Bulbispermine: A Crinine-Type Amaryllidaceae Alkaloid Exhibiting Cytostatic Activity toward Apoptosis-Resistant Glioma Cells

Article abstract of DOI:10.1002/cmdc.201100608

The Amaryllidaceae alkaloid bulbispermine was derivatized to produce a small group of synthetic analogues. These, together with bulbispermine's natural crinine-type congeners, were evaluated invitro against a panel of cancer cell lines with various levels of resistance to pro-apoptotic stimuli. Bulbispermine, haemanthamine, and haemanthidine showed the most potent antiproliferative activities as determined by the MTT colorimetric assay. Among the synthetic bulbispermine analogues, only the C1,C2-dicarbamate derivative exhibited notable growth inhibitory properties. All active compounds were found not to discriminate between the cancer cell lines based on the apoptosis sensitivity criterion; they displayed similar potencies in both cell types, indicating that the induction of apoptosis is not the primary mechanism responsible for antiproliferative activity in this series of compounds. It was also found that bulbispermine inhibits the proliferation of glioblastoma cells through cytostatic effects, possibly arising from rigidification of the actin cytoskeleton. These findings lead us to argue that crinine-type alkaloids are potentially useful drug leads for the treatment of apoptosis-resistant cancers and glioblastoma in particular.

Full text of DOI:10.1002/cmdc.201100608

MED
DOI: 10.1002/cmdc.201100608
Bulbispermine: A Crinine-Type Amaryllidaceae Alkaloid
Exhibiting Cytostatic Activity toward Apoptosis-Resistant
Glioma Cells
Giovanni Luchetti,[a] Robert Johnston,[b] Vꢀronique Mathieu,[c] Florence Lefranc,[c, d]
Kathryn Hayden,[b] Anna Andolfi,[e] Delphine Lamoral-Theys,[f] Mary R. Reisenauer,[b]
Cody Champion,[a] Stephen C. Pelly,[g] Willem A. L. van Otterlo,[g] Igor V. Magedov,[a]
Robert Kiss,[c] Antonio Evidente,[e] Snezna Rogelj,[b] and Alexander Kornienko*[a]
The Amaryllidaceae alkaloid bulbispermine was derivatized to
produce a small group of synthetic analogues. These, together
with bulbispermine’s natural crinine-type congeners, were eval-
uated in vitro against a panel of cancer cell lines with various
levels of resistance to pro-apoptotic stimuli. Bulbispermine,
haemanthamine, and haemanthidine showed the most potent
antiproliferative activities as determined by the MTT colorimet-
ric assay. Among the synthetic bulbispermine analogues, only
the C1,C2-dicarbamate derivative exhibited notable growth in-
hibitory properties. All active compounds were found not to
discriminate between the cancer cell lines based on the apop-
tosis sensitivity criterion; they displayed similar potencies in
both cell types, indicating that the induction of apoptosis is
not the primary mechanism responsible for antiproliferative ac-
tivity in this series of compounds. It was also found that bul-
bispermine inhibits the proliferation of glioblastoma cells
through cytostatic effects, possibly arising from rigidification of
the actin cytoskeleton. These findings lead us to argue that cri-
nine-type alkaloids are potentially useful drug leads for the
treatment of apoptosis-resistant cancers and glioblastoma in
particular.
Introduction
The anticancer properties of plants belonging to the Amarylli-
daceae family were known as early as the fourth century B.C.,
when Hippocrates of Cos used oil from the daffodil Narcissus
poeticus L. for the treatment of uterine tumors.[1] Using related
biosynthetic pathways, these plants produce a large number
of diverse alkaloids and related non-basic metabolites, the anti-
cancer potential of which is being explored by many research
groups worldwide.[2] Thus, various studies have identified
promising anticancer effects exhibited by the isocarbostyrils
narciclasine and pancratistatin, as well as the pyrrolophenan-
thridine alkaloid lycorine (Figure 1). These natural products
have therefore been under intense scrutiny and are currently
being pursued as clinical candidates for cancer treatment.[3]
Although the ethano-bridged phenanthridine alkaloids of
the crinine type (Figure 2) have also been reported to display
cytotoxic[4] and apoptosis-inducing[5] properties, their anticanc-
er potential remains largely unexplored, and the mechanism(s)
underlying their effects on cancer cells have not been investi-
gated.
We recently reported the results of biochemical experiments
that provide mechanistic insight into the antiproliferative ef-
[a] G. Luchetti, C. Champion, Prof. I. V. Magedov, Prof. A. Kornienko
Department of Chemistry
New Mexico Institute of Mining and Technology, Socorro, NM 87801 (USA)
E-mail: akornien@nmt.edu
[b] R. Johnston, K. Hayden, M. R. Reisenauer, Prof. S. Rogelj
Department of Biology
New Mexico Institute of Mining and Technology, Socorro, NM 87801 (USA)
[c] Prof. V. Mathieu, Prof. F. Lefranc, Prof. R. Kiss
Laboratoire de Toxicologie, Facultꢀ de Pharmacie
Universitꢀ Libre de Bruxelles (ULB), 1050 Brussels (Belgium)
[d] Prof. F. Lefranc
Service de Neurochirurgie, Hꢁpital Erasme, Brussels (Belgium)
[e] Dr. A. Andolfi, Prof. A. Evidente
Dipartimento di Scienze del Suolo della Pianta dell’Ambiente e delle-
Produzioni Animali
Universitꢂ di Napoli Federico II, Via Universitꢂ 100, 80055 Portici (Italy)
[f] D. Lamoral-Theys
Laboratoire de Chimie BioAnalytique, Toxicologie et Chimie Physique-
Appliquꢀe
Facultꢀ de Pharmacie, Universitꢀ Libre de Bruxelles (ULB), 1050 Brussels
(Belgium)
[g] Dr. S. C. Pelly, Prof. W. A. L. van Otterlo
Figure 1. Structures of Amaryllidaceae anticancer constituents pursued as
clinical candidates for cancer treatment.
Department of Chemistry and Polymer Sciences, Stellenbosch University
Stellenbosch, Western Cape (South Africa)
ChemMedChem 2012, 7, 815 – 822
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
815

A. Kornienko et al.
MED
fects induced by narciclasine and lycorine.[3a,b,d] We showed
that at therapeutic concentrations, both narciclasine and lycor-
ine do not induce apoptosis in cancer cells, but rather exhibit
cytostatic effects by targeting the actin cytoskeleton. This
mode of action explains their promising activities against
cancer cells that are resistant to pro-apoptotic stimuli and
therefore representative of tumors associated with poor prog-
noses such as melanoma, glioblastoma, and non-small-cell
lung cancer. Herein we show that the crinine-type alkaloids ex-
hibit useful cytostatic properties as well, and that these effects
are most likely caused by targeting the actin cytoskeleton or-
ganization in cancer cells. These findings warrant further inves-
tigation of the crinine-type alkaloids as promising potential
agents for the treatment of apoptosis-resistant cancers, such
as glioblastoma.[6]
ture by NOESY (Figure 3). Finally, in terms of natural product
isolation, additional crinine-type alkaloids, namely haemantha-
mine, haemanthidine, buphanamine, buphanisine, and ambel-
line (Figure 2) were obtained with procedures used previously
by our research group as detailed below in the Experimental
Section.
Figure 3. NOE correlations in compound 1, confirming the stereochemical
relationships.
Results and Discussion
Isolation and structure confirmation
Synthesis
The alkaloid bulbispermine (Figure 2), previously isolated from
various plants of the Amaryllidaceae family,[7] has been report-
ed to have cytotoxic activity against HL-60 cells.[7c] We seren-
dipitously isolated a significant quantity of this alkaloid from
Zephyranthes robustus while investigating this plant as a poten-
tial source of pancratistatin.[8] Although this plant had not
been previously reported as a source of bulbispermine, it ap-
peared to contain significant quantities of this alkaloid
(70 mgkgÀ1 of fresh bulbs). However, because of problems
with plant availability, we subsequently switched to the known
source of bulbispermine, Crinum bulbispermum (160 mgkgÀ1 of
fresh bulbs).[7b] Although the NMR spectra of isolated bulbis-
permine were consistent with those reported previously,[9] we
were concerned with their similarity to the NMR spectroscopic
data of the stereoisomeric alkaloids, 11-hydroxyvittatine and
hamayne (Figure 2). Therefore, we obtained unambiguous
proof of stereochemistry by converting a portion of bulbisper-
mine to its diacetate (compound 1) and analyzing the struc-
After isolating sufficient quantities of bulbispermine, we per-
formed its chemical derivatization and subsequently obtained
a small group of analogues. Thus, the hydroxy groups in bul-
bispermine were acetylated and propionylated to give esters
1 and 2 (Scheme 1). The reaction with one equivalent of para-
bromobenzoyl chloride resulted in selective esterification of
the more sterically accessible allylic hydroxy at C3. Notably, the
other hydroxy group at C11 is located vicinal to the C10b qua-
ternary carbon center, thus facilitating this outcome. As these
derivatizations lead to the introduction of hydrophobic groups
on the two hydroxy groups, we next sought to derivatize them
as polar carbamates. This was done by first allowing bulbisper-
mine to react with trichloroacetyl isocyanate to give intermedi-
ate 4, which was then converted into dicarbamate 5 through
basic hydrolysis. Our attempts to selectively dihydroxylate the
double bond in diacetate 1 were unsuccessful, and we ob-
tained tetraacetate 6 as a mixture of diastereomers at C1 and
C2 (Scheme 1).
Additional bulbispermine derivatization (Scheme 2) involved
selective oxidation of the allylic alcohol to give C3-ketone 7,
hydrogenation of the double bond to provide C1,C2-saturated
analogue 8, and lastly quaternization of the nitrogen atom to
result in ammonium iodide salt 9.
Biochemical experiments
The synthesized compounds were next evaluated for in vitro
growth inhibition using the MTT colorimetric assay[3a,l] against
a panel of five cancer cell lines. This included two that are re-
sistant to pro-apoptotic stimuli: human U373 and T98G glio-
blastoma (GBM, from astroglial origin);[3a,10a,b] two apoptosis-
sensitive tumor models: human Hs683 anaplastic oligodendro-
glioma[3a,10b] and HeLa cervical adenocarcinoma; and an addi-
tional human U87 glioblastoma cell line, the apoptosis-resist-
ance properties of which have not yet been investigated.[10c]
Analysis of the antiproliferative data reveals that active com-
Figure 2. Structures of selected crinine-type alkaloids.
816
www.chemmedchem.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2012, 7, 815 – 822

Bulbispermine as a Cytostatic Agent
MED
ture for antiproliferative activity
in the crinine-type alkaloids. Of
the synthetic bulbispermine de-
rivatives, the only notable activi-
ty is observed for compound 5,
in which the C3,C11-hydroxy
groups are converted into polar
carbamates, demonstrating that
derivatization of the hydroxy
groups with hydrophobic moiet-
ies leads to an apparent loss of
activity. Also of interest is the im-
portance of the C1,C2-olefinic
functionality, because analogue
8, in which the double bond is
hydrogenated, is inactive. Finally,
the lack of anticancer activity for
analogues 7–9 is consistent with
previously published observa-
tions that similar modifications
to the lycorine skeleton result in
inactive compounds.[3d,11]
Scheme 1. Derivatization of bulbispermine hydroxy groups. Reagents and conditions: a) Ac2O, py, DMAP, RT, over-
night, 78%; b) (CH3CH2CO)2O, py, RT, overnight, 82%; c) p-Br-BzCl, py, DMAP, MeCN, RT, overnight, 48%; d) Cl3C(C=
O)N=C=O, CH2Cl2, 08C, 20 min, 75%; e) K2CO3, MeOH/H2O, RT, 3 h, 98%; f) 1. OsO4, NMO, acetone/H2O, RT, 6 h,
2. Ac2O, py, DMAP, RT, overnight, 59%.
We also made use of comput-
er-assisted phase-contrast mi-
croscopy (quantitative videomi-
croscopy) to analyze the principal mechanism of action associ-
ated with bulbispermine’s in vitro growth inhibitory effects, as
first revealed by the MTT colorimetric assay. Figure 4 shows
that bulbispermine inhibits cancer cell proliferation without in-
ducing cell death when assayed at its IC50 in vitro growth in-
hibitory value in U373 GBM cells or even at sevenfold its IC50
antiproliferative value in U87 GBM cells. Based on the phase-
contrast images obtained by quantitative videomicroscopy, we
calculated the global growth rate (GGR), which corresponds to
the ratio of the mean number of cells present in the last image
captured in the experiment (conducted at t=72 h) to the
number of cells present in the first image (at t=0 h). We divid-
ed this ratio obtained in the bulbispermine-treated experiment
by the ratio obtained in the control. Lycorine, which we estab-
lished earlier to be a cytostatic compound in these cell lines,[3d]
was used as a positive control. A GGR value of 0.6 in both cell
lines means that 60% of cells grew in the bulbispermine-treat-
ed experiment relative to the control over an observation
period of 72 h.
Scheme 2. Additional bulbispermine derivatization. Reagents and conditions:
a) Dess–Martin, PhCH3, 658C, 2 h, 38%; b) H2, Pd/C, MeOH, RT, overnight,
74%; c) MeI, MeCN, RT, 2 h, 97%.
pounds (bulbispermine, haemanthamine, haemanthidine, and
synthetic derivatives 1 and 5) do not discriminate between the
cancer cell lines based on the apoptosis sensitivity criterion
and display similar potencies in both cell types, indicating that
apoptosis induction is not the primary mechanism responsible
for antiproliferative activity in this series of compounds, at
least in solid cancers (Table 1). Furthermore, three of the natu-
ral alkaloids containing the a-C11,C12-ethano bridge (bulbis-
permine, haemanthamine, and haemanthidine) are all active, in
contrast to their counterparts incorporating this C11–C12 subu-
nit at the b position (buphanamine, buphanisine, and ambel-
line). These observations are consistent with findings reported
earlier,[4i,11] and they point to the criticality of this structural fea-
Our previous work with narciclasine[3a] and lycorine[3d] re-
vealed that their cytostatic effects on cancer cells occur mainly
during cytokinesis, through an increase in the rigidity of the
actin cytoskeleton. Our experiments with bulbispermine and
its active analogue 5 also revealed that at their in vitro growth
inhibitory IC50 concentrations, these compounds markedly in-
crease the levels of polymerized actin in Hs683 glioma cells
(green fluorescence in Figure 5Cb,Db) relative to control (Fig-
ure 5Ab). These effects of increased rigidity of the actin cytos-
keleton induced by bulbispermine and its synthetic analogue 5
thus closely resemble those of narciclasine (Figure 5Bb).
Correlations of the differential cellular sensitivities in the US
National Cancer Institute (NCI) 60-cell-line screen confirmed
ChemMedChem 2012, 7, 815 – 822
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
817

A. Kornienko et al.
MED
probably interacts with other lower-affinity targets, which are
not necessarily the same for rapamycin and pancratistatin.
Table 1. Evaluation of natural and synthetic crinine-type compounds
against cancer cells resistant or sensitive to pro-apoptotic stimuli.
Compd
GI50 [mm][a]
Sensitive
Computer modeling
Resistant
T98G U373
Unknown
U87
Hs683
HeLa
The similarity of the effects on cancer cells exhibited by the
previously investigated narciclasine and lycorine, as well as the
currently described crinine-type alkaloids, gives rise to an im-
portant question of whether the structural skeletons associated
with these three types of Amaryllidaceae constituents share
the same chemical space and are capable of targeting the
same intracellular macromolecules. For the purposes of investi-
gating the structural relationship between narciclasine, lycor-
ine, and bulbispermine (Figure 7), we undertook a conforma-
tional search to find the lowest-energy conformation for each
structure, and these were superimposed. It is clear that the
three structures have in common the hydrogen bond acceptor
methylenedioxy moiety, the aromatic region, and the position-
ing of the nitrogen atom. Interestingly, in all three compounds
this nitrogen would serve as a potential hydrogen bond donor
group, as in narciclasine it is found as the amide ÀNH, and in
lycorine and bulbispermine it will be in the protonated form at
physiological pH. Therefore, it is highly probable that these
three regions represent the common pharmacophore for the
three structures in their mode of action. Diversification in the
structures is tolerated in the hydroxy-containing portions of
the three compounds, indicating that this region is possibly di-
rected away from the putative binding site(s).
bulbispermine
haemanthamine
haemanthidine
buphanamine
buphanisine
ambelline
1
2
3
4
5
6
7
8
9
9
8
14
38
6
7
11
3
4
8
3
ND
9
6
6
>100
>100
>100
98
>100
>100
>100
91
>100
>100
>100
>100
>100
>100
>100
ND
ND
ND
ND
>100
ND
ND
ND
>100
>100
>100
63
>100
>100
>100
50
>100
>100
>100
>100
>100
>100
>100
90
>100
>100
>100
46
>100
>100
>100
>100
>100
>100
>100
74
>100
>100
>100
15
>100
>100
>100
>100
ND
[a] Values represent the average of two independent experiments per-
formed in vitro; ND: not determined.
this proposed mode of action for crinine-type alkaloids. Al-
though bulbispermine has not been tested at NCI, data are
available for its congener haemanthamine (Figure 2). Thus, the
“Compare Correlation Coefficients” indicate that at its mean
GI50 value (16 mm) characteristic of growth inhibition (Figure 6),
the differential cellular sensitivities for haemanthamine are sim-
ilar to those of known actin cytoskeleton modulators rapamy-
cin[12] and isocarbostyril pancratistatin[3b] (the correlation coeffi-
cients are 0.5 and 0.35, respectively). However, the uniformity
of the mechanism among these compounds disappears if the
correlations are performed at their mean LC50 values, character-
istic of cell death (no correlation obtained for 100 mm hae-
manthamine). Therefore, at cytotoxic doses, haemanthamine
Conclusions
Malignant gliomas, which include anaplastic astrocytoma, oli-
godendroglioma, oligoastrocytoma, and glioblastoma, account
for more than 50% of all primary brain tumors.[6] Of these, glio-
blastoma multiforme represents the highest grade of malig-
Figure 4. Cellular imaging of bulbispermine (bulb.) against A) U373 and B) U87 glioma cells (GGR=global growth rate), illustrating a non-cytotoxic yet cyto-
static, antiproliferative mechanism at MTT colorimetric assay-related IC50 (U373) as well as 7ꢂIC50 (U87) concentrations.
818
www.chemmedchem.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2012, 7, 815 – 822

Bulbispermine as a Cytostatic Agent
MED
Figure 7. Superimposition of narciclasine (green), lycorine (orange), and bul-
bispermine (purple), showing high similarity in the methylenedioxy, aromat-
ic, and nitrogen portions of the structures.
nancy and is associated with dismal prognoses. Glioblastoma
cells display resistance to apoptosis, which contributes to their
poor response to conventional chemotherapy with pro-apop-
totic agents.[6] In our previous work we identified the Amarylli-
daceae small-molecule constituents narciclasine and lycorine
as promising anticancer agents that exhibit useful antiprolifera-
tive effects toward glioblastoma cells resistant to pro-apoptotic
stimuli. Herein we demonstrate that another group of Amarylli-
daceae constituents, namely crinine-type alkaloids, are also po-
tentially useful drug leads for the treatment of apoptosis-resist-
ant cancers and glioblastoma in particular. We show that a rep-
resentative crinine-type alkaloid bulbispermine inhibits the
proliferation of glioblastoma cells through cytostatic effects
possibly arising from the rigidification of the actin cytoskele-
ton. Computer-assisted superimposition of the lowest-energy
conformations of narciclasine, lycorine, and bulbispermine indi-
cates that these three molecules, representative of three major
Amaryllidaceae anticancer scaffolds, occupy similar regions in
chemical space and they indeed could be targeting the same
intracellular macromolecules to induce cytostatic effects in
cancer cells.
Figure 5. Effects on actin cytoskeleton organization by bulbispermine and
analogue 5 at their in vitro growth inhibitory (IC50) concentrations in Hs683
glioma cells. A) control (untreated) conditions (Aa: bright field; Ab: actin cy-
toskeleton). B) 100 nm positive control narciclasine-treated conditions (Ba:
bright field; Bb: actin cytoskeleton). C) 10 mm bulbispermine-treated condi-
tions (Ca: bright field; Cb: actin cytoskeleton). D) 50 mm analogue-5-treated
conditions (Da: bright field; Db: actin cytoskeleton). Narciclasine (Bb), bulbis-
permine (Cb), and analogue 5 (Db) rigidify the actin cytoskeleton by increas-
ing the amount of polymerized actin (white arrows).
Experimental Section
Chemistry
General: All commercially obtained reagents were used without
purification unless otherwise noted. THF and toluene were distilled
from sodium benzophenone-ketyl prior to use. CH2Cl2 was distilled
from CaH2 and kept under Ar. Reactions were monitored by TLC
(EM Science, Silica Gel 60 F254, 250 mm) and visualized with UV light
or I2. When appropriate, reactions were run under N2 or Ar in oven-
dried glassware using standard syringe, cannula, and septa tech-
niques. Flash chromatography was performed on silica gel (Silica
gel, 32–63 mm, 60 ꢃ pore size). 1H and 13C NMR spectra were re-
corded on JEOL 300 MHz and Bruker 400 MHz spectrometers.
HRMS was performed at the University of New Mexico mass spec-
trometry facility.
Figure 6. Correlations of the differential cellular sensitivities in the NCI 60-
cell-line screen using the COMPARE algorithm. “Compare Correlation Coeffi-
cients” were generated by a computerized pattern-recognition algorithm
and serve as an indication of similarities in differential cellular sensitivities or
characteristic fingerprints for each compound. Haemanthamine was used as
a seed to find significant correlations with the anticancer agents in the NCI
Standard Compound Database, containing rapamycin and pancratistatin at
the GI50, TGI, and LC50 levels (for definitions of these parameters, see DTP
Human Tumor Cell Line Screen: http://dtp.nci.nih.gov/branches/btb/
ivclsp.html; accessed February 16, 2012). At the GI50 level (16 mm), the corre-
*
*
lations identified rapamycin ( ) and pancratistatin ( ) as highly ranked
agents among the compounds in the database with Compare Correlation
Coefficients of 0.5 and 0.35, respectively. Whereas correlations at the TGI
level (80 mm) were worse but still significant (Compare Correlation Coeffi-
cients of 0.37 and 0.34), the LC50 levels (100 mm) were not correlated at all.
Alkaloids: Bulbispermine was isolated from dried bulbs of Crinum
bulbispermum as previously reported.[7b] Haemanthamine was iso-
lated from Egyptian Pancratium maritinum.[13] Ambelline, buphana-
ChemMedChem 2012, 7, 815 – 822
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
819

A. Kornienko et al.
MED
1
mine, and buphanisine were generously supplied by Prof. Henry M.
Fales, Department of Health, Education and Welfare, Bethesda, MD
(USA). Haemanthidine was purified from dried bulbs of Lycoris
aurea.[14] The purity of the samples was confirmed by TLC, mp, op-
tical rotation, 1H and 13C NMR, NOESY, and ESIMS analyses.
nish 8.6 mg of mono para-bromobenzoate (48%). H NMR (CDCl3,
300 MHz): d=8.94 (d, 2H, J=5.1 Hz, PhH), 8.59 (d, 2H, J=5.1 Hz,
PhH), 6.82 (s, 1H, H-10), 6.51 (s, 1H, H-7), 6.39 (d, 1H, J=12.0 Hz,
H-1), 6.24 (d, 1H, J=12.0 Hz, H-2), 5.94 (s, 2H, -OCH2O-), 5.75–5.66
(m, 1H, H-3), 4.36 (d, 1H, J=18.0 Hz, H-6b), 4.09–4.05 (m, 1H, H-
11), 3.71 (d, 1H, J=18.0 Hz, H-6a), 3.49–3.34 (m, 2H, H-12), 2.46–
2.32 (m, 2H, H-4), 2.25–2.16 ppm (m, H-4); 13C NMR (CDCl3,
300 MHz): d=165.6, 146.7, 146.5, 134.1, 131.8, 131.4, 126.7, 125.4,
107.1, 103.3, 101.1, 80.3, 71.3, 66.1, 63.6, 61.4, 50.3, 31.1, 30.0 ppm;
HRMS m/z (ESI+) calcd for C23H20BrNO5 [M+H]+ 472.0583, found
472.0568.
1
Bulbispermine: H NMR (CD3OD, 300 MHz): d=6.85 (s, 1H, H-10),
6.52 (s, 1H, H-7), 6.23 (dd, 1H, J=9.1 Hz, 2.2 Hz, H-2), 6.04 (d, 1H,
9.1 Hz, H-1), 5.87 (s, 2H, -OCH2O-), 4.36–4.29 (m, 1H, H-3), 4.27 (d,
1H, J=16.7 Hz, H-6b), 3.95 (dd, 1H, J=8.3 Hz, 6.0 Hz, H-11), 3.73
(d, 1H, J=16.7 Hz, H-6a), 3.44 (1H, dd, J=13.6 Hz, 7.1 Hz, H-4a),
3.25–3.17 (m, 2H, H-12), 2.15–2.03 (m, 1H, H-4), 1.98–1.91 ppm (m,
1H, H-4); 13C NMR (CD3OD, 400 MHz): d=148.5 (C9), 148.0 (C8),
137.4 (C10a), 136.7 (C1), 125.2 (C6a), 124.2 (C2), 107.9 (C10), 104.4
(C7), 102.4 (-OCH2O-), 80.4 (C11), 68.0 (C3), 67.7 (C4a), 63.3 (C12),
60.9 (C6), 51.6 (C10b), 33.8 ppm (C4).
Compound 4: To
a
solution of bulbispermine (5.0 mg,
0.0174 mmol) in CH2Cl2 (0.174 mL) cooled to 08C was added tri-
chloroacetyl isocyanate (4.1 mL, 0.0348 mmol). The reaction was
stirred at 08C for 20 min, and the CH2Cl2 was evaporated under re-
duced pressure. The crude material was purified by PTLC using
10% MeOH in CH2Cl2 resulting in 8.4 mg imidocarbamate (75%).
1H NMR (CD3OD, 400 MHz): d=7.03 (s, 1H, H-10), 6.67 (s, 1H, H-7),
6.46 (d, 1H, J=10.4 Hz, H-1), 6.15 (d, 2H, J=10.4 Hz, H-2), 5.97 (s,
-OCH2O-), 5.54–5.52 (m, 1H, H-3), 5.13–5.11 (m, 1H, H-11), 4.56 (d,
1H, J=15.9 Hz, H-6b), 4.05 (d, 1H, J=15.9 Hz, H-6a), 3.83–3.67 (m,
3H), 2.42–2.34 ppm (m, 2H, H-4); HRMS m/z (ESI+) calcd for
C18H20N3O6 (hydrolysis product [M+H]+) 374.1352, found 374.1346.
Compound 1: To
a
solution of bulbispermine (10.0 mg,
0.0348 mmol) in pyridine (3.0 mL) was added Ac2O (2.0 mL) and
a catalytic amount of DMAP (0.1 mg, 0.9 mmol). The reaction was
stirred at room temperature overnight, and the following day was
co-evaporated several times with toluene to remove pyridine. The
resulting residue was purified by preparative thin-layer chromatog-
raphy (PTLC) using CH2Cl2/MeOH (98:2), yielding 5.1 mg of the di-
acetate (78%). 1H NMR (CDCl3, 300 MHz): d=6.86 (s, 1H, H-10),
6.47 (s, 1H, H-7), 6.22 (dd, 1H, J=10.4 Hz, 2.2 Hz, H-1), 5.91 (s, 2H,
-OCH2O-), 5.86 (d, 1H, J=10.4 Hz, H-2), 5.48–5.42 (m, H-3b), 4.99–
4.31 (m, 1H, H-11exo), 3.34 (d, 1H, J=17.0 Hz, H-6b), 3.71 (d, 1H,
J=17.0 Hz, H-6a), 3.40 (d, 2H, J=5.2 Hz, H-12), 3.28 (dd, 1H, J=
13.5 Hz, H-4a), 2.12–2.17 (m, 2H, H-4), 2.09 (s, -OAc), 2.04 ppm (s,
OAc); 13C NMR (CDCl3, 300 MHz): d=170.6 (C=O), 170.3 (C=O),
146.9 (C9), 146.8 (C8), 134.0 (C10a), 131.4 (C1), 126.3 (C6a), 125.6
(C2), 106.8 (C10), 103.9 (C7), 101.1 (-OCH2O-), 80.4 (C11), 70.0 (C3),
66.2 (C4a), 61.1 (C6), 49.4 (C10b), 29.8 (C4), 21.4 (CH3), 21.3 ppm
(CH3); HRMS m/z (ESI+) calcd for C20H22NO6 [M+H]+ 372.1447,
found 372.1449.
Compound 5: To
a
solution of bulbispermine (5.0 mg,
0.0174 mmol) in CH2Cl2 (0.174 mL) cooled to 08C was added tri-
chloroacetyl isocyanate (4.1 mL, 0.0348 mmol). After being stirred at
08C for 20 min, the CH2Cl2 was evaporated. The resulting residue
was dissolved in MeOH (0.1 mL) and cooled to 08C. H2O (0.5 mL)
and K2CO3 (0.19 g, 1.38 mmol) was added, and the cooling bath
was removed. After stirring at room temperature for 3 h, MeOH
was evaporated, and the resulting aqueous layer was extracted
with CH2Cl2. The combined organic extracts were dried (Na2SO4)
and concentrated to afford 4.7 mg of the dicarbamate (74%).
1H NMR (CDCl3, 400 MHz): d=6.95 (s, 1H, H-10), 6.59 (s, 1H, H-7),
6.35 (d, 1H, J=10.8 Hz, H-1), 5.97 (d, 1H, J=10.8 Hz, H-2), 5.96 (s,
2H, -OCH2O-), 5.32–5.28 (m, 1H, H-3), 4.39 (d, 1H, J=16.5 Hz, H-
6b), 3.85 (d, 1H, J=16.5, H-6a), 3.67 (s, 2H, H-12), 3.56 (dd,1H, J=
14.2 Hz, 7.7 Hz, H-4a), 3.45–3.40 (m, 1H, H-11), 2.15–2.07 ppm (m,
2H, H-4); HRMS m/z (ESI+) calcd for C18H20N3O6 [M+H]+ 374.1352,
found 374.1352.
Compound 2: To
a
solution of bulbispermine (5.0 mg,
0.0174 mmol) in pyridine (1.5 mL) was added propionic anhydride
(1.0 mL). The reaction mixture was stirred at room temperature
overnight, and the following day was co-evaporated with toluene
several times to remove pyridine. The resulting residue was puri-
fied by PTLC using CH2Cl2/MeOH (98:2), yielding 5.6 mg of the di-
1
propionate (82%). H NMR (CDCl3, 400 MHz): d=6.89 (s, 1H, H-10),
Compound 6: Diacetate 1 (5.0 mg, 0.0174 mmol) was dissolved in
acetone (6 mL) and a solution of OsO4 (0.25 mL, 2.5% wt solution
in tBuOH) and NMO (0.038 g, 0.284 mmol) in H2O (1.5 mL) were
added over 5 min. The solution was stirred at room temperature
for 6 h after which the solvents were evaporated under reduced
pressure. To the obtained slurry were added EtOAc (2 mL) and sa-
turated NH4Cl (1 mL), followed by washing with saturated NaCl
(1 mL). The extraction was repeated twice with EtOAc (2 mL), and
the combined organic layers were dried with MgSO4. The solvent
was removed under reduced pressure and subjected to PTLC using
CH2Cl2/MeOH (4:1), yielding an inseparable mixture of diastereo-
mers. The mixture was subjected to standard acetylation condi-
tions (2.2 equiv Ac2O, pyridine, DMAP) in an attempt to resolve the
mixture which yielded 4.9 mg (59%) of another inseparable mix-
ture of the diastereomeric tetraacetates 6 that were biologically
evaluated as a mixture; HRMS m/z (ESI+) calcd for C24H28NO10 [M+
H]+ 490.1713, found 490.1704.
5.50 (s, 1H, H-7), 6.23 (d, 1H, J=10.4 Hz, H-1), 5.93 (s, 2H, -OCH2O-
), 5.87 (d, 1H, J=10.4 Hz, H-2), 5.51–5.47 (m, 1H, H-3), 5.03–5.00
(m, 1H, H-11), 4.39 (d, 1H, J=17.2 Hz, H-6b), 3.77 (d, 1H, J=
17.2 Hz, H-6a), 3.46–3.44 (m, 2H, H-12), 3.33 (dd, 1H, J=13.7 Hz,
3.04 Hz, H-4a), 2.40–2.31 (m, 4H, -CH2CO-), 2.23–2.19 (m, 1H), 1.19–
1.14 ppm (m, 6H, -CH3); 13C NMR (CDCl3, 400 MHz): d=174.0 (C=O),
173.6 (C=O), 147.0 (C9), 146.8 (C8), 131.5 (C10a), 125.3 (C2), 106.8
(C10), 104.0 (C7), 101.1 (-OCH2O-), 79.8 (C11), 69.6 (C3), 66.1 (C4a),
60.9 (C12), 60.2 (C6), 49.4 (C10b), 29.6 (C4), 28.0, 27.9, 9.2, 9.1 ppm;
HRMS m/z (ESI+) calcd for C22H25NO6 [M+H]+ 400.1760, found
400.1761.
Compound 3: A solution of para-bromobenzoyl chloride in MeCN
(0.5 mL, 0.0163 mmol) was added dropwise to a solution of bulbis-
permine (10.0 mg, 0.0384 mmol) in pyridine (0.5 mL). The reaction
mixture was stirred at room temperature overnight, and the result-
ing solution was diluted with a 5% (w/v) solution of NaHCO3
(1 mL). The aqueous solution was extracted from with EtOAc (3ꢂ
1 mL), and the organic layers were combined and dried with
MgSO4. After removal of the solvent, the residue was purified
using PTLC and with the eluent system CH2Cl2/MeOH (95:5) to fur-
Compound 7: An oven-dried flask was charged with bulbispermine
(10.0 mg, 0.0348 mmol) and Dess–Martin periodinane (0.044 g,
0.103 mmol) and was subjected to several cycles of vacuum-argon
purge. Dry toluene (2 mL) was added, and the reaction suspension
820
www.chemmedchem.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2012, 7, 815 – 822

Bulbispermine as a Cytostatic Agent
MED
was heated at 658C for 2 h until the substrate was completely con-
sumed. The reaction mixture was concentrated under reduced
pressure, and the resulting residue was purified by PTLC using
CH2Cl2/MeOH (9:1), yielding 3.7 mg of the allylic ketone (38%).
1H NMR (CD3OD, 400 MHz): d=7.69 (d, 1H, J=10.2 Hz, H-2), 7.11 (s,
1H, H-10), 6.69 (s, 1H, H-7), 6.24 (d, 1H, J=10.2 Hz, H-1), 5.97 (d,
2H, J=5.4 Hz, -OCH2O-), 4.58 (d, 1H, J=17.1 Hz, H-6b), 3.98 (d, 1H,
J=17.1 Hz, H-6a), 3.91 (dd, 1H, J=13.2 Hz, 6.3 Hz, H-4a), 3.80 (d,
1H, J=18.6 Hz), 3.51 (d, 1H, J=18.6 Hz), 3.25 (d, 1H, J=11.6 Hz),
2.76–2.61 ppm (m, 2H, H-4). 13C NMR (CD3OD, 400 MHz): d=206.5
(C3), 197.0 (C9), 148.1 (C8), 143.7 (C10a), 130.7 (C1), 126.9 (C6a),
125.7 (C2), 107.0 (C10), 103.3 (-OCH2O-), 101.5 (C11), 66.0 (C4a),
58.9 (C12), 58.3 (C6), 48.5 (C10b), 39.4 ppm (C4); HRMS m/z (ESI+)
calcd for C16H16NO4 [M+H]+ 286.1079, found 286.1046.
l 630 nm). One set of experimental conditions included six repli-
cates. Each experiment was carried out twice.
Cell proliferation was visualized by computer-assisted phase-con-
trast microscopy.[3a,d,15] The fibrillar actin cytoskeleton organization
was determined by computer-assisted fluorescence microscopy.[3a,d]
Phallacidin conjugated with the green-fluorescent Alexa Fluor488
(Molecular Probes Inc., Eugene, OR, USA) was used to label the fi-
brillar actin.
Computer modeling
Conformations of narciclasine, lycorine, and bulbispermine were
generated using Accelrys Discovery Studio 3.1. All conformers were
minimized, and structures with a RMSD>0.2 were retained. For
the purposes of the overlay, the lowest-energy conformation was
selected for each structure, and superimposition using tethers was
employed to bring into alignment the dioxolane and aromatic re-
gions of the structures.
Compound 8: A solution of bulbispermine (10.0 mg, 0.0348 mmol)
in MeOH (4.2 mL) was charged with Pd/C 5% (11.5 mg) and
purged with an H2 balloon. The stirred reaction mixture was kept
under an H2 atmosphere overnight at room temperature, after
which the solution was filtered through Celite and the solvent
evaporated under reduced pressure. The resulting white solid was
purified by PTLC using CH2Cl2/MeOH (9:1) to yield 7.4 mg of re-
duced product (74%). 1H NMR (CD3OD, 400 MHz): d=6.76 (s, 1H,
H-10), 6.51 (s, 1H, H-7), 5.89 (s, 2H, -OCH2O-), 4.30 (d, 1H, J=
16.7 Hz, H-6b), 4.04–4.03 (m, 1H, H-3), 3.75 (d, 1H, J=16.7 Hz, H-
6a), 3.67–3.61 (m, 1H, H-11), 3.40–3.39 (m, 1H, H-4a), 3.24 (dd, 1H,
J=11.3 Hz, 5.1 Hz, H-2), 3.09 (dd, 1H, J=12.8 Hz, 4.84 Hz, H-2), 2.68
(dd, 1H, J=14.0 Hz, 4.0 Hz, H-1), 2.12–2.03 (m, 2H, H-4), 1.99–
1.78 ppm (m, 3H); 13C NMR (CD3OD, 400 MHz): d=146.9 (C8), 146.2
(C9), 138.7 (C10a), 124.6 (C6a), 105.7 (C7), 103.2 (C10), 100.8
(-OCH2O-), 81.2 (C11), 68.6 (C3), 66.9 (C4a), 62.0 (C12), 59.8 (C6),
45.8 (C10b), 35.4 (C4), 32.1 (C2), 25.6 ppm (C1); HRMS m/z (ESI+)
calcd for C16H20NO4 [M+H]+ 290.1392, found 290.1399.
Acknowledgements
This work is supported by the US National Institutes of Health
(NIH) (RR-16480 and CA-135579) under the BRIN/INBRE and AREA
programs. R.K. is a director of research with the Fonds National
de la Recherche Scientifique (FNRS, Belgium). Grants from the
Italian Ministry of University and Research are gratefully acknowl-
edged. This work was also supported by the National Research
Foundation (Pretoria, South Africa), and faculty and departmen-
tal funding from Stellenbosch University (Stellenbosch, South
Africa). R.J. and S.R. thank Dr. Michaelann Tartis (New Mexico In-
stitute of Mining and Technology) for her advice.
Compound 9: To
a
solution of bulbispermine (5.0 mg,
0.0174 mmol) in dry MeCN (0.3 mL) was added 1.1 equiv MeI. After
5 min, the reaction turned milky, and was reacted at room temper-
ature until starting material had been totally consumed after 2 h.
The resulting precipitate was filtered and washed with cold MeCN,
followed by CHCl3, resulting in 7.3 mg of a white solid ammonium
Keywords: actin cytoskeletons
· alkaloids · apoptosis ·
bulbispermines · cancer · glioblastomas
1
salt (97%). H NMR (CD3OD, 400 MHz): d=7.02 (s, 1H, H-10), 6.71
(s, H-7), 6.23–6.17 (m, 2H, H-1, H-2), 6.01 (s, 2H, -OCH2O-), 4.69 (d,
1H, J=15.8 Hz, H-6a), 4.62 (s, 1H), 4.44–4.39 (m, 1H, H-3), 3.79–
3.75 (m, 2H), 3.35 (s, 3H, N-CH3), 2.50–2.40 ppm (m, 2H, H-4);
HRMS m/z (ESI+) calcd for C17H20NO4 [M]+ 302.1392, found
302.1385.
[1] a) J. B. Gardeil (transl. ), Traduction des Oeuvres Mꢀdicales d’Hippocrate,
Vol. 4, Fages, Meilhec et Cie, Toulouse, 1801; b) Z. Jin, Nat. Prod. Rep.
2011, 28, 1126.
[2] A. Kornienko, A. Evidente, Chem. Rev. 2008, 108, 1982.
[3] a) L. Ingrassia, F. Lefranc, J. Dewelle, L. Pottier, V. Mathieu, S. Spiegl-Krei-
necker, S. Sauvage, M. El Yazidi, M. Dehoux, W. Berger, E. Van Quaque-
beke, R. Kiss, J. Med. Chem. 2009, 52, 1100; b) G. Van Goietsenoven, J.
Hutton, J. P. Becker, B. Lallemand, F. Robert, F. Lefranc, C. Pirker, G. Van-
denbussche, P. Van Antwerpen, A. Evidente, W. Berger, M. Prꢀvost, J. Pel-
letier, R. Kiss, T. G. Kinzy, A. Kornienko, V. Mathieu, FASEB J. 2010, 24,
4575; c) F. Lefranc, S. Sauvage, G. Van Goietsenoven, V. Mꢀgalizzi, D. La-
moral-Theys, O. Debeir, S. Spiegl-Kreinecker, W. Berger, V. Mathieu, C.
Decaestecker, R. Kiss, Mol. Cancer Ther. 2009, 8, 1739; d) D. Lamoral-
Theys, A. Andolfi, G. Van Goietsenoven, A. Cimmino, B. Le Calvꢀ, N. Wau-
thoz, V. Mꢀgalizzi, T. Gras, C. Bruyꢀre, J. Dubois, V. Mathieu, A. Kornien-
ko, R. Kiss, A. Evidente, J. Med. Chem. 2009, 52, 6244; e) N. Evdokimov,
D. Lamoral-Theys, V. Mathieu, A. Andolfi, S. C. Pelly, W. A. L. van Otterlo,
I. V. Magedov, R. Kiss, A. Evidente, A. Kornienko, Bioorg. Med. Chem.
2011, 19, 7252; f) G. R. Pettit, B. Orr, S. Ducki, Anti-Cancer Drug Des.
2000, 15, 389; g) G. R. Pettit, N. Melody, M. Simpson, M. Thompson,
D. L. Herald, J. C. Knight, J. Nat. Prod. 2003, 66, 92; h) J. McNulty, A.
Thorat, N. Vurgun, J. J. Nair, E. Makaji, D. J. Crankshaw, A. C. Holloway, S.
Pandey, J. Nat. Prod. 2011, 74, 106; i) A. McLachlan, N. Kekre, J. McNulty,
S. Pandey, Apoptosis 2005, 10, 619; j) S. Vshyvenko, J. Scattolon, T. Hud-
licky, A. E. Romero, A. Kornienko, Bioorg. Med. Chem. Lett. 2011, 21,
4750; k) J. Collins, U. Rinner, M. Moser, T. Hudlicky, I. Ghiviriga, A. E.
Biochemical experiments
The overall growth level of human cancer cell lines was deter-
mined using the colorimetric MTT (3-[4,5-dimethylthiazol-2-yl]di-
phenyltetrazolium bromide; Sigma, Belgium) assay.[3a,d] Briefly, the
cell lines were incubated for 24 h in 96-microwell plates (at a con-
centration of 10000–40000 cells per mL culture medium depend-
ing on the cell type) to ensure adequate plating prior to cell
growth determination. The assessment of cell population growth
by means of the MTT colorimetric assay is based on the capacity of
living cells to convert the yellow compound MTT into a blue prod-
uct, formazan, by a reduction reaction occurring in the mitochon-
dria. The number of living cells after 72 h culture in the presence
(or absence: control) of the various compounds is directly propor-
tional to the intensity of the blue, which is quantitatively measured
by spectrophotometry—in our case with a Model 680XR instru-
ment (Bio-Rad, Nazareth, Belgium)—at l 570 nm (with reference at
ChemMedChem 2012, 7, 815 – 822
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
821

A. Kornienko et al.
MED
Romero, A. Kornienko, D. Ma, C. Griffin, S. Pandey, J. Org. Chem. 2010,
75, 3069; l) T. Mosmann, J. Immunol. Methods 1983, 65, 55.
[8] G. R. Pettit, V. Gaddamidi, G. M. Cragg, J. Nat. Prod. 1984, 47, 1018.
[9] E. E. Elgorashi, S. E. Drewes, J. van Staden, Phytochemistry 2001, 56, 637.
[10] a) N. Belot, S. Rorive, I. Doyen, F. Lefranc, E. Bruyneel, R. Dedecker, S.
Micik, J. Brotchi, C. Decaestecker, I. Salmon, R. Kiss, I. Camby, Glia 2001,
36, 375; b) F. Branle, F. Lefranc, I. Camby, J. Jeuken, A. Geurts-Moespot,
S. Sprenger, F. Sweep, R. Kiss, Cancer 2002, 95, 641; c) B. Le Calvꢀ, M.
Rynkowski, M. Le Mercier, C. Bruyꢄre, C. Lonez, T. Gras, B. Haibe-Kains,
G. Bontempi, C. Decaestecker, J. M. Ruysschaert, R. Kiss, F. Lefranc, Neo-
plasia 2010, 12, 727; d) I. Camby, C. Decaestecker, F. Lefranc, H. Kaltner,
H. J. Gabius, R. Kiss, Biochem. Biophys. Res. Commun. 2005, 335, 27.
[11] a) A. Evidente, M. R. Cicala, G. Randazzo, R. Riccio, G. Calabrese, R. Liso,
O. Arrigoni, Phytochemistry 1983, 22, 2193; b) A. Evidente, O. Arrigoni, R.
Liso, G. Calabrese, G. Randazzo, Phytochemistry 1986, 25, 2739.
[12] D. K. Sandsmark, H. Zhang, B. Hegedus, C. L. Pelletier, J. D. Weber, D. H.
Gutmann, Cancer Res. 2007, 67, 4790.
[13] A. H. Abou-Donia, A. A. Azim, A. S. El Din, A. Evidente, M. Gaber, A,
Scopa, Phytochemistry 1992, 31, 2139.
[14] Y. Yang, S. X. Huang, Y. M. Zhao, Q. S. Zhao, H. D. Sun, Helv. Chim. Acta
2005, 88, 2550.
[15] a) N. Belot, R. Pochet, C. W. Heizmann, R. Kiss, C. Decaestecker, Biochim.
Biophys. Acta Proteins Proteomics 2002, 1600, 74; b) O. Debeir, P.
Van Ham, R. Kiss, C. Decaestecker, IEEE Trans. Med. Imaging 2005, 24,
697.
[4] a) Y. H. Kim, E. Y. Park, M. H. Park, U. Badarch, G. M. Woldemichael, J. A.
Beutler, Biol. Pharm. Bull. 2006, 29, 2140; b) A. Evidente, A. S. Kireev,
A. R. Jenkins, A. E. Romero, W. F. A. Steelant, S. Van Slambrouck, A. Kor-
nienko, Planta Med. 2009, 75, 501; c) L. Z. Lin, S. F. Hu, H. B. Chai, T. Pen-
gsuparp, J. M. Pezzuto, G. A. Cordell, N. Ruangrungsi, Phytochemistry
1995, 40, 1295; d) J. Hohmann, P. Forgo, J. Molnar, K. Wolfard, A.
Molnar, T. Thalhammer, I. Mathe, D. Sharples, Planta Med. 2001, 68, 454;
e) E. Furusawa, H. Irie, D. Combs, W. C. Wildman, Chemotherapy 1980,
26, 36; f) B. Weniger, L. Italiano, J. P. Beck, J. Bastida, S. Bergonon, C.
Codina, A. Lobstein, R. Anton, Planta Med. 1995, 61, 77; g) M. A. Abd Eil
Hafiz, M. A. Ramadan, M. L. Jung, J. P. Beck, R. Anton, Planta Med. 1991,
57, 437; h) K. Likhitwitayawuid, C. K. Angerhofer, H. Chai, J. M. Pezzuto,
G. A. Cordell, J. Nat. Prod. 1993, 56, 1331; i) G. Van Goietsenoven, A. An-
dolfi, B. Lallemand, A. Cimmino, D. Lamoral-Theys, T. Gras, A. Abou-
Donia, J. Dubois, F. Lefranc, V. Mathieu, A. Kornienko, R. Kiss, A. Evi-
dente, J. Nat. Prod. 2010, 73, 1223; j) A. F. S. Silva, J. P. de Andrade,
K. R. B. Machado, A. B. Rocha, M. A. Apel, M. E. G. Sobral, A. T. Henriques,
J. A. S. Zuanazzi, Phytomedicine 2008, 15, 882.
[5] a) C. Griffin, N. Sharda, D. Sood, J. Nair, J. McNulty, S. Pandey, Cancer Cell
Int. 2007, 7, 10; b) J. McNulty, J. J. Nair, C. Codina, J. Bastida, S. Pandey,
J. Gerasimoff, C. Griffin, Phytochemistry 2007, 68, 1068.
[6] F. Lefranc, J. Brotchi, R. Kiss, J. Clin. Oncol. 2005, 23, 2411.
[7] a) O. B. Abdel-Halim, T. Morikawa, S. Ando, H. Matsuda, M. Yoshikawa, J.
Nat. Prod. 2004, 67, 1119; b) E. E. Elgorashi, S. E. Drewes, J. van Staden,
Phytochemistry 1999, 52, 533; c) M. Jitsuno, A. Yokosuka, H. Sakagami, Y.
Mimaki, Chem. Pharm. Bull. 2009, 57, 1153.
Received: December 29, 2011
Revised: February 12, 2012
Published online on March 2, 2012
822
www.chemmedchem.org
ꢁ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2012, 7, 815 – 822