Structure-activity relationship analysis of novel derivatives of narciclasine (an Amaryllidaceae isocarbostyril derivative) as potential anticancer agents

Article abstract of DOI:10.1021/jm8013585

Narciclasine (1) is a plant growth regulator that has been previously demonstrated to be proapoptotic to cancer cells at high concentrations (≥ 1 μM). Data generated in the present study show that narciclasine displays potent antitumor effects in apoptosis-resistant as well as in apoptosis-sensitive cancer cells by impairing the organization of the actin cytoskeleton in cancer cells at concentrations that are not cytotoxic (IC ₅₀ values of 30-90 nM). The current study further revealed that any chemical modification to the narciclasine backbone generally led to compounds of variable stability, weaker activity, or even the complete loss of antiproliferative effects in vitro. However, one hemisynthetic derivative of narciclasine, compound 7k, demonstrated by both the intravenous and oral routes higher in vivo antitumor activity in human orthotopic glioma models in mice when compared to narciclasine at nontoxic doses. Narciclasine and compound 7k may therefore be of potential use to combat brain tumors.

Full text of DOI:10.1021/jm8013585

1100
J. Med. Chem. 2009, 52, 1100–1114
Structure-Activity Relationship Analysis of Novel Derivatives of Narciclasine
(an Amaryllidaceae Isocarbostyril Derivative) as Potential Anticancer Agents
Laurent Ingrassia,†,‡ Florence Lefranc,†,§,|, Janique Dewelle,‡ Laurent Pottier,| Ve´ronique Mathieu,| Sabine Spiegl-Kreinecker,
Se´bastien Sauvage,‡ Mohamed El Yazidi,‡ Mischae¨l Dehoux,‡ Walter Berger,# Eric Van Quaquebeke,‡ and Robert Kiss*,|,
Unibioscreen SA, 40 AVenue Joseph Wybran, 1070 Brussels, Belgium, SerVice de Neurochirurgie, Cliniques UniVersitaires de Bruxelles,
Hoˆpital Erasme, UniVersite´ Libre de Bruxelles, Brussels, Belgium, Laboratoire de Toxicologie, Institut de Pharmacie, UniVersite´ Libre de
Bruxelles, Campus de la Plaine, CP205/1, BouleVard du Triomphe, 1050 Brussels, Belgium, Department of Neurosurgery, Wagner Jauregg
Hospital, Linz, Austria, Institute of Cancer Research, Department of Medicine I, Medical UniVersity of Vienna, Austria
ReceiVed October 27, 2008
Narciclasine (1) is a plant growth regulator that has been previously demonstrated to be proapoptotic to
cancer cells at high concentrations (g1 µM). Data generated in the present study show that narciclasine
displays potent antitumor effects in apoptosis-resistant as well as in apoptosis-sensitive cancer cells by
impairing the organization of the actin cytoskeleton in cancer cells at concentrations that are not cytotoxic
(IC50 values of 30-90 nM). The current study further revealed that any chemical modification to the
narciclasine backbone generally led to compounds of variable stability, weaker activity, or even the complete
loss of antiproliferative effects in vitro. However, one hemisynthetic derivative of narciclasine, compound
7k, demonstrated by both the intravenous and oral routes higher in vivo antitumor activity in human orthotopic
glioma models in mice when compared to narciclasine at nontoxic doses. Narciclasine and compound 7k
may therefore be of potential use to combat brain tumors.
Introduction
human diseases.11,12 This is especially true in the cancer field,
where about half the drugs are of natural origin13,14 and include
two of the most important anticancer agents taxol and taxotere.14
The antitumor potential of Amaryllidaceae isocarbostyril deriva-
tives, the most widely known of which are narciclasine (1),
lycoricidine, and pancratistatin (Figure 1), has long been studied
and has been recently reviewed.15,16 Extracts of Narcissus bulbs
have been known for millennia from folk medicine for their
antitumor effects. The powerful anticancer properties of Narcis-
sus poeticus L. were already known to Hippokrates of Kos (ca.
460-370 BCE), who recommended narcissus oil for the
treatment of uterine tumors.16 His successors, the ancient Greek
physicians Pedanius Dioscorides (ca. 40-90 AD) and Scranus
of Ephesus (98-138 AD), used this therapy to treat various
types of cancer.16 In 1877, a first Amaryllidaceae derivative,
lycorine, was isolated from Narcissus pseudonarcissus16 and
since then more than 100 have been isolated from Amarylli-
daceae species.15,16 Kornienko and Evidente16 reported that it
was likely that isocarbostyril constituents of Amaryllidaceae,
such as narciclasine, pancratistatin, and their congeners, were
the most important ingredients responsible for the therapeutic
benefits of these plants in the folk medicine treatment of cancer.
Narciclasine’s (1) anticancer activity was evidenced for the first
time by Ceriotti et al. in 1967, who described the compound as
antimitotic and displaying colchicine-like effects.17 Narciclasine
and pancratistatin also emerged as interesting antitumor drugs
in the NCI database.15 Carrasco et al.18 reported that narciclasine
interacts with the 60S eukaryotic ribosome subunit and inhibits
the peptide bond formation step in eukaryotic protein synthesis.
More recently, McLachlan et al.19 demonstrated that pancrat-
istatin, whose chemical structure is very close to that of
narciclasine (Figure 1), induces rapid apoptosis in SHSY-5Y
neuroblastoma cells. We ourselves recently reported that nar-
ciclasine at concentrations g1 µM appears to induce marked
apoptosis-mediated cytotoxic effects in human carcinoma cells
but not in normal fibroblasts by activation of the death receptor
pathway.20
About 90% of cancer patients die from their metastases or
from the dramatic invasion of vital organs. Gliomas,1 melano-
mas,2 esophageal cancer,3 pancreatic cancer,4 hormone refrac-
tory prostate cancer,5 and non-small-cell lung cancers (NSCL-
Ca)6 are among those associated with the most dismal prognoses
because of their inherent resistance to apoptosis, especially in
their metastatic form. Unfortunately, most of the agents used
today by clinicians to combat cancer are pro-apoptotic drugs,
which are rather ineffective in combating apoptosis-resistant
metastatic cancers.7 Great hope is therefore being invested in
molecules able to kill metastatic or highly invasive cancer cells
by nonapoptotic means, notably for example pro-autophagic
drugs.7 Effort is also being made to discover and develop new
types of compounds that may be able to selectively decrease
the migration of cancer cells, notably metastatic cancer cells,
and increase their sensitivity to cytotoxic drugs with either pro-
apoptotic and/or pro-autophagic effects.8-10
Natural products have played a highly significant role in the
discovery and development of new drugs for the treatment of
* To whom correspondence should be addressed. Phone: +32 477 62
20 83. Fax: +32 23 32 53 35. E-mail: rkiss@ulb.ac.be.
†
The first two authors contributed equally to this work.
Unibioscreen SA.
Service de Neurochirurgie, Cliniques Universitaires de Bruxelles,
‡
§
Hoˆpital Erasme, Universite´ Libre de Bruxelles.
|
Laboratoire de Toxicologie, Institut de Pharmacie, Universite´ Libre de
Bruxelles.
Department of Neurosurgery, Wagner Jauregg Hospital.
Institute of Cancer Research, Department of Medicine I, Medical
#
University of Vienna.
R.K. is a Director of Research at the Belgian National Fund for
Scientific Research (FNRS, Belgium). F.L. is a Clinical Research Fellow
with the FNRS.
a Abbreviations: GBM, glioblastoma; HUVEC, human umbilical vein
endothelial cell; MOMP, mitochondrial outer membrane permeabilization;
NOAEL, no-adverse-effect level; NOEL, no-effect dose level; NSCLC, non-
small-cell lung cancer; OMM, outer mitochondrial membrane; SAR,
structure-activity relationship; TMZ: temozolomide.
10.1021/jm8013585 CCC: $40.75
2009 American Chemical Society
Published on Web 01/26/2009

SAR Analysis of NoVel DeriVatiVes of Narciclasine
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4 1101
Figure 1. Chemical structures of narciclasine, lycoricidine, and pancratistatin.
The aim of the present study was to confirm the anticancer
effects of narciclasine in apoptosis-resistant as well as in apoptosis-
sensitive cancer cells and to determine whether the compound’s
activity results from pro-apoptotic effects or from impairment of
the organization of the actin cytoskeleton of cancer cells, a process
which would have a marked impact on cancer cell proliferation
(cytokinesis) and migration.21,22 In addition, given the total
synthesis of narciclasine is still very complex and requires too many
steps to be developed commercially,15,16 the current study reports
on the synthesis of novel analogues derived from narciclasine itself,
which may be amenable to large-scale preparation.16,23,24 A series
of 26 novel analogues25 was obtained with the further aim of
determining their activity in various in vitro and where appropriate
in vivo tumor models and to further establish the structure-activity-
relationship (SAR) within the series and the underlying mechanism
of action of narciclasine and its analogues.
Chemistry. For the extraction and purification of narciclasine
(1; Figure 1), several processes can be used16,26 depending on
the plant species. Described here is the outline procedure used
for the Narcissus “Carlton”. More details are provided in the
Experimental Section. An ethanolic extract of fresh bulbs of
the Narcissus “Carlton” (Ernest Turc, Angers, France) was
evaporated to leave an aqueous residue, which was itself
extracted with dichloromethane to remove nonpolar contami-
nants. Narciclasine (1) was then extracted from the residual
aqueous phase with ethyl acetate. After evaporation of the
solvent, the residue was further purified by column chroma-
tography (see Experimental Section). Narciclasine (1) was
obtained as the pure product after methanol crystallization of
the eluted fraction containing the compound. The spectral
properties (1D NMR, 2D NMR and mass spectra) of the isolated
product were identical to those previously published26 and
confirmed the identity of the product as narciclasine (1). Using
this procedure, 110 mg of narciclasine (1) was obtained from 1
kg of bulbs, which is of the same order as previously
reported.15,16
To perform SAR analysis of novel narciclasine related
products, all reactive positions of narciclasine (R1 with or
without the double bond (dotted line in the right-hand structure
of Figure 2), R2, R3, R4, R5, and X) have been modified by
hemisynthesis (Figure 2). These chemical modifications, which
are summarized in Table 1, involved a limited number of
chemical steps, which are described in detail in the Experimental
Section or the Supporting Information.
To obtain esters in position R1 and R5 (Scheme 1), we first
synthesized the well-known acetonide intermediate (2) starting
from narciclasine.27 This product (2) was then acylated to give
esters (3a-3e) as intermediates. The acetonide function of these
compounds was then deprotected by acidic conditions to provide
esters (6a-7e). To transform the diester (6a) into the monoester
(7a), a selective hydrolysis of the phenol ester (reaction (e),
Scheme 1) was used as previously published.28 To obtain
Figure 2. General structure of hemisynthesized derivatives of narci-
clasine (1).
modified products in position R4 and R5, we first synthesized
the triacetate ester of narciclasine (4) as described in the
literature28 (Scheme 1). Compound (4) was then alkylated or
coupled with various reagents to provide derivatives (5f-5o).
After alkaline hydrolysis of ester functions, we obtained
derivatives (7g-8n). Starting from product (7o), deprotection
of one benzyl group by sodium iodide furnished the compound
(8o).
To reduce the carbonyl lactam function of narciclasine, we
also started the synthesis from the acetonide derivative of
narciclasine (2). The alcohol function in position 2 of this
compound was then protected by the silyl ether (TBDMS) to
obtain product 9 (Scheme 2). The reduction by LiAlH4 followed
by the deprotection of the silyl ether gave the compound (10).
Finally, the acetonide function of this compound was then
deprotected by acidic conditions to provide derivative (11) in
which the carbonyl function of narciclasine had been reduced
(Scheme 2).
To verify the influence of the carbonyl function on the
cytotoxicity of this class of compound, we modified the amino
group of product (11). To achieve this, we first synthesized urea
or amide derivatives starting from compound 9 (Scheme 3).
After reduction of the carbonyl and reaction with trichloroacetyl
isocyanate, this intermediate gave derivative (12). After depro-
tection and hydrolysis, this furnished the urea (13). After the
same reduction of the carbonyl followed by reaction with acetyl
chloride, the intermediate gave compound (14). The acetonide
function of this latter product was then deprotected to provide
the amide (15) (Scheme 3).
To obtain derivatives (18) and (19), catalytic hydrogenation
in the presence of triethylammonium formate was performed
to provide after acetonide protection compounds (16: cis) and
(17: trans; Scheme 4). It is interesting to note this hydrogenation
gave around 50% of the trans-derivative and 50% of the cis-
derivative. These hydrogenation conditions are the best devel-
oped to date to obtain the trans-derivative starting from
narciclasine. Swern oxidation followed by reductive amination
with propylamine and deprotection of the acetonide function
provided final products (18) and (19) starting from compounds
(16) and (17; Scheme 4).

1102 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Ingrassia et al.
Table 1. Novel Narciclasine Derivatives and Their In Vitro Antiproliferative Activity against Human Cancer Cell Linesd

SAR Analysis of NoVel DeriVatiVes of Narciclasine
Table 1. Continued
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4 1103
a Chemical modification to each of 1’s reaction sites. For these derivatives, the dotted line in Figure 2 represents a double bond. b The in vitro antiproliferative
activities of the compounds are reported as IC50 values (nM) determined using the MTT colorimetric assay. IC50 values refer to the concentration needed to
decrease the overall growth of cell lines by 50% over three days on culturing in the presence of the drug. This was carried out on six different human cancer
cell lines including PC-3 prostate, U373 GBM, BxPC3 pancreas, LoVo colon, A549 NSCLC, and MCF-7 breast. c The stability of products was measured
by HPLC analysis following incubation in a physiological solution at 37 °C over 7 days. Results are expressed as the % of the incubated compound
recovered. d nd: not determined.
Scheme 1a
a (a) 2,2-Dimethoxypropane, PTSA, DMF. (b) Acyl anhydride or acyl chloride, pyridine or DMF. (c) TFA, H2O, THF. (d) Acetic anhydride, DMAP,
pyridine. (e) H2O, pyridine, 110 °C. (f) Alkyl halide, K2CO3, DMF, or acetonitrile for (5k) and 5l: 1-(R)-bromoperbenzoylglucose, AgOTf, allylsilane, 4 Å
molecular sieves, CH2Cl2/toluene; for 5o: dibenzyl phosphite, CCl4, DIPEA, DMAP, acetonitrile. (g) K2CO3, H2O, MeOH. (h) LiOH, H2O, dioxane, 0 °C.
(i) NaI, acetone.
For most compounds, aqueous stability was determined at
physiological pH and 37 °C (Table 1). Some compounds to be
orally administered were also tested at pH 2.
Pharmacology. Characterization of the Pro-Apoptotic
Effects Induced by Narciclasine (1) in Human Cancer
Cells. As stated previously, we recently demonstrated that the
in vitro growth inhibitory (antiproliferative) activity of high
narciclasine (1) concentrations (g1 µM) in carcinoma cells
(cancer cells of epithelial origin) relates to pro-apoptotic effects
resulting from compound-mediated triggering of the activation
of the initiator caspases of the death receptor pathway.20
However, Table 1 shows that the in vitro IC50 growth inhibitory
activity of narciclasine (1) in human cancer cells was well below
1 µM and in fact in the range 30-90 nM. Narciclasine (1)

1104 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Ingrassia et al.
Scheme 2a
a (a) 2,2-Dimethoxypropane, PTSA, DMF. (b) TBDMSCl, imidazole, DMF. (c) LiAlH4, THF. (d) TBAF, THF. (e) H2SO4, H2O, THF/CH2Cl2.
Scheme 3a
a (a) LiAlH4, THF. (b) Trichloroacetyl isocyanate, MEK, 0 °C. (c) TBAF, THF. (d) K2CO3, H2O, MeOH. (e) TFA, H2O, THF. (f) Acetyl chloride, TEA,
CH2Cl2. (g) TBAF, THF.
displayed similar in vitro growth inhibitory activity in carcinoma
cells highly sensitive to pro-apoptotic insult such as human
MCF-7 breast20 and PC-3 prostate cancer cells20 and apoptosis-
resistant human U373 glioblastoma (GBM) cells.29
mitochondrial membrane (OMM).20 MOMP is lethal because
it leads to the release of diverse apoptotic effectors contained
in the mitochondrial transmembrane space and compromises
mitochondrial energy metabolism.20 Accordingly, use was made
of the JC-1 dye to measure ∆Ψm in untreated and narciclasine-
treated U373 and MCF-7 cancer cells (positive control20). Figure
3B shows that narciclasine induced a concentration-dependent
decrease in ∆Ψm in MCF-7 cells but not in U373 cells. In
addition, narciclasine (1) provoked marked pro-apoptotic and
pro-necrotic processes in MCF-7 (Figure 3Ca) but not in U373
cells (Figure 3Cb).
Furthermore, the IC50 growth inhibitory value of narciclasine
(1) in the human A549 NSCLC cell line, which is also resistant
to various pro-apoptotic insults,32-34 was similar to those
observed in apoptosis-sensitive MCF-7 and PC-3 cancer cells
(Table 1). Taken together, these data strongly suggest that
narciclasine (1) (i) exerts its antitumor effects through nonapo-
ptotic mechanisms when evaluated in vitro at its IC50 growth
inhibitory values and (ii) displays similar IC50 growth inhibitory
values in apoptosis-sensitive and apoptosis-resistant cancer cells.
Narciclasine (1) Impairs Cancer Cell Proliferation and
Migration, and to a Lesser Extent Fibroblast Cell Prolifera-
tion and Migration. The in vitro scratch wound assay in
conjunction with computer-assisted phase-contrast microscopy
Figure 3A shows that human Hs683 GBM cells that are 1p19q
deleted29 (oligodendrogial in origin30) and sensitive in vivo to
pro-apoptotic drugs29 can be challenged at the level of p53
activation by a pro-apoptotic insult induced by adriamycin
(ADR; used as a positive pro-apoptotic control) but not by
temozolomide (TMZ; a pro-autophagic drug7 used here as a
negative pro-apoptotic control). TMZ can induce late apoptosis
events in cancer cells but only after they have been treated for
at least 5 days with TMZ.31 The sensitivity of Hs683 GBM
cells to the pro-apoptotic insult of adriamycin is similar to that
observed in MCF-7 cells (Figure 3A), which are also sensitive
to pro-apoptotic drugs.20 In sharp contrast, p53 is mutated and
constitutively activated (Figure 3A) in human U373 GBM cells
(astroglial in origin,30 not deleted with respect to 1p and 19q29
and not sensitive in vivo to pro-apoptotic drugs29), a feature
that suggests that these cells should be apoptosis-resistant.
Mitochondrial outer membrane permeabilization (MOMP) is
a crucial step in the apoptotic process and is often concomitant
with a decrease in mitochondrial membrane potential (∆Ψm)
as a direct consequence of the loss of integrity in the outer

SAR Analysis of NoVel DeriVatiVes of Narciclasine
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4 1105
Scheme 4a
a (a) Triethylammonium formate, Pd/C, CH2Cl2/EtOH. (b) 2,2-Dimethoxypropane, PTSA, DMF. (c) DMSO, oxalyl chloride, Et3N, CH2Cl2, -78°C. (d)
Propylamine, NaBH(OAc)3, DCE. (e) TFA, H2O, THF.
enabled the quantitative determination of both cell proliferation
and migration.22 Human U373 GBM and PC-3 prostate cancer
cells and normal Ccd-25-Lu fibroblasts were grown in vitro to
confluence and scratch wounds made by creating a linear
denuded region using a pipet tip as illustrated at the top of Figure
4 (CT 0 h). The three cell lines studied reached confluence after
72 h. One image was taken every four minutes for each
experimental treatment of the three cell lines (control, 10 and
100 nM incubation with narciclasine). The percentages of wound
area recolonized were calculated every 12 h for 72 h as indicated
in Figure 4. The lower narciclasine concentration did not retard
wound healing. At 100 nM, narciclasine (1) displayed greater
inhibitory effects in U373 compared to PC-3 cancer cells despite
IC50 growth inhibitory values in these two cell lines being 0.03
µM (Table 1). Thus, narciclasine (1) appears to exert its
antitumor effects independent of apoptosis status: resistant
(U37329) versus sensitive (PC-320) cells. We had observed
previously that normal human fibroblasts appear ∼250 fold less
sensitive to the antiproliferative/migratory effects of narciclasine.
The compound also does not induce apoptosis in these cells
probably due to the absence of death receptor pathway activa-
tion.20 However, the data in Figure 4 reveal that 100nM
narciclasine transiently impaired the wound healing process of
normal human Ccd-25-Lu fibroblasts.
Structure-Activity Relationship Analyses. Table 1 reveals
that almost none of the chemical modifications to narciclasine
(1) improved its antiproliferative IC50 values as revealed by the
MTT colorimetric assay. Only esters of the hydroxyl at position
R1 (7c-7e) apparently maintained/possibly improved the in
vitro antitumor activity of narciclasine (Table 1), while ester
(7a) had a ∼10 fold weaker activity. This type of modulation
of lycoricidine (Figure 1) had previously been reported to give
similar results.15,16,35 This may not be surprising given that
certain derivatives with ester function at R1 were revealed to
be appreciably unstable in aqueous media at 37 °C (Table 1).
The decomposition of these products determined by HPLC
analysis (Table 1) gave narciclasine and, more surprising,
narciprimine (for the chemical structure, see ref 16). Therefore,
the antiproliferative activity observed with these ester derivatives
may merely reflect in part the activity of the released parent,
narciclasine. Narciprimine was obtained by elimination of the
ester group in the allylic position (good leaving group) and
subsequent aromatization of ring C. Diesters on positions R1
and R5 (6a and 6b, Table 1) resulted in growth inhibitory
activity on cancer cells similar or weaker than corresponding
monoester derivatives (7a, 7c). These latter data suggested that
a free phenolic hydroxyl group at position R5 provides optimal
antitumor activity.15,16 Given the ready instability of monoesters
(elimination of the ester group), further synthetic effort was
directed to obtaining more stable products, notably amino
derivatives at position R1 (weaker leaving group). To provide
this kind of compound with a propylamino function, the double
bond (dotted line, Figure 2) had to be removed because all
activated intermediates with double bonds were unstable. Two
amino compounds (18 and 19) were stable in aqueous media at
37 °C but failed to demonstrate significant antitumor activity
in vitro (Table 1). This result was perhaps not surprising for
the cis-derivative (18) because it is known that the trans-B/C
ring junction (without the double bond between C-10a and C1)
is essential to maintain potent cytotoxicity for this type of
compound.15,16 In contrast, the absence of antitumor activity
for the trans-derivative (19) was more surprising given this
product does not have the double bond but has the required
trans-B/C ring junction to maintain cytotoxic activity. As
position R1 tolerates steric hindrance (given that product (7c)
Narciclasine (1) Impairs Actin Cytoskeleton Organiza-
tion in Cancer Cells. A common feature that can impair both
cell proliferation and migration (as revealed in dynamic scratch
wound assays on U373 GBM cells (Figure 4) when no cell death
is evidenced (Figure 3Cb)) relates to disorganization of the actin
cytoskeleton.1,22 At 100 nM, narciclasine markedly increased
the proportion of fibrillary polymerized actin (green fluores-
cence) constituting the actin cytoskeleton in both PC-3 prostate
cancer (Figure 5A) and U373 GBM (Figure 5B) cells but not
in Ccd-25-Lu normal fibroblasts (Figure 5C). Quantitative
determination using computer-assisted fluorescence microscopy
(Figure 5D) revealed that 100 nM narciclasine indeed markedly
increased the levels of fibrillary actin in PC-3 and U373 cancer
cells (p < 0.001) but not in Ccd-25-Lu normal cells (p > 0.05).
This increase renders the actin cytoskeleton in cancer cells more
rigid, a process that can lead to the impairment of growth (Table
1) and migration (Figure 4).

1106 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Ingrassia et al.
Figure 3. (A) Western blotting analyses of p53 expression after having cultured Hs683 (oligodendroglioma), U373 (astroglioma), and MCF-7
(breast) cancer cells for 24 h in the presence of 10 µM temozolomide (TMZ: a pro-autophagic drug) and 5 µM adriamycin (ADR: a pro-apoptotic
drug). (B) Involvement of mitochondria in apoptotic cell death induced by narciclasine: JC-1 staining of MCF-7 and U373 cells either untreated (“0
µM”) or compound treated with 0.1 or 1 µM for 24 h. Red fluorescence resulting from JC-1 staining is specific to mitochondria with an intact
mitochondrial membrane potential of ∆Ψm. Mitochondrial outer membrane permeabilization, which is a direct consequence of the loss of integrity
in the outer mitochondrial membrane, is a crucial step in the apoptotic process and is visualized through the loss of JC-1 staining in MCF-7 cancer
cells only.
with a benzoyl group reveals potent cytotoxicity), compound
(19)’s lack of antitumor activity could not be attributed to the
steric hindrance of the propylamino group. It could however,
result from inappropriate stereochemistry at position 2 during
its hemisynthesis: R-configuration for (19) in contrast to the
S-configuration for narciclasine (1). The C-2 configuration seems
therefore to be essential for the antitumor activity of this type
of compound. The hydroxyl functions at positions R2 and R3

SAR Analysis of NoVel DeriVatiVes of Narciclasine
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4 1107
Figure 4. Scratch wound assays of human PC-3 prostate cancer cells, U373 glioblastoma cells, and normal Ccd-25-Lu lung fibroblasts treated for
72 h with 10 or 100 nM narciclasine. The percentage wound healing which occurred over a 72 h period of observation was quantitatively determined
by means of computer-assisted videomicroscopy as detailed previously.40,47
also seem essential on analyzing the antitumor activity obtained
for 9, 10, and 20, as already reported in the literature.15,16,36,37
All modifications made to positions R4 and X led to compounds
devoid of antitumor activity in vitro (5f, 7i, 7m, 10, 11, 13,
and 15). The lactam carbonyl function with no substitutions on
the nitrogen atom seems therefore to be necessary for antitumor
activity. The activities of compounds with modulation of
position R5 were found to be at least 100 times less potent (5h,
7g, 7h, 7k, 8n) than narciclasine or even inactive (5g, 7i, 7j).
These data appear to confirm that the phenolic hydroxyl group
at position R5 of narciclasine is mandatory for antitumor
activity.15,16
Narciclasine (1) Displays Higher In Vitro Antiprolifera-
tive Effects in Cancer than Normal Cells. The growth
inhibitory effects of narciclasine (1) on 19 human and 3 rodent
cancer cell lines, and 6 normal cell lines including 3 fibroblast

1108 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Ingrassia et al.
Figure 5. Qualitative (A-C) and quantitative (D) determination of narciclasine-induced effects on actin cytoskeleton organization in human PC-3
prostate cancer cells (A), U373 glioblastoma cells (B), and normal Ccd-25-Lu lung fibroblasts (C). Fibrillar actin and globular actin appear as green
and red fluorescence, respectively. The levels of fibrillar actin was quantitatively determined by means of computer-assisted fluorescence microscopy
as detailed previously.47
and 3 human umbilical vein endothelial cell (HUVEC) lines
were determined (Table 2). The data show that narciclasine (1)
displayed potent antiproliferative activity independent of the
tumor type and independent of whether the cells were of human
(IC50 range 5-99 nM) or rodent origin (IC50 range 28-35 nM).
Narciclasine (1) exerted markedly weaker effects on normal
fibroblasts than on tumor cells (IC50 values g 317 nM). The
compound’s antiproliferative effect on HUVEC cells was closer
to that observed in tumor cells (IC50 range 87-94 nM). These
data thus suggest that in addition to displaying potent antitumor
effects, narciclasine (1) could also display antiangiogenic effects.
The fact that the compound displays antiproliferative and
antimigratory (data not shown) effects on tumor cells and
HUVECs, could relate at least partly to the fact that actin, a
clear target for narciclasine (Figure 5), is key to migration and
proliferation processes in both tumor cells and endothelial
cells.21,22
which are well documented in vivo, are believed to operate in
the case of narciclasine 4-O-ꢀ-D-glucopyranoside, which has
shown comparable anticancer activity to narciclasine.38 Although
similar mechanisms that convert 7k into narciclasine could
operate for the esters 4, 6a, 7a-7e, and the glycosyl derivatives
5k and 5l, among the prodrugs synthesized, compound 7k which
displays 92% stability at physiological pH 7.4 (Table 1) and is
<20% degraded in 1 h at pH 2 (Figure 6C) was considered to
be a suitable candidate (Figure 6) for further evaluation. High
chemical stability at pH 7.4 and pH 2 were considered important
to the potential pharmaceutical development of respectively iv
and oral formulations of these narciclasine prodrugs.
Figure 6A illustrates narciclasine (1) plasma concentrations
after either single administration of the compound at 10 mg/kg
po (open circles) or 5 mg/kg iv (crosses) or repeat administration
at 5 mg/kg iv for 5 days (5ix1w) to female B6D2F1 (Charles
River) mice. Post oral dose, narciclasine (1) revealed peak
plasma concentrations of ∼300 ng/mL, and an overall systemic
exposure reflected in an AUC0-24 h of 50129 ng·min/mL. When
compared to the post iv AUC0-24 h of 81132 ng·min/mL, this
confirmed the oral bioavailability of narciclasine (1) in mice to
be 32%. Post iv, narciclasine was cleared rapidly, with a terminal
elimination half-life of 66 min. Furthermore, comparison post
iv of terminal elimination half-life values after single (1ix1w;
66 min) versus repeat (5ix1w; 58 min) administration indicated
no marked change (day 1 versus day 5), suggesting that
Pharmacokinetics of Narciclasine (1) and its Prodrug
(7k) in Vivo in Healthy Mice. As indicated above, most
hemisynthesized derivatives of narciclasine in the current study
resulted in products with generally lower inhibitory effects than
narciclasine on both the proliferation and migration of cancer
cells. Accordingly, it was considered appropriate to evaluate
potential prodrugs of narciclasine with respect to their ability
to demonstrate notably improved oral bioavailability and activity
in vivo. Indeed the mechanisms involved in prodrug cleavage

SAR Analysis of NoVel DeriVatiVes of Narciclasine
Table 2. Determination of Antiproliferative IC50 Values for Narciclasine in 6 Normal and 22 Cancer Cell Linesa
human carcinoma human normal
histological group cell lines cell lines
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4 1109
rodent carcinoma
IC50 (nM)
histological group
IC50 (nM)
histological group cell lines IC50 (nM)
central nervous system U373
29
98
48
40
22
8
lung fibroblast
WI-38
Ccd-25-Lu<
WS-1
317
CNS
C6
9 L
B16F10
28
32
35
U-87
T98G
Hs683
Daoy
>10000
>10000
87
97
94
skin fibroblast
vein endothelial cells HUVEC (HTG04)
HUVEC (HTG06)
melanoma
head and neck
melanoma
FaDu
HUVEC (HTG07)
RPMI 2650
Detroit 562
SKMEL-28
HT144
G361
14
6
37
38
99
25
28
28
94
32
29
45
5
C32
prostate
pancreas
colorectal
PC-3
BxPC-3
LoVo
HCT-15
A549
lung
breast
MCF-7
Mda-Mb-231
mean ( SEM
38 ( 6
31
5
b
207
94
>10000
32 ( 2
32
28
median
IC50 min
IC50 max
99
35
a U373 (ECACC code 89081403), U-87 (ECACC code 89081402), T98G (ATCC code CRL-1690), G361 (ATCC code CRL-1424), C32 (ECACC code
87090201), and C6 (ATCC code CCL-107) cell lines were cultured in MEM medium supplemented with 5% heat inactivated fetal bovine serum. Hs683
(ATCC code HTB-138), FaDu (ATCC code HTB-43), RPMI 2650 (ATCC code CCL-30), Detroit 562 (ATCC code 138, SKMEL-28 (ATCC code HTB-72),
PC-3 (DSMZ code ACC465), BxPC-3 (ECACC code 93120816), LoVo (DSMZ code ACC350), HCT-15 (DSMZ code ACC357), A549 (DSMZ code
ACC107), MCF-7 (DSMZ code ACC115), 9L (ATCC code CRL-2200), and B16F10 (ATCC code CRL-6475) cell lines were cultured in RPMI medium
supplemented with 10% heat inactivated fetal bovine serum. HT-144 (ATCC code HTB-63), WI-38 (ATCC code CCL-75), and Ccd-25-Lu (ATCC code
CCL-215) were cultured in MEM medium supplemented with 10% heat inactivated fetal bovine serum and 100 µM nonessential amino acids. Daoy (ATCC
code HTB-186) and WS-1 (ECACC code 88021104) were cultured in MEM medium supplemented with 10% heat inactivated fetal bovine serum. Endothelial
cells obtained from the veins of human umbilical cords were cultured in EGM-2 MV Bullet Kit. MEM and RPMI cell culture media were supplemented with
4 mM glutamine, 100 µg/mL gentamicin, and penicillin-streptomycin (200 U/mL and 200 µg/mL). b The mean IC50 value could not be determined as one
or more of the corresponding data points were higher than the threshold value.
Table 3. Summary of Narciclasine Toxicity in Female Wistar Rats
narciclasine administered 5 times per week for 3 weeks at dose levels
assessment
25 mg/kg/day
10 mg/kg/day
3 mg/kg/day
no deaths
1 mg/kg/day
no deaths
general observations 8/11 rats found dead on
day 3 and the remaining
all rats sacrificed on day 3 for
ethical reasons
animals sacrificed for
ethical reasons
clinical observations days 1-3: piloerection and
days 1-3: piloerection, diarrhea,
hunched posture,
no clinical
signs
diarrhea.
hunched posture, and lethargy
piloerection from 2nd
week onward with signs
appearing about 1 h postdose
slight BW loss during
week 1 and week 2
body weight (BW)
no data available
severe BW loss (11%) on day 3
no BW loss
blood cell parameters alterations in red and white
alterations in red and white blood
cell parameters
increase in white blood cell
minimal increases in
cholesterol, triglycerides,
phospholipids, potassium,
and calcium
and biochemistry
blood cell parameters.
count but decrease in
red blood cell count and
hematocrit; increase in alanine
aminotransferase activity, decrease concentrations
in creatinine, sodium, and
chloride concentrations but
increases in cholesterol,
triglyceride, phospholipid,
potassium, and calcium
levels.
macroscopy
microscopy
stomach abnormalities, red foci stomach abnormalities, but to
irregular surface
no findings
on thymus, thymic atrophy,
and autolysis
lesser extent than at 25 mg/kg/day, of the fore stomach
emaciation, and thymic atrophy
(5/5 rats) and foci in
the stomach (1/5 rats)
keratinized forestomach
acanthosis and hyperkeratosis
(5/5 rats), with mild erosion
(1/5 rats)
glandular gastric lesions of
individual cell necrosis in
superficial mucosa
glandular gastric lesions of
individual cell necrosis in
superficial mucosa; inflammatory
and degenerative cecal mucosal
lesions (2/5 rats),
minimal acanthosis in the
keratinized forestomach mucosa
(2/5 rats)
keratinized forestomach
with lesions of acanthosis
(1/5 rats)
narciclasine (1) has no effects on CYP enzyme activity with
respect to its own metabolism.
Figure 6B illustrates plasma concentrations of compound 7k
after single iv administration of 5 mg/kg of 7k (black circles)
and of compound 7k (black triangles) and narciclasine (1; open
triangles) after single oral administration of 10 mg/kg of 7k to
female B6D2F1 mice. The data indicate that compound 7k is
incompletely transformed into narciclasine (1) when adminis-
tered orally. For 7k, its AUC0-24 h post iv was determined to be
254261 ng·min/mL and its AUC0-72 h post oral administration

1110 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Ingrassia et al.
Figure 6. Narciclasine pharmacokinetics and in vitro metabolism and stability. (A) Mean plasma concentrations of narciclasine (1) after a single
administration to mice (n ) 3) iv or po doses at 5 and 10 mg/kg, respectively, formulated in 5% 2-hydroxypropyl-ꢀ-cyclodextrine. Plasma
concentrations of 1 were determined using a validated LC-MS/MS method with a linear range of 2.5-1000 ng/mL (0.01 to 3.26 µM). (B) Plasma
concentrations of narciclasine (1) after single oral administration of compound 7k at a dose level of 10 mg/kg (open triangles) and of compound
7k after single iv administration of 5 mg/kg (black circles) or a single oral administration of 10 mg/kg (black triangles) of compound 7k formulated
in 5% HPBCD. (C) In vitro release of narciclasine (black circles) from compound 7k (open circles) in aqueous media at 37 °C and pH 2. Kinetic
degradation of compound 7k was followed using an HPLC-UV method. (D) In vitro metabolic stability of narciclasine (1) using liver microsomes
from different species. The metabolism of narciclasine (1) was followed using a LC-MS method.
was 29740 ng·min/mL to give an overall oral bioavailability
body weight, food consumption, biochemical parameters, mac-
roscopy at termination, organ weights, and histopathology on
an extensive list of tissues. The resulting data, which are
summarized in Table 3, reveal significant toxicity and/or lethality
at the two highest dose levels. For the regimen used, the no-
effect dose level (NOEL) could not be established for narci-
clasine (1) and is <1 mg/kg/day po, while the no-adverse-effect
level (NOAEL) was defined to be 1 mg/kg/day po, with the
minimal acanthosis reactive changes and minor variations in
some biochemistry parameters observed at this dose level
considered to be nonadverse. We accordingly used this NOAEL
of 1 mg/kg/day po to compare the in vivo antitumor effects of
narciclasine (1) and of compound 7k, as detailed below.
Characterization of the In Vivo Anti-Tumor Activity of
Narciclasine (1) and its Prodrug (7k) in Human Xenograft
Models. As already indicated earlier, the human A549 NSCLC
model resists various pro-apoptotic insults32-34 and we accord-
ingly used this model to investigate the in vivo antitumor effects
of narciclasine (1). We orthotopically grafted A549 NSCLC cells
in two distinct locations, either into the lungs of immunocom-
promized mice as detailed previously32,39 or into their brains
to mimic brain metastases. Taxol was used as a reference
compound. Figure 7A shows that a single iv administration of
narciclasine (1) per week at 1 mg/kg for five consecutive weeks
(1ix5w schedule), did not increase the survival of mice (black
squares) compared to control animals (black circles). Taxol
administered iv at 20 mg/kg according to this 1ix5w schedule
(stars) significantly increased the survival of A549 NSCLC
xenograft-bearing mice (Figure 7A). The reverse was observed
of 5.8% for 7k. The AUC0-72 of narciclasine (1) post oral
h
administration of compound 7k (10 mg/kg) was calculated to
be 54812 ng ·min/mL, which when dose normalized (for
differences in molecular weight between compound 7k and
narciclasine) and compared to the narciclasine (5 mg/kg) post
iv AUC0-24 h of 81132 ng·min/mL gives an improved absolute
bioavailability of ∼52% for narciclasine (1) after oral admin-
istration of 7k.
Figure 6C illustrates the in vitro release of narciclasine (1;
black circles) from compound 7k (open circles) in aqueous
media at 37 °C and pH 2. Narciclasine (1) was the major product
detected in the incubation and conversion was complete after
24 h under these conditions.
Figure 6D reveals that narciclasine (1) appears to be ap-
preciably metabolically stable in mouse, rat, dog, and human
hepatic microsomes although there appears to be a slight decline
(e20%) in rodent species after 45 min incubation. This may
indicate that cytochrome P450 isozymes may not play a major
role in the metabolism of the compound.
Narciclasine (1) Toxicity in Rats. Narciclasine (1) formu-
lated in 15% (w/v) aqueous 2-hydroxypropyl-ꢀ-cyclodextrine
(HPBCD) was administered by oral gavage for 5 consecutive
days (Monday-Friday) for 3 weeks to female Wistar rats at
dose levels of 0 (group 1), 1 (group 2), 3 (group 3), 10 (group
4), and 25 (group 5) mg/kg/day. Each group consisted of five
animals. In addition, satellite animals were allocated to each
treatment group for toxicokinetic assessment of drug exposure.
The following investigations were undertaken: clinical signs,

SAR Analysis of NoVel DeriVatiVes of Narciclasine
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4 1111
Figure 7. Activity in orthotopic models of human NSCLC and glioma in immunocompromized mice. (A-B) The impact of iv administered
narciclasine (1) and taxol on the survival of immunocompromized mice orthotopically grafted into their lungs (A) or brains (B) with human A549
NSCLC.32 Taxol was administered once per week (Monday) at 20 mg/kg for five consecutive weeks with treatment starting on the 7th day post-
tumor graft, while narciclasine was administered to the same schedule but at 1 mg/kg. (C) The impact of iv administered narciclasine (1) and
compound 7k on the survival of immunocompromized mice orthotopically grafted into their brains with human Hs683 anaplastic oligodendroglioma
cells (3 × 104 cells/graft).29,30,40,41 The two compounds were administered five times a week (Monday-Friday) at 1 mg/kg for three consecutive
weeks, with treatment starting on the 7th day post-tumor graft. (D) The impact of orally administered narciclasine (1) and compound 7k on the
survival of immunocompromized mice orthotopically grafted into their brains with human GL-19 glioblastoma cells (2 × 106 cells/graft). The two
compounds were administered once a week (Monday) at 1 mg/kg for five consecutive weeks, with treatment starting on the 7th day post-tumor
graft.
when A549 NSCLC cells were grafted into the brains of
immunocompromized mice (Figure 7B). These data suggest that
narciclasine (1) could readily penetrate the blood-brain barrier
(Figure 1). We therefore then made use of two human glioma
models to compare the antitumor effects contributed in vivo by
narciclasine (1) and its prodrug 7k. The first was the Hs683
anaplastic oligodendroglioma model29,30 that grows and devel-
ops very aggressively in the brains of immunocompromized
mice.40,41 We accordingly used an intensive iv treatment
schedule of “5ix3w” 1 mg/kg/day for narciclasine (1) and
compound 7k. While narciclasine (1) failed to significantly
increase the survival of Hs683 xenograft-bearing mice, com-
pound 7k did (Figure 7C). The second model evaluated GL-19
was a GBM primoculture established in our facilities as detailed
previously.42 This model was used to compare the oral in vivo
antitumor effects of narciclasine (1) and its prodrug 7k. The
data in Figure 7D show that narciclasine administered orally at
1 mg/kg/day using a 1ix5w schedule did not significantly
increase the survival of GL-19 GBM orthotopic xenograft-
bearing mice, while its prodrug 7k did.
toward normal and cancer cells. Narciclasine-mediated pro-
apoptotic effects in carcinoma cells at high concentrations result
from the activation of the initiator caspases of the death receptor
pathway (caspase-8 and -10), at least in human MCF-7 breast
and PC-3 prostate carcinoma cells.16 However, it should be
emphasized that the concentration of 1 µM used in this previous
study20 is ∼20 times higher than the mean antiproliferative IC50
value against cancer cells in vitro, which is around 50 nM. The
data from the present study also show that even at 1 µM
narciclasine does not elicit apoptosis and/or necrosis in apop-
tosis-resistant U373 GBM cells (Figure 3). At 50 nM, narci-
clasine blocks mitosis in carcinoma and glioma (and also
melanoma: data not shown) cells by a mechanism of action that
is not yet fully elucidated. Quantitative videomicroscopy has
revealed that in fact narciclasine markedly impairs cancer cell
migration; not only of PC-3 prostate (Figure 4) and MCF-7
breast cancer (data not shown) cells but also of U373 GBM
cells (Figure 4), a feature which is paralleled by loss of tumor
cell polarity (data not shown). We report in the present study
that narciclasine induces a rapid increase in F-actin concentration
(principally at the cortical cell location) in U373 GBM and PC-3
prostate cancer cells but not in normal fibroblasts (Figure 5).
This sharp increase in the concentration of F-actin can in turn
rigidify the actin cytoskeleton and thus impair both cell
proliferation and migration. Preliminary data from our group
also reveal that the narciclasine-induced increase in F-actin is
Discussion
We have recently shown that narciclasine when used at 1
µM in vitro induces marked apoptosis-mediated cytotoxic effects
in human carcinoma cells but not in normal fibroblasts.20 Griffin
et al.43 also observed selective cytotoxicity of pancratistatin

1112 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Ingrassia et al.
associated with a dramatic modification in serine/threonine
phosphoprotein expression levels (manuscript submitted for
publication). In addition, as altered cofilin expression or activity
controlled by phosphorylation are associated with actin dynam-
ics, which in turn are associated with normal44 and cancer45
cell proliferation and migration, the effects of narciclasine on
cofilin phosphorylation status have been investigated. We have
observed that narciclasine treatment significantly and rapidly
increases cofilin phosphorylation (manuscript submitted for
publication). Our data are in accord with the pioneering work
of Ceriotti17 and Carrasco et al.,18 as it has become evident over
recent years that a large fraction of mRNA is tightly associated
with the cytoskeleton.46 Whereas microtubules (polymers of
tubulin) are involved in RNA-cytoskeletal association in large
cells like oocytes, neurons, or oligodendrocytes, microfilaments
(polymers of actin) play the major role in smaller somatic cell
types.46 Association of RNA with cytoskeletal filaments is
clearly required for mRNA transport but also appears to be
crucial for efficient protein synthesis.46
many types of cancers, including apoptosis-resistant cancers
associated with dismal prognoses, as the compound impairs actin
cytoskeleton organization. The actin cytoskeleton is implicated
in both cell proliferation and cell migration (including the
metastatic process). Narciclasine is significantly more toxic to
cancer than normal cells. The current study reveals that chemical
modification to the narciclasine backbone led to a certain
instability among synthesized esters and even among stable
derivatives only led to weakly active molecules or even to
complete loss of antitumor activity in vitro. In contrast, one
narciclasine prodrug (compound 7k) demonstrated higher in vivo
antitumor activity than narciclasine itself.
Experimental Section
Chemistry. Extraction and Isolation of narciclasine (1). Fresh
bulbs of the narcissus “Carlton” (500 g) were homogenized and
ethanol (1 L) added. The mixture was stirred mechanically at room
temperature for 24 h. After filtration, the residue was resuspended
in ethanol (1 L) for 2 h and then refiltered. The filtrates from these
two steps were then combined and concentrated under reduced
pressure until the ethanol was essentially removed (residual volume
of aqueous extract ∼200 mL). This was then extracted 3 times
initially with CH2Cl2 (3 × 200 mL) and then 3 times with AcOEt
(3 × 200 mL). The AcOEt extract (not dried over magnesium
sulfate) was concentrated under reduced pressure to obtain 700 mg
of a brown residue. This process was repeated on another nine
batches of narcissus bulbs of 500 g each to obtain 7 g of the brown
residue. This was resuspended after sonication in a minimum
volume of methanol and applied to a silica gel (400 g, 40-63 µm)
flash chromatography column (7 cm in diameter), which was eluted
with CH2Cl2 (2.5 L) and then CH2Cl2/MeOH (3 L 9:1 v/v). The
relevant product fractions as determined by TLC (Rf: 0.6 using
CH2Cl2/MeOH (8:2 v/v)) were combined and the solvent evaporated
to give 1.5 g of a brown solid. This solid was resuspended in a
maximum 50 mL of methanol and then sonicated and the resulting
solution maintained at ambient temperature for 3 days. The
precipitate that formed was then filtered to obtain 550 mg of fine
needles of pure narciclasine (1). The overall yield was 0.011%
starting from fresh bulbs. The product’s physical and spectral
properties were identical to those previously published.23,26 (i) mp:
250-252 °C. (ii) TLC Rf: 0.6 (CH2Cl2/MeOH: 8:2 v/v). (iii) RP-
HPLC (C18): system 1, Rt ) 5.08 min, and purity, 99.9%, and
system 5, Rt ) 6.15 min and purity: 99.9%. (iv) 1H NMR (DMSO-
d6): δ 13.25 (s,1H exchangeable with D2O, PhOH), 7.89 (s, 1H
exchangeable with D2O, NH), 6.86 (s, 1H, PhH), 6.16 (m, 1H,
dCH), 6.09 (s, 2H, O-CH2-O), 5.19 (m, 2H exchangeable with
D2O, OH), 5.03 (m, 1H exchangeable with D2O, OH), 4.19-3.71
(m, 4H, HC-2, HC-3, HC-4, HC-4a); 13C NMR: (DMSO-d6): δ
169.50 (C-6), 152.90 (C-9), 145.39 (C-10b), 133.99 (C-7), 132.68
(C-8), 129.82 (C-10a), 125.28 (C-1), 106.11 (C-6a), 102.64 (C-
11), 96.40 (C-10), 72.94 (C-3), 69.41 (C-4), 69.71 (C-2), 53.44
(C-4a) and 1H-1H COSY, HMBC, HMQC, NOESY spectra confirm
all hydrogen and carbon attributions. (v) MS (EI): m/z 307 (M+•),
271, 247, 218 and MS (ESI): m/z 308 (MH+).
As recently emphasized by Kornienko and Evidente,16 the
early activity studies conducted by Ceriotti17 in 1967 revealed
that sc narciclasine injections at a dose of 0.9 mg/kg to mice
bearing the aggressive sarcoma 180 model resulted in complete
disappearance of mitoses 4 h after administration. Thus, the
impairment of mitotic processes seems to be one of the major
endpoints of the compound’s antitumor effects, as indicated by
the pioneering investigations of Ceriotti17 and our current study.
The current study has additionally confirmed the oral NOAEL
in rats to be 1 mg/kg/day using a schedule of 5ix3w (Table 3).
The NOAEL dose should be >1 mg/kg/day with respect to a
treatment schedule involving narciclasine administration once
a week for 3-5 consecutive weeks. Toxicology evaluations are
ongoing in healthy mice to define the NOAEL in this species.
Narciclasine has been demonstrated to be orally available with
an absolute bioavailability of 32% determined in mice.
This current study has failed to deliver novel narciclasine
derivatives displaying higher antiproliferative activity in vitro
than narciclasine itself (Table 1). In this study, narciclasine
administered iv at 1 mg/kg using a “1ix5w” schedule, failed to
increase the survival of mice bearing orthotopic lung grafts of
A549 NSCLC. However, the compound was significantly active
using the same dose and schedule if A549 NSCLC cells were
grafted into the brain of mice (mimicking brain metastases), a
model in which taxol, which does not appreciably cross the
blood-brain barrier, failed to demonstrate any activity. This
latter result may indicate the ready penetration of narciclasine
into the brain. Studies to confirm this are ongoing.
Given the failure to obtain novel derivatives with increased
potency, a prodrug strategy was also investigated in an attempt
to enhance the in vivo activity of narciclasine itself. The
antitumor effects in vivo of narciclasine (1) and its prodrug 7k
were determined in two human GBM models (Hs683 and GL-
19) at a dose level of 1 mg/kg/day and a treatment schedule of
“5ix3w” either iv or oral. While narciclasine (1) failed to
significantly increase the survival of GBM xenograft-bearing
mice in either model at this dose and schedule, compound 7k
significantly increased survival times in both models. The
prodrug strategy would appear to have improved the in vivo
antitumor activity of narciclasine.
Glycosylated Products 5k and 5l. Narciclasine 2,3,4-(triacetate)
4 (150 mg, 346 µmol) and 2,3,4,6-tetra-O-benzoyl-R-D-glucopy-
ranosyl bromide (455 mg, 690 µmol) were dissolved in extra dry
CH2Cl2. This mixture was stirred and cooled to -78 °C. Allyltri-
methylsilane (315 µL, 1972 µmol) and silver trifluoromethane
sulfonate (175 mg, 690 µmol) in 5 mL of dry toluene were added.
The temperature was slowly allowed to rise to room temperature
without removing the cooling bath. After 17 h of reaction, the
mixture was filtered through celite and evaporated to a reduced
volume. Purification of this crude mixture by silica gel chroma-
tography with cyclohexane/AcOEt (7:3 to 1:1 v/v) gave 5k as a
white solid (143 mg) and 5l as a white powder (92 mg).
Post oral dose, prodrug 7k was found to increase the absolute
bioavailability of narciclasine to ∼52%.
In conclusion, narciclasine, a plant growth regulator whose
antitumor effects have been evidenced in traditional medicine
going back millennia, could represent a novel weapon against
Acetic acid 2,3-diacetoxy-7-O-(2′,3′,4′,6′-tetra-O-benzoyl-R-D-
glucopyranosyl)-6-oxo-2,3,4,4a,5,6-hexahydro-[1,3]dioxolo[4,5-
j]phenanthridin-4-yl ester (5k): (i) Yield: 41%. (ii) TLC Rf: 0.43

SAR Analysis of NoVel DeriVatiVes of Narciclasine
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4 1113
1
(CH2Cl2/MeOH: 96:4 v/v). (iii) H NMR (DMSO-d6): δ 8.54 (s,
form part of the legends to the figures. The rest of the information
is provided in the Supporting Information. The potential survival
gain obtained using narciclasine and its derivatives was evaluated
by means of survival curve analysis.9,10,29,31,39
In Vivo Determination of Narciclasine (1) and its Prodrug
(7k) Plasma Concentrations. The experimental protocols are
detailed in the Supporting Information.
Statistical Analyses. Statistical comparison of control and treated
groups was initially undertaken with the Kruskal-Wallis test (a
nonparametric one-way analysis of variance). Where this revealed
significant differences, the Dunn multiple comparison procedure
(2-sided test) was applied. However, this was adapted to the special
case of comparing treatment and control groups in which only (k
- 1) comparisons were undertaken among the k groups tested by
the Kruskal-Wallis test (instead of the possible k(k - 1)/2
comparisons considered in the general procedure). The levels of
statistical significance associated with the %T/C (test/control)
survival indices were determined by using Gehan’s generalized
Wilcoxon test. The correlation between numerical variables was
analyzed by means of the nonparametric Spearman correlation test.
All these statistical analyses were carried out using Statistica
(Statsoft, Tulsa, OK).
1H, NH), 7.90-7.41 (m, 20H, -OC-PhH), 7.09 (s, 1H, PhH), 6.59
(t, 3J3′-4′ ) 3J3′-2′ ) 9.6 Hz, 1H, HC-3′), 6.45 (d, 3J1′-2′ ) 3.3 Hz,
1H, HC-1′), 6.26 (m, 1H, )CH), 6.10 (s, 1H, O-CH-O), 6.03 (s,
3
3
1H, O-CH-O), 5.73 (t, J4′-3′ ) J4′-5′ ) 9.9 Hz, 1H, HC-4′),
5.65 (dd, 3J2′-3′ ) 10.2 Hz, 3J2′-1′ ) 3.9 Hz, 1H, HC-2′), 5.31-5.21
(m, 3H, HC-2, HC-3, HC-5′), 4.99 (m, 1H, HC-4), 4.45-4.38 (m,
3H, HC-4a, H2C-6′), 2.09 (s, 3H, CH3CO), 2.05 (s, 3H, CH3CO),
2.02 (s, 3H, CH3CO) and 13C NMR (DMSO-d6): δ 169.98 (OdC-
O), 169.49 (OdC-O), 169.25 (OdC-O), 165.01-164.80 (4C,
OdC-O benzoate), 161.99 (C-6), 151.42 (C-9), 138.73 (C-10b),
137.33 (C-7), 134.89 (C-8), 133.83-133.34 (m, Ph benzoate),
132.00 (C-10a), 129.23-128.43 (Ph benzoate), 117.59 (C-1),
114.39 (C-6a), 102.44 (C-11), 100.35 (C-1′), 99.57 (C-10),
71.14-62.73 (8C, C-3, C-4, C-2, C-2′, C-3′, C-4′, C-5′, C-6′), 49.00
(C-4a), 20.89 (CH3CO), 20.51 (CH3CO), 20.42 (CH3CO). (iv) MS
(ESI): m/z 1034 (MNa+) 1012 (MH+), 579.
2,3,4-Trihydroxy-7-(3,4,5-trihydroxy-6-hydroxymethyl-tetrahy-
dro-pyran-2-yloxy)-3,4,4a,5-tetrahydro-2H-[1,3]dioxolo[4,5-j]phenan-
thridin-6-one (7k): Compound 5k (113 mg, 112 µmol) and K2CO3
(20 mg, 144 µmol) were added to MeOH (15 mL) containing two
drops of water and stirred at ambient temperature for 16 h. The
solvent was removed under vacuum and the residue purified by
column chromatography using (CH2Cl2/MeOH/H2O 70:30:5 v/v/
v) to give 7k as a white solid (44 mg): (i) Yield: 84%. (ii) TLC Rf:
0.62 (CH2Cl2/MeOH/H2O 70:30:5 v/v/v). (iii) RP-HPLC (C18):
system 3, Rt ) 10.28 min and purity, 98.3% and system 6, Rt )
Acknowledgment. This work was partially supported by
grants awarded by the “Re´gion de Bruxelles-Capitale” (Brussels,
Belgium) and by the Fonds Yvonne Boe¨l (Brussels, Belgium).
We greatly thank Jean-Franc¸ois Gaussin, Ste´phanie Thomas,
Gwenae¨l Dielie, Lise Wlodarczak, and Gentiane Simon for their
excellent technical assistance.
1
9.12 min and purity, 99.7%. (iv) H NMR (DMSO-d6): δ 7.72 (s,
1H, NH), 7.13 (s, 1H, HC-10), 6.17 (m, 1H, dCH), 6.11 (s, 1H,
O-CH-O), 6.08 (s, 1H, O-CH-O), 5.47 (m, 1H, OH), 5.25 (d,
3J1′-2′ ) 3.6 Hz, 1H, HC-1′), 5.02 (m, 2H, OH), 4.41-3.10 (m,
14H, OH, CH2, HC-2′, HC-3′, HC-4′, HC-5′, HC-2, HC-3, HC-4,
HC-4a); 13C NMR (DMSO-d6): δ 163.36 (C-6), 151.42 (C-9),
141.36 (C-10b), 140.12 (C-7), 133.93 (C-8), 130.69 (C-10a), 124.25
(C-1), 114.33 (C-6a), 103.02, 102.24, 100.50 (C-1′, C-10, C-11),
74.03, 73.77, 72.34, 71.70, 69.05, 68.92, 68.80 (C-2′, C-3′, C-4′,
C-5′, C-3, C-4, C-2), 60.16 (CH2-O), 52.70 (C-4a). (v) MS (ESI)
m/z 492 (MNa+).
Supporting Information Available: The results of the analysis
of all compounds synthesized and evaluated in this study with the
exception of 1, 5k, and 7k (HPLC, 1H NMR, 13C NMR, and mass
spectra). This material is available free of charge via the Internet
at http://pubs.acs.org.
References
(1) Lefranc, F.; Brotchi, J.; Kiss, R. Possible future issues in the treatment
of glioblastomas: special emphasis on cell migration and the resistance
of migrating glioblastoma cells to apoptosis. J. Clin. Oncol. 2005,
23, 2411–2422.
(2) Eberle, J.; Kurbanov, B. M.; Hossini, A. M.; Trefzer, U.; Fecker, L. F.
Overcoming apoptosis deficiency of melanomashope for new thera-
peutic approaches. Drug Resist. Updates 2007, 10, 218–234.
(3) Malthaner, R. A.; Collin, S.; Fenlon, D. PreoperatiVe Chemotherapy
for Resectable Thoracic Esophageal cancer. ReView, The Cochrane
Library; John Wiley & Sons: New York, 2007; pp 1-34.
(4) Giovannetti, E.; Mey, V.; Nannizzi, S.; Pasqualetti, G.; Del Tacca,
M.; Danesi, R. Pharmacogenetics of anticancer drug sensitivity in
pancreatic cancer. Mol. Cancer Ther. 2006, 5, 1387–1395.
(5) Uzzo, R. G.; Haas, N. B.; Crispen, P. L.; Kolenko, V. M. Mechanisms
of apoptosis resistance and treatment strategies to overcome them in
hormone-refractory prostate cancer. Cancer 2008, 112, 1660-1671.
(6) Shivapurkar, N.; Reddy, J.; Chaudhary, P. M.; Gazdar, A. F. Apoptosis
and lung cancer: a review. J. Cell Biochem. 2003, 88, 885–898.
(7) Lefranc, F.; Facchini, V.; Kiss, R. Proautophagic drugs: a novel means
to combat apoptosis-resistant cancers, with a special emphasis on
glioblastomas. Oncologist 2007, 12, 1395–1403.
(8) Lefranc, F.; James, S.; Camby, I.; Gaussin, J. F.; Darro, F.; Brotchi,
J.; Gabius, H. J.; Kiss, R. Combined cimetidine and temozolomide,
compared with temozolomide alone: significant increases in survival
in nude mice bearing U373 human glioblastoma multiforme orthotopic
xenografts. J. Neurosurg. 2005, 102, 706–714.
(9) Joseph, B.; Darro, F.; Behard, A.; Lesur, B.; Collignon, F.; Decae-
stecker, C.; Frydman, A.; Guillaumet, G.; Kiss, R. 3-Aryl-2-quinolone
derivatives: synthesis and characterization of in vitro and in vivo
antitumor effects with emphasis on a new therapeutical target
connected with cell migration. J. Med. Chem. 2002, 45, 2543–2555.
(10) Ingrassia, L.; Nshimyumukiza, P.; Dewelle, J.; Lefranc, F.; Wlodarc-
zak, L.; Thomas, S.; Dielie, G.; Chiron, C.; Zedde, C.; Tisne`s, P.; van
Soest, R.; Braeckman, J. C.; Darro, F.; Kiss, R. A lactosylated steroid
contributes in vivo therapeutic benefits in experimental models of
mouse lymphoma and human glioblastoma. J. Med. Chem. 2006, 49,
1800–1807.
Pharmacology. In Vitro Pharmacology: Cell Lines. Human
cancer cell lines were obtained from the American Type Culture
Collection (ATCC, Manassas, VA), the European Collection of Cell
Culture (ECACC, Salisbury, UK), and the Deutsche Sammlung von
Mikroorganismen and Zellkulturen (DSMZ, Braunschweig, Ger-
many). The code numbers and histological types of all the cell lines
used in the current study are detailed in Table 2. The GL-19 GBM
primoculture is part of a collection of several hundred GBM
primocultures established within the Department of Neurosurgery
of the Wagner Jauregg Hospital (Linz, Austria).
Biological, Biochemical and Molecular Biology Related
Experiments. The effects of narciclasine and its derivatives on:
(1) The overall growth level of human cancer cell lines was
determined using the colorimetric MTT (3-[4,5-dimethylthiazol-
2yl]-diphenyl tetrazolium bromide, Sigma, Belgium) assay.9,10,20,21
(2) Cell proliferation (mitosis) and cell migration were determined
by means of computer-assisted phase contrast microscopy.8-10,21,22
(3) The actin cytoskeleton organization was determined by means
of computer-assisted fluorescence microscopy.21 (4) Apoptosis and
necrosis were determined by means of flow cytometry as detailed
elsewhere.33
In Vivo Testing. All the in vivo experiments described in the
present study were performed on the basis of authorization no.
LA1230509 of the Animal Ethics Committee of the Belgian Federal
Department of Health, Nutritional Safety, and the Environment.
In Vivo Toxicology Study of Narciclasine (1) in Rats. The
study was performed according to ICH Harmonized Tripartite
Guidelines: Nonclinical Safety Studies for the Conduct of Human
Clinical Trials for Pharmaceuticals (16 July 1997) and undertaken
at Notox B.V. (s’-Hertogenbosch, The Netherlands: Notox project
487389 and Unibioscreen reference no. 4818-2-5-001).
In Vivo Evaluation of Anti-Tumor Activity. The details of
each experiment are provided in the Discussion Section and also
(11) Koehn, F. E.; Carter, G. T. The evolving role of natural products in
drug discovery. Nat. ReV. Drug DiscoVery 2005, 4, 206–220.

1114 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Ingrassia et al.
(12) Newman, D. J.; Cragg, G. M. Natural products as sources of new
drugs over the last 25 years. J. Nat. Prod. 2007, 70, 461–477.
(13) Mann, J. Natural products in cancer chemotherapy: past, present and
(32) Mathieu, A.; Remmelink, M.; D’Haene, N.; Penant, S.; Gaussin, J. F.;
Van Ginckel, R.; Darro, F.; Kiss, R.; Salmon, I. Development of a
chemoresistant orthotopic human non-small cell lung carcinoma model
in nude mice. Cancer 2004, 101, 1908–1918.
(33) Mijatovic, T.; Mathieu, V.; Gaussin, J. F.; De Neve, N.; Ribaucour,
F.; Van Quaquebeke, E.; Dumont, P.; Darro, F.; Kiss, R. Cardenolide-
induced lysosomal membrane permeabilization demonstrates thera-
peutic benefits in experimental human non-small cell lung cancers.
Neoplasia 2006, 8, 402–412.
(34) Mijatovic, T.; Op De Beeck, A.; Van Quaquebeke, E.; Dewelle, J.;
Darro, F.; de Launoit, Y.; Kiss, R. The cardenolide UNBS1450 is
able to deactivate nuclear factor kappaB-mediated cytoprotective
effects in human non-small cell lung cancer cells. Mol. Cancer Ther.
2006, 5, 391–399.
(35) Pettit, G. R.; Eastham, S. A.; Melody, N.; Orr, B.; Herald, D. L.;
McGregor, J.; Knight, J. C.; Doubek, D. L.; Pettit III, G. R.; Garner,
L. C.; Bell, J. A. Isolation and structural modification of 7-deoxynar-
ciclasine and 7-deoxy-trans-dihydronarciclasine. J. Nat. Prod. 2006,
69, 7–13.
future. Nat. ReV. Cancer 2002, 2, 143–148.
(14) Altmann, K. H.; Gertsch, J. Anticancer drugs from naturesnatural
products as a unique source of new microtubule-stabilizing agents.
Nat. Prod. Rep. 2007, 24, 327–357.
(15) Ingrassia, L.; Lefranc, F.; Mathieu, V.; Darro, F.; Kiss, R. Amaryl-
lidaceae isocarbostyril alkaloids and their derivatives as promising
antitumor agents. Transl. Oncol. 2008, 1, 1–13.
(16) Kornienko, A.; Evidente, A. Chemistry, biology, and medicinal
potential of narciclasine and its congeners. Chem. ReV. 2008, 108,
1982–2014.
(17) Ceriotti, G. Narciclasine: an antimitotic substance from narcissus bulbs.
Nature 1967, 213, 595–596.
(18) Carrasco, L.; Fresno, M.; Vazquez, D. Narciclasine: an antitumoral
alkaloid which blocks peptide bound formation by eukaryotic riobo-
somes. FEBS Lett. 1975, 52, 236–239.
(19) MacLachlan, A.; Kekre, N.; McNulty, J.; Pandey, S. Pancratistatin: a
natural anticancer compound that targets mitochondria specifically in
cancer cells to induce apoptosis. Apoptosis 2005, 10, 619–630.
(20) Dumont, P.; Ingrassia, L.; Rouzeau, S.; Ribaucour, F.; Thomas, S.;
Roland, I.; Darro, F.; Lefranc, F.; Kiss, R. The Amaryllidaceae
isocarbostyril narciclasine induces apoptosis by activation of the death
receptor and/or mitochondrial pathways in cancer cells but not in
normal fibroblasts. Neoplasia 2007, 9, 766–776.
(21) Hayot, C.; Debeir, O.; Van Ham, P.; Van Damme, M.; Kiss, R.;
Decaestecker, C. Characterization of the activities of actin-affecting
drugs on tumor cell migration. Toxicol. Appl. Pharmacol. 2006, 211,
30–40.
(22) Decaestecker, C.; Debeir, O.; Van Ham, P.; Kiss, R. Can anti-migratory
drugs be screened in vitro? A review of 2D and 3D assays for the
quantitative analysis of cell migration. Med. Res. ReV. 2007, 27, 149–
176.
(23) Kireev, A. S.; Nadein, O. N.; Agustin, V. J.; Bush, N. E.; Evidente,
A.; Manpadi, M.; Ogasawara, M. A.; Rastogi, S. K.; Rogelj, S.; Shors,
S. T.; Kornienko, A. Synthesis and biological evaluation of aromatic
analogues of condritol F, L-chiro-inositol, and dihydroconduritol F
structurally related to the Amaryllidaceae anticancer constituents. J.
Org. Chem. 2006, 71, 5694–5707.
(36) McNulty, J.; Mao, J.; Gibe, R.; Mo, R.; Wolf, S.; Pettit, G. R.; Herald,
D. L.; Boyd, M. R. Studies directed towards the refinement of the
pancratistatin cytotoxic pharmacophore. Bioorg. Med. Chem. Lett.
2001, 11, 169–172.
(37) Rinner, U.; Hillebrenner, H. L.; Adams, D. R.; Hudlicky, T.; Pettit,
G. R. Synthesis and biological activity of some structural modifications
of pancratistatin. Bioorg. Med. Chem. Lett. 2004, 14, 2911–2915.
(38) Abou-Donia, A. H.; De Guilio, A.; Evidente, A.; Gaber, M.; Habib,
A.-A.; Lanzetta, R.; Seif El Din, A. Narciclasine-4-O-ꢀ-D-glucopy-
ranoside, a glucosyloxy amidic phenanthridone derivative from
Pancratium maritimum. Phytochemistry 1991, 30, 3445–3448.
(39) Van Quaquebeke, E.; Mahieu, T.; Dumont, P.; Dewelle, J.; Ribaucour,
F.; Simon, G.; Sauvage, S.; Gaussin, J. F.; Tuti, J.; El Yazidi, M.;
Van Vynckt, F.; Mijatovic, T.; Lefranc, F.; Darro, F.; Kiss, R. 2,2,2-
Trichloro-N-({2-[2-(dimethylamino)ethyl]-1,3-dihydro-1H-benzo[de-
]isoquinolin-5yl}carbamoyl)acetamide (UNBS3157), a novel nonhe-
matotoxic naphthalimide derivative with potent antitumor activity.
J. Med. Chem. 2007, 50, 4122–4134.
(40) Le Mercier, M.; Lefranc, F.; Mijatovic, T.; Debeir, O.; Haibe-Kains,
B.; Bontempi, G.; Decaestecker, C.; Kiss, R.; Mathieu, V. Evidence
of galectin-1 involvement in glioma chemoresistance. Toxicol. Appl.
Pharmacol. 2008, 229, 172–183.
(41) Le Mercier, M.; Mathieu, V.; Haibe-Kains, B.; Bontempi, G.;
Mijatovic, T.; Decaestecker, C.; Kiss, R.; Lefranc, F. Knocking down
galectin-1 in human Hs683 glioblastoma cells impairs both angiogen-
esis through ORP150 depletion and endoplasmic reticulum stress
responses. J. Neuropathol. Exp. Neurol. 2008, 67, 456–469.
(42) Berger, W.; Spiegl-Kreinecker, S.; Buchroithner, J.; Elbling, L.; Pirker,
C.; Fischer, J.; Micksche, M. Overexpression of the human major vault
protein in astrocytic brain tumor cells. Int. J. Cancer 2001, 94, 377–
382.
(24) Nadein, O. N.; Kornienko, A. An approach to pancratistatins via ring-
closing metathesis: efficient synthesis of novel 1-aryl-1-deoxycondu-
ritols F. Org. Lett. 2004, 6, 831–834.
(25) Ingrassia, L.; Wlodarczak, L.; Thomas, S.; Van Quaquebeke E.; Van
den Hove, L.; Kiss, R.; Darro, F. New isocarbostyril alkaloid
derivatives. Patent WO2008043846, 2008.
(26) Evidente, A. Narciclasine 1H- and 13C-NMR data and a new improved
method of preparation. Planta Med. 1991, 57, 293–294.
(43) Griffin, C.; Sharda, N.; Sood, D.; Nair, J.; McNulty, J.; Pandey, S.
Selective cytotoxicity of pancratistatin-related natural Amaryllidaceae
alkaloids: evaluation of the activity of two new compounds. Cancer
Cell Int. 2007, 7, 10.
(27) Pettit, G. R.; Melody, N.; Herald, D. L. Antineoplastic agents: 450.
Synthesis of (+)-pancratistatin from (+)-narciclasine as relay. J. Org.
Chem. 2001, 66, 2583–2587.
(44) Ghosh, M.; Song, X.; Mouneimne, G.; Sidani, M.; Lawrence, D. S.;
Condeelis, J. S. Cofilin promotes actin polymerization and defines the
direction of cell motility. Science 2004, 304, 743–746.
(45) Yap, C. T.; Simpson, T. I.; Pratt, T.; Price, D. J.; Maciver, S. K. The
motility of glioblastoma tumour cells is modulated by intracellular
cofilin expression in a concentration-dependent manner. Cell Motil
Cytoskeleton 2005, 60, 153–165.
(28) Pettit, G. R.; Orr, B.; Ducki, S. Synthesis of pancratistatin prodrugs.
WIPO patent WO02085848, October 31, 2002.
(29) Branle, F.; Lefranc, F.; Camby, I.; Jeuken, J.; Geurts-Moespot, A.;
Sprenger, S.; Sweep, F.; Kiss, R.; Salmon, I. Evaluation of the
efficiency of chemotherapy in in vivo orthotopic models of human
glioma cells with and without 1p19 deletions and in C6 rat orthotopic
allografts serving for the evaluation of surgery combined with
chemotherapy. Cancer 2002, 95, 641–655.
(30) Belot, N.; Rorive, S.; Doyen, I.; Lefranc, F.; Bruyneel, E.; Dedecker,
R.; Micik, S.; Brotchi, J.; Decaestecker, C.; Salmon, I.; Kiss, R.;
Camby, I. Molecular characterization of cell substratum attachments
in human glial tumors relates to prognostic features. Glia 2001, 36,
375–390.
(46) Jansen, R. P. RNA-cytoskeletal associations. FASEB J. 1999, 13, 455–
466.
(47) Me´galizzi, V.; Mathieu, V.; Mijatovic, T.; Gailly, P.; Debeir, O.; De
Neve, N.; Van Damme, M.; Bontempi, G.; Haibe-Kains, B.; Decae-
stecker, C.; Kondo, Y.; Kiss, R.; Lefranc, F. 4-IBP, a sigma1 receptor
agonist, decreases the migration of human cancer cells, including
glioblastoma cells, in vitro and sensitizes them in vitro and in vivo to
cytotoxic insults of proapoptotic and proautophagic drugs. Neoplasia
2007, 9, 358–369.
(31) Roos, W. P.; Batista, L. F.; Naumann, S. C.; Wick, W.; Weller, M.;
Menck, C. F.; Kaina, B. Apoptosis in malignant glioma cells triggered
by the temozolomide-induced DNA lesion O6-methylguanine. Onco-
gene 2007, 26, 186–197.
JM8013585