Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Investigating the function and epigenetic regulation of ABCA3, a novel LUAD tumor suppressor gene
(USC Thesis Other)
Investigating the function and epigenetic regulation of ABCA3, a novel LUAD tumor suppressor gene
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
1
Investigating the Function and Epigenetic
Regulation of ABCA3, a Novel LUAD Tumor
Suppressor Gene
A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the Requirements for the degree
MASTER OF SCIENCE
(Biochemistry and Molecular Biology)
August 2015
Ramya S Ankala
2
DEDICATION
I dedicate this dissertation to my husband, Partha without whom this would have never been
possible and for his constant love and support. My sister, Smitha for never leaving my side
and her positivity. Finally to coffee that gave me company through many long nights.
3
ACKNOWLEDGMENTS
I am truly grateful to my supervisor Dr. Ite Laird-Offringa for accepting me into her
lab and for her guidance. She is not only an excellent mentor in guiding with my lab work but
also has inspired me to maintain a perfect balance in life. I would like to thank Dr.
Weisenberger for his expertise and his precious time. I would also like to acknowledge my
gratitude to Dr. Tokes for accepting me into the master’s program and his constant support
and advice.
I owe an immense debt of gratitude to Evelyn Tran for her limitless devotion of her
time, patience, training and knowledge she passed on to me. Without her guidance, support
and faith she had in me, this would not have been possible. I would like to thank my fellow
lab mates for always being helpful, their insightful advice and comments made during lab
meetings.
I am very much grateful to both my father-in-law and mother-in-law for their undying
love, words of encouragement and financial support throughout my masters. I am thankful to
my parents, my family and my grandparents who taught me how to make right choices.
I cannot end without thanking my roommates, Yogitha Tammana and Renu
Kanakamedala who made these two years a fun and memorable ride. Last but not the least I
thank my best friends, Jasmitha Mareedu, Swetha Miryala and Srujana Matta for their love
and constantly inspiring me to be kind and positive.
4
TABLE OF CONTENTS
DEDICATION 2
ACKNOWLEDGEMENTS 3
TABLE OF CONTENTS 4
LIST OF TABLES 7
LIST OF FIGURES 8
ABBREVATIONS 10
ABSTRACT 12
1. INTRODUCTION 13
1.1. Lung Cancer 13
1.2. Lung Adenocarcinoma 14
1.3. Epigenetics 15
1.3.1. DNA Methylation and Gene Expression 16
1.3.2. Perturbation of DNA methylation in Cancer 17
1.4. Genome scale analysis of DNA methylation and mRNA expression in LUAD 19
1.4.1. Identification of Differentially Methylated Genes 19
1.4.2. Identification of Functional Relevance 20
1.4.3. Top Candidate Genes Identified Through Integrative Analysis 21
1.5. ATP-Binding Cassette, Subfamily A, Member 3 (ABCA3) 22
1.5.1. ABCA3-Normal Role and Function in Lung 23
1.5.2. Implications in Lung Disease 24
1.5.3 Recent Studies and Mouse Models of ABCA3 25
1.5.4. ABCA3 in other Cancers 27
1.6. Regulation of ABCA3 28
1.6.1. TTF-1/Nkx2.1 (Thyroid Transcription Factor-1)
Regulation of ABCA3 28
1.6.2. SREBPs (Serum Response Element Binding Proteins)
Regulation of ABCA3 29
1.6.3. STAT3 (Signal Transducer and Activator of Transcription)
Regulation of ABCA3 29
5
1.7. ABCA3 Locus in the Genome 30
1.8. ABC Family Genes with Clinical Relevance in Cancer 31
2. SPECIFIC AIMS 34
3. MATERIALS & METHODS 36
3.1. Cell Lines 36
3.2. Buffers and Solutions 36
3.3. Protein and RNA Assays 37
3.3.1. Protein Assay 37
3.3.2. Transient Transfection of Expression Constructs 37
3.3.3. Protein Extraction 37
3.3.4. Western Blot Analysis 38
3.3.5. RNA Extraction 39
3.3.6. cDNA Synthesis 39
3.3.7. qPCR Analysis 39
3.4. Cloning of ABCA17 putative promoter 40
3.4.1. PCR Amplification of ABCA17 Constructs 40
3.4.2. Plasmid and ABCA17 Constructs (C1 & C2) Digestion 41
3.4.3. Ligation of Plasmid and ABCA17 Constructs 41
3.4.4. PIR1 Bacterial Cell Transformation 41
3.4.5. Mini Prep for Positive Colonies and Test Digest 41
3.4.6. Construction of ABCA17 full length construct C3 42
3.4.7. Sequencing of the Positive Clone ABCA17 Constructs 42
3.4.8. Fixing Mutations for C1 & C2 Constructs 42
3.4.9. Maxi Culture of the Positive Clones 43
3.5. Methylation Assays 43
3.5.1. In Vitro DNA Methylation by MSssI 43
3.5.2. Luciferase Assays 43
6
4. RESULTS 43
4.1. Validation of Endogenous ABCA3 Protein Expression in Different Cell Lines 44
4.2 Overexpression of ABCA3 in HEK293T 45
4.2.1. ABCA3 Expression Constructs 45
4.2.2. Detection of ABCA3 Protein Overexpression by Western Blot Analysis 46
4.2.3. qPCR Analysis To Detect Overexpression at RNA Level 46
4.3. In-vitro Methylation and Promoter Studies on ABCA3 and ABCA17P Constructs 47
4.4. Screening and selection of lung cancer cell lines 49
4.5. Promoter Construct Luciferase Reporter Assays 49
4.5.1. Design & Development of ABCA3 and ABCA17P
Promoter Constructs 49
4.5.2. Validating In-vitro Methylated ABCA17P & ABCA3 Promoter
Constructs 51
4.6. Methylated ABCA17 promoter constructs show transcriptional activity 53
4.7. Mining of Reduced Representation Bisulfite Sequencing (RRBS) data on the
UCSC Genome Browser reveals a dichotomous DNA methylation pattern within
the ABCA3-ABCA17P CGI 54
4.8. CpG promoter in other clinically relevant ABC genes do not possess differential
methylation pattern 55
5. DISCUSSION AND FUTURE DIRECTIONS 57
6. BIBLIOGRAPHY 60
7
LIST OF TABLES
Table 1 ABC transporter family genes indicated in cancers
Table 2 List of cell lines used and their properties
Table 3 List of qPCR primer sequences used in qPCR analysis
Table 4 PCR primer sequences for ABCA17 constructs C1 and C2
Table 5 Antibodies used in the validation of endogenous ABCA3 protein expression
8
LIST OF FIGURES
Figure 1.1 Histopathology of lung cancer subtypes
Figure 1.2 Main anatomical features of the distal lung
Figure 1.3 Consequences of aberrant DNA methylation
Figure 1.4 Revised Knudson’s two-hit hypothesis
Figure 1.5 Volcano plot of the differential DNA methylation analysis
Figure 1.6 Starburst plot integrating differential DNA methylation and gene expression
analysis
Figure 1.7 Correlation DNA methylation and gene expression of ABCA3
Figure 1.8 Biosynthesis of pulmonary surfactant and protein structure of ABCA3
containing Extracellular Domains and Nucleotide Binding Domains
Figure 1.9 Surfactant homeostasis in normal lung versus RDS lung
Figure 1.10 Histopathology (A) and electron microscopy (B) of ABCA3 deficient lung
tissue with ILD
Figure 1.11 Histology of fetal lungs of mice littermates
Figure 1.12 EM analysis of ultrastructure of lungs from 6-month old Abca3
+/+
and
Abca3
+/-
mice littermates
Figure 1.13 Proposed model of transcriptional regulation on ABCA3 gene expression
Figure 1.14 UCSC genome browser snapshot of the ABCA3 & ABCA17 locus
Figure 4.1 Western blot analysis of ABCA3 using different antibodies in various cell
lines
Figure 4.2 PCMV6-Entry Vecotr
Figure 4.3 Transfection of HEK293T cells with ABCA3 expression construct using
Lipofectamine 2000
Figure 4.4 Overexpression of ABCA3 in HEK293T cells
9
Figure 4.5 Orientation of ABCA3 gene locus represented through UCSC genome
browser
Figure 4.6 qPCR analysis of ABCA17P expression over a range of cell lines and whole
lung RNA
Figure 4.7 Wiggle tracks for ABCA3 in lung cancer cell lines from RNA seq data analysis
Figure 4.8 Schematic representation of CpGL vector, design of ABCA3 and ABCA17P
construct and location of the constructs over ABCA3 gene locus
Figure 4.9 In vitro methylation workflow and the mechanism of CpG methylation using
MSssI CpG methyltransferase in the presence of S-Adenosyl Methionine
(SAM)
Figure 4.10 Gel electrophoresis image of NotI-digested methylated and unmethylated
ABCA3 and ABA17 promoter constructs.
Figure 4.11 Luciferase assay for methylated and unmethylated ABCA3 and ABCA17P
promoter constructs in H522, H2228 and H1944 cell lines
Figure 4.12 Reduced Representation Bisulfite-sequencing data from ENCODE cancer cell
lines and normal cell lines at the Abca3-ABCA17P locus
Figure 4.13 Promoter CGI’s of four different ABC gene loci
10
ABBREVATIONS
SCLC Small Cell Lung Cancer
NSCLC Non-Small Cell Lung Cancer
ADC Adenocarcinoma
ATII Alveolar Epithelial Type II
TSG Tumor Suppressor Genes
KRAS Kirsten Rat Sarcoma viral oncogene homolog
BRAF v-raf murine sarcoma viral oncogene homolog B
EGFR Epidermal Growth Factor Receptor
NF1 Neurofibromatosis type I
ALK Anaplastic Lymphoma Kinase
NTL Non Tumor Lung tissue
ABC ATP-binding cassette
ABCA3 ATP-binding cassette, subfamily A, member 3
PC Phosphotidyl Choline
PG Phosphotidyl Glycerol
RDS Respiratory Distress Syndrome
ILD Interstitial Lung Disease
SP Surfactant Protein
BPD Broncho Pulmonary dysplasia
MDR Multi Drug Resistance
AML Acute Myeloid Leukemia
TTF1/Nkx2.1 Thyroid Transcription Factor-1
SREBP Serum Response Element Binding Proteins
11
STAT Signal Transducer and Activator of Transcription
JAK Janus Kinase
IL-6 Interleukin-6
PI3K/AKT Phosphoinositide-3-Kinase/Protein Kinase B (AKT)
TF Transcription Factor
ABCA17P ATP-binding cassette, subfamily A, member 17 Psuedogene
MDR1/PGY1 Multi Drug Resistance-1/P-Glycoprotein-1
MXR Mitoxantrone Resistance half-transporter
MRP1 Multi Drug Resistance protein 1
MDR3 Multi Drug Resistance-3
ABCB ATP-binding cassette, sub-family B
ABCC ATP-binding cassette, sub-family C
ABCD ATP-binding cassette, sub-family D
ABCG ATP-binding cassette, sub-family G
rAT2 Rat AT2 cell lysate
rLL Rat lung lysate
hLL Human lung lysate
IRES Internal Ribosomal Entry Site
GFP Green Fluorescence Protein
qPCR Quantitative PCR
RPKM Reads Per Kilo base per Million reads
TSS Transcription Start Site
12
ABSTRACT
Lung cancer accounts for the majority of cancer deaths in the US and worldwide. Lung
adenocarcinoma (LUAD) is the most common histological subtype of non-small cell lung
cancer (NSCLC). Recent genome-wide profiling studies of LUAD tumors have identified
numerous genetic alterations, particularly those involved in the RTK/RAS/RAF pathway.
This had led to clinically relevant molecular subtyping of many LUAD cases, however a
significant number still have unidentified driver genes. Since epigenetic alterations also
underlie lung cancer development, a previous study by Selamat et al. performed a genome-
scale integrative analysis using DNA methylation and mRNA expression data to identify
potential epigenetic driver genes. Results from this study found ABCA3 to be one of the top
candidate genes that was hypermethylated and down-regulated. ABCA3, a lipid transporter
protein highly expressed in lung alveolar epithelial type 2 cells, plays an important role in
regulating lung surfactant homeostasis and lamellar body biogenesis. Infants with
homozygous ABCA3 mutations usually die at birth due to surfactant deficiency. Some ABCA3
mutations results in interstitial lung disease (ILD) and respiratory distress syndrome (RDS).
Interestingly, ABCA3 heterozygous knockout mice exhibit hyperplasia of the lungs. With such
compelling clinical relevance in lung function, we speculate ABCA3 to be a potential tumor
suppressor gene (TSG) which becomes epigenetically deregulated by aberrant methylation
of its promoter. In addition, the complex gene structure of the ABCA3 locus may offer some
fresh insight into how this critical gene is regulated transcriptionally. Overexpression and
knockdown studies are underway to address the function of ABCA3 in LUAD. To address
ABCA3 regulation, we have performed preliminary luciferase assays to interrogate the
transcriptional activity of different regions surrounding the ABCA3 promoter. Our data show
that methylation of the region -32 to +2842 bp relative to the ABCA3 transcription start site
results in higher reporter activity than the unmethylated counterpart. This region overlaps a
part of the ABCA3 CpG island which shows differential methylation between tumor and
normal cell lines based on reduced representation bisulfite sequencing (RRBS) data from
ENCODE. Results from this study may offer new insights into the role of ABCA3 in disease and
broaden our current understanding of differential methylation within promoter CpG islands.
13
1. INTRODUCTION
1.1. Lung Cancer:
Lung cancer is the leading cause of cancer deaths in the US among both men and
women
1,2
. According to the American Cancer Society as of 2015, there were about 221,200 new
cases and an estimated 158,040 deaths from lung cancer
3
. While tobacco smoking is by far the
leading risk factor for lung cancer – more than 80% of lung cancer deaths result from smoking –
exposure to other carcinogens such as asbestos and radon gas
3
, as well as air pollution
3,4
also pose
as significant risks for this cancer. In addition, studies within the last decade have implicated
genetics in lung cancer risk
5
. Based on these factors, 10-15% of lung cancer cases are found in
non-smokers
6
.
Lung cancer is often considered a ‘silent killer’
7
because it is typically diagnosed at a late
stage when the disease has advanced
3
. This has necessitated the implementation of at-risk
population screenings
8
and development of early detection tools
9
. However, even when
accounting for early and late stages of lung cancer, the overall 5-year survival rate is
approximately 17%
10
, lower than that for late stage breast cancer
11
. Thus, there is strong interest
in investigating the molecular etiology of lung cancer.
An Introduction to Lung Cancer by Dr. Weiss (2010) http://cancergrace.org/lung/2010/04/05/an-introduction-to-lung-cancer/
Fig 1.1: Histopathology of Lung Cancer subtypes. Lung Cancer subtypes are categorized mainly
based on the pathology of the lung tissue sample from biopsy. L to R: Lung adenocarcinoma,
squamous cell carcinoma, large cell carcinoma and small cell carcinoma.
Lung cancer is divided into two major categories: Small Cell Lung Cancer (SCLC) which
accounts for 10-15% of all lung cancers and Non-Small Cell Lung Cancer (NSCLC) which accounts
for majority of cases (85-90%). NSCLC occurs in both smokers and never smokers and is further
classified histologically into adenocarcinoma (~40%), squamous cell carcinoma (~30%) and
large cell carcinoma (~10%). The remaining 5% comprise other carcinoids and neuroendocrine
tumors (Fig 1.1)
3
.
14
1.2. Lung Adenocarcinoma:
Lung adenocarcinoma (LUAD) is the most common lung cancer subtype among non-
smokers and accounts for the majority of lung carcinomas in women
12
. LUAD typically arises in
the distal part of the lung in alveoli sacs found at the terminal ends of branching bronchioles
where gas exchange occurs
13
(Fig 1.2). LUAD tumors are highly histologically heterogeneous,
exhibiting diverse growth patterns, cellular morphologies, and invasiveness based on
hematoxylin and eosin (H&E) and immunohistochemical stainings
4
. As a result of this
heterogeneity, a multidisciplinary effort has been made to systematically classify LUAD tumors to
aid in therapeutic decisions and improve molecular and prognostic correlations
4
. The main
classes of LUAD are atypical adenomatous hyperplasia (AAH), adenocarcinoma in situ (AIS),
minimally invasive adenocarcinoma (MIA) and invasive adenocarcinomas
4
.
Junquiera's Basic Histology: Text and atlas, 12th Edition
Fig 1.2: Main anatomical features of the distal lung. LUAD is thought to arise mainly occurs from
the alveolar epithelial type II cells (ATII cells) located in the alveoli of the lungs.
Much like their histology, LUAD tumors exhibit complex genomic heterogeneity.
Advances in next-generation sequencing technologies and other genome-wide profiling
15
platforms have revealed critical molecular features of LUAD
15
. Since the initial discovery of
KRAS and EGFR mutations in 1984
16
and 2004
17
, respectively, hundreds of genetic mutation
profiles have been generated from primary human LUAD samples, identifying recurrent
mutations, rearrangements, and amplifications in oncogenes such as HER-2, ROS-1, and
MET
15,18,19
, as well as loss-of-function mutations and deletions in tumor suppressor genes
such as TP53, STK11, RB1, NF1, and CDKN2A
20
. In addition, by massive parallel sequencing
of 183 LUAD tumors and matched normal tissues, Imielinski et al. were able to identify novel
mutations in the U2AF1 splicing factor that were significantly associated with reduced
survival; mutations in RBM10 RNA-binding protein and ARID1A chromatin-remodeling
protein were also significantly enriched in a fraction of cases
21
. Many groups working on
NSCLC have identified that lung ADC is majorly associated with genetic mutations in genes
such as KRAS, BRAF, EGFR, NF1, genes involved in RTK/RAS/RAF pathway and ALK fusions
8
.
These and other genome-wide profiling studies have aided in the molecular subtyping of a
large proportion of LUAD cases, allowing clinicians to determine which patients will benefit
from a particular targeted therapy
18
. However, three obstacles to treatment remain: 1) the
number of actionable driver genes is limited to currently available drugs, 2) the identification
of a driver does not guarantee it can be easily exploited therapeutically and 3) mutational
analyses of a subset of LUAD cases have not revealed obvious genetic drivers. Possible
solutions to these problems may lie in not only sequencing more LUAD/non-tumor pairs, but
also in integrating multiple layers of regulatory information. Thus, researchers are looking to
the epigenome as a potential key to unlocking the complex molecular events underlying
tumorigenesis. Adenocarcinoma tumors are heterogeneous with a wide range of molecular
alterations that respond differently to different treatments. Though general clinical tumor
diagnosis is based on these histological features, for personalized therapeutic decisions,
molecular subtyping is crucial. Next generation sequencing and other high-throughput
genomic profiling platforms help identifying clinically relevant molecular aberrations within
lung tumors
7
.
1.3. Epigenetics:
Epigenetics is the study of heritable regulatory information layered “on top” of or
“above” (epi-) our genome. The epigenome refers to the collection of modifications to the
DNA which do not alter the underlying genetic code. These modifications are divided into
four main categories: DNA methylation, covalent histone modifications, non-covalent
16
mechanisms such as incorporation of histone variants and nucleosome remodeling, and non-
coding RNA
22
. Coordination among these modifications is crucial for proper gene expression
during development and in fully matured organisms. For example, the establishment of
asymmetric chromatin landscapes in paternal and maternal genomes in the early zygote is
important for timely activation of key pluripotency genes, disruption of which impairs
development
23
. In the adult lung, for instance, perturbation of histone acetylation has been
linked to disease severity of chronic obstructive pulmonary disease (COPD)
24
. Based on these
studies, it is clear that in order to understand normal and disease states, epigenetic
mechanisms must be investigated. In present study, we will focus on the most extensively
studied epigenetic modification – DNA methylation. The following sections will discuss its
effect on gene expression, its role in cancer in general, and a recent study from the Laird-
Offringa lab investigating its changes in LUAD.
1.3.1. DNA Methylation and Gene Expression:
DNA methylation is the addition of a methyl group (–CH3) on the 5-position
of the cytosine ring (5-mC) in a CpG dinucleotide. It is catalyzed by DNA methyltransferases
(DNMTs) DNMT1, DNMT3a and DNMT3b using S-adenosyl-l-methionine (SAM) as a methyl
group donor
25
. DNMT1 is the main DNA methyltransferase which maintains methylation
patterns during DNA replication; DNMT3a and 3b are de novo DNA methyltransferases that
function independently of replication
26
. Ample studies have demonstrated the importance of
DNA methylation in that normal cellular physiology and homeostasis as evidenced by the
early lethality of mice lacking DNMTs
28
. DNA methylation is also important for X-
chromosome inactivation, gene silencing, and genomic stability
29
.
The relationship between DNA methylation and gene expression has been well-
studied, particularly at gene promoters. The majority of the mammalian genome is CpG poor
except for regions of high CpG density called CpG islands (CGIs), which are found at the 5’ end
of all constitutively expressed genes and ~40% of tissue-specific genes
29 ,30
. In normal cells,
CGIs are unmethylated, except for ~6% of CGIs that become methylated in a tissue-specific
manner during development or differentiation
31
; the remaining genome is highly
methylated
26,27,29,32
. Promoter CGI methylation can suppress transcriptional activity by either
directly inhibiting binding of sequence-specific transcription factors (TFs) or by mediating
binding of methyl-CpG binding proteins (MBDs) to induce chromatin compaction
22,26
. A
17
caveat to these mechanisms is that DNA methylation is now thought be a secondary event to
changes in the histone code: nucleosome positioning is considered the epigenetic “door”
which “closes” to initiate gene silencing and methylation is thought to be the “lock” which
maintains the repressed state
33
. When promoter CGIs are unmethylated, they are
transcriptionally permissive, and serve as TF ‘‘landing lights”
30
. However, the absence of CGI
methylation does not necessarily indicate active transcription; most CGIs are unmethylated
regardless of gene expression. Studies in macrophages suggest that this may be due to an
inherent chromatin structure that predisposes CGIs toward promoter activity
29
. It has also
been proposed that short, abortive transcripts generated by RNA polymerase II at CGI
promoters protect CGIs from DNA methyltransferases, thereby preventing methylation
29
.
Alternatively, H3K4me3, a signature chromatin mark at CGIs, may interfere with DNA
methyltransferase activity
34
. The complex interplay between DNA methylation and the
histone code in regulating gene expression underscores the importance of methylation in the
normal cell state.
1.3.2. Perturbation of DNA methylation in Cancer:
Cancer is a multifaceted disease initiated by both genetic and epigenetic events.
Numerous studies have repeatedly found aberrant DNA methylation patterns in cancer cell
lines and primary cancer tissues compared to normal cells and tissues. In a variety of cancer
specimen, locus-specific DNA hypermethylation and global DNA hypomethylation have been
observed
26
, suggesting that silencing of tumor-suppressor genes, overexpression of proto-
oncogenes, and reactivation of transposable elements
35
are important features of cancer. For
example, some colorectal carcinomas exhibiting microsatellite instability also showed
promoter hypermethylation of the mismatch repair gene hMLH1. Treatment of colorectal cell
lines containing hypermethylated hMLH1 with 5’-aza-2’-deoxycytidine, a demethylting agent,
restored hMLH1 expression and mismatch repair activity, indicating that hypermethylation
of hMLH1 was the primary inactivating event
36
. In lung adenocarcinoma, hypomethylation
of the long nuclear interspersed element 1 (LINE-1) family of transposons was statistically
significantly associated with higher histological grade and poor prognosis
37
.
18
DNA methylation can also contribute to cancer development by causing loss of
heterozygosity (LOH). LOH occurs when a particular gene locus heterozygous for a
deleterious mutation acquires an additional alteration at the remaining wild-type allele,
resulting in inactivation of that gene
38
. Since promoter methylation can be associated with
transcriptional silencing, this epigenetic modification can therefore act as the additional “hit”
in cancer, according to Knudson’s “two-hit” hypothesis
39
(Fig 1.4)
Modified from Illumina Sequencing Methods: Field guide to Methylation methods
http://www.illumina.com/content/dam/illumina-marketing/documents/products/other/field_guide_methylation.pdf
Fig 1.3: Consequences of aberrant DNA methylation. Promoter methylation can have a
significant effect on gene expression. In euchromatin, promoter hypomethylation exists and
the genes are expressed while in heterochromatin promoters are hypermethylated and the
genes are silenced. In cancer, global hypomethylation and promoter hypermethylation can
alter the expression patterns of important genes involved in cancer.
19
Peter A. Jones and Peter W. Laird. Nature genetics, Vol 21 (1999), pp. 163-167-Figure 2
Fig.1.4: Revised Knudson’s two-hit hypothesis. The two green boxes shown at the top are the two
active alleles of a TSG. The first hit for gene inactivation shown is a localized mutation on the left
or by transcriptional repression by DNA methylation on the right. The second hit is shown by
either LOH or DNA methylation that ultimately causes transcriptional silencing.
1.4. Genome scale analysis of DNA methylation and mRNA expression in LUAD:
As mentioned earlier, despite numerous genome-wide profiling studies, few additional driver
genes have been identified in LUAD tumors. However, DNA methylation-based profiling is a
promising approach, as it has identified epigenetic subtypes in several cancers
40
. In the following
sections, a study by Selamat et al.
41
(2012) using genome-scale analyses of methylation and
expression data from LUAD tumors will be discussed. The goal of the study was to identify DNA
methylation events with potential functional significance.
The authors profiled a sample set of 59 microdissected matched LUAD and adjacent non-tumor
lung (agjNTL) tissues from both smokers (29) and non-smokers (30) using the Illumina Infinium
HumanMethylation27 platform for methylation profiling and the Illumina Human WG-6 v3.0
Expression BeadChips for expression profiling
1.4.1. Identification of differentially methylated genes:
Two-dimensional hierarchical clustering of the top 5000 probes most variable across the
data set was performed on 117 samples (tumor and AdjNTL for which data passed quality
20
control). Substantial differences in the overall DNA methylation profiles of tumor and AdjNTL
were observed. Analysis of probe intensities between tumor and non-tumor samples using a false
discover rate (FDR) of Q<0.05 and a minimum median β-value difference of 20% resulted in 681
probes (520 genes) statistically significantly hypermethylated in tumors and 275 probes (247
genes) statistically significantly hypomethylated in tumor (Fig 1.5). These observations support
the fact that significant differences in DNA methylation are seen between tumors and AdjNTL.
Selamat et al. Genome Res, Vol 22 (2012), pp. 1197-1211
Fig 1.5: Volcano plot of the differential DNA methylation analysis.
X-axis: Median β-value difference (median Tumor-median NTL); Y-axis: Q-values for each
probe (-1 x log10 scale). The vertical dotted lines mark 20% change in β-values; the
horizontal dotted line marks the significance cut-off.
1.4.2. Identification of functional relevance:
Although DNA methylation differences have been identified, not all of these changes
are necessary driver events; in most cases, they are passenger alterations which have
occurred as a result of cell derangement. In order to determine whether the identified DNA
methylation changes had functional consequences, the authors integrated mRNA expression
data from the same set of matched LUAD tumor/non-tumor pairs. Out of the 349 differentially
expressed genes, 164 of these were statistically significantly hypermethylated and
downregulated (23%), while 57 genes were significantly hypomethylated and upregulated
21
(BH adjusted p-value of 0.05). The starburst plot below divides the genes into four categories
(Fig.1.6):
i. Genes which are hypermethylated and downregulated in tumors (red)
ii. Genes which are hypomethylated and upregulated in tumors (green)
iii. Genes which are hypermethylated and upregulated in tumors (blue)
iv. Genes which are hypomethylated and down-regulated in tumors (orange)
Results from this integrative analysis suggests that differential methylation could have
consequences on the expression of important genes either through activation or repression.
Selamat et al. Genome Res, Vol 22 (2012), pp. 1197-1211
Fig 1.6: Starburst plot integrating differential DNA methylation and gene expression analysis.
X-axis: DNA methylation Q-values; Y-axis: BH adjusted P-values (Scale: (-1 x log 10))
1.4.3. Top candidate genes identified through integrative analysis:
To identify the top changing genes, a two-fold cutoff of average gene expression was
applied. A total of 45 genes were identified to be coordinately hypermethylated and
downregulated in tumors, and 16 genes were coordinately hypomethylated and upregulated.
DNA methylation status between tissues from smokers and never-smokers were also
examined through correlation analyses, which showed very similar DNA methylation profiles
22
Interestingly, the correlation between smoking status and methylation change was higher
(though not significant) for NTL.
The correlation plot shown in Fig.1.7 corresponds to one of the top hypermethylated and
downregulated candidate genes, ABCA3 (ATP-binding cassette, subfamily A, member 3: ATP-
binding cassette transporter) which has a critical role in lung development and surfactant
metabolism in humans
21
.
Selamat et al. Genome Res, Vol 22 (2012), pp. 1197-1211
Fig. 1.7: Correlation DNA methylation and gene expression of ABCA3 in LUAD.
X-axis: DNA methylation; Y-axis: Gene expression in tumors and normal tissues
T-Tumor (Red); N-Normal Tissue (Blue)
Results from this integration analysis identified a novel candidate gene ABCA3
which is epigenetically deregulated. It is the top candidate gene that was hypermethylated
and downregulated. The probe used for identifying ABCA3 falls within the CGI that spans
the promoter region of ABCA3. This information fits very well into the classical paradigm of
DNA methylation regulation. The study lays the groundwork for us to investigate the role
and involvement of ABCA3 in LUAD development and progression as a candidate driver
gene.
1.5. ATP-Binding Cassette, Subfamily A, Member 3 (ABCA3):
Transport of different substrates are fundamentally mediated by one of the four
classes four of membrane-bound transport proteins: ion channels; transporters; aquaporins;
and ATP-powered pumps
42
. The ATP-binding cassette (ABC) superfamily, an example of ATP-
powered pumps are ubiquitous set of membrane transporter proteins involved in
23
transporting a diverse set of substrates across membranes as well as influx or efflux of
cells
42,43
.
1.5.1. ABCA3-Normal Role and Function in Lung:
As one of the genes belonging to the ABC subfamily A, ABCA3 is involved in regulation of
membrane trafficking, surfactant production, and transportation of lipids such as
phosphotidyl choline and cholesterol. Predominantly expressed in the lung, ABCA3 can also
be found in liver, stomach, kidney, adrenal, pancreas, trachea, and brain
44
. ABCA3 is localized
to the limiting membrane of lamellar bodies of type II cells of the lung and mediates in
transportation of lipids and other fats into it the lamellar body. There the lipids which interact
with surfactant proteins and to produce form pulmonary surfactants which then are
exocytosed to the alveolar cavity (Fig.1.8). Besides surfactant production, ABCA3 protein also
appears to be involved in the formation of normal lamellar bodies.
L: Junquiera's Basic Histology: Text and atlas, 12
th
Edition (www.accessmedicine.com), R: Modified from Bruder et al. Modern Pathology (2007)
20, 1009–1018- Figure 5
Fig. 1.8: Biosynthesis of pulmonary surfactant and protein structure of ABCA3 containing
Extracellular Domains and Nucleotide Binding Domains. R: ABCA3 role in surfactant
production through lamellar bodies. L: Protein structure of the membrane protein located at
the limiting membrane of lamellar bodies. It contains ECD and NCD that help in binding to
different substrates.
ABCA3 expression is increased prior to birth and is hormonally induced at the same
time as the surfactant proteins A–D suggesting developmental regulation
45
. Studies also
24
suggest that ABCA3 intermembrane transport of phosphatidylcholine (PC) and
phosphatidylglycerol (PG) is significantly elevated in type II epithelial cells relative to whole
lung. In addition, transfer activity increases commensurate with surfactant synthesis before
birth
46
.
1.5.2. Implications in Lung Disease:
Mutations in ABCA3 are profoundly associated with neonatal respiratory disorders.
ABCA3 deficiency appears be the most common cause of genetic surfactant dysfunction in
humans that can present at any age and may lead to death. Homozygous or compound
heterozygous ABCA3 mutations causing fatal surfactant deficiency in newborns can be
categorized as type I for abnormal intracellular trafficking and type II for decreased ATP
hydrolysis activity. More than 150 different variants of ABCA3 mutants have been
documented and are broadly involved in two clinically important neonatal lung diseases:
Respiratory Distress Syndrome (RDS) and Interstitial Lung Disease (ILD).
Respiratory Distress Syndrome (RDS): The multilayered lipid rich coating in the alveoli is
primarily composed of four different types of surfactant proteins: hydrophilic proteins SP-A
and SP-D and the hydrophobic proteins SP-B and SP-C. These are is produced in the
endoplasmic reticulum of the pulmonary type II alveolar cells and stored in the lamellar
bodies (Fig.1.9). Surfactants play a crucial role in lowering and stabilizing the alveolar
surface tension and preventing end expiratory collapse
46
. ABCA3 mutations result in the
influx of abnormal lipids and aberrant formation of surfactants causing RDS. It is a severe
defect as the infants will not survive beyond 3 to 6 months of life and the only curative
treatment is lung transplantation.
25
Kong Chen & Jay K Kolls. Nature Medicine, Vol 16 (2010), pp. 1078–1079-Figure 1
Fig 1.9: Surfactant homeostasis in normal versus RDS lung. Pulmonary surfactant in green.
Interstitial Lung Disease (ILD): ILD refers to a heterogeneous group of disorders that affects
the interstitium mainly concerning the alveolar epithelium of the lung. It occurs both in
neonates and in adults mainly differing in the course and prognoses of the disease
47
. Unlike
RDS, surfactant deficiency is observed with a milder phenotype mainly causing interstitial
thickening. Studies that did electron microscopic investigation on ABCA3 mutations
identified abnormal lamellar body pattern with tightly packed concentric membranes and
distinctive electron dense aggregates of 'fried-egg' appearance (Fig.1.10)
48
.
Fig 1.10: Histopathology (A) and electron microscopy (B) of ABCA3
deficient lung tissue with ILD
1.5.3 Recent Studies and Mouse Models of ABCA3:
Although the exact role of ABCA3 in surfactant metabolism is not known, many
studies support its role in the surfactant lipid transport
28
. With the findings that ABCA3
26
deficiency leads to severe lung diseases it remains a candidate gene to be investigated further.
Lawrence et al. identified high levels of ABCA3 expression in injured lung and regenerating
epithelial tubules in infants with broncho pulmonary dysplasia (BPD)
49,50
. Alveolar septal
thickening was observed in histopathology samples of alveoli by Baughman and colleagues.
The alveolar septum was uniform with presence of type II cell hyperplasia and filled with
granular proteins
32
. Following investigations performed by Inagaki et al. have firmly
established the importance of ABCA3 in the formation, maturation and overall lamellar body
biogenesis
51
.
Recently, Ban et al.
51
and Fitzgerald et al.
13
generated Abca3 null (Abca3-/-) mice that died
within an hour after birth due to respiratory failure. An important investigation that lead to
the finding that that ABCA3 is crucial for lamellar body biogenesis (Fig 1.12), surfactant
protein-B processing and overall lung development was done by Cheong et al.
52
The study
involved looking at the effects of homozygous Abca3
-/-
and heterozygous Abca3
+/-
knock-out
mice obtained through targeted disruption of the ABCA3 gene
21
. The most interesting result
as seen in Fig.1.11, the homozygous Abca3 -/- knockout mice exhibited regions of hyperplasia
in their lungs. This paved the path in hypothesizing a potential role of ABCA3 in lung
adenocarcinoma. The study also involved EM analysis of the ultrastructure of lamellar bodies
was also done in the Abca3
-/-
knockout mice which suggested ABCA3 is important for LB
formation. Also, mature and non-reduced levels of SP-B was significantly decreased in the
Abca3
-/-
knockout mice when compared with Abca3+/- and Abca3+/+ mice.
Cheong et al. J. Biol. Chem. 2007, 282:23811-23817
Fig 1.11: Histology of fetal lungs of mice littermates. Wild type mice showed normal alveolar
spaces (left) while homozygous knockouts (right) had uniformly thickened septa.
Heterozygous mice (middle) showed a variable morphology with some areas of thickening,
suggesting the wild type copy of ABCA3 could become silenced, perhaps by methylation.
27
Cheong et al. J. Biol. Chem. 2007, 282:23811-23817
Fig 1.12: EM analysis of ultrastructure of lungs from 6-month old Abca3+/+ and Abca3+/-
mice littermates. Pointed arrow indicates the difference in lamellar body biogenesis between
ABCA3 +/+ and ABCA3 +/- fetal lungs of mice littermates.
1.5.4. ABCA3 in other Cancers:
ABCA3 has been directly implicated in being involved in Acute Myeloid Leukemia
(AML). In contrast to the genome analysis of lung adenocarcinoma where ABCA3 expression
is decreased it is over expressed in AML. Overexpression of ABCA3 results in subcellular drug
sequestration to lysosomes thus inducing a multidrug resistance phenotype. Several studies
gathered evidence on ABCA3 expression in all major types of malignant
lymphohematopoietic diseases. It is speculated that overexpression of ABCA3 impacts
cellular detoxification of all drugs indicating its potential in protecting cells against the toxic
effects from an array of cytotoxic drugs
55,56
.
Similar to type II alveolar epithelial cells of the lung, mammary gland epithelial cells
show elevated rate of lipid secretion and transport mediated by ATP binding cassette
transporters. Schimanski et al. compared protein expression patterns of ABCA3 between
normal and breast cancer samples. Differential expression patterns were observed with
significantly diminished levels of ABCA3 in the breast cancer samples. Through multivariate
analysis, low expression of ABCA3 was identified to be an adverse risk factor for tumor
recurrence and seems to be associated with poor prognosis
56
.
Inferring from the genome scale DNA methylation analysis that ABCA3 is a top
candidate in being hyper methylated and down-regulated gene, and considering its clinical
relevance in lung diseases, observed phenotypes of the knockout mice and direct
involvement in different types of cancers, we hypothesized that ABCA3 loss of function
28
through epigenetic deregulation might be implicated in either development or progression
of lung adenocarcinoma.
1.6. Regulation of ABCA3:
Ultrastructural analysis of the Abca3-/- lungs of mice reported loss of mature lamellar
bodies and surfactant deficiency in the alveolar space. Similar observations were made in
human infants with varied ABCA3 mutations. Although very little is known about the
regulation mouse ABCA3 (mABCA3) mRNA dramatically increases in the developing fetal
lung. Generally, the last trimester of gestation is crucial for pulmonary surfactant maturation
which is correlated with an increased expression of surfactant proteins and surfactant
phospholipids. Such coordination is crucial for packaging of surfactant proteins and lipids
together which is essential for the transition to air breathing at birth
57
.
Regulation of surfactant protein expression in respiratory epithelial cells is
dependent on a number of transcription factors which are less well-understood. Studies on
mice suggest that the members of ABC family of lipid transport proteins are regulated either
directly or indirectly by SREBPs
58,59
, TTF-1 or Nkx2.1
58,61
, STAT3
60
, FOXA2, GATA-6, NFATc3,
and C/EBPa
57
. Potential DNA binding sites for these transcription factors, which influence
lung developmental and gene expression, are present at the distal promoter region of the
ABCA3 gene. The upstream region of ABCA3 consists of several CpG islands and lacks a TATA
box. This supports the possibility that epigenetic deregulation of CGI at the promoter region
of ABCA3 can affect the binding of important factors and cause changes in its expression
levels and results in downstream pathway alterations.
1.6.1. Thyroid transcription factor-1 (TTF-1/Nkx2.1/TITF1) Regulation of ABCA3:
TTF-1/Nkx2.1 is a homeodomain-containing transcription factor important in
regulating genes involved in morphogenesis and differentiation of the thyroid, lung and
ventral forebrain. About 85% of lung adenocarcinomas bear this factor as a marker. In the
lung, Nkx2.1 essentially controls the expression of SP-A, SP-B, SP-C, CCSP (Clara cell secretory
protein) and ABCA3 genes. Whitsett et al. conducted promoter studies to investigate the
potential regulatory elements in the mABCA3. The distal 5` upstream region (bp -2591 to -
1102) of ABCA3 gene serves as a binding site for TTF1/Nkx 2.1.
29
1.6.2. SREBPs (Serum Response Element Binding Proteins) Regulation of ABCA3:
SREBPs, a family of transcription factors is known to influence lipid homeostasis in
the lung. Potential serum response element sequences were detected in a proximal (bp -1102
to +11) region of the mABCA3 promoter. The mammalian genome encodes three SREBP
isoforms: SREBP-1a, SREBP-1c, and SREBP-2. The proximal region of the ABCA3 distinct
promoter region regulates cell selective and lipid-sensitive responses and is also the site for
SREBP-1c binding and getting activated by SCAP (SREBP cleavage activating protein). In vivo
studies involving deletion of SCAP in respiratory epithelial cells inhibited SREBP and
mABCA3 mRNAs indicating regulatory regions of the mABCA3 gene mediate tissue-selective
and lipid-sensitive regulation to influence surfactant homeostasis.
1.6.3. STAT3 (Signal Transducer and Activator of Transcription) Regulation of ABCA3:
Another study by Whitsett et al. demonstrated that mABCA3 mRNA levels were
significantly decreased upon deletion of STAT3 in the respiratory epithelium. The STAT3 -/-
mice resulted in lamellar body functional abnormalities and intracellular site of surfactant
lipid storage (Fig.1.13). The study proposed a model emphasizing the role of STAT3 in the
regulation of mABCA3. STAT3 which is induced by IL-6 and activated via the JAK pathway is
important in turn enhances SREBP1c to activate Abca3 gene expression which is involved in
surfactant homeostasis
62
.
Modefied from Whitsett et al. Am J Respir Cell Mol Biol Vol 38 (2008), pp 551–558
Fig 1.13: Proposed model of transcriptional regulation on ABCA3 gene expression. Loss of
function of ABCA3 will lead to disruption of lung surfactant homeostasis.
30
From these promoter studies and knockout mouse models, it is evident that the CGI
promoter region plays an important role in regulating ABCA3 gene expression. The CpG island
also contains potential binding sequences recognized by TF’s important for tissue specific
surfactant production. Given that the sequence similarity of mouse and human ABCA3 is high, it
can be predicted that the human ABCA3 promoter region also holds the same regulatory
properties as in mice and the CGI could potentially play a role in surfactant homeostasis.
1.7. ABCA3 Locus in the Genome:
The novel human ABCA3 transporter gene is located on the chromosome 16p13.3. The
gene provides instructions for making a membrane transporter protein involved in surfactant
production and is ubiquitously expressed in lungs, glandular tissues and the heart. Klugbauer and
Hofmann initially isolated ABCA3 cDNA clones and identified that the expressed protein is a 1704-
amino acid polypeptide containing two homologous repeats, each with 6 putative transmembrane
helices and an ATP-binding cassette motif
63
. Connors et al.
64
and Yamano et al.
65
independently
identified highest ABCA3 expression occurs in the lung. Mulugeta et al. observed ABCA3 mRNA in
lung ATII cells of human, rat, and mouse, and in various rat tissues. The protein is highly conserved
among these species and the expression was highest in lungs before birth.
As the ATP-binding cassette transporter genes represent the largest family of
transporters containing ABC genes, and are divided into seven subfamilies based on amino acid
sequence similarities and phylogeny. As they are involved in genetic diseases and other
complications, it is important to understand the gene evolution in order to develop animal models
and perform functional studies. The most notable expansion involving multiple gene duplication
and deletion events were identified in the ABCA3-like genes – ABCA14, ABCA15 and ABCA16.
These genes occur in mouse, rat and dog genomes but are absent in humans
66
.
Ban et al.
51
made ABCA3 -/- knockout mice that contributed significant detail regarding
the functionality and the unique gene structure of ABCA3. It was identified that ABCA17 is present
in rodent genomes and is highly expressed in mice testes where it is a sperm-specific transporter
regulating intracellular lipid metabolism
46
. Later, Piehler et al. performed rodent studies on
ABCA17 and noted interesting facts in comparison with the human orthologue of ABCA17
66
. The
study reported that the human ABCA17 is a transcribed non processed pseudogene while the
rodent ABCA17 gene codes for a 1733 amino acid transporter protein. Human ABCA17P
nucleotide sequence displays 58% identity with mouse and 51% identity with rat with relatively
low overall sequence homologies. But when the exons were compared, the UCSC BLAT genome
browser analysis reveals significantly higher homologies. This information supports the fact that
31
human ABCA17P was once transcriptionally active. ABCA17P locus (Fig.1.14) on the
chromosome revealed that it borders ABCA3 but runs in the opposite direction, suggesting that
both the genes originated by an ancestral gene duplication. Such orientation of a pseudogene
occurring next to its functional gene is unique to human ABCA3-ABCA17P locus and has not been
reported for any other genes.
The ABCA3 locus analyzed in the UCSC genome browser emphasized some distinguishing
characteristics with respect to its orientation with its adjacent pseudogene ABCA17P. It is
interesting to note that the genes run in a head to head orientation along the genome. Another
important feature of this locus is that the 5`ends of both the genes share a common CpG island
(Fig 1.14). ABCA3-ABCA17P locus is also one of the first indications that a progenitor gene and its
pseudogene can overlap.
Fig 1.14: UCSC genome browser snapshot of the ABCA3 & ABCA17 locus. ABCA3 and ABCA17P
run in opposite direction along the genome. The genes share a common CpG island (-1040 to
+1417 in respect to ABCA3 TSS).
1.8. ABC family genes with clinical relevance in cancer:
The involvement of many ABC transporters in genetic diseases has directed a lot of
attention to these genes. Because the transporters play an important role in the efflux of a
wide variety of substrates, results from the loss of function or absence of the gene are
diverse
42
.
32
Table 1: ABC transporter family genes indicated in cancers
Many ABC genes are extensively characterized to play a role in the drug resistance of
malignant cells, host detoxification and protection against xenobiotic substances
47
. Some
important genes that are involved in cancer are listed in the Table 1.
The expression levels of these proteins in humans have important consequences for
an individual’s susceptibility to certain drug-induced side effects, interactions, and treatment
efficacy. With respect to regulatory mechanisms, it is interesting to look for any ABC genes
that are influenced by epigenetic regulation similar to ABCA3, which was hyper methylated
and down-regulated from the analyses conducted by Selamat et al.
Epigenetic regulation mechanisms for MDR1/ABCB1 and ABCB4 genes have been
discovered. Henrique et al. investigated epigenetically silenced MDR1 promoter and its effect
on tumor growth
68
. MDR1 promoter composed of two promoters, a major
downstream/proximal (DSP) and a minor upstream (USP), encompasses three CpG islands.
The transcriptional activity is regulated both by DNA methylation and histone modification,
which when altered, result in varied levels of hyperplasia in the prostate tissue
48
. Aberrant
DNA methylation of the CGI spanning the ABCB1 promoter region was identified in 39% of
primary lung cancer and in 20% of head and neck cancer tissues. Savai & Dammann et al.
reported overexpression of ABCB4 significantly suppressed colony formation and
proliferation of lung cancer cells
69
.
33
Because the ABCA3 locus includes part of the ABCA17P pseudogene (only 175 bp
distance between exon 1 of each gene), , it can be informative to know about existing ABC
genes that have similar gene structures or have nearby pseudogenes. At least nine
pseudogenes have been detected in the ABC family. These pseudogenes are a result of either
partial or full duplication, insertions, or deletions of various sizes. ABCC6, ABCC13 and
ABCD1 pseudogenes, although functional in some organisms like rodents and mice, are no
longer functional in humans due to acquired mutations
23
.
Based on the findings that ABCA3 is significantly hypermethylated and
downregulated in a subset of human LUAD tumors and that ABCA3 heterozygous knockout
mice exhibit hyperplasia, we hypothesize that ABCA3 may be a potential tumor suppressor
gene. In this study, we propose to test the functional relevance of perturbing ABCA3
expression in LUAD cell lines and to investigate in detail the regulation of ABCA3 expression
by DNA methylation.
34
2. SPECIFIC AIMS
LUAD, which is the most common histological subtype of non-small cell lung cancer
(NSCLC), is a heterogeneous and complex disease. With the advent of high-throughput
genomic profiling platforms, identifying driver gene mutations such as those in EGFR, BRAF,
ROS1 and others within tumors is now possible and is being used for molecular subtyping of
NSCLC. The role of epigenetics in cancer has become well-appreciated; therefore it is likely
that some of the yet unidentified drivers may be epigenetically altered rendering molecular
subtyping by mutation status alone insufficient. A previously published study by Selemat et
al. identified candidate epigenetic driver genes by integrative analysis of DNA methylation
and mRNA expression data. Through this analysis, ABCA3 was identified as a candidate driver
gene that is hypermethylated and downregulated in LUAD. Loss of function of ABCA3 is
known to cause surfactant deficiency leading to various lung diseases. However it has not yet
been implicated in lung cancer.
In this study, we aim to determine whether ABCA3 is a potential TSG involved in
development and progression of LUAD. In parallel we will investigate the effects of DNA
methylation on transcriptional regulation of ABCA3
AIM 1: To determine whether ABCA3 has a functional relevance using LUAD cell lines.
Because ABCA3 is a potential TSG, I expect that overexpression of ABCA3 in cancer cell lines
lacking its expression will decrease the cancer cells’ transformed phenotype. In contrast, by
knocking down ABCA3 in cells expressing it, I expect to see exacerbation of the cancer
phenotype, for example increased cell proliferation. The effect of ABCA3 in these cell lines
will be measured by soft agar assay, cell proliferation assay, and cell migration/invasion
assays.
AIM 2: To determine the effect of DNA methylation on transcription at the ABCA3 locus.
The ABCA3 locus contains not only the ABCA3 gene, but also its neighboring pseudogene
ABCA17P. The first exon of ABCA17P lies in close proximity to exon1 of ABCA3: 176 bp
upstream of the ABCA3 transcription start site. Thus, the ABCA17P promoter region will also
be considered in this aim. Constructs spanning different regions of the promoter CGI of
ABCA3 and ABCA17P will be cloned into a CpG-less, promoter-less vector and subjected to
35
in-vitro methylation. A luciferase reporter assay will then be performed on both methylated
and unmethylated constructs to measure transcriptional activity driven by these regions.
Results from these experiments may determine whether CGI methylation of ABCA17P plays
a role in the epigenetic regulation of ABCA3.
36
3. MATERIALS & METHODS
3.1. Cell Lines:
All cells were obtained from ATCC and were cultured at 37ºC and 5% CO 2 atmosphere.
To passage these cell lines, cells were washed with sterile PBS, trypsinized and split in 1:5 ratios
(Table 2).
Cell Lines Tissue Disease Culture Medium Growth
Properties
H522 Lung LuAD-Stage 2 RPMI with 10% FBS
and 1% P/S
Adherent
H2228 Lung LuAD RPMI with 10% FBS
and 1% P/S
Adherent
H1944 Lung (derived from
metastatic site)
LuAD-Stage 3B RPMI with 10% FBS
and 1% P/S
Adherent
H23 Lung LuAD RPMI with 10% FBS
and 1% P/S
Adherent
HEK 293T Fetal Kidney Normal DMEM with 10% FBS
and 1% P/S
Adherent
MCF7 Breast (derived from
metastatic site)
Breast Cancer DMEM with 10% FBS
and 1% P/S
Adherent
FBS: Fetal Bovine Serum; P/S: Penicillin/Streptomycin
Table 2: List of cell lines used and their culture properties
3.2. Buffers & Solutions
Agar Plates: LB Agar (Fisher Bioreagents) was prepared per manufacture’s protocol and
autoclaved. Zeomycin was added to the autoclaved agar to get a final concentration of 100mg/ml.
Liquid agar was poured into the plates (about 20ml per plate), allowed to cool down and solidify
which are used for bacterial transformation.
Liquid Broth (LB): LB Broth (Fisher Bioreagents) was prepared as per the manufacturer’s
protocol and Zeomycin was added to get a final concentration of 25mg/ml and 2ml was
distributed among test tubes to grow mini and maxi prep cultures.
37
1% Agarose Gel:1gm of Agarose was added in 100ml of 1X TAE buffer and heated until the
agarose is completely melted. 8ul of EtBR was added and the solution was casted on a plate,
allowed to cool and solidify.
50X TAE Buffer: 242g Tris Base, 57.1 ml Glacial Acetic Acid and 100ml 0.5M EDTA are brought
up to 1litre using dH 2O.
10X Running Buffer (per one liter) 10X Transfer Buffer (per one liter)
30g Tris 30.2 g Tris base
144g Glycine 144g glycine
10g SDS
1X Transfer Buffer (per one liter) 10X TBST Wash Buffer (per one liter)
100ml 10X Transfer buffer 10ml 1M Tris pH 8.0
200ml of methanol 50ml 3M NaCl
Made up to 1L with dH20 5ml 10% Tween 20
All buffers and solutions used in western blot are 1X dilutions.
Milk-TBST Blocking Solution: 5gm of dried milk powder added in 100ml of 1X TBST
Cell Lysis Buffer:
0.5M Tris, pH 6.8 – 1.25ml
10% SDS – 2ml
Glycerol – 1ml
Volume brought up to 10ml using dH20
3.3. Protein and RNA Assays
3.3.1. Protein Assay:
Using Reagents A, B & S from the Bio Rad DC Protein Assay Kit, an assay was performed per
manufacturer’s protocol to estimate the protein concentrations of cell lysates. The absorbance
was observed at a wavelength of 750nm using a spectrophotometer (Beckman and Coulter).
3.3.2. Transient Transfection of Expression Constructs:
Approximately 300,000 cells were plated on a 6-well plate media and grown overnight. The media
was then is replaced with 1ml of optimem media and the cells were transfected with pCMV6
Vector backbone, IRES-GFP Vector backbone, ABCA3-GFP and FAM83A-GFP in individual wells
using lipofectamine. Then the plate was incubated at 37C, 5% Co2 for 5hrs and the optimum
media is replaced with the corresponding media used for the cell line and incubated at 37C, 5%
CO 2 for 48 hrs. Transfection efficiency was determined through GFP fluorescence using
fluorescent microscopy.
38
3.3.3. Protein Extraction:
For HEK293T lysate, the cells were harvested after 48hrs and for A549 lysate, the cells were
harvested after 24hrs. The transfected cells were pelleted at 1.5rpkm for 5mins using sterile
1XPBS. In general, 300ml of Cell Lysis Buffer with protease inhibitor was added for cell pellet size
of 2mm. The volume of cell lysis buffer added was adjusted accordingly for different cell lines.
This was followed by shearing the cell pellet using 3mm sterile syringe until a homogenous, non-
viscous mixture was obtained. The protein concentration of the cell lysate was determined by
performing Bradford assay using Bio Rad DC Protein Assay Kit. Samples are later frozen at -80°C
for long-term storage.
3.3.4. Western Blot Analysis:
Samples are prepared according to the desired protein concentration by using the cell lysates, 5ul
of 5X protein loading dye and the final volume was made up to 25ul with cell lysis buffer. This
protein mixture in denatured at 100
o
C for 5min on a heat block. Once denatured, the samples
were loaded on to 8% SDS gel. 7ul of Precision Plus Protein Kaleidoscope Ladder (Biorad) was
used for protein size identification. As a negative control, lysate from cells transfected with IRES-
GFP backbone and CMV6-GFP backbone vectors were used. The SDS-PAGE was run at 120V for
90mins. A wet transfer was set up by preparing a gel-membrane sandwich using sponge and
paper on either sides and was tightly clamped ensuring no air bubbles occur. The transfer is taken
place at 100V on ice or at 4
o
C. The membrane was then blocked using 5% milk-TBST for 1hr at
room temperature followed by thorough rinsing with 1X TBST to remove milk residues. Primary
antibody incubation was done overnight at 4
o
C using Mouse-Monoclonal ANTI-FLAG M2 antibody
at 1:10,000 dilution. The blot was then washed 3 times with 1X TBST for 10mins each at room
temperature followed by secondary antibody incubation using anti-mouse IgG (Santa Cruz
Biotechnology) at 1:5000 dilution for 1hr at room temperature. Subsequently, the blot was
washed 3 times using 3% milk-RBST, 1.5% milk-TBST and 1X TBST for 10mins each at room
temperature. Pierce ECL Western Blotting Substrate was added on to the blot at 1:1 ratio and then
the bands were identified using GelDoc Imager (Biorad) with a 60sec exposure. Images were
captured for every sec and the best defining image was used for analysis. B-Actin (Abcam) was
used as an internal control on a stripped blot obtained by incubating the blot with 1X stripping
buffer (Biorad) for 15mins. Primary B-Actin (Abcam) at a dilution 1:1000 was incubated
overnight followed by three 1X TBST washes, 10mins each at room temperature. Secondary
antibody anti-rabbit IgG (Santa Cruz Biotechnology) was incubated at 1:5000 dilution for 1hr at
room temperature followed by the same protocol as before. Images were captured for every 5sec
of a 60sec exposure and the best defining image was used for analysis.
39
3.3.5. RNA Extraction:
The transfected cells were harvested for RNA using Trizol-chloroform extraction. After
appropriate incubation the transfected 6-well plate was placed on ice in order to inactivate any
activities occurring in the cells. Extraction was performed in a sterile RNase free environment.
The media in each well was drained out and cells are rinsed with 1ml of 1X PBS. One ml of trizol
was added per well with high confluence and the cells are lysed by pipetting the cells up and down.
The homogenized cell suspension was transferred into a tube and 200ul of chloroform was added
per 1ml of trizol. The samples were vortexed at maximum speed using vortex for 15sec and then
subjected to centrifugation at 14000rpm for 20mins at 40C. The mixture was separated into two
phases out of which the aqueous phase was pipetted into a fresh tube by angling the tube at 45°.
500ul of cold isopropanol was added for 1ml of trizol used and centrifuged at 14000rpm at 40C.
The supernatant was removed and the pellet was washed using cold 70% ethanol and again
centrifuged at 14000rpm for 20min at 40C. The RNA pellet was then air dried for 1hr and
resuspended in 120ul of PCR water using filter tips. The concentration was determined using
Nanodrop (Thermo Scientific).
3.3.6. cDNA Synthesis:
To convert RNA to cDNA, iScript cDNA Synthesis Kit (BIORAD Cat. No. 170-8891) was used. The
procedure was performed in a sterile RNase free environment. The reaction setup was made using
extracted RNA at a concentration of 1ug, 4ul of 5x iScript reaction mix, 1ul iScript reverse
transcriptase and nuclease free water was used to make up the volume to 20ul. The reaction
protocol involves incubation of the reaction mix at 25C for 5mins followed by 42C for 30mins and
85C for 5mins. Samples can be frozen at -20C for long term usage.
3.3.7. q-PCR:
Then, total cellular RNA was isolated from the cells using Trizol reagent (Invitrogen). The mRNA
copies were quantified using a real-time RT-PCR kit (Takara, Japan) on a Bio-Rad iQ5 Sequence
Detection System (Bio-Rad, Hercules, CA, USA). Primers used are as follows:
Gene Forward Reverse
ABCA3 5'-GACGGGACATGCTGTTTCTA-3' 5'-CAGGGGCAACTAAAGCCATA-3'
FAM83A 5'-AGAAGGCTGGGGCTCATTTG-3' 5'-AGGGGCCATCCACAGTCTTC-3'
ACTIN 5`-GTTGAGAACCGTGTACCATGT-3` 5`TTCCCACAATTTGGCAAGAGC-3`
ABCA17 5`-ATTGATGATCCCAAAGCTTTCTAC-3` 5`- CACCAGCCTTCTATGTCATCTT-3`
Table 3: List of qPCR primer sequences used in qPCR analysis
40
Actin was used to calculate the delta Ct values (average expression of the construct/average
expression of Actin). The delta Ct values were then normalized for ABCA3 & FAM83A expression
using the IRES-GFP control mRNA expression. Finally the fold difference was calculated by using
this formula: 2^-(ddCt) and finally the graphs are plotted with constructs on X-axis and fold
difference on Y-axis.
3.4. Cloning of the ABCA17 putative promoter:
3.4.1. PCR Amplification of ABCA17 Constructs:
Two different ABCA17 constructs that vary in size are amplified from human lung genomic DNA
using Phusion® High Fidelity Polymerase from NEB. Appropriate primers were used
corresponding to each construct at 10uM concentrations. To this mixture, 10ul of 5X GC rich
buffer along with 15ul 5mM betaine, 1.5ul of DMSO and 1ul of 10mM dNTP’s were added and
made up to a 50ul final volume using nuclease free PCR water. A gradient PCR amplification was
performed using Bio Rad PCR machine where the samples are initiated at 98ºC for 30sec. The
reaction is followed by 27 cycles of denaturation at 98ºC for 30sec followed by gradient annealing
ranging from 55ºC to 65ºC and extension at 72ºC for 6mins. The final extension occurs at 72ºC for
10mins and final cooling done at 4ºC. Once the amplification is completed, 1ulof the PCR reaction
along with 4ul of 10X gel loading buffer and 10ul of dH 2O was run on 1% agarose gel at 90V for
approximately 30mins. 3ul of 1Kb DNA ladder was loaded next to the PCR reaction to identify
positive band using Bio Rad gel imager.
ABCA17 Construct 1 (C1) - 1845bp
Primers Sequence Strand
5-Abca17_SpeI TGTTGTactagtAGAGCCGTCCCCAAAGTTCCAGAATG Sense
3-Abca17mid_NcoI ACAACAccatggTACTCGCAGGGTGGCTGACTG Anti-sense
ABCA17 Construct 2 (C2) - 2874bp
Primers Sequence Strand
5-Abca17mid_SpeI TGTTGTactagtTGCAACCCGCTACTGGCGGA Sense
3-hAbca17NcoIv2 TGTTGTccatggTCACCATCCACTATAATGTTTACC
ATCAATCTTGAGAG
Anti-sense
Table 4: PCR primer sequences for ABCA17 constructs C1 and C2
41
3.4.2. Plasmid and ABCA17 Constructs (C1 & C2) Digestion:
Both pCpGL-basic vector and the constructs were subjected to double digestion at a concentration
of 1ug using 0.5ul of HF-SpeI and HF-NcoI restriction enzymes and 5ul of Cutsmart buffer from
NEB. The final volume was brought up to 50ul using dH 2O and the digest reactions were incubated
at 37C for 2hrs. This was followed by running the digested samples on 1% agarose gel and as a
negative control, undigested plasmid was also ran along with the digested samples. Bands of
interest were observed (1855bp & 3844bp for C1 and 2879bp & 3844bp for C2) and extracted
using Gel Extraction Kit (Qiagen) as per the manufacturer’s protocol.The constructs were eluted
in 30ul dH 2O.
3.4.3. Ligation of Plasmid and ABCA17 Constructs:
Using T4-Ligase (NEB), the digested plasmid and the inserts (PCR amplified constructs) were
ligated at two different ratios L1-1:7 & L2-1:5. The reaction also includes 1ul of 10X ligation buffer
(NEB) and a final volume of 15ul was brought up by dH 2O. Samples were incubated at 16C
overnight in Bio Rad PCR machine.
3.4.4. PIR1 Bacterial Cell Transformation:
Transformation was done using 5ul of the ligated product and 5ul of PIR1 competent cells that
are compatible with the CpG less vector. The samples are then incubated on ice for 30mins. Then,
cells are subjected to heat shock at 42C for 40sec and immediately placing them on ice for 2min
to ensure the intake of plasmid by PIR1 cells. To these samples, 400ul of LB broth was added and
incubated at 37C. After 35min incubation at 37C the cells were spin down at 10,500rpm for 30sec
and 340ul of the supernatant was removed and the cells are re-suspended. The re-suspended cells
are now carefully plated in a sterile environment on zeomycin containing agar plates. The plate
are then incubated overnight at 37C and observed for colonies after 16hrs.
3.4.5. Mini Prep for Positive Colonies and Test Digest:
The colonies observed on the plates were screened for positive clones containing the plasmid.
Hence colonies that are picked randomly are cultured in 2ml LB containing Zeomycin at a
concentration of 25mg/ml. These cultures were grown overnight at 37C. Mini Prep Plasmid
Extraction Kit (Qiagen) was used according to the protocol and finally the extracted plasmid DNA
was eluted in 30ul dH 2O. These samples are subjected to a test digest to screen for positive clones
using 0.25ul of restriction enzymes SpeI and AlwNI (NEB) in a total volume of 25ul that includes
2.5ul Cutsmart buffer (NEB) and H 2O. Positive clones were selected based on the presence of
bands at 1198bp & 4501bp for C1 and 2496bp & 4227bp for C2 on a 1% agarose gel.
42
3.4.6. Construction of ABCA17 full length construct C3:
The individual constructs C1 & C2 were subjected to double digestion using HF-SpeI & HF-NotI
and cutsmart buffer, incubated at 37C for 2hrs and then purified and eluted in 30ul dH 2O. These
purified digests were then ligated using T4 Ligase. Samples were incubated at 16C overnight in
Bio Rad PCR machine which were later transformed into PIR1 bacterial cells. Positive clones were
screened by performing mini prep Mini Prep Plasmid Extraction Kit (Qiagen) on selected colonies
and the DNA was eluted in 30ul dH 2O. A test digest using SpeI & AflII was performed and
corresponding clones that have bands at 2243bp & 6097bp were identified positive containing
the C3 construct.
3.4.7. Sequencing of the Positive Clone ABCA17 Constructs:
IDT sequencing platform was used to sequence both the constructs extracted from positive clones
in order to avoid any mutations. Samples were prepared using appropriate DNA concentrations
measured by Template Total Mass/Sample DNA concentration and 25pmol primer was added.
The volume was made up to 15ul using dH 2O. The resulting sequence from IDT was then aligned
with the construct using serial cloner 2.6.1 and mutation free constructs are selected.
Primer Sequences:
CpG Forward: 5`-TAGCAATCAATATTGAAAACCAG-3`
5-Abca17mid_SpeI: 5`-ACTAGTTGCAACCCGCTACTGGCGGA-3`
Ab3-2_seqf: 5`-ATGCTCATTACGGAGACTAGGC-3`
Ab17_2800seqF: 5`-AATGACTCTGCGCACGTGCGTTTC-3`
CpG Reverse: 5`-GCCTTTCTTAATATTCTTGGCA-3`
3.4.8. Fixing Mutations for C1 & C2 Constructs:
The CpG rich constructs C1 & C2 individually acquired mutations which were fixed using the
mutation free C3 clone. Based on the location of mutation restriction sites SpeI and NotI for C1 &
NotI and NCoI for C2 were selected. Desired bands were isolated through gel extraction and
ligated with each other following the previous protocol. Both C1 & C2 ligated products were
transformed into PIR1 cells and screened for positive clones. Thus, obtained clones were sent off
for sequencing and analysis resulted in selection of mutation free C1 & C2 constructs.
43
3.4.9. Maxi Culture of the Positive Clones:
Once the mutation free positive clone was selected, the corresponding bacterial colony cultured
in 250ml LB containing 4X zeomycin antibiotic and was grown overnight at 37C. Using Qiagen
Plasmid Maxi Kit, the plasmid DNA was extracted and eluted in 300ul PCR water which is again
subjected to a test digest as previously mentioned in order to make sure we have the desired
construct.
3.5. In Vitro DNA Methylation by MSssI Methylase:
Promoter constructs were incubated for 24 h at 37 °C with SssI methylase (New England BioLabs)
at 1 unit/μg pDNA in PCR water and NEBuffer2 (New England BioLabs) along with 10X 32μm S-
adenosylmethionine. MSssI was replenished after 24hrs and incubated again at 37C overnight.
The methylated DNA was extracted using phenol-choloroform-isoamyl alcohol method and were
eluted in 40ul of PCR water. Methylation was confirmed using NotI digestion, which is sensitive
to CpG dinucleotide methylation by CpG Methyltransferase.
3.5.2. Luciferase Assays:
H522, H2228 and H1944 were transfected with 2ug of ABCA3 and ABCA17 constructs
individually. The media was changed after overnight incubation. Luciferase activity was assayed
18–36 h later using the Dual-luciferase reporter assay system (Promega) with the CpGL vector
without promoter constructs as the internal reporter control. For the constructs, CpGL vector
contains the ABCA17 promoter from positions +435 to +2283 in C1 and +435 to +3307 in C2 and
C3 from +435 to +4925. For ABCA3 constructs, promoter is located from positions +430 to +2412
in Ab3-1, +430 to +2896 in Ab3-2 and +430 to +4639 in Ab3-1-2.
44
4. RESULTS
4.1. Validation of endogenous ABCA3 protein expression in different cell lines:
As an intracellular lipid transporter specifically expressed in lung ATII cells, endogenous
ABCA3 protein expression was set out to be analyzed in a range of cell lines using western blot
analysis. Analysis was not only limited to the selected cell lines but was also performed on whole
cell lysates of A549, human lung tissue, mouse lung tissue lysate and HeLA. Out of these whole cell
lysates HeLA, H522, H2228, A549, human lung lysates and mouse lung lysates were expected to
have endogenous ABCA3 expression while H1944 and PC3 cells served as negative controls (RNA
Seq data). Cell lysates were obtained and assayed for protein concentration. All samples are used
at a concentration of 25ug in a total volume of 25ul.
As seen in the following figure (Fig 4.1) many non-specific bands were observed in all the
cell lines. PC3 cells assumed to be negative for ABCA3 expression (based on RNA Seq data
performend in house) also showed a non-specific band while H1944 a second negative control
did not show any band consistent with the RNA seq data. Hence no commercial antibody was
acceptable for western blot use.
Fig 4.1: Western blot analysis of ABCA3 using different antibodies in various cell lines. A) Primary
ABCA3 C-terminal antibody from Santa Cruz (1:150), B) Primary ABCA3 N-terminal antibody
from Abcam (1:1000), C) Primary ABCA3 C-terminal antibody from Abcam (1:1000), D) Primary
ABCA3 antibody from Covance (1:500 diln)
M: Precision protein ladder marker. rAT2: Rat AT2 cell lysate, rLL: Rat lung lysate,
hLL: Human lung lysate. Arrows indicate approximate location of the expected ABCA3 band.
Bands observed are non-specific and not the band of interest.
45
Antibody Catalog Number Epitope Dilution Isotype
Santa Cruz
Biotech
sc-48442 C-Terminal 1:150 Goat IgG
Santa Cruz
Biotech
sc-48444 N-Terminal 1:150 Goat IgG
Covance MMS-645R C-Terminal 1:500 Mouse IgG2a
ABCAM ab187755 N-terminal 1:1000 Polyclonal Rabbit
Table 5: Antibodies used in the validation of endogenous ABCA3 protein expression
4.2 Overexpression of ABCA3 in HEK293T:
4.2.1. ABCA3 Expression Constructs:
The expression construct purchased from Origene was remodeled using a polylinker
with feasible restriction sites to insert IRES-eGFP fragment. The IRES-eGFP fragments were
amplified from the lentivector LeGO iG (Courtesy of Dr. Kate Lawrenson of Gayther Lab, USC)
and ligated to the plasmid which contains a CMV promoter. ORFs cloned will be expressed as a
tagged protein with a C-terminal Myc-DDK tag that is used for detecting transgene protein
expression. The plasmid also contains ampicillin as well as Geneticin (G418) selection marker
that enables selection in culture. The control expression construct was generated by subcloning
a polylinker in a way that ABCA3 was removed from the plasmid while restoring all other
important sites and tags.
Fig 4.2: The pCMV6 entry vector plasmid that was used to generate the ABCA3 expression
constructs. An IRES-eGFP fragment was cloned downstream of the Myc and FLAG (DDK) tags (See
Materials and Methods).
46
4.2.2. Detection of ABCA3 protein overexpression by Western Blot Analysis:
In order to test for ABCA3 protein expression from the constructs HEK293T cells that do
not express any ABCA3 were selected to prevent endogenous background expression. The cells
were transiently transfected with 2ug of ABCA3 expression construct using Lipofectamine that
resulted in a 70% transfection efficiency (Fig.4.3). The cells were grown for 24hrs and then
harvested for RNA and protein. In order to confirm the ABCA3 overexpression at protein level,
western blot analysis was performed using Anti-Flag antibody (Sigma) at 1-10000 dilution that
recognizes the Flag-tag (Sequence: DYKXXD) present in the construct backbone. The flag tag is
one of the most commonly used hydrophilic octapeptide tags and here it is attached to ABCA3 at
its C-terminal end. Using this antibody a band was observed at 192 kDa as shown in Fig 4.4B.
4.2.3. qPCR analysis to detect overexpression at RNA level:
In order to detect mRNA expression, RNA was harvested from both the cell lines using
trizol and subjected to cDNA synthesis. Later, qPCR was performed using the primers specific for
ABCA3. Actin was used to calculate the delta Ct values which were normalized using IRES-GFP
vector. Approximately 27000 fold change relative to empty vector was observed for ABCA3 thus
confirming overexpression of the construct.
Fig. 4.3. Transfection of HEK293T cells with ABCA3 expression construct using Lipofectamine
2000. BF: Bright field microscopy image of HEK293T cells, GFP: GFP fluorescence captured by
fluorescent microscopy. Merge: Overlay of bright field and fluorescence images representing the
efficiency of transfection. From the merge image, transfection efficiency was estimated to be
about 70% using Lipofectamine-2000.
47
Fig 4.4. Overexpression of ABCA3 in HEK293T cells. A) Western blot of ectopically expressed
FLAG-tagged ABCA3 using anti-FLAG antibody (1:10000). 25ug protein was loaded in each lane.
Actin was used as loading control. B) qPCR analysis of transfected cells showing robust
upregulation of ABCA3 mRNA. ABCA3 levels were normalized to actin and presented as fold
change over empty control.
From these analyses, we conclude that ABCA3 is expressed in the transfected cells. In
order to use these constructs in functional analysis we established cell lines that stably express
ABCA3. The cell lines were chosen based on the endogenous ABCA3 expression determined from
RNA seq data analyzed in house. Fig.4.7 indicates the cell lines determined to be used in the study.
The stable cell lines were generated through stable transfection of ABCA3 and initial set of
positive cells were selected using GFP florescence. In order to determine if the cells have stably
integrated the foreign DNA, selective pressure is applied by growing the cells in media containing
Geneticin (G418). The selectable marker in the construct if integrated into the genome helps the
cell grow and hence these cells are specifically selected and expanded to become stable cell line
expressing ABCA3. This is in progress and data related is not shown here.
4.3. In-vitro DNA Methylation and promoter studies on ABCA3 and ABCA17P constructs:
UCSC genome browser was used to analyze the ABCA3 gene locus. Previous studies by
Piehler et al. on the ABCA3 locus represented interesting findings in respect to its orientation with
its adjacent pseudogene ABCA17P. It was interesting to note that the 5`-ends of both of these
genes are located in a close proximity (176bp apart,) and run in a head to head orientation along
the genome (Fig.4.5). Another important feature of this locus is that the 5`ends of both the genes
share a large common CpG island (-1040 to +1417 in respect to ABCA3 TSS). ABCA3-ABCA17P
48
locus is also one of the first indications that the promoters of progenitor gene and its pseudogene
can overlap.
Fig 4.5: Orientation of ABCA3 gene locus represented through UCSC genome browser. The blue
tracks indicate gene products of ABCA3 and ABCA17P. The green bar indicates the CpG island.
ABCA3 and ABCA17P runs opposite to each other along the genome. The genes share a common
CpG island (-1040 to +1417 in respect to ABCA3 TSS).
Due to the presence of a CpG island at this overlap region, we speculated that ABCA17P
might have potential regulatory effects on ABCA3. Studies on pseudogenes that indicated
regulatory function are usually expressed as non-coding RNA. Although, from systematic analysis
of protein expression data available publicly, ABCA17P is known to be expressed at low levels in
testis, kidney, brain, lung, prostate, lymph nodes and secretory glands such as salivary and
adrenal glands, its expression and occurrence in cancer is not well defined. Peihler et al.
demonstrated that ABCA17P expression is extremely low in human lung tissue through northern
blot analysis. Hence to set out to validate the expression levels of ABCA17P in both whole lung
extract and different cancer cell lines available in house, a qPCR analysis was carried out using
primers specific for ABCA17P on the samples indicated below. Out of the 7 cell lines used, K562
(chronic myelogenous leukemia) was used as a positive control and the expression was
normalized using its delta Ct as a base line (Fig. 4.6).
Fig.4.6: qPCR analysis of ABCA17p expression over a range of cell lines and whole lung
RNA. Samples include K562 (leukemia (CML)) as a positive control, H2228, H1944, H23, H522,
LUAD cell lines; RWPE1 (immortalized prostate cell line) & whole lung lysate (human).
0
1
2
3
4
5
6
7
8
9
10
K562 H2228 RWPE1 WHOLE
LUNG
H1944 H23 H522
49
4.4. Screening and selection of lung cancer cell lines:
RNA Seq is a powerful technique that provides information regarding the transcriptional
activity of gene of interest. Quantifying gene expression, identification of alternatively spliced
genes and detection of allele-specific expression are a few of many of its applications
74
. Gene-level
expression measurements are reported in reads per kilo base per million reads (RPKM) which
also reports for length normalization. Depending on the RPKM’s of each gene, expression levels
can be estimated and used.
Fig 4.7: ABCA3 RNA-seq expression data of LUAD cell lines used in the present study. RNA-
seq processed reads were visualized using Integrated Genomics Viewer (IGV).
Blue track indicates ABCA3 gene structure. Red wiggle tracks indicate LUAD cell lines H522 &
H2228 expressing ABCA3. Green tracks indicate LUAD cell lines H1944 & H358 that do not
express ABCA3. (Image courtesy of Dr. Crystal Marconett)
RNA-Seq data were generated by Dr. Crystal Marconette from ILO lab, USC on all the lung
cancer cell lines available in house. Using Integrative Genomics Viewer (IVG) wiggle tracks were
generated for many cell lines out of which tracks H23 (RPKM: 18.95), H522 (RPKM: 11.9) and
H2228 (RPKM: 9.54) were identified to be the cell lines with endogenous ABCA3 expression (Fig.
4.7). As a negative control, H1944 (RPKM: 1.44) and H358 (RPKM: 0.82) were selected.
4.5. Promoter Construct Luciferase Reporter Assays:
4.5.1. Design & Development of ABCA3 and ABCA17P Promoter Constructs:
Epigenetic events such as DNA hypermethylation of promoters are commonly observed
in cancer
51
. From genome scale DNA methylation analysis
20
part of the overlapping CpG Island in
the promoter region of ABCA3 and ABCA17P locus is hyper methylated. We hypothesized that
this methylation which spans over the adjacent ABCA17 pseudogene promoter region could be
50
associated with regulation of ABCA3 in an orientation dependent manner. In order to investigate
effects of ABCA17P DNA methylation on ABCA3, luciferase constructs were designed and
constructed.
The promoter CGI spans about 2.4kb over the ABCA3-ABCA17P gene locus, overlapping
the first exons of both the genes that run opposite directions. Three different lengths of constructs
were generated each for ABCA3 and ABCA17P using the CGI as a boundary and depending on the
feasibility of potential primers. ABCA3 whole construct (Ab3-1-2) extends from -1569 to +2640
relative to its TSS. The remaining two constructs represent the first 1983bp (Ab3-1) and
remaining 2467bp (Ab3-2) of the whole construct Ab3-1-2 that run from -1569 to +416 and -174
to +2640 respectively. A similar strategy was applied to ABCA17P where the whole construct (C3)
is 4491bp long and spans the entire CGI and extends from -1650 to +2842 relative to ABCA3 TSS.
The remaining two constructs represent 1849bp (C1) and 2876bp (C2) of the C3 construct that
run from –1650 to +194 and -31 to +2842, respectively.
pCpGL - Basic vector:
pCpGL basic vector (shown below in Fig 4.8) was used to clone the different promoter constructs
spanning the CpG island over ABCA17. The vector backbone contains R6K gamma origin of
replication and a Zeocin resistance gene. The vector including its multiple cloning site (MCS) is
completely free of CpG dinucleotides.
51
Fig 4.8: Schematic representation of CpGL vector, design of ABCA3and ABCA17P constructs and
location of the constructs over ABCA3 gene locus.
A) Design of CpGL Vector from Klug et al. (2006), B) Schematic of ABCA17P promoter constructs
C1, C2, C3 and ABCA3 promoter constructs Ab3-1, Ab3-2 and Ab3-1-2, C) UCSC genome browser
annotation track to display ABCA3-ABCA17P gene locus on human chromosome 16. ABCA3 and
ABCA17P promoter constructs aligned at ABCA3-ABCA17P gene locus. The blue tracks indicate
gene products of ABCA3 and ABCA17P. The green bar indicates CpG island. Aligned ABCA17P
constructs include Ab17-C1, C2 & C3 and ABCA3 constructs include Ab3-1, Ab3-1 & Ab3-1-2.
Colors of each construct in UCSC genome browser annotation track correspond to the schematic
displayed in panel B.
4.5.2. Validating In-vitro Methylated ABCA17P & ABCA3 promoter constructs:
All the promoter-luci constructs were cloned accordingly (refer methods) into CpG less
vector and subjected to methylation by CpG Methyltransferase (M.SssI). It is one of the basic
epigenetic tools used to methylate all cytosine nucleotides occuring in unmethylated or
hemimethylated DNA in a 5’-CpG-3’ context. Methylation prevents the action of some methylation
sensitive restriction enzymes (Ex: HpaII/NotI) which are used to test whether the DNA
methylation of constructs occurred or not.
52
Fig 4.9: In vitro methylation workflow and the mechanism of CpG methylation using MSssI CpG
methyltransferase in the presence of S-Adenosyl Methionine (SAM). “Construct” refers to either
ABCA3 or ABCA17P promoter constructs used for in vitro methylation.
Fig 4.10: Gel electrophoresis image of NotI-digested methylated and unmethylated ABCA3 and
ABA17 promoter constructs. Methylated and unmethylated ABCA3 and ABCA17P constructs
were digested with NotI for 2hrs at 37C to test for in vitro methylation. The lanes under
53
methylated represent supercoiled and undigested constructs. CpGL vector is undigested in both
methylated and unmethylated samples. Methylated samples represent undigested constructs.
Unmethylated ABCA17P C1, C2 & C3 bands observed at 5699bp, 6723bp & 8340bp respectively
while ABCA3 3-1 & 3-1-2 observed at 5826bp & 8508bp.
T The treated constructs are then transiently transfected both into ABCA3
positive
cell
lines H522, H2228 and ABCA3 negative
cell line H1944. The positive cell colonies sere used to
ensure that all necessary transcription factors would be present. This was followed by performing
a luciferase assay to check the promoter activities of both the methylated and unmethylated
constructs.
4.6. Methylated ABCA17 promoter constructs show transcriptional activity:
To explore the effect of DNA methylation on transcriptional activity, two cell lines H522, H2228,
which have endogenous ABCA3 expression, and H1944, were transiently transfected with
methylated and unmethylated ABCA3 3-1, 3-2, 3-1-2 and ABCA17P C1, C2 and C3 constructs. As
the CpGL vector contains CpG free luciferase, a reporter assay would provide information
regarding the transcriptional activity without skewing the results. The luciferase reporter assay
was performed after 24hrs and the activities of each constructs were normalized using the
unmethylated CpG less vector. Using unmethylated CpG less vector gave a good comparison of the
differences in activities between the methylated and unmethylated constructs. Traditionally, in
vitro DNA methylation of promoter constructs result in suppression of transcriptional activity
when compared with unmethylated constructs.
H522 & H2228: Methylated ABCA3 constructs followed the normal pattern of low transcriptional
activity in comparison with the unmethylated counterparts. In contrast, we observed that the
methylated ABCA17P C2 and C3 constructs were not silenced, but instead some activity was
observed for both H522 and H2228. However, the activity was weaker than their counter
unmethylated constructs. It is interesting to note that the 2.5kb C2 construct contains both the
promoter region and CGI spanning in a reverse orientation and it is unique to observe an activity
upon methylation at this region. Statistical significance was not calculated as the experiment was
only replicated twice, but as observed in Fig 4.11, both the cell lines display similar patterns of
activity.
H1944: On contrary, no specific activity for any of the ABCA17 constructs was observed in. This
was expected as H1944 as it does not possess endogenous ABCA3 expression and may not contain
the required transcription factor machinery. The activities of constructs followed the normal
trend in which methylated samples showed very low (or) no activity when compared to the
54
unmethylated constructs. Similar assay is intended to be performed on another cell line H358
which also does not have endogenous ABCA3 expression and we would expect a similar pattern
of activity in these cell lines.
Fig 4.11: Luciferase assay for methylated and unmethylated ABCA3 and ABCA17P
promoter constructs in H522, H2228 & H1944 cell lines. X-axis indicates the fold change in
activity relative to CpGL vector, Y-axis indicates the constructs which are color coded accordingly
as previously represented in Fig 4.8, Grey bars indicate unmethylated constructs and black bars
indicate methylated constructs.
4.7. Mining of Reduced Representation Bisulfite Sequencing (RRBS) data on the UCSC
Genome Browser reveals a dichotomous DNA methylation pattern within the ABCA3-
ABCA17P CGI:
In parallel to the determining the activity of methylated constructs, data mining of the CGI was
done to compare the DNA methylation status between cancer cell lines and normal cell lines. From
such mining of the RRBS data, which is publicly available on UCSC genome browser (Fig.4.12), an
interesting observation was made regarding the pattern of methylation with in the CGI. The CpG
island that spans over ABCA17P which possess promoter sequences of ABCA3 is noted to be
hypermethylated while the CpG island heading towards the 5` end of ABCA3 was not methylated
at all. Such variable patterns of cytosine methylation at different CpG sites within the island
Fold change relative to CpGL empty vector
55
indicate the presence of potential primary sites of methylation which may be critical for the
maintenance of gene repression.
Fig 4.12: Reduced Representation Bisulfite-sequencing data from ENCODE cancer cell lines
and normal cell lines at the Abca3-ABCA17P locus. The blue tracks indicate gene products of
ABCA3 and ABCA17P. The green bar indicates the CpG island. The arrow points out the significant
methylation pattern difference existing within the CpG island. The tracks below CGI indicate RRBS
methylation status for each cell line (Green: Unmethylated; Red: Methylated). Black line across
the cell lines distinguishes cancer versus normal cell lines. Cancer cell lines include K562
(Leukemia), HeLA-S3 (Cervical Cancer), HepG2 (Liver Cancer), A549 (LuAD) and MCF-7 (Breast
Cancer). Normal cell lines include AG044550 (Lung fibroblast), IMR90 (Lung fibroblast), Fibrobl
(Skin fibroblasts), HMEC (Breast), Lung BR (Lung Bronchial Cells), PrEc (Prostate), SAEC (Small
Airway Epithelial Cells)
4.8. CpG promoter in other clinically relevant ABC genes do not possess differential
methylation pattern:
Following this observation, we wanted to identify if such methylation pattern is unique only to
ABCA3 or is a common pattern among genes belonging to the ABC family. In order to identify the
methylation patterns, clinically relevant genes of ABCA family were selected. ABCB1, ABCG2,
ABCC1 and ABCB4 are important in general lung function and also are involved in causing
multidrug resistance. In the UCSC genome browser, all the genes were identified to have CGI
promoters but none of them appeared significantly hypermethylated in cancer (Fig. 4.13). Within
the CpG no peculiar methylation pattern has been observed similar to ABCA3. However, the
coverage of the RBSS was poor and idd not show the whole CGIs. Thus, it is difficult to conclude
whether the ABCA3 DNA methylation pattern seen in cancer cell lines is unique.
56
Fig 4.13: Promoter CGI’s of four different ABC gene loci. The Genome Browser annotation
tracks displayed for four different genes ABCB1 (chromosome 7), ABCG2 (chromosome 4), ABCC1
(chromosome 16) and ABCB4 (chromosome 7). The blue tracks indicate multiple gene products
of the gene. The green bar indicates the CpG island. The annotation track indicates no significant
differential methylation occurring in any of the promoter CGI of the displayed genes. From top to
bottom each row of the RRBS data indicates the following cell lines: K562 (Leukemia), HeLA-S3
(Cervical Cancer), HepG2 (Liver Cancer), A549 (LuAD) and MCF-7 (Breast Cancer). Normal cell
lines include AG044550 (Lung fibroblast), IMR90 (Lung fibroblast), Fibrobl (Skin fibroblasts),
HMEC (Breast), Lung BR (Lung Bronchial Cells), PrEc (Prostate), SAEC (Small Airway Epithelial
Cells). RRBS methylation status represented as Green: Unmethylated and Red: Methylated.
57
5. DISCUSSION AND FUTURE DIRECTIONS
DNA methylation of CpG-rich promoter regions is known to cause gene silencing by
either directly hindering transcription factor binding or indirectly by recruiting MBD
proteins to alter the local chromatin structure
75
. Though CGI methylation might not always
be an initiating event in gene silencing, it surely plays a role in locking the gene in a silent
state.
The goal of the present study was to determine whether ABCA3 plays a role in LUAD
development as a potential tumor suppressor gene. Although much has been reported on
ABCA3 in lung disease, its functional contributions to lung cancer has never been
documented. Interestingly in AML, ABCA3 upregulation is a poor prognostic factor by
inducing multidrug resistance
61
, whereas based on findings by Selamat et al., upregulation of
ABCA3 may be beneficial to patients with LUAD. This apparent incongruity may mean that
ABCA3 has far broader cellular roles than regulating lung surfactant homeostasis and
lamellar body biogenesis.
Hanahan and Weinberg
76
proposed that the transition from a normal cell to cancer
cell occurs through any of six widely accepted “hallmarks” of cancer. In light of this concept,
we intended to determine whether cell proliferation, anchorage-independent growth, and
invasiveness changed upon perturbation of ABCA3 in LUAD cell lines. The availability of RNA-
seq data for a panel of LUAD cell lines allowed us to preliminarily select cell lines appropriate
for overexpression and knockdown studies. However, because subsequent experiments
would interrogate ABCA3 function, its expression at the protein level needed to be validated.
This proved to be the most challenging and limiting factor in the study because none of the
commercially available primary antibodies we tested succeeded in reproducibly detecting
ABCA3. We are currently testing an ABCA3 antibody approved for immunofluorescence
staining as an alternative approach to detecting endogenous levels of ABCA3. In lieu of using
a protein-specific antibody, we opted to express and detect by Western blot FLAG-tagged
ABCA3. We successfully detected the tagged protein in HEK293T cells using an anti-FLAG
antibody and are currently optimizing transfection conditions for LUAD cell lines.
58
Due to the obstacles faced in optimizing the ABCA3 Western blots, we were not yet
able to establish stable LUAD cell lines overexpressing ABCA3 at the time of this report.
However, steps are in place to generate these cell lines as well as those with stable
knockdown of ABCA3.
In addition to investigating the function of ABCA3 in LUAD, we concurrently
interrogated the transcriptional regulation of ABCA3. The finding that ABCA3 is
hypermethylated in LUAD suggests that DNA methylation plays a role in regulating its
expression. The ABCA3 locus presented interesting characteristics with respect to its
orientation and its promoter CGI overlap with its neighboring pseudogene which run head to
head along the genome. The CGI spans the 5` region of ABCA17P which is also the promoter
of ABCA3. Aberrant methylation in lung adenocarcinoma samples was observed at this
region, potentially affecting ABCA3 expression and the various mechanisms involved ought
to be explored. Though some consider pseudogenes biologically irrelevant, sequences within
them are well preserved suggesting the existence of gene elements that play an important
cellular role.
Most of the thesis work was focused on identifying the effect of promoter CGI
methylation on transcriptional activity of ABCA3. Reporter gene assays were performed
using in vitro methylated constructs in three cell lines H522, H2228 and H1944. Since
expression of a gene indicates that the full repertoire of TFs required for its transcription is
also present, we expected the levels of reporter activity of our various ABCA3 and ABCA17P
constructs to vary based on ABCA3 status. In the cell lines H522 and H2228 that have
endogenous ABCA3 expression as expected, the ABCA3 methylated samples showed
significantly repressed activity compared to their unmethylated counterparts. This silencing
can be recapitulated using the CpG-less vector system described by Klug et al. In contrast,
two of the ABCA17 constructs, C2 (+-32 to +2842 bp relative to the ABCA3 transcription start
site) and C3 (-1650 to 2842 bp relative to the ABCA3 transcription start site) spanning the
CpG island were not robustly silenced. In case of methylated ABCA17P constructs, two out of
the three (C2 and C3) interestingly showed transcriptional activity. In H1944 cells that do not
have endogenous ABCA3 expression, neither of the methylated nor unmethylated constructs
showed activity.
59
Although the difference in activities of the methylated versus unmethylated construct
in ABCA17P C2 and C3 cannot be statistically calculated (insufficient replicates), the trend in
their transcriptional activity in both H522 and H2228 cells is remarkable. This may indicate
the presence of potential regulatory elements that may lie within the CGI.
Another interesting observation to note is that differential methylation seems to
occur in part of the CpG island, as evidenced by RRBS data from ENCODE (UCSC genome
browser, hg19). This strongly suggests that partial methylation of a CGI has the potential to
deregulate gene expression. As far as we know, our report is the first account of this sort of
dichotomous methylation pattern and warrants further study.
In order to verify and validate if similar methylation patterns observed by RRBS
occuring in LUAD, LUAD cell lines available in- house will be subjected to Methylight
77
. The
mechanism that lies behind such differential methylation can be further investigated by
transfection of methylated and unmethylated hybrid constructs reflecting a range of specific
methylation patterns. Results of the reporter assays from these transfections could tell us
whether a unique CGI methylation pattern is required for suppression or activation of
transcriptional activity.
Another interesting observation is the presence of enhancer-blocking transcription
factor CTCF at ABCA17P locus. Studies by Wade et al. indicate that upon aberrant
methylation CTCF binding does not occur and the methylated regions can help in
transcriptional activity or mediate other additional TF’s in binding at this region. On
observing DNAse hypersensitivity tracks in UCSC genome browser, it can be speculated that
methylated ABCA17-CGI locus holds a potential to act like an enhancer and activate another
gene located distantly. Activation of such genes could in turn be involved in transcriptional
silencing of the potential TSG ABCA3. It can also be speculated that heterozygous ABCA3+/-
individuals are more susceptible to developing lung ADC as they have haploinsufficiency for
ABCA3 and silencing the other allele through any of these mechanism could give an extra hit
in developing the disease.
60
BIBLIOGRAPHY
1. American Cancer Society Cancer Facts & Figures 2014
http://www.cancer.org/acs/groups/content/@research/documents/webcontent/a
cspc-042151.pdf
2. Jemal A, Siegel R and Naishadham D: Cancer statistics 2013. CA Cancer; J Clin 2013,
Vol 63(1), pp. 11–30
3. American Cancer Society Non-small cell lung cancer key statistics web page:
http://www.cancer.org/cancer/lungcancer-non-smallcell/detailedguide/non-
small-cell-lung-cancer-key-statistics
4. American Cancer Society. “World Health Organization: Outdoor Air pollution Causes
Cancer.” Stacy Simon (2013)
http://www.cancer.org/cancer/news/world-health-organization-outdoor-air-
pollution-causes-cancer
5. Yokota J, Shiraishi K, Kohno T. Genetic basis for susceptibility to lung cancer: Recent
progress and future directions. Adv Cancer Res. Vol 109, 51-72 (2010)
6. Jonathan M. Samet, Erika Avila-Tang, Paolo Boffetta, Lindsay M. Hannan, Susan
Olivo-Marston, Michael J. Thun, and Charles M. Rudin. Lung cancer in never
smokers: clinical epidemiology and environmental risk factors. Clin Cancer Res. Vol
15(18), 5626–5645 (2009)
7. The Maimonides Cancer Center web page. Lung Cancer main web page, “Welcome to
the Thoracic Oncology Program at Maimonides Cancer Center.”
http://www.maimonidesmed.org/CancerCenter/LungCancer.aspx
8. Marcus MW, Raji OY, Field JK. Lung cancer screening: identifying the high risk
cohort. J Thorac Dis. Vol 7(Suppl 2), S156-62 (2015)
9. Fred R. Hirsch, Wilbur A. Franklin, Adi F. Gazdar, and Paul A. Bunn Jr. Early
Detection of Lung Cancer. Clinical Perspectives of Recent Advances in Biology and
Radiology. Clin Cancer Res. Vol 7(1), 5-22 (2001)
10. SEER Cancer Statistics Factsheets: Lung and Bronchus Cancer. National Cancer
Institute: http://seer.cancer.gov/statfacts/html/lungb.html
11. American Cancer Society web page. Breast cancer survival rate by stage:
http://www.cancer.org/cancer/breastcancer/detailedguide/breast-cancer-
survival-by-stage
61
12. Chandra P. Belani, Sherry Marts , Joan Schiller, Mark A. Socinski. Women and
lung cancer: Epidemiology, tumor biology, and emerging trends in clinical research.
Lung Cancer. Vol 55, 15-23 (2007)
13. Smith LJ, McKay KO, van Asperen PP, Selvadurai H, Fitzgerald DA. Normal
development of the lung and premature birth. Paediatr Respir Rev. Vol 11(3), 135-42
(2010)
14. Travis WD, Brambilla E, Noguchi M, et al. International association for the study of
lung cancer/American thoracic society/European respiratory society international
multidisciplinary classification of lung adenocarcinoma. J Thorac Oncol. Vol 6, 244-85
(2011)
15. Zhao Chen and Christine M. Fillmore et al. Non-small-cell lung cancers: a
heterogeneous set of diseases. Nature Reviews Cancer. Vol 14, 535-546 (2014)
16. Santos E, Martin-Zanca D, Reddy EP, Pierotti MA, Della Porta G, Barbacid M.
Malignant activation of a K-ras oncogene in lung carcinoma but not in normal tissue
of the same patient. Science. Vol 223(4637): 661-4 (1984)
17. Paez JG, Jänne PA, Lee JC, Tracy S, Greulich H, Gabriel S, Herman P, Kaye FJ,
Lindeman N, Boggon TJ, Naoki K, Sasaki H, Fujii Y, Eck MJ, Sellers WR, Johnson
BE, Meyerson M. EGFR mutations in lung cancer: correlation with clinical response
to gefitinib therapy. Science. Vol 304(5676), 1497-500 (2004)
18. Grzegorz J. Korpanty , Donna M. Graham, Mark D.Vincent and Natasha B. Leighl.
Biomarkers that currently affect clinical practice in lung cancer: EGFR, ALK, MET,
ROS-1, and KRAS. Front Oncol. Vol 4(204), 2014
19. The Cancer Genome Atlas Research Network: Comprehensive molecular profiling of
lung adenocarcinoma; Nature, Vol 511 (2014), pp. 543-550
20. Ding, Getz and Wheeler et al. Somatic mutations affect key pathways in lung
adenocarcinoma. Nature. Vol 455, 1069-1075 (2008)
21. Marcin Imielinski, Alice H. Berger, Peter S. Hammerman et al. Mapping the
Hallmarks of Lung Adenocarcinoma with Massively Parallel Sequencing. Cell. Vol 150,
1107-1120, (2012)
22. Shikhar Sharma, Theresa K. Kelly and Peter A. Jones. Epigenetics in Cancer.
Carcinogenesis. Vol 31 (no.1), 27–36 (2010)
23. Irene Cantone and Amanda G Fisher. Epigenetic programming and reprogramming
during development. Nature Structural & Molecular Biology. Vol 20, 282–289 (2013)
62
24. Ito K, Ito M, Elliott WM, Cosio B, Caramori G, Kon OM, Barczyk A, Hayashi S,
Adcock IM, Hogg JC, Barnes PJ. Decreased histone deacetylase activity in chronic
obstructive pulmonary disease. N Engl J Med. Vol 352(19), 1967-76 (2005)
25. Cooper et al. Molecular Biology of lung cancer; J Thorac Dis, Vol 5-Suppl 5 (2013),
pp. S479–S490
26. Keith D Roberstson and Peter A Jones. DNA methylation: past, present and future
directions. Carcinogenesis, Vol 21 no 3, 461-467 (2000)
27. Miho M. Suzuki and Adrian Bird: DNA methylation landscapes: provocative insights
from epigenomics; Nature Reviews, Vol 9, 465-476 (2008)
28. Kulis M, Esteller M. DNA methylation and cancer. Adv Genet. Vol 70, 27-56 (2010)
29. Aimee M. Deaton and Adrian Bird: CpG islands and the regulation of transcription.
Genes & Development. Vol 25, 1010-1022 (2011)
30. Robert S. Illingworth and Adrian P. Bird: CpG islands – ‘A rough guide’-Mini review.
FEBS Letters. Vol 583, 1713–1720 (2009)
31. A Portela, J Liz, V Nogales, F Setien, A Villanueva, and M Esteller. DNA methylation
determines nucleosome occupancy in the 5’-CpG islands of tumor suppressor genes.
Oncogene. Vol 32, 5421–5428 (2013)
32. Peter A. Jones and Daiya Takai. The Role of DNA Methylation in Mammalian
Epigenetics. Science. Vol 293, 1068-1070 (2001)
33. Peter A Jones. DNA methylation and cancer. Oncogene. Vol 21, 5358-5360 (2002)
34. Ooi SK, Qiu C, Bernstein E, Li K, Jia D, Yang Z, Erdjument Bromage H, Tempst P,
Lin SP, Allis CD, et al. DNMT3L connects unmethylated lysine 4 of histone H3 to de
novo methylation of DNA. Nature. Vol 448, 714–717 (2007)
35. Maria Hatziapostolou and Dimitrios Iliopoulos. Epigenetic aberrations during
oncogenesis. Cell. Mol. Life Sci. Vol 68, 1681-1702 (2011)
36. Herman JG, Umar A, Polyak K, Graff JR, Ahuja N, Issa JP, Markowitz S, Willson
JK, Hamilton SR, Kinzler KW, Kane MF, Kolodner RD, Vogelstein B, Kunkel TA,
Baylin SB. Incidence and functional consequences of hMLH1 promoter
hypermethylation in colorectal carcinoma. Proc Natl Acad Sci USA. Vol 95(12) 6870-
5 (1998)
37. Ikeda K, Shiraishi K, Eguchi A, Shibata H, Yoshimoto K, Mori T, Baba Y, Baba H,
Suzuki M. Long interspersed nucleotide element 1 hypomethylation is associated
63
with poor prognosis of lung adenocarcinoma. Ann Thorac Surg. Vol 96(5),1790-4
(2013)
38. Genetics Home Reference web page. National Institutes of Health:
http://ghr.nlm.nih.gov/glossary=lossofheterozygosity
39. Knudson, A. G. Jr. Mutation and cancer: statistical study of retinoblastoma. Proc. Natl
Acad. Sci. USA. Vol 68, 820–823 (1971)
40. C. Sawan et al. Epigenetic drivers and genetic passengers on the road to cancer.
Mutation Research. Vol 642, 1–13 (2008)
41. Suhaida A. Selamat et al. Genome-scale analysis of DNA methylation in lung
adenocarcinoma and integration with mRNA expression. Genome Res. Vol 22, 1197-
1211 (2012)
42. Vasiliou et al. Human ATP-binding cassette (ABC) transporter family. Hum
Genomics, Vol 3(3), 281–290 (2009)
43. Michael Dean and Tarmo Annilo. Evolution of the ATP-binding cassette (ABC)
transporter superfamily in vertebrates. Annu Rev Genomics Hum Genet, Vol 6, 123–
42 (2005)
44. Stahlman, Besnard, Wert, Weaver, Dingle, Xu, von Zychlin, Olson and Whitsett.
Expression of ABCA3 in Developing Lung and Other Tissues. Journal of
Histochemistry & Cytochemistry, Vol 55(1), 71–83 (2007)
45. Tarling E. J et al. Role of ABC transporters in lipid transport and human disease.
Trends in Endocrinology and Metabolism, Vol 24, No. 7, 342-350 (2013)
46. Jesús Perez-Gil and Timothy E. Weaver. Pulmonary Surfactant Pathophysiology:
Current Models and Open Questions. Physiology, Vol 25, 132–141 (2010)
47. J.E. Bullard, S.E. Wert, and L.M. Nogee. ABCA3 Deficiency: Neonatal Respiratory
Failure and Interstitial Lung Disease. Seminars in Perinatology, Volume 30(6), 327–
334 (2006)
48. Elisabeth Bruder et al. Ultrastructural and molecular analysis in fatal neonatal
interstitial pneumonia caused by a novel ABCA3 mutation. Modern Pathology, Vol 20,
1009–1018 (2007)
49. Jeffrey A. Whitsett et al. Expression of ABCA3 in Developing Lung and Other Tissues.
J Histochem Cytochem, Vol 55(1), 71-83 (2007)
64
50. Lawrence M. Nogee et al. Heterozygosity for ABCA3 Mutations Modifies the Severity
of Lung Disease Associated with a Surfactant Protein C Gene (SFTPC) Mutation.
Pediatric Research, Vol 62(2), 176-179 (2007)
51. Nobuhiro Ban, Yoshihiro Matsumura, Hiromichi Sakai, Yasukazu Takanezawa,
Mayumi Sasaki, Hiroyuki Arai§ and Nobuya Inagaki. ABCA3 as a Lipid
Transporter in Pulmonary Surfactant Biogenesis. The Journal of Biological Chemistry.
Vol 282, 9628-9634 (2007)
52. Cheong et al. ABCA3 is critical for lamellar body biogenesis in vivo. J. Biol. Chem. Vol
282, 23811-23817 (2007)
53. Daniel Steinbach, Jean-Pierre Gillet, Axel Sauerbrey, et al. ABCA3 as a possible
cause of drug resistance in childhood acute myeloid leukemia. Clin Cancer Res. Vol 12,
4357-4363 (2006)
54. S. Schimanski et al. Expression of the lipid transporters ABCA3 and ABCA1 is
diminished in human breast cancer tissue. Horm Metab Res, Vol 42(2) (2010), pp.
102-109
55. Vale´rie Besnard, Yan Xu, and Jeffrey A. Whitsett. Sterol response element binding
protein and thyroid transcription factor-1 (Nkx2.1) regulate Abca3 gene expression.
Lung Cell Mol Physiol, Vol 293, 1395–1405 (2007)
56. Horton et al. Combined analysis of oligonucleotide microarray data from transgenic
and knockout mice identifies direct SREBP target genes. PNAS, Vol 100, 12027–12032
(2003)
57. Iwamoto N, Abe-Dohmae S, Sato R and Yokoyama S. ABCA7 expression is
regulated by cellular cholesterol through the SREBP2 pathway and associated with
phagocytosis. J Lipid Res, Vol 47, 1915–1927 (2006)
58. Bohinski et al. The lung-specific surfactant protein B gene promoter is a target for
thyroid transcription factor 1 and hepatocyte nuclear factor 3, indicating common
factors for organ-specific gene expression along the foregut axis. Mol Cell Biol, Vol 14,
5671–5681 (1994)
59. Mario deFelice et al. TTF-1 phosphorylation is required for peripheral lung
morphogenesis, perinatal survival, and tissuespecific gene expression. J Biol Chem,
Vol 278, 35574–35583 (2003)
60. Matsuzaki, Besnard, Clark, et al. STAT3 regulates ABCA3 in type II cells. Am J Respir
Cell Mol Biol, Vol 38, 551–558 (2008)
65
61. Norbert Klugbauer and Franz Hofmann. Primary structure of a novel ABC
transporter with a chromosomal localization on the band encoding the multidrug
resistance-associated protein. FEBS Letters. Volume 391(1–2), 61–65 (1996)
62. Timothy D. Connors et al. The Cloning of a Human ABC Gene (ABC3) Mapping to
Chromosome 16p13.3. Genomics. Volume 39(2), 231–234 (1997)
63. Yamano et al. ABCA3 is a lamellar body membrane protein in human lung alveolar
type II cells. FEBS Letters. Volume 508(2), 221–225 (2001)
64. Mulugeta S, Gray JM, Notarfrancesco KL, Gonzales LW, Koval M, Feinstein SI,
Ballard PL, Fisher AB, Shuman H. Identification of LBM180, a lamellar body limiting
membrane protein of alveolar type II cells, as the ABC transporter protein ABCA3. J
Biol Chem. Vol 277(25), 22147-55 (2002)
65. Karobi Moitra and Michael Dean. Evolution of ABC transporters by gene
duplication and their role in human disease. Biol. Chem. Vol. 392, 29–37 (2011)
66. A. Sparreboom et al. Pharmacogenomics of ABC transporters and its role in cancer
chemotherapy. Drug Resistance Updates. Vol 6, 71–84 (2003)
67. Henrique et al. Epigenetic regulation of MDR1 gene through post-translational
histone modifications in prostate cancer. BMC Genomics. Vol 14 (2013)
68. Rajkumar Savai and Reinhard H. Dammann et al. ABCB4 is frequently
epigenetically silenced in human cancers and inhibits tumor growth. Scientific
Reports. Vol 4(6899), 1-9 (2014)
69. Keith D. Robertson. DNA methylation and human disease. Nature Reviews. Vol 6,
597-610 (2005)
70. Marta Kulis and Manel Esteller. Advances in Genetics. Vol 70, 27–56 (2010)
71. Robert P. Baughman and Roland M. du Bois. Diffuse Lung Disease: A Practical
Approach. Springer Science & Business Media. (2011)
72. Melanie Ehrlich. DNA hypomethylation in cancer cells; Epigenomics. Vol 1(2), 239-
259 (2009)
73. Freeman et al. ABCA3 inactivation in mice causes respiratory failure, loss of
pulmonary surfactant and depletion of lung phosphatidyl glycerol. J Lipid Res. Vol 48,
621–632 (2007)
74. Kimberly R. Kukurba and Stephen B. Montgomery. RNA Sequencing and Analysis.
Cold Spring Harb Protoc. (2015)
66
75. Stephen B Baylin. DNA methylation and gene silencing in cancer. Nature Clinical
Practice-Oncology. Vol 2, Supplement 1, S4-S11 (2005)
76. Douglas Hanahan and Robert A. Weinberg. Hallmarks of Cancer: The Next
Generation. Cell. Vol, 646-674 144 (2011)
77. Peter W Laird et al. MethyLight: a high-throughput assay to measure DNA
methylation. Nucleic Acids Research. Vol 28(8) (2000)
Abstract (if available)
Abstract
Lung cancer accounts for the majority of cancer deaths in the US and worldwide. Lung adenocarcinoma (LUAD) is the most common histological subtype of non-small cell lung cancer (NSCLC). Recent genome-wide profiling studies of LUAD tumors have identified numerous genetic alterations, particularly those involved in the RTK/RAS/RAF pathway. This had led to clinically relevant molecular subtyping of many LUAD cases, however a significant number still have unidentified driver genes. Since epigenetic alterations also underlie lung cancer development, a previous study by Selamat et al. performed a genome- scale integrative analysis using DNA methylation and mRNA expression data to identify potential epigenetic driver genes. Results from this study found ABCA3 to be one of the top candidate genes that was hypermethylated and down-regulated. ABCA3, a lipid transporter protein highly expressed in lung alveolar epithelial type 2 cells, plays an important role in regulating lung surfactant homeostasis and lamellar body biogenesis. Infants with homozygous ABCA3 mutations usually die at birth due to surfactant deficiency. Some ABCA3 mutations results in interstitial lung disease (ILD) and respiratory distress syndrome (RDS). Interestingly, ABCA3 heterozygous knockout mice exhibit hyperplasia of the lungs. With such compelling clinical relevance in lung function, we speculate ABCA3 to be a potential tumor suppressor gene (TSG) which becomes epigenetically deregulated by aberrant methylation of its promoter. In addition, the complex gene structure of the ABCA3 locus may offer some fresh insight into how this critical gene is regulated transcriptionally. Overexpression and knockdown studies are underway to address the function of ABCA3 in LUAD. To address ABCA3 regulation, we have performed preliminary luciferase assays to interrogate the transcriptional activity of different regions surrounding the ABCA3 promoter. Our data show that methylation of the region -32 to +2842 bp relative to the ABCA3 transcription start site results in higher reporter activity than the unmethylated counterpart. This region overlaps a part of the ABCA3 CpG island which shows differential methylation between tumor and normal cell lines based on reduced representation bisulfite sequencing (RRBS) data from ENCODE. Results from this study may offer new insights into the role of ABCA3 in disease and broaden our current understanding of differential methylation within promoter CpG islands.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
DNA methylation markers for blood-based detection of small cell lung cancer in mouse models
PDF
Establishing the non-coding RNA LINC00261 as a tumor suppressor in lung adenocarcinoma
PDF
DNA methylation changes in the development of lung adenocarcinoma
PDF
CpG methylation profiling in lung cancer cell lines, tumors and non-tumors
PDF
DNA methylation inhibitors and epigenetic regulation of microRNA expression
PDF
Modeling lung adenocarcinoma progression in vitro using immortalized human alveolar epithelial cells
PDF
Transcriptional regulation by epigenetic mechanisms
PDF
Investigating the role of SASH1 gene located on chromosome 6 in ovarian cancer
PDF
Application of tracing enhancer networks using epigenetic traits (TENET) to identify epigenetic deregulation in cancer
PDF
Evaluating the impact of long non-coding RNAs on tumor mutational burden in cancer
PDF
Epigenetic regulation of non CPG island gene promoters
PDF
Identification of DNA methylation markers in diffuse large B-cell lymphoma
PDF
The kinetic study of engineered MBD domain interactions with methylated DNA: insight into binding of methylated DNA by MBD2b
PDF
Identification of novel epigenetic biomarkers and microRNAs for cancer therapeutics
PDF
Tight junction protein CLDN18.1 attenuates malignant properties and related signaling pathways of human lung adenocarcinoma in vivo and in vitro
PDF
Effects of chromatin regulators during carcinogenesis
PDF
Determinination of the causal potential of histone modifications on transcription and chromatin structure
PDF
Do ZFX and ZNF711 regulate the same genes in HEK293T cells?
PDF
CpG poor promoter SULT1C2 regulated by DNA methylation and is induced by cigarette smoke condensate in lung cell lines
PDF
Functional characterization of a prostate cancer risk region
Asset Metadata
Creator
Ankala, Ramya S.
(author)
Core Title
Investigating the function and epigenetic regulation of ABCA3, a novel LUAD tumor suppressor gene
School
Keck School of Medicine
Degree
Master of Science
Degree Program
Biochemistry and Molecular Biology
Publication Date
07/30/2015
Defense Date
06/11/2015
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
ABCA17p,ABCA3,epigenetic regulation,epigenetics,lung adenocarcinoma,OAI-PMH Harvest,psuedogenes
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Laird-Offringa, Ite A. (
committee member
), Tokes, Zoltan A. (
committee member
), Weisenberger, Daniel (
committee member
)
Creator Email
ramyaswetha.ars@gmail.com,rankala@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-614637
Unique identifier
UC11303924
Identifier
etd-AnkalaRamy-3750.pdf (filename),usctheses-c3-614637 (legacy record id)
Legacy Identifier
etd-AnkalaRamy-3750.pdf
Dmrecord
614637
Document Type
Thesis
Format
application/pdf (imt)
Rights
Ankala, Ramya S.
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
Tags
ABCA17p
ABCA3
epigenetic regulation
epigenetics
lung adenocarcinoma
psuedogenes