Roles of epithelial-mesenchymal transition and niche in tumorigenesis of tumor-initiating cells

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Request accessible transcript

Transcript (if available)

Content
i

Roles of epithelial-mesenchymal transition and niche in tumorigenesis of tumor-initiating
cells
By
Dinesh Babu Uthaya Kumar

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSTIY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
Master of Science
(Molecular Microbiology and Immunology)

AUGUST 2014

Copyright 2014
Dinesh Babu Uthaya Kumar

ii

ACKNOWLEDGEMENTS
This dissertation would not have been possible without the guidance, support, and generous help
of dozens of people within USC and beyond.
First and Foremost, a fervid thanks to my advisor, the experienced and erudite Dr. Kegio Machida
for his tremendous guidance, innumerable support, and sharing his knowledge throughout my stay in his
lab. His dedication, hard work, and ethics lured me and most of his lab members to adore the research in
his lab. I’m most grateful to him for having trust in me and encouraging me at every possible step.
I am most obliged to Ms. Chialin Chen, senior PhD student of Dr.Machida’s lab without whom
my research wouldn’t have been possible. She most gracefully taught, trained, and guided me in all the
possible techniques, tricks and tips in order be a definite researcher. I would like to thank all the former
and current Dr. Machida’s lab members for the support, cooperation and friendly environment. Mr. Jian-
Chang Liu (Felix), Dr. Douglas Feldman, Mr. Rajeshwar Nityanandam, Ms. Ambika Ramrakhian, and
Mr. Jordan Pedersen. Also, my sincere thanks to Dr. Hidekazu Tsukamoto’s Lab members Mr. Feng Chi
and Mr. Raul Lazaro for their help and assistance with reagents and animal work.
I would like to thank Ms. Michelle Mac Veigh from the cell and Tissue Imaging core of the USC
Research Center for Liver Diseases, and the USC Norris Comprehensive Cancer Center Cell and Tissue
Imaging core for her immense guidance with all the imaging techniques. My extended appreciation to the
supporting staff of my department especially, Ms. Silvina Campos for her untiring and patient help with
all possible means.
My Sincere gratitude and thanks goes to the Lee Summer Fellowship Committee members Dr.
Hidekazu Tsukamoto, and Dr. Keane Lai. A special thanks to Dr. Keane Lai for his continuous support,
invaluable advice, guidance and motivation.

iii

Finally I’ll end with my heartfelt gratitude with where I started: my parents, my sister, and my
friends in India and in Los Angeles who have always cared and stood by me.

iv

TABLE OF CONTENTS
ACKNOWDEGMENTS ii
LIST OF TABLES vi
LIST OF FIGURES vii
ABSTRACT Study-1 ix
ABSTRACT Study-2 x

Chapter 1: Overview and Introduction
1.1 Hepatocelluar Carcinoma and its risk factors 1
1.2 Tumor-Initiating Cells 1

Chapter 2: TLR4-NANOG pathway activates Twist1 for Liver oncogenesis
initiated by Tumor-Initiating Cells
2.1 Introduction 3
2.2 Materials and Methods 4
2.3 Results
2.3.1 HCFD promotes liver oncogenesis in NS5A Tg mice in TLR4-dependent
manner
16
2.3.2 HCFD and alcohol additively promote the liver tumor incidence 18
2.3.3 Twist1 is identified as one of the most conspicuously upregulated genes 20
2.3.4 TLR4 regulates Twist1 transcription 20
2.3.5 Twist1 knockdown reduces TIC self-renewal, migration, and
tumorigenesis
21
2.3.6 NANOG and STAT3 expression regulate Twist1 23
2.3.7 Mouse and human HCC have accentuated expression of TLR4, p-STAT3,
and TWIST1

26
2.3.8 TWIST1 overexpression promotes tumor formation 30
2.4 Discussion 32

v

Chapter 3: Cell-Cell competition between tumor-initiating cells and Hepatic Stellate induces
fusion with Kupffer cells, leading to metastatic colonization in alcoholic liver cancer.
3.1 Introduction 34
3.2 Materials and Methods 35
3.3 Results
3.3.1 Hepatic progenitor cell’s niche increases the tumorigenicity of TICs 39
3.3.2 Niche induces cellulocytosis of Kupffer cells with TICs 41
3.3.3 Accentuated Expression of CD133 and F4/80 in metastatic organs 43
3.3.4 CD47 and FGFR2 crucial role in the regulating TICs Niche 44
3.3.5 Cell-Cell competition leads to HSCs apoptosis 46
3.4 Discussion 47

BIBLIOGRAPHY 50

vi

LIST OF TABLES
Table 2.1 : In vitro Mutagenesis primer sets 7-8
Table 2.2 : Antibodies list 9-10
Table 2.3: qRT-PCR primer sets 10-11
Table 2.4 : ChIP-qPCR and Sequential ChIP primer sets 15

vii

LIST OF FIGURES
Figure 2.1 NS5A Tg mice fed HCFD+LPS developed frequent tumor development,
including adenomas and cystic tumors in the livers and pancreatic tumors.
17

Figure 2.2 NS5A Tg mice fed with Ethanol + Western diet developed an additive
effect of tumor incidence to significant increase in body weight and endotoxin level
in plasma.

19

Figure 2.3. Twist1 is required in EMT of TICs, down regulation of the same affects
their cancer-initiating property.
22

Figure 2.3.1 Twist1 is required in EMT of TICs 23

Figure 2.4 NANOG and STAT3 influence TWIST1 promoter activation in Ns5a
TICs.
25

Figure 2.4.1 TWIST1 promoter activation in Huh7 26

Figure 2.5 Induction of TLR4/NANOG/P-STAT3/TWIST1 pathway components
in Mouse HCC.
27

Figure 2.5.1 Induction of TLR4/NANOG/P-STAT3/TWIST1 pathway components
in HCFD NS5A Tg mouse.
28

Figure 2.5.2 Accentuated protein expression in Human HCC+Obesity Samples 29

Figure 2.6 Twist1 over expression drives tumor growth independent of Tlr4. 31

viii

Figure 2.7 A schematic representation of the hypothetical model 33

Figure 3.1 Niche increases the tumorigenicity of TICs 40

Figure 3.1.1 Niche induces increase metastatic ability in TICs 41

Figure 3.2 Niche induces fusion of Kupffer cells with TICs 42

Figure 3.3 Metastatic TICs has accentuated expression of F4/80 and CD133 43

Figure 3.4 CD47 and FGF2 expression levels 44

Figure 3.4.1 CD47 and FGF2 regulate proliferation and tumor growth 45

Figure 3.5 Activated Stellate cells of the Niche undergoes apoptosis 46

Figure 3.6 The Proposed model, a cell-cell communication between Tumor-
initiating Cells and its Niche.

48

ix

Abstract Study – 1 (Chapter 2)
BACKGROUND & AIMS: Alcohol and obesity cause steatohepatitis through activation of Toll-like
receptor-4 (TLR4) signaling and enhance hepatitis C virus (HCV) associated liver carcinogenesis. The
HCV NS5A protein ectopoically upregulates TLR4 in hepatocytes, generates TLR4/NANOG dependent
liver tumor initiating stem cell-like cells (TICs), and induces liver tumor in alcohol fed transgenic (Tg)
mice. These TICs have robust and selective expression of the leptin receptor, and increased
phosphorylation and activation of STAT3. However, whether the TLR4-NANOG pathway promotes an
oncogenic signaling in other etiological mouse models and patients is ambiguous. METHOD: We sought
to determine whether a Western diet high in cholesterol, and saturated fat (HCFD) that causes obesity can
also synergistically induce liver tumors via TLR4 signaling. RESULTS: We have identified the TLR4-
NANOG oncogenic pathway in the genesis of TICs, and liver tumor in alcohol and/or HCFD fed NS5A
Tg mice. An sensitized response in TICs to LPS and/or Leptin resulted in an enchanced Twist1 Promoter
activity, which was abrogated by silencing: Tlr4, Nanog or Stat3. We provide evidence of NANOG and
STAT3 sharing their binding site on Twist1 promoter to transactive it. CONCLUSION: Taken together,
TLR4-NANOG axis promotes liver tumorigenesis/metastasis through accentuated mesenchymal
phenotype with elevated Twist1 expression upon exposure to HCV and alcohol/high-fat diets. The TLR4
pathway serves as a novel therapeutic target for HCC.

x

Abstract – Study 2 (Chapter 3)
BACKGROUND & AIMS: Evidence that TLR4/NANOG dependent liver tumor initiating stem cell-like
cells (TICs) induces liver tumor in Western diet high in cholesterol and saturated fat (HCFD)/alcohol fed
transgenic (Tg) mice (Chen, Tsukamoto et al. 2013, Chen, Lai et al. 2013). These TICs have robust
characteristics of distal organ metastasis and tumor development (in similar or dissimilar germ line
lineage). However, whether this characteristics of TICs is due to their microenvironment (dissimilar
lineage) crosstalk in mouse models and patients is obscure. METHOD: We sought to determine whether
the Hepatic Progenitor cell’s Niche acts as an intricate factor in regulating the homeostasis of TICs and
can also synergistically induce their metastatic characteristics. RESULTS: We have identified the
Hepatic Stellate cells (HSCs), and the Kupffer cells (KCs) are involved in the genesis of extremely
metastatic TICs via fusion (cytogenetic analysis) between KCs and TICs. A sensitized Fibroblast growth
factor 2 (FGF) emanated from the TIC’s Niche acts as a cue for a Cell-Cell competition between TICs
and HSCs leading to HSCs apoptosis. Enhanced expression of CD47 (don’t eat me signal) ligand in TICs
ensures cellocytosis with KCs via its receptor SIRP-α rather undergoing phagocytosis. Abrogation of
CD47 or Fibroblast growth factor receptor 2 (FGFR2) in TICs prevented incidence of metastasis and
tumor growth. CONCLUSION: Here we provide evidence that the TIC’s Niche serves as an intrinsic
factor for its phenotypic trait - an extremely metastatic cell which can develop tumor in diverse germline
lineage. The CD47 & FGFR2 pathway serves as a novel therapeutic target for HCC.

1

Chapter 1
Overview and Introduction
1.1 Hepatocellular Carcinoma and its risk factors
Hepatocellular carcinoma (HCC) is the sixth most common primary neoplasm and third largest cause of
cancer-related deaths in the world with more than 700 000 cases are diagnosed yearly
1
. Hepatitis C virus
(HCV) and hepatitis B virus (HBV) are the main factors causing HCC, concepts which have been well
studied
2
. However, environmental factors, such as alcohol and high fat diet (diabetes), also facilitate liver
tumor incidence
3
. Some reports and our studies reveal that the synergism of alcohol, diabetes and viral
hepatitis has a 48-fold increase of HCC incidence in experimental mice compared to that of healthy mice
4-
6
. There are a limited number of studies in HCC using genomewide surveys. Out of 125 surgically excised
HCC in French patients for copy number analysis and performed whole-genome sequencing in 24 HCC in
which the largest subset had alcoholic cirrhosis
3, 7
whereas, whole-genome sequencing in 27 HCC from
Japanese patients, which were largely due to HBV and HCV
8
. However, there is still not an effective
method to treat this disease.

1.2 Tumor-Initiating Cells
Despite recent discoveries that have led to the development of new therapies for cancer, treatment
options are still limited for many types - particularly a basal subtype breast cancer, HCC where prognosis
remains poor. In addition, the ability to manage tumor recurrence and metastasis following successful
initial induction of remission continues to be a challenge to treat the disease. The experimental
demonstration of tumor-Initiating cells (popularly known as cancer stem cells) in several human tumors in
recent years supports tumor hierarchy as a fundamental concept in tumor biology and promises a new
cellular target for anticancer drug discovery
9-12
. In leukemia-initiating cells the clonal progression to
cancer could operate through the stem cell compartment
13
. Although the paths for developing agents that
2

target tumor-initiating cells, still a conundrum. The cancer stem cell hypothesis provides an important
framework for drug discovery and cancer treatment, with the potential to find novel antitumor activities.

3

Chapter 2
TLR4-NANOG pathway activates Twist1 for Liver oncogenesis
Initiated by Tumor-Initiating Cells
2.1 Introduction
The ample epidemiological evidence supports an underlying pathophysiological connection
among HCV, diabetes, and hepatocarcinogenesis
4, 14-17
. Diabetes and viral hepatitis synergistically
increase the risk for HCC
16
, and the HCC odds ratio increases from 8.6-fold to 47.8-fold as a result of
concomitant diabetes in HBV or HCV infected patients
16
. Likewise, a high-fat diet and obesity
superimposed by HCV infection leads to an increased incidence of overt diabetes
18
, potentially
establishing a self-reinforcing oncogenic cycle. Obesity and diabetes resulting from a high-fat diet are
associated with elevated levels of hepatic portal and/or systemic gut-derived bacterial endotoxin that
stimulates expression of proinflammatory cytokines and inflammation in the liver and fat tissues leading
to liver injury
19, 20
.

Hepatitis C virus (HCV) affects more than 170 million people worldwide, making it a leading
international health burden
2, 21
. HCV is a major cause of HCC, the sixth most common cancer in the world
and the third leading cause of cancer mortality. HCC has a low five-year survival rate due to the lack of
therapeutic options. Understanding the molecular mechanisms of HCV-induced hepatocarcinogenesis is
required for the eventual development of improved therapeutic modalities for this disease
22
. HCV
contains a 9.5-kb single-stranded positive-sense RNA genome which encodes a polyprotein that is
processed into multiple proteins by cellular and viral proteases. The HCV NS5A protein, a major target of
therapeutic efforts, may interact with an interferon-induced, double-stranded RNA-activated protein
kinase PKR
23
, thus accounting for the resistance of most HCV strains to interferon treatment. NS5A has
also been shown to have a cryptic trans-acting activity for some cellular gene promoters
24
.
4

We recently demonstrated that HCV infection and associated expression of the NS5A protein
leads to excessive TNFα production, fulminant hepatitis and a six fold increase in mortality in response to
LPS. These effects are mediated through increased expression of the innate immune receptor TLR4, and
an activation of associated downstream signaling
25
that is required for the increased inflammation
observed in HCV-infected livers. Increased TLR4 signaling in NS5A-positive hepatocytes following
chronic and excessive alcohol consumption have had been found to underlie the generation and expansion
of highly malignant, CD133+/NANOG+ liver TICs in alcohol-associated hepatocarcinogenesis.
Nevertheless, the significance of TLR4 in hepatocarcinogenesis associated with the concurrence of
obesity and HCV infection has not been directly addressed.

The long-term consumption of a high fat/cholesterol diet (HFCD) leads to elevated levels of
serum fatty acids and to overgrowth of gut bacterial flora, resulting in increased inflammation, oxidant
stress and aggravated liver disease
26
. Saturated fatty acids may play a central role in this process through
TLR4-dependent induction of the pro-inflammatory mediator cyclooxygenase-2
27
. In our recent finding,
we emphasized the key role of Nanog in LPS-TLR4 axis
12
, and P-STAT3 via Leptin-Leptin receptor (OB-
R) signaling
28
as a consequence of alcohol/HCFD on mice models. In light of these findings, we reasoned
that TLR4, NANOG, and P-STAT3 may also mediate synergism between HCV and diabetes in the
pathogenesis of HCC.

2.2 MATERIALS AND METHODS
Human subjects
For immunostaining and immunoblotting analysis of TWIST1, TLR4, NANOG, and P-STAT3 in human
HCC, necropsy or surgically excised HCC tissues from 8 patients with or without HCV infection, with or
5

without a history of alcoholism, with or without Obesity/Diabetes/ BMI >30 were obtained as cryo-
preserved samples and paraffin embedded tissue sections according to the approved University
Institutional Review Board (IRB) protocol. Many of the specimens were obtained from Liver Tissue Cell
Distribution System at University of Minnesota. Samples were obtained from both genders between the
ages of 42-80. Histologically, they all had varying degrees of steatosis (microvesicular and
macrovesicular) and inflammation in addition to different stages of HCC. Completely normal liver tissues
from 2 patients with accidental death or stroke, but without an apparent liver pathology, were also
obtained for immunostaining or immunoblotting.

Mice
In the animal studies, mice expressing the HCV core gene genotype 1b under control of the human
elongation factor (EF) 1a promoter were generated and bred at the USC transgenic mouse facility (8-13
and 8-20 lines). The primary mouse fibroblast cultures were prepared from both core transgenic mouse
and littermate embryos by trypsinizing the embryonic tissue and plating the dissociated cells. Littermates
on a mixed C57BL core transgenic and Tlr4-/- mice (Jackson Lab) were intercrossed at least six time.
Lieber-DeCarli diet containing 3.5% ethanol or isocaloric dextrin (Bioserv, Frenchtown, NJ) was fed to
all the mice in the alcohol feeding arm of the experiment. High-cholesterol high-fat diet was modified
from TD.03350 (Harkan Teklad, Inc.) as previously described
29, 30
.
Cells
Hepatocytes were isolated using the two step collagenase perfusion method
31
. The viability of the
hepatocytes exceeded 70% in all experiments as determined by Trypan-blue exclusion. Cells were plated
at 3  10
5
or 2  10
6
hepatocytes per 6-well or 10 cm Petri dish, respectively.

6

Isolation of mouse TICs using MACS
Tumor-initiating cells (TICs) were isolated from the liver tumor acquired from our novel animal model,
HCV-NS5A transgenic mice with high cholesterol, high fat diet (HCFD) /Alcohol long-term (12 months)
feeding. Tumor samples of liver tumors from mice were collected directly after surgical removal and were
mechanically dissociated, digested in collagen IV (BD Biosciences) and dispase (sigma) mixture, and
incubated at 37 ℃ for 2 hours. Single cell suspensions were incubated with CD133 microbeads (Miltenyi)
and separated using an auto-MACS device (Miltenyi), according to manufacturer's protocol. Isolated
TICs were maintained in Dulbecco’s modified Eagle’s medium nutrient mixture F-12 (DMEM/F12)
containing 10% fetal bovine serum (FBS), 1% Nucleosides, 1 µM dexamethasone, epidermal growth
factor (EGF), 1 µg/ml penicillin, 1 µg/ml streptomycin and 1% nonessential amino acid (NEAA).
CD133+ TICs and CD133- control cells were cryo-preserved in 60% FBS, 20% DMEM/F12, and 20%
DMSO.

Plasmids, lentivirus and retrovirus vectors, and production of lenti & retro viruses
The Ns5a expression plasmid was constructed by inserting HCV Ns5a cDNA downstream of the CMV
promoter into pcDNA3.1 (Invitrogen). All retroviruses were based on lentivirus (pPAX2: Addgene) or
MMTV vectors (pVPack-GP: Stratagene). Lentivirus vectors were prepared by standard procedures in
HEK293T cells. Three plasmids, packaging vector pPAX2 (Addgene), ecotropic envelope gene (VSV
glycoprotein) expression vector pMDV (Addgene), and shRNA expression cassette, were transfected in
HEK293T cells using BioT transfection reagent (Bioland Scientific LLC). Retroviruses expressing Stat3C
and Stat3D were produced in Phoenix cells/ HEK293T
32
. After 48 hr transfection, the virus supernatants
were harvested, purified, mixed with polybrene (4µg/ml), and used to infect Huh7 cells / TICs. The virus
supernatant was tested using LentiX-gostick (Clontech) for their titer. Human GIPZ lentiviral shRNAmir
target gene set was used for Homo sapiens toll-like receptor 4 (TLR4) (RHS4531-NR_024169,
7

RHS4430-98525129, RHS4430-98843572, and RHS4430-99137800) (Open Biosystems). To increase
silencing effects and to reduce off-target effects, a combination of shRNA lentiviruses were used to knock
down target genes. The TWIST1 luciferase constructs were obtained from Nakamura’s Lab
33

Site-directed mutagenesis
Site-directed mutagenesis was performed according to the PCR-based mutagenesis kit (Quik Change site-
directed mutagenesis kit, Stratagene, USA) with Advantage polymerase (Clontech). Sequences containing
a AATGG motif are bound by NANOG and TTCCTATAA motif are bound by stat3 in vitro
34, 35
. The
fragment -209/+1 of the Twist1 gene, containing putative NANOG binding sites (5’-TAAT (G/T) (G/T)-
3’ or 5'-[CG][GA][CG]C[GC]ATTAN[GC]-3') and Stat3 binding sites (5-. TTC(C/T) N (A/G) GAA-3),
was inserted into a pGL3 basic vector (Twist-1-209luc). The Nanog consensus sequence AATGG and
Stat3 consensus sequence TTCCTATAA was mutated utilizing a mutagenic forward primer and a reverse
primer as previously described
35
. The mutated sequences were confirmed by DNA sequencing. Primers
used are listed in Table 2.1.

Gene Forward primer (5’-3’) Reverse primer (5’-3’)
Nanog- mut1
Proximal
GTT TGG GAG GAC GAA GGA
GAC CCC GAG GAA GG

CCT TCC TCG GGG TCT CCT
TCG TCC TCC CAA AC
Nanog- mut 2
Distal
AGG TCG TTT TTG CCT GGT
TTG GGA GGA CG

CGT CCT CCC AAA CCA GGC
AAA AAC GAC CT
Stat3- mut1
Proximal
TTT CCT ATA AAA CAT GAT TAC
GTC CCT CCT CCT CAC G

CGT GAG GAG GAG GGA CGT
AAT CAT GTT TTA TAG GAA A

8

Stat3- mut 2
Distal
CTG GAA AGC GGA AAC TAT
GAT TAC GAA CTT CGA AAA
GTC CC

GGG ACT TTT CGA AGT TCG
TAA TCA TAG TTT CCG CTT
TCC AG

Table 2.1: In vitro Mutagenesis primer sets

Histology & Immunohistochemistry
Five micrometer-sections were stained with hematoxylin and eosin (H&E) or processed for other staining.
Tissue samples were fixed on 10% neutral buffered formalin, embedded in paraffin, sectioned, and
stained with hematoxylin and eosin or immunohistochemistry staining was performed using primary
antibodies against Nanog (Rabbit ab80892, Abcam), p-stat3 (Rabbit #9134, Cell Signaling technology),
TLR4 (Mouse monoclonal antibody, SC293072, Santa Cruz), TLR4 (goat sc-8694,Santa Cruz
Biotechnology), or Twist1 (Rabbit polyclonal antibody, sc-15393, Santa Cruz Biotechnology), based on
the standard protocol with their respective secondary antibodies. Slides were then mounted using xylene
based mounting media including hematoxylin for nuclei counterstaining (Vector Laboratories) according
to the manufacturer’s recommendations. The staining was subjected to morphometric analysis. To
determine the specificity of immunohistochemistry staining, serial sections were similarly processed,
except primary antibodies were omitted in controls. The area of interest were quantified using
MetaMorph software.

Confocal immunofluorescent microscopy
Immunofluroscence staining of cryosections or paraffin sections was performed using primary antibodies
against NANOG (Rabbit ab80892, Abcam), P-Stat3 (Rabbit #9134, Cell Signaling technology), TLR4
(Mouse monoclonal antibody, SC293072, Santa Cruz), TLR4 (goat sc-8694, Santa Cruz Biotechnology),
9

or TWIST1 (Rabbit polyclonal antibody, sc-15393, Santa Cruz Biotechnology) (refer Table 2.2), based on
the standard protocol with their respective secondary antibodies. Slides were mounted using the mounting
media, including DAPI for nuclei counterstaining (Vector Laboratories) according to the manufacturer’s
recommendations. The staining was subjected to morphometric analysis. To determine the specificity of
immunofluorescent staining, serial sections were similarly processed, except primary antibodies were
omitted in controls. Fluorescence images were captured on a Zeiss confocal microscope LSM510, using
sequential acquisition to give separate image files. The degree of staining was categorized by the extent
and intensity of the staining. Image analysis of nuclear translocation was performed using Metamorph or
ImageJ v3.91 software (http://rsb.info.nih.gov/ij). Ten high power fields were selected for analysis of
each stain. To avoid being biased by the TLR4 and Twist1 staining, each field was selected by viewing
nuclear (DAPI) staining only to identify near-confluent cells and thereby maximize the cell numbers
included in the analysis. The sections were then evaluated and photographed under a fluorescence
microscope and expression of TLR4, P-stat3, Nanog, and Twist1 were correlated. Quantitative
fluorescence data were exported from ImageJ generated histograms in Microsoft Excel software for
further analysis and presentation.

Antibody Manufacturer
TWIST1
SC-15393 (Santa Cruz
Bio Technology)

NANOG
ab80892 Abcam

STAT3
#9139S (Cell Signaling)

P-STAT3
#9134S (Cell Signaling)

TLR4
SC-10741 (Santa Cruz Bio Technology)
TAK1
SC-7162 (Santa Cruz Bio Technology)
TRAF6
SC-7221 (Santa Cruz Bio Technology)
10

IKK-B
SC-8014 (Santa Cruz Bio Technology)
P-IKK-B
#2694 (Cell Signaling)
Β-ACTIN
A5441 SIGMA

Table 2.2: Antibodies list

Quantitative Real-Time PCR (RT-PCR)
Total RNA was extracted by using TRIzol Reagent (Invitrogen) and purified using the RNeasy mini kit
(QIAGEN) according to the manufacturer’s protocol. RNA concentration and purity were determined by
A 260 and A 260/A 280 ratios, respectively. The RNA samples were treated with DNase I (Invitrogen) to
remove residual traces of DNA. cDNA was obtained from 1 µg of total RNA, using SuperScript III
reverse transcriptase (Invitrogen) and random primers in a final volume of 10 µl. Quantitative real-time
PCR was performed on an ABI 7300 HT Real-Time PCR machine using 2X SYBR Green Master Mix
(Applied Biosystems). Cycle conditions of all reactions are 1 cycle at 50°C for 2 minutes, followed by 1
cycle at 95°C for 10 minutes, followed by 40 cycles at 95°C for 15 seconds and 60°C for 1 minute.
Specificity of PCR products was tested by dissociation curves. Gene expression was determined relative
to β-actin control via the ΔCt method. cDNAs were amplified by PCR using the primer pairs listed in the
Table 2.3.
Gene Forward primer (5’-3’) Reverse primer (5’-3’)
Twist-1 AGA TGT CAT TGT TTC CAG AGA TTA GTT ATC CAG CTC CAG AGT
Nanog AGG GTC TGC TA TGA GAT GCT CAA CCA CTG GTT TTT CTG CCA
Stat3 GCC ACG TTG GTG TTT CAT AAT C TTC GAA GGT TGT GCT GAT AGA G
11

Tlr4 ATG GCA TGG CTT ACA CCA CC GAG GCC AAT TTT GTC TCC ACA
E-cad CTG CTG CTC CTA CTG TTT CTA C TCT TCT TCT CCA CCT CCT TCT
N-cad CAG GGT GGA CGT CAT TGT AG AGG GTC TCC ACC ACT GAT TC
Gapdh TGG ATT TGG ACG CAT TGG TC TTT GCA CTG GTA CGT GTT GAT

Table 2.3: qRT-PCR primer sets

Gene array analysis of liver tumors
To perform gene array and proteomic analyses for identification of anti-apoptotic or proto-oncogenic
proteins, we prepared serial cytosections of the livers, stained them with H&E, and collected hepatocytes
with normally appearing, dysplastic, or transformed morphology by using laser-capture microscopy as
described
36-38
from livers of mice. To identify changes associated with synergism of alcohol, comparative
analysis was done in the cells isolated from non-Tg mice vs. Tg mice fed alcohol. A gene microarray
analysis requires a minimum of 100-200 cells and proteomic analysis requires approximately 50,000-
100,000 cells for each cell phenotype
39
. The cells were lysed for RNA or protein extraction for gene chip
analysis and 1D gel MS/MS analysis
36, 37, 39-41
. The cells collected from each group of three animals were
isolated for RNA or protein individually and later pooled to achieve a collection of sufficient amounts of
samples. For gene profiling, the Affymetrix mouse gene chip was used and analysis was performed in the
Genome Core Facility at Los Angeles Children’s Hospital.

Endotoxin measurement.
For endotoxin measurements, blood was collected from inferior vena cava with pyrogen-free heparin as
12

previously described
42
. Extreme care was provided to eliminate pyrogen and endotoxin contamination of
all surgical instruments and laboratory supplies. Blood samples were transferred to appropriate glass tubes
made pyrogen-free by heating at 180°C for 24 hours. Pyrogen-free water was supplied by the
manufacturer. Just before assay, plasma samples were diluted and heated to 75°C for 10 minutes to
denature endotoxin-binding proteins that can mask the endotoxin detection. Levels of endotoxin were
measured using the Limulus amebocyte lysate pyrogen test and a kinetic program (Kinetic test, Kinetic-
QCL, Santa Clara, CA; Biowhittaker). The threshold of detection is 0.1 pg/ml.
.
Proliferation Assay
TICs were initially seeded at 5x10
4
cells per well in 6-well plates. Cell number and viability were
measured at day 0, 2, 3, and 4 by the countess automated cell counter (Invitrogen) with trypan blue
exclusion. All experiments were repeated three times.

Wound Healing (Migration) Assay
Cells were seeded in 6-well plate and cultured until fully confluent. The confluent cell monolayer was
slightly and quickly wounded with a linear scratch by a sterile 200 µl pipette tip. The debris was removed
and the edge of the scratch was smoothed with PBS washing. The open gap was inspected and
photographed microscopically (10X object, Nikon) at 1 and 24 hours
43
.

Soft-agar Colony Formation Assay
Cells (2,500) were seeded in 0.35% agarose in TIC growth medium on a layer of 0.5% agar in the TICs
growth medium. Cells were incubated for 14 days at 37 ℃ in a humidified atmosphere at 5% CO 2 in air
13

and 500 µl TIC growth medium were added twice a week. At the end of the incubation period, colonies
were stained with crystal violet (CV) followed by scanning for colony counts. The CV stain was also read
at OD540.

Spheroid Assay
50 of TICs were seeded onto Ultra low attachment 96-well plates (Corning Inc.), followed by incubating
at 37 ℃ in a humidified atmosphere at 5% CO 2 in the air for 14 days and 100 µl TIC growth medium were
added twice a week. The number of colonies were counted under optical microscope and the proliferation
was measured using Luminescent Cell Viability Assay (Promega) followed by manufacturer’s
instructions. All experiments were repeated at least in 24 wells.

Immunoblotting
Total cell lysates were obtained by lysing the cells in cold NP-40 buffer (150 mM NaCl, 1.0% NP-40,
10% Glycerol, and 50mM Tris, pH 8.0) containing complete protease inhibitor mixture (Roche) for 1 h on
ice, followed by centrifugation at 14,000 RPM for 15 min and collection of supernatant. Protein
concentration was determined using the Dc protein assay Kit (Bio-Rad) and the supernatant was mixed
with 6X Laemmli sample buffer. Proteins were separated on 10% SDS-PAGE and transferred to
nitrocellular membranes (Thermo). The membranes were blocked with 5% non-fat milk and 0.1% tween-
20 for 1 h, followed by incubation with the primary antibodies, Twist1 (Santa Cruz Biotechnology), E-
Cadherin (BD Biosciences), N-Cadherin (Santa Cruz Biotechnology), TLR4 (Santa Cruz Biotechnology),
Nanog (abcam), p-stat3 (Cell signaling Technology) and β-actin (sigma) (all in 1:1,000 dilution) at 4 ℃
overnight. Horseradish peroxidase–conjugated IgG (Santa Cruz Biotechnology; 1:2,000) was used to treat
the membranes for 1 hour at room temperature, and enhanced with a SuperSignal® West Pico
14

Chemiluminescent substrate (Thermo). The bands were detected in Premium Clear Blue X-Ray films
(Bioland Scientific LLC).

Reporter Assays
Early passage liver TICs obtained from NS5A transgenic mice (fewer than ten passages in culture) were
cultured in six-well plates and cotransfected using BioT (Bio land Scientific) with 1 μg luciferase reporter
and 50 ng (SV40) Renilla luciferase expression vector to control for transfection efficiency. Forty-eight
hours after transfection, cells were lysed in 1x passive lysis buffer, and luciferase activity was measured
using the Dual-Glo Luciferase System (Promega) using a Lumat LB9501 luminometer (Berthold). At
least three independent biological replicates were used for this experiment. The TWIST1 plasmid
constructs were obtained from Nakamura et al.,
33
. The data shown represents the mean ± SD,

Subcutaneous xenograft transplantation of the TICs into immunodeficient mice
Cells (10,000) in 100 µl solution were mixed with 100 µl Matrigel (BD Biosciences) and injected into the
dorsal flanks of NOG mice. Mice were anesthetized with ketamine (80 mg/kg) and xylazine (10 mg/kg)
through I.P. during the procedure. The tumor volume was measured with a caliper and calculated
according to the formula V=a × b
2
/2, where “V” represents tumor volume, “a” presents the largest, and
“b” the smallest superficial diameter
44
. All the animal experiments were approved by the IACUC
Committee of the University of Southern California.
Tumor collection and analysis
Tumor-bearing animals were sacrificed at day 35, and tumors were collected and measured as the volume
and weight. The tumor tissues were divided for (1) fixation with neutrally buffered 10% formalin for
15

H&E staining and histological evaluation of the tumor; (2) fixation with 4% paraformaldehyde followed
by sucrose treatment for subsequent immune-staining; and (3) snap-freezing for mRNA and protein
analysis of the targeted genes with shRNA.
Live animal imaging
The tumor processing was monitored using noninvasive imaging by whole-body GFP imaging utilizing
the bioluminescence imaging system (IVIS 200 Imaging Series, Xenogen) at day 21 and 35.
Chromatin Immunoprecipitation (ChIP) and Re-ChIP
Anti-Nanog (Abcam) and Anti Stat3 (Cell signaling technology) monoclonal antibody was used to
immunoprecipitate sonicated chromatins prepared from TICs post LPS and Leptin treatment. The ChIP
was performed using preimmune IgG as specificity control. Immunoprecipitated DNA was quantified for
twist1 promoter using qRT-PCR primers which are listed in Table 2.4. The Re-ChIP/ Sequential ChIP
analysis was performed according to the manufacture’s protocol (Active Motif Re-ChIP IT®).
Gene Forward primer (5’-3’) Reverse primer (5’-3’)
Nanog- Proximal
Binding Site
ATG GTT TGG GAG
GAC GAG TTA

AAA GTT TCC GCT TTC CAG
TCC
Stat3 - Distal
Binding Site
GGA CTG GAA AGC
GGA AAC T

GCA GAC TTG GAG GCT CTT
ATA C
Stat3 – Proximal
Binding Site
GCC AGG TCG TTT
TTG AAT GG

CGT GCA GGC GGA AAG TTT
GG
Specificity
control -1
CCC AGC AAT CCC
AAA TCG G

CAG CAA TGG CAA CAG CTT
CTA
Specificity
control -2
CTC ACG TCA GGC CAA TGA
GAG AGC TGC AGA CTT GGA G

Table 2.4: ChIP-qPCR and Sequential ChIP primer sets

16

Statistical Analysis
Experimental data are presented as the mean ±standard deviation (SD). All statistical analysis was
performed using a two-tailed Student’s t test and Chi squared test. Differences were considered
statistically significant when P values were less than 0.05. Error bars reflect standard errors.
2.3 Results
2.3.1 HCFD promotes liver oncogenesis in NS5A Tg mice in TLR4-dependent manner
The long-term consumption of a high-fat diet elevates plasma levels of gut-derived bacterial
endotoxin
45
. Based on our previous finding of increased hepatocyte TLR4 expression in the NS5A-Tg
mice
5
, the co-receptor for endotoxin, we postulated that the synergism between HCV and obesity in liver
disease progression may involve TLR4-dependent signaling. We therefore employed a knockout strategy
to test the role of TLR4 in this interaction between NS5A and obesity. Liver specific NS5A Tg and wild
type (wt) mice with or without TLR4 deficiency (Tlr4-/-) were maintained on normal chow or a HCFD
supplemented with or without LPS for 12 months (Fig. 2.1 A). HCFD feeding resulted in obesity in both
WT and NS5A Tg mice, and remarkably this effect was prevented by TLR4 deficiency in both genotypes
(Fig. 2.1 A and 2.1 F). HCFD feeding caused 39% liver tumor incidence in NS5A Tg mice, but only 6%
in WT mice, and this increment was abrogated in HCFD-fed Tlr4-/-NS5A Tg mice deficient in TLR4
(Fig. 2.1 A and 2.1 C). Conversely, LPS supplementation in the HCFD (100 mg/kg) further increased the
incidence to 47% in NS5ATg mice (Fig. 2.1 A). As expected, HCFD increased plasma endotoxin and
Leptin levels in all genotype groups (Fig. 2.1 B). Activation of TLR4 signaling was assessed by co-IP of
TRAF6-TAK1 and immunoblotting of p-IKK- ; evident of this was only observed in HCFD-fed NS5A
Tg mouse livers (Fig. 2.1 D, 2.1 E) but not the chow. Liver samples collected from chow-fed WT mice
subjected to a single dose of intravenous injection of LPS (2 mg/kg) served as a positive control for these
TLR4 activation parameters (last three lanes of Fig. 2.1 D and 2.1 E). Collectively, these results
17

demonstrate that the Western diet and HCV NS5A synergistically induces liver tumors in a manner
dependent on TLR4 much like the previously demonstrated alcohol effect
5
.

Figure 2.1 NS5A Tg mice fed HCFD+LPS developed frequent tumor development, including
adenomas and cystic tumors in the livers and pancreatic tumors. (A) Summary of Tlr4 disruption in
HCV Ns5a infected mice fed with HCFD or HCFD+LPS or control diet for 12 months reduces tumor
incidence; N, number of experimental mice; *P<0.05, **P<0.01, ***P<0.005 (B) Expression of Endotoxin
and Leptin level in the plasma of HCFD and control diet mice cohorts. (C) Gross images of non-
pathological Liver and pathological liver tumor with multiple nodules from control (1, 2) and HCFD (3-6)
/ HCFD+LPS (7) diet fed mice cohorts, respectively. (D) IP, showing increased levels of Phosphorylated
IKK-β in HCFD Ns5a liver tissue lysate with or without LPS treatment, as compared to the control Wt and
controlNs5a Tg cohorts. Cell Protein extracts were incubated with anti-TAK1 primary antibody, targeted
with agarose beads, washed, and the associated proteins were resolved on a gel. Pulled down, the protein
complex was then immunoblotted to detect the specific association with TAK-1, TRAF6, where IgG was
18

used as Isotype control. Injection of LPS (2 mg/kg) into mouse liver serves as a positive control for the
interaction shown in the last lane. (E) Tlr4 knockout Tg mice cohorts with or without HCV Ns5a Tg fed
with or without HCFD had statistically significant reduction in their body weight, as compared to their
control Wt andNs5a Tg cohorts (F) IKK-β and Phosphorylated IKK-β Protein analysis, showing increased
levels of Phosphorylated IKK-β in HCFD Ns5a liver lysates with or without LPS treatment, as compared
with Control diet or WT mice using immunoblot.

2.3.2 HCFD and alcohol additively promote the liver tumor incidence
We next tested whether a combination of HCFD and alcohol further increases the liver tumor
incidence in NS5A Tg mice. After 12 months of feeding ethanol containing HCFD, 14% of WT mice
developed liver tumors, while 53% of NS5A mice had tumors (Fig. 2.2 A). Feeding of regular high fat
ethanol diet or HCFD resulted in 25% or 36% tumor incidence in NS5A Tg mice, as seen in the preceding
experiment. By combining HCFD and alcohol into the diet of the NS5A Tg mice, we observed maximum
liver tumor incidence. As expected, endotoxemia is evident in these mice with heightened over expression
and activation of TLR4 (Fig. 2.2 B).
19

Figure 2.2 NS5A Tg mice fed with Ethanol + Western diet developed an additive effect of tumor
incidence to significant increase in body weight and endotoxin level in plasma.
(A) Summary of how NS5A Tg mice fed with Ethanol + Western diet developed an additive effect of
tumor incidence with significant increase in body weight and endotoxin level in plasma compared to the
control cohorts, *P<0.05 (B) Upper panel showing, EtOH+WD feeding to NS5a Tg mice, increased the
expression of TLR4 protein in liver whole protein lysate (lane 7&8) compared to the WT (lane 5&6), in
control diet fed cohorts, the NS5a Tg mice had mediocre expression of TLR4 (lane 3&4 ) compared the
WT control (lane 1&2). The expression level was analyzed using immunoblot against α-TLR4 primary
antibody. Lower panel shows, protein complex pull down assay (IP) performed on previously explained
two diverse diets and Tg mice cohorts. The liver protein extracts were incubated with anti-TAK1 primary
antibody, targeted with agarose beads, washed, and the associated proteins were resolved on a gel.
Immunoblot analysis detected the specific association of TAK-1, TRAF6, and NS5a Tg mice fed with
EtOH+WD. (C) Summary of microarray analysis, showing increased Twist1 gene in Ns5a+HCFD, as
compared to their control (D) Twist1 mRNA was analyzed using qRT-PCR in Huh7 cells. Data normalized
to GAPDH are expressed as the fold change compared with their respective controls *P<0.05 (E) TWIST1
mRNA analysis on Huh7 cells with or without TLR4 sh-RNA followed by treatment with or without
LPS(F) TLR4 presence but not silencing augments TWIST1 Luc promoter activity induced by LPS. *P <
0.05.

20

2.3.3 Twist1 is identified as one of the most conspicuously upregulated genes
To understand the molecular mechanisms of enhanced liver oncogenesis in NS5A vs. WT mice,
both fed HCFD, we performed RNA microarray analysis on livers from both groups. This analysis
identified 131 and 43 genes differentially upregulated and downregulated in NS5A mice, respectively.
Some of the highly upregulated genes in different functional categories are listed in Fig. 2.2 C, including
the stem cell marker Nanog and the oncogene Igf2bp3. Both Nanog and lgf2bp3 were found to be critical
in self-renewal and tumorigenic activity of TICs isolated from liver tumors of alcohol-fed NS5A mice
5
.
Twist1, known for its role in epithelial mesenchymal transition (EMT) and tumor metastasis
46-48
, was also
shown to be highly upregulated. Our qPCR analysis confirmed Twist1 induction in HCFD-fed NS5A Tg
mice, as compared to HCFD-fed WT mice or chow-fed (control) NS5A Tg mice (Fig. 2.2 D). Further,
this induction was abrogated in Tlr4-/-NS5A Tg mice, suggesting TLR4 dependency in Twist1 expression
(Fig. 2.2 D).

2.3.4 TLR4 regulates Twist1 transcription
To confirm the role of TLR4 in the regulation of TWIST1, Huh7 cells were transduced with an
empty or NS5A expression plasmid together with a specific sh-RNA for TLR4 or scrambled sh-RNA in
the presence or absence of LPS in the media. The LPS treatment up regulated TWIST1 in NS5A
transduced cells, but not in vector-transduced cells, and this induction was significantly reduced by TLR4
knockdown with the sh-RNA. A similar and more conspicuous blocking effect was observed for TWIST1
induction in LPS treated NS5A transduced cells by expression of a dominant negative TLR4 lacking the
cytoplasmic domain (TLR4 Cyt) vs. WT TLR4 (TLR4+) (Fig. 2.2 E). To test this TLR4-dependent
regulation of TWIST1 at the promoter level, Huh7 cells were transfected with a TWIST1 proximal
promoter (-700/-1) luciferase plasmid with or without identical manipulations for expression of NS5A,
21

and TLR4. Results from this analysis were a mirror image of TWIST1 mRNA data (Fig. 2.2 F), suggesting
that TLR4-dependent regulation of TWIST1 expression is largely transcriptional.

2.3.5 Twist1 knockdown reduces TIC self-renewal, migration, and tumorigenesis
Next, we assessed the functionality of Twist1 in TICs by knocking down its expression with sh-
RNA. Twist1 knockdown did not affect TLR4 or NANOG expression (Fig. 2.3 A), but upregulated
Albumin and E-cadherin expression while downregulating N-cadherin expression (Fig. 2.3 B). It altered
TIC morphology from a spindle (mesenchymal) shape to a tadpole-like (epithelial) shape (Fig. 2.3 C,
insert) and increased the cell size assessed by flow cytometry (Fig. 2.3.1 A). It also reduced cell
proliferation (Fig. 2.3.1 B), colony formation in soft agar (Fig. 2.3 C), and spheroid formation in
methylcellulose culture (Fig. 2.3.1 C), as well as suppressed migration, assessed by a scratch method (Fig.
2.3 D). Finally, we tested the effects of Twist1 knockdown on the tumor-initiating capacity of TICs by
monitoring tumor growth in NOG mice subcutaneously transplanted with TICs over a 35-day period.
Although the volume of the tumor formed was not statistically different with TICs transduced with
scrambled vs. Twist1 shRNA at day 15 post-transplantation (data not shown), the tumor volume and
weight were significantly reduced at day 35 (Fig. 2.3 E-1&2). Optical-image analysis of live tumor-
bearing mice also confirmed that the tumor size had been reduced by Twist1 knockdown (Fig. 2.3 E-
3&4).
22

Figure 2.3. Twist1 is required in EMT of TICs, down regulation of the same affects their cancer-
initiating property. (A) Immunoblot analysis confirms decrease in TWIST1 protein in TICs from HCFD
Ns5a mice upon Sh-RNA treatment but, unaltered expression in hypothesized upstream proteins NANOG
, TLR4 and, the control housekeeping gene β-ACTIN (B) The RNA expression profile of Twist1,
Albumin, N-cad and E-cad which partake in EMT (C) The morphology of sh-Twist1 (2) transduced TICs
takes up a tadpole shape when compared to scrambled TICs (1) which is more of a parenchymal cell
phenotype. In vitro oncogenesity was tested via colony formation, silencing of Twist1 (4) in TICs
significantly reduces their colony formation ability in contrast to their control, n=3 (3) and their data are
summarized as graphical annotation. (D) Twist1 silenced TICs have a diminished ability to migrate in
comparison to their scramble control, shown using In-vitro cell migration assay, n=3. Images were
captured at 0 hours and 24 hours post a fine line of scratch made using a 100ul tip (E) Tumor volume (1)
and weight (2) of subcutaneously injected TICs with Twist1 silencing is reduced, in comparison to the
scrambled control. Tumor volume was measured at day 21 and 35 post-transplantation, the tumor weight
was measured at day 35 when tumor bearing mice were sacrificed. (3) Image of subcutaneous tumors on
mice at day 35 post-transplantation. (4) Subcutaneous tumor growth was monitored by noninvasive
bioluminescence imaging, n=8 NOG mice for each cohort, *P<0.05.

23

Figure 2.3.1 Twist1 is required in EMT of TICs. (A) Flow cytometry analysis indicating the change in
cell type post Twist1 knockdown in TICs (B) Analysis of cell number and viability post infection with
Lentivirus comprising sh-Twist1 or sh-Scramble into TICs; significant decrease in both cell number and
viability was seen with the infection of the former compared to the later. *P < 0.05, n=5. (C) Sphere
formation assay demonstrated the significant decrease in the number of colonies formed with the Twist1
gene is diminished in TICs *P < 0.05, n=3.

2.3.6 NANOG and STAT3 expression regulate Twist1
We next investigated the molecular mechanisms responsible for TLR4-dependent activation of
the Twist1 proximal promoter by mutation analysis using TICs transduced with scrambled or Tlr4
shRNA. This analysis revealed that the region between -51 and -209 upstream of the transcription
initiation site is essential for the basal and TLR4-dependent activity (Fig. 2.4 A). The same was
24

performed in huh7 cells only to see a mirror image of the results obtained from TICs (Fig. 2.4.1). A
deletion of a segment between -102 and -74 also markedly reduced the activity, suggesting this region
contains critical cis-elements. We speculated that NANOG, and STAT3 downstream of LPS-TLR4, and
Leptin-OB-R axis respectively
5, 12, 28
could trans-activate the Twist1 promoter. Through Tess™ and
Transfac™ analysis, we identified potential two NANOG and two STAT3 binding sites in this region. To
test this notion, we performed in vitro mutagenesis on the respective NANOG and STAT3 binding sites
and showed that the STAT3-2 (distal STAT3 site) and NANOG-1 (proximal NANOG site) sites are
critical for the Twist1 promoter activity as mutation of each site markedly attenuated the responsiveness
to either ligand (LPS and Leptin) (Fig. 2.4 B). Also, using TICs infected either with lentivirus expressing:
Nanog Sh-RNA / Tlr4 Sh-RNA or retrovirus expressing the dominantly negative STAT3 (Stat3D) /
constitutively active STAT3 (Stat3c) under the influence of LPS and/or Leptin. We demonstrated that
abrogation of these key cellular signals too will significantly contribute to the attenuated Twist1 promoter
activity. (Fig. 2.4 C). Next we examined the enrichment of NANOG and STAT3 at these sites by ChIP-
qPCR (Fig. 2.4 D). Both NANOG and STAT-3 were enriched at NANOG-1 and STAT3-2 sites,
suggesting these two transcription factors bind to these two sites to cooperatively transactivate the Twist1
proximal promoter in response to LPS or leptin. To validate this speculation Re-CHIP/ Sequential ChIP
analysis was performed and evidently shown to have increased Nanog and Stat3 binding expression (Fig.
2.4 E).
25

Figure 2.4 NANOG and STAT3 influence TWIST1 promoter activation in Ns5a TICs. (A) TWIST1
promoter analysis with deletion constructs demonstrates the importance of the proximal segment (–209/–
1) in LPS-induced promoter activity in TLR4-transduced Ns5a TICS cells. (B) Nanog and Stat3
26

dependence of the TWIST1 promoter activity, demonstrated by abrogating the potential binding site of
NANOG and STAT3 on the TWIST1 (-209/-51) promoter using in-vitro mutagenesis. (C) Also similar
effects were observed by abrogating Tlr4 and Nanog using shRNA, Stat3 using Stat3D, and
overexpressing Stat3 using Stat3C, followed by treatment with LPS (10ug/ml)/media or Leptin
(5ug/ml)/media, n=3 (D) Schematic diagrams of TWIST1 promoter depicting the locations of NANOG
consensus binding sequences (yellow scripts), and STAT3 consensus binding sequences (Green scripts),
and the specificity control areas analyzed by ChIP (white scripts). Immediate below the schematic
representation is NANOG ChIP-qPCR data (black bar graphs) for TICs post LPS and Leptin treatment,
with NANOG, and STAT3 enrichment; n=3. Showing Induction of NANOG, and STAT3 using STAT3
ChIP-qPCR (blue bar graphs), n=3. (E) Re-CHIP/ Sequential ChIP qPCR analysis showing significant
fold enrichement of NANOG and STAT3 in TICs post LPS and Leptin treatment; n=3.

Figure 2.4.1 TWIST1 promoter activation in Huh7. TWIST1 promoter analysis with deletion constructs
demonstrates the importance of the proximal segment (–209/–51) in LPS-induced promoter activity in
Huh7 cells.

2.3.7 Mouse and human HCC have accentuated expression of TLR4, p-STAT3, and TWIST1
To understand the mechanistic link of the LPS-TLR4-NANOG axis and Lepin-OB-R-pSTAT3
axis influence on TWIST1 in HCC, immunoblot analysis was performed on liver tumors from HCFD-fed
NS5A Tg mice vs. normal liver from chow-fed control mice. As speculated, TLR4, STAT3, p-STAT3,
and TWIST1 were all upregulated in the tumor lysates (Fig. 2.5 A) and a mirror image of that was seen in
qRT-PCR analysis (Fig. 2.5 B-D). Immunostaining of these liver tumor sections also confirms increased
expression of TLR4, TWIST1, and p-STAT3, and their co-localization (Fig. 2.5.1 A and B). We next
27

assessed the clinical relevance of our finding by analyzing the expression of these proteins in HCC
samples from obese HCV patients as compared to their non-cancerous liver tissues by
immunohistochemistry. NANOG, TLR4, p-STAT3, and TWIST1 immunofluorescent staining were
detected and co-localized only in cancerous tissues (Fig. 2.5.2 A). TWIST1 nuclear staining was more
prevalent in HCC sections, but very minimally detected in normal liver tissues (Fig. 2.5.2 B). TWIST1
expression observed both at nuclei and also at pri-nuclei is quite the daunting factor, we observed
multiple cells expressing TWIST1 either at pri-nuclei or cytoplasmic region (data not shown).

Figure 2.5 Induction of TLR4/NANOG/P-STAT3/TWIST1 pathway components in Mouse HCC.
(A) TLR4, STAT3, P-STAT3, NANOG and, TWIST1 protein levels are increased in HCC specimens
from HCFD Ns5a Tg mice, as compared with cirrhotic or healthy livers, N = 5 samples/cohort, n=3 (B)
Twist1 (C) Nanog (D) Stat3 mRNA profiling showing significant increase in HCFD Ns5a Tg mice Liver
tumor, in contrast with the healthy livers, *p<.05, N=8 samples/cohort, n=3.

28

Figure 2.5.1 Induction of TLR4/NANOG/P-STAT3/TWIST1 pathway components in HCFD NS5A
Tg mouse. (A) Immunofluorescent microscope demonstrates higher expression of TWIST1, TLR4,
NANOG and P-STAT3, which are often colocalized in HFCD Ns5a Tg specimens, as compared with
control diet liver tissue. Original magnification, 40x oil, N=15 samples/cohort, n=3 (B)
Immunohistochemical staining showing increased TWIST1 expressing cells in HFCD Ns5a Tg
specimens, as compared with control diet liver tissue, 40X magnification, N=15 samples/cohort, n=3. (C)
Quantification of the IHC data using Metamorph.

29

Figure 2.5.2 Accentuated protein expression in Human HCC+obesity samples (A)
Immunofluorescent microscopic analysis demonstrates higher expression of TLR4, NANOG and, P-
STAT3, which are often colocalized with TWIST1 in patient HCC specimens, as compared with
noncancerous liver tissue (Original magnification, 40x oil. Red boxed are cropped images) N=7
samples/cohort, n=3 (B) Immunohistochemical staining showing increased TWIST1 expressing cells in
patient HCC specimens, as compared with noncancerous liver tissue, Red boxed are 40X magnification,
N=7 samples/cohort, n=3 (C) Immunohistochemical staining showing increased TLR4 and P-STAT3
expressing cells in patient HCC specimens, as compared with noncancerous liver tissue, Red boxed are
40X magnification, N=7 samples/cohort, n=3.

30

2.3.8 TWIST1 overexpression promotes tumor formation
Our results indicate Twist1 silencing reduces TIC-derived tumorigenesis (Fig. 2.3 E), and that
Twist1 is downstream of TLR4 (Fig. 2.4). We then asked whether Twist1 overexpression can promote
tumor formation, and metastasis, and how Tlr4 silencing can influence these outcomes. To address these
questions, we transplanted TICs infected with lentivirus expressing scrambled or Tlr4 sh-RNA and an
empty or Twist1 expressing retrovirus vector into NOG mice. Overexpression of Twist1 indeed promoted
the tumor growth, and significantly increased the final tumor volume and weight (Fig. 2.6 A-C).
Concomitant Tlr4 silencing reduced the final tumor volume and tended to decrease the final tumor weight
(P=0.05). This was surprising as TLR4 is upstream of Twist1, but underscores the importance of other
TLR4-dependent oncogenic pathways. Twist1 overexpression also increased metastasis to the lung
compared to the liver (Fig. 2.6 D).
31

Figure 2.6 Twist1 over expression drives tumor growth independent of Tlr4. (A) Sh-scramble + over
expression of Twist1, Sh-Tlr4 + over expression of Twist1, Sh-scramble + retro vector, and Sh-Tlr4 +
retro vector cells were injected subcutaneously into the flanks of NOG mice (1 million cells/ site) and
tumor volume were observed to be increased with Twist1 over expression in TICs with or without Tlr4
silencing. (B) Tumor volume at day 15, 25 and, 30 showed an increase in tumor volume with Twist1 over
expression when compared with their respective controls. (C) Tumor weight at day 30 when tumor
32

bearing mice were sacrificed. (D) Metastasis to liver and lung organs post subcutaneous injection from
previously described 4 different cohorts.

Discussion
The findings presented in this study describe an unexpected convergence of recent research on NANOG
and STAT3 pathways. We have identified an important and functional link between the NANOG pathway
involving LPS-TLR4, and the STAT3 pathway involving Leptin and its receptor. These two pathways
work together to activate the Twist1 gene, a master regulator of EMT, in the maintenance of stemness of
Tumor-initiating stem-like cells (TICs); TICs a small percentage of cells with stem-like properties
residing in a tumor, have been documented in a variety of cancerous tissues
49, 50
. EMT remodels cells
and tissues during embryogenesis, wound healing, and simultaneously with the acquisition of malignant
traits
51, 52
. We demonstrated increased incidence of tumor formation in NS5A Tg mice fed HCFD (Fig.
2.1). Tumor formation (Fig. 2.2) was further promoted in the NS5A Tg mice fed both Alcohol and western
diet. The canonical TAK-1 and TRAF6 signaling is activated in both the HCFD as well as HCFD+
alcohol fed NS5A Tg mouse models, indicating TLR4 is ectopically expressed in the parenchymal cells
with or without LPS treatment (Fig. 2.1 & 2.2).
We observed from the RNA microarray analysis that Twist1, a master regulator of EMT
47, 53, 54
, was
increased 11-fold in NS5A-TICs (Fig. 2.2 C). Feeding of HCFD supplemented activates Tlr4, and Nanog
signaling (Fig. 2.2 C), along with increased Leptin and endotoxin levels in the plasma (Fig. 2.1). Previous
RNA microarray on TICs from alcohol fed NS5A Tg mouse models
5
, which lacked induction of Twist1,
led us to hypothesize that for activation of Twist1 in TICs from HCFD NS5A Tg model, the adipose tissue
derived Leptin- pSTAT3 axis is needed along with the TLR4-NANOG axis. This notion further led us to
analyze the NANOG and STAT3 binding sites on the Twist1 promoter (Fig. 2.4). However, both NANOG
proximal and STAT3 distal mutants tend to attenuate the luciferase signal in response to Leptin and LPS
stimulation, respectively. We think this is due to the complexity of the two binding proteins being in an
33

immediate proximity to each another. Which was further confirmed through reCHIP qPCR analysis (Fig.
2.4 E). Watt et al., showed that Nanog might interact with Stat3 to regulate its own gene expression
55
.
We believe this interaction potentially causes induction of Twist1. In mouse and human tissue sections,
IHC and IF staining against TWIST1, TLR4, P-STAT3, NANOG further illustrated their co-localization.
Very few cells had cytoplasmic and peri-nuclear expression of TWIST1 in IF compared to IHC, which
had more of nuclear staining. Similar expression profiling has been previously described by Sun et al.
48

Moreover, we observed that over-expression of Twist1 without Tlr4 can independently drive the tumor
formation and metastasis (Fig. 2.6).

Figure 2.7 A schematic representation of the hypothetical model proposed link between oncogenic
TLR4/NANOG signaling, OB-r/P-STAT3, 9 and an effective TWIST1 pathway in generating TICs with
EMT characteristics. Ectopic upregulation of TLR4 and its activation by LPS induce the pluripotency
factor NANOG, other stem cell genes, and self-renewal of TICs. On the other hand, upregulation of
Adipose cells and its increased secretion of Leptin induces P-STAT3. NANOG and P-STAT3 induce
TWIST1, which in turn causes increased EMT characteristics in TICs.

In conclusion, stemness markers NANOG and STAT3 promote elevated levels of Twist1 in TICs through
LPS-TLR4 and Leptin-OB-R that promotes malignancy and mesenchymal phenotype (Fig. 2.7).
34

Chapter 3
Cell-Cell competition between tumor-initiating cells and Hepatic Stellate induces fusion
with Kupffer cells, leading to metastatic colonization in alcoholic liver cancer.

3.1 Introduction
The Tumor-Initiating cells (TICs) are a small population of cells in the tumor microenvironment,
which not only possess normal stem cell properties: self-renewal, and differentiation; but the property of
tumor-initiation and maintenance of self-growth
56
. Nevertheless, the origin of Tumor-Initiating cells is
still unclear
57,58
. Previous reports so far indicate that Tumor-Initiating cells are dedifferentiated from
cancer cells, however others suggest that Tumor-Initiating cells are differentiated from normal stem
cells
57,58
. Whatsoever the resource of TICs are; TICs have to develop with their niche. Known stem cell’s
niche functions are: maintaining the stem cell quiescence, and providing proliferation or differentiation-
inducing signals
59
. Niche secrets signals that support Tumor-Initiating cells self-renewal and proliferation;
abrogating this niche signals could inhibit Tumor-Initiating cells expansion
60
. However, the mechanism of
the interaction between TICs and their niche is undisclosed, and poorly understood.

These homologous traits between the stem cells and the tumor-Initiating cells encouraged us to study the
stem cell niche which probably supports TICs development; either directly or indirectly. The composition
of the stem cell niche is usually more than one cell type; for example, the hepatic progenitor (hepatoblast)
cell’s niche is composed of: Hepatic Stellate Cells (HSCs), Endothelial Cells (ECs), Hepatocytes,
Cholangiocytes, Kupffer Cells (KCs), pit cells (liver-specific natural killer (NK) cells), and inflammatory
cells
59, 61
. Moreover, all of these cells could interact and cross-talk with HPCs, influencing their
proliferation and differentiation processes
59
.The complex nature of many cell types involved makes it an
unfathomable mystery. Nevertheless, understanding the molecular mechanisms of this interaction
between TICs and its niche will be a key step to help evaluate HCC's development, and which further can
35

be applied to cancer therapy. In this study we show evidence that a part of hepatoblast's stem cell niche:
the HSCs, & the KCs is able to support TICs expansion. Cytogenetic analysis of male TICs (XY)
cocultured with Female Niche (XX) demonstrated a cell-cell fusion (XXXY) between KCs and TICs.

Macrophages fuse with themselves to differentiate into multinucleated osteoclasts or giant cells through
CD47 and SIRP-α signaling
62
. CD47 (don’t eat me signal) is ubiquitously expressed at low levels on
normal cells, and multiple tumors express increased levels of CD47 compared with their normal cell
counterparts
63, 64
. Though the cue for this fusion was oblivious, we examined the Fibroblast growth factor
(FGF) role in this process. FGF family are believed to play critical roles during organogenesis and
carcinogenesis via signaling between epithelial and stromal compartments
65
. Since, FGF2 stimulates
embryonic fibroblasts (EFs) to secrete Activin A (ActA)
66
. ActA belongs to the TGF-b family of ligands
and promotes the activation of the SMAD2/3 transcription factors, which is considered beneficial for
hESC self-renewal
67-69
. These known cascade of events led to the hypothesis that, CD47 & SIRP-α
interaction could play a critical role in this fusion process and cue for this cellulocytosis is FGF secretion
from the microenvironment. Here, we identified that FGFR2, and CD47 enriched TICs, share its receptor
(FGFR2) to the emanated FGF which engender them to share their ligand (CD47) to the SIRP-α receptor
of the Kupffer cells for cellulocytosis. We propose that anti-cancer drugs targeting this niche mediated
fusion could/will be a new therapeutic approach for the treatment of HCC.

3.2 Material and Methods
Isolation of Tumor-Initiating Cells
Tumor-initiating Cells (TICs) were isolated from liver tumor acquired from our novel animal
model, HCV-NS5A double Tg mice with high cholesterol, high fat diet (HCFD)/Alcohol long-term (12
months) feeding. Liver tumors from mice obtained the post-surgical removal were mechanically
dissociated, and digested in a cocktail of Dispase (Sigma), Dnase I (Sigma), and Collagenase (Sigma).
36

Then, incubated at 37 ℃ for 2 hours. Single cell suspensions were incubated with CD133 microbeads
(Miltenyi) and separated using an auto-MACS device (Miltenyi), according to manufacturer's
instructions
12
.TICs were maintained in Dulbecco’s modified Eagle’s medium nutrient mixture F-12
(DMEM/F12) containing 10% fetal bovine serum (FBS), Nucleosides, 1 µM dexamethasone, mouse
epidermal growth factor (mEGF), 1 µg/ml penicillin, 1 µg/ml streptomycin and nonessential amino acid
(NEAA).

Lentivirus production and TIC Labeling
Isolated TICs were labeled with DsRed via lentivirus infection. Lentivirus vectors were prepared by
standard procedures in HEK293T cells. Three plasmids, packaging vector pPAX2, ecotropic envelope
gene expression vector pMD2.G, and the Lentivirus vector, DsRed, shCD47 or shFGFR2, were
transfected in HEK293T cells by lipid-based transfection reagent BioT (Bioland). The Lentivirus
supernatant was harvested at 48 hours of transfection filtered through a 0.45µm filter and aliquots were
stocked at -80℃. The concentrated viral stocks were generated by ultracentrifugation. TICs were infected
with 10-30 M.O.I. of lentivirus mixed with polybrene (6µg/mL). Hygromycin B (30µg/ml) was used to
select DsRed virus infected TICs while puromycin (10µg/ml) was used to select shCD47 and shFGFR2
virus infected TICs.

Isolation and Culturing of Hepatic Stellate Cells and Kupffer Cells
Hepatic stellate cells (HSCs) and kupffer cells (KCs) were isolated from Collagen-α1 (I) GFP mice
which were injected twice per week, subcutaneously, with CCl 4 in mineral oil (2µl of CCl 4/ gm of mouse
and mineral oil 1:1 mixture) for a total of 6 weeks to induce liver fibrosis and activate HSCs. The HSCs,
isolated from the mineral oil injected Collagen-1(I) GFP mice, were used as control. HSCs and KCs were
37

isolated by in situ digestion of the liver and arabinogalactan gradient ultracentrifugation as reported
previously. The purity of the cells was examined by phase-contrast microscopy, and the viability was
examined by Trypan blue exclusion. Isolated cells were maintained in Dulbecco’s modified Eagle’s
medium, containing 10% fetal bovine serum (FBS), 1 µg/ml penicillin, 1 µg/ml streptomycin and
nonessential amino acid (NEAA).

In vivo tumor formation assay
The isolated HSCs (1×10
4
cells) and/or KCs (1×10
4
cells) were subcutaneously co-injected with
DsRed labeled TICs (2,500 or 1×10
4
cells) into NOG mice (NOD/Shi- scid/IL-2Rγnull, a severely
immunodeficient mouse). HSCs (1×10
4
cells), KCs (1×10
4
cells) and/or TICs (1×10
4
cells) were also used
for in vitro co-culture.
The tumor volume was measured at day 21 and 35 post-transplantation with caliper and calculated by
using the formula: V=a×b
2
/2, where “V” represents tumor volume, a presents the largest superficial
diameter, and “b” presents the smallest superficial diameter. The tumor bearing NOG mice were
sacrificed at day 35, the tumors were collected and the tumor weight also measured.

Plasmids, Lentivirus, and Retrovirus Vectors
Lentivirus vectors were prepared by standard procedures in HEK293T cells. Three plasmids,
packaging vector pPAX2 (Addgene), ecotropic envelope gene expression vector pMD2.G (Addgene), and
the cassettes of shRNA, Scramble, Twist1, CD47, and FGFR2 (Open Biosystems), were transfected in
HEK293T using BioT transfection reagent (Bioland Scientific LLC). After 48 hours of transfection, the
virus supernatants were harvested, purified and mixed with polybrene (8 µg/mL) when used to infect
TICs. The infected cells were then screened with 10µg/ml puromycin and 10µg/ml.
38

Mouse CD47 were
AGGCCAAGTCCAGAAGCATTC and AATCATTCTGCTGCTCGTTGC
63

Mouse FGFR2: sense, 59-AAATACCAAATCTCCCAACC-39; antisense, 59-
GCCGCTTCTCCATCTTCT-39
70

Cytogenetics
TICs were plated at a density of 1*10
3
cm
22
on T75 flasks (Falcon) in TICs medium. After 42 h, colcemid
(Sigma) was added at a final concentration of 150 ng ml
-1
. After 4–6 h, cells were digested with trypsin,
placed in hypotonic medium consisting of 5% FCS and 75mM Kcl, and fixed to slides. The slides were
rinsed with PBS, fixed with 3:1 methanol: acetic acid solution for 80min, submerged in 70% acetic acid
for 2 s, and then allowed to dry overnight. The next day, the slides were washed with 2 X SSC at 37 C for
30 min, and then treated with 10% pepsin (Sigma) diluted in 40ml of 0.01M HCl at 37 C for 20 min. The
slides were then washed with PBS for 5 min, placed in post-fixation solution (1ml 37% formaldehyde,
0.18 g MgCl2, 39ml
PBS) for 5 min and dehydrated with 70%, 80% and 100% ethanol for 2 min each at room temperature.
Immediately before hybridization, the slides were denatured in a 70% formamide solution, pH 7.0 (35 ml
formamide, 10 ml £ 10 SSC and 5ml ddH20) at 72 8C for 4min.

Fluorescence in situ hybridization (FISH)
Meanwhile, paints for the mouse X (FITC) and Y (Cy3) chromosomes (Vysis) were mixed,
denatured at 75 8C for 10 min, and preannealed at 37 8C for 30 min. The probes were then added onto the
slides and hybridization was carried out using HYBrite (Vysis; 72 8C denaturation temperature, 2min
denaturation time, and reannealing at 37 8C overnight). The next day, the slides were washed with 2X
SSC at 73 8C for 5min, with 2X SSC/0.1% NP-40 (Sigma) at room temperature for 1 min, and then
39

counterstained with 125 mg ml21 4,6-diamidino-2-phenylindole (DAPI) II (Vysis). Cells were observed
using a Nikon E800 fluorescence microscope, and captured using CytoVision software from Applied
Imaging.

Statistical analysis
Student’s t-test was used to analyze data using Statistical software.

3.3 Results

3.3.1 Hepatic progenitor cell’s niche increases the tumorigenicity of TICs
Primarily in order to access the role of Niche in TICs tumorigenicity the collagen-α1 (I) GFP
mice derived HSCs – activated (CCl4) or quiescent (Oil) were mixed with TICs (CD133+) or CD133-
cells or PIL4 cells pre-injection and delivered to the NOG mice subcutaneously. A statistical increase in
the tumor volume was observed when the activated HSCs was co-injected with TICs which was absent in
the control (Fig. 3.1A). A synergistic effect in the tumor volume was observed when CCl4 collagen-α1 (I)
GFP mice derived KCs injected along with activated HSCs and TICs (Fig. 3.1B) and as well as in the
incidence of metastasis to distal organs of dissimilar germ line lineage (Fig. 3.1.1). In order to assess the
role of the Niche ex vivo, all the three cells: HSCs (LX2), Macrophages (RAW 264.7) and, TICs were
examined for the colony and sphere formation assay. Consistent with the in vivo results increased number
of colonies (Fig. 3.1C) and Sphere (Fig. 3.1D) was observed when all the three cells co-cultured.
Microscopy analysis of the activated HSCs (GFP) and KCs (GFP) (collagen-α1 (I) GFP mice) and DS red
labelled TICs, grown either by direct co-culture or by indirect means (Boyden chamber 6 micron)
unexpectedly showed double positive cells in the direct co culture method (Fig. 3.1E-9).
40

Figure 3.1 Niche increases the tumorigenicity of TICs (A) Comparison of tumor size in different cell
types (Activated Stellate cells, TICs, and PIL4) and cell number. (B) Tumor volume with respect to
different cell types. Where, all the three cell type together causes maximum volume incidence. Upper
panel gross image of the tumor and lower panel indicating a graphical representation of the observed
volume, *p<0.05. Under in vitro conditions, (C) Colony formation assay showed increased number of
colonies when all the three cells types were grown together, *p<0.05, n=3. (D) Spheroid formation assay
showing increase in the number of spheres formed when all the three cells types were co-cultured,
*p<0.05, n=3. (E) HSCs and Kupffer cells isolated from Col-GFP Tg mice (in green fluorescence) and
TICs labeled with DS red show fusion in-vitro. Live cell images of five days co-cultured cells. (1) HSCs
single culture (2) TICSs and HSCs indirect co-culture (3) TICSs and HSCs direct co-culture (4) KCs
41

single culture (5) TICSs and KCs indirect co-culture (6) TICSs and KCs direct co-culture (7) TICSs
single culture (8) TICSs, HSCs and KCs indirect co-culture (9) TICSs and KCs direct co-culture. 10X
microscopy.

Figure 3.1.1 Niche induces increase metastatic ability in TICs. Tumor incidence percentage in
different organs (similar and dissimilar germ line lineage) post sub cutaneous injection in NOG mice
indicates that the Niche promotes metastasis of TICs to different germ line lineage.

3.3.2 Niche induces cellulocytosis of Kupffer cells with TICs
To confirm our ex vivo demonstration of the two merged cells (Ds red labelled TICs and GFP
labelled Niche) we then performed flow cytometry analysis on the cells isolated either from subcutaneous
tumor or metastatic liver tumor. Cells were isolated via MACS™ based on their cell surface markers
CD133 (TICs marker) or CD11b (Macrophage marker) (Fig. 3.2 A-4). CD133+ cells sorted from
subcutaneous tumor showed 18.89% double positivity with F4/80 (Fig. 3.2 A-1 upper panel) and 6.17%
with GFP-Niche (Fig. 3.2 A-1 lower panel). This was consistently seen in CD11b sorted cells from
metastatic liver tumor 7.29% (Fig. 3.2 A-2 upper panel) and 4.77% (Fig. 3.2 A-2 lower panel). The
double positive cells (Fig. 3.1 E-9) were then sorted using FACSAria™ and maintained in vitro with
hygromycin B as selection marker (Fig. 3.2 B 1-3). Colonies formed out of these sorted cells to strongly
express the double positivity (Fig. 3.2 B 4-6). Further to assert these findings, cytogenetic analysis were
42

performed on fused and control cells. TICs obtained from male (XY) mice and Niche from female (XX)
mice were co-cultured and FISH analysis was performed. Increase in the number chromosomes post
fusion (Fig. 3.2 C) and the expression of XXXY (Fig. 3.2 D) confirms the occurrence fusion.

Figure 3.2 Niche induces fusion of Kupffer cells with TICs. (A) Flow cytometry analysis of KCs and
TICs in subcutaneous tumor and metastatic liver tumor. APC-F4/80, DS red-TICs, FITC-GFP labelled
Niche (1) MACS, CD133 sorted cells from subcutaneous tumor (2) MACS, CD11b (Macrophage1
marker) sorted cells from subcutaneous tumor (3) MACS, CD11b (Macrophage1 marker) sorted cells
43

from metastatic liver tumor (4) Schematic representation of the protocol (B) Live cell images of cell-
fusion after sorting. Triple co-cultured cells (Ds red labelled TICs (male mice), HSCs, and GFP labelled
KCs (female mice)) were sorted using FACSAria™ based on their expression of CD133 (2) and F4/80
(1). The isolated double positive cells (3) were screened with hygromycin B; 10X microscopy. Colony
formation assay of these sorted cells expressing double positivity (4-6) (C) Cytogenetic analysis of TIC's
metaphase spreading (left) and sorted cell (right) demonstrates increase in chromosome number. (D)
Increased X (red) and Y (green) chromosomes indicates the fusion of cells via Fish analysis post co-
culture of the three cell type in-vivo.

3.3.3 Accentuated Expression of CD133 and F4/80 in metastatic organs
Further in order to confirm the presence of these double positive population of cells we then
performed Immunofluorescence staining on the metastatic liver tumor obtained from the NOG mice
which was subcutaneously injected with TICs and its niche. This demonstrated the presence of double
positive cells - relatively rare in number (Fig 3.3A Top most panel) compared to the (Fig. 3.3A third
panel from the top) expression of individual CD133 and αSMA (activated HSCs cell marker).

Figure 3.3 Metastatic TICs has accentuated expression of F4/80 and CD133. (A) Subcutaneous tumor
formed from co-injection of TICs and its Niche show expression of markers of two different germ line
lineages.

44

3.3.4 CD47 and FGFR2 crucial role in the regulating TICs Niche
The role of CD47 in anti-phagocytosis is well documented in many tumors
71-73
. To understand
whether this plays a critical role in the fusion process of TICs with Kupffer cells, we hypothesized that by
blocking CD47 expression in TICs (Fig. 3.4 A) which is expressed by 41.96% of the CD133+ cells the
tumorigenicity of TICs will reduce. Profiling of mRNA, protein and viability are demonstrated (Fig. 3.4.1
A1-3, B) post knocking down of CD47 via Lentivirus sh-RNA. Here we show evidence that the CD47
abrogated TICs when injected along with their niche have a statistically significant reduction in the tumor
volume compared to their control cohorts (Fig. 3.4.1 C 1-3). We then analyzed the potential cue for this
cascade of events leading to fusion. ELISA assay (Fig. 3.4 B) demonstrates a prominent level of FGF2
secretion by the Niche. To comprehend this, loss of function assay of FGFR2 on TICs was designed using
Lentivirus sh-RNA. Profiling of mRNA, protein, viability, tumor size and weight (post subcutaneous
injection along with their niche) are demonstrated (Fig. 3.4.1 A4-6, B, and C1-3).

Figure 3.4 CD47 and FGF2 expression levels (A) Flow cytometry analysis establishes accentuated
expression of CD47 (Don’t eat me signal) in TICs. (B) ELISA demonstrating increased secretion of Fgf2
in the conditioned media post co-culture of all the 3 cell types.

45

Figure 3.4.1 CD47 and FGF2 regulate proliferation and tumor growth (A) RNA and protein
expression levels of CD47 (1) and FGF2R (2) in TICs (Scrambled and post Lentivirus mediated
knockdown). (B) In vitro, aberration of cd47 or fgf2r in TICs decreased cell proliferation in co-culture
model. (C) Live Mouse imaging of Ds Red labelled TICs and GFP expressing (Niche; HSCs & KCs) at
21 days post co-injection shows both TICs, and its Niche are alive after co-injection (1). Knocking down
46

of cd47 or fgf2r gene expression in TICs reduces subcutaneously co-injected tumor volume (2) and tumor
weight (3).

3.3.5 Cell-Cell competition leads to HSCs apoptosis
Both survival and apoptosis of HSC are regulated by growth factors expressed during fibrotic
liver injury
74
. To understand the underpinning events leading to the fusion, and we speculated the role
HSCs in the microenvironment. What astonished us is the HSCs death when co-cultured along with TICs
(Fig 3.5A). We then investigated the role of apoptosis in HSCs, via Flow cytometry analysis (Annexin V
– apoptosis marker) we demonstrated that the HSCs under apoptosis in vivo when co-cultured with TICs
and KCs (Fig. 3.5B).

Figure 3.5 Activated Stellate cells of the Niche undergoes apoptosis (A) A decrease in viable Carbon
tetrachloride activated Hepatic stellate cells was observed post co-culture with TICs (upper panel); n=3.
(B) Flow cytometry analysis of AnnexinV - DAPI (apoptosis marker) staining on the co-cultured cells
shows evidence of Apoptosis in activated Hepatic Stellate cell; n=3.

47

Discussion
In this report, we identified TICs fuse with KCs cells through programmed cell death of HSCs
leading to highly metastatic trait of TICs. We show evidence that CD47 and FGFR2 play a crucial role in
this process.
TICs share with stem cells the capacity to self-renew and to migrate. When TICs leave their primary
tissue to invade another one, they have become metastatic. Metastasis is the most devastating attribute of
cancer, being the spread of tumor cells to distal organs of either same or dissimilar germ line lineage in
which they proliferate. This is associated with enhanced motility of TICs and with the capacity of these
cells to evade the immune system. Thus, metastatic cells appear to have at least one of the functional
characteristics of macrophages, which is mobility
62
. One among the famous notion is the abortive
digestion of apoptotic tumor cells might be a potential mechanism for hybrid formation
75
. Evidence
suggests that somatic/tumor cells fuse with macrophages, to create diversity, metastasis, chromosomal
aberration and epigenetic regulation
76
. Also, the fusion between hepatocytes and bone marrow derived
macrophages shed light on the somatic cell fusion in case of liver injury
77-79
. The findings presented in
this study show evidence that in the liver TICs form hybrids with Kupffer cells (Fig. 3.2 D), we believe
that this influence the TICs to be highly metastatic by nature (Fig. 3.1.1). The sequential events in this
process are schematically represented (Fig 3.6).
48

Figure 3.6 The Proposed model, a cell-cell communication between Tumor-initiating Cells and its
Niche.

The exudation of FGF2 from the surrounding niche provides signaling cues to HSCs and TICs to undergo
Cell-cell competition mediated apoptosis of HSCs. The loser cell (HSC) debris are phagocytized by the
resident liver macrophages the KCs whereas, the winner cell (TIC) undergo fusion (cellulocytosis) via the
CD47 – SIRP-α signaling cascade (Fig. 3.4.1 C 1-3).
49

Prospective of this research would be to manifest the role of CD47 and FGFR2 in fusion via cytogenetic
analysis (FISH) and analyze the chromosomal aberrations (SKY). And to show the devoid nature of
apoptosis in HSCs under the influence of abrogated FGFR2 in TICs.

50

BIBLIOGRAPHY

[1] Forner A, Llovet JM, Bruix J: Hepatocellular carcinoma. Lancet 2012, 379:1245-55.
[2] Okuda K: Hepatocellular carcinoma. Journal of hepatology 2000, 32:225-37.
[3] Bruix J, Gores GJ, Mazzaferro V: Hepatocellular carcinoma: clinical frontiers and perspectives. Gut
2014, 63:844-55.
[4] Hassan MM, Hwang LY, Hatten CJ, Swaim M, Li D, Abbruzzese JL, Beasley P, Patt YZ: Risk factors
for hepatocellular carcinoma: synergism of alcohol with viral hepatitis and diabetes mellitus. Hepatology
(Baltimore, Md) 2002, 36:1206-13.
[5] Machida K, Tsukamoto H, Mkrtchyan H, Duan L, Dynnyk A, Liu HM, Asahina K, Govindarajan S,
Ray R, Ou JH, Seki E, Deshaies R, Miyake K, Lai MM: Toll-like receptor 4 mediates synergism between
alcohol and HCV in hepatic oncogenesis involving stem cell marker Nanog. Proc Natl Acad Sci U S A
2009, 106:1548-53.
[6] Yuan JM, Govindarajan S, Arakawa K, Yu MC: Synergism of alcohol, diabetes, and viral hepatitis on
the risk of hepatocellular carcinoma in blacks and whites in the U.S. Cancer 2004, 101:1009-17.
[7] Guichard C, Amaddeo G, Imbeaud S, Ladeiro Y, Pelletier L, Maad IB, Calderaro J, Bioulac-Sage P,
Letexier M, Degos F, Clement B, Balabaud C, Chevet E, Laurent A, Couchy G, Letouze E, Calvo F,
Zucman-Rossi J: Integrated analysis of somatic mutations and focal copy-number changes identifies key
genes and pathways in hepatocellular carcinoma. Nature genetics 2012, 44:694-8.
[8] Fujimoto A, Totoki Y, Abe T, Boroevich KA, Hosoda F, Nguyen HH, Aoki M, Hosono N, Kubo M,
Miya F, Arai Y, Takahashi H, Shirakihara T, Nagasaki M, Shibuya T, Nakano K, Watanabe-Makino K,
Tanaka H, Nakamura H, Kusuda J, Ojima H, Shimada K, Okusaka T, Ueno M, Shigekawa Y, Kawakami
Y, Arihiro K, Ohdan H, Gotoh K, Ishikawa O, Ariizumi S, Yamamoto M, Yamada T, Chayama K,
Kosuge T, Yamaue H, Kamatani N, Miyano S, Nakagama H, Nakamura Y, Tsunoda T, Shibata T,
Nakagawa H: Whole-genome sequencing of liver cancers identifies etiological influences on mutation
patterns and recurrent mutations in chromatin regulators. Nature genetics 2012, 44:760-4.
51

[9] Bonnet D, Dick JE: Human acute myeloid leukemia is organized as a hierarchy that originates from a
primitive hematopoietic cell. Nature medicine 1997, 3:730-7.
[10] Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF: Prospective identification of
tumorigenic breast cancer cells. Proc Natl Acad Sci U S A 2003, 100:3983-8.
[11] Choi SA, Lee JY, Phi JH, Wang KC, Park CK, Park SH, Kim SK: Identification of brain tumour
initiating cells using the stem cell marker aldehyde dehydrogenase. European journal of cancer (Oxford,
England : 1990) 2014, 50:137-49.
[12] Chen CL, Tsukamoto H, Liu JC, Kashiwabara C, Feldman D, Sher L, Dooley S, French SW, Mishra
L, Petrovic L, Jeong JH, Machida K: Reciprocal regulation by TLR4 and TGF-beta in tumor-initiating
stem-like cells. The Journal of clinical investigation 2013, 123:2832-49.
[13] Barabe F, Kennedy JA, Hope KJ, Dick JE: Modeling the initiation and progression of human acute
leukemia in mice. Science (New York, NY) 2007, 316:600-4.
[14] Peters MG, Terrault NA: Alcohol use and hepatitis C. Hepatology (Baltimore, Md) 2002, 36:S220-5.
[15] Donato F, Gelatti U, Limina RM, Fattovich G: Southern Europe as an example of interaction
between various environmental factors: a systematic review of the epidemiologic evidence. Oncogene
2006, 25:3756-70.
[16] Song SJ, Poliseno L, Song MS, Ala U, Webster K, Ng C, Beringer G, Brikbak NJ, Yuan X, Cantley
LC, Richardson AL, Pandolfi PP: MicroRNA-antagonism regulates breast cancer stemness and metastasis
via TET-family-dependent chromatin remodeling. Cell, 154:311-24.
[17] Lai MS, Hsieh MS, Chiu YH, Chen TH: Type 2 diabetes and hepatocellular carcinoma: A cohort
study in high prevalence area of hepatitis virus infection. Hepatology (Baltimore, Md) 2006, 43:1295-
302.
[18] Shintani Y, Fujie H, Miyoshi H, Tsutsumi T, Tsukamoto K, Kimura S, Moriya K, Koike K: Hepatitis
C virus infection and diabetes: direct involvement of the virus in the development of insulin resistance.
Gastroenterology 2004, 126:840-8.
52

[19] Liu TX, Becker MW, Jelinek J, Wu WS, Deng M, Mikhalkevich N, Hsu K, Bloomfield CD, Stone
RM, DeAngelo DJ, Galinsky IA, Issa JP, Clarke MF, Look AT: Chromosome 5q deletion and epigenetic
suppression of the gene encoding alpha-catenin (CTNNA1) in myeloid cell transformation. Nature
medicine 2007, 13:78-83.
[20] Ribeiro PS, Cortez-Pinto H, Sola S, Castro RE, Ramalho RM, Baptista A, Moura MC, Camilo ME,
Rodrigues CM: Hepatocyte apoptosis, expression of death receptors, and activation of NF-kappaB in the
liver of nonalcoholic and alcoholic steatohepatitis patients. Am J Gastroenterol 2004, 99:1708-17.
[21] Yao F, Terrault N: Hepatitis C and hepatocellular carcinoma. Curr Treat Options Oncol 2001, 2:473-
83.
[22] Crippin JS, McCashland T, Terrault N, Sheiner P, Charlton MR: A pilot study of the tolerability and
efficacy of antiviral therapy in hepatitis C virus-infected patients awaiting liver transplantation. Liver
Transpl 2002, 8:350-5.
[23] Gale MJ, Jr., Korth MJ, Tang NM, Tan SL, Hopkins DA, Dever TE, Polyak SJ, Gretch DR, Katze
MG: Evidence that hepatitis C virus resistance to interferon is mediated through repression of the PKR
protein kinase by the nonstructural 5A protein. Virology 1997, 230:217-27.
[24] Kato N, Lan KH, Ono-Nita SK, Shiratori Y, Omata M: Hepatitis C virus nonstructural region 5A
protein is a potent transcriptional activator. J Virol 1997, 71:8856-9.
[25] Machida K, Cheng KT, Sung VM, Levine AM, Foung S, Lai MM: Hepatitis C virus induces toll-like
receptor 4 expression, leading to enhanced production of beta interferon and interleukin-6. J Virol 2006,
80:866-74.
[26] Feingold KR, Grunfeld C: Role of cytokines in inducing hyperlipidemia. Diabetes 1992, 41 Suppl
2:97-101.
[27] Lee JY, Sohn KH, Rhee SH, Hwang D: Saturated fatty acids, but not unsaturated fatty acids, induce
the expression of cyclooxygenase-2 mediated through Toll-like receptor 4. J Biol Chem 2001, 276:16683-
9.
53

[28] Feldman DE, Chen C, Punj V, Tsukamoto H, Machida K: Pluripotency factor-mediated expression
of the leptin receptor (OB-R) links obesity to oncogenesis through tumor-initiating stem cells. Proc Natl
Acad Sci U S A, 109:829-34.
[29] Van Heek M, Compton DS, France CF, Tedesco RP, Fawzi AB, Graziano MP, Sybertz EJ, Strader
CD, Davis HR, Jr.: Diet-induced obese mice develop peripheral, but not central, resistance to leptin. The
Journal of clinical investigation 1997, 99:385-90.
[30] Haluzik M, Gavrilova O, LeRoith D: Peroxisome proliferator-activated receptor-alpha deficiency
does not alter insulin sensitivity in mice maintained on regular or high-fat diet: hyperinsulinemic-
euglycemic clamp studies. Endocrinology 2004, 145:1662-7.
[31] Seglen PO: Preparation of isolated rat liver cells. Methods Cell Biol 1976, 13:29-83.
[32] Saitoh S, Akashi S, Yamada T, Tanimura N, Kobayashi M, Konno K, Matsumoto F, Fukase K,
Kusumoto S, Nagai Y, Kusumoto Y, Kosugi A, Miyake K: Lipid A antagonist, lipid IVa, is distinct from
lipid A in interaction with Toll-like receptor 4 (TLR4)-MD-2 and ligand-induced TLR4 oligomerization.
Int Immunol 2004, 16:961-9.
[33] Ohkuma M, Funato N, Higashihori N, Murakami M, Ohyama K, Nakamura M: Unique CCT repeats
mediate transcription of the TWIST1 gene in mesenchymal cell lines. Biochem Biophys Res Commun
2007, 352:925-31.
[34] Jauch R, Ng CK, Saikatendu KS, Stevens RC, Kolatkar PR: Crystal structure and DNA binding of
the homeodomain of the stem cell transcription factor Nanog. J Mol Biol 2008, 376:758-70.
[35] Cheng GZ, Zhang WZ, Sun M, Wang Q, Coppola D, Mansour M, Xu LM, Costanzo C, Cheng JQ,
Wang LH: Twist is transcriptionally induced by activation of STAT3 and mediates STAT3 oncogenic
function. J Biol Chem 2008, 283:14665-73.
[36] Marko-Varga G, Berglund M, Malmstrom J, Lindberg H, Fehniger TE: Targeting hepatocytes from
liver tissue by laser capture microdissection and proteomics expression profiling. Electrophoresis 2003,
24:3800-5.
54

[37] Michel C, Desdouets C, Sacre-Salem B, Gautier JC, Roberts R, Boitier E: Liver gene expression
profiles of rats treated with clofibric acid: comparison of whole liver and laser capture microdissected
liver. Am J Pathol 2003, 163:2191-9.
[38] Espina V, Milia J, Wu G, Cowherd S, Liotta LA: Laser capture microdissection. Methods Mol Biol
2006, 319:213-29.
[39] Wulfkuhle JD, Sgroi DC, Krutzsch H, McLean K, McGarvey K, Knowlton M, Chen S, Shu H, Sahin
A, Kurek R, Wallwiener D, Merino MJ, Petricoin EF, 3rd, Zhao Y, Steeg PS: Proteomics of human breast
ductal carcinoma in situ. Cancer Res 2002, 62:6740-9.
[40] Wulfkuhle JD, Liotta LA, Petricoin EF: Proteomic applications for the early detection of cancer. Nat
Rev Cancer 2003, 3:267-75.
[41] Espina V, Mehta AI, Winters ME, Calvert V, Wulfkuhle J, Petricoin EF, 3rd, Liotta LA: Protein
microarrays: molecular profiling technologies for clinical specimens. Proteomics 2003, 3:2091-100.
[42] Mathurin P, Deng QG, Keshavarzian A, Choudhary S, Holmes EW, Tsukamoto H: Exacerbation of
alcoholic liver injury by enteral endotoxin in rats. Hepatology (Baltimore, Md) 2000, 32:1008-17.
[43] Rodriguez LG, Wu X, Guan JL: Wound-healing assay. Methods Mol Biol 2005, 294:23-9.
[44] Carlsson G, Gullberg B, Hafstrom L: Estimation of liver tumor volume using different formulas - an
experimental study in rats. J Cancer Res Clin Oncol 1983, 105:20-3.
[45] Erridge C, Attina T, Spickett CM, Webb DJ: A high-fat meal induces low-grade endotoxemia:
evidence of a novel mechanism of postprandial inflammation. Am J Clin Nutr 2007, 86:1286-92.
[46] Kang Y, Massague J: Epithelial-mesenchymal transitions: twist in development and metastasis. Cell
2004, 118:277-9.
[47] Yang J, Mani SA, Donaher JL, Ramaswamy S, Itzykson RA, Come C, Savagner P, Gitelman I,
Richardson A, Weinberg RA: Twist, a master regulator of morphogenesis, plays an essential role in tumor
metastasis. Cell 2004, 117:927-39.
55

[48] Sun T, Zhao N, Zhao XL, Gu Q, Zhang SW, Che N, Wang XH, Du J, Liu YX, Sun BC: Expression
and functional significance of Twist1 in hepatocellular carcinoma: its role in vasculogenic mimicry.
Hepatology (Baltimore, Md), 51:545-56.
[49] Farnie G, Clarke RB: Breast stem cells and cancer. Ernst Schering Found Symp Proc 2006:141-53.
[50] Reyal F, Luton D, Feraud O, Doridot V, Multon O, Oury JF, Blot P, Sibony O: Monochorionic
monoamnionic twins: vaginal delivery. J Perinat Med 2001, 29:446-9.
[51] Nawshad A, Lagamba D, Polad A, Hay ED: Transforming growth factor-beta signaling during
epithelial-mesenchymal transformation: implications for embryogenesis and tumor metastasis. Cells
Tissues Organs 2005, 179:11-23.
[52] Thiery JP: Epithelial-mesenchymal transitions in development and pathologies. Curr Opin Cell Biol
2003, 15:740-6.
[53] Mani SA, Guo W, Liao MJ, Eaton EN, Ayyanan A, Zhou AY, Brooks M, Reinhard F, Zhang CC,
Shipitsin M, Campbell LL, Polyak K, Brisken C, Yang J, Weinberg RA: The epithelial-mesenchymal
transition generates cells with properties of stem cells. Cell 2008, 133:704-15.
[54] Morel AP, Lievre M, Thomas C, Hinkal G, Ansieau S, Puisieux A: Generation of breast cancer stem
cells through epithelial-mesenchymal transition. PLoS One 2008, 3:e2888.
[55] Torres J, Watt FM: Nanog maintains pluripotency of mouse embryonic stem cells by inhibiting
NFkappaB and cooperating with Stat3. Nat Cell Biol 2008, 10:194-201.
[56] Pardal R, Clarke MF, Morrison SJ: Applying the principles of stem-cell biology to cancer. Nat Rev
Cancer 2003, 3:895-902.
[57] Clarke MF, Fuller M: Stem cells and cancer: two faces of eve. Cell 2006, 124:1111-5.
[58] Lobo NA, Shimono Y, Qian D, Clarke MF: The biology of cancer stem cells. Annual review of cell
and developmental biology 2007, 23:675-99.
[59] Gaudio E, Carpino G, Cardinale V, Franchitto A, Onori P, Alvaro D: New insights into liver stem
cells. Digestive and liver disease : official journal of the Italian Society of Gastroenterology and the
Italian Association for the Study of the Liver 2009, 41:455-62.
56

[60] Clarke MF, Dick JE, Dirks PB, Eaves CJ, Jamieson CH, Jones DL, Visvader J, Weissman IL, Wahl
GM: Cancer stem cells--perspectives on current status and future directions: AACR Workshop on cancer
stem cells. Cancer Res 2006, 66:9339-44.
[61] Meurer SK, Tihaa L, Lahme B, Gressner AM, Weiskirchen R: Identification of endoglin in rat
hepatic stellate cells: new insights into transforming growth factor beta receptor signaling. J Biol Chem
2005, 280:3078-87.
[62] Vignery A: Macrophage fusion: are somatic and cancer cells possible partners? Trends in cell
biology 2005, 15:188-93.
[63] Jaiswal S, Jamieson CH, Pang WW, Park CY, Chao MP, Majeti R, Traver D, van Rooijen N,
Weissman IL: CD47 is upregulated on circulating hematopoietic stem cells and leukemia cells to avoid
phagocytosis. Cell 2009, 138:271-85.
[64] Majeti R, Chao MP, Alizadeh AA, Pang WW, Jaiswal S, Gibbs KD, Jr., van Rooijen N, Weissman
IL: CD47 is an adverse prognostic factor and therapeutic antibody target on human acute myeloid
leukemia stem cells. Cell 2009, 138:286-99.
[65] Abate-Shen C, Shen MM: FGF signaling in prostate tumorigenesis--new insights into epithelial-
stromal interactions. Cancer cell 2007, 12:495-7.
[66] Greber B, Lehrach H, Adjaye J: Fibroblast growth factor 2 modulates transforming growth factor
beta signaling in mouse embryonic fibroblasts and human ESCs (hESCs) to support hESC self-renewal.
Stem cells (Dayton, Ohio) 2007, 25:455-64.
[67] James D, Levine AJ, Besser D, Hemmati-Brivanlou A: TGFbeta/activin/nodal signaling is necessary
for the maintenance of pluripotency in human embryonic stem cells. Development (Cambridge, England)
2005, 132:1273-82.
[68] Brons IG, Smithers LE, Trotter MW, Rugg-Gunn P, Sun B, Chuva de Sousa Lopes SM, Howlett SK,
Clarkson A, Ahrlund-Richter L, Pedersen RA, Vallier L: Derivation of pluripotent epiblast stem cells
from mammalian embryos. Nature 2007, 448:191-5.
57

[69] Vallier L, Alexander M, Pedersen RA: Activin/Nodal and FGF pathways cooperate to maintain
pluripotency of human embryonic stem cells. Journal of cell science 2005, 118:4495-509.
[70] Chikazu D, Hakeda Y, Ogata N, Nemoto K, Itabashi A, Takato T, Kumegawa M, Nakamura K,
Kawaguchi H: Fibroblast growth factor (FGF)-2 directly stimulates mature osteoclast function through
activation of FGF receptor 1 and p42/p44 MAP kinase. J Biol Chem 2000, 275:31444-50.
[71] Chao MP, Jaiswal S, Weissman-Tsukamoto R, Alizadeh AA, Gentles AJ, Volkmer J, Weiskopf K,
Willingham SB, Raveh T, Park CY, Majeti R, Weissman IL: Calreticulin is the dominant pro-phagocytic
signal on multiple human cancers and is counterbalanced by CD47. Science translational medicine 2010,
2:63ra94.
[72] Chao MP, Tang C, Pachynski RK, Chin R, Majeti R, Weissman IL: Extranodal dissemination of
non-Hodgkin lymphoma requires CD47 and is inhibited by anti-CD47 antibody therapy. Blood 2011,
118:4890-901.
[73] Chao MP, Weissman IL, Majeti R: The CD47-SIRPalpha pathway in cancer immune evasion and
potential therapeutic implications. Current opinion in immunology 2012, 24:225-32.
[74] Issa R, Williams E, Trim N, Kendall T, Arthur MJ, Reichen J, Benyon RC, Iredale JP: Apoptosis of
hepatic stellate cells: involvement in resolution of biliary fibrosis and regulation by soluble growth
factors. Gut 2001, 48:548-57.
[75] Pawelek JM: Tumour cell hybridization and metastasis revisited. Melanoma research 2000, 10:507-
14.
[76] Duelli D, Lazebnik Y: Cell fusion: a hidden enemy? Cancer cell 2003, 3:445-8.
[77] Wang X, Willenbring H, Akkari Y, Torimaru Y, Foster M, Al-Dhalimy M, Lagasse E, Finegold M,
Olson S, Grompe M: Cell fusion is the principal source of bone-marrow-derived hepatocytes. Nature
2003, 422:897-901.
[78] Willenbring H, Bailey AS, Foster M, Akkari Y, Dorrell C, Olson S, Finegold M, Fleming WH,
Grompe M: Myelomonocytic cells are sufficient for therapeutic cell fusion in liver. Nature medicine
2004, 10:744-8.
58

[79] Vassilopoulos G, Wang PR, Russell DW: Transplanted bone marrow regenerates liver by cell fusion.
Nature 2003, 422:901-4.

Asset Metadata

Creator Uthaya Kumar, Dinesh Babu (author)

Core Title Roles of epithelial-mesenchymal transition and niche in tumorigenesis of tumor-initiating cells

Contributor Electronically uploaded by the author (provenance)

School Keck School of Medicine

Degree Master of Science

Degree Program Molecular Microbiology and Immunology

Publication Date 01/24/2015

Defense Date 05/27/2014

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag HCFD,NANOG,OAI-PMH Harvest,TLR4,tumor-initiating stem-like cells,TWIST1

Format application/pdf (imt)

Language English

Advisor Machida, Keigo (committee chair), Ou, J.-H. James (committee member), Stiles, Bangyan L. (committee member)

Creator Email dineshbabu2989@gmail.com,uthayaku@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-447164

Unique identifier UC11287636

Identifier etd-UthayaKuma-2739.pdf (filename),usctheses-c3-447164 (legacy record id)

Legacy Identifier etd-UthayaKuma-2739.pdf

Dmrecord 447164

Document Type Thesis

Format application/pdf (imt)

Rights Uthaya Kumar, Dinesh Babu

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Abstract (if available)

Abstract Abstract Study – 1 (Chapter 2) ❧ BACKGROUND & AIMS: Alcohol and obesity cause steatohepatitis through activation of Toll‐like receptor‐4 (TLR4) signaling and enhance hepatitis C virus (HCV) associated liver carcinogenesis. The HCV NS5A protein ectopoically upregulates TLR4 in hepatocytes, generates TLR4/NANOG dependent liver tumor initiating stem cell‐like cells (TICs), and induces liver tumor in alcohol fed transgenic (Tg) mice. These TICs have robust and selective expression of the leptin receptor, and increased phosphorylation and activation of STAT3. However, whether the TLR4‐NANOG pathway promotes an oncogenic signaling in other etiological mouse models and patients is ambiguous. METHOD: We sought to determine whether a Western diet high in cholesterol, and saturated fat (HCFD) that causes obesity can also synergistically induce liver tumors via TLR4 signaling. RESULTS: We have identified the TLR4‐NANOG oncogenic pathway in the genesis of TICs, and liver tumor in alcohol and/or HCFD fed NS5A Tg mice. A sensitized response in TICs to LPS and/or Leptin resulted in an enchanced Twist1 Promoter activity, which was abrogated by silencing: Tlr4, Nanog or Stat3. We provide evidence of NANOG and STAT3 sharing their binding site on Twist1 promoter to transactive it. CONCLUSION: Taken together, TLR4‐NANOG axis promotes liver tumorigenesis/metastasis through accentuated mesenchymal phenotype with elevated Twist1 expression upon exposure to HCV and alcohol/high‐fat diets. The TLR4 pathway serves as a novel therapeutic target for HCC. ❧ Abstract – Study 2 (Chapter 3) ❧ BACKGROUND & AIMS: Evidence that TLR4/NANOG dependent liver tumor initiating stem cell‐like cells (TICs) induces liver tumor in Western diet high in cholesterol and saturated fat (HCFD)/alcohol fed transgenic (Tg) mice. These TICs have robust characteristics of distal organ metastasis and tumor development (in similar or dissimilar germ line lineage). However, whether this characteristics of TICs is due to their microenvironment (dissimilar lineage) crosstalk in mouse models and patients is obscure. METHOD: We sought to determine whether the Hepatic Progenitor cell’s Niche acts as an intricate factor in regulating the homeostasis of TICs and can also synergistically induce their metastatic characteristics. RESULTS: We have identified the Hepatic Stellate cells (HSCs), and the Kupffer cells (KCs) are involved in the genesis of extremely metastatic TICs via fusion (cytogenetic analysis) between KCs and TICs. A sensitized Fibroblast growth factor 2 (FGF) emanated from the TIC’s Niche acts as a cue for a Cell‐Cell competition between TICs and HSCs leading to HSCs apoptosis. Enhanced expression of CD47 (don’t eat me signal) ligand in TICs ensures cellocytosis with KCs via its receptor SIRP‐α rather undergoing phagocytosis. Abrogation of CD47 or Fibroblast growth factor receptor 2 (FGFR2) in TICs prevented incidence of metastasis and tumor growth. CONCLUSION: Here we provide evidence that the TIC’s Niche serves as an intrinsic factor for its phenotypic trait—an extremely metastatic cell which can develop tumor in diverse germline lineage. The CD47 & FGFR2 pathway serves as a novel therapeutic target for HCC.