저 시-비 리- 경 지 2.0 한민

는 아래 조건 르는 경 에 한하여 게

l 저 물 복제, 포, 전송, 전시, 공연 송할 수 습니다.

다 과 같 조건 라야 합니다:

l 하는, 저 물 나 포 경 , 저 물에 적 된 허락조건 명확하게 나타내어야 합니다.

l 저 터 허가를 면 러한 조건들 적 되지 않습니다.

저 에 른 리는 내 에 하여 향 지 않습니다.

것 허락규약(Legal Code) 해하 쉽게 약한 것 니다.

Disclaimer

저 시. 하는 원저 를 시하여야 합니다.

비 리. 하는 저 물 리 목적 할 수 없습니다.

경 지. 하는 저 물 개 , 형 또는 가공할 수 없습니다.

약학석사학위논문

CD133-mediated TM4SF5

Expression Enhances Growth and

Survival of Liver Cancer Spheres

CD133-의존적 TM4SF5 발현에

따른 간암 Sphere의 증식 및 생존

에 대한 연구

2017년 2월

서울대학교 대학원

약학과 의약생명과학전공

김 소 미

CONTENTS

ABSTRACT∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙1

INTRODUCION…………………………………………………………………∙4

MATERIALS AND METHODS………………………………………………6

RESULTS

1. Correlation between TM4SF5 and CD133 was determined∙∙∙∙∙∙∙10

2. CD133 regulates the expression of TM4SF5∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙14

3. C-terminus of CD133 is important for the regulation of TM4SF5

and Wnt3a/β-catenin transcriptionally induce TM4SF5∙∙∙∙∙∙∙∙∙∙∙∙18

4. CD133-mediated TM4SF5 induction promotes sphere

formation∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙22

5. TSAHC, a specific TM4SF5 inhibitor, abolishes the sphere

formation and growth∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙26

6. Clinical relevance of the CD133/TM4SF5/CD44 linkage in

recurrence free survival∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙31

7. Summary of the present study∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙∙35

DISCUSSION…………………………………∙∙………………………………37

REFERENCES…………………………………………………………………41

국문초록…………………………………………………………………………47

LIST OF FIGURES

Figure 1. Correlation between TM4SF5 and CD133 was determined.

Figure 2. CD133 regulates the expression of TM4SF5

Figure 3. C-terminus of CD133 is important for the regulation of

TM4SF5 and Wnt3a/β-catenin transcriptionally induce TM4SF5

Figure 4. CD133-mediated TM4SF5 induction promotes sphere

formation

Figure 5. TSAHC, a specific TM4SF5 inhibitor, abolishes the sphere

formation and growth

Figure 6. Clinical relevance of the CD133/TM4SF5/CD44 linkage in

recurrence free survival

Figure 7. Summary of the present study

- 1 -

ABSTRACT

CD133-mediated TM4SF5 expression

enhances growth and survival of liver cancer

spheres

Somi Kim

College of Pharmacy

The Graduate School

Seoul National University

Transmembrane 4 L six family member 5 (TM4SF5) is a member of

the tetraspanin L6 superfamily and highly expressed in various

cancers including hepatocarcinoma. Previous study showed that

TM4SF5 physically interacts with CD44, one of the cancer stem cell

(CSC) markers, and this interaction leads to self-renewal stemness

and activates circulating properties via an involvement of Epithelial-

Mesenchymal Transition (EMT) phenotypes. In this study, I

investigated how TM4SF5 and CD44 could be regulated while leading

- 2 -

to their collaboration for cellular stemness-like and circulating

properties. Since CD133 is also well-known to be a cancer stem cell

surface marker and a pentaspan transmembrane protein expressed

by a broad range of cancers including hepatocarcinoma, I studied here

how the expression of TM4SF5 and CD44 are regulated by CD133

related to their stemness-like and circulating properties. In this

study, I found that overexpression of CD133 in liver cancer cells

increases expression of TM4SF5 and CD44. However, neither

overexpression nor knockdown of TM4SF5 affects CD133

expression. As overexpression of CD44 does not change the

expression of CD133 and TM4SF5, our results suggest that CD133

regulates TM4SF5 and CD44 as an upstream molecule of them. Then,

two tyrosine residues (Y828 and Y852) present in the cytoplasmic

tail of CD133 were substituted by a phenylalanine residue individually,

or simultaneously, Y828/852F, or the c-terminal cytoplasmic domain

of CD133 (52 amino acids) was completely deleted. Under both

conditins, c-Src and Akt activities and expression level of TM4SF5

decreased, compared to CD133 wildtype. Topflash reporter assays

showed that the transcriptional activity of β-catenin by CD133

mutants decreased, compared to the wildtype. TM4SF5 reporter

assays also showed that the activated β-catenin promoted TM4SF5

- 3 -

expression subsequently, suggesting that CD133 regulates TM4SF5

expression by c-Src, Akt, and β-catenin signaling cascades. In 3D

aqueous condition, CD133-overexpressing cells formed more and

bigger spheres than cells expressing C-terminal mutants of CD133.

In addition, treatment of TSAHC, a specific TM4SF5 inhibitor,

showed that CD133-overexpressing cells showed more effective

inhibitory response to form spheres upon TSAHC treatment,

suggesting that CD133 regulates TM4SF5 expression required for

sphere formation and growth. Taken together, the expression of

TM4SF5 increase by CD133 expression. Tyrosine phosphorylation

and activation of c-Src, Akt, and β-catenin signaling pathways

triggers liver cancer cells to adopt characteristics of circulating

tumor cells.

Key words: TM4SF5, CD133, β-catenin, Src, Akt, sphere formation

and growth

Student number: 2015-21866

- 4 -

INTRODUCTION

The transmembrane 4 L six family member 5 (TM4SF5) is

tetraspanin and highly expressed in various cancers including

hepatocarcinoma. Its overexpression in liver cancer cells promotes

cell proliferation, migration, and invasion [1-4]. Previous study

showed that TM4SF5 interacts with CD44, one of the cancer stem

cell markers, and promotes self-renewal and circulating capacities

through signaling linkage via STAT3/Twist1/Bmi1 [5]. This study

suggested that TM4SF5-mediated EMT confers “stemness” to HCC,

and that TM4SF5 might serve as a potential therapeutic target in HCC.

According to the previous studies, TM4SF5 is known to be associated

with tumor growth, but its modulating mechanism is not yet fully

understood. Therefore, based on this previous study, we focused

here on the implication of another cancer stem cell marker, CD133.

CD133 (prominin-1), is a transmembrane glycoprotein that typically

localizes at membrane protrusions [6-12]. CD133 was identified as

a novel cell surface marker for liver stem and progenitor cells [6-7].

CD133 is predicted to be pentaspan transmembrane protein and it is

- 5 -

tyrosine-phosphorylated at residues 828 and 852. In addition, Src

tyrosine kinase is responsible for the phosphorylation of CD133 and

subsequent induction of signal transduction cascades, suggesting that

the c-terminal cytoplasmic domain of CD133 plays an important role

in the regulation of its functions [13, 14]. In addition to a marker of

cancer stem cells (CSCs), numerous reports suggest that CD133 is

a marker of poor prognosis in various cancers including

hepatocarcinoma and CD133 contributes to cancer initiation, invasion,

and tumor growth. Also, CD133+ cells display resistance in multiple

drugs and radiotherapy [15-17].

In this study, we investigated what regulates TM4SF5 expression

and found a positive correlation between CD133 and TM4SF5. We

suggest here that CD133 regulates TM4SF5 expression and CD133-

mediated TM4SF5 expression affects growth and survival in 3D

aqueous conditions. Our findings in the present study suggest that

CD133 upregulates TM4SF5 expression by activation of c-Src, Akt,

and β-catenin. Eventually, CD133-mediated TM4SF5 activation is

critical for sphere growth and survival in 3D aqueous environment.

- 6 -

Material & Methods

1. Cell culture

Parental SNU449, SNU761, and Hep3B cells were cultured in RPMI-

1640 (WelGene Inc.) containing 10% FBS and 1%

penicillin/streptomycin (GenDEPOT Inc.) at 37°C in 5% CO2. Huh7

cell was cultured in DMEM (WelGene Inc) under the same condition.

SNU449, SNU761, Huh7, and Hep3B cells were stably transfected

with CD133-eGFP and selected by puromycin (2ug/ml). Cells were

transiently transfected with CD133-eGFP by lipofectamine.

2. Spheroid Formation

Cells were collected and washed by PBS 2 times to remove serum,

then suspended in serum free DMEM/F12 supplemented with 1%

penicillin/streptomycin (GenDEPOT Inc.), 2% B27 supplement

(Invitrogen). 25ng/ml of hEGF and hbFGF (Peprotech) were added

every 2nd days. The cells were subsequently cultured in ultra-low

attachment 6-wellplates (Corning Inc. Corning, NY, USA) at a

density of no more than 5x103 cells/well.

- 7 -

3. RT-PCR

Total RNA was extracted from cells using TRIzol (Invitrogen),

according to the manufacturer’s protocol. Total RNA (500ng) was

reversely transcribed using amfiRivert Platinum cDNA Synthesis

master mix (GenDEPOT). Primers used for PCR were indicated in

the Table 1. cDNA was subject to reverse-transcription polymerase

chain reaction with Dream Taq Green PCR master mix (Thermo

scientific).

TM4SF5 forward CTGCCTCGTCTGCATTGTGG

reverse CAGAAGACACCACTGGTCGCG

CD133 forward GAGTCGGAAACTGGCAGATAG

reverse CACAGAGGGTCATTGAGAGATG

CD44 forward CGGACACCATGGACAAGTTT

reverse GAAAGCCTTGCAGAGGTCAG

β-actin forward TGACGGGGTCACCCACACTGTGCCCATCTA

reverse CTAGAAGCATTTGCGGTGGACGACGGAGGG

4. Western Blots

The cells were grown in 6-well culture plates and harvested at 80%

- 8 -

of confluency, before preparation of whole cell lysates with modified

RIPA buffer (Lee et al.,2008). Primary antibodies for immunoblot

were used as follows: CD133 (Miltenyi), CD44 (Santa Cruz,

Biotechnology), phospho-Y416 Src (Cell Signaling), c-Src (Santa

Cruz, Biotechnology), phosphor-Y473 Akt (Santa Cruz,

Biotechnology), Akt (Santa Cruz, Biotechnology), α-tubulin

(Sigma).

5. Mutagenesis

Mutations of CD133 cDNA (Y828F, Y852F, Y828/852F) were

engineered by pfu polymerase (Stratagene). After mutant strand

synthesis, the template plasmid was removed by DpnI digestion.

Mutations were confirmed by direct sequence analyses.

Y828F forward GAATGGATTCGGAGGACGTGTTCGATGATGTTGAAACTATAC

reverse GTATAGTTTCAACATCATCGAACACGTCCTCCGAATCCATTC

Y852F forward GGTTATCATAAAGATCATGTATTTGGTATTCACAATCCTGTTATG

reverse CATAACAGGATTGTGAATACCAAATACATGATCTTTATGATAACC

C-ter del forward CTCTAATTTTTGCGGTAAAGCTTCGAATTCTGCAG

reverse CTGCAGAATTCGAAGCTTTACCGCAAAAATTAGAG

- 9 -

6. Luciferase Reporter Gene Assay

pTOP-FLASH vector and CD133-eGFP vector were transiently

co-transfected into Hep3B and Huh7 cells. Twenty-four hours

after transfection, luciferase activities of cell extracts were

measured with Luciferase kit (Promega).

7. Flow Cytometry

The cells were labeled with the following antibodies: CD133

(Miltenyi), CD90(Biolegend). Labeled cells were detected using a

FACSCalibur (BD Biosciences). Only second antibody labeled cells

were used as controls. A dead cell removal procedure was

performed before antibody labeling and flow cytometry analysis.

8. Statistical Methods

A-two-tailed unpaired Student ’ s t-test was performed to

determine significance of difference between two groups. A p-value

less than 0.05 was considered statistically significant

- 10 -

Results

1. Correlation between TM4SF5 and CD133 was determined.

As TM4SF5 renders self-renewal capacity to HCC and promotes

sphere formation by interaction with CD44 [5], we examined whether

TM4SF5 has any correlation with other cancer stem cell markers.

CD133 and CD90 are used to define cancer stem cells in multiple

human epithelial cancers including liver and breast cancer, so we

examined the expression of these markers in SNU449 stable cells

with TM4SF5 overexpression [33, 34]. The expression of these

markers was assessed by flow cytometry analysis. The result shows

that the signal of CD133 was not detectable in TM4SF5-expressing

cells (Fig 1A). In addition, the signal of CD90 was also not detectable

in TM4SF5-expressing cells as well (Fig 1B), suggesting that

TM4SF5 does not influence the expression of cancer stem cell

markers such as CD133 and TM4SF5. In opposition to the effect of

TM4SF5 on the expression of cancer stem cell markers, we also

examined whether cancer stem cell markers affect the expression of

TM4SF5. CD133 was sorted into CD133+ and CD133 – populations

and the expression of TM4SF5, and Stat3, Bmi1, and Twist1 was

confirmed. Regarding Stat3, Bmi1 and Twist1, it was reported that

- 11 -

TM4SF5 rendered stemness to HCC through Stat3, Twist1, and Bmi1

signaling pathway [5]. Our results show that the expression of

TM4SF5 increases in the CD133+ population compared to CD133-

population. Also, the expression of pY705Stat3, Bmi1, and Twist1

increases in CD133 +population compared to the CD133- population

(Fig1C). In addition to the CD133 subpopulation, CD133 was

suppressed by shCD133 and the expression patterns of TM4SF5,

Stat3, Bmi1, and Twist1 were consistent with previously observed

CD133 subpopulation patterns.

- 12 -

Primary Ab : mouse anti-human CD133 or CD90 Secondary Ab : PE-anti mouse

P Cp

Tp T3 T7 T10 T16

CD133

Co

un

ts

A

Tp T3 T7 T10 T16

P Cp N138A N155Q NA/NQ

CD90

Co

un

ts

B

C

sh

Sc

ram

sh

CD

13

3

Huh7

TM4SF5

-actin

CD133

CD

133

+

CD

133-

Huh7

STAT3

Bmi1

pY705

STAT3

Twist1

- 13 -

Figure 1. Correlation between TM4SF5 and CD133 was determined.

(A) Flow cytometry histograms showing CD133 expression of

SNU449 parental (P), and SNU449 stable cells without TM4SF5 (Cp),

and with TM4SF5 cell cones (Tp, T3, T7, T10 and T16). Mouse-

anti human CD133 was used for primary antibody, and PE-anti

mouse was used for secondary antibody. (B) Flow cytometry

showing CD90 expression of SNU449 parental (P), SNU449 stable

cells without TM4SF5 (Cp), with TM4SF5 N-glycosylation mutation

(N138A, N155Q, NA/NQ), and with TM4SF5 cell clones (Tp, T3, T7,

T10, T16) (C) Expression of CD133, TM4SF5, STAT3, Twist1 and

Bmi1 from Huh7 CD133+, CD133- and shCD133.

- 14 -

2. CD133 regulates the expression of TM4SF5.

To identify the role of CD133 in liver cancers and its contribution to

TM4SF5 expression, liver cancer cells were screened first. The

positive correlation between cd133 and tm4sf5 expression was

identified (Fig 2A). Huh7 and Hep3B cells present a relatively high

expression level of endogenous CD133 and they also have relatively

high expression of endogenous TM4SF5. In our study, SNU449 and

SNU761 liver cancer cells were used as the cells whose endogenous

level of CD133 is relatively low, and Huh7 and Hep3B as the cells

whose endogenous level of CD133 is relatively high. Next, When

CD133 was overexpressed, the expression level of TM4SF5 and

CD44 increased at both mRNA and protein levels (Fig 2B). However,

neither overexpression nor knockdown of TM4SF5 affected CD133

expression, suggesting that CD133 is a possible upstream regulator

of TM4SF5 and regulates its expression (Fig 2C). Moreover,

overexpression of CD44 did not change the expression level of

CD133 and TM4SF5, which suggest that CD133 differentially

regulates TM4SF5 and CD44 in separate ways (Fig 2D).

- 15 -

A

B

- 16 -

C

D

- 17 -

Figure 2. CD133 regulates the expression of TM4SF5.

(A) mRNA levels of tm4sf5, cd133, and cd44 in different liver cancer

cells. (B) Expression levels of TM4SF5, CD133 and CD44 (mRNA in

upper panel and protein lower panel) in CD133 overexpressing cells.

Negative control is a TM4SF5 control cell and positive control are

TM4SF5 overexpressing cells. (C) Expression of TM4SF5, CD133

and CD44 (mRNA in upper panel and protein in lower panel) in

TM4SF5 overexpressing cells. (D) Expression of TM4SF5, CD133

and CD44 (mRNA in upper panel and protein in lower panel) in CD44

overexpressing cells.

- 18 -

3. C-terminus of CD133 is important for the regulation of TM4SF5

and Wnt3a/β-catenin transcriptionally induce TM4SF5.

According to the previous reports, CD133 signaling enhances

tumorigenic potential through c-Src, PI3K/Akt pathways by tyrosine

phosphorylation within the cytosolic tail [29-32]. Therefore, in the

present study, the C-terminus of CD133 was mutated to identify its

significance for downstream signaling leading to TM4SF5 expression.

Tyrosine residues at 828 and 852 of the c-terminus of CD133 were

substituted by phenylalanine (Y828F, Y852F) and two residues were

substituted by phenylalanine simultaneously (Y828/852F). In

addition to the point-mutated CD133 isoforms, c-terminus of CD133

was deleted (c-ter deletion mutant). The result shows that when it

is point mutated or c-terminus was deleted, CD133 mediated

expression of p473Akt and pY416c-Src decreased (Fig 3A).

Interestingly, the expression of TM4SF5 decreased with

overexpression of both mutants of CD133, compared to CD133 WT,

indicating that the c-terminus of CD133 is critical for the regulation

of TM4SF5 expression.

It was also reported that CD133 increased β-catenin activation by

inhibition of GSK3β [27]. Therefore, we used a Topflash luciferase

- 19 -

construct to confirm the transcriptional activation of β-catenin by

CD133. This luciferase reporter ene is driven by a β-catenin/TCF

complex responsive element. Our results show that CD133 induces

the transactivation potential of β-catenin. By inhibition of PI3K

(LY294002) and GSK3β (LiCl), its activity decreases dose-

dependently (Fig 3B). Also, when β-catenin activity was stimulated

by wnt3a conditioned media, TM4SF5 was transcriptionally induced

which suggests that CD133 regulates TM4SF5 by β-catenin

activation (Fig 3C).

- 20 -

(-) (+) (-) (+) (-) (+)0.0

0.5

1.0

1.5

2.0

2.5 *

12hr 24hr

Wnt3a-CM

Huh7

48hr

*

TM4SF5 Luciferase assay

Rela

tive L

um

inescen

ce U

nit

s

0

2

4

6

8

10***

***

***

*

Huh7 TopFlash Luciferase Assay

1

ns

Topflash

CD133

LY294002

LiCl

- + + + + + + +

- - + + + + + +

- - -- - 10uM - -- - - - - - 2mM

50uM

10mM

Rela

tive L

um

inescen

ce U

nit

s

A

B C

- 21 -

Figure 3. C-terminus of CD133 is important for the regulation of

TM4SF5 and Wnt3a/β-catenin transcriptionally induce TM4SF5.

(A) Immunoblots with the C-terminus mutants of CD133 for pY416c-

Src, Akt and TM4SF5. (B) Transactivation by β-catenin/TCF

complex. The reporter gene bearing the TCF-driving promoter (TOP)

was transfected to CD133 overexpressing cells. Luciferase activity

was measured 48hr after transfection. Data are the mean ±s.d. of

triplicate. Asterisk indicates the significance of the difference

between control and CD133 WT cells treated with LY294002 (PI3K

inhibitor) and LiCl (GSK3β inhibitor). *, ** or *** depicts statistically

significant difference at p<0.05, 0.005, or 0.001. (C) Transactivation

of TM4SF5 by Wnt3a conditioned medium.

- 22 -

4. CD133-mediated TM4SF5 induction promotes sphere formation.

Next, we investigated whether CD133-mediated TM4SF5 is

associated with sphere forming ability in 3D aqueous condition.

CD133 overexpressing cells were plated in low adhesion culture

dishes, leading sphere formation after 7 days. The diameter for each

spheres was analyzed by a software. Our results showed that

CD133+ cells formed more and bigger spheres compared to CD133-

cells (Fig4A). Moreover, CD133-overexpressing cells form more

and bigger spheres (Fig4B and Fig4C). These spheres were collected

and extracted proteins were immunoblotted to confirm the

downstream signaling events. Spheres derived from CD133-

overexpressing cells induced TM4SF5 expression and as well c-Src

and Akt activaton (Fig4D). In addition, with C-terminus mutants of

CD133 (CD133 Y828F and Y852F), reduced sphere formation and

growth were observed compared to the spheres derived from CD133

WT cells, suggesting the importance of the CD133 c-terminus in

sphere formation and growth (Fig4E).

- 23 -

0.0

0.2

0.4

0.6

0.8** ***

control

(n=93)

control

(n=160)

CD133

(n=113)

CD133

(n=173)

Huh7 Hep3B

Sp

hero

id D

iam

ete

r (u

m)

B C

Huh7 CD133+

Huh7 CD133-

A

Huh7

0

70

40 30

20

10

50

60

No

. o

f s

ph

ere

s

- 24 -

D

E

- 25 -

Figure 4. CD133-mediated TM4SF5 induction promotes sphere

formation.

Sphere growth. Cells (5X103 cells in each well) were cultured for 7

days with serum-free conditioned medium and image of spheres

were obtained. Representative images are shown. (A) Sphere

formation assays show that CD133+ cells form more spheres

compared to CD133- cells. (B) Sphere formation assays showing

that CD133-overexpressing cells form more and bigger spheroids.

(C) The diameter of the spheres was measured by Image J software

and asterisks indicate the significance of the difference between the

control and CD133 WT expressing cells (p<0.05, ANOVA). (D)

CD133 overexpressing spheres induce TM4SF5 expression and its

downstream signals such as c-Src and Akt. (E) Sphere growth in the

control cells, CD133 WT, and the mutants of CD133 tyrosine

phosphorylation (CD133 Y828F and Y852F)

- 26 -

5. TSAHC, a specific TM4SF5 inhibitor, abolishes the sphere formation

and growth

To identify the importance of TM4SF5 in sphere formation, TM4SF5

was overexpressed in CD133- cells and a sphere formation assay

was conducted. The result shows that overexpression of TM4SF5 in

CD133- cells leads to formation of more and bigger spheres

compared to the CD133- control population (Fig 5A). These results

suggests that TM4SF5 contributes sphere forming ability of CD133-

cells and plays a critical role in sphere formation and growth. Next,

we examined the importance of TM4SF5 in sphere formation using a

chemical inhibitor of TM4SF5, TSAHC. An anti-TM4SF5 small

synthetic compound [TSAHC; 4’-(p-toluenesulfonylamido)-4-

hydroxychalcone] was treated dose dependently to CD133

overexpressed spheres in Huh7 and Hep3B cells. Previous results

demonstrated that TSAHC inhibits TM4SF5 by interruption of N-

glycosylation or structural integrity of the EC2 domain [20]. The

result was that CD133 overexpressing cells showed a more effective

response against TSAHC (1uM or 3uM) by reducing the sizes of the

spheres (Fig 5B and Fig 5C). Although the IC50 of TSAHC against

TM4SF5 was evaluated to be approximately 5uM in 2D cell cultures

conditions, even a small concentration of TSAHC (1uM or 3uM) could

- 27 -

reduce the sizes of the spheres in 3D aqueous spheroid culture

systems. Next, we further examined the TSAHC for the blockade of

CD133-mediated TM4SF5 effects by treating TSAHC analog control

compounds (PKH208, PKH502). These analog control compounds

lack of p-toluensulfonylamido groups [28], so the treatment of

control compounds results in loss of specificity [20]. In the present

study, when these compounds were treated, they did not reduce the

sizes of spheres, suggesting that inhibition of TM4SF5 is critical for

sphere formation and growth and might provide a clue to develop a

novel strategy for liver cancer therapy (Fig 5D).

- 28 -

B

C

B

A

Co

ntr

ol

TM

4S

F5

Hu

h7 C

D1

33

- Day 4 Day 10 Day 7

- 29 -

DMSO PKH-208 (1uM) PKH-502 (1uM)

Hep3B

ctrl

Hep3B

CD133

D

- 30 -

Figure 5. TSAHC, a specific TM4SF5 inhibitor, abolishes sphere

formation and growth

Sphere formation assay. Cells (5X103cells in each well) were

cultured for 7 days with serum-free conditioned medium and images

of spheres were obtained. Representative images are shown. (A)

Sphere formation assay showing that TM4SF5 plays an important

role in sphere formation of Huh7 CD133- cells. (B) Sphere formation

assay showing that TSAHC inhibits sphere formation of CD133-

overexpressing Huh7 cells. TSAHC was treated at Day0 (1uM, and

3uM respectively in DMSO). (C) Sphere formation assay showing

that TSAHC inhibits sphere formation of CD133-overexpressing

Hep3B cells. TSAHC was treated at Day 0 (1uM, and 3uM

respectively in DMSO). (D) Sphere formation assay showing that

TSAHC analog control compounds do not affect sphere formation and

growth (PKH-208 and PKH-502 respectively).

- 31 -

6. Clinical relevance of the CD133/TM4SF5/CD44 linkage in

recurrence free survival

We next examined the possible clinical significance and relevance of

the CD133/TM4SF5/CD44 axis in recurrence free survival of liver

cancer patients. It was assessed by using a clinical database,

SurvExpress. Kaplan-Meier analysis using TCGA liver cancer

patients revealed that the subjects who highly express TM4SF5 are

more likely to experience recurrence event compared to the lower

risk groups with lower expression levels(Figure 6A). However, the

subjects who highly express CD133 or CD44 have comparable

frequencies to experience recurrence events, compared to low risk

groups (data not shown, p=0.382, P=0.7412, respectively). Also, the

subjects who highly express CD133 and TM4SF5 are more likely to

experience recurrence event compared to the lower risk groups

(Figure 6B). Same pattern was observed in the subjects who highly

express TM4SF5 and CD44 (Figure 6D). However, there was no

significant difference in the possibilities of recurrence events

between the high and low risk groups who express CD133 and CD44

(Figure 6C). Interestingly, the subjects who highly express

CD133/TM4SF5/CD44 are more likely to experience recurrence

events compared to the low risk groups, supporting the previous

- 32 -

results that CD133-mediated TM4SF5 expression and possibly

CD44 is critical in HCC progression.

- 33 -

Recurrence Free survival

TM4SF5

P=0.006673

Recurrence Free survival

CD133/TM4SF5

P=0.03937

Recurrence Free survival

CD133/CD44

P=0.4273

Recurrence Free survival

TM4SF5/CD44

P=0.006673

Recurrence Free survival

CD133/TM4SF5/CD44

P=0.01787

A B

C D

E

- 34 -

Figure 6. Clinical relevance of CD133/TM4SF5/CD44 linkage in

recurrence free survival

Recurrence free survival of liver cancer patients. Kaplan-Meier

survival analysis was performed using SurvExpress database.

Graphs represent the Kaplan-Meier curve for risk groups,

concordance index, and p-value of the log-rank testing equality of

recurrence free survival. Red and Green curves denote High- and

Low- risk groups respectively. (A) Recurrence free survival (RFS)

probabilities of the patients with high and low expressions of

TM4SF5. (B) RFS probabilities of the patients with high and low

expression of CD133 and TM4SF5. (C) RFS probabilities of the

patients with high and low expression of CD133 and CD44. (D) RFS

of the patients with high and low expression of TM4SF5 and CD44.

(E) RFS of the patients with high and low expression of CD133,

TM4F5, and CD44. Database is from TCGA-Liver cancer (Original,

Quantile normalized).

- 35 -

7. Summary of the present study

Our present results have shown that c-terminus of CD133 is tyrosine

phosphorylated at residue 828 and 852 by c-Src. Upon

phosphorylation, pro-oncogenic pathways promote tumor growth

such as activation of PI3K and Akt signaling. Activated Akt inhibits

GSK3β and promotes β-catenin release from its complex and

translocation into the nucleus[27]. Finally, β-catenin transactivates

TM4SF5. Therefore, it is likely that CD133 plays a pivotal role in the

regulation of TM4SF5 expression through c-Src/Akt and β-catenin

activation.

It was also previously reported that TM4SF5 modulates c-Src

activity during TM4SF5-mediated invasion through a TM4SF5/c-

Src/EGFR signaling pathway [26]. Therefore, results suggests that

CD133 and TM4SF5 are closely related to each other and c-Src can

be an important mediator to regulate CD133-mediated TM4SF5

expression. Thus, our present findings suggest that CD133-

mediated TM4SF5 expression might be a therapeutic target for

hepatocarcinoma.

- 36 -

Figure 7. Summary of the present study.

CD133 mediates TM4SF5 expression by continual activation of c-

src, Akt and β-catenin. And this induced TM4SF5 expression

signaling enhances sphere growth and survival in 3D aqueous

condition.

- 37 -

Discussion

TM4SF5 is highly expressed in various cancers including

hepatocarcinoma and plays roles in aberrant cell proliferation and

enhanced metastatic potential [21]. Previous study shows that the

interaction of TM4SF5 and CD44 promotes self-renewal and

circulating capacities in liver cancer cells [5]. Despite its tumorigenic

characteristics, we have always encountered the question of what

regulates TM4SF5 expression, because its modulating mechanism

was not fully understood.

In the present study, we describe a positive correlation between

CD133 and TM4SF5. Furthermore CD133 mediates TM4SF5

expression by a c-Src, Akt, and β-catenin signaling cascade,

enhancing sphere growth and survival in 3D aqueous condition.

One of the characteristics of cancer stem cells is the capacity to form

spheres in a low adhesion environment [22]. Interestingly, our

previous study showed that the expression of TM4SF5 in

hepatocytes promotes sphere formation in a non-adhesive condition,

and its suppression or mutation in the expression of TM4SF5 at N-

- 38 -

glycosylation residues block the sphere formation and growth [5]. In

the study, it was shown that the interaction of TM4SF5 and CD44 is

observable in TM4SF5-positive spheroids, and the disruption

between two molecules by anti-TM4SF5 reagent (TSAHC, [20])

abolishes sphere formation, suggesting that TM4SF5 is important for

sphere formation and self-renewal capacity.

It was previously reported that CD133 might be an interesting CSC

marker [22-25], and CD133+ cells display tumorigenic activity,

self-renewal capacity and a chemo-resistant phenotype [15-16].

Therefore, it is likely that elucidation of the molecular mechanisms

underlying sphere formation and self-renewal capacity can be

connected to the interaction between CD133 and TM4SF5.

In this study, we demonstrated that CD133 upregulates TM4SF5

expression, but that TM4SF5 does not affect CD133 expression in

the other way, suggesting that CD133 is a possible upstream

regulator of TM4SF5 and regulates its expression. In the same

context, overexpression of CD44 does not change the expression

level of CD133 and tm4sf5, which suggests that cd133 regulates

tm4sf5 and cd44 expression in separate ways.

- 39 -

The functional role of the c-terminus of CD133 in the regulation of

TM4SF5 expression was determined by generating the mutants of

tyrosine phosphorylation at residues 828 and 852 and c-terminus

deletion mutant. It was previously described that phosphorylation of

CD133 at tyrosine-828 and tyrosine-852 by Src kinase is critical

for CD133 signaling [13]. The result suggests that this c-terminus

of CD133 is important for the regulation of TM4SF5 work as a

platform for the recruitment of signaling intermediates leading to

activation of c-Src, Akt, β-catenin. By using a the Topflash

luciferase reporter gene assay, the β-catenin transcriptional

activation by CD133 was confirmed and TM4SF5 was induced

transcriptionally by wnt3a conditioned media, suggesting that CD133

regulates TM4SF5 via β-catenin activation.

In sphere formation assay, it was observed that CD133-mediated

TM4SF5 induction promotes sphere formation and growth, and C-

terminus of CD133 is important for sphere formation. Interestingly,

when the anti-TM4SF5 reagent, a small synthetic compound TSAHC

[20] was used to treat the spheres, CD133-overexpressing spheres

- 40 -

showed a more effective response against TSAHC compared to its

control. However, control analog compounds of TSAHC was treated

on spheres, there was no change in the size of spheres, indicating

that blocking of TM4SF5 is critical for sphere formation and growth

and TM4SF5 could be a potential therapeutic target for

hepatocarcinoma.

From the clinical point of view and to validate our in cellulo data, the

subjects who highly express CD133/TM4SF5/CD44 will be more

likely to experience recurrence event compared to the lower risk

groups. Tese patient data support that CD133-mediated TM4SF5

and CD44 is critical in HCC progression

Taken all together, the current study showed that CD133 regulates

TM4SF5 expression via its tyrosine phosphorylation followed by

subsequent activation of c-Src, Akt, and β-catenin. Consequently,

CD133-mediated TM4SF5 expression signaling enhances sphere

growth and survival in 3D aqueous condition.

- 41 -

References

1. Lee, S.Y., et al., Focal adhesion and actin organization by a cross-

talk of TM4SF5 with integrin alpha2 are regulated by serum

treatment. Exp Cell Res, 2006. 312(16): p.2983-99.

2. Lee, S.A., et al., Tetraspanin TM4SF5 mediates loss of contact

inhibition through epithelial-mesenchymal transition in human

hepatocarcinoma. J Clin Invest, 2008. 118(4): p. 1354-66

3. Lee, S.A., et al., Transmembrane 4 L six family member 5

(TM4SF5) enhances migration and invasion of hepatocytes for

effective metastasis. J Cell Biochem, 2010. 111(1) p. 59-66

4. Jung, O., et al., Tetraspan TM4SF5-dependent direct activation

of FAK and metastatic potential of hepatocarcinoma cells. J cell

Sci, 2012. 125(Pt 24): p. 5960-73

5. Lee D., et al., Interaction of tetranspan(in) TM4SF5 with CD44

promotes self-renewal and circulating capacities of

hepatocellular carcinoma cells. Hepatology, 2015. 61(6): 1978-

1997

- 42 -

6. Lardon J., et al., Stem cell marker prominin-1/AC133 is

expressed in duct cells of the adult human pancreas. Pancreas.

2008; 36:e1-6.

7. Yin, A.H., et al., AC133, a novel marker for human hematopoietic

stem and progenitor cells. Blood 90, 1997. 5002-5012

8. Miraglia, S., et al., A novel five-transmembrane hematopoietic

stem cell antigen: isolation, characterization, and molecular

cloning. Blood 90, 1997. 5013-5021

9. Mizrak D., et al., CD133:molecule of the moment. The Journal of

pathology. 2008; 214:3-9

10. Corbeil D., et al., Rat prominin, like its mouse and human

orthologues, is a pentaspan membrane glycoprotein. Biochemical

and biophysical research communications. 2001; 285:939-944.

11. Neuzil J., et al., Tumour-initiating cells vs. cancer ‘stem’ cells

and CD133: what’s in the name? Biochemical and biophysical

research communications: 2007; 355:855-859

- 43 -

12. Shmelkov SV., et al., AC133/CD133/Prominin-1. The

international journal of biochemistry & cell biology. 2005;

37:715-719

13. Boivin D., et al., The stem cell marker CD133 (Prominin-1) is

phosphorylated on cytoplasmic tyrosine-828 and tyrosine 852 by

Src and Fyn tyrosine kinases. Biochemistry. 2009, 48, 3998-

4007

14. Wei Y., et al., Activation of PI3K/Akt pathway by CD133-p85

interaction promotes tumorigenic capacity of gloma stem cells.

Proceedings of the National Academy of Sciences of the United

States of America. 2013; 110:6829-6834.

15. Singh SK., et al., Cancer stem cells in nervous system tumors.

Oncogene 2004; 23: 7267-7273.

16. Bao S., et al., Glioma stem cells promote radioresistance by

preferential activation of the DNA damage response. Nature 2006;

444: 756-760

17. Vermeulen L., et al., Single-cell cloning of colon caner stem cells

reveals a multi-lineage differentiation capacity. Proceedings of

- 44 -

the National Academy of Sciences of the United States of America.

2008; 105: 13427-13432

18. Lee, T.K., et al., CD24(+) liver tumor-initiating cells drive self-

renewal and tumor initiation through STAT3-mediated NANOG

regulation. Cell Stem Cell, 2011. 9(1):p. 50-63

19. Yang, W., et al., Wnt/beta-catenin signaling contributes to

activation of normal and tumorigenic liver progenitor cells. Cancer

Res, 2008. 68(11): p. 4287-95.

20. Lee, S.A., et al., Blockade of four-transmembrane L6 family

member 5(TM4SF5)-mediated tumorigenicity in hepatocytes by

a synthetic chalcone derivative. Hepatology, 2009. 49(4): p.

1316-25.

21. Lee, S.A., et al., Modulation of signaling between TM4SF5 and

integrins in tumor micoenvironment, Front Biosci-Landmrk,

16(2011) 1752-1758

22. Reya T., et al., Stem cells, cancer and cancer stem cells. Nature

2001; 414:105-111.

23. Ricci-Vitiani L., et al., Identification and expansion of human

colon-cancer-initiating cells. Nature 2007; 445:111-115

- 45 -

24. Miki J,. et al., Identification of putative stem cell markers, CD133

and CSCR4, in hTERT-immortalized primary non-malignant and

malignant tumor-derived human prostate epithelial cell lines and

in prostate cancer specimens. Cancer Res 2007; 67:3153-3161

25. Todaro M., et al., Colon cancer stem cells dictate tumor growth

and resist cell death by production of interleukin-4. Cell Stem

Cell 2007; 1:389-402

26. Jung OS., et al., The COOH-terminus of TM4SF5 in hepatoma

cell lines regulates c-Src to form invasive protrusions via EGFR

Tyr845 phosphorylation, Bba-Mol Cell Res, 1833 (2013) 629-

642

27. Fang D., et al., Phosphorylation of beta-catenin by AKT

promotes beta-catenin transcriptional activity. K Biol Chem 2007;

282:11221-11229

28. Kumar SK, Hager E, Pettit C, Gurulingappa H, Davidson NE, Khan

SR. Design, synthesis, and evaluation of novel boronic-chalcone

derivatives as antitumor agents. J Med Chem 2003;46:2813-

2815.

- 46 -

29. Takenobu H., et al., CD133 suppresses neuroblastoma cell

differentiation via signal pathway modification. Oncogene 2011;

30: 97-105

30. Dubrovska A., et al., The role of PTEN/Akt/PI3K signaling in the

maintenance and viability of prostate cancer stem-like cell

populations. Proc Natl Acad Sci USA 2009; 106: 268-273

31. Hu L., et al., In vivo and in vitro ovarian carcinoma growth

inhibition by a phosphatidylinositol 3-kinase inhibitor

(LY294002). Clin Cancer Res 2000; 6:880-886

32. Uddin S., et al., Role of phosphatidylinositol 3’-kinase/AKT

pathway in diffuse large B-cell lymphoma survival. Blood 2006;

108: 4178-4186.

33. Lee, T.K., et al., CD24+ liver tumor-initiating cells drive self-

renewal and tumor initiation through STAT3-mediated NANOG

regulation. Cell Stem Cell, 2011. 9(1):p.50-63

34. Al-Hajj, M., et al., Prospective identification of tumorigenic

breast cancer cells. Proc Natl Acad Sci USA, 2003.

100(7):0.3938-9.

- 47 -

요 약 (국문초록)

Transmembrane 4 L6 family member 5 (TM4SF5)는 간암을 포함한

다양한 암 종에서 높게 발현되는 막단백질로서, 선행 연구를 통해

TM4SF5 단백질이 암줄기세포 표지자들 중 하나인 CD44 단백질과 결

합하여 EMT (Epithelial-Mesenchymal transition, 상피-중배엽 세포

전이)현상을 통해, 간암세포의 자가복제능력에 중요한 역할을 한다는 것

을 밝혀졌다. 본 연구에는 이 막단백질을 조절하는 인자가 무엇인 지를

알고자 연구하는 과정 중에 TM4SF5 및 CD44가 또 다른 암줄기세포

표지자인 CD133에 의한 연쇄적 신호전달과정을 통해 발현이 조절되고,

3차원적 배양액 환경에서 sphere의 형성 및 증식에 영향을 준다는 것을

확인하였다. CD133은 세포막을 다섯 번 통과하는 단백질 수용체로, 간

암을 포함한 다양한 종류의 암에서 많이 발현하며 암줄기세포 표지자로

많이 알려져 있다. 간암에서 CD133의 높은 발현은 종양의 개시 및 전

이에 영향을 미치고, 항암제의 저항성을 높여 치료를 어렵게 하는 것으

로 확인된 바 있다. 간암 세포에 CD133을 과발현하면 TM4SF5와

CD44 단백질의 발현이 따라서 증가하였지만, TM4SF5를 과발현하도록

한 세포에서는 CD133과 CD44 단백질의 발현이 증가하거나 감소하지

않았다. 또한 TM4SF5 단백질 발현의 저하시키거나, CD44를 과발현하

더라도 간암세포에서 CD133 단백질의 발현에 변화가 없는 것으로 보아,

- 48 -

CD133이 TM4SF5와 CD44의 상위 조절인자의 역할을 하는 것으로 확

인되었다. 이전 연구에 따르면, CD133의 세포질쪽 C-꼬리부분에는 두

개(828 및 852번)의 타이로신에 인산화가 일어나는 데, 세포질에 존재

하는 c-Src가 타이로신 인산화를 일으키고, 차례로 PIP3 합성과 AKT

를 활성화를 유발하는 것으로 알려져 있다. 활성화된 Akt는 GSK3β의

활동을 억제를 유도하여 β-catenin이 핵 안으로 들어가 세포의 생존을

촉진시키는 신호전달들의 전사를 위한 활성을 유도할 수 있는 것으로 알

려진 바 있다. 이 때, Y828F 혹은 Y858F CD133 mutant와 두 타이로신

을 동시에 페닐알라닌으로 치환환 CD133 Y828F/Y852F mutant, 그리

고 세포질 쪽 C-꼬리부분(52개의 아미노산)을 완전 제거한 CD133 c-

terminal deletion mutant를 발현할 경우, wildtype CD133의 경우에 대

비하여 TM4SF5의 발현 정도가 줄어든 것을 확인할 수 있었다. 또한

β-catenin의 전사적 활성도를 topflash reporter assay를 통해 확인한

경우, CD133의 mutant들의 경우에서, 하위 신호인 β-catenin의 전사

적 활성도가 wildtype CD133에 대비하여 감소하는 것을 확인하였다.

한편 이렇게 활성화된 β-catenin이 TM4SF5의 발현을 증가시키는 것

은 TM4SF5의 promoter reporter assay를 통해 확인할 수 있었다. 따

라서, 이러한 결과는 CD133이 c-Src, Akt, β-catenin의 연쇄적 신호

전달 경로를 통해 TM4SF5의 발현을 조절한다는 것을 보여준다. 이러

한 CD133의 과다발현된 간암세포들은 혈청 성분과 부착 환경이 미약한

수용성 3D (aqueous 3D)상황에서 sphere를 더 잘 형성하고 증식이 더

- 49 -

향상되었고, CD133의 mutant들을 발현하는 간암 세포의 경우에는

sphere의 형성과 증식이 훨씬 저하되었다. 한편, CD133이 과 발현된 세

포에 anti-TM4SF5 reagent인 소분자화합물 TSAHC를 처리하였을 경

우, DMSO 처리 대조군에 비해 sphere 형성 및 증식이 억제되는 것을

확인하였다 (IC50 = 0.3 M). 이는 CD133에 의해 TM4SF5의 발현이

조절되며, TM4SF5 단백질의 기능이 sphere를 형성하는데 중요한 역할

을 한다는 것을 제시한다. 이와 같은 결과들을 종합할 때, TM4SF5 단

백질은 CD133의 타이로신인산화 과정을 통해 활성화된 c-Src, Akt, 그

리고 β-catenin를 포함하는 신호전달 체계에 의해 그 발현이 증가되고,

결과적으로 간암세포의 종양형성내지는 circulatory 종양세포로서의 특

성을 더욱 더 가지도록 한다는 것을 시사한다고 판단된다.

주요어 : TM4SF5, CD133, β-catenin, Src, Akt, sphere formation and

growth,

학번 : 2015-21866