Anticancer activity of vitamin e-derived compounds in murine c6 glioma cells

Article abstract of DOI:10.1002/cmdc.200900492

A small library of vitamin E analogues, with a free acid group linked to the chroman core through an amide, ether or ester bond, was synthesized and evaluated for their anticancer potency on C6 murine glioma cells. Several compounds showed antitumor activity better than temozolomide, the mostly commonly used drug in the treatment of gliomas.

Full text of DOI:10.1002/cmdc.200900492

MED
DOI: 10.1002/cmdc.200900492
Anticancer Activity of Vitamin E-Derived Compounds in Murine C6 Glioma
Cells
Francesco Mazzini,*[a] Michele Betti,*[b] Barbara Canonico,[b] Thomas Netscher,[c] Francesca Luchetti,[b]
Stefano Papa,[b] and Francesco Galli[d]
Vitamin E (VE)[1] is a family of eight structurally related com-
pounds (four tocopherols and four tocotrienols) derived from
chroman-6-ol. Originally discovered as a dietary factor essential
in several cancer cell models, showing high selectivity for ma-
lignant cells and low toxicity to normal cells.[10,11] It was pro-
posed that the chroman ring of 2 is responsible for the activa-
tion of specific signaling pathways, such as PP2A/PKC, leading
to cell-cycle arrest or apoptosis, while the presence of the suc-
cinyl ester further increases these biological responses leading
to cell and mitochondria membrane destabilization.[9d] Further-
more, methylation of the terminal free carboxylic group leads
to loss of proapoptotic activity.[12] The limiting factor for the
clinical application of 2, in particular regarding oral administra-
tion, is that the ester bond is prone to cleavage by esterases,
releasing a-tocopherol, which is not active as an anticancer
agent. In that regard, a-tocopheryloxyacetic acid (3, aTEA) is a
promising VE analogue, which, unlike 2, has an acetic acid
moiety attached to the chroman hydroxy group via a non-hy-
drolyzable ether bond. Compound 3 was shown to be very ef-
fective in terms of selectivity and potency in suppressing the
growth of various tumors, both in vitro and in vivo.[13,14]
Recently, other VE analogues with non-hydrolyzable bonds
were reported with enhanced proapoptotic activity against
Jurkat, U937 and human Meso-2 malignant mesothelioma cells
compared to compound 2.[15,16] While structurally related to 2,
the ester bond is replaced by an amide bond, the precursors
being a- and d-tocopheramine. Apart from the expected in-
creased stability, a substantial improvement in the apoptotic
activity is observed when going from the ester to the amide
analogues and introducing an olefinic bond in the acid moiety.
Those results suggest that the amide modification can be used
as a starting point for the design of new VE analogues with
potent antitumor activity.
for reproduction in rats,[2] VE has many more important biolog-
ical roles,[3] such as the scavenging of reactive oxygen and ni-
trogen species, thus protecting the organism against the
attack of free radicals,[4] and the modulation of cellular signal-
ing,[5] enzymatic activity and gene expression[6] in antioxidant
and non-antioxidant manners.[7]
In recent years, evidence has accumulated on the antitumor
and anti-inflammatory activity of VE metabolites[8] and synthet-
ic derivatives.[9] The succinic monoester of a-tocopherol (1), a-
tocopheryl succinate (aTOS, 2), a representative VE analogue,
has been reported as a potent cytostatic and cytotoxic agent
The molecular mechanism that causes the induction of
apoptosis by VE analogues is still unclear. Monosuccinate 2
leads to destabilization of lysosomes and mitochondria by
sphingomyelinase activation, ultimately targeting mitochondria
through the generation of reactive oxygen species (ROS),[17] af-
fording the release of cytochrome c and the activation of pro-
apoptotic proteins, such as caspase 9 and the Bcl-2 protein
family.[9d] Apoptosis induced by the ether analogue 3 seems to
proceed through another mechanism involving death receptor
Fas signaling, the activation of JNK and its substrate c-Jun,[18]
truncation of Bid, conformational change of Bax, and activation
of caspases 9 and 3.[19]
[a] Dr. F. Mazzini
Dipartimento di Chimica e Chimica Industriale, Universitꢀ di Pisa
Via Risorgimento 35, 56126 Pisa (Italy)
Fax: (+39)0502219260
E-mail: lcap@ns.dcci.unipi.it
[b] Dr. M. Betti, B. Canonico, F. Luchetti, S. Papa
Dipartimento di Scienze dell’Uomo, dell’Ambiente e della Natura, Universitꢀ
degli Studi di Urbino “Carlo Bo”, Via Cꢀ le Suore 2, 61029 Urbino (Italy)
Fax: (+39)0722304226
E-mail: michele.betti@uniurb.it
Our research involves the systematic study of the application
of VE and related compounds in the treatment of glioblasto-
ma—one of the most frequently occurring brain tumors, ac-
counting for ~12–15% of all brain tumors.[20] Glioblastomas are
among the most aggressive malignant human tumors, charac-
terized by a diffuse local invasion of the normal parenchyma
that renders complete surgical resection of cancerous tissue
[c] Dr. T. Netscher
Research and Development, DSM Nutritional Products
P.O. Box 2676, 4002 Basel (Switzerland)
[d] Dr. F. Galli
Dipartimento di Medicina Interna, Sez. di Biochimica Applicata e Scienze
Nutrizionali, Universitꢀ di Perugia, Via del Giochetto, 06126 Perugia (Italy)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.200900492.
540
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 540 – 543

extremely difficult. Furthermore, the currently available drugs
are of poor therapeutic potential.[21] Thus, novel therapeutic
strategies are required to improve the poor prognosis associat-
ed with this kind of tumor.
in the presence of N,N-diisopropylethylamine, but the subse-
quent alkaline hydrolysis with methanolic KOH provided the
acid 14 in low yield together with compound 15 as the major
product. However, this apparent drawback turned out to give
us another VE analogue for the screening test. The ether ana-
logues 18, 20, 3 and 21 were prepared by reacting 1 with
NaH, followed by the addition of methyl (E)-4-bromo-2-bute-
noate, methyl 4-(bromomethyl)benzoate and ethyl bromoace-
tate, respectively. The subsequent alkaline hydrolysis of the
ester group afforded the acid analogues. Fumaric monoester
17 was obtained by isomerization of 16 by refluxing in CCl4 in
the presence of a catalytic amount of benzyltrimethylammoni-
um tribromide. A stoichiometric amount of the latter reagent
was used in the bromination of 9, affording amide 13. Deriva-
tive 10 was obtained by acid-catalyzed isomerization of 9.
Monoester analogues of amides 11, 12 and 13 proved to be
unstable under purification conditions and could, therefore,
not be isolated.
In previous studies, we demonstrated that 13’-carboxy-d-to-
cotrienol, g-tocopherol, a- and g-(2’-carboxyethyl)-6-hydroxy-
chromans (metabolites of a- and g-tocopherol, respectively)
act as antiproliferative agents on murine glioma C6 cells.[9d,22]
This biological activity is a consequence of cell-cycle arrest in
G0/G1 phase, which is mediated by decreased expression of
cyclin E and cyclin-dependent kinases 2 and 4, and the phos-
phorylative activation of p27 (specific inhibitor of cells entering
S phase). Thus, considering the perspective of a possible che-
motherapeutic application of VE in human gliomas and the
promising results provided by the VE amides discussed above,
we wanted to investigate the activity of the VE amides and
other analogues on a glioma C6 cell model. Therefore, we pre-
pared a small library of VE derivatives with a free acid group
linked to the chroman core via an amide, ether or ester bond,
and assayed them for antiproliferative and apoptotic activity.
By applying the general synthetic strategy shown in
Scheme 1, VE analogues were obtained by reacting a-toco-
pheramine 7[23] or the a-tocopheryloxyl anion with the appro-
priate cyclic anhydride or bromoester (see Table 1 for struc-
tures). In the latter case, saponification of the ester afforded
the acid derivative. Monoester 16 was obtained in improved
yields by generating the a-tocopheryloxyl anion in situ using
NaH rather than using Et3N and DMAP under classic conditions,
whereas the latter procedure conveniently provided amides 8,
9, 11 and 12. Amino acid 14 was synthesized by a similar
approach by reacting 7 with methyl (E)-4-bromo-2-butenoate
All synthesized VE analogues were evaluated in an MTT
assay[24] for their effect on C6 cell viability. The data showed a
marked reduction in cell viability for all VE analogues when
tested at 1 mm, while the tocopherols (1, 4–6) provided poor
to low antiproliferative activity at the same concentration.
Most of the VE analogues exhibited submicromolar IC50 values
(Table 1), and in many cases the values were lower than that of
temozolomide 22, the most commonly used drug in the treat-
ment of gliomas.
In contrast to the results reported using other tumor cell
lines,[15] the substitution of the ester with the amide or amino
bond was slightly detrimental to the cytotoxic activity of these
compounds in C6 cells (8 vs 2, 14 vs 18, 9 vs 16). Moreover,
the introduction of electron
withdrawing groups in the to-
copheramine derivatives (11–13
and 15) further decreased the
antiproliferative activity of these
compounds.
The presence of an olefinic
double bond between the free
acid group and the chroman
ring in the ester and ether ana-
logues did not improve activity
compared with the very good
potency exhibited by com-
pounds 2, 3 and 21. Conversely,
the results provided by com-
pounds 19 and 20 were quite
promising, as, to the best of our
knowledge, they are the first ex-
amples of succinate-like VE ana-
logues with an aromatic group
introduced in the acid portion.
In fact, the aromatic group can
be further functionalized easily,
allowing the development of
other VE derivatives with poten-
Scheme 1. General synthetic strategy for the preparation of VE analogues.
tial antitumor activity.
ChemMedChem 2010, 5, 540 – 543
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
541

MED
the VE amide analogues, the
best apoptosis (A) to necrosis
(N) ratio was exhibited by cyto-
static compound 12 (A/N=4.3),
while for VE ester and ether an-
alogues, a ratio of 13.5 was pro-
vided by compound 19, fol-
lowed by compound 16 (A/N=
6.4) and 3 (A/N=5.0). These
three compounds also showed
the highest pro-apoptotic activi-
ty (3, 75%; 19, 65%; 16, 58%).
Conversely, compound 2 caused
the highest necrotic effect
(85%) at the concentrations
tested in this study. This result
is in contrast to data reported
in other cell lines.[10]
Table 1. Individual IC50 values for the effect of vitamin E compounds on C6 cell viability determined by the
MTT test.[a]
Compd
VE analogue
IC50 ÆSD [mm] Compd
VE analogue
IC50 ÆSD [mm]
1.28Æ0.11
tocopherols
15
1
>50
>50
ester VE analogues
4
2
0.11Æ0.01
5
6
2.00Æ0.13 16
0.21Æ0.02
0.85Æ0.07
>50
17
To investigate the effect of
these compounds on the cell
cycle, further cytofluorimetric
analysis was carried out (fig-
ure 1S, Supporting Information).
This analysis revealed that com-
pounds 9 and 11–15 essentially
caused a cytostatic effect (cell
count: 40–60%, Table 2), com-
pounds 7, 8, 10, 17 and 18
showed moderate antiprolifera-
tive activity (cell count: 30–
40%, Table 2), while compounds
2, 3, 16, 19, 20 and 21 proved
to be very effective inhibitors of
cell growth (cell count: <30%,
amino and amide VE analogues
ether VE analogues
7
0.41Æ0.04
3
0.36Æ0.04
0.74Æ0.06
8
0.53Æ0.05 18
1.04Æ0.08 19
0.68Æ0.07 20
9
0.52Æ0.04
0.45Æ0.04
10
11
12
1.57Æ0.14 21
0.19Æ0.02
0.91Æ0.08 temozolomide
Table 2), succinate
the highest value.
2 showed
13
14
0.86Æ0.09 22
1.25Æ0.10
0.74Æ0.08
In conclusion, a series of VE
analogues, half of them never
described before, were assessed
for their anticancer activity in
murine C6 glioma cells. The
substitution of the ester or
ether bond with an amide bond
to link the free acid group to
[a] IC50 value indicates how much VE analogue is needed to reduce cell viability by 50% compared to a control.
Data represent mean values of at least five separate experiments.
Moreover, although the presence of a free carboxylic group
in VE analogues has been widely reported as a prerequisite for
antitumor activity,[9c,12] this does not appear to be the case for
compounds 19 and 20. We were also quite surprised by the
good activity afforded by tocopheramine 7, a close structural
analogue of 1, which was shown to be approximately 100-fold
more effective than the inactive compound (1).
We also investigated the number of necrotic or apoptotic
cells after 24 h of incubation with 1 mm of test compound by
cytofluorimetric analysis (Table 2). Tocopherols 1, 4–6 affected
the cell viability in a negligible way, while the other VE deriva-
tives provided a moderate to high induction of apoptosis: 18–
43% (VE amides) and 5–75% (VE esters and ethers). Among
the chroman core of VE analogues did not increase their cyto-
toxicity in this cell model. However, nine compounds exhibited
IC50 values better than temozolomide, a golden standard in
the treatment of glioblastoma. While compound 2 provided
the lowest IC50 value, this was due mostly to its necrotic activi-
ty. However, compound 3 showed a very good IC50 value and
the highest pro-apoptotic effect; similar properties were found
for compounds 16 and 21. Contrary to what has generally
been reported for VE analogues,[9c] the presence of a free acid
group did not appear to be an essential requirement for anti-
tumor activity in C6 cells, as shown by amine 7 and ester 19.
The latter showed good growth inhibition of the cancer cells
and the highest apoptosis to necrosis ratio, representing a
542
www.chemmedchem.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2010, 5, 540 – 543

Acknowledgements
Table 2. Effects of compounds 1–22 on cell death parameters of murine
C6 glioma cells.[a]
We are grateful to Dr. A. Cuzzola for MS analyses and Dr. A.
Mandoli for useful discussions. Financial support from the Minis-
tero dell’Universitꢀ e della Ricerca (Project FIRB RBPR05NWWC
and PRIN 2007 (20078TC4E5)) and from the Fondazione per la
Ricerca sulla Fibrosi Cistica (Project FFC#13/2008) is gratefully
acknowledged.
Compd
Cell count[b]
[%]
Residual cells[c] [% of total]
A/N
2
V
A
N
CTRL
tocopherols
1
4
5
6
100Æ2
94Æ1
4Æ2
2Æ1
92Æ3
91Æ3
67Æ4
71Æ2
88Æ2
84Æ3
89Æ1
72Æ3
8Æ2
9Æ2
4Æ2
7Æ3
2.0
1.3
1.8
1.3
7Æ1
4Æ2
16Æ2
12Æ3
Keywords: amides · antitumor agents · apoptosis · cancer ·
amino and amide VE analogues
7
8
9
10
11
12
13
14
15
36Æ3
32Æ2
42Æ3
38Æ3
48Æ2
44Æ4
49Æ3
44Æ4
52Æ3
46Æ5
57Æ4
33Æ4
48Æ6
50Æ5
47Æ7
73Æ3
52Æ4
57Æ4
39Æ6
33Æ4
32Æ4
35Æ6
33Æ6
43Æ8
18Æ3
31Æ5
28Æ4
15Æ5
10Æ4
35Æ4
17Æ5
17Æ6
10Æ7
9Æ3
2.6
3.3
0.9
2.1
2.8
4.3
2.0
1.8
1.9
vitamin E
[1] IUPAC-IUB Joint Commission on Biochemical Nomenclature (JCBN), Eur.
J. Biochem. 1982, 123, 473–475.
[2] H. M. Evans, K. S. Bishop, Science 1922, 56, 650–651.
[3] J.-M. Zingg, Mol. Aspects Med. 2007, 28, 400–422.
[4] R. Brigelius-Flohꢁ, M. G. Traber, FASEB J. 1999, 13, 1145–1155.
[5] F. Galli, M. C. Aisa, C. Annetti, A. Floridi in Encyclopedia of Vitamin E
(Eds.: V. R. Preedy, R. R. Watson), CABI Publishing, Oxford, 2006, Chap-
ter 31.
17Æ5
15Æ4
ester VE analogues
2
16
17
ether VE analogues
3
18
19
14Æ3
10Æ4
33Æ6
40Æ4
5Æ4
58Æ5
40Æ4
85Æ3
9Æ5
0.1
6.4
2.0
[6] J.-M. Zingg, Mol. Aspects Med. 2007, 28, 481–506.
[7] a) M. G. Traber, J. Atkinson, Free Radical Biol. Med. 2007, 43, 4–15; b) A.
Azzi, Free Radical Biol. Med. 2007, 43, 16–21.
23Æ3
37Æ2
20Æ4
[8] a) F. Galli, A. M. Stabile, M. Betti, C. Conte, A. Pistilli, M. Rende, A. Floridi,
A. Azzi, Arch. Biochem. Biophys. 2004, 423, 97–102; b) Q. Jiang, X. Yin,
M. A. Lill, M. L. Danielson, H. Freiser, J. Huang, Proc. Natl. Acad. Sci. U.S.A.
2008, 105, 20464–20469.
[9] a) J.-M. Zingg, Mini-Rev. Med. Chem. 2007, 7, 545–560; b) M. Betti, A.
Minelli, B. Canonico, P. Castaldo, S. Magi, M. C. Aisa, M. Piroddi, V. Di To-
maso, F. Galli, Free Radical Biol. Med. 2006, 41, 464–472; c) J. Neuzil, M.
Tomasetti, Y. Zhao, L F. Dong, M. Birringer, X. F. Wang, P. Low, K. Wu,
B. A. Salvatore, S. J. Ralph, Mol. Pharmacol. 2007, 71, 1185–1199; d) C.
Constantinou, A. Papas, A. I. Constantinou, Int. J. Cancer 2008, 123,
739–752.
[10] J. Neuzil, M. Tomasetti, A. S. Mellick, R. Alleva, B. A. Salvatore, M. Birring-
er, M. W. Fariss, Curr. Cancer Drug Targets 2004, 4, 355–372.
[11] a) J. Neuzil, Br. J. Cancer 2003, 89, 1822–1826; b) K. N. Prasad, B. Kumar,
X. D. Yan, A. J. Hanson, W. C. Cole, J. Am. Coll. Nutr. 2003, 22, 108–117.
[12] K. Kogure, S. Hama, M. Kisaki, H. Takemasa, A. Tokumura, I. Suzuki, K. Fu-
kuzawa, Biochim. Biophys. Acta Gen. Subj. 2004, 1672, 93–99.
[13] T. Hahn, L. Szabo, M. Gold, L. Ramanathapuram, L. H. Hurley, E. T. Akpor-
iaye, Cancer Res. 2006, 66, 9374–9378.
26Æ3
10Æ5
42Æ4
30Æ5
40Æ5
29Æ4
75Æ4
37Æ4
65Æ5
42Æ4
50Æ5
15Æ5
21Æ4
5Æ5
5.0
1.8
13.0
2.3
39Æ2
21Æ2
25Æ3
21Æ3
20
21
18Æ5
21Æ5
2.4
temozolomide
22
23Æ3
45Æ4
31Æ5
24Æ5
1.3
[a] Cell count, viability, apoptosis and necrosis were evaluated by the cy-
tofluorimetric assay after 24 h of exposure to each test compound at the
final concentration of 1 mm. Data represent mean values of at least three
separate experiments. Cell count%=percentage cell count with respect
to cell count of control-vehicle. V=viable, A=apoptotic, N=necrotic,
% RSD=percent relative standard deviation, CTRL=vehicle control.
[b] Cell count is given as percent of CTRL Æ% RSD. The number of cells
in the vehicle control doubled after 24 h, thus a cell count of 50% means
that there was no change in the number of cells after 24 h, thus a cyto-
static effect. [c] Values given are all Æ% RSD.
[14] L. Jia, W. Yu, P. Wang, B. G. Sanders, K. Kline, Prostate 2008, 68, 849–
860.
[15] A. Tomic-Vatic, J. EyTina, J. Chapman, E. Mahdavian, J. Neuzil, B. A. Salva-
tore, Int. J. Cancer 2005, 117, 188–193.
[16] J. Turꢂnek, X.-F. Wang, P. Knçtigovꢂ, S. Koudelka, L.-F. Dong, E. Vrublovꢂ,
E. Mahdavian, L. Prochꢂzka, S. Sangsura, A. Vacek, B. A. Salvatore, J.
Neuzil, Toxicol. Appl. Pharmacol. 2009, 237, 249–257.
[17] T. Weber, H. Dalen, L. Andera, A. Negre-Salvayre, N. Auge, M. Sticha, A.
Lloret, A. Terman, P. K. Witting, M. Higuchi, M. Plasilova, J. Zivny, N. Gel-
lert, C. Weber, J. Neuzil, Biochemistry 2003, 42, 4277–4291.
[18] L. Jia, W. Yu, P. Wang, J. Li, B. G. Sanders, K. Kline, Prostate 2008, 68,
427–441.
[19] W. Yu, M. C. Shun, K. Anderson, H. Chen, B. G. Sanders, K. Kline, Apopto-
sis 2006, 11, 1813–1823.
[20] C. I. Hill, C. S. Nixon, J. L. Ruehmeier, L. M. Wolf, Phys. Ther. 2002, 82,
496–502.
[21] M. C. Tate, M. K. Aghi, Neurotherapeutics 2009, 6, 447–457.
[22] F. Mazzini, M. Betti, T. Netscher, F. Galli, P. Salvadori, Chirality 2009, 21,
519–524.
[23] F. Mazzini, T. Netscher, P. Salvadori, Eur. J. Org. Chem. 2009, 2063–2068.
[24] T. Mosmann, J. Immunol. Methods 1983, 65, 55–63.
slight improvement over the corresponding acid 20. These re-
sults are very promising for the design of new VE analogues
for the treatment of glioma, considering the various possibili-
ties of functionalization of the aromatic moiety linked to the
chroman ring in analogues 19 and 20.
Quite unexpectedly, tocopheramine 7, which is not a succi-
nate-like or ether-linked VE analogue, showed good antiproli-
ferative and apoptotic properties. Notably, just substituting the
hydroxy group in the inactive derivative 1 with an amine
group led to a 100-fold increase in antitumor activity. Further
studies are currently underway to investigate the mechanism-
of-action of these VE analogues and their specific targets, and
also the therapeutic potential of the other tocopheramines
and tocotrienamines that have been recently prepared in our
laboratories.[23]
Received: December 1, 2009
Revised: January 20, 2010
Published online on February 4, 2010
ChemMedChem 2010, 5, 540 – 543
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemmedchem.org
543