhancement on the signals of H-9 at 2.56 ppm and H13 at 4.78 ppm. The overall analyses of 1D NOE enabled the
establishment of 4 as 8a-acetoxy-1a,10a-hydroxy-5,6-epoxyhirsutinolide-13-O-acetate. The relative stereochemistry of
compounds 1 and 2 should be in the form presented, but structural
confirmation will need to be obtained.
Hirsutinolides 5 and 6 were isolated as amorphous solids and
are characteristic compounds found in several Vernonia species.
Their MS spectra, as well as the 1D and 2D NMR spectroscopic data,
were in full agreement with earlier published data for the compounds diacetylpiptocarphol (or 1b,4b-Epoxy-8a,l3-diacetoxy1a,10a-dihydroxy-germacra-5E,7(11)-dien-6,12-olide) (5, Bardón
et al., 1988) and 8-acetyl-13-etoxypiptocarphol (or 1b,4b-Epoxyl3-etoxy-1a,10a-dihydroxy-8-acetoxy-germacra-5E,7(11)-dien-6,
12-olide) (6, Catalán et al., 1986). The resolution of their 1H NMR
spectra was improved by decreasing the temperature to 273 K
(see Supplementary data, Fig. S1). Compound 5 has already been
described in V. scorpioides (Bardón et al., 1988; Catalán et al.,
1986), and compound 6 was first published with the H-8 a-orientation (Catalán et al., 1986). After that, structure 6 was commonly
represented with an H-8 b-orientation (Borkosky et al., 1996;
Valdés et al., 1998).
The configurations of the asymmetric carbon atoms of the sesquiterpene skeleton are assumed to be analogous to those previously established for similar compounds described in the
literature (Valdés et al., 1998) because insufficient amount of
material hampered attempts to determine the absolute configuration of the molecules.
Some cytotoxicity has been described for SLs from V. cinerea
(Kuo et al., 2003), V. lasiopus (Koul et al., 2003) and V. amygdalina
(Jisaka et al., 1993).
The results of the in vitro tests obtained for the isolated SLs are
summarized in Table 3. Hirsutinolide 4 showed significant cytotoxicity only on HeLa cells, with an IC50 of 3.3 ± 4.0 lM. On L929 and
Author's personal copy
1542
H. Buskuhl et al. / Phytochemistry 71 (2010) 1539–1544
Table 2
NMR spectroscopic data for compounds 2 and 4.a
Carbon
2
1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
CH3COO-8
CH3COO-8
CH3COO-13
CH3COO-13
4
13
H NMR mult. (J)
1.89
3.96
1.89
2.21
C NMR
215.1
27.7
m
m
m
m
37.4
4.24 s
6.11 dd (5.5, 2.5)
2.23 dd (15.5, 5.5)
2.80 dd (15.5, 2.5)
71.2
67.0
87.6
155.1
68.0
41.8
80.2
128.2
166.7
55.0
4.97 s
1.31 s
1.36 s
2.07 s
30.3
23.9
20.4
169.1
20.6
170.2
2.14 s
b
HMBC
1
1 and 3
1 and 3
1, 2, 4 and 5
2, 4 and 15
1.44
1.91
2.11
2.31
4, 6 and 15
3.48 s
6, 7, 10 and CH3COO-8
1, 7, 8, 10 and 14
10
5.99 dd (10.2, 0.7)
2.07 dd (16.2, 0.7)
2.56 dd (16.2, 10.2)
H NMR mult. (J)
6, 7, 11, 12 and CH3COO-13
(ddd
(ddd
(ddd
(ddd
1, 9 and 10
3, 4 and 5
CH3COO-8
4.78
5.95
1.14
1.52
2.05
CH3COO-13
2.04 s
12.8,
12.5,
12.8,
12.8,
12.5, 8.6)
7.6, 0.5)
8.6, 0.5)
12.8, 7.6)
d (13.5)
d (13.5)
s
s
s
13
C NMR
108.5
32.2
34.6
78.1
69.9
90.2
158.3
68.6
40.5
77.1
132.1
166.0
55.1
25.6
26.7
20.6
169.6
20.6
170.0
HMBCb
1, 3 and 4
1, 2 and 5
2, 4, 5 and 15
4 and 6
6, 7, 9, 10, 11 and CH3COO-8
7, 10 and 14
1, 7, 8, 10 and 14
7, 11, 12 and CH3COO-13
7, 11, 12 and CH3COO-13
1, 9 and 10
3, 4 and 5
CH3COO-8
CH3COO-13
a
The experiments were recorded in CDCl3 and all NMR chemical shifts are given in ppm related to the TMS signal at 0.00 ppm as internal reference and coupling constants
(J) are given in Hz. The unambiguous 1H and 13C NMR chemical shifts assignments were established by a combination of 1D and 2D NMR including 1H–1H, one-bond and longrange 1H–13C correlation experiments.
b
Long-range 1H–13C HMBC correlations, optimized for 8 Hz, are from hydrogen(s) stated to the indicated carbon.
Table 3
SL cytotoxicity on HeLa, L929 and B16F10 cells, and the Damage Index on bone
marrow cells.
Sample
IC50
HeLaa
(lM)
IC50
L929a
(lM)
IC50
B16F10a
(lM)
Lactone mixture 1/
3
Glaucolide 2
Hirsutinolide 4
Hirsutinolide 5
Hirsutinolide 6
Paclitaxel
MMS
>100
>100
>100
2.1 ± 1.8
3.3 ± 4.0
>100
58.5 ± 15
2.5 ± 2.2
–
>100
>100
>100
>100
8.62 ± 8.8
–
>100
>100
>100
>100
6.85 ± 5.4
–
Damage
indexb
154 ± 23
137 ± 15
163 ± 27
128.5 ± 13
132.3 ± 19.5
–
189 ± 1.3
a
IC50 determined after 24-h incubation at 37 °C/5%CO2.
Damage index determined at concentrations of hirsutinolide 4, 5 and 6, glaucolide 2 and lactone mixture 1/3 of 24.3, 25.2, 26.2, 24.3 and 242 lM, respectively.
Negative control (1% DMSO) of 14.3 ± 2.1 (maximal possible score 200).
b
B16F10, the IC50 was >100 lM, with a reduction in viability of up to
20%. Hirsutinolide 6 gave an IC50 of 58.5 ± 15 lM on HeLa cells and
also showed low toxicity on L929 and B16F10 cells (IC50 > 100 lM).
Hirsutinolide 6 was apparently more cytotoxic than hirsutinolide 5
in HeLa cells (IC50 > 100 lM).
Hirsutinolide 6 has an ethoxy group on C-13 and hirsutinolides
4 and 5 presented acetoxy groups. Kuo et al. (2003), comparing
hydroxylated with acetylated SLs in C-13, observed lower cytotoxicity for the acetylated SLs. The major difference between hirsutinolide 4 and 5 relates to the presence of an epoxy group (4) or a
double bond (5) across C5–C6. In fact, the presence of epoxy groups
(as in glaucolide 2) could be related to a more significant cytotoxic
effect. This hypothesis was confirmed by comparing the cytotoxicity of glaucolide 2 (IC50 2.1 ± 1.8 lM) with that of hirsutinolide 4
(IC50 3.3 ± 4.0 lM) in HeLa cells, with these two cytotoxicities
being statistically similar to the IC50 of paclitaxel (2.5 ± 2.2 lM)
in HeLa cells.
Pillay et al. (2007) observed that a sesquiterpene lactone (4,5aepoxy-6a-hydroxy-1(10)E,11(13)-germacradien-12,8a-olide) was
highly cytotoxic to CHO cells (IC50 2.2 lg/mL). Dirsch et al.
(2000), evaluating glaucolide A, suggested that the a-methylenec-lactone group also makes a strong contribution to biological
activity.
The genotoxicity of the isolated compounds was verified using
bone marrow cells. The results obtained from this assay show a
tendency of these compounds to damage DNA, but mutagenic
properties were not evaluated.
The negative and positive controls used in the genotoxicity test
were DMSO (1%) and methyl methane sulfonate (MMS, 40 lM)
with respective damage indices (DI) of 14.3 ± 2.1 and 189 ± 1.3.
Glaucolide 2, hirsutinolide 5 and hirsutinolide 6 showed DI values
that were statistically lower than MMS (p < 0.05), but all the compounds registered DI values when compared with the negative
control (p < 0.01) indicating genotoxicity (Table 3).
These results cannot be directly related to a mutagenic potential in vivo, as mutagenic assays must be performed to confirm
the genotoxicity and mutagenic potential. The results obtained
here suggest that at least part of the cytotoxicity might be related
to genotoxic aspects (Collins and Dusinská, 2002).
Burim et al. (1999) observed that glaucolide B, isolated from V.
eremophila, had an increase in chromosomal aberrations in lymphocytes at 4 and 8 lg/mL, as well as cytotoxicity at a concentration of >8 lg/mL. Glaucolide B at 160–640 mg/kg, however, did
not significantly increase the frequency of chromosomal aberrations in mouse bone marrow cells, nor did it affect cell division
in in vivo experiments using BALBc mice.
3. Experimental
3.1. General experimental procedures
Melting points were determined using a QUIMIS Q-340S23
micromelting point apparatus. All NMR data were recorded in
CDCl3 at 295 K for compounds 1–4, and 295 and 273 K for
compounds 5 and 6, respectively, using a Bruker AVANCE 400
Author's personal copy
H. Buskuhl et al. / Phytochemistry 71 (2010) 1539–1544
spectrometer operating at 9.4 T. The 1H and 13C isotopes were observed at 400 and 100 MHz, respectively. One-bond and long-range
1
H–13C correlation (HSQC and HMBC) experiments were optimized
for an average coupling constant 1J(C,H) of 140 Hz and LRJ(C,H) of
8 Hz, respectively. All 1H and 13C NMR chemical shifts (d) are given
in ppm related to TMS as an internal reference at 0.00 ppm, and the
coupling constant (J) in Hz. All of the pulse programs employed
were supplied by Bruker. Low-resolution ESI-MS and ESI-MS/MS
data were acquired in the negative ion mode using a Bruker Esquire 6000 ESI-ion trap instrument, whereas the high-resolution
HR-TOF-MS measurements were carried out using a hybrid quadrupole reflector orthogonal time-of-flight high-resolution Bruker
Q-TOF mass spectrometer equipped with an electrospray source.
3.2. Plant material
Flowers and leaves of V. scorpioides were collected from wild
specimens of ‘‘restinga” forest (a distinct type of coastal tropical
and subtropical moist broadleaf forest) in Navegantes in November
2006, and identified by Dr. Ana Claudia Araújo of the Universidade
do Vale do Itajaí. A voucher specimen (M. Biavatti 11) was deposited at the Barbosa Rodrigues Herbarium (HBR), Itajaí, Santa Catarina, Brazil.
3.3. Extraction and isolation
Fresh flowers and leaves of V. scorpioides (3 kg) were extracted
with EtOH (6 L) at room temperature, in the absence of light, for
7 days. After solvent reduction to 1/6 of the initial volume under
reduced pressure and the addition of H2O (500 mL), the extract
was submitted to liquid–liquid fractioning using solvents with
increasing polarities. This procedure produced n-hexane (1.2 g),
CH2Cl2 (2.3 g), and EtOAc (1.4 g) fractions. The CH2Cl2 fraction
was initially subjected to silica gel CC (60–230 mesh) eluted with
n-hexane (500 mL), followed by CH2Cl2 (800 mL), EtOAc (700 mL)
and MeOH (400 mL), yielding one subfraction for each of the four
solvents. Additional chromatography separation of the CH2Cl2 subfractions (420 mg) were carried out by silica gel CC (230–400
mesh), using n-hexane with increasing concentrations of EtOAc
(0–50% EtOAc) as eluent, and collecting eluent in 30–50 mL portions to furnished 42 fractions. Fractions 17–24 were combined
and subjected to silica gel CC using n-hexane:ethyl acetate (7:3)
to yield compound 6 (4.5 mg, 8-acetyl-13-etoxypiptocarphol
(Catalán et al., 1986) and ethyl caffeate (11.86 mg, Su et al.,
2008). Fractions 20–26 were combined and subjected to MPLC on
a silica gel Lobar B column (Merck) using n-hexane:EtOAc (8:2)
to yield compounds 1 and 3 (15 mg) as a mixture, as well as compound 5 (6 mg, diacetylpiptocarphol, Bardón et al., 1988).
Chromatographic separation of the EtOAc subfraction (1.7 g) by
silica gel CC (230–400 mesh) using n-hexane with increasing concentrations of acetone (from 10% to 50%), and collecting eluant
fractions of 30–50 mL, furnished 82 fractions. Fractions 20–25
were combined and subjected to silica gel CC as above, yielding a
mixture of SLs (163 mg), which was subjected to MPLC on a silica
gel Lobar B column (Merck) using n-hexane:acetone (1:1) yielding
compounds 2 (12 mg) and 4 (25 mg). From fractions 26–61, luteolin (10 mg) and apigenin (5 mg) were obtained after successive silica gel CC, as described above.
3.4. 8a,13-Diacetoxy-1a,10a,-5b,6b-diepoxygermacra-7(11)-en-12olide (1)
Amorphous solid, mp 98–100 °C; for 1H and 13C NMR spectroscopic data, see Table 1; ESI-MS m/z 411.5 [M H] ; ESI-MS/MS
(daughter ions, 25%) m/z 369 (30), m/z 351 [M H AcOH] (60),
1543
309 (83), 291 (50); HR-TOF-MS (ESI positive) m/z 413.1350
[M+H]+ (calcd for C19H24O10+H, 413.14477).
3.5. 10a,4a-Dihydroxy-5b,6b-isoglaucolide B (2)
Colorless gum; for 1H and 13C NMR spectroscopic data, see Table 1; ESI-MS m/z 411.2 [M H] ; ESI-MS/MS (daughter ions,
25%) m/z 369 (80), 351 [M H AcOH] (30), 309 (20), 291 (40),
247 (70); HR-TOF-MS (ESI positive) m/z 413.1350 [M+H]+ (calcd
for C19H24O10+H, 413.14477).
3.6. 8a,13-Diacetoxy-1a,10a,-1b,5b-diepoxygermacra-7(11)-en-12olide (3)
Amorphous solid, mp 98–100 °C; for 1H and 13C NMR spectroscopic data, see Table 1; ESI-MS m/z 411.5 [M H] ; ESI-MS/MS
(daughter ions, 25%) m/z 369 (30), m/z 351 [M H AcOH] (60),
309 (83), 291 (50); HR-TOF-MS (ESI positive) m/z 413.1350
[M+H]+ (calcd for C19H24O10+H, 413.14477).
3.7. 8a,13-Diacetoxy-1a,10a-hydroxy-5,6-epoxy-hirsutinolide (4)
Colorless gum; for 1H and 13C NMR spectroscopic data, see Table 1; ESI-MS m/z 411.2 [M H] ; ESI-MS/MS (daughter ions,
25%) m/z 369 (100), 351 [M H AcOH] (40), 309 (38), 291 (50),
247 (45); HR-TOF-MS (ESI positive) m/z 413.1350 [M+H]+ (calculated for C19H24O10+H, 413.14477).
3.8. Computational methods
All calculations were carried out using exchange–correlation
functional B3LYP (Becke, 1993) with 6-31+G(d,p) (Rassolov et al.,
2001) using the Gaussian03 suite of programs (Frisch et al.,
2003). The structures presented in Fig. 2 were obtained using
ChemCraft 1.5 (http://www.chemcraftprog.com).
3.9. Cytotoxicity assay
Cell viability was assessed by 3-(4,5-dimethylthiozol-2-yl)-2,5diphenyltetrazolium bromide (MTT) assays to determine IC50
concentrations of the studied agents, as described previously
(Mosmann, 1983). The full procedure is supplied as Supplementary
data.
3.10. Comet assay
In vitro genotoxicity was assessed using bone marrow cells obtained from the femur of Mus musculus (Swiss). The UNIVALI protocol 236/07 was used, as described previously (Tice et al., 2000),
and the full procedure is supplied as Supplementary data.
3.11. Statistical analysis
The results are expressed as mean values ± SD from three separate experiments. The IC50 values, i.e., the concentration necessary
for 50% inhibition, were calculated from the dose response curves
using non-linear regression analysis that gave a percentage of the
inhibition values. Group differences were determined by analysis
of variance (ANOVA). When statistically significant differences
were indicated by ANOVA, the values were compared by the Tukey
test. The differences were considered statistically significant from
the controls at p < 0.05.
Author's personal copy
1544
H. Buskuhl et al. / Phytochemistry 71 (2010) 1539–1544
Acknowledgements
The authors are grateful to CNPq and CAPES for their financial
support. G.F.C. thanks CENAPAD – SP for the allocated computational time and resources.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.phytochem.2010.06.007.
References
Alarcon, M.C., Callegari Lopes, J.L., Herz, W., 1990. Glaucolide B, a molluscicidal
sesquiterpene lactone, and other constituents of Vernonia eremophila. Planta
Med. 56, 271–273.
Barbosa, L.C., Costa, A.V., Pilo-Veloso, D., Lopes, J.L., Hernandez-Terrones, M.G., KingDiaz, B., Lotina-Hennsen, B., 2004. Phytogrowth-inhibitory lactones derivatives
of glaucolide B. Z. Naturforsch. [C] 59, 803–810.
Bardón, A., Catalán, C.A.N., Gutierrez, A.B., Herz, W., 1988. Piptocarphol esters and
other constituents from Vernonia cognata. Phytochemistry 27, 2989–2990.
Becke, A.D., 1993. Density-functional thermochemistry. III. The role of exact
exchange. J. Chem. Phys. 98, 5648–5652.
Borkosky, S., Alvarez Valdés, D., Bardón, A., Díaz, J.G., Herz, W., 1996. Sesquiterpene
lactones and other constituents of Eirmocephala megaphylla and Cyrtocymura
cincta. Phytochemistry 42, 1637–1639.
Burim, R.V., Canalle, R., Lopes, J.L.C., Takahashi, C.S., 1999. Genotoxic action of the
sesquiterpene lactone glaucolide B on mammalian cells in vitro and in vivo.
Gent. Mol. Biol. 22, 401–406.
Cabrera, A.L., Klein, R.M., 1980. Compostas: 3. Tribo: Vernoniae. Fl. Ilustr. Catarin.
354–355.
Campos, M., Oropeza, M., Ponce, H., Fernandez, J., Jimenez-Estrada, M., Torres, H.,
Reyes-Chilpa, R., 2003. Relaxation of uterine and aortic smooth muscle by
glaucolides D and E from Vernonia liatroides. Biol. Pharm. Bull. 26, 112–115.
Catalán, C.A.N., de Iglesias, D.I.A., Kavka, J., Sosa, V.E., Herz, W., 1986. Sesquiterpene
lactones and other constituents of Vernonia mollissima and Vernonia squamulosa.
J. Nat. Prod. 49, 351–353.
Chea, A., Hout, S., Long, C., Marcourt, L., Faure, R., Azas, N., Elias, R., 2006.
Antimalarial activity of sesquiterpene lactones from Vernonia cinerea. Chem.
Pharm. Bull. 54, 1437–1439.
Chen, X., Zhan, Z.J., Zhang, X.W., Ding, J., Yue, J.M., 2005. Sesquiterpene lactones with
potent cytotoxic activities from Vernonia chinensis. Planta Med. 71, 949–954.
Collins, A.R., Dusinská, M., 2002. Oxidation of cellular DNA, measured with the
comet assay. In: Armstrong, D. (Ed.), Oxidative Stress Biomarkers and
Antioxidant Protocols. Humana Press, New Jersey, pp. 147–159.
Costa, F.B., Terfloth, L., Gasteiger, J., 2005. Sesquiterpene lactone-based classification
of three Asteraceae tribes: a study based on self-organizing neural networks
applied to chemosystematics. Phytochemistry 66, 345–353.
Dirsch, V.M., Stuppner, J., Ellmerer-Müller, E.P., Vollmar, A.M., 2000. Structural
requirements of sesquiterpene lactones to inhibit LPS-induced nitric oxide
synthesis in RAW 264.7 macrophages. Bioorg. Med. Chem. 8, 2747–2753.
Drew, M.G.B., Hitchman, S.P., Mann, J., Lopes, J.L.C., 1980. X-ray crystal structure of
the sesquiterpene lactone scorpioidine. J. Chem. Soc., Chem. Commun. 17, 802–
803.
Freire, M.F.I., Abreu, H.S., Cruz, L.C.H., Freire, R.B., 1996. Inhibition of fungal growth
by extracts of Vernonia scorpioides (Lam.) Pers. Microbiology 27, 1–6.
Frisch, M.J. et al., 2003. Gaussian 03, Revision A.1. Gaussian Inc., Pittsburgh, PA.
Jakupovic, J., Baruah, R.N., Chau Thi, T.V., 1985. New vernolepin derivatives from
Vernonia glabra and glaucolides from Vernonia scorpioides. Planta Med. 5, 378–
380.
Jiménez, M., Ortega, A., Navarro, A., Maldonado, E., Van Calsteren, M.R., Jankowski,
C.K., 1995. Reaction of the molluscicide glaucolide B with bentonite. J. Nat. Prod.
58, 424–427.
Jisaka, M., Ohigashi, H., Takegawa, K., Huffman, M.A., Koshimizu, K., 1993.
Antitumoral and antimicrobial activities of bitter sesquiterpene lactones of
Vernonia amygdalina, a possible medicinal plant used by wild chimpanzees.
Biosci. Biotechnol. Biochem. 57, 833–834.
Kos, O., Castro, V., Murillo, R., Poveda, L., Merfort, I., 2006. Ent-kaurane glycosides
and sesquiterpene lactones of the hirsutinolide type from Vernonia triflosculosa.
Phytochemistry 67, 62–69.
Koul, J.L., Koul, S., Singh, C., Taneja, S.C., Shanmugavel, M., Kampasi, H., Saxena, A.K.,
Qazi, G.N., 2003. In vitro cytotoxic elemanolides from Vernonia lasiopus. Planta
Med. 69, 164–166.
Kuo, Y.H., Kuo, Y.J., Yu, A.S., Wu, M.D., Ong, C.W., Yang Kuo, L.M., Huang, J.T., Chen,
C.F., Li, S.Y., 2003. Two novel sesquiterpene lactones, cytotoxic vernolide-A and B, from Vernonia cinerea. Chem. Pharm. Bull. 51, 425–426.
Leite, S.N., Palhano, G., Almeida, S., Biavatti, M.W., 2002. Wound healing activity and
systemic effects of Vernonia scorpioides extract in guinea pig. Fitoterapia 73,
496–500.
Mosmann, T., 1983. Rapid colorimetric assay for cellular growth and survival:
application to proliferation and cytotoxicity assays. J. Immunol. Methods 65,
55–63.
Padolina, W.G., Yoshioka, H., Nakatani, N., Mabry, T.J., Monti, S.A., Davis, R.E., Cox,
P.J., Sim, G.A., Watson, W.H., Wu, I.B., 1974. Glaucolide-A and -B, new
germacranolide-type sesquiterpene lactones from Vernonia (Compositae).
Tetrahedron 30, 1161–1170.
Pagno, T., Blind, L.Z., Biavatti, M.W., Kreuger, M.R., 2006. Cytotoxic activity of the
dichloromethane fraction from Vernonia scorpioides (Lam.) Pers. (Asteraceae)
against Ehrlich’s tumor cells in mice. Braz. J. Med. Biol. Res 39, 1483–1491.
Pillay, P., Vleggaar, R., Maharaj, V.J., Smith, P.J., Lategan, C.A., Chouteau, F., Chibale,
K., 2007. Antiplasmodial hirsutinolides from Vernonia staehelinoides and their
utilization towards a simplified pharmacophore. Phytochemistry 68, 1200–
1205.
Rassolov, V.A., Ratner, M.A., Pople, J.A., Redfern, P.C., Curtiss, L.A., 2001. 6-31G* basis
set for third-row atoms. J. Comp. Chem. 22, 976–984.
Su, Y.O., Zhang, W.D., Zhang, C., Liu, R.H., Shen, Y.H., 2008. A new caffeic ester from
Incarvillea mairei var. grandiflora (Wehrhahn) Grierson. Chin. Chem. Lett. 19,
829–831.
Tice, R.R., Agurell, E., Anderson, D., Burlinson, B., Hartmann, A., Kobayashi, H.,
Miyamae, Y., Rojas, E., Ryu, J.C., Sasaki, Y.F., 2000. Single cell gel/comet assay:
guidelines for in vitro and in vivo genetic toxicology testing. Environ. Mol.
Mutag. 35, 206–221.
Valdés, D.A., Bardón, A., Catalán, C.A.N., Gedris, T.E., Herz, W., 1998. Glaucolides,
piptocarphins and cadinanolides from Lepidaploa remotiflora. Biochem. Syst.
Ecol. 26, 685–689.
Warning, U., Jakupovic, J., Bohlmann, F., 1987. Scorpiolid, ein neuer
sesquiterpenlacton-typ aus Vernonia scorpioides. Liebigs Ann. Chem., 467.
Williams, R.B., Norris, A., Slebodnick, C., Merola, J., Miller, J.S., Andriantsiferana, R.,
Rasamison, V.E., Kingston, D.G., 2005. Cytotoxic sesquiterpene lactones from
Vernonia pachyclada from the Madagascar rainforest. J. Nat. Prod. 68, 1371–
1374.