Inhibition of LEF1-mediated DCLK1 by Niclosamide ......Nov 16, 2018 · Committee (IACUC) of the Gwangju Institute of Science and Technology (GIST-2017-038). All work . related to

1

Inhibition of LEF1-mediated DCLK1 by Niclosamide Attenuates Colorectal Cancer Stemness

So-Yeon Park1,2,*, Ji-Young Kim3,*, Jang-Hyun Choi1,*, Jee-Heun Kim1, Choong-Jae Lee1,Pomila

Singh4, Shubhashish Sarkar4, Jeong-Heum Baek5, and Jeong-Seok Nam1,2,6†

1School of Life Sciences, Gwangju Institute of Science and Technology, Gwangju 61005, Republic of Korea

2Cell Logistics Research Center, Gwangju Institute of Science and Technology, Gwangju 61005, Republic of

Korea

3Laboratory Animal Resource Center, Gwangju Institute of Science and Technology, Gwnagju 61005, Republic

of Korea

4Department of Neuroscience and Cell Biology, University of Texas Medical Branch, Galveston, Texas, USA

5Division of Colon and Rectal Surgery, Department of Surgery, Gil Medical Center, Gachon University College

of Medicine, Incheon, Republic of Korea

6Silver Health Bio Research Center, Gwangju Institute of Science and Technology, Gwangju 61005, Republic of

Korea

Running Title: Niclosamide targets DCLK1-mediated cancer stem functions.

Keywords: Niclosamide; Wnt signaling; Colorectal cancer; Stemness; Doublecortin-like kinase1

*Equal contribution

†Correspondence: [email protected] (e-mail), +82-62-715-2893 (phone), +82-62-715-2484 (FAX)

Disclosure of Potential Conflicts of Interest: No potential conflicts of interest were disclosed.

Word count: 150-word translational relevance; 221-word abstract; 5,000-word main text with 6

figures

Research. on January 26, 2021. © 2018 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited. Author Manuscript Published OnlineFirst on November 16, 2018; DOI: 10.1158/1078-0432.CCR-18-1232

http://clincancerres.aacrjournals.org/

2

Translational Relevance

Constitutive activation of Wnt signaling has been implicated in cancer stem cells (CSCs) leading to

cancer initiation and progression. A number of Wnt inhibitors have been proposed as CSC-targeting

drugs; yet a lack of understanding of molecular regulation on CSC survival and self-renewal remains

obstacle. Niclosamide inhibits multiple components of Wnt signaling and CSC, thus we explored the

downstream targets of niclosamide by using public genomic databases, human colorectal cancer (CRC)

cells, and CRC mouse models. We found that doublecortin-like kinase1 (DCLK1)-B is predominantly

expressed in CRC cells, and it is more enriched in CSCs than in non-CSCs. DCLK1-B is

transcriptionally activated by lymphoid enhancer-binding factor 1 (LEF1) and this LEF1/DCLK1-B

axis is critical for CSC survival and self-renewal activity. Disruption of the LEF1/DCLK1-B axis by

niclosamide can impair the tumor-initiating and survival potential of CSCs. These findings between

LEF1 and DCLK1-B may improve understanding the signaling network of CSCs.




3

Abstract

Purpose: Niclosamide, an FDA-approved anthelmintic drug, has been characterized as a potent Wnt

inhibitor that can suppress tumor growth and cancer stem-like cell (CSC) populations. However, the

underlying molecular mechanisms remain poorly understood. The current study aimed to examine

how Wnt inhibition by niclosamide preferentially targets CSCs.

Experimental Design: The mechanistic role of niclosamide in CSC inhibition was examined in

public databases, human colorectal cancer (CRC) cells, CRC xenografts, and azoxymethane/dextran

sulfate sodium (AOM/DSS)-induced CRC model.

Results: Niclosamide suppresses CSC populations and their self-renewal activities in CRC cells, and

this CSC-targeting effect leads to irreversible disruption of tumor-initiating potential in vivo.

Mechanistically, niclosamide downregulates multiple signaling components of the Wnt pathway,

specifically lymphoid enhancer-binding factor 1 (LEF1) expression, which is critical for regulating

stemness. Subsequently, we identified that the doublecortin-like kinase1 (DCLK1)-B is a target of

LEF1 and upregulates cancer stemness in CRC cells. We first documented that niclosamide blocks the

transcription of DCLK1-B by interrupting the binding of LEF1 to DCLK1-B promoter. DCLK1-B

depletion impairs cancer stemness resulting in reduced survival potential and increased apoptosis, thus

sensitizing CRC to chemoradiation.

Conclusions: Disruption of the LEF1/DCLK1-B axis by niclosamide eradicates cancer stemness and

elicits therapeutic effects on CRC initiation, progression, and resistance. These findings provide a

preclinical rationale to broaden the clinical evaluation of niclosamide for the treatment of CRC.




4

Introduction

Colorectal cancer (CRC) is a major health problem worldwide owing to its high prevalence

and mortality rates (1,2). While earlier diagnosis by advanced technology and new treatment regimens

have considerably improved the survival of CRC patients in the past decades, nearly 50% of CRC

patients still face recurrence at local or distant sites after conventional therapy (3). Initially,

conventional therapy kills most cancer cells and immediately shrinks tumor while sparing a small

subpopulation, collectively referred to as cancer stem-like cells (CSCs), which re-initiate tumor

growth and colonize distant organs resulting in relapse and metastasis (4,5). CSCs engage in self-

renewal, induce tumors at low-cell density, and produce tumors with differentiated and heterogeneous

cell profiles. An increasing number of studies suggest that CSCs promote cancer progression at

various stages including tumor initiation, growth, invasion, and metastasis, and notably, they are more

resistant to anticancer therapy than are differentiated non-CSCs (5). In this regard, compounds or

designed drugs that hinder CSC-specific pathways or interfere with CSC-specific targets have been

recently recognized as promising combinatorial therapies for permanent cure. To actualize this

possibility, we need to understand molecular mechanism by which CSCs maintain stemness and

promote resistance leading to recurrence after chemoradiotherapy.

CSCs display many features of embryonic or tissue stem cells and typically demonstrate

persistent activation of one or more highly conserved signal transduction pathways involved in

development and tissue homeostasis, including Notch, Hedgehog, and Wnt pathways. In particular,

the aberrant activation of Wnt/β-catenin signaling pathway is involved in both development and

progression of CRC. During colon carcinogenesis, hyperactivation of Wnt/β-catenin signaling drives

adenomatous polyp formation (6,7), and differentiated colon epithelial cells with constitutive

activation of Wnt/β-catenin signaling can re-acquire stem cell-like properties and give rise to poorly

differentiated CRC (8,9). CRC tissues display more intense nuclear accumulation of β-catenin than do

normal tissues. Moreover, CSC populations harbor more enhanced transcriptional activity of Wnt/β-




5

catenin signaling and express higher levels of Wnt target gene than non-CSCs (10,11). This activated

Wnt/β-catenin signaling is critical for CSC to self-renew and repopulate tumor, resulting in

chemoresistance and metastasis (10,12-15). Thus, Wnt signaling inhibitors are gaining interest as

potential CSC-targeting agents and are under preclinical or clinical investigation (16-18).

Thus far, Wnt/β-catenin inhibitors have not been approved in clinics, and only few, such as

the anti-frizzled7 (FZD7) antibody vantictumab (NCT01345201) and the porcupine inhibitor LGK974

(NCT01351103), have made it to early clinical trials. Clinical trials of new lead compounds are

expensive and lengthy and need incremental process. In this regard, repurposing the existing Food and

Drug Administration (FDA)-approved drugs is being considered a promising strategy because finding

new uses for old drugs is much faster and more economical than developing a new drug from scratch;

additionally, because existing drugs have known pharmacokinetic and safety profiles and have often

been approved for human use, any newly identified uses can be rapidly evaluated in clinical trials.

Niclosamide, a teniacide from anthelmintic family that is especially effective against cestodes, has

been approved for use in humans for nearly 50 years. In the past decade, niclosamide has been

identified as a potential anticancer agent by multiple independent high-throughput screening efforts.

The Wnt-targeting effect of niclosamide was first identified through high-throughput screening of a

library of FDA-approved drugs that detect frizzled1 (FZD1) endocytosis in human osteosarcoma (19).

Accumulating evidence suggests that niclosamide interrupts Wnt/β-catenin pathway by targeting

multiple signaling components such as Dvl2 protein (20,21), β-catenin (20,21), β-catenin/TCF

complex formation (21), and LRP6 (22-24). Collectively, in contrast to other Wnt inhibitors,

niclosamide can target various components involved in Wnt/β-catenin pathway, thus it is a strong

multitargeting Wnt inhibitor (25). Wnt/β-catenin inhibition by niclosamide has shown anticancer

effects including tumor cell growth inhibition and tumor burden reduction in various types of cancer

such as colorectal (20,21), prostate (22), breast cancer (23), and glioblastoma (24). More recently,

accumulating evidence has suggested that niclosamide can be a potential CSC-targeting drug. A drug

screening of a LOPAC chemical library first identified that niclosamide efficiently suppressed




6

spheroid formation of side populations in breast CSCs (26). Likewise, niclosmiade has been reported

to decrease CSC population, such as Aldefluor+ melanoma CSCs (27) and CD133+ ovarian CSCs

(28). Although these previous data suggest that niclosamide can reduce CSC populations by

interrupting Wnt/β-catenin pathway, how Wnt inhibition by niclosamide results in the reduction of

CSC populations remains elusive. Therefore, we need to identify the molecular mechanism of CSC

inhibition by niclosamide to enable the use of niclosamide in clinics.

Taking all of these findings from previous studies into consideration, we hypothesized that

Wnt-targeting niclosamide can attenuate CSC properties in CRC and that its effects can possibly be

extended as a potential adjuvant therapeutics with chemoradiotherapy to reduce therapy resistance in

CSCs. Mechanistically, multitargeting Wnt/β-catenin by niclosamide leads to the reduction in

lymphoid enhancer factor 1 (LEF1) expression, which is correlated with therapy resistance and poor

prognosis in CRC patients. By genomic analysis of metastatic CRC patient tumors, we identified a set

of enriched LEF1 target genes among the stemness-related genes, and discovered that doublecortin-

like kinase1 (DCLK1) displayed the most dramatic decrease following niclosamide treatment.

Moreover, we found that colorectal CSCs predominantly expressed DCLK1-B rather than DCLK1-A.

Mechanistically, DCLK1-B transcription is directly activated by Wnt/β-catenin signaling and LEF1

mediates Wnt-induced CSC properties by enhancing proliferation, survival, anti-apoptosis, and self-

renewal potential of CSCs. Our results thus provide a good understanding of how niclosamide

selectively targets CSCs, suggesting DCLK1-B as a potential CSC-specific target for preventing

cancer progression and therapy resistance.




7

Materials and Methods

Ethics, cell culture and reagents

All animal experiments were carried out according to the Institutional Animal Care and Use

Committee (IACUC) of the Gwangju Institute of Science and Technology (GIST-2017-038). All work

related to human tissues was preapproved by Institutional Review Board (IRB) at Gwangju Institute

of Science and Technology (#20170410-BR-28-03-02) and the Lee Gil Ya Cancer and Diabetes

Institute of Gachon University (GCIRB-2013-66), and was conducted in accordance with the Helsinki

Declaration. Informed consent forms were signed and obtained from all subjects prior to participation.

Detailed information of CRC patient samples, cell lines, including from where and when they were

obtained, and reagents were described in Supplementary Materials and Methods.

Luciferase reporter assay

The luciferase reporter experiments using TopFlash/FopFlash plasmids were performed as reported

previously (29). Promoter-reporter constructs for DCLK1-B and DCLK1-A are described in previous

report (30). To measure luciferase activity, luminescence was detected by a luminometer (Glomax,

Promega) according to manufacturer's recommendations.

Statistical analysis

All statistical data are expressed as mean ± standard deviation (SD) (n=3). Statistical comparisons

were determined by Student’s t-test between two groups or by one-way ANOVA with Dunnett’s

multiple comparison among multiple groups. In case of in vivo experiments, the number of mice was

indicated in each legends. Log-rank test was used for Kaplan-Meier analyses. *, **, and *** indicate

p < 0.05, p < 0.01, and p < 0.001, respectively.




8

More information about methods, including the cell proliferation assay, the limiting dilution assay, the

establishment of knockdown/knockout cell lines, is provided in Supplementary Materials and

Methods.




9

Results

Niclosamide inhibits cancer stemness and therapy resistance by inhibiting Wnt/β-catenin signaling

in CRC cells

To examine whether niclosamide inhibits the CRC proliferation, we evaluated the half maximal

inhibitory concentration (IC50) values in multiple cells including HCT116, HT29, SW480, and patient-

derived CRC cells (P#21257113 and P#14005083). Patient-derived CRC cells were isolated from

primary tumors of CRC patients and pathologically validated based on immunohistochemical markers.

They were determined to express the colon-specific cytokeratin (CK) pattern, CK20+/CK7- (31), but

not to express the fibroblast marker vimentin (32) (Supplementary Figure 1A). The IC50 values of

niclosamide were below 6 μM in all tested CRC cells, while niclosamide showed minimal effect in a

normal colon epithelial cell line (CCD-18Co) even at a maximum concentration 20 μM (Figure 1A).

These data are consistent with previous report that niclosamide does not exert a significant effect

against proliferation in normal cells (20), suggesting that niclosamide possesses a therapeutic window

for its antitumor effects. Next, to determine whether niclosamide inhibits Wnt/β-catenin signaling, we

performed TOP/FOP assay at lower concentrations than IC50 (Figure 1B). Indeed, niclosamide dose-

dependently decreased the transcriptional activity of Wnt/β-catenin signaling by more than 50% at

lower concentrations than IC50. At approximately one-third of IC50 concentration (0.4 μM in HCT116

and 2 μM in HT29), niclosamide decreased Wnt/β-catenin transcriptional activity by 74.5% and 63.3%

in HCT116 and HT29 cells, respectively. We used this concentration in functional and molecular

analyses to minimize the effect on cellular proliferation. Previously, we have discovered that Wnt/β-

catenin signaling is activated to a greater extent in tumor cells than in normal cells and that it is

further enhanced in CSC populations for regulating self-renewal activity and chemoresistance

(29,33,34). Therefore, we hypothesized that niclosamide might target CSCs by inhibiting Wnt/β-

catenin signaling. Consistently, niclosamide treatment effectively decreased various CSC populations




10

such as LGR5+, CD44v6+, and Aldefluor+ CSCs in multiple CRC cells (Figure 1C, Supplementary

Figure 1B and 1C). Additionally, niclosamide inhibited the self-renewal ability of CSC populations

(Figure 1D and Supplementary Figure 1D). Moreover, niclosamide-exposed CSCs could not fully

restore their self-renewal activity even under niclosamide-free conditions. These effects were

accompanied by a shift of mRNA patterns in which a set of stemness-related transcription factors was

decreased, while differentiation marker ANPEP was enhanced by niclosamide (Supplementary Figure

1E). Next, we investigated whether CSC-targeting niclosamide could suppress the therapy resistance

of CSCs. Indeed, the surviving CRC cells following 5-fluorouracil (5-FU) or radiation exposure, the

first line therapies for CRC, displayed the increased CD44v6+ CSC populations (Figure 1E and

Supplementary Figure 1F) with the enhanced self-renewal activities (Figure 1F and Supplementary

Figure 1G). Surprisingly, combinatorial treatment of niclosamide prevented the increase of stemness

in surviving cells following conventional anticancer. Of note, these CSC-targeting effects of

niclosamide finally resulted in sensitization of CRC cells to chemotherapy (Figure 1G and 1H) and

radiotherapy (Figure 1I). Therefore, niclosamide inhibits CSC-like properties by inhibiting Wnt/β-

catenin signaling and thus can be a potential agent to overcome therapy resistance in CRC cells.

LEF1, a regulator of cancer stemness, is a potent target of niclosamide

Next, to investigate the molecular mechanism of niclosamide, we performed qPCR screening

of Wnt/β-catenin pathway genes in niclosamide-treated HCT116 cells (Figure 2A). Various Wnt

components were suppressed by niclosamide, and LEF1 showed the most potent reduction following

niclosamide treatment. In further experiments, niclosamide dose-dependently reduced both mRNA

and protein levels of LEF1 in multiple CRC cells (Figure 2B and 2C). Consistently,

immunofluorescence analysis revealed that LEF1 were increased in nuclear regions by Wnt activation,

and niclosamide could reduce both the basal and Wnt-induced nuclear LEF1 (Figure 2D). In fact,

histopathological analysis determined that colon epithelial cells expressed LEF1 more intensely in




11

cancerous tissues than in normal tissues (Figure 2E). And fluorescence-activated cell sorting (FACS)

analysis indicated that LEF1 were significantly enriched in LGR5+, CD44v6+, and ALDH1A1+ CSC

populations (Figure 2F). These data suggest that LEF1 expression is higher in cancer than in normal

tissue and further enriched in CSC populations. Therefore, we investigated whether LEF1 was critical

for regulating CSC functions. First, we established LEF1 knockout HCT116 cells using the CRISPR-

Cas9 genome editing system. We generated multiple clones of LEF1-knockout cells by using two

different gRNA sequences. LEF1 knockout clone #1 (sequence I) and clone #3 (sequence II) were

validated to lose LEF1 expression (Supplementary Figure 2A), thus used for further analysis. We

discovered that various CSC populations were significantly decreased in both LEF1-knockout clones

(Figure 2G). Moreover, the reduction of tumor sphere forming efficiency was observed in LEF1-

knockout cells, convincing the impaired self-renewal activity in the absence of LEF1 (Figure 2H).

Also, we found that high LEF1 expression was associated with poor prognosis such as low relapse-

free survival and overall survival in CRC patients (Figure 2I), in consistent with previous report that

high expression of LEF1 serves as a poor prognostic marker in CRC patients (35). Additionally, we

found that LEF1 expression was more elevated in metastatic CRC than in non-metastatic CRC (Figure

2J), and that chemoresistant CRC tumors displayed higher expression of LEF1 than chemosensitive

CRC tumors (Figure 2K). Given that metastasis and chemoresistance are typical features of CSCs

resulting in poor prognosis in CRC patients (4,5), these findings suggest that LEF1 is positively

correlated with cancer stemness in CRC patients. Taken together, we firstly documented that LEF1, a

potent target of niclosamide, is highly expressed in various CSC populations and it is required for

maintaining stemness.

DCLK1 is a target of LEF1 and positively correlated with cancer progression and CSCs in CRC

patients.

To elucidate the LEF1-mediated downstream target of niclosamide, we compared gene expression




12

profiles between metastatic and non-metastatic primary CRC samples (GSE41258) and applied gene

expression data to gene set enrichment analysis. In these computations, a group of genes that are

upregulated by LEF1 were significantly enriched in metastatic CRCs (false discovery rate q-value <

0.005) (Figure 3A). Twenty LEF1 target genes were determined as the leading-edge subset which

contributed the most to the LEF1 enrichment signal in metastatic CRCs. Next, we compared these

genes with a set of stemness-related genes (n = 3401), which were generated by combining multiple

gene sets from 3 categories, namely, adult stem cell-, embryonic stem cell-, and CSC-related genes.

Seven LEF1 target genes were related to stemness and up-regulated in metastasis CRC versus non-

metastatic CRC tissues (Figure 3B). Among them, DCLK1 was decreased the most in a dose-

dependent manner after niclosamide treatment (Figure 3B). In silico analysis showed that DCLK1

expression was up-regulated in chemoresistant CRC cells (Figure 3C) and clinical evidence strongly

suggested that DCLK1 expression is positively correlated with LEF1 expression in CRC patients

(GSE37892 and GSE14333), and with poor prognosis in CRC patients (GSE17538 and GSE14333)

(Figure 3D and 3E). Additionally, the stem cell transcription factor POU5F1 was positively correlated

with DCLK1 in CRC patient tumors (GSE17538 and GSE14333) (Supplementary Figure 3A).

Consistent with genomic analyses, immunohistochemical results revealed that the number of

DCLK1+ cells was markedly increased in the epithelial region of cancerous tissues than in that of

normal tissues (Figure 3F). Previously, DCLK1+ CRC cells were determined to be CSCs originating

from LGR5+ stem cells, suggesting that DCLK1 can be a potential CSC-specific target (8). Our

comprehensive genomic analysis indicating DCLK1 as a potential target of niclosamide that regulates

Wnt-induced stem-like properties in CRC was thus promising. Therefore, we decided to perform

further investigations on DCLK1 and its regulation by niclosamide. According to UniProt database

(https://www.uniprot.org), there are two types of DCLK1 protein isoforms generated by two distinct

promoter regions (Supplementary Figure 3B) (36-38). The A promoter (α-promoter) regulates the

transcription of ~82 kDa DCLK1 (DCLK1-A), which contains two N-terminal doublecortin (DC)

domains, a C-terminal serine/threonine kinase domain, and a middle serine/proline rich domain. The




13

B promoter (β-promoter) exists in downstream of the type A promoter and it regulates transcription of

~47 kDa DCLK1 (DCLK1-B), which lacks two N-terminal DC domains. Thus, we quantified DCLK1

isoforms in various cells (Figure 3G). Western blot analysis determined that all CRC cells

predominantly expressed DCLK1-B rather than DCLK1-A, in accordance with previous report that

DCLK1-B is highly expressed in human CRC cells and tissues (30). Indeed, DCLK1-B promoter was

about 30-fold more activated than that of DCLK1-A promoter in HCT116 cells, suggesting that the

major expression of DCLK1-B may be derived from its highly active transcription (Supplementary

Figure 3C). Consistently, we confirmed that DCLK1-B was principally expressed in CRC tissues and

its expression was elevated in CRC tissues than normal tissues (Figure 3H and Supplementary Figure

3D), as described in previous report (30). Then, we enriched CSCs by culturing multiple CRC cells in

suspension condition where CSCs were allowed to survive and proliferate to form spheres while non-

CSCs were not. Then, immunofluorescence revealed that CSC-enriched spheres expressed DCLK1 at

higher levels than did monolayer bulk cancer cells (Figure 3I). In multiple CRC cells, we newly

discovered that DCLK1-B was far more predominant than DCLK1-A in CSCs. Additionally, DCLK1-

A was also increased in CRC tissues than in normal tissue (Figure 3H and Supplementary Figure 3D)

and it was further enhanced in CSCs (Figure 3J) in accordance with previous reports that DCLK1-A

marks CSCs contributing to CRC progression in multiple CRC animal models (8,39). However, in

these previous reports, DCLK1-B has been overlooked, and more recently, overexpression of

DCLK1-B has emerged as a prognostic factor in CRC patients (30). Consistently, we observed the

principal expression of DCLK1-B in colorectal CSCs. Therefore, we tried to figure out whether

DCLK1-B is critical for cancer stemness and whether it could be a target for niclosamide in CRC cells.

Niclosamide effectively inhibits DCLK1-B expression via the Wnt/β-catenin-LEF1 axis

First, we investigated whether DCLK1-B was regulated by Wnt/β-catenin signaling using multiple

CRC cells. DCLK1-B was increased following Wnt activation at both mRNA and protein levels




14

(Figure 4A). Next, since comprehensive genomic analysis demonstrated that DCLK1 might be a

target of LEF1, we validated the molecular relationship between LEF1 and DCLK1-B using LEF1-

knockout HCT116 cells (Figure 4B). LEF1 was up-regulated by Wnt3a in wild-type and control

(empty vector-transfected) cells, while it was not in LEF1-knockout cells. Interestingly, DCLK1-B

expression was significantly diminished in LEF1-knockout cells and the up-regulation of DCLK1-B

by Wnt3a was disrupted in the absence of LEF1, suggesting that DCLK1-B was regulated by Wnt/β-

catenin signaling and LEF1 might be involved in DCLK1-B regulation. Next, we discovered that

DCLK1-B was significantly decreased by niclosamide at both mRNA and protein levels in a dose-

dependent manner (Figure 4C and 4D). Given that niclosamide treatment led to a significant

suppression in DCLK1-B promoter activity (Figure 4E), niclosamide might regulate DCLK1-B at

transcription level. Therefore, we subsequently investigated whether Wnt/β-catenin signaling directly

regulates the transcription of DCLK1-B on its promoter region. To this end, we searched the LEF1-

binding sites on DCLK1-B promoter using a prediction tool, ALLGEN PROMO database version

3.0.2 (http://alggen.lsi.upc.es/cgi-bin/promo_v3/promo/promoin it.cgi?dir DB=TF_8.3). We found

three potential LEF1-binding sites at -1743, -1595, and -1412 bp (Figure 4F). Then chromatin

immunoprecipitation assay revealed that LEF1 bound to DCLK1-B promoter between –1895 and –

1329 regions. Notably, niclosamide inhibited the binding of LEF1 to DCLK1-B promoter (Figure 4G).

Collectively, we firstly demonstrated that niclosamide could target DCLK1-B through abrogating its

transcription on its promoter region. In further investigation, we found that two siRNAs targeting

LEF1 (siLEF1) showed different efficacy on LEF1-knockdown, and that the more the siLEF1

decreased LEF1 expression, the more DCLK1-B expression was reduced (Supplementary Figure 4A).

Similarly, DCLK1-B reduction was far more significant in LEF1-knockout cells than in LEF1-

knockdown cells. Thus, these data convinced our theory that DCLK1-B expression is dependent on

LEF1 expression. In fact, LEF1 mediates nuclear responses to Wnt signals by forming the

transcriptional regulatory complexes with other components such as β-catenin and T-cell transcription

factor (TCF) families. Indeed, we found that silencing of β-catenin or TCF4 reduced DCLK1-B




15

expression (Supplementary Figure 4B and 4C), proposing that a LEF1/β-catenin/TCF4 complex may

be involved in DCLK1-B regulation. Also, DCLK1-A was suppressed by silencing of LEF1, β-catenin,

or TCF4 (Supplementary Figure 4A-4C). These data were in consistent with previous report that

DCLK1-A is transcriptionally regulated by a direct binding of LEF1/β-catenin/TCF4 complex to its

DCLK1-A promoter (30). Furthermore, we firstly determined that Wnt3a ligand could up-regulate

DCLK1-A in multiple CRC cells (Supplementary Figure 4D) and provided new evidences that

niclosmide could suppress DCLK1-A at transcription level (Supplementary Figure 4E-4G). Taken

together, our data newly discovered that DCLK1 is a downstream target of Wnt/β-catenin signaling,

therefore, a potent target of niclosamide.

DCLK1-B expression is a critical driver of initiation and maintenance of CRC stemness

To examine whether DCLK1-B is critical for regulating cancer stemness in colorectal CSCs, we

designed siRNAs specifically targeting DCLK1-B and validated that two siRNA sequences

(siDCLK1-B #1 and siDCLK1-B #2) efficiently decreased DCLK1-B in CRC cells without affecting

DCLK1-A (Supplementary Figure 5A-5C). Limiting dilution assay demonstrated that DCLK1-B-

knockdown impaired self-renewal abilities of CSCs (Figure 5A), resulting in a significant decrease in

survival potential of CSCs (Figure 5B and Supplementary Figure 5D). Moreover, DCLK1-B-

knockdown increased AnnexinV+ apoptotic cells (Figure 5C) by activating apoptotic cascades, such

as cleaved poly ADP-ribose polymerase (PARP) and cleaved caspase 3 (Supplementary Figure 5E and

5F). In addition, CSC inhibitory effects of DCLK1-B-knockdown led to sensitization of CRC cells to

anticancer therapy such as 5-FU and radiation (Figure 5D and 5E). Next, to estimate in vivo function

of DCLK1-B, we established DCLK1-B-knockout HCT116 cells by CRISPR-Cas9 system. Multiple

DCLK1-B-knockout clones were generated using two different gRNA sequences (Supplementary

Figure 5G). Clone #1 (sequence I) and clone #3 (sequence II) were validated to lose DCLK1-B

expression without affecting DCLK1-A. In similar with DCLK1-B-knockdown effects (Figure 5A),




16

the frequency of self-renewing CSC and the proportion of CSC populations were diminished by

DCLK1-B-knockout (Supplementary Figure 5H and 5I). Next, we evaluated in vivo CSC properties of

DCLK1-B-knockout cells (clone #1) using two different CRC mouse models. Since growing evidence

suggests that CSCs display greater tumorigenic and metastatic potential in vitro and in vivo than non-

CSC cancer cells (4), we estimated both tumor-initiating and metastatic potential of DCLK1-B-

knockout cells. First, we subcutaneously transplanted limiting dilutions of cell preparations into mice

and found that the frequency of tumor-initiating cells was decreased with a statistically significance

(Figure 5F, Supplementary Figure 5J). Moreover, DCLK1-B-knockout significantly reduced primary

tumor volume (Figure 5G, Supplementary Figure 5K). Next, we used a tail vein injection model

which is a common model for studying lung metastasis. Tail vein injection of HCT116 cells resulted

primarily in pulmonary metastases. Interestingly, DCLK1-B-knockout diminished both the number

and size of metastatic colonies on lungs (Figure 5H and 5I), suggesting that DCLK1-B expression

appears to be critical for founding of metastatic colonies and their subsequent robust outgrowth.

Lastly, we examined whether the reduction of DCLK1-B mediates CSC-inhibitory effects of

niclosamide by using DCLK1-B-overexpressing cells (Supplementary Figure 5L). DCLK1-B-

overexpressing cells displayed the increased CD44v6+ CSC population and self-renewal activities.

More importantly, DCLK1-B overexpression rescued cancer stemness under niclosamide treatment

(Figure 5J and Figure 5K), suggesting that suppressive effects of niclosamide on CSCs are achieved,

at least in part, through disruption of DCLK1-B expression. Taken together, our results indicated that

DCLK1-B, a downstream target of niclosamide, is critical for maintaining CSC populations and CSC

properties both in vitro and in vivo.

Niclosamide exerts a potent in vivo antitumor effect in both HCT116 xenografts and AOM/DSS-

induced spontanenous CRC models

Following mechanistic studies of niclosamide on CSC regulation, we investigated whether




17

niclosamide can be a potential therapeutic agent for CRC using multiple in vivo systems. First, to

investigate the tumor growth-inhibitory effect of niclosamide, we subcutaneously xenografted

HCT116 cells and treated mice with niclosamide (10 mg/kg or 40 mg/kg) when the mean primary

tumor volume reached 100 mm3. Compared to vehicle, niclosamide significantly inhibited primary

tumor growth (Figure 6A-6C). In addition, 10 mg/kg seemed to be sufficiently potent in exerting a

maximum effect on primary tumor growth, since there was no significant difference between 10

mg/kg and 40 mg/kg niclosamide-treated groups. Also, we evaluated therapeutic effect of niclosamide

in patient-derived tumor xenograft (PDX) model (Supplementary Figure 6A-6C), which was

authenticated by short tandem repeat profiling analysis (Supplementary Figure 6D). The growth of

PDX tumors were also inhibited by niclosamide treatment (40 mg/kg). Second, we used the classical

model of spontaneous CRC, AOM/DSS model, to evaluate whether niclosamide can be used as a

chemopreventive agent against CRC development (Figure 6D). The niclosamide treatment (40 mg/kg)

from week 1 to 13 significantly blocked CRC development, indicating the decreased number of

colorectal polyps than control mice (Figure 6E). Additionally, AOM/DSS treatment significantly

reduced colon length, which is a classical symptom of inflammation; however, co-treatment with

niclosamide blocked the AOM/DSS-induced shortening of colon (Figure 6F). Microscopic

observations revealed that niclosamide efficiently blocked the AOM/DSS-induced development of

intermediate- or high-grade dysplasia (Figure 6G). Immunofluorescence analysis demonstrated that

AOM/DSS-induced dysplastic tissues expressed higher levels of DCLK1 than did normal tissues, and

niclosamide reduced the number of DCLK1+ dysplastic regions. Next, to investigate whether this

anticancer effect of niclosamide is derived from its CSC-targeting effect, we isolated HCT116 cells

from vehicle- or niclosamide-treated primary tumors and evaluated their self-renewal ability through

sphere-forming assays (Figure 6H). HCT116 cells from niclosamide-treated mice did not form

tumorspheres as efficiently as HCT116 cells from vehicle-treated mice did. These in vivo results are

consistent with our in vitro data, showing that niclosamide-exposed CRC cells could not restore their

self-renewal ability even under niclosamide-free conditions (Figure 1D). Immunohistochemical




18

analyses demonstrated that CD44v6+ CSC populations were decreased in niclosamide-treated primary

tumors (Figure 6I), indicating the reduced expression of stemness-related transcription factor OCT4

(Figure 6J). In the same manner, in vivo limiting dilution assay revealed that niclosamide treatment

irreversibly reduced the frequency of tumor-initiating cells in PDX tumors (Supplementary Figure 6E)

resulting in reduction of tumor growth (Supplementary Figure 6F). Accordingly, the molecular targets

of niclosamide, LEF1 and DCLK1, were decreased in niclosamide-treated primary tumors (Figure 6K,

6L, Supplementary Figure 6G, and 6H). Thus, CSC-targeting niclosamide efficiently attenuated both

tumor growth and AOM/DSS-induced colonic dysplasia, suggesting that the inhibition of

Wnt/LEF1/DCLK1 axis might be a new therapeutic strategy for CRC treatment as well as

chemoprevention of CRC development.




19

Discussion

The use of niclosamide, an anthelmintic family drug, has now been extended to multiple disease

models including Helicobacter pylori infection, Parkinson’s disease, and severe acute respiratory

syndrome (40-42). Recently, this antiparasitic drug has been proposed as a promising anticancer agent

in several types of cancer due to its remarkable ability to inhibit tumor growth (20-24). However, the

potential mechanism of niclosamide during CRC initiation and progression is still elusive. CSCs are

believed to initiate cancer and be resistant to chemoradiation by demonstrating higher survival

potential and lower apoptosis than non-CSCs, thereby resulting in metastasis and poor prognosis (43).

In this study, we reported that niclosamide efficiently decreases therapy resistance in CRCs by

reducing CSC populations and their self-renewal activity, thereby attenuating the survival potential of

CSCs following chemoradiation (Figure 1C-1I). These findings potentially broaden the clinical

application of niclosamide not only to shrink tumors but also to improve the efficacy of current

chemoradiotherapy.

Intensive investigations are currently focusing on targeting CSCs; however, therapeutic strategies

targeting CSCs are still limited because CSC markers are often shared by normal stem cells (NSCs).

In 2012, DCLK1 was first identified as a CSC-specific marker through in vivo lineage-tracing

experiments in CRC mouse models (39). DCLK1 is expressed in differentiated tuft cells but not in

NSCs in the normal intestine, while DCLK1 is detected in CSCs that continuously produce tumor cell

progeny in the intestinal polyps in ApcMin/+ mice. Currently, accumulating evidence suggests that long-

lived DCLK1+ tuft cells may be responsible for CRC development by participating as tumor-initiating

populations with persistent Wnt activation (8). DCLK1 was determined to be responsible for cancer

progression via the enhancement of survival (44), self-renewal (44), and epithelial-mesenchymal

transition (45,46). Moreover, clinical studies in CRC patients demonstrated that DCLK1 was

expressed in low-grade adenomas, and its levels increased with worsening severity of dysplasia (47).

DCLK1 expression was highly observed in advanced adenomas, which have a clinically higher




20

malignant potential in CRC patients (47).

There are two types of DCLK1 protein isoforms generated by two distinct promoter regions,

DCLK1-A and DCLK1-B promoter. In this study, we firstly documented that colorectal CSCs

predominantly expressed DCLK1-B rather than DCLK1-A (Figure 3J), and validated that deletion of

DCLK1-B could abrogate CSC populations and their stem-like functions such as self-renewal,

survival, and anti-apoptotic potential (Figure 5A-5I). Given that recent clinical investigations provide

more evidences that prognostic characteristics of DCLK1 may be derived from DCLK1-B rather than

DCLK1-A (30,48,49), our data suggest that DCLK1-B might be a promising target for eliminating

CSCs. Moreover, our demonstration of niclosamide targeting Wnt/LEF1/DCLK1-B axis suggests a

new therapeutic strategy for CSC inhibition, and further investigation on DCLK1-B may be helpful

for developing other Wnt inhibitors and CSC-targeting drugs.

Several previous reports have shown that niclosamide effectively reduces tumor burden in various

type of cancer including breast cancer, melanoma and CRC (20-24). In consistent with these reports,

we validated that niclosamide serves as a potent anticancer drug by inhibiting tumor growth (Figure

6A-6C). Also, we newly documented that in vivo treatment with niclosamide irreversibly impaired the

self-renewal activity of CSCs in primary tumors resulting in a reduction in CSC populations (Figure

6H and 6I). In fact, niclosamide treatment has been demonstrated to exhibit in vivo anti-tumor activity

in HCT116 subcutaneous tumors when it was treated four days after tumor cell inoculation (20),

suggesting that niclosamide abrogates primary tumor formation. In this study, we used the same

HCT116 subcutaneous model, but we started the niclosamide treatment after the primary tumor

volume reached 100 mm3, to mimic clinical circumstance of chemotherapy (Figure 6A). Our data

showed that niclodsamide treatment inhibited primary tumor growth even when it was started after the

formation of primary tumor burden. Thus, our HCT116 model may provide preclinical evidence that

niclosamide helps the treatment of CRC patients for primary tumor shrinkage. In another CRC model,

we documented that niclosamide delayed AOM/DSS-induced development of colon dysplasia,




21

suggesting that niclosamide can work as an inhibitor of cancer progression (Figure 6D-6G). Under

normal conditions, DCLK1 is expressed in tuft cells at a relatively low frequency, while the number

of DCLK1+ cells expands in dysplasia region of AOM/DSS-treated mice. This result is consistent

with previous reports that DCLK+ cells give rise to cancer by producing cancer cells during colon

carcinogenesis. Recently, nonsteroidal anti-inflammatory drugs (NSAIDs), such as celecoxib,

rofecoxib, and valdecoxib, have been reported to be used as cancer chemopreventive agents, which

act by blocking inflammatory signaling that contributes to colon carcinogenesis (50). In this study, we

showed that in vivo treatment of niclosamide reduced the number of DCLK1+ polyps during

AOM/DSS-induced colon carcinogenesis. Therefore, we propose that similar to NSAIDs, niclosamide

might work as chemopreventive drug for CRC.

In summary, we revealed that niclosamide exerts in vivo effects against both colon

carcinogenesis and tumor growth by targeting the Wnt/LEF1/DCLK1-B axis-mediated CSC

properties. Niclosamide inhibits certain CSC functions including survival, anti-apoptosis, and self-

renewal, resulting in a reduction in CSC populations. Moreover, CSC-targeting niclosamide

successfully sensitizes CRC to chemoradiation. These findings provide a preclinical rationale to

broaden the clinical evaluation of niclosamide for CRC treatment.




22

Additional information

Supplementary information including the Supplementary Table Legends, the Supplementary Figure

Legends, and Supplementary Materials and Methods, is attached as a separate file.

Authors’ Contribution

Conception and design: J.-S. Nam, J.-Y. Kim, S.-Y. Park

Development of methodology or material support: J.-S. Nam, P. Singh, S. Sarkar, J.-H. Baek

Acquisition of data (provided animals, provided facilities, etc): J.-H. Choi, S.-Y. Park, J.-H. Kim

Analysis and interpretation of data: J.-Y Kim, J.-H. Kim, C.-J. Lee, J.-H. Choi, S.-Y. Park

Writing, review, and/or revision of the manuscript: J.-S. Nam, S.-Y. Park, J.-Y. Kim

Study supervision: J.-S. Nam

Acknowledgements

We thank Heo Sukyung for establishing the immunohistochemistry assay and Kim Hyeseon for

assistance with Western blotting and cell maintenance.

Grant support

This research was supported by a grant of the Korea Health Technology R&D Project through the

Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare,

Republic of Korea (grant number: HI15C2056). Additionally, this work was supported by the

National Research Foundation of Korea Grant funded by the Korean Government (NRF-




23

2017R1E1A1A01075125), by a grant from the Cell Logistics Research Center of the National

Research Foundation of Korea (NRF-2016R1A5A1007318), and by a Gwangju Institute of Science

and Technology (GIST) Research Institute (GRI) grant funded by the GIST in 2018.




24

References

1. Arnold M, Sierra MS, Laversanne M, Soerjomataram I, Jemal A, Bray F. Global patterns and

trends in colorectal cancer incidence and mortality. Gut 2016:gutjnl-2015-310912.

2. Siegel RL, Fedewa SA, Anderson WF, Miller KD, Ma J, Rosenberg PS, et al. Colorectal

cancer incidence patterns in the United States, 1974–2013. JNCI: Journal of the National

Cancer Institute 2017;109(8).

3. Kanwar SS, Poolla A, Majumdar AP. Regulation of colon cancer recurrence and development

of therapeutic strategies. World J Gastrointest Pathophysiol 2012;3(1):1.

4. Zarour LR, Anand S, Billingsley KG, Bisson WH, Cercek A, Clarke MF, et al. Colorectal

cancer liver metastasis: evolving paradigms and future directions. Cell Mol Gastroenterol

Hepatol 2017;3(2):163-73.

5. Batlle E, Clevers H. Cancer stem cells revisited. Nat Med 2017;23(10):1124.

6. Anderson EC, Wong MH. Caught in the Akt: regulation of Wnt signaling in the intestine.

Gastroenterology 2010;139(3):718-22.

7. Novellasdemunt L, Antas P, Li VS. Targeting Wnt signaling in colorectal cancer. A review in

the theme: cell signaling: proteins, pathways and mechanisms. American Journal of

Physiology-Cell Physiology 2015;309(8):C511-C21.

8. Westphalen CB, Asfaha S, Hayakawa Y, Takemoto Y, Lukin DJ, Nuber AH, et al. Long-lived

intestinal tuft cells serve as colon cancer–initiating cells. The Journal of clinical investigation

2014;124(3):1283-95.

9. Schwitalla S, Fingerle AA, Cammareri P, Nebelsiek T, Göktuna SI, Ziegler PK, et al.

Intestinal tumorigenesis initiated by dedifferentiation and acquisition of stem-cell-like

properties. Cell 2013;152(1):25-38.

10. Vermeulen L, Felipe De Sousa EM, Van Der Heijden M, Cameron K, De Jong JH, Borovski T,

et al. Wnt activity defines colon cancer stem cells and is regulated by the microenvironment.

Nat Cell Biol 2010;12(5):468.




25

11. Shenoy AK, Fisher RC, Butterworth EA, Pi L, Chang L-J, Appelman HD, et al. Transition

from colitis to cancer: high Wnt activity sustains the tumor-initiating potential of colon cancer

stem cell precursors. Cancer Res 2012;72(19):5091-100.

12. Chikazawa N, Tanaka H, Tasaka T, Nakamura M, Tanaka M, Onishi H, et al. Inhibition of

Wnt signaling pathway decreases chemotherapy-resistant side-population colon cancer cells.

Anticancer Res 2010;30(6):2041-8.

13. Mohammed MK, Shao C, Wang J, Wei Q, Wang X, Collier Z, et al. Wnt/β-catenin signaling

plays an ever-expanding role in stem cell self-renewal, tumorigenesis and cancer

chemoresistance. Genes & diseases 2016;3(1):11-40.

14. Uchida H, Yamazaki K, Fukuma M, Yamada T, Hayashida T, Hasegawa H, et al.

Overexpression of leucine‐rich repeat‐containing G protein‐coupled receptor 5 in colorectal

cancer. Cancer Sci 2010;101(7):1731-7.

15. Emons G, Spitzner M, Reineke S, Möller J, Auslander N, Kramer F, et al. Chemoradiotherapy

Resistance in Colorectal Cancer Cells is Mediated by Wnt/β-catenin Signaling. Mol Cancer

Res 2017;15(11):1481-90.

16. Kahn M. Can we safely target the WNT pathway? Nature reviews Drug discovery

2014;13(7):513.

17. Koury J, Zhong L, Hao J. Targeting signaling pathways in cancer stem cells for cancer

treatment. Stem Cells Int 2017;2017.

18. de Sousa e Melo F, Vermeulen L. Wnt signaling in cancer stem cell biology. Cancers (Basel)

2016;8(7):60.

19. Chen M, Wang J, Lu J, Bond MC, Ren X-R, Lyerly HK, et al. The anti-helminthic

niclosamide inhibits Wnt/Frizzled1 signaling. Biochemistry 2009;48(43):10267-74.

20. Osada T, Chen M, Yang XY, Spasojevic I, Vandeusen JB, Hsu D, et al. Antihelminth

compound niclosamide downregulates Wnt signaling and elicits antitumor responses in

tumors with activating APC mutations. Cancer Res 2011;71(12):4172-82.




26

21. Sack U, Walther W, Scudiero D, Selby M, Kobelt D, Lemm M, et al. Novel effect of

antihelminthic Niclosamide on S100A4-mediated metastatic progression in colon cancer. J

Natl Cancer Inst 2011;103(13):1018-36.

22. Lu W, Lin C, Roberts MJ, Waud WR, Piazza GA, Li Y. Niclosamide suppresses cancer cell

growth by inducing Wnt co-receptor LRP6 degradation and inhibiting the Wnt/β-catenin

pathway. PLoS One 2011;6(12):e29290.

23. Londoño-Joshi AI, Arend RC, Aristizabal L, Lu W, Samant RS, Metge BJ, et al. Effect of

niclosamide on basal-like breast cancers. Mol Cancer Ther 2014;13(4):800-11.

24. Wieland A, Trageser D, Gogolok S, Reinartz R, Höfer H, Keller M, et al. Anticancer effects

of niclosamide in human glioblastoma. Clin Cancer Res 2013;19(15):4124-36.

25. Li Y, Li P-K, Roberts MJ, Arend RC, Samant RS, Buchsbaum DJ. Multi-targeted therapy of

cancer by niclosamide: A new application for an old drug. Cancer Lett 2014;349(1):8-14.

26. Wang Y-C, Chao T-K, Chang C-C, Yo Y-T, Yu M-H, Lai H-C. Drug screening identifies

niclosamide as an inhibitor of breast cancer stem-like cells. PLoS One 2013;8(9):e74538.

27. Zhou J, Jin B, Jin Y, Liu Y, Pan J. The antihelminthic drug niclosamide effectively inhibits the

malignant phenotypes of uveal melanoma in vitro and in vivo. Theranostics 2017;7(6):1447.

28. Arend RC, Londoño-Joshi AI, Gangrade A, Katre AA, Kurpad C, Li Y, et al. Niclosamide and

its analogs are potent inhibitors of Wnt/β-catenin, mTOR and STAT3 signaling in ovarian

cancer. Oncotarget 2016;7(52):86803-15.

29. Jang G-B, Hong I-S, Kim R-J, Lee S-Y, Park S-J, Lee E-S, et al. Wnt/β-catenin small-

molecule inhibitor CWP232228 preferentially inhibits the growth of breast cancer stem-like

cells. Cancer Res 2015;75(8):1691-702.

30. O’Connell MR, Sarkar S, Luthra GK, Okugawa Y, Toiyama Y, Gajjar AH, et al. Epigenetic

changes and alternate promoter usage by human colon cancers for expressing DCLK1-

isoforms: Clinical Implications. Sci Rep 2015;5:14983.

31. O'Sullivan B, Brierley JD, D'Cruz A, Fey M, Pollock RE, Vermorken J, et al. UICC manual




27

of clinical oncology. John Wiley & Sons; 2015.

32. Goodpaster T, Legesse-Miller A, Hameed MR, Aisner SC, Randolph-Habecker J, Coller HA.

An immunohistochemical method for identifying fibroblasts in formalin-fixed, paraffin-

embedded tissue. J Histochem Cytochem 2008;56(4):347-58.

33. Kim J-Y, Lee H-Y, Park K-K, Choi Y-K, Nam J-S, Hong I-S. CWP232228 targets liver cancer

stem cells through Wnt/β-catenin signaling: a novel therapeutic approach for liver cancer

treatment. Oncotarget 2016;7(15):20395.

34. Jang G-B, Kim J-Y, Cho S-D, Park K-S, Jung J-Y, Lee H-Y, et al. Blockade of Wnt/β-catenin

signaling suppresses breast cancer metastasis by inhibiting CSC-like phenotype. Sci Rep

2015;5:12465.

35. Wang W-J, Yao Y, Jiang L-L, Hu T-H, Ma J-Q, Ruan Z-P, et al. Increased LEF1 expression

and decreased Notch2 expression are strong predictors of poor outcomes in colorectal cancer

patients. Dis Markers 2013;35(5):395-405.

36. Engels BM, Schouten TG, van Dullemen J, Gosens I, Vreugdenhil E. Functional differences

between two DCLK splice variants. Mol Brain Res 2004;120(2):103-14.

37. Omori Y, Suzuki M, Ozaki K, Harada Y, Nakamura Y, Takahashi E-i, et al. Expression and

chromosomal localization of KIAA0369, a putative kinase structurally related to Doublecortin.

J Hum Genet 1998;43(3):169-77.

38. Shang L, Kwon Y-G, Nandy S, Lawrence DS, Edelman AM. Catalytic and regulatory

domains of doublecortin kinase-1. Biochemistry 2003;42(7):2185-94.

39. Nakanishi Y, Seno H, Fukuoka A, Ueo T, Yamaga Y, Maruno T, et al. Dclk1 distinguishes

between tumor and normal stem cells in the intestine. Nat Genet 2013;45(1):98.

40. Tharmalingam N, Port J, Castillo D, Mylonakis E. Repurposing the anthelmintic drug

niclosamide to combat Helicobacter pylori. Sci Rep 2018;8(1):3701.

41. Barini E, Miccoli A, Tinarelli F, Mulholand K, Kadri H, Khanim F, et al. The Anthelmintic

Drug Niclosamide and its Analogues Activate the Parkinson's Disease Associated Protein




28

Kinase PINK1. ChemBioChem 2017.

42. Wu C-J, Jan J-T, Chen C-M, Hsieh H-P, Hwang D-R, Liu H-W, et al. Inhibition of severe

acute respiratory syndrome coronavirus replication by niclosamide. Antimicrob Agents

Chemother 2004;48(7):2693-6.

43. Abdullah LN, Chow EK-H. Mechanisms of chemoresistance in cancer stem cells. Clinical

and translational medicine 2013;2(1):3.

44. Chandrakesan P, Yao J, Qu D, May R, Weygant N, Ge Y, et al. Dclk1, a tumor stem cell

marker, regulates pro-survival signaling and self-renewal of intestinal tumor cells. Mol

Cancer 2017;16(1):30.

45. Chandrakesan P, Panneerselvam J, Qu D, Weygant N, May R, Bronze M, et al. Regulatory

roles of Dclk1 in epithelial mesenchymal transition and cancer stem cells. Journal of

carcinogenesis & mutagenesis 2016;7(2).

46. Liu W, Wang S, Sun Q, Yang Z, Liu M, Tang H. DCLK1 Promotes Epithelial‐Mesenchymal

Transition via the PI3K/Akt/NF‐κB Pathway in Colorectal Cancer. Int J Cancer 2017.

47. Takiyama A, Tanaka T, Kazama S, Nagata H, Kawai K, Hata K, et al. DCLK1 Expression in

Colorectal Polyps Increases with the Severity of Dysplasia. In Vivo 2018;32(2):365-71.

48. Sarkar S, O'Connell MR, Okugawa Y, Lee BS, Toiyama Y, Kusunoki M, et al. FOXD3

Regulates CSC Marker, DCLK1-S, and Invasive Potential: Prognostic Implications in Colon

Cancer. Mol Cancer Res 2017;15(12):1678-91.

49. Sarkar S, Popov VL, O'Connell MR, Stevenson HL, Lee BS, Obeid RA, et al. A novel

antibody against cancer stem cell biomarker, DCLK1-S, is potentially useful for assessing

colon cancer risk after screening colonoscopy. Lab Invest 2017;97(10):1245.

50. Thun MJ, Henley SJ, Patrono C. Nonsteroidal anti-inflammatory drugs as anticancer agents:

mechanistic, pharmacologic, and clinical issues. J Natl Cancer Inst 2002;94(4):252-66.




29

Figure legends

Figure 1. Niclosamide attenuates stemness and therapy resistance in CRC cells

(A) Proliferation inhibition by niclosamide treatment was determined for 48 hours by the MTT assay

in multiple CRC cells, patient-derived primary cells (P#21257113 and P#14005083), and normal

epithelial cells (CCD-18Co). (B) Both HCT116 cells and HT29 cells were transiently transfected with

TOPflash and FOPflash vectors and treated with either niclosamide or a Wnt activator (LiCl, 20 mM)

for 18 hours. Luciferase activity was normalized by β‐galactosidase activity. (C) Reduction in CSC

population following niclosamide treatment was determined in HCT116 cells by FACS analysis based

on various CSC markers including LGR5, CD44v6, and Aldefluor. (D) Self-renewal activity of both

HCT116 cells and HT29 cells were assessed by tumorsphere-forming assays. Primary tumorspheres

treated with niclosamide were collected and dissociated into single cells. These cells were

subsequently re-plated in culture dishes without additional treatment to form secondary tumorspheres.

Scale bar represents 100 μm. (E and F) HCT116 cells were treated with 5-FU (100 nM) or exposed to

radiation (2 Gy) with or without niclosamide 0.2 μM. Afther 48 hours, cells were subjected to (E)

FACS analysis to estimate the proportion of CD44v6+ CSCs and (F) their self-renewal capacity was

determined by sphere-forming assays. (G and H) HCT116 cells were treated with various

concentrations of (G) oxaliplatin or (H) 5-FU with or without niclosamide treatment. Cell viability

was measured by MTT assays to determine whether niclosamide sensitizes cells to anti-cancer drugs.

(I) HCT116 cells were exposed to various doses of radiation from 1 Gy to 10 Gy with or without

niclosamide treatment. Proportion of surviving cells was determined by counting the number of

surviving colonies. Bar graphs represent the mean ± SD, and statistical analyses were performed by

one-way ANOVA with Dunnett’s multiple comparison. *, **, and *** indicate p < 0.05, p < 0.01, and

p < 0.001, respectively.




30

Figure 2. Niclosamide inhibits cancer stemness by attenuating LEF1 in CRC cells

(A) Inhibitory effect of niclosamide on the expression of Wnt/β-catenin pathway-related genes was

determined by RT-PCR in HCT116 cells after 48-hour treatment of niclosamide. Quantitation was

performed by RT-PCR, and data are presented as fold change relative to control (log2). (B and C)

Inhibitory effect of niclosamide on LEF1 expression was assessed in CRC cells through (B) real-time

PCR and (C) Western blotting in HCT116 cells after 48-hour treatment of niclosamide. (D)

Stimulatory effect of Wnt activator (LiCl, 20 mM) on nuclear LEF1 proteins was attenuated by

niclosamide treatment (0.4 μM). (E) LEF1 expression levels were visualized by immunostaining in

paired normal tissues and cancer tissues obtained from identical CRC patients. They were presented

with matched H&E images. Blue indicates nuclei, and red indicates LEF1. Scale bar represents 100

μm. (F) FACS analysis revealed LEF1 expression levels in LGR5+, CD44v6+, and ALDH1A1+ CSC

populations in HCT116 cells. (G and H) Reduction in cancer stemness in LEF1-knockout cells were

determined by evaluating (G) various CSC populations and (H) their self-renewal activity. (G) FACS

analysis demonstrated the number of LGR5+, CD44V6+, and ALDEFLUOR+ CSCs. (H) Self-renewal

activity of LEF1-knockout cells was determined by sphere-forming assays. (I) Kaplan-Meier survival

analyses were conducted in CSC patients based on LEF1 expression in two independent cohorts

(GSE39582 and GSE17538). (J) LEF1 expression were significantly elevated in metastatic CRC

patient tumors (Dukes’ stage D) versus early non-invasive CRC patient tumors (Dukes’ stage A)

(GSE14333). (K) Comparison of LEF1 expression between chemoresistant and chemosensitive CRC

cells from Oncomine (Boyer cell line). Bar graphs represent the mean ± SD. Statistical analyses were

performed by one-way ANOVA with Dunnett’s multiple comparison. *, **, and *** indicate p < 0.05,

p < 0.01, and p < 0.001, respectively.




31

Figure 3. DCLK1 is a target of LEF1 and correlates with CSCs and poor prognosis in CRC

patients

(A) GSEA of metastatic CRC patient tumors showed significant enrichment in LEF1 target genes

(FDR q-value < 0.005) (GES41258) (B) A set of leading-edge subset of LEF1 target genes and a set of

stemness-related genes (described in the Supplementary Materials and Methods) were compared, and

seven common genes were identified. mRNA levels of the seven genes were quantified by real-time

PCR following 48-hour treatment of niclosamid. (C) mRNA expression levels of DCLK1 were

compared between chemosensitive and chemoresistant CRC cells from Oncomine (Gyorffy cell line).

(D) Significant positive correlations between DCLK1 and LEF1 were observed in CRC patient tumors

(GSE37892 and GSE14333). Statistical significance was determined by correlation analysis using

GraphPad software. (E) Kaplan-Meier survival analyses were conducted based on DCLK1 expression

in two independent CRC patients cohorts (GSE17538 and GSE14333). (F) DCLK1 expression was

determined by immunofluorescence assay in patient-derived CRC tissues and the paired normal

adjacent to tumor (NAT) tissues from the identical CRC patients. (G) Protein levels of DCLK1-A (82

kDa) and DCLK1-B (47 kDa) isoforms were determined in normal mouse brain tissues, normal

human colon epithelial cells, and human CRC cells by Western blot assays. Total amount of loaded

protein is indicated above the wells. β-Actin was used as a loading control. (H) DCLK1-A and

DCLK1-B protein levels were determined in CRC tissues and the normal tissues obtained from the

identical CRC patients. (I) HCT116 cells were cultured under attached monolayer condition or sphere-

forming conditions to enrich for CSCs. Immunofluorescence assays were performed to visualize

DCLK1 expression. Blue indicates nuclei, and red indicates DCLK1. Scale bar represents 100 μm. (J)

DCLK1-A and DCLK1-B protein levels were compared between attached monolayer CRC cells and

CSC-enriched sphere CRC cells by Western blot assays. Bar graphs represent the mean ± SD, and

statistical analyses were performed by one-way ANOVA with Dunnett’s multiple comparison. *, **,

and *** indicate p < 0.05, p < 0.01, and p < 0.001, respectively.




32

Figure 4. Niclosamide effectively inhibits DCLK1-B expression via the Wnt/β-catenin-LEF1 axis

(A) Time-dependent induction of DCLK1-B protein and mRNA levels following Wnt activation in

multiple CRC cells. Cells were treated with Wnt3a (100ng/ml) in the absence of serum for indicated

times, then they were subjected to real-time PCR or Western blot analysis. (B) Protein levels of

DCLK1-B and LEF1 were determined in wild-type, control (empty vector-transfected), and LEF1-

knockout cells with or without Wnt activation (Wnt3a, 100 ng/ml, 2 hours) by Western blot analysis.

(C and D) Inhibitory effects of niclosamide on DCLK1-B expression were determined in various

human CRC cells by (C) RT-PCR and (D) Western blot assays. Cells were treated with niclosamide

for 48 hours. (E) Promoter reporter assay was performed against DCLK1-B promoter in HCT116 cells

following 24 hour-treatment of niclosamide in dose-dependent manner. (F) Predicted binding sites of

LEF1 on DCLK1-B promoter region according to ALLGEN PROMO database version 3.0.2. (G)

ChIP assay was conducted in HCT116 cells with or without niclosamide treatment (0.2 μM).

Immunoprecipitation was conducted against LEF1 and LEF1-bound DNA fragments amplified by

PCR. PCR products were confirmed by gel electrophoresis. Bar graphs represent the mean ± SD (n=3),

and statistical analyses were performed by one-way ANOVA with Dunnett’s multiple comparison. *,

**, and *** indicate p < 0.05, p < 0.01, and p < 0.001, respectively.




33

Figure 5. DCLK1-B is a critical driver of the initiation and maintenance of colorectal stemness

(A) Limiting dilution assay was performed to compare the tumor-initiating potential of CSCs in

control cells and in DCLK1-B-knockdown cells. HCT116 cells (left panel) or HT29 cells (right panel)

were transfected with non-targeting siRNA (siCTRL) or DCLK1-B-targeting siRNAs (siDCLK1-B).

Two different siRNA sequences were used. (B) Clonogenic assay was performed to compare the

survival potential in control (siCTRL-transfected) and in DCLK1-B-knockdown HCT116 cells. The

siRNA-transfected cells were seeded in 12-well plates and cultured for 2 weeks in growth media. The

number of surviving colonies were counted per wells. (C) The apoptotic cells were visualized by

staining Annexin V+ cells. Cells were transfected with the indicated siRNAs and cultured in serum-

free media for 48 hours. Apoptotic cells were measured by FACS analysis. (D) Cell viability after 48-

hour treatment with 5-FU was measured by MTT assay. IC50 values of 5-FU was evaluated by

GraphPad Prism using a non-linear regression model. (E) Cell survival fraction was determined two-

weeks after the radiation exposure (2 Gy) by clonogenic assay in control or DCLK1-B-knockdown

HCT116 cells. (F and G) Limiting dilution assay was performed as PDX models which were

described above. DCLK1-B knockout or control cells were subcutaneously inoculated with Matrigel

into the inguinal folds of the NSG mice at four doses (1×103, 1×104, 1×105, and 1×106, 8-10

replicates/group). Tumor formation was observed for 3 weeks following inoculation. (G) Primary

tumor volumes were presented as a box and whisker plot. Boxes represent the 25th to 75th percentiles,

and horizontal lines within the boxes represent the median (control cells: n=8, DCLK1B-KO cells:

n=10). (H and I) DCLK1-B-knockout HCT116 or control cells were inoculated into mouse tail vein

and the number of metastatic nodules were counted (n=6/group). (H) Representative image of Indian

ink-stained lungs were provided and the number of metastatic nodules were counted. (I)

Representative image of H&E-stained lungs shows metastatic colonies on lungs. (J and K)

Niclosamide-induced cancer stemness was restored by DCLK1-B overexpression. Control (empty

vector-transfected) or DCLK1-B-overexpressing HCT116 cells were treated by 0.4 μM niclosamide

for 48 hours, and then (J) CD44v6+ CSC populations were measured by FACS analysis and (K) their




34

self-renewal activities were determined by analyzing the tumor sphere-forming efficiency. Bar graphs

represent the mean ± SD. Statistical analyses were performed by one-way ANOVA with Dunnett’s

multiple comparison. *, **, and *** indicate p < 0.05, p < 0.01, and p < 0.001, respectively.




35

Figure 6. Niclosamide exerts a potent in vivo antitumor effect in both HCT116 xenografts and

AOM/DSS-induced spontaneous CRC models

(A-C) Therapeutic effect of niclosamide was evaluated in CRC xenograft models. HCT116 cells were

subcutaneously inoculated into NPG mice. When the volume of primary tumors reached

approximately 100 mm3, tumor-bearing mice were treated with niclosamide (10 mg/kg or 40 mg/kg,

daily, intraperitoneal injection) or vehicle (PBS) (n = 6/group). (A) Primary tumor volume was

measured twice per week until the day of sacrifice. (B) On the day of sacrifice, all primary tumors

were isolated, and (C) the primary tumor weights were evaluated. (D-G) Therapeutic effect of

niclosamide was examined in an AOM/DSS-induced CRC mouse model. (D) C57BL/6 mice were

intraperitoneally injected with AOM (10 mg/kg). One week later, the mice were given 2.0% DSS in

the drinking water for a week. Then, the mice were given drinking water without DSS for 2 weeks.

This cycle was repeated three times. After injection with AOM, the mice were intraperitoneally

treated with niclosamide (20 mg/kg) or vehicle twice a day for 12 weeks (n = 6/group) and sacrificed

on the 13th week. On the day of sacrifice, (E) the number of intestinal polyps and (F) colon length

were measured. (G) DCLK1 expression levels in intestinal polyps were visualized by

immunofluorescence and presented with matched H&E images. Blue indicates nuclei, and red

indicates DCLK1. (H) CRC cells were isolated form vehicle- or niclosamide-treated HCT116

xenografts, and the tumor repopulating potential was evaluated by sphere-forming assay. (I) CD44v6+

CSC populations within primary tumors of HCT116 xenografted mice were visualized by

immunofluorescence assay, and the intensity of CD44v6 expression was quantified using Image-Pro

Plus software. Nuclei were counterstained with DAPI, and a matched region of an H&E-stained

section is presented. Scale bar represents 100 μm. (J-L) Visualization and quantification were

performed as above against various target proteins including (J) OCT4, (K) LEF1, and (L) DCLK1.

Bar graphs represent the mean ± SD. Statistical analyses were performed by one-way ANOVA with

Dunnett’s multiple comparison among more than three groups or Student’s t-test between two groups.

*, **, and *** indicate p < 0.05, p < 0.01, and p < 0.001, respectively.






















Published OnlineFirst November 16, 2018.Clin Cancer Res So-Yeon Park, Ji-Young Kim, Jang-Hyun Choi, et al. Attenuates Colorectal Cancer StemnessInhibition of LEF1-mediated DCLK1 by Niclosamide

Updated version

10.1158/1078-0432.CCR-18-1232doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2018/11/16/1078-0432.CCR-18-1232.DC1

Access the most recent supplemental material at:

Manuscript

Authorbeen edited. Author manuscripts have been peer reviewed and accepted for publication but have not yet

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

Subscriptions

Reprints and

[email protected] at

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

Permissions

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

.http://clincancerres.aacrjournals.org/content/early/2018/11/16/1078-0432.CCR-18-1232To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-18-1232

http://clincancerres.aacrjournals.org/content/suppl/2018/11/16/1078-0432.CCR-18-1232.DC1

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

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/early/2018/11/16/1078-0432.CCR-18-1232


Documents

Inhibition of LEF1-mediated DCLK1 by Niclosamide ......Nov 16, 2018 · Committee (IACUC) of the Gwangju Institute of Science and Technology (GIST-2017-038). All work . related to