-
Curcumin Inhibits the Sonic HedgehogSignaling Pathway and
Triggers Apoptosisin Medulloblastoma Cells
Maha H. Elamin,1 Zakia Shinwari,1 Siti-Faujiah Hendrayani,1
Hindi Al-Hindi,2 Essam Al-Shail,3
Yasser khafaga,4 Amani Al-kofide,5 and Abdelilah
Aboussekhra1,6*1Department of Biological and Medical Research, King
Faisal Specialist Hospital and Research Center, Riyadh, Saudi
Arabia2Department of Pathology, King Faisal Specialist Hospital and
Research Center, Riyadh, Saudi Arabia3Department of Neurosciences,
King Faisal Specialist Hospital and Research Center, Riyadh, Saudi
Arabia4Department of Radio-Oncology, King Faisal Specialist
Hospital and Research Center, Riyadh, Saudi Arabia5Department of
Oncology, King Faisal Specialist Hospital and Research Center,
Riyadh, Saudi Arabia6College of Medicine, Al-Faisal University,
Riyadh, Saudi Arabia
Medulloblastoma is an aggressive primary brain tumor that arises
in the cerebellum of children and young adults.The Sonic Hedgehog
(Shh) signaling pathway that plays important roles in the pathology
of this aggressive disease is apromising therapeutic target. In the
present report we have shown that curcumin has cytotoxic effects
on
medulloblastoma cells. Curcumin suppressed also cell
proliferation and triggered cell-cycle arrest at G2/M
phase.Moreover, curcumin inhibited the Shh–Gli1 signaling pathway
by downregulating the Shh protein and its mostimportant downstream
targets GLI1 and PTCH1. Furthermore, curcumin reduced the levels of
b-catenin, the activate/phosphorylated form of Akt and NF-kB, which
led to downregulating the three common key effectors, namely
C-myc,N-myc, and Cyclin D1. Consequently, apoptosis was triggered
by curcumin through the mitochondrial pathway viadownregulation of
Bcl-2, a downstream anti-apoptotic effector of the Shh signaling.
Importantly, the resistant cellsthat exhibited no decrease in the
levels of Shh and Bcl-2, were sensitized to curcumin by the
addition of the Shh
antogonist, cyclopamine. Furthermore, we have shown that
curcumin enhances the killing efficiency of nontoxic dosesof
cisplatin and g-rays. In addition, we present clear evidence that
piperine, an enhancer of curcumin bioavailability inhumans,
potentiates the apoptotic effect of curcumin against
medulloblastoma cells. This effect was mediated
through strong downregulation of Bcl-2. These results indicate
that curcumin, a natural nontoxic compound,represents great promise
as Shh-targeted therapy for medulloblastomas. � 2009 Wiley-Liss,
Inc.
Key words: chemosensitization; Bcl-2; piperine;
radiosensitization
INTRODUCTION
Medulloblastoma, aggressive tumor of the cere-bellum, is the
most common malignant brain tumorin children. It represents �20% of
pediatric intra-cranial neoplasms [1]. Medulloblastoma, like all
theother types of cancer, results from alterations inthe
equilibrium between cell growth and cell death,which drives the
proliferation of cerebellar granuleneuronal precursors (CGNP) [2].
This equilibrium isunder the control of different metabolic
pathways.The most important medulloblastoma-related carci-nogenesis
pathways are Sonic Hedgehog (Shh),WNT/b-catenin, and Akt/nuclear
factor-kB (NF-kB)[3]. The Shh signaling pathway plays important
rolesin the proliferation of the neuronal precursor of
thecerebellum and in the genesis of medulloblastoma[4,5]. Shh
induces the transcription of two importanttranscription factors
Glioma-associated oncogenehomolog 1 (GLI1) and patched homolog 1
(PTCH1)in many cell types [6]. Furthermore, Shh induces
theexpression of different important oncoproteins,
including the cell-cycle proteins N-myc, C-myc,and Cyclin D1
[1,7]. Like Shh, N-myc activity isnecessary for both normal and
neoplastic cerebellargrowth [8–10]. Amplification of N-myc and/or
C-mycoccurs in 5–10% of medulloblastoma cases [11]. Themyc family
of proteins acts as transcription regulatorsthat play key roles in
cell-cycle progression, trans-formation, and angiogenesis [12].
Amplifications aswell as high transcriptional levels of MYCC
areassociated with an unfavorable survival outcome of
MOLECULAR CARCINOGENESIS 49:302–314 (2010)
� 2009 WILEY-LISS, INC.
Abbreviations: Shh, Sonic Hedgehog; NF-kB, nuclear
factor-kB;GLI1, Glioma-associated oncogene homolog 1; PTCH1,
patchedhomolog 1; DMSO, dimethyl sulfoxide; PBS,
phosphate-bufferedsaline; PI, propidium iodide; IR, ionizing
radiation.
*Correspondence to: King Faisal Specialist Hospital and
ResearchCenter, BMR, MBC # 03-66, PO Box 3354, Riyadh 11211,
SaudiArabia.
Received 4 June 2009; Revised 13 October 2009; Accepted29
October 2009
DOI 10.1002/mc.20604
Published online 18 December 2009 in Wiley
InterScience(www.interscience.wiley.com)
-
medulloblastoma patients [13,14]. The other impor-tant mediator
of the Shh pathway is the anti-apoptosis Bcl-2 protein, which is
frequently overex-pressed in medulloblastomas [15,16]. The
upregula-tion of Bcl-2 and the consequent inhibition ofapoptosis in
Shh-dependent medulloblastoma isAkt/NF-kB-related [16]. In fact
several lines ofevidence suggest that the Atk/NF-kB pathway
syner-gizes with Shh to promote aggressive medulloblas-toma
[16,17].
Currently, medulloblastoma patients are besttreated with
surgery, craniospinal radiotherapy andchemotherapy [18–21].
However, these aggressiveregimens are associated with serious
long-term sideeffects [1,20]. Furthermore, surgery cannot
beachieved in all cases. Therefore, several new ther-apeutic
solutions are currently under investigation.These include the
replacement of combined radio-and chemotherapy by potent and more
specificchemotherapy [3,22]. Thereby, there is continuousdemand for
the identification and the developmentof nontoxic and efficient
anti-medulloblastomaagents. In line with this, several dietary
constituentshave been studied and have in fact shown
greatpreventive and anti-cancer properties, without theadverse side
effects of the currently used chemo-therapeutic agents [23,24].
Curcumin (diferuloyl-methane) is one of the most widely
characterizedphytochemicals. It is a yellow natural
productextracted from the rhizoma of curcuma longa (tur-meric) that
has been used as food spice and colorant.Curcumin is a
polyphenolic, nontoxic, pharmaco-logically active substance that
has anti-oxidant, anti-inflammatory, and antiseptic activities.
Thereby, ithas been used for centuries as therapeutic agentagainst
various diseases [25]. Curcumin has variousanti-cancer
characteristics that include the inhib-ition of cell proliferation
and angiogenesis as well asthe induction of cell death in tumor
cells [26].However, curcumin has only marginal effect onvarious
normal cell types [27]. Furthermore, it hasbeen recently shown that
curcumin induces immu-norestoration in tumor-bearing animals [28],
indi-cating that the inclusion of this natural product
intherapeutic regimens against cancer should bebeneficial for great
proportion of cancer patients. Infact, curcumin has already been
the subject of severalclinical trials for potential use as
chemotherapeuticagent [25].
In this study we investigated the effect of curcuminon
medulloblastoma cells, and we have shown thatthis natural nontoxic
agent inhibits the Shh–Gli1signaling and has great
anti-medulloblastoma effects.
MATERIALS AND METHODS
Cell Lines, Chemicals, and Cell Culture
DAOY cell line was obtained form ATCC, whilethe other
medulloblastoma cells are primary cells
cultured in the laboratory. Cisplatin (cis
diammine-dicloroplatinum II), Curcumin, and piperine (Sigma,St.
Louis, MO) were dissolved in dimethyl sulfoxide(DMSO) and used at
the indicated concentrations.
For the development of primary cell cultures, apiece of the
medulloblastoma sections, obtainedfrom consented patients, was
first placed in a drop ofcomplete media (DMEM:F12, 50:50 medium
supple-mented with 15% NBCS and 1% antibiotics) beforebeing minced
with scalpel blades to very small pieces(
-
Immunoblotting
SDS–PAGE was performed using 12% separatingminigels and equal
amounts of protein extract(30 mg) were loaded. After protein
migration andtransfer onto polyvinylidene difluroide
membrane(PVDF), the membrane was incubated overnightwith the
appropriate antibodies:
NF-kB (F-6), Shh (H-160), b-catenin (9F2), N-myc(3C 165), Cyclin
D1 (HD11), Survivin (C-19), Atk(b-1), p-Akt (104A282), Bcl-2 (C-2),
Bax (B-9),pro-caspase3 (H-277), pro-caspase9 (F-7), PARP(H-250),
NF-M (1A2), HuD (E1), Nestin (10C2),GAPDH (FL-335) and b-actin
(C-11) and C-myc(C-19) from Santa Cruz, CA.
RNA Extraction, cDNA Synthesis, and RT-PCR Assay
Total RNA was extracted using the Tri1 Reagent(Sigma) and the
yield was quantitated spetrophoto-metrically. Following the
manufacturer’s instruc-tions, single stranded cDNA was synthesized
using200 ng of total RNA, the MMLV reverse transcriptaseand the
oligo dT18 (Roche, San Francisco, CA). ThecDNA was amplified for 40
cycles under the follow-ing conditions: melting temperature (958C)
for 50 s,annealing temperature (548C) for 50 s, and
extensiontemperature (728C) for 1 min. The PCR products
wereseparated by electrophoresis on a 2% agarose gel at80 V for an
hour. The sequences of the primers wereas follow:
GLI1, Fw: ACC CGG GGT CTC AAA CTG; Rv: GGCTGA CAG TAT AGG CAG
AGC
PTCH1, Fw: GAC GCC GCC TTC GCT CTG; Rv:GCC CAC AAC CAA GAA CTT
GCC
b-actin, Fw: CCCAGCACAATGAAGATCAAGATCAT;Rv:
ATCTGCTGGAAGGTGGACAGCGA
Quantification of Protein and RNA Expression Levels
The expression levels of RNAs and proteinswere measured using
the densitometer (BIO-RAD,Hercules, CA; GS-800 Calibrated
Densitometer).Films were scanned and protein signal intensity
ofeach band was determined. Next, dividing theobtained value of
each band by the values of thecorresponding internal control
allowed the correc-tion of the loading differences. The fold of
inductionwas determined by dividing the corrected values
thatcorresponded to the treated samples by that of thenontreated
one (time 0).
Annexin V/PI and Flow Cytometry
Confluent cells were either treated with DMSO andused as control
or challenged with different agents,whereupon cells were incubated
in DMEM/F12medium with supplements. Detached and adherentcells were
harvested 72 h later, centrifuged and re-suspended in 1 ml of PBS.
Cells were then stained bypropidium iodide (PI) and Alexa Fluor 488
Annexin
V, using Vibrant Apoptosis Assay kit #2 (Molecularprobe, Eugene,
OR). Stained cells were analyzed byflow cytometry. The percentage
of cells was deter-mined by the FACScadibur apparatus and the
CellQuest Pro software from Becton Dickinson (San Jose,CA). For
each cell culture three independent experi-ments were performed
using 104 cells in eachexperiment.
Cell-Cycle and Cell Death Analysis by Flow Cytometry
Cells were treated with DMSO or curcumin, andthen harvested and
resuspended in 1 ml of PBS beforebeing fixed by drop wise addition
of 3 ml of 100%methanol. Fixed cells were centrifuged,
resuspendedin 50 ml of RNase (1 mg/ml) and incubated for 30 minat
room temperature, followed by addition of 1 mlof 0.1 mg/ml of PI.
Cells were analyzed for DNAcontent by flow cytometry (Becton
Dickinson). Thepercentage of cells in various cell-cycle phases
wasdetermined by using Cell Quest software (BectonDickinson).
RESULTS
Curcumin Has Cytotoxic and Anti-Proliferative Effectson
Medulloblastoma Cells
We investigated the cytotoxic effect of curcuminon different
medulloblastoma primary cells andthe DAOY cell line using the WST-1
assay. Cells wereseeded in triplicates into microtiter plates and
treatedwith increasing concentrations of curcumin for 24 h,and then
the cytotoxic effect was measured. For eachcell culture, at least
three independent experimentswere carried out. Figure 1A shows
dose-dependenteffect of curcumin on four different
medulloblastomacells (MED-1, MED-4, MED-5, and DAOY). WhileMED-1
cells exhibited high resistance to curcumin,the three other cells
showed different sensitivity tothe agent. The LC50 (the
concentration that leads to50% survival) were 20, 25, and 28mM for
the DAOY,MED-5, and MED-4, respectively (Figure 1A). Thisshows that
curcumin is cytotoxic against mostmedulloblastoma cell cultures and
that the DAOYcell line is the most sensitive.
Subsequently, the MED-5 cells were treatedwith 40mM curcumin for
different periods of time(0, 24, 48, and 72 h) and
curcumin-dependent cellkilling was assessed using flow cytometry.
Figure 1Bshows that cell death increased in a time-dependentmanner,
reaching the maximum proportion (58%)after 72 h of treatment. This
confirms the cytotoxiceffect of curcumin against medulloblastoma
cellsand shows that its effect is time dependent.
Next, we investigated the effect of curcuminon medulloblastoma
cell proliferation using theWST-1 cell proliferation assay. Cells
were seededinto microtiter plates and were either mocktreated or
challenged with 40mM of curcumin fordifferent periods of time.
Figure 1C shows thatwhile the number of nontreated cells increased
in
304 ELAMIN ET AL.
Molecular Carcinogenesis
-
a time-dependent manner, the number of curcumin-treated cells
decreased gradually reflecting curcu-min-dependent inhibition of
cell proliferation. Likefor apoptosis the effect was more
pronounced onDAOY cells, while was not significant against theMED-1
cell culture and intermediate for MED-4 andMED-5 (Figure 1C).
After showing the effect of curcumin on cellproliferation we
sought to investigate its effect onthe cell cycle of the MED-4 and
MED-5 cell culturesthat were treated with different concentrations
for3 days. Figure 1D shows a dose-dependent accumu-lation of cells
at the G2/M phase of the cell cyclereaching after 3 days of
treatment 59% and 63%in MED-4and MED-5, respectively. Similar
proportionof G2/M cells was reached at only 20mM of curcuminin
MED-5 cells (Figure 1D). This shows that curcumintriggers G2/M
cell-cycle arrest in medulloblastomacells, which led to growth
inhibition.
Curcumin Inhibits the Sonic Hedgehog Signaling Pathwayin
Medulloblastoma Cells
The Shh signaling pathway is a major regulator ofthe equilibrium
between cell proliferation and cell
death, and therefore is implicated in the develop-ment of the
cerebellar tumor medulloblastoma[1,29]. Thereby, we sought to study
the effect ofcurcumin on this important medulloblastoma-related
pathway. To this end, MED-5 cells weretreated with 40 mM of
curcumin for different periodsof time (0–24 h), and then cellular
lysates wereprepared and used for Western blot analysis
usingspecific antibodies and GAPDH as internal control.Figure 2A
shows that the expression level of theShh protein was downregulated
12.5-fold after 8 hof treatment, reaching a level as low as 8% of
thebasal level. To further appreciate the curcumin effecton the
oncogenic Shh signaling pathway, we studiedits effect on the direct
downstream targets of the Shhprotein, namely PTCH1 and GLI1.
Therefore, MED-5cells were either sham treated or challengedwith 40
mM of curcumin for 8 h. Total RNA waspurified and the levels of the
PTCH1 and the GLI1mRNAs were assessed by RT-PCR. Figure 2B
showsthat the level of both genes decreased significantlyin
response to curcumin treatment. Indeed, thelevel of GLI1 decreased
more than five times andthe PTCH1 level decreased more than
2-fold
Figure 1. Cytotoxic and anti-proliferative effects of curcumin
onmedulloblastoma cells. (A) Exponentially growing cells were
culturedin 96-well plates and treated with the indicated curcumin
concen-trations for 72 h. Cell death was analyzed using the WST-1
assay. Thearrows indicate the LC50 for each cell line and the error
barsrepresent standard deviations. (B) MED-5 cells were treated
with40mM of curcumin and reincubated for the indicated periodsof
time. Cell death was assessed by flow cytometry. The numbers in
the boxes represent the proportion of cell death. (C) Cells
werecultured in 96 wells plates and challenged with curcumin (40
mM) forthe indicated periods of time, and then cell proliferation
wasassessed by the WST-1 assay. Dashed lines: treated cells,
continueslines, nontreated cells. (D) Cells were either mock
treated orchallenged with the indicated concentrations of curcumin
for 72 h.The cell cycle status was analyzed by flow cytometry. The
numbersrepresent the proportions of G2/M cells.
ANTI-MEDULLOBLASTOMA EFFECTS OF CURCUMIN 305
Molecular Carcinogenesis
-
as compared to the control nontreated cells(Figure 2B).
To further elucidate the effect of curcumin on theShh signaling
pathway, we studied the effect ofcurcumin on the main Shh and Gli1
downstreamtargets, N-myc, C-myc, and Cyclin D1. Cyclin D1and N-myc
are important mediators of Shh-inducedproliferation and
carcinogenesis [30]. Importantly,curcumin led to a sharp decrease
in the expressionof these three proteins during the first 4 h of
treat-ment (Figure 2A). The levels of C-myc, Cyclin D1, andN-myc
were reduced 14-, 11-, and 6-fold, respectively(Figure 2A). Since
the anti-apoptosis Bcl2 proteinis also an important mediator of Shh
in medullo-blastoma [15] and its transcriptionally upregulatedby
Shh through the Gli1 transcription factor [31],we sought to
investigate the effect of curcuminon this protein. Figure 2A shows
50% decrease inthe level of Bcl-2 after 8 h of treatment.
Theseresults present the first evidence that curcumininhibits the
Shh–Gli1 signaling in medulloblastomacells.
Next, we investigated whether the inhibition ofthe Shh signaling
pathway affected the levels ofb-catenin and NF-kB, two major
transcription factorsthat are implicated in medulloblastoma
carcino-genesis and interact with Shh [16,17,32]. Interest-
ingly, 8 h of curcumin treatment led to a sharpdecrease in the
level of the NF-kB protein andthe phosphorylated/active form of the
Akt kinase,reaching levels that represents only 6% and 35% ofthe
basal level, respectively (Figure 2C). Likewise,the level of the
transcription factor b-cateninwas strongly downregulated in
response to curcuminreaching a level 10-fold lower than the
corres-ponding basal level, after only 4 h of treatment(Figure
2C).
Among the cancer markers that are under thecontrol of the NF-kB
pathway there is the anti-apoptosis protein survivin. Figure 2C
shows thatcurcumin treatment led to the downregulation ofsurvivin
as well. The reduction in the level of thisprotein was time
dependent in medulloblastomacells reaching a level 12-fold lower
after 24 h oftreatment (Figure 3C).
It is noteworthy that the effect of curcumin onthese proteins
and pathways was also observed in theDAOY cell line (data not
shown), indicating that thiseffect could be considered as general
to mostmedulloblastoma cells.
Together, these results indicate that curcuminhas great
inhibitory effect on the Shh signaling path-way and its downstream
medulloblastoma-driveneffectors.
Figure 2. Curcumin inhibits the Shh–Gli1 signaling pathway.MED-5
cells were not treated or challenged with curcumin (40 mM)and then
reincubated for the indicated periods of time. (A and C)Whole cell
extracts were prepared and used for immunoblot analysisusing the
indicated antibodies. (B) Total RNA was extracted from
these cells, and the cDNA was synthesized and used to evaluate
themRNA expression for the indicated genes using the
RT-PCRtechnique. b-Actin was used as internal control. The numbers
underthe bands represent the corresponding expression levels
ascompared to the basal level (time 0, control).
306 ELAMIN ET AL.
Molecular Carcinogenesis
-
Curcumin Triggers Apoptosis Through the MitochondrialPathway in
Medulloblastoma Cells
Since the inhibition of the Shh-dependent signal-ing pathway
triggers apoptosis in medulloblastoma[33,34], we investigated
whether curcumin triggersapoptosis in these cells. To this end, the
Annexin V/PI staining technique followed by flow cytometry
was used. Sixty percent confluent cells were treatedwith
different concentrations of curcumin for 3 days,and then were
stained and sorted. Figure 3A showsfour groups of cells, viable
cells that excluded bothAnnexin V and PI (Annexin V�/PI�), bottom
left;early apoptotic cells that were only stained withAnnexin V
(Annexin Vþ/PI�), bottom right; lateapoptotic cells that were
stained with both Annexin
Figure 3. Curcumin triggers apoptosis through the internal
path-way in medulloblastoma cells. Sub-confluent cells were either
mocktreated or challenged with the indicated concentrations of
curcuminfor 72 h and then cell death was analyzed using the Annexin
V/PIflow cytometry assay. (A) Charts, with the numbers indicating
theproportion of apoptotic and necrotic cells. (B) Graph
representing thedose-dependent apoptosis in the indicated cells.
(C) Histogram, cellswere treated with 40 mM curcumin for 72 h. M
for MED. (D) The
MED-5 cells were treated with 40mM curcumin for the
indicatedperiods of time, and then cell extracts were prepared and
used forimmunoblotting analysis using the indicated antibodies. the
numbersunder the bands represent the corresponding expression
levels ascompared to the basal level (time 0). (E) Graph showing
the Bax/bcl-2ratio. Error bars represent standard deviations of
three differentexperiments.
ANTI-MEDULLOBLASTOMA EFFECTS OF CURCUMIN 307
Molecular Carcinogenesis
-
V and PI (Annexin Vþ/PIþ), top right and necroticcells that were
only stained with PI (Annexin V�/PIþ), top left. The proportion of
apoptosis wasconsidered as the sum of both early and lateapoptosis
after deduction of the proportion ofspontaneous apoptosis. Curcumin
triggered essen-tially apoptosis in all medulloblastoma cells.
Thiseffect increased in a dose-dependent manner, andcells showed
different responses, with the highestproportion of apoptosis
induced in the DAOY cellline (Figure 3A and B). This parallels the
cytotoxicresult presented in Figure 1A, indicating thatthe
curcumin-dependent cytotoxicity is mediatedmainly through the
apoptotic cell death pathway,with only marginal necrosis. It is
noteworthy thatthere is a great correlation between the proportion
ofcell death detected by WST-1 (Figure 1A) and byAnnexin V (Figure
3A and B).
To shed more light on the effect of curcumin onmedulloblastoma
cells we treated 11 medulloblas-toma cells (10 primary cultures and
the DAOY cellline) with 40 mM of curcumin for 3 days and then
theapoptotic response was assessed. Figure 3C showsthat curcumin
has differential effect on medullo-blastoma cells. While eight
cells exhibited highsensitivity with more than 40% cell death
throughapoptosis, three cell cultures showed high resistanceto the
drug. MED-5, MED-7, and DAOY were themost sensitive with 80% cell
death, whilst MED-1,and MED-9 were the most resistant with about
30%apoptotic cells (Figure 3C).
To confirm the induction of apoptosis by curcu-min in
medulloblastoma cells and determine theapoptotic route that
curcumin-dependent Shhinhibition activates, the MED-5 cells were
treatedwith 40 mM of curcumin and harvested afterdifferent time
periods (0, 24, 48, and 72 h). Wholecell extracts were prepared and
50 mg of extractedproteins were used to evaluate the levels of
differentpro- and anti-apoptotic proteins using the immuno-blot
technique and specific antibodies. b-Actin andGAPDH were used as
internal controls. First, weassessed the effect of curcumin on the
caspase-3 andPARP proteins (two principal markers of
apoptosis).Figure 3D shows that the level of pro-caspase 3decreased
fivefolds after 48 h of curcumin treatment.Concomitantly, the level
of the cleaved form of thePARP protein increased significantly
after 72 h oftreatment. This clearly shows the induction
ofapoptosis by curcumin in medulloblastoma cells.Next, we assessed
the levels of Bax and Bcl-2 proteinsand have found that while the
level of Bax increasedin a time-dependent manner, the level of
Bcl-2decreased sharply after only 24 h of treatment andthen
increased slightly (Figure 3D). This resultedafter 48 h of
treatment in 15-fold increase in theBax/Bcl-2 ratio (Figure 3E),
showing that curcumintriggers apoptosis through the mitochondrial
path-way. To confirm this, we assessed the level of the pro-
caspase 9 in these cells, and showed that the level ofthis
protein decreased in a time-dependent mannerreaching a level more
than threefold lower after 72 hof treatment (Figure 3D). Together,
these resultsdemonstrated that curcumin triggers apoptosisin
medulloblastoma cells through the internalmitochondrial pathway via
Bcl-2 decrease.
The Shh Antagonist Cyclopamine SensitizesCurcumin-Resistant
Medulloblastoma Cells
To further elucidate the role of the Shh pathway
incurcumin-dependent induction of apoptosis inmedulloblastoma
cells, we investigated the effect ofcurcumin on the Shh pathway in
the curcumin-resistant MED-1 cells. Figure 4A shows that
curcumindid not significantly affect the expression of the Shhand
its important downstream protein Cyclin D1 inMED-1 cells. This
indicates that the resistance ofthese cells to curcumin could be
due to the inabilityof curcumin to inhibit the Shh signaling
pathway. Toinvestigate this possibility we explored
whethercyclopamine, a natural antagonist of the Shh signal-ing
pathway [35], could sensitize MED-1 cells tocurcumin.
Cells were treated with curcumin (20mM), cyclop-amine (10 and
20mM), and the combinations ofboth agents, for 72 h and then the
proportion ofapoptosis was determined. Figure 4B shows thatwhile
the effect of the single agents is only marginalthe combination of
curcumin and cyclopaminetriggered high proportion of apoptotic
cells. Indeed,more than 80% of cells died through apoptosis
whencyclopamine (20 mM) was added to curcumin (20 mM)(Figure 4C).
Similar results were obtained with theMED-9 cells (data not shown).
This clearly shows thatMED-1 and MED-9 cells are resistant to
curcumin-dependent inhibition of the Shh signaling and thatcurcumin
and cyclopamine have synergistic effect ininducing apoptosis in
medulloblastoma cells.
Interestingly, in the curcumin-resistant MED-1cells, wherein
curcumin increased slightly the levelof the Shh protein, the levels
of Cyclin D1 and Bcl-2were not downregulated as in the
curcumin-sensitiveMED-5 cells (Figure 2) but rather increased
followingtreatment with curcumin (40mM) (Figure 4D). Thisshows that
the pro-apoptotic effect of cucumin inmedulloblastoma cells could
be mediated throughthe downregulation of Shh and the
anti-apoptosisprotein Bcl-2. Similarly, the anti-apoptosis
proteinsurvivin was also highly upregulated followingtreatment of
MED-1 cells with curcumin (Figure4D), which provides further
explanation to the highresistance of these cells to curcumin.
Curcumin Potentiates the Apoptotic Effects ofCisplatin and
g-Rays
Ionizing radiation (IR) and cisplatin are the majortherapeutic
agents for medulloblastoma [1]. There-fore, we sought to test the
ability of curcumin, as
308 ELAMIN ET AL.
Molecular Carcinogenesis
-
inhibitor of the Shh pathway, to potentiate the effectof the DNA
damaging agents g-rays and cisplatinin triggering apoptosis in
medulloblastoma cells.Therefore, MED-5 cells were either mock
treated andused as control or challenged with curcumin(20mM),
g-rays (5 Gy), cisplatin (6 mM), and thecombination of curcumin
with cisplatin or withg-rays, and then the proportion of induced
apoptosiswas assessed by the Annexin V/flow cytometrytechnique
(Figure 5A and B). Importantly, whilethe proportion of apoptotic
cells was only marginalin response to each of the single agents,
thecombination of curcumin with cisplatin or withIR triggered
apoptosis in more than 70% of cells,showing a great synergistic
effect between curcuminand both cisplatin and IR (Figure 5A and B).
Thisclearly shows that curcumin can potentiate the effectof both
g-rays and cisplatin in inducing apoptosis inmedulloblastoma cells.
Similar results were obtainedwith the DAOY cell line (data not
shown). Toelucidate the molecular basis of this potentiationwe
investigated the effect of the single agents and thecombinations on
the levels of Bax and Bcl-2 in MED-
5 cells using the immunoblotting technique.Figure 5C shows that
while the effect of curcuminand cisplatin on the level of Bcl-2 was
only marginal,g-rays downregulated Bcl-2 level 5-fold, 48 h
pos-tirradiation. Interestingly, the combination of cur-cumin with
cisplatin led to a time-dependentdownregulation of Bcl-2 reaching a
level 10-foldlower after 72 h of treatment. Likewise,
curcuminenhanced the g-ray effect on Bcl-2 leading to a
level20-fold lower only 24 h postirradiation (Figure 5C).On the
other hand, the level of Bax did notsignificantly change following
all the treatments(data not shown). These results show that
curcuminpotentiates the action of cisplatin and g-rays intriggering
cell death through apoptosis by enhanc-ing the downregulation of
the anti-apoptosis Bcl-2protein.
Piperine Potentiates the Effect of Curcumin inTriggering
Apoptosis in Medulloblastoma Cells
It has been previously shown that piperineincreases the curcumin
bioavailability by 2000%[36]. To test whether this natural product
can have
Figure 4. Cyclopamine sensitizes curcumin-resistant
medulloblas-toma cells. (A and D) Curcumin (40 mM) treated MED-1
cells wereincubated for the indicated periods of time and proteins
wereextracted and used for immunoblot analysis using the
indicatedantibodies. The numbers under the bands represent the
correspond-ing expression levels as compared to the basal level
(time 0). (B) MED-
1 cells were either mock treated or challenged with curcumin
orcyclopamine or the combination of both. Cells were then
reincu-bated for 72 h and cell death was assessed by Annexin V/PI
flowcytometry. The numbers in the charts represent the proportions
ofnecrotic and apoptotic cells. (C) Histogram showing the
proportionof apoptotic cells.
ANTI-MEDULLOBLASTOMA EFFECTS OF CURCUMIN 309
Molecular Carcinogenesis
-
enhancer effect on curcumin-dependent apoptosisinduction, both
products were used separately or incombination against MED-1,
MED-4, MED-5, andDAOY cells, and apoptosis was assessed as
describedabove. We have found that piperine triggers apop-tosis in
medulloblastoma cells in a dose-dependentmanner. In fact 100 and
200 mM of piperine increasedthe proportion of apoptotic cells from
10% to 31%and 57%, respectively (data not shown). Further-more,
piperine enhanced the killing effect of curcu-min on all cells
tested (Figure 6A and B). The effectcould be considered as additive
for MED-4 andDAOY. However, it is rather synergistic for MED-1and
MED-5 cells. Importantly, the combination ofboth agents triggered
apoptosis in more than 70% ofthe highly resistant MED-1 cell
culture (Figure 6B).This suggests that the combination is very
effectiveagainst resistant medulloblastoma cells.
To elucidate the molecular basis of the potentia-tion of the
curcumin effect by piperine, we inves-tigated the effect of the
combination on the level ofthe Bax and Bcl-2 proteins in the
curcumin-resistant
MED-1 cells. Figure 6C shows that while the effect ofthe single
agents is insignificant on the expressionlevels of Bax and Bcl-2,
the combination of curcuminand piperine led to 4-folds increase in
the level of Baxand 100-fold decrease in the level of Bcl-2,
whichresulted in 400-fold increase in the ratio Bax/Bcl-2(Figure
6D). This shows that piperine enhances theapoptotic effect of
curcumin by increasing its effectin downregulating the
anti-apoptosis protein Bcl-2.
DISCUSSION
In the present report, we present clear evidencethat curcumin
could constitute a potent anti-medulloblastoma agent for the
following reasons:
(1) Curcumin has an outstanding safety profile.Indeed, different
phase I clinical trials haveshown that curcumin is safe when
consumed atdoses as high as 12 g per day for 3 months[25,37].
(2) Curcumin crosses the brain blood barrier andreaches the
brain. Indeed, it has been shown that
Figure 5. Curcumin enhances the cytotoxic effect of cisplatin
andg-rays through strong Bcl-2 decrease. MED-5 cells were either
mocktreated or challenged with curcumin (20 mM) or cisplatin (6 mM)
or g-rays (5 Gy) or the combination of curcumin with either
cisplatin or g-rays. Cells were then reincubated for different
periods of time asindicated. Cell death was assessed by Annexin
V/PI flow cytometry
and the level of Bcl-2 was evaluated by immunoblotting. (A)
Chartswith the numbers representing the proportions of necrotic
andapoptotic cells. (B) Histogram showing the proportion of
apoptoticcells. The error bars represent standard deviations. (C)
Immunoblots,the numbers under the bands represent the
correspondingexpression levels as compared to basal level (time
0).
310 ELAMIN ET AL.
Molecular Carcinogenesis
-
curcumin can reach different regions of the brainincluding the
cerebellum in rats [38] andmice [39–41]. In fact, accumulating data
invarious experimental models have shown thatdietary curcumin has
neuroprotective effectsand therefore is a strong candidate for use
inthe prevention or treatment of major disablingage-related
neurodegenerative diseases, likeAlzheimer’s, Parkinson’s, and
stroke. Thesepromising results have already led to ongoingpilot
clinical trials [42].
(3) We have shown here that curcumin inhibitsmedulloblastoma
cell proliferation by enablingthe arrest of the cell-cycle at G2/M
phase.
(4) Curcumin is known to influence several bio-chemical and
molecular pathways that playimportant roles in the development and
pro-gression of various types of cancer [37,43]. In thepresent
study we present the first evidence thatcurcumin inhibits the Shh
signaling. Indeed,curcumin down regulated the Shh protein andalso
its main targets in the pathway Gli1 and
PTCH1 (Figure 2). The Shh pathway playsessential roles in the
cerebellar developmentand neoplastic transformation [4,5].
Further-more, activation of the Shh signaling
promotesmedulloblastoma from both neuronal progeni-tors and stem
cells [44,45]. Moreover, blockade ofthe Shh pathway led to
medulloblastoma growthinhibition [33]. Thereby, the
curcumin-depend-ent inhibition of Shh signaling could be of
greattherapeutic value against medulloblastomas.Besides, we have
shown that curcumin down-regulates the expression of many
importantcancer proteins implicated in different signalingpathways
that have been found deregulatedin various cancers, including
medullo-blastoma. Indeed, curcumin inhibited two otherimportant
medulloblastoma-dependent carci-nogenesis pathways Akt/NF-kB and
b-catenin(Figure 2C). In fact it has been recently shownthat the
AKT signaling pathway is activated inmedulloblastomas, which
exhibited significantexpression of phosphorylated Akt [46]. In
the
Figure 6. Piperine potentiates the pro-apoptotic effect of
curcu-min through strong downregulation of Bcl-2 in
medulloblastomacells. Cells were either mock treated or challenged
with curcumin(20 mM) or piperine (100 mM) or the combination of
curcumin withpiperine. (A) Cells were then reincubated for 72 h and
cell death wasassessed by Annexin V/PI flow cytometry. Charts, the
numbersrepresent the proportions of necrotic and apoptotic cells.
(B)
Histogram showing the proportion of apoptotic cells. The error
barsrepresent standard deviations. (C) MED-1 cells were reincubated
forthe indicated periods of time and then cell lysates were
prepared andused for immunobloting analysis using the indicated
antibodies. Thenumbers under the bands represent the
induction/reduction levels ofthe Bax and Bcl-2 proteins. (D) Graphs
showing the Bax/Bcl2 ratios.
ANTI-MEDULLOBLASTOMA EFFECTS OF CURCUMIN 311
Molecular Carcinogenesis
-
present study we have shown that curcumindownregulates the
phosphorylated/active formof Akt, which is able to increase the
incidence ofShh-induced medulloblastoma [17]. Similarly,the level
of the oncogenic transcription factorb-catenin was also
downregulated by curcuminin medulloblastoma cells. Several lines
ofevidence indicate functional interactionbetween the two pathways
Shh and Wnt/b-catenin [47]. It has been recently shown thatthe
Wnt/b-catenin signaling is required for Hhpathway-driven
tumorigenesis [32]. Therefore, itis very important to block Wnt
signaling, givenits activation in medulloblastomas [21].
Interestingly, the activation of these three path-ways
Akt/NF-kB, Shh, and Wnt/b-catenin convergein the upregulation of
C-myc, N-myc, and Cyclin D1,three oncoproteins that play key roles
in thedevelopment of medulloblastomas [1,7,10,29]. Highlevels of
these three oncoproteins were consideredto be bad prognosis because
they are related tounfavorable therapeutic outcome [19]. N-myc
proto-oncogene, a member of the myc family of transcrip-tional
activators [48], is required for cerebellargranule neuron precursor
proliferation [8,9], andtherefore it is an important player in the
carcino-genesis of medulloblastoma. Our data show thatcurcumin has
strong inhibitory effect on theseproteins in medulloblastoma cells,
which furthersupports the inhibition of the three
importantmedulloblastoma-driven pathways and the
efficientanti-medulloblastoma effect of curcumin. The factthat
these pathways have anti-apoptosis effects,suggests that the
pro-apoptotic action of curcuminagainst medulloblastoma cells is
mediated throughthe inhibition of these pathways and the
down-stream anti-apoptosis protein Bcl-2.
(5) We have also shown here that in mostcases, curcumin
triggered cell death in highproportions of medulloblastoma cells
mainlythrough the apoptotic pathway. This seems totake place by
downregulating different anti-apoptosis proteins, including Bcl-2
and survivin.Survivin is a potent anti-apoptosis protein that
isdifferentially expressed in cancer and thereforeconstitutes an
important anti-cancer target [49].Moreover, high expression of
survivin playsimportant role in resistance to chemo-
andradiotherapy and has been shown to be relatedto unfavorable
outcome for medulloblastomas[50]. Importantly, our data have shown
that 24 hof curcumin treatment can reduce survivin to30% of the
basal level in medulloblastoma cells.Similar effect has been shown
on different othercancer cell lines [25]. Interestingly, when
thelevels of Bcl-2 and survivin were assessed incurcumin-resistant
medulloblastoma cells MED-
1 and MED-9 an increase rather than decreasewas observed.
Concomitantly, no decrease in thelevel of the Shh protein and its
downstreamtarget Cyclin D1 was observed, suggesting
thatcurcumin-dependent induction of apoptosis ismediated through
Shh/Bcl-2 downregulation.Indeed, the natural Shh antagonist
cyclopamine,sensitized the MED-1 cells to curcumin(Figure 4).
These results also showed that these two plant-derived
Shh-antagonist products have synergisticeffect against
medulloblastoma, and this could be ofgreat value for the treatment
of these aggressivetumors. Indeed, both agents target pathways that
arecrucial for medulloblastoma survival. Furthermore,cyclopamine
has been considered for the treatmentof medulloblastoma patients
[51,52].
(6) Curcumin sensitized medulloblastoma cells to g-rays and
cisplatin, two widely used agents for thetreatment of these
neoplasms [1,20]. Indeed, thecombination of curcumin with nontoxic
doses ofeach agent led to 80% cell death in
differentmedulloblastoma cells. This effect has beenmediated
through a strong downregulation ofthe anti-apoptosis protein Bcl-2
(Figure 5C). Ithas been recently shown that curcuminenhanced the
anti-tumor effects of gemcitabineand radiation in the PC3 human
prostatecancer cell line through downregulating theMDM2 oncogene
[53] and pancreatic cancercells in vitro and in vivo [54]. In
anotherstudy, curcumin potentiated taxol-inducedapoptosis by
downregulating NF-kB and Akt[55]. Together, these results show that
curcumincan be used to potentiate the activity andreduce the
nondesirable side effects of somechemotherapeutic agents and
IR.
(7) Curcumin-dependent induction of apoptosiswas enhanced by
piperine, an enhancer ofcurcumin bioavailability. The fact that
curcumincan induce the expression of different importanttumor
suppressor proteins and inhibit theexpression of various key
oncoproteins indicatesthat this agent could be considered as great
anti-medulloblastoma product. However, curcuminis limited in its
clinical utility owing to its lowbioavailability. Indeed, several
reports haveshown that curcumin is rapidly degraded. Majorreasons
contributing to the low plasma andtissue levels of curcumin appear
to be due topoor absorption, rapid metabolism, and rapidsystemic
elimination [56]. Interestingly, it hasbeen shown that piperine,
another naturalpolyphenolic nontoxic agent that inhibits hep-atic
and intestinal glucuronidation, can signifi-cantly enhance the
curcumin absorption leveland bioavailability in both rats and
humans by
312 ELAMIN ET AL.
Molecular Carcinogenesis
-
154% and 2000%, respectively [36]. Importantly,we have shown
here that piperine is cytotoxicand enhances also the killing effect
of curcuminagainst medulloblastoma cells. Strikingly, piper-ine
sensitized the cell cultures that were resistantto the killing
effect of curcumin. We have alsoshown that the synergistic effect
of curcuminand piperine was mediated through down regu-lating the
most important anti-apoptosis pro-tein, Bcl-2. Indeed, the
combination of bothagents decreased by 100-fold the level of
thisprotein (Figure 6). This finding is of greatimportance, since
it has been recently shownthat Bcl-2 is an important mediator of
the Shhactivity in medulloblastoma [15]. The fact thatcurcumin
downregulates both Shh and Bcl-2suggests that this agent in
combination withpiperine could constitute a great
anti-medullo-blastoma agent.
Together, these findings provide clear evidencethat curcumin
inhibits the Shh–Gli1 signaling andtriggers cell growth inhibition
and the inductionof cell death through the internal
apoptoticpathway in medulloblastoma cells. In addition, itenhances
the killing effect of cisplatin and g-raysand targets pathways
crucial for tumor survival.Furthermore, piperine enhances curcumin
anti-cancer effect showing that the combination of thesetwo natural
and safe products could be of great valueif included in the
medulloblastoma therapeuticregimens.
ACKNOWLEDGMENTS
We are thankful to Dr. K. Al-Hussein and P.S.Manogaran for their
help with the flow cytometry.This work was performed under the RAC
proposal #2050016.
REFERENCES
1. Jozwiak J, Grajkowska W, Wlodarski P. Pathogenesis
ofmedulloblastoma and current treatment outlook. Med ResRev
2007;6:869–890.
2. Hanahan D, Weinberg RA. The hallmarks of cancer.
Cell2000;100:57–70.
3. Gilbertson RJ. Medulloblastoma: Signalling a change
intreatment. Lancet Oncol 2004;5:209–218.
4. Wechsler-Reya RJ, Scott MP. Control of neuronal
precursorproliferation in the cerebellum by Sonic Hedgehog.
Neuron1999;22:103–114.
5. Ho KS, Scott MP. Sonic hedgehog in the nervous
system:Functions, modifications and mechanisms. Curr Opin
Neuro-biol 2002;12:57–63.
6. Hooper JE, Scott MP. Communicating with Hedgehogs. NatRev Mol
Cell Biol 2005;6:306–317.
7. Kenney AM, Widlund HR, Rowitch DH. Hedgehog and PI-3kinase
signaling converge on Nmyc1 to promote cell cycleprogression in
cerebellar neuronal precursors. Development2004;131:217–228.
8. Kenney AM, Cole MD, Rowitch DH. Nmyc upregulationby sonic
hedgehog signaling promotes proliferation in
developing cerebellar granule neuron precursors. Develop-ment
2003;130:15–28.
9. Knoepfler PS, Cheng PF, Eisenman RN. N-myc is essentialduring
neurogenesis for the rapid expansion of progenitorcell populations
and the inhibition of neuronal differ-entiation. Genes Dev
2002;16:2699–2712.
10. Hatton BA, Knoepfler PS, Kenney AM, et al. N-myc is
anessential downstream effector of Shh signaling during bothnormal
and neoplastic cerebellar growth. Cancer Res2006;66:8655–8661.
11. Aldosari N, Bigner SH, Burger PC, et al. MYCC and
MYCNoncogene amplification in medulloblastoma. A fluorescencein
situ hybridization study on paraffin sections from theChildren’s
Oncology Group. Arch Pathol Lab Med 2002;126:540–544.
12. Vita M, Henriksson M. The Myc oncoprotein as a
therapeutictarget for human cancer. Semin Cancer Biol
2006;16:318–330.
13. Scheurlen WG, Schwabe GC, Joos S, Mollenhauer J,Sorensen N,
Kuhl J. Molecular analysis of childhood primitiveneuroectodermal
tumors defines markers associated withpoor outcome. J Clin Oncol
1998;16:2478–2485.
14. Grotzer MA, Hogarty MD, Janss AJ, et al. MYC messengerRNA
expression predicts survival outcome in childhoodprimitive
neuroectodermal tumor/medulloblastoma. ClinCancer Res
2001;7:2425–2433.
15. Bar EE, Chaudhry A, Farah MH, Eberhart CG. Hedgehogsignaling
promotes medulloblastoma survival via BclII. Am JPathol
2007;170:347–355.
16. McCall TD, Pedone CA, Fults DW. Apoptosis suppression
bysomatic cell transfer of Bcl-2 promotes Sonic hedgehog-dependent
medulloblastoma formation in mice. Cancer Res2007;67:5179–5185.
17. Rao G, Pedone CA, Del Valle L, Reiss K, Holland EC, Fults
DW.Sonic hedgehog and insulin-like growth factor signalingsynergize
to induce medulloblastoma formation from nestin-expressing neural
progenitors in mice. Oncogene 2004;23:6156–6162.
18. Gajjar A, Chintagumpala M, Ashley D, et al.
Risk-adaptedcraniospinal radiotherapy followed by high-dose
chemo-therapy and stem-cell rescue in children with newlydiagnosed
medulloblastoma (St Jude Medulloblastoma-96): Long-term results
from a prospective, multicentre trial.Lancet Oncol
2006;7:813–820.
19. Brandes AA, Paris MK. Review of the prognostic factors
inmedulloblastoma of children and adults. Crit Rev OncolHematol
2004;50:121–128.
20. Rood BR, Macdonald TJ, Packer RJ. Current treatment
ofmedulloblastoma: Recent advances and future challenges.Semin
Oncol 2004;31:666–675.
21. Rossi A, Caracciolo V, Russo G, Reiss K, Giordano
A.Medulloblastoma: From molecular pathology to therapy.Clin Cancer
Res 2008;14:971–976.
22. Grill J, Sainte-Rose C, Jouvet A, et al. Treatment
ofmedulloblastoma with postoperative chemotherapy alone:An SFOP
prospective trial in young children. Lancet
Oncol2005;6:573–580.
23. Garg AK, Buchholz TA, Aggarwal BB. Chemosensitizationand
radiosensitization of tumors by plant polyphenols.Antioxid Redox
Signal 2005;7:1630–1647.
24. Surh YJ. Cancer chemoprevention with dietary
phytochem-icals. Nat Rev Cancer 2003;3:768–780.
25. Goel A, Kunnumakkara AB, Aggarwal BB. Curcumin
as‘‘Curecumin’’: From kitchen to clinic. Biochem
Pharmacol2007;19:19.
26. Aggarwal BB, Kumar A, Bharti AC. Anticancer potential
ofcurcumin: Preclinical and clinical studies. Anticancer
Res2003;23:363–398.
27. Syng-Ai C, Kumari AL, Khar A. Effect of curcumin on
normaland tumor cells: Role of glutathione and bcl-2. Mol
CancerTher 2004;3:1101–1108.
ANTI-MEDULLOBLASTOMA EFFECTS OF CURCUMIN 313
Molecular Carcinogenesis
-
28. Bhattacharyya S, Mandal D, Saha B, Sen GS, Das T, Sa
G.Curcumin prevents tumor-induced T cell apoptosis
throughStat-5a-mediated Bcl-2 induction. J Biol Chem
2007;282:15954–15964.
29. Raffel C. Medulloblastoma: Molecular genetics and
animalmodels. Neoplasia 2004;6:310–322.
30. Oliver TG, Grasfeder LL, Carroll AL, et al.
Transcriptionalprofiling of the Sonic hedgehog response: A critical
role forN-myc in proliferation of neuronal precursors. Proc Natl
AcadSci USA 2003;100:7331–7336.
31. Bigelow RL, Chari NS, Unden AB, et al.
Transcriptionalregulation of bcl-2 mediated by the sonic hedgehog
signal-ing pathway through gli-1. J Biol Chem
2004;279:1197–1205.
32. Hoseong Yang, S, Andl, T, Grachtchouk, V, et al.
Patho-logical responses to oncogenic Hedgehog signaling in skinare
dependent on canonical Wnt/beta-catenin signaling. NatGenet
2008;40:1130–1135.
33. Berman DM, Karhadkar SS, Hallahan AR, et al.
Medullo-blastoma growth inhibition by hedgehog pathway
blockade.Science 2002;297:1559–1561.
34. Romer JT, Kimura H, Magdaleno S, et al. Suppression of
theShh pathway using a small molecule inhibitor
eliminatesmedulloblastoma in Ptc1(þ/�)p53(�/�) mice. Cancer
Cell2004;6:229–240.
35. Chen JK, Taipale J, Cooper MK, Beachy PA. Inhibition
ofHedgehog signaling by direct binding of cyclopamine toSmoothened.
Genes Dev 2002;16:2743–2748.
36. Shoba G, Joy D, Joseph T, Majeed M, Rajendran R, SrinivasPS.
Influence of piperine on the pharmacokinetics ofcurcumin in animals
and human volunteers. Planta Med1998;64:353–356.
37. Strimpakos AS, Sharma RA. Curcumin: Preventive
andtherapeutic properties in laboratory studies and clinical
trials.Antioxid Redox Signal 2008;10:511–545.
38. Bala K, Tripathy BC, Sharma D. Neuroprotective and
anti-ageing effects of curcumin in aged rat brain
regions.Biogerontology 2006;7:81–89.
39. Yang F, Lim GP, Begum AN, et al. Curcumin inhibitsformation
of amyloid beta oligomers and fibrils, bindsplaques, and reduces
amyloid in vivo. J Biol Chem 2005;280:5892–5901.
40. Ryu EK, Choe YS, Lee KH, Choi Y, Kim BT. Curcumin
anddehydrozingerone derivatives: Synthesis, radiolabeling,
andevaluation for beta-amyloid plaque imaging. J Med
Chem2006;49:6111–6119.
41. Pan MH, Huang TM, Lin JK. Biotransformation of
curcuminthrough reduction and glucuronidation in mice. Drug
MetabDispos 1999;27:486–494.
42. Cole GM, Teter B, Frautschy SA. Neuroprotective effects
ofcurcumin. Adv Exp Med Biol 2007;595:197–212.
43. Thangapazham RL, Sharma A, Maheshwari RK. Multiplemolecular
targets in cancer chemoprevention by curcumin.AAPS J
2006;8:E443–E449.
44. Schuller U, Heine VM, Mao J, et al. Acquisition of
granuleneuron precursor identity is a critical determinant
ofprogenitor cell competence to form Shh-induced medullo-blastoma.
Cancer Cell 2008;14:123–134.
45. Yang ZJ, Ellis T, Markant SL, et al. Medulloblastoma can
beinitiated by deletion of Patched in lineage-restricted
progen-itors or stem cells. Cancer Cell 2008;14:135–145.
46. Hartmann W, Digon-Sontgerath B, Koch A, et al.
Phospha-tidylinositol 3’-kinase/AKT signaling is activated in
medullo-blastoma cell proliferation and is associated with
reducedexpression of PTEN. Clin Cancer Res 2006;12:3019–3027.
47. Taipale J, Beachy PA. The Hedgehog and Wnt
signallingpathways in cancer. Nature 2001;411:349–354.
48. Henriksson M, Luscher B. Proteins of the Myc
network:Essential regulators of cell growth and differentiation.
AdvCancer Res 1996;68:109–182.
49. Altieri DC, Survivin, cancer networks and
pathway-directeddrug discovery. Nat Rev Cancer 2008;8:61–70.
50. Haberler C, Slavc I, Czech T, et al.
Histopathologicalprognostic factors in medulloblastoma: High
expression ofsurvivin is related to unfavourable outcome. Eur J
Cancer2006;42:2996–3003.
51. Romer J, CurranT.
Targetingmedulloblastoma:Small-moleculeinhibitors of the Sonic
Hedgehog pathway as potential cancertherapeutics. Cancer Res
2005;65:4975–4978.
52. Bar EE, Stearns D. New developments in
medulloblastomatreatment: The potential of a cyclopamine-lovastatin
combi-nation. Expert Opin Investig Drugs 2008;17:185–195.
53. Li M, Zhang Z, Hill DL, Wang H, Zhang R. Curcumin, a
dietarycomponent, has anticancer, chemosensitization, and
radio-sensitization effects by down-regulating the MDM2 onco-gene
through the PI3K/mTOR/ETS2 pathway. Cancer
Res2007;67:1988–1996.
54. Kunnumakkara AB, Guha S, Krishnan S, Diagaradjane P,Gelovani
J, Aggarwal BB. Curcumin potentiates antitumoractivity of
gemcitabine in an orthotopic model of pancreaticcancer through
suppression of proliferation, angiogenesis,and inhibition of
nuclear factor-kappaB-regulated geneproducts. Cancer Res
2007;67:3853–3861.
55. Bava SV, Puliappadamba VT, Deepti A, Nair A, KarunagaranD,
Anto RJ. Sensitization of taxol-induced apoptosis bycurcumin
involves down-regulation of nuclear factor-kap-paB and the
serine/threonine kinase Akt and is independentof tubulin
polymerization. J Biol Chem 2005;280:6301–6308.
56. Anand P, Kunnumakkara AB, Newman RA, Aggarwal
BB.Bioavailability of curcumin: Problems and promises. MolPharm
2007;4:807–818.
314 ELAMIN ET AL.
Molecular Carcinogenesis