CurcuminPromotesAutophagicSurvivalofaSubsetofColon Cancer ... · antitumor properties of curcumin....

Prevention and Epidemiology

CurcuminPromotesAutophagic Survival of aSubset ofColonCancer Stem Cells, Which Are Ablated by DCLK1-siRNA

Carla Kantara1,2, Malaney O'Connell1, Shubhashish Sarkar1, Stephanie Moya1,Robert Ullrich2, and Pomila Singh1,2

AbstractCurcumin is known to induce apoptosis of cancer cells by different mechanisms, but its effects on cancer stem

cells (CSC) have been less investigated. Here, we report that curcumin promotes the survival of DCLK1-positivecolon CSCs, potentially confounding application of its anticancer properties. At optimal concentrations,curcumin greatly reduced expression levels of stem cell markers (DCLK1/CD44/ALDHA1/Lgr5/Nanog) inthree-dimensional spheroid cultures and tumor xenografts derived from colon cancer cells. However, curcuminunexpectedly induced proliferation and autophagic survival of a subset of DCLK1-positive CSCs. Spheroidcultureswere disintegrated by curcumin in vitro but regrewwithin 30 to 40 days of treatment, suggesting a survivalbenefit from autophagy, permitting long-term persistence of colorectal cancer. Notably, RNA interference–mediated silencing of DCLK1 triggered apoptotic cell death of colon cancer cells in vitro and in vivo, and abolishedcolorectal cancer survival in response to curcumin; combination of DCLK1-siRNA and curcumin dramaticallyreversed CSC phenotype, contributing to attenuation of the growth of spheroid cultures and tumor xenografts.Taken together, our findings confirm a role of DCLK1 in colon CSCs and highlight DCLK1 as a target to enhanceantitumor properties of curcumin. Cancer Res; 74(9); 2487–98. �2014 AACR.

IntroductionColorectal cancer incidence remains very high in the

United States and Western world (1), and even thoughseveral antibodies/molecules have been developed for treat-ing colorectal cancers, surgical removal and chemo/radio-therapy remains standard of care. Colon cancer stem cells(CSC) are believed to be resistant to chemo/radiotherapy,and a major cause of relapse (2). Several colorectal cancerstem cell markers, with extracellular domains, have beenidentified, including CD44 (3), Lgr5 (4, 5), and DCLK1 (6, 7).We recently reported that immortalized embryonic epithe-lial cells, induced to overexpress progastrin (HEKmGAS),developed tumorigenic/metastatic potential (8). Trans-formed HEKmGAS CSCs coexpressed stem cell markersDCLK1/CD44, while nontumorigenic (HEKC) cells did not(8). Normal stem cells (NSC) from colonic crypts, positive forDCLK1/Lgr5, also do not coexpress CD44 (9). Here, weconfirmed colorectal cancer cells (HCT-116, DLD-1, and

HT-29) coexpress stem cell markers DCLK1/CD44, similarto tumorigenic HEKmGAS cells (8).

Curcumin (active ingredient in turmeric powder), isolatedfrom Curcuma longa (10), is an important ingredient of Asianfoods. Curcumin has potent anti-inflammatory, antibacterial,and anticancer effects (10), but lacks solubility in aqueoussolutions; bioavailable formulations have been developed (11)and are being examined in phase I/II clinical trials (Dr. L.Helson, personal communication; ref. 12). Curcumin is nontoxicat high doses (12 g/d) and targets multiple oncogenic pathways(10, 12–14). However, proapoptotic effects of curcumin areattenuated by autocrine growth factors (14). Thus, subpopula-tions of cancer cells may escape inhibitory effects of curcumin;this concept was extended to CSCs in the current studies.

Curcumin induces caspase-3–dependent and -independentapoptosis due to autophagy (13, 15). Autophagy represents adouble-edged sword that causes either cell death or survival ofcells (16).Majority of the studies suggest that curcumin-inducedautophagy is a prodeath signal (13, 17, 18). However, curcumin-induced autophagy also allows tumor-initiating cells to surviveand eitherdifferentiate (19), or becomesenescent/quiescent (20,21). Possible autophagic effects of curcumin on CSCs remainunknown. Inhibitory effects of curcumin have been reported onCSCs (22–25). However, response of CSCs, positive for DCLK1/CD44/Lgr5, to curcumin remains unknown. Here, we examinedinhibitory effects of curcumin against colon cancer cells in vitroand in vivo, in relation to effects on apoptosis/autophagy/proliferation of DCLK1/CD44/Lgr5þ stem cells. Spheroidalregrowth assay was used to examine resistance of CSCs tocurcumin. Our results suggest the novel possibility thatDCLK1þ cells survive curcumin-induced autophagy.

Authors' Affiliations: Departments of 1Neuroscience andCell Biology and2Sealy Cancer Center, University of Texas Medical Branch Health, Galves-ton, Texas

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Corresponding Author: Pomila Singh, Department of Neuroscience andCell Biology, University of TexasMedical Branch, 10.104Medical ResearchBuilding, 301 University Boulevard, Route 1043, Galveston, TX 77555-1043. Phone: 409-772-4842; Fax: 409-772-3222; E-mail:[email protected]

doi: 10.1158/0008-5472.CAN-13-3536

�2014 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 2487

on November 6, 2020. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst March 13, 2014; DOI: 10.1158/0008-5472.CAN-13-3536

http://cancerres.aacrjournals.org/

Because a subpopulation of DCLK1þ cells survived inhibitoryeffects of curcumin, we examined inhibitory efficacy of DCLK1-siRNA�curcumin, against growth of HCT-116 cells in vitro andin vivo. Our studies demonstrate that DCLK1-siRNA inducesapoptosis of colon cancer cells/tumors in the absence of autop-hagy; combination of curcuminþDCLK1-siRNA inducedmassiveapoptotic/autophagic cell death, resulting in almost completeloss of stem cell populations expressing DCLK1/CD44/Lgr5.

Materials and MethodsMaterials and methods used in the current study are similar

to that described previously (8). Reagents used are detailed inSupplementary Methods (A). HCT-116, DLD-1, and HT-29colon cancer cells from American Type Culture Collection(ATCC) were maintained in Dulbecco's Modified Eagle Medi-um as previously described (26). These cells were purchased in1990s from ATCC and were authenticated by BioSynthesisDNA Identity Center in 2012.

Treatment of cellsSubconfluent cells in monolayer cultures were treated with

either dimethyl sulfoxide (DMSO) or nontargeting siRNA (con-trols) or optimally effective concentrations of curcumin(25mmol/L)�DCLK1-siRNA (100 nmol/L), based onpreliminarystudies with increasing concentrations of these agents. After 24to 48 hours of treatment, cells were processed for measuringviability/proliferation/apoptosis/autophagy by publishedmeth-ods (8, 17, 18), as detailed in Supplementary Methods (B).

Viability/proliferation. Briefly, viability/proliferation ofcells was measured by Trypan blue exclusion test and MTTassay, respectively, as described previously (8).

Apoptosis/autophagy/proliferation. To assess the per-centage of cells undergoing apoptosis/autophagy/proliferation,control and treated cells/spheroids, dissociated enzymaticallyand cytospun onto slides, were fixed and processed for stainingwith antibodies against apoptoticmarker (activated caspase-3),autophagic marker (LC3A/B-I/II), and proliferation marker[proliferating cell nuclear antigen (PCNA)]. Live cells were alsostained with Acridine Orange to visualize autophagic vesicles.

Colon cancer cells grown as primary/secondaryspheroids

Colon cancer cells were grown as primary/secondary spher-oids, as previously described by us (8). Primary spheroids weretreated on day 6 after seeding the wells (at which time well-formed primary-spheroids were present), with optimally effec-tive concentrations of curcumin�DCLK1-siRNA, as describedin legend of Fig. 2A. For generating secondary spheroids, wellscontaining primary spheroids were enzymatically dissociated,and approximately 5,000 cells replated in low-attachmentplates as described in Supplementary Methods (C).

Relapse experiment. For the relapse experiment (Fig. 2),control/treated primary spheroids were dissociated andreplated as secondary spheroids and imaged daily with whitelight microscopy. In some experiments, spheroids were pro-cessed for paraffin embedding, followed by hematoxylin andeosin (H&E) staining/immunohistochemistry/immunofluo-rescence staining, as previously described (8). Control/treated

spheroids were also processed for either Western blot analysisor cell viability, as described above.

FACsorted colon cancer cells and analysis for stem cellmarkers

Colon cancer cellswereFACsorted andanalyzed for stemcellmarkers, as described previously (8) and detailed in Supple-mentary Methods (D). Briefly, subconfluent cells were har-vested and processed for labeling with fluorophore-taggedantibodies against DCLK1/Lgr5, FACsorted into distinct popu-lations of positive/negative cells, cytospun and fixed on slides,and processed for immunofluorescence staining with anti-CD44 antibody and 40,6-diamidino-2-phenylindole (DAPI).Images acquired with an epifluorescent microscope were ana-lyzed using METAMORPH, v6.0 software (Molecular Devices).

Cells grown as subcutaneous xenografts in athymic(SCID/nude) mice

Cells were grown as subcutaneous xenografts in athymic(SCID/nude) mice as described previously (8) and detailed inSupplementary Methods (E). Briefly, 5 � 106 cells were inoc-ulated subcutaneously in both flanks of female athymic mice.One week after injection, xenografts were visible on both sides.Mice, bearing tumors, were randomly divided into groups of3 each. Mice were injected every 2 days with either 0.01%DMSO (Control), curcumin (3mg/100mL in 0.01%DMSO¼ 100mmol/L), nontargeting siRNA (control; 100 (nmol/L)/100 mLPBS), DCLK1-siRNA (0.5 pmol/100mL in PBS¼ 100 nmol/L), orcurcumin (100 mmol/L) and DCLK1-siRNA (100 nmol/L).Tumor volume was measured every other day. Mice weresacrificed 3 weeks after initiating treatment, and tumors wereremoved and weighed and processed for Western blot/immunofluorescence analysis.

Western blot analysis of cells growing either as 2Dcultures, 3D spheroids, or xenografts

Western blot analysis of cells growing either as two-dimen-sional (2D) cultures, three-dimensional (3D) spheroids, or xeno-grafts was conducted as detailed in SupplementaryMethods (F).Briefly, cells/tumorswere processed forWestern blot analysis aspreviously described (8). Blots were cut into horizontal stripscontaining target or loading control proteins and processed fordetection of antigen–antibody complexes by chemilumines-cence. Membrane strips containing target/loading control pro-teins were simultaneously exposed to autoradiographic films. Incases in which limited samples were analyzed for multipleproteins, loading control b-actin wasmeasured in a correspond-ing sample containing equivalent protein. In a few cases, b-actinwas stripped to measure target protein with equivalent proteinwith similar molecular mass within the same membrane. Rel-ative band density on scanned autoradiograms was analyzedusing ImageJ program (rsbweb.nih.gov/ij/download) and exp-ressed as a ratio of b-actin in corresponding samples.

Transient transfection of cell/spheroids with double-stranded siRNA oligonucleotide

Transient transfectionof cell/spheroidswithdouble-strandedsiRNAoligonucleotide is detailed inSupplementaryMethods (G)

Kantara et al.

Cancer Res; 74(9) May 1, 2014 Cancer Research2488




andwas conducted as previously described (8). Transfected cellsin 2D were propagated in normal growth medium containing10% fetal calf serum, and growthwas examined after 48 hours inanMTT assay. Transfected spheroids weremaintained in spher-oid medium as described in Supplementary Methods (C).

Statistical analysis of dataData are presented as mean� SEM of values obtained from

four to six samples from two to three experiments/mice. Totest for significant differences between means, nonparametricMann–Whitney test was used using StatView 4.1 (AbacusConcepts, Inc.); P values were considered statistically signifi-cant if less than 0.05.

ResultsDCLK1þ colon CSCs coexpress CD44Monolayer cultures of colon cancer cell lines (HCT-116/

DLD-1/HT-29) were analyzed by either immunofluorescenceor FACSorting for expression levels of CSC markers, DCLK1/CD44/Lgr5, as previously described (8). On an average, 2% to3% of cells expressed stem cell markers (data not shown). We

recently reported that transformed/tumorigenic embryonicepithelial cells coexpressed stem cell markers DCLK1/CD44,unlike isogenic nontumorigenic cells (8). Human colon cancercell lines were similarly positive for transformed phenotype(representative data from HCT-116 cells are presented inSupplementary Fig. S1i). Majority of FACSorted DCLK1þ cells(>80%) coexpressed CD44, whereas FACSorted Lgr5þ cells didnot (Supplementary Fig. S1ii and S1iii). Surprisingly, a largenumber of CD44þ cells cosorted with Lgr5þ cells (Supplemen-tary Fig. S1i and S1iii), suggesting that a subpopulation of Lgr5þ

cells may be tightly adherent to CD44þ cells. Lineage-tracingstudies in the future may allow us to determine whether theadherent CD44þ cells perhaps represent daughter progenitorcells, derived from Lgr5þ cells. CD44þ cells, which cosortedwith Lgr5þ cells, did not coexpress DCLK1 (data not shown),unlike coexpression of CD44 by a majority of DCLK1þ cells(described above). CSCs, positive for either DCLK1 or Lgr5,were mostly present along outer edges of spheroids, derivedfrom colon cancer cells (Supplementary Fig. S2A); CD44þ cells,on the other hand, were distributed throughout the spheroids,providing further evidence that cells positive for only CD44

BA

C

Di) ii)

i) ii)

iii)

H&

E

Curcumin-treated HCT-116 spheres

Time

(h) 0 24 48 72

CD44

CD44

Co

ntr

ol

25

mm

ol/

L

Cu

rcu

min

CD44LGR5DCLK1

48 h

*

1

0.5

0

Control 25 mmol/L

Curcumin

Ab

so

rba

nc

e (

56

0 n

m)

DCLK1 LGR5

DCLK1 LGR5

HCT-116 spheres

Control Curcumin

Activated

caspase-3

b-Actin

17 kDa

42 kDa

300

200

100

0

# o

f Tu

mo

rosp

here

s

Control10 mmol/L 25 mmol/L

25 m

mo

l/L

10 m

mo

l/L

Co

ntr

ol

Cu

rcu

min

Cu

rcu

min

(DM

SO

)

Curcumin Curcumin

*

Control 25 mmol/L Curcumin

Activated caspase-3 Activated caspase-3×20

×20

×20

×20

×10

×10

×10

×20

Figure 1. A and B, curcumin inhibitsgrowth of HCT-116 cells/spheroidsand reduces expression of stemcell markers. A, MTT assay results(absorbance at 560 nm) withcontrol/curcumin–treated HCT-116 cells in culture. Data, mean �SEM of 8 wells/two experiments.B, representative images (one ofthree experiments) of 2D cells,treated with curcumin/vehicle for48 hours, stained with indicatedstem cell markers. C and D,curcumin disintegrates spheroidsand induces apoptosis. Ci, equalnumber of HCT-116 spheroids/well, imaged at 48 hours aftertreatment, as shown. Cii, number ofHCT-116 spheroids after 48 hoursof treatment. Data, mean � SEM ofspheroids in 24 wells from oneof two experiments. Ciii,representative H&E images ofsingle HCT-116 spheroids sectionsafter 0 to 72 hours of curcumintreatment. D, Western blot (i) andstaining (ii) data for activatedcaspase-3 from a representative oftwo experiments with spheroidsafter 48 hours of treatment. Imagesin B, Cii, and iii are digitallyenhanced. �, P < 0.05 versuscontrol values.

Effects of DCLK1-siRNA�Curcumin on Cancer Stem Cells

www.aacrjournals.org Cancer Res; 74(9) May 1, 2014 2489




may perhaps represent daughter progenitor cells. Coexpres-sion of CD44 and DCLK1 was evident in cells along outer edgesof spheroids, while coexpression of Lgr5 and CD44 was lessfrequent (Supplementary Fig. S2Bi and S2Bii).

Curcumin attenuates growth of HCT-116 cells/spheroidsassociated with loss of stem cell markers

Curcumin (25 mmol/L) was optimally effective in reducinggrowth of HCT-116 cells in 2D cultures by >50% (Fig. 1A),resulting in reduced expression of stem cell markers DCLK1/Lgr5/CD44 (Fig. 1B). Low doses of curcumin (10 mmol/L) didnot significantly reduce number of spheroidal growths/well,but had morphologic effects (Fig. 1Ci and ii). Curcumin (25mmol/L), reduced total number of tumorospheres by >60%/

well, associated with disintegration of spheroids (Fig. 1Ciand ii), in a time-dependent manner (Fig. 1Ciii), along withcaspase-3 activation (Fig. 1Di and ii).

Regrowth (relapse) of curcumin-treated HCT-116spheroids

HCT-116 primary spheroids growing in 24-well plates weretreated on day 6 with either control vehicle or 25 mmol/L ofcurcumin for 48 hours (Fig. 2Ai). Primary spheroids on day 8were dissociated and replated as secondary spheroids. By day 4,cells from the control group started growing as secondaryspheroids, whereas curcumin-treated cells did not (Fig. 2Ai).For approximately 28 days, secondary spheroids did not formfrom curcumin-treated samples, while control samples

A

B

C

siRNA

DCLK1

b-Actin b-Actin

b-Actin

Control DCLK1

HCT-116 spheres

Primary spheroids Secondary spheroids

i)

No

treatment

Control

siRNA

DCLK1

siRNA

(100 nmol/L)

ii)

i)

i)

LGR5

DCLK1

HCT-116 control

ii)

LGR5

DCLK1

Day 4 Day 30

HCT-116 curcumin

Day 45Day 4 Day 30 Day 45

95 kDa

47 kDa

42 kDa

95 kDa

47 kDa

42 kDa

ii)

48 h H&E

100

75

50

25

0Day 4 Day 30 Day 45

Control

25 μmol/L Curcumin

Curcumin+DCLK1siRNA%

Cel

l via

bilit

y

Secondary spheroids

a a a

*

*

47 kDa

42 kDa

×10

×10

×10 ×10

×10 ×10

×10 ×10

×10 ×10

×10

×10

×10 ×20

×20

×20 ×20

×20

×20

Control

Curcumin

Day 6 Day 8 Day 4 Day 30 Day 45

Day 45Day 30Day 4Day 8Day 6

Figure 2. AandB, regrowth/relapseand reexpression of stem cellmarkers in curcumin-treated HCT-116 spheroids. Equal number ofHCT-116 cells, seeded in 24 wellplates, were treated on day 6 withor without curcumin (25 mmol/L;single arrow). Ai, after 48 hours,primary spheroids/cells wereharvested and plated assecondary spheroids (two arrows).Representative images ofsecondary spheroids/cells imageduntil day 45. Aii, cell viability (%) ofsecondary spheroids/cells; data,mean � SEM of 8 wells/twoexperiments. �, P < 0.05 versuscontrol values; a, P < 0.05 versuscurcumin values. Bi and ii,representative Western blot datafrom secondary spheroids/cells induplicate wells derived fromcontrol/curcumin–treatedsamples. C, treatment of HCT-116spheroids with DCLK1-siRNA.Representative Western blot data(i) and images (ii) of HCT-116primary spheroids treated withcontrol or DCLK1-siRNA (100nmol/L) for 48 hours.Representative H&E staining ofspheroid sections shown in thelast panels.

Kantara et al.





developed dense secondary spheroids. By days 28 to 30, smallspheroidal structures appeared in wells containing curcumin-treated samples (Fig. 2Ai). By day 45, curcumin-treated sphe-roidal cells had regrown as secondary spheroids, suggestingthat a subset of stem cells survived curcumin. Secondary cells/spheroids were isolated as single cells at indicated days andanalyzed for cell viability (Fig. 2Aii). Surprisingly, approximate-ly 25% of curcumin-treated cells were viable on day 4 afterreplating (Fig. 2Aii), increasing dramatically by days 30 to 45,matching regrowth of treated-cells as secondary spheroids(Fig. 2Ai and ii). Relative levels of Lgr5 remained stable insecondary spheroids from control wells, while relative levels ofDCLK1 increased 2- to 3-fold by days 30 to 45 (Fig. 2Bi and ii).Curcumin treatment of primary spheroids resulted in almostcomplete attenuation of Lgr5, but low levels of DCLK1remained (Fig. 2Bii). These results suggest that a subset ofDCLK1þ cells is resistant to curcumin. At day 30, after replat-ing, relative levels of Lgr5/DCLK1 were increased in curcumin-treated spheroids; by day 45, levels had increased 2- to 4-fold(Fig. 2Bii). Because curcumin attenuated Lgr5 expression but

not DCLK1, we used DCLK1-siRNA for targeting DCLK1þ

CSCs.

DCLK1-siRNA targets DCLK1 expression and inducesdisintegration of HCT-116 spheroids

HCT-116 spheroids were treated with either control orDCLK1-siRNA. DCLK1-siRNA (100 nmol/L) attenuated DCLK1expression in HCT-116 spheroids (Fig. 2Ci), disintegratingspheroids within 48 hours (Fig. 2Cii); lower concentrationswere less effective (data not shown). Control-siRNA had noeffects. Surprisingly DCLK1-siRNA was more effective thancurcumin, while combination of DCLK1-siRNAþcurcuminwas significantly more effective than either agent alone, inboth 2D (Fig. 3Ai) and 3D (Fig. 3Aii and iii).

Inhibitory effects of curcumin�DCLK1-siRNA againstgrowth of HCT-116 xenografts in vivo

Mice inoculated subcutaneously with 5� 106 HCT-116 cellshad palpable tumors by day 7, andwere injected on ventral sideof tumors with either control or DCLK1-siRNA�curcumin

C

B

A iii)i)

3D spheroids

**

* ; a; b

i)

**

*; a

2D monolayer

ii)

**

*

HCT-116 xenografts

, a, b

,a

; a; b

48 h

ii)

2.5

2

1.5

1

0.5

0

Ab

so

rba

nc

e (

56

0 n

m)

Ab

so

rba

nc

e (

56

0 n

m)

Control

Control

Control Control

Control

Control

Control

Control

(DMSO)

(DMSO)

(DMSO)

siRNA

siRNA

siRNA

siRNA

siRNA

siRNA

siRNA

siRNA+

siRNA+

siRNA

Curcumin

Curcumin

Curcumin Curcumin+

curcumin

curcumin

Curcumin

curcumin

DCLK1

DCLK1

DCLK1 siRNA

DCLK1

DCLK1 DCLK1

DCLK1DCLK1 siRNA

25 μmol/L

25 μmol/L

25 mmol/L

25 mmol/L

(DMSO)

+

1.2

0.8

0.4

0

Tu

mo

r vo

lum

e (

mm

3)

Weeks

1,000

500

0

0 1 2 3 4

Treatment

Inoculation

Sacrifice Treatments

Control (DMSO)

Control siRNA

Curcumin

Curcumin +

DCLK1 siRNA

DCLK1 siRNA

Tu

mo

r w

eig

ht

(g) 0.4

0.3

0.2

0.1

0

×10

×10

×10

×10

×10

×20

×20

×20

×20

×20

Figure 3. A–C, inhibitory effects ofcurcumin (25 mmol/L) � DCLK-siRNA (100 nmol/L) on growth ofHCT-116 cells, growing either as2D (Ai)/3D (Aii, iii) in vitro or asxenografts in vivo (B andC). Ai–Aiii,cells/spheroids treated for 48hours as shown. Controls treatedwith 0.01% DMSO or control-siRNA (100 nmol/L). Growth ofcells/spheroids (absorbance at560 nm) was analyzed asdescribed in SupplementaryMethods (C). Data,mean�SEMof8 to 12 wells/two experiments for2Dcells (Ai) and 3D-spheroids (Aii).Representative images ofspheroids are presented in Aiii.B and C, athymic mice (3/group)inoculated bilaterally with 5 � 106

HCT-116 cells and treated withcurcumin � DCLK1-siRNA [asdescribed in SupplementaryMethods (E)]. B, representativeimages of tumor-bearing mice atday 28. Ci, tumor volumesmeasured at indicated time points.Cii, tumors weights (g) at time ofsacrifice. Data in Ci/ii, mean �SEM of six tumors/three mice.�, P < 0.05 versus control values;a,P < 0.05 versus curcumin values;b, P < 0.05 versus DCLK1-siRNAvalues.






every second day for 3 weeks. Representative tumor-bearingmice, from the five treatment groups, are shown at day 28in Fig. 3B. Tumor size (volume) continued to enlarge inmice inthe order of control (DMSO) ¼ control-siRNA > curcumin >DCLK1-siRNA (Fig. 3Ci). However, in curcuminþDCLK1-siRNA group, the preformed tumors actually began to shrinkin size (Fig. 3Ci). After 3 weeks of treatment, mice wereeuthanized and tumor weights noted. Tumor weights in con-trol versus treated mice followed a similar pattern describedabove for tumor size (Fig. 3Cii). Surprisingly, DCLK1-siRNAwas more effective than curcumin against growth of HCT-116cells/tumors (Fig. 3Ai and ii, B, and C), suggesting a functionalrole of DCLK1 in proliferative/tumorigenic potential of CSCs.The latter possibility was confirmed in the relapse experimentwith spheroids. Primary spheroids treated with curcu-minþDCLK1-siRNA did not re-form secondary spheroids evenafter 60 days of replating (data not shown), and cell viabilityremained <5% after replating (Fig. 2Aii).

Inhibitory efficacy of curcumin on relative expression ofstem cell markers/transcription factors in HCT-116 cells

Autocrine progastrin exerts growth-promoting effects oncolon cancer cells by activating NF-kBp65/b-catenin and upre-gulating relative expression of DCLK1/Lgr5/CD44 (9, 26–29).Control and curcumin-treated monolayer cultures (M), spher-oids (S), and tumors (T) were harvested and analyzed byWestern blot analysis for the above indicated proteins. Repre-sentative Western blot data are presented in SupplementaryFig. S3Ai and S3Aii; data from several blots are presented as apercentage change in ratio of relative levels of target proteins/b-actin in control versus curcumin-treated cells (Supplemen-tary Fig. S2B). Results confirmed that curcumin attenuatesactivation of b-catenin/NF-kBp65 and reduces relative expres-sion of stem cell markers. Surprisingly, even though curcuminreduced expression of indicated proteins by >40% to 90%(Supplementary Fig. S3), inhibitory efficacy on growth of2Dcells/spheroids/tumors was <50% (Fig. 3A–C). Because cur-cumin induces autophagy (13, 15–20), which can lead to eithercell survival or cell death (16), we next examined autophagy/apoptosis in response to curcumin�DCLK1-siRNA.

Curcumin induces autophagy and apoptosis of coloncancer cells while DCLK1-siRNA only induces apoptosis

Curcumin-treated HCT-116 cells were analyzed for relativelevels of autophagic markers (LC3A/BI-II and Beclin-1) andactivated caspase-3. Representative Western blot data areshown in Fig. 4Ai and ii. Data from several blots are presentedas percentage change in the ratio of relative levels of targetproteins/b-actin in control versus curcumin-treated cells (Fig.4Aiii and iv). Curcumin increased relative levels of LC3-Ibetween 12 and 24 hours, which was processed to generateLC3-II (Fig. 4Ai and ii), confirming previous reports (20).Beclin-1 (autophagy-latent gene, ATG6), required to initiateautophagosome formation (30), also increased in a time-dependent manner in response to curcumin (Fig. 4Aii and iv).Acidic vesicular organelles (AVO) stain orange/red with acri-dine orange and specify autophagy (31). Formation of AVOswas confirmed in HCT-116 cells in response to curcumin in a

dose- and time-dependent manner (Supplementary Fig. S4Ai–S4Aiv). Increased expression/formation of LC3-I/II in responseto curcumin was confirmed by immunofluorescent staining ofHCT-116 cells in culture (Fig. 4B). A significant percentage ofLC3þ cells in curcumin-treated samples coexpressed activatedcaspase-3 (Fig. 4B, merged images), suggesting that curcumin-induced autophagy results in apoptotic death of many HCT-116 cells. DCLK1-siRNA significantly increased staining foractivated caspase-3, but not LC3 (Fig. 4B). Combination ofcurcuminþDCLK1-siRNA significantly increased staining forLC3/activated caspase-3, suggesting that combined regimenmay synergistically induce autophagy/apoptosis, whereinmajority of autophagic cells go through apoptosis (Fig. 4B).

PCNAþ cells are mainly present at outer edges of controlHCT-116 spheroids (Fig. 4C), similar to staining pattern ofDCLK1/Lgr5 (Supplementary Fig. S2A and S2B). Curcumin-treated HCT-116 spheroids become positive for PCNA andLC3-I/II in an overlapping area of spheroids (Fig. 4C), suggest-ing that curcumin-induced autophagy is associated with bothapoptosis (Fig. 4B) and proliferation (Fig. 4C).

To further examine cell death/cell survival role of autophagy,HCT-116 cells were treated with an inhibitor of autophagy (3-methylalanine, 3-MA; Supplementary Fig. S4B). Inhibitoryeffects of curcumin were partially reversed by 3-MA, but notto control levels (Supplementary Fig. S4B), suggesting thatcurcumin-induced autophagy results in both survival/apopto-sis. 3-MA had insignificant effects on cells treated with DCLK1-siRNA or DCLK1-siRNAþcurcumin (Supplementary Fig. S4B),confirming that autophagy in DCLK1-siRNAþcurcumin–trea-ted cells is mainly linked to apoptosis.

HCT-116 xenografts, harvested from control/treated mice,were processed for H&E/immunofluorescence (Fig. 5). TheH&E sections from curcumin-treated tumors demonstratedunique hollow circular areas, surrounded by concentriclayers of cells (Fig. 5Ai), not seen in other groups. Asobserved in 2D/3D cells in vitro (Fig. 4B and C), control/DCLK1-siRNA–treated tumor sections were largely negativefor LC3 (Fig. 5Aii and iii). Curcumin-treated tumor sectionsdemonstrated strong LC3 staining in concentric layers ofcells surrounding the hollow areas, which seemed freeof nucleated cells; combined treatment with curcu-minþDCLK1-siRNA significantly augmented LC3 staining,but the hollow areas, surrounded by concentric layers ofcells, were not evident any longer (Fig. 5Aii). PCNAþ cellswere present along the edges of control tumors, but rela-tively absent in tumors treated with DCLK1-siRNA or cur-cuminþDCLK1-siRNA (Fig. 5Aii). Curcumin-treated tumorsections, on the other hand, demonstrated PCNA staining inconcentric layers of cells surrounding the hollow areas (Fig.5Aii), similar to PCNA staining pattern seen in curcumin-treated spheroids (Fig. 4C). Enhanced images from Fig. 5Aiiare presented in Supplementary Fig. S5A to present thestaining of LC3 and PCNA more clearly. A significant num-ber of LC3-expressing cells coexpressed PCNA (Supplemen-tary Fig. S5A, yellow color in merged images), while cellspositive for LC3 and PCNA in control tumor sections weredistinct and separate (Supplementary Fig. S5A). Percentagestaining for LC3/PCNA per tumor section from 10 to 15

Kantara et al.





sections/4 to 6 tumors/3 mice was quantified as described inlegend of Supplementary Fig. S5, and presented as bargraphs in Fig. 5Aiii. Control and DCLK1-siRNA–treatedtumors were minimally (�1%) LC3þ, while LC3 staining oftumor sections treated with curcumin/curcuminþDCLK1-siRNA increased several fold. Control tumor sections werepositive for PCNA staining in 4% to 5% of the area, while <1%to 1.5% area of DCLK1-siRNA–treated tumors were PCNAþ

(Fig. 5Aiii). An unexpected finding was that the percentageof PCNAþ cells in curcumin-treated tumors increased 2-foldfrom control levels; DCLK1-siRNAþcurcumin attenuatedPCNA staining to <1% (Fig. 5Aiii). Activation of caspase-3was minimal in control tumors (�1%) but significantlyincreased in tumors treated with either curcumin (�7%),DCLK1-siRNA (�6–7%), or DCLK1-siRNAþcurcumin(�9%; Fig. 5Bi and ii). Importantly, none of the PCNAþ cellswere positive for activated caspase-3 (Fig. 5Bi), unlike cost-aining of LC3 with PCNA in curcumin-treated samples (Fig.5Aii).

Treatment of tumors with curcumin�DCLK1-siRNAreverses transformed phenotype of CSCs

Tumor sections presented in Fig. 5Aii and Bi were alsoprocessed forDCLK1/Lgr5/CD44 staining. Representative datafromone tumor/group of a total of four to six tumors/group arepresented in Fig. 6A and Supplementary Fig. S5B. Intensity ofstaining for all three stem cell markers was strongest at theedges of tumors. Control tumors demonstrated significantcoexpression of DCLK1/CD44 and Lgr5/CD44 (Fig. 6A andSupplementary Fig. S5B), confirming transformed phenotypeof CSCs. Treatment with curcumin�DCLK1-siRNA causedcomplete attenuation of transformed phenotype, with negli-gible coexpression of CD44 with DCLK1/Lgr5 (Fig. 6A andSupplementary Fig. S5Bi and S5Bii). Surprisingly, curcumin-treated tumor sections demonstrated DCLK1 staining in con-centric layers of cells surrounding the hollow areas (Fig. 6A andSupplementary Fig. S5Bi); these layers of cells were alsopositive for LC3 and PCNA staining, as described above forimages presented in Fig. 5Aii. The latter findings suggest the

B

Control

(DMSO)

Curcumin

Control

siRNA

DCLK1

siRNA

Curcumin

+ DCLK1 siRNA

LC3Activated

caspase-3 Merge

< 1% < 1%

~ 20% ~ 10%

< 1% < 1-2%

<1% ~ 20%

~35% ~50%

C

Aii)i)

iv)iii)

Co

ntr

ol

25

mm

ol/

L C

urc

um

in

25 mmol/L Curcumin25 mmol/L Curcumin

HCT-116HCT-116

LC3 ILC3 II

b-Actin b-Actin

16 kDa60 kDa

15 kDa

42 kDa

14 kDa

42 kDa

0 h0 h 12 h 24 h 48 h

2 h 4 h 6 h 12 h 24 h48 hBeclin

Activated caspase-3

Activated caspase-3100

80

60

40

20

0

120

100

80

60

40

20

00 h 2 h 4 h 6 h 12 h 24 h 48 h 0 h 12 h 24 h 48 hTime (h) Time (h)

% C

han

ge in

th

e r

ati

o o

f

% C

han

ge in

th

e r

ati

o o

f

targ

et

pro

tein

s/b

-acti

n

targ

et

pro

tein

s/b

-acti

n

LC3-I LC3-II Beclin

×20 ×20 ×20

×20×20

×20 ×20 ×20

×20×20×20

×20 ×20 ×20

×20

×20 ×20

×20

Figure 4. A–C, curcumin inducesautophagy/apoptosis in HCT-116cells/spheroids while DCLK1-siRNA induces apoptosis. Ai and ii,representative Western blot datafrom one of three experimentswith curcumin-treated HCT-116cells in 2D; LC3-I/II (Ai) and Beclin-1/activated caspase-3 (Aii) asshown. Aiii–iv, % change inWestern blot data from all threeexperiments in Ai–ii. Ratio of targetprotein/b-actin at 0 hoursassigned 0% value; ratios atincreasing time points presentedas % of 0-hour value.B, immunofluorescent staining forLC3-I/II and activated caspase-3from representative control/treated cells, cultured oncoverslips, from one of twoexperiments. Yellow color inmerged images, costaining ofLC3/activated caspase-3 [�%cells stained for indicated protein(s) shown in each panel]. C,representative immunofluorescentstaining for LC3 and PCNA incontrol/treated HCT-116spheroidal sections from one oftwo experiments. Arrows, stainingfor indicated proteins.






novel possibility that autophagic cells present among theconcentric layers of cells, surrounding the hollow areas mayrepresent a subpopulation of DCLK1þ cells, which retain thepotential to proliferate (as suggested by PCNA-labeling of thesecells; Fig. 5Aii and Supplementary Fig. S5A). Curcumin-treatedtumor cells also continued to express CD44, unlike DCLK1-siRNA–treated tumors (Fig. 6A and Supplementary Fig. S5B).Curcumin significantly reduced expression of Lgr5, whileDCLK1-siRNA was much less effective (Fig. 6A and Supple-mentary Fig. S5B). CurcuminþDCLK1-siRNA was most effec-tive in attenuating DCLK1/Lgr5 staining along with significantloss in CD44 staining (Fig. 6A and B and Supplementary Fig.S5B). Tumors were also processed for Western blot analysis.Data fromall tumors are presented as percentage change in theratio of target protein/b-actin, wherein ratio for control sam-

ples was arbitrarily assigned 100% value (Fig. 6B). Values fromcontrol groups (vehicle or control siRNA) were almost iden-tical; therefore, a single bar for control values is shown in Fig.6B. Curcumin significantly reduced relative levels of stem cellmarkers (DCLK1/Lgr5/ALDHA1) and pluripotent marker(Nanog), associated with a significant loss in levels of activatedNF-kBp65s276 and total b-catenin, with a 3-fold increase inlevels of activated caspase-3 and LC3-II (Fig. 6B; onlychanges in LC3-II are shown). DCLK1-siRNA attenuatedrelative levels of DCLK1/ALDHA1/Nanog, but had insignif-icant effects on Lgr5. DCLK1 had insignificant effects onactivated NF-kB, but significantly reduced total b-catenin,resulting in significantly increasing activated caspase-3(apoptotic pathway), with no LC3II (autophagy). Curcu-minþDCLK1-siRNA was most effective in attenuating

×10 ×10 ×10

×10

×10 ×10 ×10

×10 ×10 ×10

×10 ×10 ×10

×10 ×10 ×10

×10 ×10 ×10

×10 ×10 ×10

×10 ×10

×10 ×10 ×10

×10 ×10

×20 ×20 ×20

A

B

Control

(DMSO)

Curcumin

Control ×4

×4

siRNA

DCLK1

siRNA

Curcumin

+

DCLK1

siRNA

LC3 PCNA Merge

ii)

iii)

*

*

*

*

i)

LC3 PCNA

% S

tain

ing

/se

cti

on

* ;b

;b;b

;b;aa

ii)

**

*

*

**

Activated caspase-3 PCNA

% S

tain

ing

/se

cti

on

;b

;a

;b

;b

Control

(DMSO)

Curcumin

Control

siRNA

DCLK1

siRNA

Curcumin

+

DCLK1

siRNA

Activated

caspase-3 PCNA Mergei)

Figure 5. A and B, effect ofcurcumin�DCLK1-siRNA onproliferation/apoptosis/autophagyof HCT-116 xenografts in vivo.Tumors from Fig. 3 experimentswere processed for paraffinembedding/staining or Westernblot (Fig. 6B). Ai, representativeH&E sections from control/curcumin–treated xenografts.Aii, representative tumorsections stained for LC3/PCNA(magnified images presented inSupplementary Fig S5A). 5Aiii, %area stained for indicated proteins.Data, mean � SEM of % stainingof sections from three to sixtumors/group [described inSupplementary Methods (E)]. Bi,representative sections stained foractivated caspase-3/PCNA. 5Bii,% staining/section, analyzed asdescribed above. �,P<0.05 versuscorresponding control values;a, P < 0.05 versus correspondingcurcumin values; b,P < 0.05 versuscorresponding DCLK1-siRNAvalues.

Kantara et al.





relative levels of DCLK1/Lgr5/ALDHA1/Nanog/cellularb-catenin, resulting in a robust activation of both cas-pase-3 and LC3, suggesting that autophagic response toDCLK1-siRNAþcurcumin likely leads to cell death ratherthan survival, unlike the response to curcumin alone.

DiscussionA novel finding of the current study is that a subset of

DCLK1þ colon CSCs is resistant to inhibitory effects of curcu-min, and that DCLK1-siRNA ismore effective than curcumin inreducing tumor mass in vivo. Conventional anticancer thera-pies (radio/chemotherapy) primarily kill rapidly proliferatingcancer cells that form bulk of the tumors, but are believed tospare relatively quiescent CSCs. Dietary agents, such as cur-cumin, are believed to suppress self-renewal of CSCs, thus,sensitizing drug-resistant tumors (2, 23). A curcumin analog,

G0-Y030, inhibited tumorosphere formation from ALDHA1þ/CD133þ colon CSCs (22). We used DCLK1/Lgr5/CD44 asmarkers of colon CSCs, as they mark both normal and cancerintestinal/pancreatic stem cells (4–6, 32, 33). The spheroidrelapse assay provided the first evidence that a subset ofDCLK1þ cells may be resistant to curcumin, becoming quies-cent/dormant for a period of time before re-forming spheroids(Fig. 2A andB), which is a hallmark of stem cells. Resistance of asubset of esophageal squamous carcinoma cells to curcuminhas been reported; but the authors concluded that curcumineliminates CSCs as ALDHA1þ/CD44þCSCswere eliminated bycurcumin (24). Another dietary agent was reported to signif-icantly target DCLK1þ CSCs, via the Notch signaling pathway(34). Our findings, however, suggest that DCLK1þ CSCs are noteliminated by curcumin.

Multiple signaling pathways are inhibited by curcuminin epithelial cancers, resulting in apoptotic death (10).

DCLK1 LGR5 ALDHA1

Nanog p-NFκ Bp65ser276 Total ββ-catenin

Activated caspase-3

B% Change in the ratio of target proteins/β-actin in HCT-116 tumors with

curcumin ± DCLK1-siRNA

LC3-II

A

×10 ×10 ×10 ×10

×10 ×10 ×10 ×10

Figure 6. A and B, effect ofcurcumin�DCLK1-siRNA on stemcell populations in HCT-116xenografts. Xenograft sectionswere processed as describedin Fig. 5. A, representative mergedimages of sections costained withDCLK1/CD44 or Lgr5/CD44.Images with single antibodies arepresented in Supplementary Fig.S5Bi and S5Bii. Blue, DAPI;yellow, stem cell populationscoexpressing indicated markers.B, mean � SEM of Western blotdata from three to six tumors/group, presented as % change inratio of target protein/b-actin.Ratio of control samples arbitrarilyassigned 100% values; ratio oftreated samples expressed as %of control. �, P < 0.05 versuscontrol values.






Besides apoptosis, curcumin induces caspase-3–independentapoptosis (autophagy) in cancer cells (15). Autophagy culmi-nates in either cell death or survival/quiescence/differentia-tion of tumor cells (15, 21, 18). Curcumin-induced autophagy incancer cells is mainly reported to result in cell death (12, 18).Our results, however, suggest that while curcumin inducesapoptotic/autophagic cell death of many cancer cells/CSCs, itseems to induce autophagic survival/quiescence of a subset ofDCLK1þ CSCs (Figs. 2A and B, 5Aii, and 6A), representing anovel aspect of our findings.

Autophagy has been described as a prosurvival mecha-nism that acts as a cellular switch between apoptosis andquiescence/senescence (21). Specific pathways, activated inautophagic cells in response to curcumin, protect cells fromcell death and allow cells to differentiate or become quies-cent (19, 20). Curcumin induces differentiation of autopha-gic glioma-initiating/embryonic stem cells (19, 35). Similarto our findings, Mosieniak and colleagues (20) observed thata subpopulation of curcumin-treated autophagic colon can-cer cells survived, becoming senescent/quiescent; theauthors, however, did not examine possible regrowth ofcells beyond 72 hours. Our results suggest that curcumin-induced quiescent colon cancer cells may represent DCLK1þ

CSCs, which become dormant for a period of time, followedby reformation of spheroids.

Apoptotic and autophagic cell death are not mutuallyexclusive but can induce cell death simultaneously and coop-eratively (36), which may explain coexpression of apoptotic/autophagic markers within same cells in response to curcumin(Fig. 4B). Even though autophagy and apoptosis can occur insame cells, autophagy in response to curcumin is believed tobe due to endoplasmic reticulum (ER) stress, independent ofapoptosis (37). Autophagy either allows ER-stressed cells tosurvive or drives them toward apoptosis (37); it remains to bedetermined whether ER stress plays a role in survival of asubset of DCLK1þ colon cancer cells, in response to curcumin.Besides ER stress, curcumin also induces reactive oxygenspecies (ROS), which can result in autophagy (17). It is thuspossible that a subset of DCLK1þ cells are resistant to curcu-min-induced ROS and/or ER stress, and spared from autop-hagic cell death. A complex cross-talk between several signal-ing pathways is believed to dictate the outcomeof autophagy incancer cells, which can either survive or proceed to cell death,as recently reviewed (38). In future studies, we will examine therole of some of these pathways in allowing a subpopulation ofDCLK1þ cells to go through autophagy, associated with sur-vival, rather than apoptosis.

Results of our in vivo studies also suggest that long-termtreatment of tumors with curcumin perhaps allows a subsetof DCLK1þ CSCs to proliferate and maintain tumor mass,which may explain the disconnect between potent inhibitoryeffects of curcumin against multiple growth-promotingpathways (current studies and refs. 10, 13, 14), but less thanoptimal inhibitory effects against tumor growth (Fig. 3B andC). Previously also, curcumin was reported to increase asubset of colon cancer cells in S-phase (20); based on ourresults, we believe that the PCNAþ cells likely representDCLK1þ CSCs, which survive curcumin. Results of our

preliminary studies further suggest that DCLK1þ CSCs alsosurvive other insults, such as radio/chemotherapy (unpub-lished data from our laboratory), strongly suggesting thatDCLK1 expression in CSCs is perhaps linked to chemore-sistance of CSCs, allowing cells to survive autophagy by asyet unknown mechanisms.

Because DCLK1þ cells seemed to be resistant to curcuminand other chemotherapeutic insults as well, we examinedinhibitory effects of DCLK1-siRNA. Downregulation of DCLK1expression significantly reduces growth of colon cancer cells invitro and in vivo (7, 39), as confirmed by us (Fig. 3). A surprisingand unexpected finding was that DCLK1-siRNA was moreeffective than curcumin in reducing size/weight of coloncancer tumors growing as subcutaneous xenografts (Fig.3C). Combination of DCLK1-siRNAþcurcumin was even moreeffective than either agent alone, and caused preformedtumors to lose tumor mass (Fig. 3C). Synergistic inhibitoryeffects of curcumin against cancer cells/CSCs from manydifferent organs have been previously reported with chemo-therapeutic agents (reviewed in ref. 23). Here, we demonstratefor thefirst time, synergistic inhibitory effects of curcuminwithsiRNA molecules against a stem cell marker, DCLK1. Althoughcurcumin treatment resulted in both apoptotic cell death andautophagic survival, DCLK1-siRNA caused only apoptotic celldeath of colon cancer cells (Figs. 4B and 5B). RNA interferenceagainst other CSC markers, such as CD44, also inhibit prolif-eration and induce apoptosis of colon cancer cells (3), suggest-ing that DCLK1 and CD44 may represent functional CSCmarkers, which play an important role in maintaining prolif-erative potential of cancer cells. Many reports strongly suggestthat DCLK1 expression is critically required for maintainingtumorous growths in many different organs, including intes-tines and pancreas (6, 7, 33, 39, 40), as further confirmed in thecurrent studies.

Combined regimen of curcuminþDCLK1-siRNA augment-ed both apoptotic/autophagic cell death pathways, with nosign of proliferation/survival of CSC populations. It is there-fore speculated that addition of DCLK1-siRNA overcomesresistance of a subset of DCLK1þ cells to curcumin, resultingin possible elimination of CSCs, as suggested by the resultsof the relapse assay (Fig. 2Aii); however, it remains possiblethat a subset of quiescent CSCs, positive for other stem cellmarkers, such as Lgr5, remain dormant/undetectable. Ourresults suggest that while curcumin targets Lgr5þ CSCs,DCLK1-siRNA does not eliminate LGR5þ CSCs (Fig. 6). Onthe basis of our results, it is proposed that combination ofcurcuminþDCLK1-siRNA may therefore be effective towardeliminating CSCs positive for Lgr5/DCLK1/CD44. In a pre-vious study, we had reported that addition of p38MAPK mayovercome resistance of insulin—like growth factor-II (IGF-II)–expressing colon cancer cells to curcumin (14). On thebasis of recent findings, as discussed above, it may bepossible to overcome resistance of a subset of DCLK1þ

CSCs against curcumin by adding inhibitors of ROS, and/orautophagy, to avoid possible deleterious effects of down-regulating DCLK1 in normal intestinal stem cells and neu-roprogenitor cells (6, 41, 42). Thus, in the future, it may bepossible to avoid toxic effects of radio/chemotherapy by

Kantara et al.





using combinatorial strategies with nontoxic agents (such ascurcumin and DCLK1-siRNA), which target CSCs.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: C. Kantara, R. Ullrich, P. SinghDevelopment of methodology: C. Kantara, M. O'Connell, S. Sarkar, P. SinghAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): C. Kantara, S. MoyaAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): C. Kantara, M. O'Connell, S. Sarkar, P. Singh

Writing, review, and/or revision of the manuscript: C. Kantara, M. O'Con-nell, S. Sarkar, R. Ullrich, P. SinghStudy supervision: R. Ullrich, P. Singh

Grant SupportThis work was supported by NIH grants CA97959 and CA114264 to P. Singh

and NASA grants NNX09AM08G and NNJ04HD83G to R. Ullrich.The costs of publication of this article were defrayed in part by the payment of

page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received December 12, 2013; revised February 11, 2014; accepted March 3,2014; published OnlineFirst March 13, 2014.

References1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013. CA Cancer

J Clin 2013;63:11–30.2. Ning X, Shu J, Du Y, Ben Q, Li Z. Therapeutic strategies targeting

cancer stem cells. Cancer Biol Ther 2013;14:295–303.3. Park YS, Huh JW, Lee JH, Kim HR. shRNA against CD44 inhibits cell

proliferation, invasion andmigration, and promotes apoptosis of coloncarcinoma cells. Oncol Rep 2012;27:339–46.

4. Schepers AG, Snippert HJ, StangeDE, van denBornM, van Es JH, vandeWetering M, et al. Lineage tracing reveals Lgr5þ stem cell activity inmouse intestinal adenomas. Science 2012;337:730–5.

5. Kemper K, Prasetyanti PR, De Lau W, Rodermond H, Clevers H,Medema JP. Monoclonal antibodies against Lgr5 identify humancolorectal cancer stem cells. Stem Cells 2012;30:2378–86.

6. May R, Riehl TE, Hunt C, Sureban SM, Anant S, Houchen CW.Identification of a novel putative gastrointestinal stem cell and ade-noma stem cell marker, doublecortin and CaM kinase-like-1, followingradiation injury and in adenomatous polyposis coli/multiple intestinalneoplasia mice. Stem Cells 2008;26:630–7.

7. Nakanishi Y, Seno H, Fukuoka A, Ueo T, Yamaga Y, Maruno T, et al.Dclk1 distinguishes between tumor and normal stem cells in theintestine. Nat Genet 2013;45:98–103.

8. Sarkar S, KantaraC,Ortiz I, Swiercz R, Kuo J, DaveyR, et al. Progastrinoverexpression imparts tumorigenic/metastatic potential to embryon-ic epithelial cells: phenotypic differences between transformed andnontransformed stem cells. Int J Cancer 2012;131:E1088–99.

9. Sarkar S, Swiercz R, Kantara C, Hajjar KA, Singh P. Annexin A2mediates up-regulation of NF-kB, b-catenin, and stem cell in responseto progastrin in mice and HEK-293 cells. Gastroenterology 2011;140:583–95.e4.

10. GuptaSC,PrasadS,KimJH,PatchvaS,WebbLJ, Priyadarsini IK, et al.Multitargeting by curcumin as revealed by molecular interaction stud-ies. Nat Prod Rep 2011;28:1937–55.

11. Helson L, Bolger G, Majeed M, Vcelar B, Pucaj K, Matabudul D.Infusion pharmacokinetics of Lipocurc� (liposomal curcumin) and itsmetabolite tetrahydrocurcumin in Beagle dogs. Anticancer Res2012;32:4365–70.

12. Hatcher H, Planalp R, Cho J, Torti FM, Torti SV. Curcumin: fromancient medicine to current clinical trials. Cell Mol Life Sci 2008;65:1631–52.

13. Aoki H, Takada Y, Kondo S, Sawaya R, Aggarwal BB, Kondo Y.Evidence that curcumin suppresses the growth of malignant gliomasin vitro and in vivo through induction of autophagy: role of Akt andextracellular signal-regulated kinase signaling pathways. Mol Phar-macol 2007;72:29–39.

14. Singh P, Sarkar S, Umar S, Rengifo-Cam W, Singh AP, Wood TG.Insulin-like growth factors are more effective than progastrin in revers-ing proapoptotic effects of curcumin: critical role of p38MAPK. Am JPhysiol Gastrointest Liver Physiol 2010;298:G551–62.

15. O'Sullivan-Coyne G, O'Sullivan GC, O'Donovan TR, Piwocka K,McKenna SL. Curcumin induces apoptosis-independent death inoesophageal cancer cells. Br J Cancer 2009;101:1585–95.

16. White E, DiPaola RS. The double-edged sword of autophagy modu-lation in cancer. Clin Cancer Res 2009;15:5308–16.

17. Kim JY, Cho TJ, Woo BH, Choi KU, Lee CH, Ryu MH, et al. Curcumin-induced autophagy contributes to the decreased survival of oralcancer cells. Arch Oral Biol 2012;57:1018–25.

18. Li B, Takeda T, Tsuiji K, Wong TF, Tadakawa M, Kondo A, et al.Curcumin induces cross-regulation between autophagy and apopto-sis in uterine leiomyosarcoma cells. Int J Gynecol Cancer 2013;23:803–8.

19. Zhuang W, Long L, Zheng B, Ji W, Yang N, Zhang Q, et al. Curcuminpromotes differentiation of glioma-initiating cells by inducing autop-hagy. Cancer Sci 2012;103:684–90.

20. Mosieniak G, Adamowicz M, Alster O, Jaskowiak H, SzczepankiewiczAA, Wilczynski GM, et al. Curcumin induces permanent growth arrestof humancolon cancer cells: link between senescence and autophagy.Mech Ageing Dev 2012;133:444–55.

21. Patschan S, Chen J, Polotskaia A, Mendelev N, Cheng J, Patschan D,et al. Lipid mediators of autophagy in stress-induced prematuresenescence of endothelial cells. Am J Physiol Heart Circ Physiol2008;294:H1119–29.

22. Lin L, LiuY, LiH, Li PK, Fuchs J,ShibataH, et al. Targeting coloncancerstem cells using a new curcumin analogue, GO-Y030. Br J Cancer2011;105:212–20.

23. Saha S, Adhikary A, Bhattacharyya P, DAS T, Sa G. Death by design:where curcumin sensitizes drug-resistant tumours. Anticancer Res2012;32:2567–84.

24. Almanaa TN, GeuszME, Jamasbi RJ. Effects of curcumin on stem-likecells in human esophageal squamous carcinoma cell lines. BMCComplement Altern Med 2012;12:195

25. Yu Y, Kanwar SS, Patel BB, Nautiyal J, Sarkar FH, Majumdar AP.Elimination of colon cancer stem-like cells by the combination ofcurcumin and FOLFOX. Transl Oncol 2009;2:321–8.

26. Singh P, Owlia A, Varro A, Dai B, Rajaraman S, Wood T. Gastrin geneexpression is required for the proliferation and tumorigenicity of humancolon cancer cells. Cancer Res 1996;56:4111–5.

27. Singh P, Sarkar S, Kantara C, Maxwell C. Progastrin peptides increasethe risk of developing colonic tumors: impact on colonic stem cells.Curr Colorectal Cancer Rep 2012;8:277–89.

28. Umar S, Sarkar S, Cowey S, Singh P. Activation of NF-kappaB isrequired for mediating proliferative and antiapoptotic effects of pro-gastrin on proximal colonic crypts of mice, in vivo. Oncogene2008;27:5599–611.

29. Umar S, Sarkar S, Wang Y, Singh P. Functional cross-talk betweenbeta-catenin and NFkappaB signaling pathways in colonic crypts ofmice in response to progastrin. J Biol Chem 2009;284:22274–84.

30. He C, Levine B. The Beclin 1 interactome. Curr Opin Cell Biol 2010;22:140–9

31. Paglin S, Hollister T, Delohery T, Hackett N, McMahill M, Sphicas E,et al. A novel response of cancer cells to radiation involves autophagyand formation of acidic vesicles. Cancer Res 2001;61:439–44.

32. Barker N, van Es JH, Kuipers J, Kujala P, van denBornM,CozijnsenM,et al. Identification of stem cells in small intestine and colon by markergene Lgr5. Nature 2007;449:1003–7.

33. Bailey JM, Alsina J, Rasheed ZA, McAllister FM, Fu YY, Plentz R, et al.DCLK1 marks a morphologically distinct subpopulation of cells with






stem cell properties in preinvasive pancreatic cancer. Gastroenterol-ogy 2014;146:245–56.

34. Ponnurangam S, Mammen JM, Ramalingam S, He Z, Zhang Y,Umar S, et al. Honokiol in combination with radiation targets notchsignaling to inhibit colon cancer stem cells. Mol Cancer Ther 2012;11:963–72.

35. Mujoo K, Nikonoff LE, Sharin VG, Bryan NS, Kots AY, Murad F.Curcumin induces differentiation of embryonic stem cells throughpossible modulation of nitric oxide-cyclic GMP pathway. Protein Cell2012;3:535–44.

36. Maiuri MC, Zalckvar E, Kimchi A, Kroemer G. Self-eating and self-killing: crosstalk between autophagy and apoptosis. Nat Rev Mol CellBiol 2007;8:741–52.

37. Basile V, Belluti S, Ferrari E, Gozzoli C, Ganassi S, Quaglino D, et al.Bis-dehydroxy-curcumin triggers mitochondrial-associated cell deathin human colon cancer cells through ER-stress induced autophagy.PLoS ONE 2013;8:e53664.

38. Wang S, Yu Q, Zhang R, Liu Bo. Core signaling pathways of survival/death in autophagy-related cancer networks. Int J Biochem Cell Biol2011;43:1263–66.

39. Sureban SM,MayR,Mondalek FG,QuD, PonnurangamS, Pantazis P,et al. Nanoparticle-based delivery of siDCAMKL-1 increases micro-RNA-144 and inhibits colorectal cancer tumor growth via a Notch-1dependent mechanism. J Nanobiotechnology 2011;9:40.

40. Verissimo CS,Molenaar JJ,Meerman J, Puigvert JC, Lamers F, KosterJ, et al. Silencing of themicrotubule-associated proteins doublecortin-like and doublecortin-like kinase-long induces apoptosis in neuroblas-toma cells. Endocr Relat Cancer 2010;17:399–414.

41. Matsumoto N, Pilz DT, Ledbetter DH. Genomic structure, chromo-somal mapping, and expression pattern of human DCAMKL1(KIAA0369), a homologue of DCX (XLIS). Genomics 1999;56:179–83.

42. Liu JS, Schubert CR, Fu X, Fourniol FJ, Jaiswal JK, Houdusse A, et al.Molecular basis for specific regulation of neuronal kinesin-3 motors bydoublecortin family proteins. Mol Cell 2012;47:707–21.


Kantara et al.




2014;74:2487-2498. Published OnlineFirst March 13, 2014.Cancer Res Carla Kantara, Malaney O'Connell, Shubhashish Sarkar, et al. Cancer Stem Cells, Which Are Ablated by DCLK1-siRNACurcumin Promotes Autophagic Survival of a Subset of Colon

Updated version

10.1158/0008-5472.CAN-13-3536doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2014/03/13/0008-5472.CAN-13-3536.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/74/9/2487.full#ref-list-1

This article cites 42 articles, 11 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/74/9/2487.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/74/9/2487To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-13-3536

http://cancerres.aacrjournals.org/content/suppl/2014/03/13/0008-5472.CAN-13-3536.DC1

http://cancerres.aacrjournals.org/content/74/9/2487.full#ref-list-1

http://cancerres.aacrjournals.org/content/74/9/2487.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/74/9/2487


CurcuminPromotesAutophagicSurvivalofaSubsetofColon Cancer ... · antitumor properties of curcumin....

Documents

Transcript of CurcuminPromotesAutophagicSurvivalofaSubsetofColon Cancer ... · antitumor properties of curcumin....